Aci sp-223-2004

daneshlink.
com
Investigating Concrete:
Selected Works of
Bryant and Katharine Mather
--``,```,`,`,`,,,,`,`````,`,,`,,-`-`,,`,,`,`,,`---
Together, Katharine and Bryant Mather have contributed over I 00 years of scientific work for concrete
research and development activity for the U.S. Army Corps of Engineers and have participated
in many technical organizations, such as the American Concrete Institute (ACI), the American
Society for Testing and Materials (ASTM), the Transportation Research Board (TRB), and others.
Editors:
Celik Ozyildirim
Shuaib Ahmad American Concrete Institute®
Advancing concrete knowledge
Copyright American Concrete Institute
Provided by IHS under license with ACI
en 'l'l'-J
Licensee=University of Texas Revised Sub Account/5620001114, User=erur, ert
No reproduction or networking permitted without license from IHS Not for Resale, 01/26/2015 02:30:03 MST
Daneshlink.com
daneshlink.com
--``,```,`,`,`,,,,`,`````,`,,`,,-`-`,,`,,`,`,,`---

Provided by IHS under license with ACI Licensee=University of Texas Revised Sub Account/5620001114, User=erur, ert
Daneshlink.com
daneshlink.com
ACI Special Publication
INVESTIGATING CONCRETE
--``,```,`,`,`,,,,`,`````,`,,`,,-`-`,,`,,`,`,,`---
Selected Works
of Bryant and Katharine Mather
Editors:
Celik Ozyildirim
Shuaib Ahmad
American Concrete Institute@;

Advancing concrete knowledge
ACI SP-223

Daneshlink.com
daneshlink.com
First printing, October 2004
Copyright © 2004
AMERICAN CONCRETE INSTITUTE
P.O. Box 9094
Farmington Hills, Michigan 48333-9094
All rights reserved, including rights of reproduction and use in any form or by any means, including the making of
copies by any photo process, or by any electronic or mechanical device, printed or written or oral, or recording for
sound or visual reproduction or for use in any knowledge or retrieval system or device, unless permission in writing
is obtained from the copyright proprietors.
Cover photo: Tuscaloosa Dam
Printed in the United States of America
Editorial production: Shannon B. Banchero
Library of Congress Control Number: 2004114043

ISBN: 0-87031-158-1
--``,```,`,`,`,,,,`,`````,`,,`,,-`-`,,`,,`,`,,`---

Daneshlink.com
daneshlink.com
ACKNOWLEDGEMENTS
This commemorative publication is the result of contributions from a number of individuals. Ed

Nawy originated the idea of commemorative compilations to honor individuals who have made
significant contributions to concrete technology through their distinguished work and contributions.
The idea was embraced by the Technical Activities Committee (TAC) of ACI. For this
commemorative publication, Bryant Mather worked with Celik Ozyildirim and Shuaib Ahmad to
compile a list of Bryant and Katharine's publications. This was a daunting task.
Gratitude and appreciation are extended to external reviewers Richard Gaynor, Terence Holland,
Steven Kosmatka, Joseph Lamond, and James Pierce, who offered very valuable comments and
suggestions.
The editors express their deep appreciation to Shannon Banchero of the Engineering Staff at ACI
for working with various publishers and compiling the selected papers for this book. Her meticulous
work is greatly appreciated.
Finally, appreciation is extended to all those publishers and organizations that permitted ACI to
reproduce the original papers and articles of Katharine and Bryant Mather in this book. They are as
follows: American Society for Testing and Materials, Elsevier, Transportation Research Board,
National Ready-Mixed Concrete Association, RILEM, and Thomas Telford.
--``,```,`,`,`,,,,`,`````,`,,`,,-`-`,,`,,`,`,,`---
3
Daneshlink.com
daneshlink.com
--``,```,`,`,`,,,,`,`````,`,,`,,-`-`,,`,,`,`,,`---

Daneshlink.com
daneshlink.com
FOREWORD
In October 2000, the American Concrete Institute's Technical Activities Committee approved
publication of a commemorative compilation honoring Bryant Mather for his significant
contributions to concrete technology. This publication, Investigating Concrete-Selected Works ofBryant
and Katharine Mather, is the result of a cooperative effort by Mather, Celik Ozyildirim, and ACI Chief
Engineer Shuaib Ahmad.
--``,```,`,`,`,,,,`,`````,`,,`,,-`-`,,`,,`,`,,`---
It is divided into seven parts: potential of concrete; cementitious materials; aggregates; durability of
concrete; petrographic examination of aggregates and concrete; specifications for use of concrete in
transportation structures; and research and emerging technologies. This publication also includes an
appendix with a bibliography of selected works of Bryant and Katharine Mather. Bryant helped in
developing a comprehensive list of their publications. Per Bryant's request, Katharine was added to
the title and several of her papers were added to this publication. The total number of publications
was in excess of 800. The list was reduced to 400 by identifying papers dealing with concrete
materials only. Bryant, Celik Ozyildirim, and Shuaib Ahmad met at ACI Headquarters to further
reduce the list to a more manageable 164 publications. This reduced list of 164 citations is included
in the appendix of this publication.
Together, Katharine and Bryant Mather have contributed over 100 years of scientific work for
concrete research and development activity for the U.S. Army Corps of Engineers and have
participated in many technical organizations, such as the American Society for Testing and Materials
(ASTM), Transportation Research Board (fRB), American Concrete Institute (ACI), and others.
Their work has significantly contributed to the advancement and better understanding of concrete.
On September 25,2001, the U.S. Army Engineer Research and Development Center renamed its
structures laboratory building the Katharine and Bryant Mather Building.
Bryant started his career in August 1941, when he accepted a job as a junior geologist with the
Central Concrete Laboratory at the U.S. Military Academy at West Point, N.Y. Not long thereafter,
Bryant began to examine aggregate to associate rock types with "soundness."
Katharine started her career at the same laboratory in April 1942. She also had assignments
performing microscopic petrography. She, like Bryant, had more semester hours of petrography
during her graduate work at Johns Hopkins University than any other geological specialty. Both
Bryant and Katharine were assigned to work on the preparation of the Handbook for Concrete and
Cement-a compilation of test methods, specifications, terminology, and standards in concrete
technology developed or modified and adopted for use by the Corps of Engineers.
Throughout their careers, Bryant and Katharine made important contributions to concrete
petrography. They co-authored a paper on petrographic methods for concrete aggregates, which
was the basis for ASTM C 295, adopted in 1954. As of 2002, ASTM C 295 still cites this paper as its
first reference. Katharine Mather compiled the contributions made by the use of light microscopy
to concrete research, which was published as part of a symposium, sponsored by the ASTM
committee on microscopy. The paper was selected by ASTM Committee C 09, Concrete and
Concrete Aggregates, to receive the Sanford E. Thompson Award for 1961.

5
Daneshlink.com
daneshlink.com
Katharine later collaborated with Tom Kennedy to produce petrographic data on concrete exposed
to freezing and thawing at Treat Island, Maine. She compared the data with specimens of the same
batches subjected to freezing and thawing in the laboratory. This paper was published in theACI
Journal and was selected for the Wason Medal in 1955.
When the Mathers joined the Corps of Engineers' Concrete Laboratory in 1941, the principal
research effort at the laboratory was the Cement Durability Program, in which durability primarily
referred to resistance to freezing and thawing.
Bryant further discussed the subject of freezing and thawing in his paper "How to Make Concrete
That Will be Immune to the Effects of Freezing and Thawing," which was published in ACI SP-
122, Paul KliegerSymposium on Peiformance of Concrete.
Over the last 50 years, the contributions of Bryant and Katharine have been recognized in many
ways by various organizations, including ACI, ASTM, and TRB.
The Proceedings of the Katharine and Bryant Mather International Coiference co-sponsored by the
American Concrete Institute and published as ACI SP-100, Concrete Durabiliry: Katharine and Bryant
Mather International Conference, consists of two volumes totaling 2,179 pages. One of Bryant's last
contributions to the concrete industry was his tireless work on the revision of ACI SP-1, Concrete
Primer, which was published by the American Concrete Institute in 2002.
The editors hope that this commemorative publication to honor Katharine and Bryant Mather is
well received and provides a good comprehensive overview of their distinguished contribution to
the concrete industry.
--``,```,`,`,`,,,,`,`````,`,,`,,-`-`,,`,,`,`,,`---
6
Daneshlink.com
daneshlink.com
TABLE OF CONTENTS
Part I - Potential of Concrete ........................................................................................................... 9

Reali:;;jng the Potential of Concrete as a Construction Material. ................................................................ II
Part II - Cementitious Materials .................................................................................................. 21
Effectiveness ifMineral Admixtures in Preventing Excessive Expansion of Concrete Due to Alkali-
Aggregate Reaction....................................................................................................................................... 23
Part III - Aggregates ........................................................................................................................... 55
Concrete Aggregate: Shape, Surface, Texture, and Coatings.......................................................................... 57
Part IV- Durability of Concrete................................................................................................... 79
Reflections on Concrete Durabiliry and on International Conferences on Concrete Durabiliry......................... 81
Resistance to Freezing and Thawing
How to Make Concrete that Will Be Immune to the Effects ifFree:;;jng and Thawing.................................. 89
Crystal Growth in Entrained-Air Voids.................................................................................................. 105
Resistance to Alkali-Aggregate and Alkali-Silica Reactions
Cracking if Concrete in the Tuscaloosa Lock............................................................................................. 107
How to Make Concrete That Wzll Not Stiffer Deleterious Alkali-Szlica Reaction..................................... 127
Resistance to Sulfate Attack
Sulfate Soundness, Sulfate Attack, and Expansive Cement in Concrete.................................................... 135
Sulfate Attack on Hydraulic-Cement Concrete ......................................................................................... 143
Part V - Petrographic Examination of Aggregates and Concrete......................... I 53
Method if Petrographic Examination ifAggregates for Concrete............................................................... 155
Applications ifLight Microscopy in Concrete Research.............................................................................. 185
Part VI- Specifications for use of Concrete in Transportation Structures ..... .207
Concrete in Transportation: Desired Performance and Specifications........................................................... 209
Part VII- Research and Emerging Technologies .......................................................... 219
Research on Concrete ................................................................................................................................ 221
High-Performance Concrete in the U.S. Army Corps ifEngineers ............................................................ 231
&bert E. ue and Concrete ..................................................................................................................... 239
Self-Curing Concrete, W~ Not? .............................................................................................................. 245
Appendix - Selected Bibliography of Published Works of Katharine and
Bryant Mather.......................................................................................................................................... 247
--``,```,`,`,`,,,,`,`````,`,,`,,-`-`,,`,,`,`,,`---
7
Daneshlink.com
daneshlink.com
--``,```,`,`,`,,,,`,`````,`,,`,,-`-`,,`,,`,`,,`---

Daneshlink.com
daneshlink.com
PART 1- POTENTIAL OF CONCRETE

--``,```,`,`,`,,,,`,`````,`,,`,,-`-`,,`,,`,`,,`---
Bryant's vision of concrete materials and their potential did not have limits. He believed that
concrete was the choice material for construction. To help realize the potential of concrete as a
construction material, he considered the ethical, social, and environmental issues. He worked
endlessly to teach the professionals of the concrete community that concrete is made from local
materials and local know-how and therefore has infinite variability. He also stressed that if certain
factors are carefully considered, concrete with very uniform characteristics can be achieved and can
be made to last as long as one desires it to last.
Bryant's paper, Realiifng the Potential of Concrete as a Construction Material, articulates the potential of
concrete in a unique fashion. In this paper, Bryant concluded that the potential of concrete as a
construction material will be better realized when in his words:
1. We improve our procedures for describing what we want so we don't pay for behavior we
don't need and can't use, i.e., we learn to specify intelligently.
2. We employ readily available procedures for describing, understanding, and controlling
variability of concrete.
3. We consider the ethical aspects of construction and especially the environmental aspects of
concrete production and use.
4. We find out and act to alleviate the real-life problems of the people who are out there trying
to make good concrete.

9
Daneshlink.com
daneshlink.com
--``,```,`,`,`,,,,`,`````,`,,`,,-`-`,,`,,`,`,,`---

Daneshlink.com
daneshlink.com
Reprinted, with permission, from &aliifng the Potential of Concrete: A Construction Material in Creating With Concrete, V. 1,
©Thomas Telford Publishing, 1 Heron Quay, London E14 4JD, 1999.
REALIZING THE POTENTIAL OF CONCRETE AS A CONSTRUCTION

MATERIAL
by Bryant Mather1
ABSTRACT
Concrete is international, but made locally; has infinite variability, but can be made very uniform; and
can be made to last as long as you want it to. Therefore, what is needed to more fully realize its
potential as a construction material is to understand what we want it to do, learn how to make it so it
will do so, use available methods to restrict undesired variability, consider the ethical and environmental
aspects of its use, and help the people who are making it to do it better.
Keywords
Concrete; Durability; Environment; Ethics; Service life; Specification overkill; Statistical methods;
Uniformity; Variability
Dr. Bryant Mather is Director of the Structures Laboratory, U.S. Army Engineer Waterways
Experiment Station, 3909 Halls Ferry Road, Vicksburg, MS 39180-6199, USA. He is a Past-President of
ACI and ASTM and has served on numerous ACI and ASTM committees. He is a prolific author and
editor and has received numerous awards and honors from ACI, ASTM, and other institutions.
INTRODUCTION
When the question of "realizing the potential of concrete as a construction material" was suggested to
me as the subject of these remarks, my £rst reaction was, what is there that has been built using
--``,```,`,`,`,,,,`,`````,`,,`,,-`-`,,`,,`,`,,`---
construction materials that hasn't been built using concrete? What unrealized potential does concrete
have as a construction material? Or perhaps one fust needs to de£ne "construction"- and then
"construction material." Stein (1993) includes in "Construction Categories" a list of building types and
uses ranging from "Animal Clinic," "Bank," "Dog I<ennel," "Factory," "Garage," "Jail," "Museum,"
"Roundhouse," "Stable," and "Theater," to "Warehouse." I believe that examples of all of these have
been built using concrete.
I therefore concluded that perhaps the opportunity presented by this title is not so much to talk about
how to construct some category of construction that has never before been produced using concrete,
but rather describing how to do concrete work of the kind that the human race has been engaged in to
some degree or other for the past eight millennia (Malinowski and Friefelt 1993) but doing it better so
as to serve better, at lower social cost. We need, inter alia, to learn how to better realize the potential of
concrete as a construction material by producing it with the use of less energy and smaller emissions of
greenhouse gases.
1
U.S. Army Engineer Waterways Experiment Station, United States of America

11
Daneshlink.com
daneshlink.com
Even as was the case with these remarks, in 1987 the organizers of the First International Conference
on Concrete Durability invited Katharine Mather and me to speak at the opening session. On that
occasion, the version that was published had the tide "Reflections on Concrete Durability and on
International Conferences on Concrete Durability" (Mather and Mather 1987). In it, we reiterated a
number of well-known relevant facts about concrete. I restate them in slighdy revised form as an
introduction.
Concrete is international and as we have air to breathe. water to drink. earth to grow plants in: it
provides for all that which the 1976 National Ready Mixed Concrete Association's U.S. Bicentennial
bumper sticker said:- "Concrete- the Foundation of America." and that can now be expanded to
"Concrete- the Foundation of Civilization."
Concrete is made locally-The mixing water, with negligible exceptions (perhaps when concrete is
made on the moon); the aggregates usually; the cement often, are or are made from one or more locally
available natural raw materials, after various degrees of processing.
--``,```,`,`,`,,,,`,`````,`,,`,,-`-`,,`,,`,`,,`---
Concrete has infinite variability-The varieties of concrete that are intentionally produced are more
numerous than the known varieties of natural sedimentary rocks. Concretes have been - and will be
-required to be produced having densities from as low as 160 kg/m3 (10 lb mass/fe) to as high as
4800 kg/m3 (300 lb mass/ft\ that have unconfined compressive strengths from as low as 0.35 MPa (50
psi) to as high as 275 MPa (40,000 psi); that are of any color- white, black, pink, turquoise,
chartreuse, or as one Corps of Engineers officer demanded, "sand tan;" we await the demand for "olive
drab."
Concrete can be made to be vet;y uniform-ACI Committee 214 was organized in 1946. It told
concrete makers and users how to use strength tests to measure uniformity of concrete as produced and
still does. In 1979, ASTM Standard Method C 917 for Evaluation of Cement Strength Uniformity was
adopted. For many years ASTM Specification C 33 has put rather restrictive limits on uniformity of
grading of fine aggregate in terms of fineness modulus. There are uniformity requirements on pozzolans
given in ASTM Specification C 618.
Concrete can be made to last as long as you want it to-Even as some relatively fragile sedimentary
rocks exist that have survived some hundreds of millions of years, so there are concretes that, if they
remain in relatively mild environments, could last forever or at least as close to forever as we can
measure. The duration of the life of a concrete with its relevant properties not sufficiendy degraded so
as to impair or jeopardize its ability to serve its intended purpose, is not usually a function of the
absolute level of any universally relevant property of the concrete such as slump, density, strength,
modulus of elasticity, or color but is almost always a function of the interaction of some property or
properties, the required levels of which depend on the severity of the environment to which the
concrete will be exposed in service. One need not worry much about resistance to freezing and thawing
in the tropics, except in facilities using refrigeration; or at the poles, except in facilities using intermittent
heating. One need not worry about acid attack, abrasion attack, sulfate attack, attack by boring or
burrowing molluscs, or any of the myriad hazards described by Kleinlogel (1950), Biczok (1976), or
others except where they exist and impact the concrete of interest. And where these potentially
destructive influences can be and are brought to bear on the concrete, if their action and its intensity
and duration are property anticipated, the concrete can be rendered adequately resistant so as to merit
the designation durable, even when as demanded of us by authorities dealing with the storage of nuclear
waste, means not significandy degraded for many centuries or in some cases hundreds of centuries.

12
Daneshlink.com
daneshlink.com
With these considerations in mind, what is there to say about the unrealized potential of concrete as a
construction material? In my view, several things: First, what do we need to do that we are not doing
that, if we did it, would greatly reduce the number of cases of premature degradation of concrete in
service. Too much concrete that should last at least SO or 100 years with minimal damage is showing
serious damage in a tenth or less of that time. And, next, what do we need to do that we are not doing
that if we did it, would greatly reduce the cost of concrete by not paying extra to provide resistance to
aggressive influences that have negligible likelihood of being encountered by the concrete in its
intended service life. We now, often, pay a lot to provide frost-resistant concrete for use in regions
where natural freezing has not taken place for millions of years. If we only ordered what we need and
can benefit from, we'd have fewer things to look out for through quality control and quality assurance,
and that, too, would contribute to lowering the cost.
One important matter, therefore, is to know how properly to assess the nature and severity of the
environment of service; to get a reasonable estimate of intended service life; to get a reasonable
assessment of how much degradation is tolerable structurally and, when relevant, aesthetically. One
must then, from these and known relationships of environmental influences and concrete composition
and constitution, have and use procedures and criteria by which to select materials, proportions, and
construction practices that will, with reasonable probability of success and compliance with the owner's
budget for the work, give concrete that will not prematurely deteriorate.
Most of the knowledge to do this exists, some in relatively quite refmed states. But the manner in which
this knowledge is used in an attempt to make it an effective tool for makers and users of concrete is
primitive. Today we talk about reactive vs. non-reactive aggregate and high-alkali vs. not high-alkali
cement as if all structures, in all environments, involving all ambient temperatures and temperature
ranges, in all degrees of moistness or dryness needed the same level of protection. There is no
relaxation of the specified limit of 0.60 percent equivalent Na2 0 maximum on cement as a function of
environmental severity, degree of reactivity of aggregate, or anticipated level of permeability of
concrete. An aggregate is either reactive or not; if not, nothing need be done, no precautions taken; if
reactive, then the precautions must be taken without reference to degree of reactivity or degree of
severity of exposure.
Today we talk about need for resistance to freezing and thawing. If the concrete will ever likely freeze
when any of the spaces in it that can hold freezable water are more than 91 percent water filled, we
need air-entrainment, sound aggregate, and protection against freezing until moderate maturity is
achieved. And, air-entrainment means an air content that will give an air-void system with a bubble-
spacing factor, I, less than 0.2 mm (0.008 in.) regardless of whether we have one or one hundred cycles
of freezing and thawing per year and regardless of whether the rate of freezing is fast or slow or
whether the minimum temperature achieved 10 mm below the exposed concrete surface is minus S °C
or minus SO oc.
In 1987 (Mather and Mather), we decided to refer to these sorts of approaches as specification overkill.
It involves, much of the time, paying more than needs to be paid to buy a level of protection that, in
many cases, cannot ever be needed and from which no good can ever be derived. Ever in the preceding
sentence means not more often than once in a million years. Today we have computers that can be run
by most bright junior-high-school scientists that could be used to take the knowledge that does exist
and derive from it reasonable criteria for matching degree of precautions needed or degree of restriction
required of the contractor in his exercise of choice among materials, mixture proportions, and practices,
--``,```,`,`,`,,,,`,`````,`,,`,,-`-`,,`,,`,`,,`---

13
Daneshlink.com
daneshlink.com
so that only limited precautions and light restrictions are called for when the nature and intensity of the
anticipated environmental hazards and insults the concrete must resist are slight or of short duration in
time. We can then reserve the levels of precautions we now take for all levels of severity for those cases
where the hazard is really very severe and protection is needed for a very long time. We might well, by
so limiting these requirements in their applicability, be able to increase their degree of protection so as
further to minimize the probability that they will ever be less effective than needed. My challenge
therefore is: Because concrete need not deteriorate, ensure that it doesn't, but do so without raising the
cost more than necessary; stamp out specification overkill.
VARIABILITY
Another area in which there is significant room for improvement in concrete technology in practice in
the degree to which we realize the potential of concrete as a construction material is the application of
the theory of probability to the understanding of, specification of, and interpretation of test data on the
measured properties of concrete. The application of the theory of probability to the ready-mixed
concrete industry was described by Stanton Walker (1944, 1944a) in a section of the publication Control
ofQualiry ofReacfy-Mixed Concrete entitled "Over-Design to Meet Specifications." Tables were given
showing the average strength required for operations with different coefficients of variation so that
there will be nine chances out of 10 or 99 chances out of 100 that no test will fall below the specified
strength. He remarked that these relations "demonstrate forcibly that investment in good control is self-
liquidating." In 1946, the ACI Advisory Committee, under his chairmanship, recommended the
--``,```,`,`,`,,,,`,`````,`,,`,,-`-`,,`,,`,`,,`---
formation of a new technical committee entitled, "Evaluation of Results of Compression Tests of Field
Concrete." The mission of the committee was stated as follows: The assignment is to analyze data of
compressive tests of samples of concrete taken in the field to determine the nature and extent of
variations in them for different job conditions and to develop measures of the reliability of such tests as
bases for the design of concrete proportions. (Taken from a letter from Stanton Walker to Harvey
Whipple, November 27, 1946.)
The committee produced a standard which as of 1965 was called "ACI Standard Recommended
Practice for Evaluation of Compression Test Results of Field Concrete" (ACI 214-65), which had as its
abstract:
"Statistical methods provide valuable tools for assessing results of strength tests, and
such information is also of value in refming design criteria and specifications. The report
discusses briefly the numerous variations that occur in the strength of concrete and
presents statistical methods that are useful in interpreting these variations. Criteria are
offered that can be used to establish specifications and maintain required uniformity. An
appendix presents a simplified version of statistical quality-control procedures."
In 1973 ACI SP 37, "Realism in the Application of ACI Standard 214-65," was
published. The preface to this SP included the following:
Standard ACI 214, since its first promulgation, has produced something of a
revolution in the field of evaluation of concrete strength tests, and, when applied
properly, has resulted in considerable improvement in the quality, and especially the
uniformity, of the concrete on projects where it was used. For reasons that are
explored in some of the papers in this symposium, however, acceptance of the ideas

14
Daneshlink.com
daneshlink.com
and techniques contained in the Standard has been slow and reluctant in some
quarters. This situation has been the subject of much discussion at meetings of
Committee ACI 214 in recent years.
At the meeting in Denver, in March 1971, the conclusion was reached that what was
needed was a strong program of education. As a first step in this direction, the
decision was made to hold a symposium at the fall meeting. A committee consisting
of Howard T. Arni, Federal Highway Administration, Co-chairman; Edward A.
Abdun-Nur, Consulting Engineer, Co-chairman; Francis J. Principe, Principe-Dana,
Incorporated; and V. M. Malhotra, Canada Department of Energy, Mines, and
Resources, was appointed to organize the symposium.
The Symposium was held on November 4,1971, in Buffalo, New York.
Mr. Abdun-Nur's two papers, "Control of Concrete Mixes," which was co-authored
by Joseph J. Waddell and appeared in the March 1959 ACI Journal, and "How Good
Is Good Enough?" in the January 1962 ACI Journal, have become classics in the field
of evaluation of strength tests and are still extremely helpful to those who wish to
make use of Standard ACI 214. The committee hopes that the inclusion of these two
papers and of Standard ACI 214 itself as Appendices along with the other papers will
make this volume a very useful guide to those who are involved in problems of
evaluation of strength tests of concrete.
Regrettably, much of what was said in 1973 about the application of probability is still true. What was
said about what was needed is ... education is still true - and always will be true. We could all benefit
from reading Ed Abdun-Nur's paper on "How Good Is Good Enough?" again (Abdun-Nur 1962).
In 1993 I was invited to write a Foreword to a book called Concrete Mix Design, Qualiry Control and
Specification by Ken Day (1995); in it I said:
"I was once required to attend night courses at the laboratory in which the Director,
--``,```,`,`,`,,,,`,`````,`,,`,,-`-`,,`,,`,`,,`---
Mr. Charles E. Wuerpel, taught a course in concrete mix design. I failed to learn to
do it - too much arithmetic. This was in 1942. This book, I believe, tells the reader
that there is a great deal more arithmetic now than then, but don't worry; the
computer will take care of getting it done, provided you instruct the computer
properly.
"Somewhat later I became convinced that most of the time one could make concrete
of appropriate quality for a given use with locally available materials. I once upset
some nuclear physicists by suggesting that 'sidewalk' concrete from the local
neighborhood ready-mix plant, if uniform, would make just as good a biological
shield around a nuclear reactor as concrete with high-density aggregate imported
from a thousand miles away. I suggested that the extra volume needed would cost
less than the shipping cost of a smaller volume of expensive aggregates. Ken Day
understands this (section 1.2).
"It has been one of my theses that properly proportioned concrete cannot be over
vibrated -but one needs to proportion concrete for 'excessive' vibration if one

15
Daneshlink.com
daneshlink.com
wants to vibrate it excessively. Ken knows this too (section 1.3). I have also argued
that one can design structures (including pavements) for whatever loading is relevant
and then select a design concrete strength for that loading (compressive, flexural, or
whatever) and proportion a mixture to achieve it; but having done so, it is dumb not
to use compressive strength for routine control testing; Ken says this (section 1.7).
He knows that hot weather does not direcdy increase water demand, he just implies
it like everybody else. Unlike everybody else, however, he has a proposal ('equivalent
slump,' section 12.1) to get to the truth of the matter, as I have been urging for many
years.
"The concept of distinguishing acceptable non-conforming concrete from

unacceptable non-conforming concrete that Ken calls 'structurally defective' and
'contractually defective' needs more emphasis; and this book provides it.
"It is good to be reminded of the fact that when worrying about the 'actual' strength
of concrete in a structure it is quite proper to worry about its present strength under
some conditions and its eventual strength under others (section 5.15). It is also good
to read in section 6. 1: 'Fortunately the pressure to specify minimum cement
contents was resisted.' We had the batde of New Orleans about this in the ACI and a
rule was made that if there were to be a minimum cement content there must also be
an alternative performance requirement that if met would obviate the need for the
prescriptive one.
"I was very happy to see 'Waterproofers' put into quotation marks, because the term implies
doing something that is not possible.
"For these and other reasons I am pleased to commend this book to its readers, in spite of my
belief that, in concrete technology, mix should only be used as a verb and design should only
refer to selection of properties of concrete structural elements and hence the tide should have
been 'Concrete Mixture Proportioning."'
I am told that a new edition of this may appear in 1999.
SOCIAUETHICAL ISSUES
In April 1995 the Netherlands Cement industry in its magazine Cement included the following: "Does
the future of concrete have anything to do with ethics? Any human activity has an ethical dimension.
This is true of science and technology, as therefore for civil engineering too. The future of concrete will
not be affected so much by ethical considerations, but mainly by technical, economic, social, and
ecological developments. Nevertheless ethical guidelines can contribute to the attitudes adopted. If the
construction industry acts responsibly, with a keen realization of the contribution it can make, concrete
as a construction material will have an important role in the future. At the beginning of any technical
venture all those involved should ask themselves in principle: 'Will this do more harm than good?' In
the many cases where those involved do not consider it incumbent upon them to stand on the sidelines,
they should follow some ethical guidelines, which apply to all areas of science and technology."
--``,```,`,`,`,,,,`,`````,`,,`,,-`-`,,`,,`,`,,`---

16
Daneshlink.com
daneshlink.com
Among the questions that are properly asked about concrete as a construction material, in addition to
the questions about whether what it is to be used to build will do more harm than good - e.g., ICBMS,
nuclear power plants, gambling casinos, etc. - is the question of whether concrete or some alternative
construction material does more harm to the environment in its production and use. In the United
States, there has been created an Environmental Council of Concrete Organizations (ECCO) with its
headquarters at the Portland Cement Association. It has produced a folder called "The Surprising
Environmental Benefits of Cement" which points out that unlike building materials that rely on scarce
or non-renewable resources, concrete is made from three abundant, readily available ingredients: (1)
water, (2) aggregate (stone, sand, and gravel, and (3) portland cement (a fine gray powder made from
abundant natural materials). A growing list of recycled materials supplements these three basic
ingredients. In addition to stone, sand, and gravel, aggregate may contain some recycled materials, such
--``,```,`,`,`,,,,`,`````,`,,`,,-`-`,,`,,`,`,,`---
as slag, a by-product of steel-making. And part of the portland cement can be replaced by other
materials, such as fly ash, a by-product of coal-burning power plants.
Even the actual process of making cement uses recycled materials. High-energy wastes such as old tyres
can be safely recycled as fuel for the cement-making process. Each year, a single cement kiln can recycle
one million tyres, conserving fossil fuels and placing less demand on landfill space. It then points out
that when you consider that concrete is strong; adaptable; and resistant to fire, water, and weather;
there's little wonder it has a longer service life than wood, steel, asphalt, and other construction
products. Concrete's durability conserves resources by reducing maintenance and the need for
reconstruction. And at the end of its initial service life, concrete can be crushed and reused as aggregate
for new concrete, continuing the cycle of environmental benefits. The folder also notes that the thermal
mass of concrete buildings saves energy year-round by reducing temperature swings.
During the cooling season, the building will require cooling mainly at night - an off-peak time when
electric companies can produce power more efficiently and conserve fossils fuels. During the heating
season, a concrete building will cost less to heat than a comparable structure made of different
materials. And these benefits last as long as the building itself- which in the case of concrete
structures is a very long time.
It is concluded that concrete is one of the single most environmentally friendly construction products
currently available.
In another connection, I was once asked to speak to the question of whether or not I was aware of any
structure built of concrete by the Corps of Engineers that would have been less of an insult to its
environment had it been built of a different construction material. After deliberation and consultation, I
concluded that no Corps of Engineers concrete structure I could think of would have been more
environmentally friendly if made of some other construction material. When I made this point in a talk,
a member of the audience told me of a concrete repair to the tailrace from a penstock at a hydroelectric
power darn where when the repair was completed it was noted that fish tended to assemble close to the
new concrete surface. I have not been given any explanation of this, but I regard it as tending to
validate my conclusion. I suggest that considering ethical, social, environmental ("green") issues will, to
an increasing degree, help in the realization of the potential of concrete as a construction material.
17
Daneshlink.com
daneshlink.com
PROBLEMS
In 1997 the National Ready Mixed Concrete Association surveyed its membership on technical and
production-related problems. Highly rated problems were several quality assurance issues, including
improperly conducted fresh and hardened concrete testing and obtaining reports from commercial
laboratories. Problems relating to materials listed aggregates and cements as being variable and a lack of
communication by the material suppliers when the characteristics of these materials changed. Other
highly rated problems related to controlling air content and slump, primarily due to changing
compatibility of cements and admixtures. On specification issues, a significant problem was dealing
with unreasonable requirements, such as inconsistent prescription and performance requirements.
When asked to describe a recent technical or production problem, about 40% indicated that they were
dealing with variations in materials and mixture performance, such as setting time, slump, strength, and
air content. Another 30% were dealing with customer complaints regarding concrete flatwork and
finishing problems, often related to improper concreting practices. About 20% of the responses related
to complying with unreasonable and conflicting specification requirements, such as specifications for
specific brand materials, significandy over-designed mixtures, and other durability-related test
qualification.
About 30% of production problems related to improper functioning of plant and truck equipment and
control systems. Obtaining and retaining trained plant personnel and truck drivers was cited as one of
the primary problems. About 10% of the comments related to compliance with environmental
regulations and the problems with handling returned concrete, wash water, and storm water.
When asked about how things could be improved, 40% indicated that there needs to be some means to
educate and inform engineers, architects, and other specifiers about following national standards while
allowing flexibility for local conditions and materials; standardize the submittal process; move toward
performance-based specifications; and promote the use of concrete. About 20% indicated that testing
issues should be addressed by improving and informing about standard procedures and the qualification
of testing personnel. More educational and training programs for ready-mixed concrete producers,
including sales training; dealing with material variability; mix designs for high-performance concrete;
and other quality-control functions were stressed by about 15% of the responses. Addressing durability
issues such as scaling and other flatwork problems; alkali-silica reactivity; and concrete permeability
specifications were also mentioned. About 9% also indicated that contractor and customer education
should be made available, while another 9% indicated that effort should be expended toward updating
guides and standards in ACI, ASTM, and other organizations.
CONCLUSIONS
The potential of concrete as a construction material will be better realized when:
1. We improve our procedures for describing what we want so we don't pay for behavior we don't
need and can't use, i.e., we learn to specify intelligendy.
2. We employ readily available procedures for describing, understanding, and controlling variability
of concrete.
3. We consider the ethical aspects of construction and especially the environmental aspects of
concrete production and use.
--``,```,`,`,`,,,,`,`````,`,,`,,-`-`,,`,,`,`,,`---
18
Daneshlink.com
daneshlink.com
4. We find out and act to alleviate the real life problems of the people who are out there
trying to make good concrete.
REFERENCES
1) Abdun-Nur, E. A., 1962, "How Good Is Good Enough?" Journal, American Concrete Institute,
Proceedings, V. 59, pp. 31-46 (reprinted in ACI SP 37, 1973, pp 173-187).
2) American Concrete Institute, 1973, Realism in the Application rifACI Standard 214-65, ACI SP-37,
American Concrete Institute, Detroit, Mich., 215 pp.
3) American Concrete Institute, 1998, "Recommended Practice for Evaluation of Strength Test Results
of Concrete," ACI 214-77 (Reapproved 1997), ACI Manual rif Concrete Practice, Part 2, 14 pp.
--``,```,`,`,`,,,,`,`````,`,,`,,-`-`,,`,,`,`,,`---
4) American Society for Testing and Materials, 1998, "Standard Specification for Concrete Aggregates,"
V. 04.02: C 33, C 33-97, pp. 10-16; C 618: "Standard Specification for Coal Fly Ash or Calcined Natural
Pozzolan for Use as a Mineral Admixture in Concrete," C 618-98, pp. 298-300.
5) American Society for Testing and Materials, 1998, "Standard Test Method for Evaluation of Cement
Strength Uniformity from a Single Source," V. 04.01: C 917, C 917-95a, pp. 417-424.
6) Biczok, I., 197 6, Concrete Corrosion and Concrete Protection, 8th Ed, Akademiai Kiado, 545 pp, Budapest.
7) Kleinlogel, A., 1950, Einfliisse atgBeton and Stahlbeton, 5th Ed, revised by Walz, K., and Vierheller, H.,
Verlag von Wilhelm Ernst & Sohn, 339 pp, Berlin.
8) Malinowski, R., and Friefelt, K, 1993, "The Ubaid Plaster ofDosariyah (5-4000 Years BC)," Swedish
Council for Building Research, Prehistoric Hydraulic Mortar, D12,16 pp, Stockholm.
9) Mather, K., and Mather, B., 1987, "Reflections on Concrete Durability and on International
Conferences on Concrete Durability," Concrete Durabiliry, Scanlon,J.M., ed.,ACI SP-100, pp. 1-5,2 V.,
2179 pp., American Concrete Institute, Detroit, Mich.,.
10) National Ready Mixed Concrete Association, 1997, Technical !'!formation Letter No. 529, Silver Spring,
Md., 8pp.
11) Netherlands Cementidustrie,. 1995, "Does the Future of Concrete Have Anything to do with
Ethics?" Cement, V. 47, No.4, Apr.
12) Stein, J. S.,. 1993, Construction Glossary, John Wiley and Sons, Inc., 1137 pp., New York, N.Y.
13) Walker, S., 1944, "Application of Theory of Probability in Designing for Strength Specifications,"
Technical !'!formation Letter No. 14, National Ready Mixed Concrete Association, Washington, D.C,.
(Reprinted in Rock Products, Mar., 1944; in Concrete, May 1944).
14) Walker, S., 1944a, "Control of Quality of Ready-Mixed Concrete," National Ready Mixed Concrete
Association, Washington, D.C., 71 pp., (2nd Ed., 1945, 75 pp.; 3rd Ed., 1953, 76 pp.).

19
Daneshlink.com
daneshlink.com
--``,```,`,`,`,,,,`,`````,`,,`,,-`-`,,`,,`,`,,`---

Daneshlink.com
daneshlink.com
PART II- CEMENTITIOUS MATERIALS
Cementitious materials are an integral component of concrete. The properties of cementitious

materials influence the service life of concrete and, therefore, the service life of concrete structures.
Bryant had a keen and passionate interest in extending the service life of concrete structures. One
of the areas of his interest was alkali-aggregate reaction. He authored numerous publications dealing
with various aspects on use of pozzolans and slag related to alkali-silica reaction.
In 1959, he published a classic paper Effectiveness ofMineral Admixtures in Preventing Excessive Expansion
of Concrete Due to Alkali-Aggregate Reaction that has been selected to be reproduced in this publication.
In this paper he described the effectiveness of pozzolans or slag in preventing excessive expansion
of concrete due to alkali-silica reaction. He stated that a reduction in expansion will occur if
sufficient amounts of pozzolans are used in the concrete mixture.
The main conclusions drawn in the above-mentioned paper were:

1. The synthetic silica glass was found to be the most active admixture, only 10%
replacement by volume is needed to reduce the percentage expansion by 75% in two
weeks; next the calcined shales, requiring 19 to 29% replacement; the uncalcined
diatomite, requiring 22%; the volcanic glasses, requiring 32 to 36%; and lasdy, the slags
and fly ashes, requiring 39 to 45%.
2. It is evident from this investigation that no current chemical test could be used, with
reliance, to evaluate the effectiveness of an admixture in preventing excessive expansion.
Further work may develop a reliable chemical test, because a relation was found between
minimum percentage replacement value, fineness of the material, reactive silica, and the
amount of alkalies retained by the reaction product.
3. The mortar-bar test requires improvement to increase the precision for both intra- and
interlaboratory testing. Andreasen and Christensen* have noted that the flner fractions
(passing a No. 125 sieve) in the Pyrex aggregate may act as admixtures, and have stressed
the importance of removing these finer fractions by washing the aggregate. Other
differences in testing conditions and procedures between laboratories should be
resolved.
4. Specifications for evaluating admixtures for effectiveness in preventing excessive

expansion due to the alkali-aggregate reaction require the material to be tested at one
percentage replacement, generally 25% by volume.
* A. H. M. Andreasen and K. E. Hauland Christensen, "Investigation of the Effect of Some Pozzolans on Alkali
Reactions in Concrete," Progress Report L 1, Committee on Alkali Reactions in Concrete, Danish National Institute of
Building Research and Academy of Technical Sciences, Copenhagen (1957).
21 --``,```,`,`,`,,,,`,`````,`,,`,,-`-`,,`,,`,`,,`---

Daneshlink.com
daneshlink.com
--``,```,`,`,`,,,,`,`````,`,,`,,-`-`,,`,,`,`,,`---

Daneshlink.com
daneshlink.com
Reprinted, with permission, from Proceedings, ASTM V. 59, ©AS1M International, 100 Barr Harbor Drive, West
Conshohocken, P A 19428.
EFFECTIVENESS OF MINERAL ADMXTURES IN PREVENTING

EXCESSIVE EXPANSION OF CONCRETE DUE TO ALKALI-
AGGREGATE REACTION*+
by Leonard Pepper and Bryant Mather'
SYNOPSIS
Twenty materials, representing eight different classes of mineral admixtures, were evaluated, using
both chemical and mortar-bar test methods, for their effectiveness in preventing excessive
expansion of concrete due to alkali-aggregate reaction. It was found that the chemical tests cannot
be used with reliance to evaluate effectiveness. Each of the replacement materials evaluated will
prevent excessive expansion if a sufficient quantity is used. Correlations were found between
effectiveness and: fineness, dissolved silica, and percentage of alkali retained by reaction product.
Five of the materials tested (a fly ash, a tuff, a calcined shale, a calcined diatomite, and an uncalcined
diatomite) showed a reduction in alkalinity of 40 percent or more when tested by the quick chemical
test. All of these except the fly ash met the requirement proposed by Moran and Gilliland for the
relationship between reduction in alkalinity and silica solubility.
Six of the materials tested (two slags, a fly ash, a pumicite, and two calcined shales) reduced mortar-
bar expansion at least 75 percent with high-alkali cement and Pyrex glass aggregates when used as
50, 45, 35, and 30 percent replacements of the cement.
Calculations were made that suggest that the minimum quantity of each material required for
effective prevention of excessive expansion ranged from 10 percent for the synthetic silica glass to
45 percent for one of the slags. By groups, these calculated minimum percentages were: calcined
shales, 19 to 29; uncalcined diatomite, 22; volcanic glasses, 32 to 36; slags, 39 to 45; and fly ashes, 40
to 44.
The investigation of mineral ad-mixtures as cement-replacement materials was initiated by the Office
of the Chief of Engineers in 1950 as part of the Civil Works Investigations Program with the
purpose of ascertaining the degree to which portland cement may be advantageously replaced by
other materials, considering cost and the quality of the resulting concrete. This paper deals with that
part of the investigation that was concerned with the ability of these materials to prevent excessive
expansion of concrete due to alkali-aggregate reaction.
--``,```,`,`,`,,,,`,`````,`,,`,,-`-`,,`,,`,`,,`---
• Presented at the Sixty-second Annual Meeting of the Society, June 21-26, 1959.
+This paper is based on Technical Report No 6-481, U.S. Army Engineer Waterways Experiment Station, July, 1958.
1 Chemist, Chief of Chemistry Section, and Civil Engineer, Chief of Special Investigation Branch, respectively, U. S.
Army Engineer Waterways Experiment Station, Concrete Division, Jackson, Miss.

23
Daneshlink.com
daneshlink.com
Many tests to determine pozzolanic activity of mineral admixtures have been described (2, 6, 7).2 In
some, only the mineral admixture is involved; in others, the admixture is mixed with either lime or
portland cement. Several tests to evaluate the effectiveness of mineral admixtures in preventing
excessive expansion due to the alkali-aggregate reaction have been proposed. These tests, as well as
those for determination of alkali release and fineness, were selected for use in this investigation. The
tests used were evaluated for reliability, and the minimum percentage of the replacement material
that would effectively reduce the expansion was determined.
MATERIALS
Portland Cements
Three portland cements were used in this investigation: two type I cements, one of high alkali
content (1.07 percent alkalies calculated as soda) and one oflow alkali content (0.37 percent); and
one type II cement of medium alkali content (0.68 percent). The results of tests of these cements are
given for in Table I.
Replacement Materials
The 20 replacement materials studied represent eight classes. The classes, the specific materials
tested, and the abbreviations for the materials used herein, are as follows:
Class of Material Material

1. Granulated blast-furnace slags Slag I
Slag II
2. Natural cements Natural I (Nat I)

Natural II (Nat II)
3. Fly Ashes Fly Ash I (FA I)

Fly Ash II (FA II)
Fly Ash III (FA III)
Fly Ash IV (FA IV)
4. Natural Volcanic Gases Purnicite (Prun F)

Purnicite (Pum L)
Tuff
Obsidian (Obs)
5. Calcined opaline shales Calcined Shale (C Sh M)

Calcined Shale (C Sh N)
Calcined diatomaceous earth (Cal D)
6. Uncalcined Diatomite Uncalcined diatomaceous earth(Unc D)
7. Uncalcined Quart Silica Flour

Air Float
Ad-mix
8. Glass Synthetic, pure silica glass (SS glass)
2
The boldface numbers in parentheses refer to the list of references appended to this paper.
24
--``,```,`,`,`,,,,`,`````,`,,`,,-`-`,,`,,`,`,,`---

Daneshlink.com
daneshlink.com
The 16 materials in the first six classes were also used in the general investigation of cement-
replacement materials. The methods of processing these materials and data on their use in concrete
are given in report I of that investigation, (11) Classes 7 and 8 consist of four materials, all of which
may be regarded as essentially pure silica. The three materials in class 7 are quartz and the one in
class 8 is an amorphous silica. The results of tests of all20 materials are given in Table II.
Hydrated Lime
Some of the tests required the admixtures to be reacted with hydrated lime. The lime selected
conformed to ASTM Specifications C 6 for Normal Finishing Hydrated Lime and, in addition, it
was required that it contain not less than 75 percent CaO nor more than 5 percent MgO based on
the non-volatile portion, and that not more than 5 percent be retained on the No. 325 sieve. The
properties of the lime are given in Table II.
Aggregates
Three aggregate conditions were used in making 1 by 1 by 11-in. bars for expansion tests: Pyrex
glass, Sioux quartzite, and a Sioux-Klufa opaline quartzite combination, all previously described in
Waterways Experiment Station reports (10, 12). Both the reactive aggregate (Pyrex glass) and the
innocuous aggregate (Sioux quartzite) were graded as required by CRD-C 123 (9). The combination
of90 percent Sioux quartzite and 10 percent Klufa opaline quartzite has been found (12) to be more
--``,```,`,`,`,,,,`,`````,`,,`,,-`-`,,`,,`,`,,`---
reactive than any other combination of these materials. This combination was graded as follows:
Percentage Retained
Sieve Sioux Klufa
3/8 in. to No.4 2.5 -
No. 4to No.8 7.5 -
No. 8 to No. 16 10.0 -
No. 16 to No. 30 32.0 3.0
No. 30 to No. 50 25.6 4.9
No. 50 to No. 100 9.0 2.0
Passing No. 100 3.5 -
Total 90.1 9.9
TESTS AND RESULTS
Test for Reactivity with Sodium Hydroxide (NaOH):

Test Method--To evaluate admixtures for their effectiveness in preventing excessive expansion of
concrete, an adaptation of the quick chemical test for reactivity of aggregate with sodium hydroxide
has been used. Since some materials will harden when 25 g of the admixture is reacted with 25 ml of
1N
NaOH, it was necessary to reduce the weight of admixture used to 12.5 g. In all other respects the
test is the same as described in the Corps of Engineers version of the quick chemical test (CRD-C
128) (9).
Tests ofIndividual Materials and Blends--Each of the 19 admixtures in classes 1 through 7 and the two
type I pordand cements were tested alone. The twentieth admixture, synthetic pure silica glass, was
not tested because its great absorptive capacity precluded testing without further modification of the
test method. Each material was tested in either duplicate or triplicate.
25
Daneshlink.com
daneshlink.com
The average results of the tests of each of the admixtures and the portland cements are given in
Table III. The two type I cements gave values for dissolved silica (SJ of zero and for reduction in
alkalinity (RJ of 187 and 137. The 19 admixtures gave results for Sc ranging from 0 to 883 and for Rc
ranging from 65 to 545. The utility of this test would be greatly enhanced if either or both of the
determined functions (Sc, R) were additive. However, the results obtained from a series of tests
definitely indicate that neither function is additive and, therefore, the performance of a given blend
cannot be predicted.
Discussion ofResults-Two criteria have been applied to the test to determine the effectiveness of
admixtures in reducing expansion due to alkali-aggregate reaction. The Davis Dam specification (8)
required a minimum reduction in alkalinity (R.) of 40 percent for a material to be regarded as
effective. Of the materials tested in this investigation, the data in Table III show that only five meet
this requirement: fly ash I, tuff, calcined shale M, calcined diatomite, and uncalcined diatomite, with
R. values of 46, 41, 52, 55, and 44 percent, respectively. Calcined shale N, with an R. value of 39
percent, may be regarded as borderline. The second criterion was proposed by Moran and Gilliland
(7) who plotted the Sc and Rc values for a number of materials and found that those that will reduce
mortar expansions 75 percent or more in two weeks at 20 percent replacement by weight (25
percent by volume) will fall above or to the right of a line having the equation Rc + 2/3 630. The =
test results in Table III have been graphically represented in Fig. 1. It is evident that calcined
diatomite, calcined shale M, tuff, uncalcined diatomite, and calcined shale N meet this requirement.
Although fly ash I met the Davis Dam specifications, it does not conform to this requirement. Table
IV summarizes these data as well as the data obtained by other methods used in this investigation to
test the effectiveness of mineral admixtures in preventing expansion due to alkali-aggregate reaction.
Modified Test for Reactivity with NaOH:

Modified Method- The Bureau of Reclamation used the quick chemical test for some time to assist
in the evaluation of admixtures (8). However, it did not prove to be completely satisfactory, and in
1952, a description of a modified version of the reactivity test (6) was published. The modified test
assumes an admixture-to-cement ratio of 1 to 3 by weight and a tricalcium silicate (C3S) content of
cement of 40 percent. Therefore, 25 ml of 0.5N NAOH (instead of 25 ml of IN NAOH) are reacted
with 4.0 g of admixture and 1.5 g of calcium hydroxide (Ca(OH)z). The reaction container, the
reaction time, and the temperature are the same as for the original method. The containers are,
however, agitated throughout the reaction period by an end-over-end rotation at 30 rpm. In addition
to the determination of Sc and RoH 4on the filtrate, the net reduction in concentration of sodium
(Na) and potassium (K) was determined and also the concentration of alumina (Al20 3).
Criteria- The paper by Mielenz et al (6) correlated graphically the reduction in alkalies (RJ and the
reduction in alkalinity (Ro~ of 63 admixtures representing a wide variety of materials, with the
reduction in expansion (RJ of mortar bars containing Pyrex aggregate, high-alkali cement, and the
tested admixture at ages of 14 days, 2 months, and 1 year. The correlations were found to be
satisfactory for all three ages, but best at 14 days; therefore, only the 14-day correlations were
shown. Admixtures that reduced expansion by 75 percent in the mortar-bar test were generally
regarded as effective. The value of 75 percent reduction was, therefore, used to determine the
criteria for the chemical tests. The values of 180 or greater for reduction in alkalies and 201 or
greater for reduction in alkalinity were apparently selected from the graph as being those that will
4
In this paper, the reduction in alkalinity as determined in the modified test is designated RoH rather than R,.
--``,```,`,`,`,,,,`,`````,`,,`,,-`-`,,`,,`,`,,`---

26
Daneshlink.com
daneshlink.com
most efficiently separate admixtures that will reduce mortar-bar expansion by 75 percent or more
from those that will reduce expansion by less than 75 percent.
Statistical Determination of Criteria- The published data did not include the line of best fit, which may
substantiate the criteria. As part of this investigation, the equations of these lines, the correlation
coefficients, and the standard errors of estimate (4) were calculated and are:
Value
Equation of Regression Standard Corresponding
Correlation
Determination Test Age Error of to 75 percent
Line,• R., Coefficient
Estimate Reduction in
Expansion
Net reduction of 14 days 0.162 RA + 41.0 0.84 12.5 209.7
alkalies 2 months 0.161 RA + 41.0 0.79 15.3 211.3
1 year 0.165 RA + 33.2 0.78 16.6 253.1
Reduction in 14 days 0.154 RoH + 38.2 0.84 12.4 238.6

alkalinity 2 months 0.256 RoH + 37.6 0.82 14.3 239.7
1 year 0.165 RoH + 29.4 0.82 15.3 276.5
.. ..
a R. = reduction 1n alkalies, millieqmvalents per liter, RoH = reduction 1n alkalinity, millieqmvalents per liter, and Re =
percent reduction in mortar expansion.
The difference between the equations for the three test ages and between the corresponding correlation
coefficients are not significant The magnitude of the correlation coefficients indicates that the regression lines
are not due to random chance. Information concerning the reproducibility of the test was included with the
published data. The standard deviation for reduction in alkalies was found to vary between 7.4 and 14.5. The
criterion specified for reduction in alkalies is two to three standard deviations less than the values found by use
of the regression line to correspond to a 75 percent reduction in expansion both at 14 days and 2 months.
The criterion for reduction in alkalies, and probably for reduction in alkalinity, is too low in that many
admixtures that will reduce expansion by only 70 percent in 14 days will be evaluated as effective along with
most if not all admixtures that will reduce expansion by 75 percent or more. The criteria apparendy take into
account the lack of precision of the chemical test without considering the lack of precision in the mortar-bar
test
Additional Modification of the Test- An additional series of tests was made in which the mixtures of
replacement material and calcium hydroxide that were reacted with NaOH were proportioned to
represent specific admixture-to-cement ratios by solid volume and specific calculated C 3S contents
rather than the assumed value of 40 percent. In this series, the Si02 and Al2 0 3 determinations were
omitted and the double end point titration was used.
Blends Tested- The reactivity of admixture-calcium hydroxide mixtures representing the following
blends was determined by the procedure just as described.
--``,```,`,`,`,,,,`,`````,`,,`,,-`-`,,`,,`,`,,`---

27
Daneshlink.com
daneshlink.com
Replacement, Solid Volumea, percent

Class Replacement Material
Type I High-Alkali Type I Low-Alkali
1 Slag I 30,50, 70 50
Slag II 50 -
--``,```,`,`,`,,,,`,`````,`,,`,,-`-`,,`,,`,`,,`---
2 Nat I 20,35,50 35
Nat II 35 -
3 FAI 30,45,60 45
FAil 45 -
FA III 45 -
FAIV 45 -
4 PumF 25,35,45 35
PumL 35 -
Tuff 35 -
Obs 35 -
5 CShM 20,30,40 30
CShN 30 -
CalD 30 -
6 UncD 8, 12, 16 12
a Values taken from reference (11).
Test Results- Test results obtained for these blends and for the lime substituting for the two
cements are shown in Table V. All the results shown were obtained from duplicate determinations.
The following blends, of all those tested, were found to meet both published criteria (6) when tested
with the calcium hydroxide ratio representing the high-alkali cement: 45 percent pumicite F, 35
percent tuff, 30 and 40 percent calcined shale M, 30 percent calcined shale N, and calcined
diatomite. None of the blends made with uncalcined diatomite were found to meet the
requirements; undoubtedly, the percentage of uncalcined diatomite used was too low. Some
anomalies were also observed. The 35 percent blend of pumicite L with high-alkali cement gave
results greater than 210 for reduction in alkalinity but the results for reduction in alkalies were less
than 180; this was also found for the 45 and 60 percent of fly ash I with high-alkali cement and the
45 percent blend with low-alkali cement. The 30 percent blend of fly ash I with high-alkali cement
met the 210 criteria only when the double end point titration was used. Only the 45 percent blend of
pumicite F with high-alkali cement would be affected if the criteria were changed to those derived
from the regression line. The other blends (35 percent tuff, 30 and 40 percent calcined shale M, 30
percent calcined shale N, 30 percent calcined diatomite, all with high-alkali cement) would still be
regarded as effective inhibitors. The following general observations were apparent from this test:
(a) The amount of replacement material used must exceed a given minimum before expansion can
be effectively reduced and, for the materials tested, the minimum replacement was never lower
than 30 percent.
(b) The values for reduction in alkalies (RJ and alkalinity CRorJ generally increased with increasing
replacement.
(c) Blends made with low-alkali cement will not necessarily give results similar to those obtained
with blends made with high-alkali cement.

28
Daneshlink.com
daneshlink.com
(d) Changing the numerical values of the criteria does not materially affect the interpretation of the
results.
Mortar-Bar Test
Mortars- The only direct method available for determining the degree of expansion caused by
alkali-aggregate reaction is the mortar-bar test. Similarly, the only direct method available for
determining the effectiveness of an admixture in preventing excessive expansion due to the reaction
is to compare the expansion of bars made with the admixture and a reactive aggregate with bars
made without the admixture (control bars), all other conditions being the same. For this phase of the
investigation, each of the 14 materials in classes 1 and 3 through 6 and, the one admixture in class 8
were used both at the assumed optimum and at one half the assumed optimum percentages by solid
volume (the class 8 material was, in addition, used at one fourth optimum) with the type I, high-
alkali portland cement. Three sets of three bars each, each set made on a different date, were made
for each blend with Pyrex glass as the aggregate. The bars were fabricated in accordance with CRD-
C 123 (9) except that:
(a) The water content of the mortar was adjusted for a flow of 110 +- 5 percent by ten %-in. drops
in 6 sec.
(b) The quantity of cementing material used per batch did not equal450 g; however, the volume of
cementing material used per batch did equal that of 450 g of portland cement.
--``,```,`,`,`,,,,`,`````,`,,`,,-`-`,,`,,`,`,,`---
Method CRD-C 123 (9), with the exceptions noted above, differs from ASTM Method C 2275 from
which mortar-bar methods used by other laboratories were derived in the following essential
respects:
(a) The grading of the Pyrex aggregates differs as follows:
Sieve Size Percentage Used

Passing Retained CRD-C 123 ASTMC227
3/8-in. No.4 2.5 0
No.4 No.8 7.5 20
No.8 No.16 10.0 20
No.16 No.30 35.0 20
No. 30 No. SO 30.5 20
No. 50 No. 100 11.0 20
No.100 Pan 3.5 0
(b) CRD-C 123 provides for stripping bars from molds at an age of 44 4+-4 hr while ASTM
Method C 227 provides for stripping at 24 +- 2 hr.
In addition three sets of bars were made with the flrst admixture in each of the above classes at the
assumed optimum (11) percentage only (class 8 material was also blended at one half optimum
percentage), blended with the type I high-alkali cement and with the Sioux-Klufa combination as the
aggregate. Finally, these same admixtures at the same percentage replacement were blended with the
type II medium-alkali cement, and bars were made with Pyrex glass as the aggregate. The following
control mixtures were used, each represented by three sets of three bars.
(c) Type I high-alkali cement, Pyrex glass.

(d) Type I high-alkali cement, Sioux-Klufa combination.

29
Daneshlink.com
daneshlink.com
(e) Type I high-alkali cement, Sioux quartzite.

(f) Type II medium-alkali cement, Pyrex glass.
(g) Type II medium-alkali cement, Sioux-Klufa combination.
Tests and Results- The bars were stored for a period of 1 yr. Their length change was determined at
14 days, 6 months, and 1 yr according to Method CRD-C 123. The percentage reduction in
expansion was calculated by the following equation:
R = E, -E, xlOO
e E,
where:
Re = reduction in mortar expansion, percent,
Et = average expansion of bars containing the admixture, percent, and
Er average expansion of the applicable control bars, percent.
The average results of all these tests are shown in Table VI both as percentage expansion and
percentage reduction in expansion at bar ages of 14 days, 6 months, and 1 year.
Discussion of Test Results- The expansion due to alkali-aggregate reaction was reduced for every
combination tested except in three cases: the bars made from mortars containing blends of high-
alkali cement, both 2.5 and 5.0 percent of the synthetic silica glass with Pyrex glass aggregate (1 /4
and% optimum percentages of the admixtures), and the blend of 12 percent uncalcined diatomite
with the Sioux-Klufa aggregate. The degree of reduction of the expansion was increased by
increasing the percentages of admixture in the mortar. Maximum reduction of expansion for bars
made with Pyrex glass aggregate was found for the blend of 30 percent calcined shale with high-
alkali cement. For mortars made with the Sioux-Klufa combination as the aggregate, the maximum
reduction in expansion was found for the blend containing the synthetic silica glass at its optimum
percentage. Three of the mortars made with the medium-alkali cement showed shrinkage at the age
of 1 yr: those containing calcined shale M, slag 1, and synthetic silica glass at their optimum
percentages.
The following materials may be regarded as effective at a test age of 14 days, using the criterion of a
reduction in expansion of at least 75 percent when tested in mortars made with high-alkali cement
and Pyrex glass: slags I and II, fly ash IV, pumicite L, calcined shale M, calcined shale N, and
synthetic silica glass at their optimum percentage replacements. At test ages of 6 months and 1 yr, all
the admixtures tested (at their optimum percentage replacement), except uncalcined diatomite,
--``,```,`,`,`,,,,`,`````,`,,`,,-`-`,,`,,`,`,,`---
caused reduction in expansion of 75 percent or more. When the Sioux-Klufa combination was used
as the aggregate, slag I, calcined shale M, and synthetic silica caused reduction in expansion of 75
percent or more at all three ages and pumicite F at a test age of 14 days. All the admixtures tested
with the medium-alkali cement (Pyrex glass-aggregate) caused a reduction in expansion of 75 percent
or greater at all three ages.
Petrographic Examination of Bars- After expansion measurements were completed, condition of the
bars was determined by examination of the exterior and interior surfaces, both visually and with a
stereoscopic microscope. Each bar was rated numerically on a condition scale depending on the
presence and amount of (a) external signs of reaction such as warping, surface gel, and surface
cracking and (b) internal signs of reaction, such as gel and reacted aggregate. The scale is as follows:

30
Daneshlink.com
daneshlink.com
0 =No signs of reaction, bar as cast

1 = Slight evidence of reaction
2 = Slight evidence of reaction, signs are more abundant
3 = Moderate reaction, some damage
4 = Intermediate between conditions 3 and 5
5 = Severe reaction, damage, and deterioration
The numerical ratings of the bars are given in Table VI. The observations are summarized below:
(a) The general observations included the following:
1. There was a definite over-all relationship between the observed condition and the expansion
of the bars examined. The actual relationship was that the observable signs of alkali-
aggregate reaction increased as the expansion increased.
2. Every bar examined showed some evidence of alkali-aggregate reaction.
3. Calcium sulfoaluminate was not found in any of the bars.
4. All bars containing slag had the typical blue-green color in the interior and across the bottom
of the bar as cast. The area covered by the color varied directly with the slag content. The
color faded on exposure to air, but was still quite apparent after two months exposure. In a
previous investigation of portland blast-furnace slag cements (13), the blue-green color was
found to fade more rapidly than it did in these bars.
(b) Specific relationships observed were:
1. All the bars made with Pyrex glass as the aggregate were rated as being of condition 1 or 2.
Of the 14 admixtures in classes 1 and 3 through 6, 10 had the same rating at both the
optimum and half optimum percentages replacement. The other four were rated as showing
more reaction with increasing expansion.
2. The bars made with the Sioux-Klufa combination, as the aggregate show, in general, have
higher expansion and higher ratings, varying from 2 to 5, than those made with Pyrex glass
aggregate.
3. Both the condition scale rating and the expansion indicate that uncalcined diatomite is not
effective when used as a 12 percent replacement of high-alkali cement with the Sioux-Klufa
combination aggregate.
4. All the bars made with the synthetic silica glass required such large amounts of water in
mixing that they were abnormally weak. This made it impossible to use bar strength, which
--``,```,`,`,`,,,,`,`````,`,,`,,-`-`,,`,,`,`,,`---
normally diminishes with increased reaction, in arriving at rating conditions.
Rate of Alkali Release

In addition to the other chemical tests described previously, the rate of alkali release of the
replacement materials in the presence of lime and water was determined by the method described in
Appendix B of the paper by Moran and Gilliland (7). The concentration of the alkalies in solution
was determined at 0, 2, 7, 28, and 90 days for each of the 16 materials in classes 1 to 6 and for the 4
materials in classes 7 and 8 at 28 days only. The alkali content of each material was also determined
in accordance with CRD-C 244 (9) as well as the water-soluble alkalies as directed in CRD-C 209 (9).
The results of all these analyses are shown in Table VII. The alkali content of the solution at various
test ages is reported in terms of total alkali, as Na20, as percentage of the replacement material, and
also as a percentage of the alkali content of the material. The rate of release curves are similar for
both sets of data and are shown in Fig. 2, based on the alkali content. The rate of alkali release of the
majority of the admixtures increased after the second day, becoming zero or decreasing after 28
days. The diatomites ceased releasing alkali after 7 days, and calcined shale N absorbed alkali at a

31
Daneshlink.com
daneshlink.com
uniform rate after 7 days. For slags I, II, and calcined shale M, the rate of release was constant up to
28 days and then became negative. Tuff stopped releasing alkali after the second day. The amount of
alkali released, as a percentage of alkali content, was greatest for the natural cements, then the fly
ashes, followed by the slags and volcanic glasses. The calcined earths released the least alkali. As a
percentage of the replacement material, the greatest amount of alkali was released by the volcanic
glasses and fly ash I. The least amount was released by the calcined earths and the slags.
Mean Particle Diameter

In addition to the determination of particle size by microscopic methods(fable 11), the mean
particle diameter for each of the replacement materials was calculated as described in ASTM
Specification C 402- 58 T, Section 8(b).* The calculated values are shown in Tables II and VIII and
correspond very well to the microscopic results.
Autoclave Expansion
Pastes containing the type II cement and the replacement materials in classes 1, and 2 through 8, in
the proportion of 4 to 1 were tested for autoclave expansion. All ten bars tested were sound; the
maximum expansion was 0.13 percent for pumicite L.
--``,```,`,`,`,,,,`,`````,`,,`,,-`-`,,`,,`,`,,`---
COMPARISON OF TEST RESULTS AND CORRELATIONS
Comparison of Chemical and Mortar-Bar Test Results

Most of the results of the various tests previously described are compared in Table IV. Only the
results concerning materials in classes 1 and 3 through 6 are shown. For those tests involving blends
of an admixture and a cement or an admixture in a mortar, only the high-alkali cement and the Pyrex
glass aggregate results are shown. In addition, the criteria were rigidly interpreted to determine
whether a material would be effective; that is, when a test result was found to be one unit less than
the criterion, the admixture was judged to be noneffective. As can be seen from the table, complete
agreement between the methods was obtained only for calcined shale M, which is effective as a
replacement of 25 percent by volume or greater. If it is assumed that the percentage replacement
used for uncalcined diatomite was too low, the methods agree and uncalcined diatomite will not be
effective as replacements of 16 percent or less. It was noted previously and is also evident in Table
IV that all the materials tested, except uncalcined diatomite, caused reduction in expansion of 75
percent or greater at test ages of 6 months and 1 yr when used at their "optimum" percentage
replacement. If the 6-month and 1-yr test results are not considered, agreement between the
methods was also obtained for fly ash II, fly ash III, obsidian (all noneffective at their optimum
percentage replacement or lower), pumicite F (effective at 45 percent replacement but not at the
optimum or lower), and calcined shale N (found effective at optimum percentage replacement but
disagreement was found between the two criteria for reactivity with NaOH at 25 percent
replacement). Slags I and II and fly ash IV were found effective at their optimum percentage
replacement by the mortar-bar test but ineffective by any of the chemical tests. Tuff and calcined
diatomite were found to be effective by the chemical tests at their optimum percentage replacement
but ineffective by the mortar-bar tests at 14-day test age. The lack of agreement in the results for fly
ash I and pumicite L is between the chemical methods.
• Tentative Specifications for Raw or Calcined Natural Pozzolans for Use as Admixtures in Portland Cement
Concrete (C 402-58T), 1958, Book of ASTM Standards, Part 4, p. 519.

32
Daneshlink.com
daneshlink.com
MINIMUM PERCENTAGES OF REPLACEMENT FOR EFFECTIVENESS
For the proper employment of a criterion of effectiveness based on the ability of an admixture to
reduce expansion by a given proportion, or to a given proportion, of that which takes place when no
admixture is used, it is necessary that the control condition be one in which substantial expansion
takes place. It further follows that the effectiveness of the admixture must be evaluated under
conditions in which it serves a function other than that of a mere diluent. It has been shown for a
given case studied by U. W. Stoll at the U.S. Naval Civil Engineering Laboratories that, for a
reactive aggregate used with a cement of 1.2 percent alkalies calculated as soda, substitution of a
cement having 0.50 percent alkalies would result in 75 percent less expansion and that replacement
of 60 percent of the cement of 1.2 percent alkali content with an inert diluent should be expected to
produce the same reduction in expansion.
Effective admixtures should exhibit some beneficial effect in addition to that of dilution. To
determine the extent of this additional beneficial effect, the applicable data in Table V and VI were
plotted in Fig. 3. The curves were interpolated (or exterpolated) to determine the percentage
replacement that corresponded to the appropriate criteria for effectiveness. The interpolated values
are shown in Table IX.
--``,```,`,`,`,,,,`,`````,`,,`,,-`-`,,`,,`,`,,`---
All the admixtures tested do have some additional beneficial effect other than dilution of cement.
Some effects are manifested at different levels of replacement ranging from a minimum of 10
percent for synthetic silica glass at both 14 days and 6 months to 52 percent for fly ash III at 14
days. The agreement between the results obtained for the two test methods is very good considering
that these are interpolated results.
The minimum percentage replacement values may also be used to compare admixtures with each
other. The lower the minimum replacement value, the more effectively the admixture will prevent
expansion due to the alkali-aggregate reaction. This comparison is made in Table VIII. The
admixtures are listed in ascending order of the average minimum replacement values. The average of
the 14-day and 6-month values was used. The changes in the order would not be significant if either
the 14-day or 6-month values were used. The synthetic silica glass heads the list as most effective,
followed by the calcined shales and uncalcined diatomite, then the natural glasses, and finally the
blast-furnace slags and fly ashes.
CORRELATION OF MINIMUM REPLACEMENT VALUES WITH FINENESS, ALKALI RELEASE, AND

DISSOLVED SILICA
The relation between the interpolated minimum replacement value for each admixture and the
fineness of the admixture (mean particle diameter), the rate of alkali release at 28 and 90 days, and
the amount of silica dissolved in the quick chemical test is indicated in Table VIII. The minimum
replacement value did not correlate with the 0, 2, and 7-day alkali-release data. Figure 4 gives the
curves of best fit, the equations of these curves, the correlation indices or coefficient, and the
standard errors of estimate. The correlations, with due considerations of the small sample, are not
due to random chance. Since evidence of correlation between the independent variables was not
found, a multiple linear regression curve for the data in Table VIII was calculated with the following
results:

33
Daneshlink.com
daneshlink.com
R = 27.3-0.0158 Sc, + 1.6ln D + 3.5lnA90
where: multiple correlation coefficient= 0.951, standard error of estimate= 2.50, and beta
coefficient for Sc = 0.542,A 90 = 0.366, D = 0.109.
The sample size is too small for a sound opinion to be formed regarding the multiple correlation,
although it appears to be a good correlation. The beta coefficients indicate that the Sc value of an
admixture has the greatest effect on the minimum percentage replacement value and the mean
--``,```,`,`,`,,,,`,`````,`,,`,,-`-`,,`,,`,`,,`---
particle diameter the least. Other correlations were attempted including silica content, alkali content,
alkali release as a percentage of the amount of admixture used, all versus the minimum replacement
value, but without success. The correlations and also the lack of correlations indicate to some extent
the factors that are necessary for effective reduction of expansion due to the alkali-aggregate
reaction. Further work is necessary to determine the relative effect of lime on the reactions. The
multiple correlation does not suggest the mechanism of the reaction but rather serves to evaluate the
relative effect of the factors, when considered together, on the reaction.
REACTIONS INVOLVED IN PREVENTING EXCESSIVE EXPANSION
Many hypotheses have been advanced to explain alkali-aggregate reaction and the effectiveness of
measures to prevent its causing excessive expansion in concrete; however, a great deal of work still
remains to be done to provide a basis for an adequate hypothesis. With the thought that some
benefits might be derived from simplification of what appears to be a complex reaction, a summary
of an hypothesis concerning the reactions involved was prepared, part of which has been published
(1). It proposes a working hypothesis concerning phenomena involved in pozzolanic reaction, alkali-
aggregate reaction, and prevention of excessive expansion of concrete due to alkali-aggregate
reaction. The hypothesis was based on the assumption that all three phenomena are fundamentally
the same: the reaction of calcium, alkalies, and hydroxide ions with soluble silica. The degree of
reaction was thought to depend on the silica content, the metastability of the silica, and the fineness
of the material. This investigation substantiates part of the working hypotheses. Since precise
knowledge of the actual reactions that will cause the phenomena is lacking, the hypothesis will
certainly aid in determining the type of chemical test needed to evaluate materials with respect to the
phenomena. A chemical test to evaluate effectiveness of admixtures will have to determine, in the
presence of lime, the reactive (glassy or noncrystalline) silica content of the material, and the alkalies
that will be removed from solution. The chemical tests used in this investigation do not fulfill these
requirements since:
(a) The quick chemical test does not evaluate the reductions in alkalies directly, nor does the
reaction take place in the presence of lime.
(b) The modified quick chemical test does not correlate the effects of dissolved silica and reduction
of alkalies with effectiveness in preventing excessive expansion. In addition, both tests lack the
precision normally desired in chemical tests. The precision of the quick chemical test has been
previously examined and reported (10).
CONCLUSIONS
All the materials tested in this study will effectively reduce expansion if sufficient amounts are
employed in the mixture. All the materials exhibit a beneficial effect with regard to expansion other
than by dilution. The synthetic silica glass was found to be the most active admixture, only 10

34
Daneshlink.com
daneshlink.com
percent replacement by volume is needed to reduce the percentage expansion by 75 percent in two
weeks; next the calcined shales, requiring 19 to 29 percent replacement; the uncalcined diatomite,
requiring 22 percent; the volcanic glasses, requiring 32 to 36 percent; an lasdy, the slags and fly
ashes, requiring 39 to 45 percent.
It is evident from this investigation that neither chemical test can be used, with reliance, to evaluate
the effectiveness of an admixture in preventing excessive expansion. Further work may develop a
reliable chemical test, since a relation was found between minimum percentage replacement value,
--``,```,`,`,`,,,,`,`````,`,,`,,-`-`,,`,,`,`,,`---
fineness of the material, reactive silica, and the amount of alkalies retained by the reaction product.
The mortar-bar test requires improvement to increase the precision for both intra- and
interlaboratory testing. Andreasen and Christensen (3) have noted that the finer fractions (passing a
No. 125 sieve) in the Pyrex aggregate may act as admixtures, and have stressed the importance of
removing these finer fractions by washing the aggregate. Other differences in testing conditions and
procedures between laboratories should be resolved.
Specifications for evaluating admixtures for effectiveness in preventing excessive expansion due to
the alkali-aggregate reaction require the material to be tested at one percentage replacement,
generally 25 percent by volume. As indicated in this study, materials may be more effectively
evaluated and compared if they are tested at two or three different percentage replacements. The
minimum percentage replacement for effectiveness may then be determined by interpolation.
Pending additional development work, the criteria and procedures set forth in Specifications CRD-C
262 and Methods 263- 577 appear to be the most reasonable means of establishing the effectiveness
of a material in preventing excessive expansion due to alkali-aggregate reaction.
ACKNOWLEDGMENTS
The work reported in this paper was conducted by the Concrete Division of the U. S. Army
Engineer Waterways Experiment Station under the direction of Thomas B. Kennedy and under the
supervision of Bryant Mather, J. M. Polatty, C. H. Willatts, R. V. Tye, R. L. Curry, Leonard Pepper,
and Katharine Mather. The report on which this paper is based was prepared by Alan D. Buck, B. J.
Houston, and Leonard Pepper. Valuable comments were provided by Raymond E. Davis, Director
Emeritus, Engineering Materials Laboratory, University of California, Berkeley, Calif., Roy W.
Carlson, consulting engineer, Berkeley, Calif., and R. L. Blaine, Chief, Concreting Materials Section,
National Bureau of Standards, Washington, D. C.
REFERENCES
1) American Concrete Institute Committee 212, Committee's Closure to Discussion of"Admixtures

for Concrete." Proceedings, Journal Am. Concrete Inst., Vol. 51, December 19 55, p. 148-6.
2) Symposium on Use ofPozzolanic Materials in Mortars and Concrete, Am. Soc. Testing Mats.
(1950). (Issued as separate publication ASTM STP No. 99.)
3) A. H. M. Andreasen and K. E. Hauland Christensen, "Investigation of the Effect of Some

Pozzolans on Alkali Reactions in Concrete," Progress Report L 1, Committee on Alkali Reactions in

35
Daneshlink.com
daneshlink.com
Concrete, Danish National Institute of Building Research and Academy of Technical Sciences,
Copenhagen (1957).
4) Mordecai Ezekiel, "Methods of Correlation Analysis," John Wiley and Sons, Inc., New York, N.
Y., 2nd Edition, June 1947.
5) P. E. Irick and J. B. Blackburn, "Application of Statistical Methods to Laboratory Concrete

Freeze-Thaw Test Data," Proceedings, Highway Research Board, Vol. 34, pp. 329-367 (1955).
6) Richard C. Mielenz, Kenneth T. Greene, Elton J. Benton, and Fred H. Geier, "Chemical Test for
Alkali Reactivity of Pozzolans," Proceedings, Am. Soc. Testing Mats., Vol. 52, p. 1128 (1952).
7) W. T. Moran, and J. L. Gilliland, "Summary of Methods for Determining Pozzolanic Activity,"

Symposium on Use ofPozzolanic Materials in Mortar and Concretes. Am. Soc. Testing Mats., p. 109
(1950). (Issued as a separate publication ASTM STP No. 99.)
8) U.S. Bureau of Reclamation, "Calcined Reactive Siliceous Material for Use in Concrete,"
Specifications No. 1904, Appendix A to reference 7, Davis Dam Project, Denver, Colo. (1947)
9) U. S. Army Engineer Waterways Experiment Station, CE, Handbook for Concrete and Cement
(with quarterly supplements), Vicksburg, Miss., August 1949.
10) Leonard Pepper, "Tests for Chemical Reactivity between Alkalies and Aggregates, Quick
Chemical Test," Technical Memorandum No. 6-368, Report No. 1, Vicksburg, Miss., August 1953.
11) Bryant Mather, "The Partial Replacement of Portland Cement in Concrete," Papers on Cement
and Concrete, Am. Soc. Testing Mats., p. 37 (1958). (Issued as separate publication ASTM STP No.
205.)
12) R. V.-Tye and B. Mather, "Tests for Chemical Reactivity between Alkalies and Aggregate,
Mortar-Bar Test," CE, Technical Memorandum No. 6-368, U.S. Army Engineer Waterways
Experiment Station, Report No. 2, Vicksburg, Miss., September 1956.
13) B. Mather, "Investigation of Portland Blast-Furnace Slag Cements," Technical Report No.6-
445, U.S. Army Engineers Waterways Experiment Station, Vicksburg, Miss., Dec. 1956;Journal of
the American Concrete Institute (Proceedings), Vol. 54, pp. 205-232 (1957).
--``,```,`,`,`,,,,`,`````,`,,`,,-`-`,,`,,`,`,,`---
DISCUSSION
MR T. C. POWERS1- I have not read this paper, hence this question. Do these correlations,
pertaining to soluble silica, pertain to all the materials you described at first, or only to the selected
four that you described later?
MR BRYANT MAlliER (author)- It pertained to the 14 of the 20 materials that belong to
classes 1 and 3 through 6, as indicated by the values in Table VIII.
1
Research Counselor, Portland Cement Assn. Skokie, Ill.

36
Daneshlink.com
daneshlink.com
MR A. D. CONROW- What is the author's opinion of Pyrex glass as a material to represent a

reactive aggregate in and the reliability of results obtained with its use as a means for judging
performance if common aggregates are used?
MR MATHER- I believe that Pyrex glass provides the most standardizable reactive aggregate to
use as a representative of extreme reactivity. I do not envision the case where a mineral admixture
would be effective in reducing expansion of a high-alkali cement- Pyrex glass mortar and fail to be
effective in the field where the potential expansion was due to alkali-aggregate reaction. There are
expansions in the field, as is well known, which have not been proved to be due to alkali-aggregate
reaction. Pyrex glass with high-alkali cement would not provide a controlled situation for testing for
some other kind of expansion and the ability of something to prevent or reduce it.
MR RAYMOND E. DAVIS3 - I am wondering if the results of the Pyrex glass test do not give an
exaggerated idea of the quantity of pozzolan required to reduce the alkali aggregate expansion as
compared with the results that we get with most of the reactive aggregates.
MR MATHER- I am sure that the results do tend to give such an exaggerated idea. It will be
recalled that ASTM Specifications C 402 - 58 T provide alternative requirements for the ability of a
mineral admixture to deal with alkali reaction expansion. The test with Pyrex glass and high-alkali
cement and the requirement that a 75 percent reduction in expansion be obtained at 14 days when
25 percent by weight of the cement is replaced by the mineral admixture is the less desirable
alternative. The preferable alternative is the test with Pyrex glass and job cement and the
requirement that the 14-day expansion not exceed 0.020 percent when the amount of admixture to
be used on the job is used in the mortar. This latter approach not only allows the severity of the test
to be reduced proportionately to the degree to which the job cement has less than a maximum alkali
content but also permits the cement: admixture ratio to be that which will actually be employed.
Whether we can and should also employ a less reactive aggregate than Pyrex glass is debatable. If we
could be sure that the accelerated test gave in 14 days an expansion proportional to the ultimate
expansion of the test combination in service and could also be sure that the test aggregate
represented the most reactive condition to be encountered, then such a test aggregate -less
reactive than Pyrex -would be desirable. Obviously an admixture should not be rejected for use
because tests with a cement of 1.0 percent alkalies as Na20, Pyrex glass, and a cement:admixture
ratio of 4: 1 showed less than 75 percent reduction in expansion when it is certain that the cement
will never contain more than say 0.7 percent alkalies as Na20, the aggregates are only mildly reactive,
and the cement: admixture ratio will be 3: 1 rather than 4: 1.
2
Research Engineer, Ash Grove Lime and Portland Cement Co., Chanute, Kans.
3
Director Emeritus, Engineering Materials Laboratory, University ofCalifomia, Berkeley, Calif.
--``,```,`,`,`,,,,`,`````,`,,`,,-`-`,,`,,`,`,,`---

37
Daneshlink.com
daneshlink.com
Table I - Results of Tests of Portland Cements

Type I Cement Type II Cement
Test result
High-alkali Low-alkali Medium-alkali
Chemical Data
Major components:
SiOz, percent 20.6 20.1 23.1
Alz03, percent 5.4 6.6 5.0
Fez03, percent 3.9 3.1 3.9
CaO, percent 61.0 65.5 61.1
MgO, percent 3.9 1.2 2.8
S03, percent 2.1 2.0 2.0
Loss on ignition, percent 1.9 0.97 1.6
Minor components:
NazO, percent 0.62 0.11 0.23
KzO, percent 0.68 0.39 0.69
Total as NazO, percent 1.07 0.37 0.68
PzOs, percent 0.13 0.20 ...
Mnz03, percent 0.07 0.12 ...
Separate determinations:
Insoluble residue, percent 0.54 0.12 0.23
Chloroform solubility, percent 0.002 0.004 ...
Moisture content, percent 0.36 0.19 ...
Calculated compounds:
C3S, percent 44 56 29
CzS, percent 26 16 44
C~, percent 8 12 7
C~F, percent 12 9 12
CaS04, percent 4 3 3
Physical Data
Fineness, passing No. 325 sieve, percent 90.1 93.2 94.8
Fineness, Blaine, sq em per g 3765 3450 3590
Specific gravity 3.11 3.15 3.17
Time of setting (Gillmore):
Initial, hr:min 2:50 2:45 5:30
Final, hr:min 4:20 5:45 8:00
Autoclave expansion, percent 0.28 0.14 0.10
Normal consistency, percent 24.0 27.2 24.6
Air in mortar, percent 10.2 7.6 8.8
Comprehensive strength, psi:
3 days 1510 2600 1745
7 days 2065 3900 2665
28 days 3065 5565 5140
Heat of hydration, cal per g (70 F water-cement ratio = 0.40)
3 days
7 days 56 66 -
28 days 65 84 71
6 months 71 99 82
1 yr 86 106 -
Bleeding (water-cement ratio = 0.4): 86 111 89
Rate, ml per sq em per sec X 106
Capacity, ml per cc X 103 64 59 37
Bleeding (water-cement ratio = 0.45): 18 15 7
Rate 114 - 37
C:1pacity 29 - 7

38
--``,```,`,`,`,,,,`,`````,`,,`,,-`-`,,`,,`,`,,`---
Daneshlink.com
daneshlink.com
No reproduction or networking permitted without license from IHS
Table II - Results of test of replacement materials and hydrated lime

Granulated Natural volcanic glass
water-
Natural
quenched
cement
Fly ash
Pumicite
Calcined shales Uncalcined quartz ~ ~
iron blast-
furnace slag
~ ~4.1
Test "' e
!
....
~ .=
result :.=
~
~
~ z ~ ..
1.0 U')
~ '"0
IX>
"" IX>
~
19 19 r;- '\' C.l
~
I
~
0 ~
.es :.=
.......~
19 ~
~
"? r--
~
u Q 1;1 0 1.0
. c: ~
U')
'lil
~
....
~,...._
=:e ~
....
~
....
.... ~
.... ...~ e; € 8:
::::::
~ ~ u~ ~~
..c:!? Q
-;""
u .... ~
N~
~~ ~~
=ff{ S"?
~~ ~
--``,```,`,`,`,,,,`,`````,`,,`,,-`-`,,`,,`,`,,`---
Chemical data
Component
Si0 2,% 38.8 37.9 21.0 24.9 47.2 47.4 38.2 44.9 68.8 68.1 69.8 74.8 61.2 70.1 71.1 76.7 99.8 91.4 95.3 98.6 1.03
A}z03,% 11.3 13.3 5.2 5.7 19.5 34.0 25.7 34.0 14.8 15.8 14.9 13.8 12.5 19.3 14.9 12.0 0.16 4.8 2.6 0.0 0.93
Fe203, % 0.52 1.0 2.1 4.1 18.2 9.0 16.3 6.5 1.4 1.9 0.85 1.2 4.5 5.9 4.1 2.2 0.05 1.3 0.09 0.0 0.15
CaO, % 43.6 35.8 47.7 46.6 5.3 2.3 3.9 2.3 0.65 1.9 1.4 0.27 8.6 0.22 1.02 0.60 0.10 0.42 0.0 0.0 72.7
MgO,% 2.0 10.2 15.3 5.8 1.2 1.2 0.9 0.4 0.33 0.97 0.00 0.00 3.0 0.85 2.3 0.70 0.0 0.09 0.0 0.0 1.1
so3, % o.o8 o.o7 2.1 1.7 2.2 0.7 0.60 0.34 0.03 0.02 0.00 0.00 0.44 0.10 1.33 0.05 0.01 0.0 0.0 0.0 0.05
Loss on 1.35• 0.56• 6.2 11.2 0.8 3.9 12.2 7.9 3.9 1.99 6.6 0.69 6.8 1.26 2.5 3.2 0.13 1.8 0.74 1.2 24.3
Not for Resale, 01/26/2015 02:30:03 MST
rution, 0/o
Vl
\0 azO, 0.20 0.24 0.11 0.26 1.62 0.38 0.63 0.30 1.38 1.82 2.01 3.52 0.88 0.40 1.21 1.72 0.0 0.0 0.0 0.0 0.0
1me, 0/o
zO, 1.17 0.99 1.06 0.78 1.98 1.75 1.02 1.72 4.96 2.94 3.82 3.78 1.14 0.61 1.56 1.68 0.0 0.20 0.02 0.0 0.07
1me, 0/o
otal as 0.97 0.89 0.81 0.77 2.92 1.53 1.30 1.43 4.64 3.75 4.52 6.01 1.63 0.80 2.24 2.83 0.0 0.13 O.ot 0.00 0.05
azO,
nne, o/o
!Os, 0/o 0.02 0.02 0.01 0.15 0.23 0.14 0.26 0.51 0.02 0.07 0.11 0.04 0.29 0.14 0.27 0.29
·nz03,% 0.58 0.53 0.36 0.21 0.07 0.04 0.16 0.04 0.05 0.10 0.05 0.07 0.08 0.05 0.06 0.03
tsoluble 0.56 0.52 6.5 11.7 70.4 79.4 64.9 76.8 94.6 93.5 61.2 94.9 67.1 86.9 80.3 76.2
sidue,%
b.loro- 0.002 0.008 0.002 0.039
•rm
,Jubility,
Moisture 0.18 0.13 0.35 0.40 0.20 0.19 0.17 0.26 0.83 0.63 4.3 0.07 1.7 1.2 1.4 4.1 0.07 1.01 0.06 2.90 0.83
content, 0/o
Daneshlink.com
daneshlink.com
Table II - Results of test of replacement materials and hydrated lime, continued

water-
Natural ~ ~
quenched
iron blast-
cement
Fly ash Calcined shales Uncalcined quartz
~ ~
Test fiJ ~
Pumicite fiJ
e
~
result furnace slag OS
'"5'J> :..=
~
~ OS '"0
.~
~
Sulfide
I
~00
... ~
0.86
u~\Q,-.. u
=~e.
1.34
~;::!;
""N
0.06
u
~Lt)
........
""N
0.02
u
<
.... ...,
0.05
r--
=~ =~
0.00
00
0.13
I ";'
e:~
0.04
'f
(Z''"O
...... <
0.00
~~
~
0.00
==
E-<
0.00
,1:1
0
fiJ
0.00
.d
fiJ
'-?.d
OCil
u ::S <u Z < u <
0.00
~ 0~
0
0.00
-; 0
0.10
0
u
;§
0.09
Lt)
N~"?
~~~~~~~·a~
\Q
i-1"?
r--
><"?
:=
fiJ
£. l
sulfur,%
Total 0.45 0.53 0.43 3.17 11.1 7.22 0.02 0.04 0.08 0.04 1.10 0.07 0.30 0.15
carbon,% 3
Physical data
.j::..
ineness
0
as sing 95.8 98.8 87.9 96.5 95.5 92.8 93.6 85.1 98.6 90.8 86.2 66.9 98.1 70.5 98.8 68.5 100.0
io. 200
eve, dry,
as sing 90.2 97.2 91.0 93.2 93.5 87.7 85.8 81.5 98.1 91.4 85.5 67.2 96.8 71.5 98.6 93.0 96.9 96.0 99.9 96.7 98.0
0. 325
eve, wet,
pecific surface, cm2/g

Iaine air 360 3695 11,26 6420 3565 2940 2945 4205 4410 8340 1046 3415 1368 1177 10450 12125 4425 5500 4125 476000
~rme- 5 0 0 5 0
Ji!ity
tsher air 341 3390 9790 5830 3335 2855 2715 3960 4640 8230 1321 3850 1292 1502 12195 15345 4545 5130 4075
~rme 0 5 1 0
Jility
_Jein 279 4615 3835 1475 7030 5960 5050 5045 4505 6670 9850 4085 3360 7600 6485 12445 2595 2790 3745
hydro- 0
meter
Daneshlink.com
--``,```,`,`,`,,,,`,`````,`,,`,,-`-`,,`,,`,`,,`---
daneshlink.com

water-
Natural ~ ;;;;
quenched
iron blast-
cement
Fly ash Calcined shales Uncalcined quartz
~ ~
Test "'"' QJ
--``,```,`,`,`,,,,`,`````,`,,`,,-`-`,,`,,`,`,,`---
Pumicite
~
Ul
furnace slag
-&
~
result
~
~
o:l 't:l
<J
~
u
~~
.......
u
~~a--
....... p!:l
""N '-'
u
~;:!;
""N
u
~Ul
.......
""N
~.... ...,
r-;-
=~ =~
or
~~
c;-
~~
e< ~~
~ l::
::s
f-1
Physical data
..c"'
0
..= ":l..c:
~ ::E ~~ z ~ j ~
..., 1""1"'"
~ P-4 ~
Q
~
~ "-!
Ul
rt")
~ ~ ~~ ~ ~ ~ ·§ ~
Ia
....... rt")
r--
rt")
:-9
"'
£, !
Nitrogen 950 8100 4680 2620 7900 1600 2470 30,200 7990 6040 2593 8000 2423 1360 179,0 25950
adsorp- 0 0 0 0 00 0 0 00 00 00 0 0
tionb
Particle size (microscope)

Minimum, 0.5 0.5 -0.5 -0.5 0.5 0.5 0.5 -0.5 -0.5 0.5 -0.5 -0.5 -0.5 0.2 <0.4 0.2
j.l
~
......
[aximum, 270 400 350 250 1000 1000 2000 3000 250 500 450 1000 200 300 400 2000 100 4 60
redomina 2-9 2-15 0.5-3 0.5-3 1-10 3-13 4-20 3-14 2-20 2-14 2-8 6-15 2-6 1-7 2-10 0.5-1 0.4-0.8
t range, 1.1
lean 4.8 4.7 5.3 7.1 6.9 5.2 4.8 2.5 2.1 6.1 1.6 1.8 2.0 1.8 4.1 2.0 2.2 0.5
iameter,
Jecific 2.84 2.87 2.85 2.89 2.48 2.36 2.43 2.26 2.35 2.39 2.27 2.36 2.51 2.35 2.45 2.30 2.64 2.60 2.63 2.20
·avity
Normal 30.4 26.2 38.4 36.0 21.8 33.2 47.2 48.0 44.0 43.6 50.8 23.0 25.6 67.0 62.0 36.4 c c c d
consis-
Daneshlink.com
tency,%
daneshlink.com

water-
Natural
quenched
cement
Fly ash Calcined shales Uncalcined quartz ..,.
...., ...'?
iron blast- I
Test
result
furnace slag Pumicite ~ ~
...
ll'> "'"'OS 11.1
...... ... :§
..o...., "EJi
--``,```,`,`,`,,,,`,`````,`,,`,,-`-`,,`,,`,`,,`---
....
N
00 ..,.
... <Xl \0
I I
~ OS "tl
~
0\ <.l 11.1
~
I
~ ...., 1"- I I I ><

:-:
u~ u ~ ~ll'> z:::l os\0
~......
NI I
"tl Q
~
ll'> ·~ 1"- OS
Q~I ...........
~ <
I
u u 8'? "'
~ ~;;-
... lt: ,.C:I ..C:I
0 ::S «=I "tl
~ll'> ,-.. ,-.. ..c:"'
<.l
< ... ~~ ~
I
......... ~~ ~~ j~ ~~
>.
...... ,...N
~
......
......
,...N ...... ......
...... ......
...... e; ~
._, ...lo
....,
::s
f-< 0
c::
;:J ·~ 0
rJJl:: ~ rJl :::c:
Physical data
Air 2.8 7.5 4.9 9.6 4.9 2.6 2.7 2.8 2.7 1.7 2.0 4.1 3.2 2.5 3.6 1.6 c 5.2 2.5 d
content of
mortar,%
'leeding of paste
~ ate, 0.00 0.0001 0.0 0.000 e 0.000 f f f f f f 0.0 c c c d
f g f
tl/cm2 /g 012 17 018 119
3
apacity, 0.05 0.055 0.0 0.004 0.023 f f f f f g f f f 0.0 c d

tl/ml 5
a Values corrected for sulfur.

b Tests conducted by National Bureau of Standards.
' Insufficient quantity of material to complete all tests.
d Material is hygroscopic and gelled.
'Accurate bleeding results could not be obtained.
fUsing the specified proportions of water to solids, resultant pastes were too dry to test.
g Bleeding was too great to be measured with available apparatus.
Daneshlink.com
daneshlink.com
Table Ill- Results of Quick Chemical Test

Cements and Replacement Materials Alone
Dissolved Reduction in
Material Reduction in
Silica Sc, Alkalinity,
Alkalinity, Ra
milimoles per Rc, milimoles percent
liter per liter
Type I cement, low alkali 0 187 18

Type I cement, high alkali 0 137 14
Slag I 0 74 8
Slag II 0 68 7
Natural cement I 1 119 12
Natural cement II 0 113 11
Fly Ash I 44 456 46
Fly Ash II 158 280 28
Fly Ash III 33 352 36
Fly Ash IV 151 246 25
Pumicite F 270 167 17
Pumicite L 359 188 19
Tuff 450 406 41
Obsidian 204 65 7
Calcined Shale M 405 510 52
39
--``,```,`,`,`,,,,`,`````,`,,`,,-`-`,,`,,`,`,,`---
Calcined Shale N 883 390
Calcined diatomaceous earth 382 545 55
Uncalcined diatomaceous earth 810 430 44
SiOz flour 360 248 28
Air float 39 351 38
Ad-mix 82 85 11.6

43
Daneshlink.com
daneshlink.com
.... 600
I~I
I))
=.... "'-. _I
I))
0. aCalD
,..>--3>
"'
I))
500 J' o Cal Sh M
0
E
.E
oFAI
~~
c:0
~
oUncD
:;:::: 400 r-- 40 E!.!:cent redugion _ _ __niuff_ - - - - - - - --·-
:::; FA Ill in ATI<arrnfty CSHNo
01.1'1
~
00
z
~
Air Float
300
:c
~
0 FAIIO
z"' oFAIV oSi~ica Flour
0
~
:s
200
oPuml
"
"" "'
1-Lo
-;:a aPum F
.;<:
<c: NHi
c: 100 'N~ll1
0
:;:::: Slag I DAd-Mix
u Obs
:::; 'slag II
"2
a:
0 100 200 300 400 500 600 700 800 900
Concentration of Dissolved Silica in NaOH (1 N) Solution,mi!limoles per liter
Reactivi(y ofMaterials with Sodium Hydroxide.
--``,```,`,`,`,,,,`,`````,`,,`,,-`-`,,`,,`,`,,`---
Fig. 1 -

44
Daneshlink.com
daneshlink.com
Tabl e IV - summaryof test resu Itsoneffect1veness of adm1xtures

.
Admixture Volume Reactivity with Modified reactivity with Mortar-bar reduction in
replacement Na+ (OH)• NaOH expansion> 75 percent
material, N.+ OH- moles per
percent plus K+ liter
R.~40 R+ moles, Single Double 14 days 6 1year
percent 2/35 per liter end end months
~630 > 180 point> point>
210 210
70 - - n n n - - -
50 - - n n n e e e
Slag I
30 - - n n n - - -
25 n n - - - - n n
50 - - n n n e e e
Slag II
25 n n - - - n n n
60 - - n e e - - -
45 - - n e e n e e
Fly ash I
30 - - n n e - - -
22.5 e n - - - n n n
Fly ash II
45 - - - n n n e e
22.5 n n - - - n n n
Fly ash III
45 - - n n n n e e
22.5 n n - - - n n n
45 - - n n n e e e
Fly ash IV
22.5 n n - - - n n n
45 - - e e e - - -
35 - - n n n n e e
Pumicite F
25 n n n n n - - -
17.5 - - - - - n n n
PumiciteL
35 - - n e e e e e
17.5 n n - - - n n n
35 - - e e e n e n
Tuff
17.5 e e - - - n n n
35 - - n n n n e e
Obsidian
17.5 n n - - - n n n
40 -
--``,```,`,`,`,,,,`,`````,`,,`,,-`-`,,`,,`,`,,`---
- - e e e - -
Calcined 30 e e e e e e e e
shale M 20 - - n n n - - -
15 - - - - - n n n
Calcined 30 n e e e e e e e
shale N 15 - - - - - n n n
Calcined 30 e e e e e n e e
diatomaceous
15 - - - - - n n n
earth
16 e e n n n - - -
Uncalcined
12 - - n n n n n n
diatomaceous
earth
8 - - n n n - - -
6 - -
- - - n n n
aIn this test, the admixture 1s used by 1tself. The cntenon correlates the results With mortar-bar results m which 20
percent portland cement by weight (25 percent by volume) is replaced by the admixture.
Note: The letter n denotes that the admixture was found to be non-effective, the letter e that the admixture was found to
be effective.
45
Daneshlink.com
daneshlink.com
Tabl e V - Resu Its of mo d'f

11ed qUic
. k chem1ca Itest
Blends made with Ca(OH)2 equivalent of Type I high- Blends made with Ca(OH)2 equivalent of Type I low-
alkali cement, milliequivalents per liter alkali cement, milliequivalents per liter
Net Reduction in Reduction in
Replacement Net
Material reduction alkalinity (OH·) K+ alkalinity (OH·)
percent Reduction K+ Reduction reduction
in alkali, Single Double Single Double
in Na+ released inNa+ released in alkali,
(Na+ + end end end end
(Na+ + K+)
K+) point point point point
Cement 0 6 0 6 -2 -4 s 0 s s 1
Slag 30 34 10 24 16 32
so 25 6 19 35 59 -4 19 -23 49 73
70 10 13 -3 90 142
Slag II so 162 8 154 37 54 - - - - -
Natural 20 8 19 -11 -S -19
cement I 35 3 24 -21 45 32 -16 28 -44 41 32
so -19 -39 -58 70 73
Natural
35 -12 21 -32 35 31 - - - - -
cement II
Fly ash I 30 69 13 56 167 223
45 52 22 30 261 327 60 19 41 239 312
--``,```,`,`,`,,,,`,`````,`,,`,,-`-`,,`,,`,`,,`---
60 87 13 75 295 376
Fly ash II 45 134 18 116 170 200 - - - - -
Fly ash III 45 52 11 41 148 182 - - - - -
Fly ash IV 45 110 15 95 146 198 - - - - -
Pumicite F 25 60 45 15 104 121
35 152 48 104 154 18S 140 49 91 116 144
45 220 26 194 227 256
PumiciteL 35 199 24 175 23S 26S - - - - -
Tuff 35 224 16 208 2S7 293 - - - - -
Obsidian 35 47 33 14 50 73 - - - - -
Calcined 20 50 2 48 68 72
Shale M 30 216 0 216 238 243 151 6 145 175 172
40 325 -6 331 377 381
Calcined
30 356 0 356 386 390 - - - - -
shale N
Calcined
diatomac- 30 258 6 252 355 362 - - - - -
eous earth
Uncalcined 8 47 0 47 -2 -6
diatom- 12 43 4 39 44 37 8 23
-17 -25 36
a ceous
earth
16 95 -2 97 53 72

46
Daneshlink.com
daneshlink.com
Tabl e VI- Resu Itsof M0 rt ar· 8ar Expans1on Tests

Reduction in expansion, Warp, a
Replacement material Expansion, % 0 0
Petrographic
/o /o
rating of bar
Volume, condition
Type 14 days 6mos 1yr 14 days 6mos 1yr 1yr
%
Type I Hi£h-alkali cement
~ex ae;,-eeate
None - 0.106 0.221 0.230 - - - 0.05 2
Slag I 50 0.008 0.011 0.014 92 95 94 0.0 1
25 0.060 0.098 0.102 43 56 56 0.03 2
Slag II 50 0.021 0.025 0.028 80 89 88 0.02 2
25 0.084 0.117 0.123 21 47 47 0.04 2
Fly ash I 45 0.028 0.033 0.035 74 85 85 0.0 1
22.5 0.063 0.089 0.093 41 60 60 0.0 2
Fly ash II 45 0.031 0.032 0.037 71 86 84 0.0 1
22.5 0.069 0.090 0.095 35 59 59 0.0 1
Fly ash III 45 0.034 0.035 0.037 68 84 84 0.0 1
22.5 0.059 0.086 0.093 44 61 60 0.0 1
Fly ash IV 45 0.027 0.026 0.028 75 88 88 0.0 1
22.5 0.063 0.083 0.091 41 62 60 0.0 1
Pumicite F 35 0.035 0.040 0.045 67 82 80 0.0 2
17.5 0.059 0.082 0.087 44 63 62 0.0 2
Pumicite L 35 0.025 0.034 0.040 76 85 83 0.0 1
17.5 0.065 0.088 0.094 39 60 59 0.02 1
Tuff 35 0.028 0.052 0.060 74 76 74 0.0 1
17.5 0.072 0.155 0.170 32 30 26 0.01 2
--``,```,`,`,`,,,,`,`````,`,,`,,-`-`,,`,,`,`,,`---
Obsidian 35 0.036 0.046 0.053 66 79 77 0.0 1

17.5 0.081 0.127 0.136 24 43 41 0.03 2
Calcined 30 0.018 0.021 0.029 83 90 87 0.0 1
shale M 15 0.069 0.092 0.104 35 58 55 0.0 1
Calcined
30 0.004 0.002 0.009 96 99 96 0.0 1
shale N
15 0.044 0.057 0.067 58 74 71 0.02 1
Calcined 30 0.032 0.037 0.043 70 83 81 0.0 1
diatomaceous
15 0.071 0.104 0.110 33 53 52 0.0 1
earth
Uncalcined 12 0.077 0.123 0.134 27 44 42 0.03 1
diatomaceous
6 0.101 0.185 0.198 5 16 14 0.03 1
earth
Synthetic 10 0.024 0.053 0.086 77 76 63 0.03 1
silica glass 5 0.059 0.182 0.236 44 18 -3 0.04 1
2.5 0.101 0.260 0.293 5 -18 -27 0.05 2

47
Daneshlink.com
daneshlink.com
Table VI-continued
Reduction in expansion, Warp,a
Replacement material Expansion, % Petrographic
% %
rating of bar
Volume,
Type 0
14 days 6mos 1yr 14 days 6mos 1 yr 1yr condition
/o
Sioux-Klufa ~~e~ate
None - 0.159 0.681 0.803 - - - 0.06 5
Slag I 50 0.010 0.117 0.180 94 83 78 0.0 3
Fly ash I 22.5 0.065 0.381 0.464 59 44 42 0.04 3
Pumicite F 35 0.018 0.282 0.361 89 59 55 0.04 3
Calcined
30 0.018 0.051 0.102 89 93 87 0.02 3
shale M
Uncalcined
diatomaceous 12 0.119 0.792 0.964 25 -16 -20 0.05 5
earth
Synthetic 10 0.021 0.044 0.059 87 94 93 0.01 2
silica glass 5 0.053 0.281 0.317 67 59 61 0.03 3
Sioux ag~egate
None - 0.004 0.027 0.057 - - - 0.0 2
Type II medium-alkali cement
pYrex ag,.,egate
None - 0.046 0.068 0.075 - - 0.0 2
Slag I 50 -0.003 -0.009 -0.005 107 113 107 0.0 1
Fly ash I 45 0.008 0.005 0.008 83 93 89 0.0 1
Pumicite F 35 -0.002 -0.004 0.003 104 106 96 0.0 1
Calcined
30 0.000 -0.005 -0.002 100 107 103 0.0 1
shale M
Uncalcined
diatomaceous 12 0.001 0.002 0.004 98 97 95 0.0 1
earth
Synthetic 10
-0.019 -0.022
0.008 83 128 129 0.0 1
silica glass 0.004
5 0.016
0.006 87 94 79 0.0 1
Sioux-Klufa aggregate
None - 0.100 0.454 0.537 - - - 0.02 4
• Detemuned according to Method CRD-C123.
--``,```,`,`,`,,,,`,`````,`,,`,,-`-`,,`,,`,`,,`---

48
Daneshlink.com
daneshlink.com
60 I
IL Fl Ashesj
5c
6
.z."' 4c . . L t2<
~
..."' ~¥: ~/
....c: <.: .i;f L
u
~
c:
0
30
2•f
'~~
lr-~/,_9'~
,'~":!~ ..-
v
0
---- L
-LY.#
l!./
k"
~-~
2 7 28
·-· -
2 28 90
Age, days
_j_ _j_
I Natural Volcanic GlassesJ
.,....
Fig. 2 - Rate ofAlkali Release from Replacement Materials.
--``,```,`,`,`,,,,`,`````,`,,`,,-`-`,,`,,`,`,,`---

49
Daneshlink.com
daneshlink.com
Table VII-Rate of alkali release

Water-soluble alkali
Alkali content,a% of content, % of
Total alkali as Na20 released
Replace- replacement material replacement
ment material
material %of replacement material % of alkali content
Na20 ~0 totalb Na20 ~0 total days days
initial initial
2 7 28 90 2 7 25 90
Slag I 0.20 1.17 0.97 0.00 0.01 0.01 0.00 0.21 0.37 0.35 0 22 0.30 31 38 36
Slag II 0.24 0.99 0.89 0.00 0.00 0.00 0.00 0.17 0.22 0.19 0 19 0.21 24 25 21
Natural
0.11 1.06 0.81 0.01 0.31 0.22 0.23 0.28 0.36 0.49 0.54 28 35 44 60 67
cement I
Natural
cement 0.26 0.78 0.77 0.02 0.11 0.09 0.11 0.17 0.27 0.41 0.45 14 22 35 53 58
II
Fly ash I 1.62 1.98 2.92 0.23 0.04 0.26 0.24 0.28 0.52 1.31 1.20 8 10 18 45 41
Fly ash
0.38 1.75 1.53 0.03 0.02 0.04 0.04 0.08 0.18 0.56 0.68 3 5 12 37 48
II
Fly ash
0.63 1.02 1.30 0.02 0.01 0.03 0.03 0.07 0.16 0.48 0.62 2 5 12 37 48
III
Fly ash
0.30 1.72 1.43 0.01 0.01 0.02 0.02 0.07 0.14 0.35 0.50 1 5 10 24 35
IV
Pumicite
1.38 4.96 4.64 0.01 0.02 0.02 0.01 0.14 0.46 0.74 0.91 0 3 10 16 20
F
Pumicite
1.82 2.94 3.75 0.02 0.01 0.03 0.05 0.18 0.52 0.71 0.56 1 5 14 19 15
L
Tuff 2.01 3.82 4.52 0.02 0.03 0.04 0.75 1.09 1.08 1.13 1.11 17 24 24 25 24
Obsidia
3.52 3.78 6.01 0.01 0.00 0.01 0.01 0.19 0.51 0.97 1.11 0 3 8 16 18
n
Calcined
0.88 1.14 1.63 0.02 0.02 0.03 0.04 0.07 0.09 0.13 0.07 2 4 6 8 4
shale M
Calcined
0.40 0.61 0.80 0.00 0.02 0.01 0.00 0.03 0.12 0.08 0.05 0 4 15 10 6
shale N
Calcined
diatom a
1.21 1.56 2.24 0.03 0.01 0.04 0.03 0.18 0.34 0.37 0.33 1 8 15 16 15
ceous
earth
Uncalcin
ed
diatom a 1.72 1.68 2.83 0.04 0.00 0.04 0.02 0.02 0.12 0.10 0.10 1 1 4 4 4
ceo us
earth
Si02
flour
0.00 0.00 0.00 - - - - - - 0.00 - - - - - -
Air float 0.00 0.20 0.13 - - - - - - 0.00 - - - - - -
Ad-mix 0.00 0.02 0.01 - - - - - - 0.00 - - - - - -
Syntheti
c silica 0.00 0.00 0.00 - - - - - - 0.00 - - - - - -
glass
• Detertn111ed as descnbed 1n Method CRD-C 244.
hTotal alkali as NazO = NazO+ 0.658 K 20.
c Determined as described in Method CRD-C 209 (ASTM Methods C 114-53, 1958, Book of ASTM Standards, Part 4,
p. 63).
--``,```,`,`,`,,,,`,`````,`,,`,,-`-`,,`,,`,`,,`---

50
Daneshlink.com
daneshlink.com
Table VIII-comparison of effectiveness of admixtures and correlation of minimum replacement

vaIues w1t. h_QIh1s1ca
. I~o_Qerf1es of adm1x
. ture
Minimum% Total alkali as NazO,
Predominant Sc dissolved
replacement, Mean released, % of alkali content
Replacement range, silica,
solid volume, diameter
material microscope, millimoles
for effective calculated, f..l
f..l per liter
inhibition 28d~ 90d~
Synthetic silica 10 0.5
glass
Calcined shale 19 1.8 1 to 7 10 6 883
N
Uncalcined 22 1.8 0.5 to 1 4 4 810
diatomaceous
earth
Calcined shale 26 1.6 2 to 6 8 4 405
M
Calcined 29 2.0 2 to 10 16 15 382
diatomaceous
earth
PumiciteL 32 2.5 2 to 14 19 15 359
Pumicite F 35 4.8 2to 20 16 20 270
Tuff 36 2.1 2 to 8 25 24 450
Obsidian 36 6.1 6 to 15 16 18 204
Slag I 39 4.8 2 to 9 38 36 0
Fly ash IV 40 5.2 3 to 14 24 35 151
Fly ash I 41 5.3 1 to 10 45 41 44
Fly ash II 42 7.1 3 to 13 37 48 158
Fly ash III 44 6.9 4 to 20 37 48 33
Slag II 45 4.7 2 to 15 25 21 0

--``,```,`,`,`,,,,`,`````,`,,`,,-`-`,,`,,`,`,,`---
51
Daneshlink.com
daneshlink.com
100 lOO~--~----~----~----~
"',..
... 80
";!:
....,c
<!
0
60
h<wl>t~J<u;,.iffil
ror 7S Pemmi!Wd..Clion
! ml><P"mieyn
.:
c:
0
tl::J 40 40 41
In f:q>ansion l5
1:
a:
8,
0
$1~ i 41 J • Oh<idl.1f\ i~
• SI~U 48 t. c.t~Shdi<!M 2S
3
c: 20 V f ly Ash I 46 & (aloinc<l Sh>llo N 12
.,.:::: wfly Mh 11 -48 o C<1fti~ Didloo%Meoo> Edt1ft 32
£f A fly A~h II! )~ • llm:akii'K'd OiOl?.am=>LIS [4rth .2S
.. X 5ynft!e"tit 5llk4 ~u 10
00 "'F A~n IV 45 0 ~--~~-...;..-----"!-:::---~
20 40 80 0 20 40 60 80
Percentage Replacement of Portland Cement
100 ~--~~--~----~----~ 100r---~--y-~-----r----~
"'
1
--``,```,`,`,`,,,,`,`````,`,,`,,-`-`,,`,,`,`,,`---
80
"'
c:
",..
·~
60
5.£
c:
0
'B::l
40
~ o~l
•5~1}11
'II
~
c:
v Hytu.ni
.. fly Ash II
20
~ 4 Fly ~<shill
11. .. flyA.!hiV
Oil'vm<itl!'f
-~it<l~
0 o~--~~--~----~--~
0 20 ~ 60 80 0 20 40 60 80
Percentage Replacemen!JO,rrP_o_rt_la_n_d""'C_e_m_e_n_t...,...--....,...--....,
I
300
200
16(1
I .I
Pe«:ent
Replacement
for 180 l'er~ent
. '/ v Reduction in
./. Alkalies 36
I
i
100 . oPumM:ilc f 44- 43
v~! 1/ rt •Cail<:ir!fiiSNileM n YCaklned Sl1a.le M 21l

vtfnt.,lri~~d Dia:om~<«-u Al.lnc~k4n('d Ol<ltocruct!1)1,1~
Earth n (;,r;h Jl
20 40 60
Percentage Replacement of Portland Cement
Fig. 3 - Determination ofMinimum Percentage Replacement of Cement for Reduction in Expansion, Alkalies,
and Alkaliniry.

52
Daneshlink.com
daneshlink.com
Tabl e IX- M".

1mmum percentage rep Iacement b1y vo ume for effect1veness
Mortar-bar test Modified reactivity with NaOH
Replacement
reduction in reduction in
material 14 days 6 months average
alkalies alkalin!!Y
Slag I 41 37 39 - -
Slag II 48 42 45 - -
F1_y_ ash I 46 36 41 - 36
F1y ash II 48 36 42 - -
F1y ash III 52 36 44 - -
--``,```,`,`,`,,,,`,`````,`,,`,,-`-`,,`,,`,`,,`---
F1y ash IV 45 34 40 - -
Pumicite F 41 29 35 44 43
PumiciteL 35 28 32 - -
Tuff 36 35 36 - -
Obsidian 39 33 36 - -
Calcined shale M 28 23 26 27 28
Calcined shale N 22 16 19 - -
Calcined
diatomaceous 32 26 29 - -
earth
Uncalcined
diamotaceous 25 19 22 22 31
earth
Synthetic silica
10 10 10 - -
glass

53
Daneshlink.com
daneshlink.com
·.v.
~·
>.
v
I#" •
0 •
10
7 R =4.20-t 1034LNA21!o
S-_,=4.26
p =0.852
I :i!61llrp f/1.7Jj'J I
I I I
5 10 20
Tom! Alk~li\~> Na;P)Relea!led.pere~entof.>\l<m Caltent \uN~
5 5
~.f ~
• •
-
•/ .
4
l
' ·~ •
......
• ~
/.•
2
10 / A= 18.87 t- 12.5HN D
Sy "'4.11
p=O.~
Sy=lU
r=O.m
"'· '
~ =42.71- ().Q27 Sc
1 5o 10 20 0 200 400 000 BOO 1000

M"'m P.lilt~d~ [bm!!t~r. ~' Diswhlecl SiliC.I!I,rnillimd"" per liter fSJ
Fig. 4 - Correlation ofMinimum &placement Values with Alkali Release, Fineness, and Dissolved Silica.
--``,```,`,`,`,,,,`,`````,`,,`,,-`-`,,`,,`,`,,`---
54
Daneshlink.com
daneshlink.com
PART Ill- AGGREGATES
Aggregates make up the bulk of concrete and, therefore, influence many aspects of concrete
behavior. Lack of durability of aggregates results in detrimental consequences for the concrete.
Bryant's interest in aggregates included the physical characteristics, such as shape and surface
texture. He was also interested in other aspects such as evaluation of the coatings on the aggregates.
Variations in shape and surface texture of aggregates, or the presence of coatings on the aggregate
particles may significandy affect properties of concrete that are important to its behavior in service.
Bryant's thought was that the potential effects of these variations should be converted into useful
tools that could be used to make sound engineering judgments.
Bryant contributed a number of papers addressing the effect of aggregates on the overall behavior
and properties of concrete. His paper en tided "Concrete Aggregates: Shape, Suiface Texture, and Coatings,"
published in 1966, was selected to be reproduced in this part of the publication. In this paper,
Bryant explained that an aggregate's property is significant in concrete if it has an effect on the
behavior of concrete in service.
In the above-mentioned paper, Bryant pointed out that engineers who are required to evaluate the
concrete-making properties of aggregates or to discuss the factors responsible for variations in the
behavior of concrete are generally aware that variations in aggregate particle shape, surface texture,
and coatings may produce effects of significance to these considerations. He elaborated that
generally engineers are not provided with adequate information with which to convert their general
awareness of the potential significance of these variations into useful tools that they can employ in
making engineering judgments.
In conclusion Bryant stated that to provide such information, it will be necessary for scientists,
engineers, and technologists concerned with these questions to:
(1) develop adequate definitions and concepts of the properties,
(2) develop suitable techniques for determining the degree to which aggregates differ with respect to
--``,```,`,`,`,,,,`,`````,`,,`,,-`-`,,`,,`,`,,`---
them, and
(3) determine the effects of such differences on the behavior of concrete.

55
Daneshlink.com
daneshlink.com
--``,```,`,`,`,,,,`,`````,`,,`,,-`-`,,`,,`,`,,`---

Daneshlink.com
daneshlink.com
Reprinted, with permission, fromASIM SIP 169, ©ASTM International, 1966, 100 Barr Harbor Drive, West
Conshohocken, PA 19428.
CONCRETE AGGREGATES
SHAPE, SURFACE TEXTURE, AND COATINGS
by Bryant Mather;
INTRODUCTION
A property of aggregate is "significant" from the viewpoint of this publication only if it has a
significant effect on the behavior of portland-cement concrete in service. Similarly, a test of
aggregate is "significant" only if it yields useful information about a significant property or group of
properties. The behavior of concrete in service is determined by the interaction of certain of its
properties and the significant characteristics of the exposure to which it is subjected. Variations in
shape and surface texture of, or the presence of coatings on, aggregate particles may significandy
affect properties of concrete that are important to its behavior in service.
No ASTM methods exist by which quantitative determinations of particle shape, surface texture, or
coatings can be made. Few satisfactory data are available concerning the relations between variation
in these properties of aggregates and behavior of concrete. These properties have not been
adequately defined, and adequate definition is necessary before satisfactory information can be
obtained regarding the degree to which aggregates differ with respect to them. The nature and
degree of differences among aggregates with respect to these properties must be satisfactorily
determined before their effect on the behavior of concrete can be properly evaluated.
The absence of satisfactory data is not due to failure of previous writers to suggest the probable
significance of these properties. These properties are mentioned or discussed in the 1943 ASTM
"Report on Significance of Tests of Concrete and Concrete Aggregates" in the papers by Gilkey [1],
Hubbard [2], Kriege [3], and Lang [4]; and in the 1948 ASTM Symposium on Mineral Aggregates in
the papers by Allen [5], Rhoades and Mielenz [6], Rockwood [7], Sweet [8], and Woolf [9]. Reports
of work done in Australia, Denmark, France, Germany, Great Britain, Japan, Sweden, Switzerland,
and the United States are cited in the list of references at the end of this paper. Standardized test
--``,```,`,`,`,,,,`,`````,`,,`,,-`-`,,`,,`,`,,`---
methods have been developed by Markwick [10,11] and Shergold [12,13] of the Road Research
Laboratory in Great Britain; by Schiel [14-16], Walz [17], Schulz [18-20], Schulze [21], Pickel and
Rothfuchs [22], Pickel [23], and the Forschungsgesellschaft fiir das Strassenwesen [24], in Germany;
by von Matem [25] in Sweden; by Feret [26,27] in France; and by Goldbeck [28], Huang [29], and
the Corps of Engineers [30-32] in the United States.
The most careful studies of concepts and definitions of particle shape and surface texture have been
made by sedimentary petrologists, notably Wentworth (33-35], Wadell [36-38], Zingg [39],
Krumbein [40-42], Pettijohn [43], and Sneed and Folk [44]. However, the relation of their work to
aggregates for pordand-cement concrete has received litde attention, especially in the United States.
1
Personal Member, ASTM Research civil engineer, Concrete Div., U.S. Army Engineer Waterways Experiment
Station, Jackson, Miss.
Copyright American Concrete Institute 57

Daneshlink.com
daneshlink.com
Two relatively independent properties, sphericity and roundness, control particle shape. Sphericity is the
property that measures, depends upon, or varies with the ratio of the surface area of the particle to
its volume, the relative lengths of its principal axes or those of the circumscribing rectangular prism,
the relative settling velocity, and the ratio of the volume of the particle to that of the circumscribing
sphere. Roundness is the property the measure of which depends upon the relative sharpness or
angularity of the edges and comers of the particle.
S uiface texture is the property the measure of which depends upon the relative degree to which
particle surfaces are polished or dull, smooth or rough, and the type of roughness.
--``,```,`,`,`,,,,`,`````,`,,`,,-`-`,,`,,`,`,,`---
Coatings are layers of material, covering part or all of the surface of the particle, that have been
deposited on the particle after its formation. Coatings usually, but not necessarily, consist of
materials differing physically or chemically, or in both respects, from those composing the particle,
and are not derived from the particle itself.
Sphericity, roundness or angularity, smoothness or roughness, and coatings of a particle are the
result of the interaction of the nature, structure, and texture of the rock (or other material, for
example, slag) of which the particle consists and the forces to which it was subjected during and
after its formation. Rocks having closely spaced partings or cleavages in one or two directions tend
to yield flat or elongated particles oflow sphericity [6]. The design and operation of crushing
equipment influence sphericity of crushed particles; generally, the greater the reduction ratio, the
lower the sphericity [7,10,45-47]. Roundness is primarily a function of the strength and abrasion
resistance of the material and the amount of wear to which the particle has been subjected. Surface
texture depends on hardness, grain size, pore structure, and texture of the rock, and the degree to
which forces acting on the particle surfaces have smoothed or roughened it. Hard, dense, fine-
grained materials will generally have smooth fracture surfaces. Soft, loosely bonded coatings include
clay coatings on inadequately scrubbed gravel particles and dust coatings on crushed particles. Hard,
tightly bonded coatings are usually formed by precipitation of materials carried in solution in
groundwater on sand and gravel particles in natural deposits.
The ASTM Recommended Practice for Petrographic Examination ofAggregates for Concrete (C 295) notes that
such examinations are made to determine the physical and chemical properties of the material that
may be observed by petrographic methods and that bear on the quality of the material for its
intended use, and to describe and classify the constituents of the sample. It directs that pieces of
coarse aggregate be examined to establish whether coatings are present. If they are, it should be
determined whether the coatings consist of materials likely to be deleterious in concrete (opal,
gypsum, easily soluble salts, organic matter) and how fumly the coatings are bonded to the pieces.
The relevant features of each rock type found in the sample may include particle shape, particle
surface, grain size, texture, and structure, including observations of pore space, packing of grains,
and cementation of grains. This recommended practice points to the desirability of getting
information on these properties but does not provide procedures for obtaining such information
nor does it define the properties.

58
Daneshlink.com
daneshlink.com
MEASUREMENT OF PROPERTIES
Sphericity
Since 1942, the ASTM (C 125) Definitions ofTerms Relating to Concrete and Concrete Aggregates has
included the following statements:
"Flat piece - One in which the ratio of the width to thickness of its circumscribing rectangular
prism is greater than a specified value."
"Elongated piece - One in which the ratio of the length to width of its circumscribing rectangular
prism is greater than a specified value."
Classes proposed by Zingg based on ASTM designation C 125 using 1.5 as the "specified value."
Class Name b/a c/b Name L/W W/T
I Disks (oblate) >2/3 <2/3 flat <1.5 >1.5
Spherical neither flat
II >2/3 >2/3 <1.5 <1.5
(equiaxial) nor elongated
Blades both flat and
III <2/3 <2/3 >1.5 >1.5
(triaxial) elongated
N Rods (prolate) <2/3 >2/3 elongated >1.5 <1.5
Since these statements provide no "specified value," they are descriptions of procedures for defining
rather than defmitions. The property with which they deal is sphericity. Sphericity was defined by
Wadell [36-38] as the cube root of the ratio of the volume of the particle to the volume of the
circumscribing sphere. If d equals the nominal diameter of a particle, that is, the diameter of a sphere
of the same volume as the particle, and a equals the long dimension of the particle, that is, the
diameter of the circumscribing sphere, then
sphericity = 3
(1t! 6)d3 ~3 = -d
= 3-
3
(1t/6)a a3 a
3
Krumbein [41] pointed out that for most particles d is approximately equal to abc, the product of
the longest, intermediate and shortest axes of the particle. Therefore,
. .
sphenc1ty = 3
(1t I 6)abc = ~c
3 -
(7t/6)a3 a2
sphericity 3
clb
The ratios b/ a and c/ b were used by Zingg [39] in developing a sphericity classification that closely
approximates that stipulated in ASTM Designation C 125 which uses the ratios L/W (=a/ b) and
W/T (= b/ c). 2 If 1.5 is taken as the "specified value" in the statements in ASTM Designation C 125,
the resulting defmitions will yield the four following sphericity classes proposed by Zingg.
--``,```,`,`,`,,,,`,`````,`,,`,,-`-`,,`,,`,`,,`---
Figure 1 shows that each class developed by Zingg and stipulated in ASTM Designation C 125,
using 1.5 as the "specified value," includes particles of widely differing sphericity and that sphericity
alone fails to differentiate the kinds of departure of the particle from spherical. The dashed lines in
Fig. 1 drawn at the ratios of 1/3 and 3, defming the stippled area, indicate the classes set up by the
2
L, W, and T refer to length, width and thickness; L > W > T.

Daneshlink.com
daneshlink.com
Corps of Engineers test method [31] based on the use of 3 as the specified value. Walz (17]
proposed a plot indicating four classes based on L/T as abscissa and W/T as ordinate.
Sneed and Folk [44] found from a study of 1150 particles from the Lower Colorado River that the
frequency distribution for the ratio of L-W/L-Twas essentially normal, indicating that a majority of
the particles tended to have widths equal to about the mean of the length and thickness.
Plum [48] suggested expressing the degree of content of flat and elongated particles by the ratios a/ b
and c/ b, respectively, where a is minimum thickness, cis maximum length, and b is mean size.
Markwick [10] devised the procedure used in British Standard 812 [49] based on classifying as flat
(flaky) those particles having a thickness less than 0.6 times the mean sieve size, and as elongated
(long) those with a length greater than 1.8 times the mean sieve size. The weights of material passing
the respective gages expressed as a percentage of the sample are designated "flakiness index" and
"elongation index" [12]. Lees [50] reviewed the concepts and procedures of British Standard 812,
1960 [48] and concluded that the British Standard flakiness and elongation gages do not satisfactorily
perform the function for which they were intended. He favored the use of caliper measurements. He
recommended a return to Zingg's [39] ratio of 0.66 for both elongation and flatness-ratio boundaries
between shape categories. AASHO Designation M 80 [51] does not differentiate flat and elongated
particles since its reference is to those having "length greater than 5 times the average thickness."
Schiel [14] devised a formula involving specific gravity, settling velocity (a function of thickness),
and sieving (a function of width), the results of which are expressed as values from 100 to about 70.
Good particle shape= 100, cubic= >86.5, fairly cubical83.5 to 86.5, flaky= 80.5 to 83.5, and very
flaky = <80.5.
Feret [26,27], Pickel [23], Pickel and Rothfuchs [22], and Stern [52] have used procedures based on
the number of particles of a given sieve size that can be held in a container of a given volume. Plum
[48] stated that the number of particles of desirable shape, when compacted, needed to fill a
container of a given volume is smaller than the number needed if the particles are flat but greater
than the number needed if they are elongated.
Huang [29] proposed a procedure for determining particle index using a rhombohedron-shaped
mold into which aggregates are compacted first by 10 roddings and later by 50. Using the two unit
weight values, two values for percentage of voids are calculated, and using these and a nomograph, a
value for particle index is obtained. These values were compared with sphericity values calculated
according to Wadell [53] with results shown in Fig. 2. Huang states that particle index takes account
of shape (= sphericity), angularity, and surface texture. His work was related to soil-aggregate
mixtures. Rex and Peck [54] proposed a method for evaluating simultaneously the effects of "shape"
and "surface texture" by observing the rate at which sand of a single size fraction will run through a
3/8-in. orifice. The rate for a given sand divided by that for the same size fraction of standard
testing sand is the "time index."
The Corps of Engineers [31] and Schulze [21] have developed proportional calipers, or gages, to
permit rapid determination of dimensional ratios of particles. These instruments operate so that as
one of two jaws or slits is adjusted to a dimension of the particle, the other jaw or slit adjusts to a
given submultiple thereof. Schulze's device (Fig. 3) involves two slits having a 1:3 ratio. The Corps
of Engineers device (Fig. 4) has two jaws normally having a 1:3 ratio but capable of being set also at
1:2 or 1:5.
--``,```,`,`,`,,,,`,`````,`,,`,,-`-`,,`,,`,`,,`---

Daneshlink.com
daneshlink.com
Schulze proposed setting the gage for length and rejecting the particle if its width is less than one
third of its length. The Corps of Engineers procedure involves classification on criteria of both W/ T
and L/ W = 3. A particles is classed as flat if the ratio of width to thickness is greater than 3 and
elongated if the ratio of length to width is greater than 3.
Heywood [55] discussed a method developed by Hogan that is applicable to large particles. Surface
area is calculated from the weight of the wax coating produced by immersing the particle in molten
paraffin at a prescribed temperature. The sphericity of sand particles has also been studied by
determinations of surface area by Davies and Rees [56] using the air permeability method suggested
by Robertson and Emodi [57] based on the work of Carman [58] and Lea and Nurse [59].
Roundness
Roundness can be defined as the ratio of the average radius of curvature of the comers and edges of
the particle to the radius of the maximum inscribed circle [42]. Numerical values of roundess are less
frequendy employed than are descriptive terms based on the following categories [43]:
1) angular (litde evidence of wear),
2) subangular (evidence of some wear, faces untouched),
3) subrounded (considerable wear, faces reduced in area),
4) rounded (faces almost gone), and
5) well rounded (no original faces).
Roundess is independent of sphericity. Data are not available to indicated the relation, if any,
between empirical procedures designed to measure roundness (or angularity) of aggregate and the
property, "roundness," as defmed above. Shergold [13] suggested a procedure involving
determination by a standard method of the percentage of voids in compacted samples of individual
--``,```,`,`,`,,,,`,`````,`,,`,,-`-`,,`,,`,`,,`---
sieve fractions of coarse aggregate. He proposed the term "angularity number" = percentage voids
- 33. The angularity number ranges from 0 for well-rounded gravel to 11 for very angular material.
Goldbeck [28] described a similar method of determining voids in manufactured sands. Loudon [60]
determined an "angularity factor" based on specific surface. Powers [61] pointed out that Loudon's
angularity factor does not relate direcdy to angularity as defmed here, but is rather merely the
reciprocal of the sphericity factor. Mackey [62] reviewed various methods and recommended one
based on particle outline. He assumed a shape of 1.0 for "the perfect ellipsoid" and made deductions
for "irregularities."
Methods based on measurements of radii of curvature have been described by Wentworth [34,35]
and Waddell [36-38].
Surface texture
Jones [63] suggested that a device such as the replica surface analyzer described by Herschniann [64]
might be of some value in the study of aggregate surface texture and cited various reports on
surface texture, surface finish, and surface roughness [65-68]. Roughness of aggregate particle
surfaces should probably be expressed in terms of arithmetic average deviation of the actual surface
from the mean surface. Various types of equipment have been developed for use in the evaluation
of metal surfaces. The December, 1954, issue of Mechanical Engineering describes devices
developed by Brush Electronics Co. and General Electric Co. Other equipment is produced by the
Micrometrical Div. of the Bendix Corp. Such determinations of the effect of aggregate surface
textures as have been made have described the aggregates studied only in terms as "rough," "fairly
rough," "smooth," and "very smooth." Kriege [3] reported that in certain limestones the degree of

Daneshlink.com
daneshlink.com
roughness was found to range from 1.3 to 3.0 times the surface measured from plane polished
shapes, but he does not relate these differences to performance in concrete. Kramrisch [69]
described a procedure in which a linear profile was prepared and examined and presented results of
work on about 140 samples representing about 30 different types of rock.
In addition to the quantitative measurement of roughness, the types of roughness may be of

importance. Pettijohn [43] listed the following varieties of roughness: furrowed, grooved, scratched,
ridged, pitted, dented, striated, frosted, and etched. Blanks [70] differentiated the relative significance
of undulatory and abrupt rugosity. Knight and Knight [71] listed these types of surface texture:
glassy, smooth, granular, crystalline (fine, medium, coarse), pitted, honeycombed, and porous.
Coatings
The presence of a coating, its nature, thickness, continuity, and degree of adhesion or tenacity, can
best be established by petrographic examination. Blanks [70] listed the following types of coatings
--``,```,`,`,`,,,,`,`````,`,,`,,-`-`,,`,,`,`,,`---
noted by the U.S. Bureau of Reclamation: "silt, clay, gypsum, impure carbonates of lime and
magnesia, iron oxides, opal, manganiferous substances, soluble phosphates, or mixtures of these
substances." These are the characteristic natural coatings; artificial coatings include asphalt and
pordand-cement paste.
EFFECTS AND SIGNIFICANCE OF PROPERTIES
Sphericity
Equidimensional particles are generally preferred to flat or elongated particles for use as concrete
aggregates because they present less surface area per unit volume and generally produce tighter
packing when consolidated. Powers [61] wrote, "From some points of view, perfect sphericity is the
ideal particle shape, for spheres have the smallest specific surface area for a giver nominal size, and
an aggregation of spheres has the least percentage of voids for a given grading." For these reasons,
equidimensional particles of a given grading require a minimum of cement paste for a given degree
of workability of concrete. The extent to which the undesirability attributed to effects of flat and
elongated particles may have resulted from the use of gradings that work best with equidimensional
particles is not known. Smith and Kidd [72] described procedures used in connection with the
production and use of 2,000,000 tons of crushed stone(dolomite) aggregate in the U.S. projects on
the St. Lawrence Seaway. They stated that the percentages of each size group of coarse aggregate
were chosen to produce a combined grading agreeing as closely as practicable to an ideal grading
computed from a modification of Talbot's formula:
P=/%=(~r5
where P is the accumulative percentage passing a sieve of size d for an aggregate having a maximum
size D. The modification consisted of using 0.38 rather than 0.5 as the exponent. Smith and Kidd
stated that "an exponent of 0.3 would be used with angular flat and elongated crushed aggregate and
0.5 with spherically shaped natural gravels." They do not give data on the particle shape of the
aggregates they used nor on the variation in particle shape with difference in particle size. This,
however, does tend to conf111ll the supposition that if a grading suitable for relatively spherical
particles is employed with particles that are highly nonspherical, the results may be expected to be
less satisafctory than if a more appropriate grading had been employed.

Daneshlink.com
daneshlink.com
When the strength of the material of which the particles are composed is greater than that of the
matrix, and the bond between the particles and the matrix is well developed, it would be expected
that a concrete made with spherical aggregate particles would have less strength than one made with
nonspherical particles. Shergold's results [13] indicate that changing sphericity (from 20 to 40 per
cent flat particles) had little effect on the percentage of voids, whereas changing roundness had a
direct progressive effect.
Various investigations of the effects of flat and elongated particles have been reported [47, 73-7 6J.
Markwick [1 01 observed that "there is widespread unanimity of opinion" that "flaky and elongated
material is unsatisfactory" but "so far as can be ascertained, there are no references in the relevant
literature to actual failures attributed" to such material. Gilkey [1] stated that flat or elongated
particles may be positively objectionable either because of the uneconomic grading that they cause
or because of structural weakness introduced within the mass. Pramstig [77] described flat and
angular particles as "unfavorable" since "these require a higher percent of cement and are
detrimental to the workability of the concrete." He described "sliver segregators" that remove
unfavorably shaped particles.
Blanks [70] said, "Flat or elongated particles ... tend to decrease workability and thus to require more
sand, cement, and water. They tend to reduce bulk weight and decrease compressive strength. If flat
particles become oriented with their large faces in a horizontal position, bleed water will be
--``,```,`,`,`,,,,`,`````,`,,`,,-`-`,,`,,`,`,,`---
entrapped under them, a condition which prevents the development of good bond." Scholer [78]
regarded this latter phenomenon as likely to develop pores along which weathering may progress.
Mercer [79], on the other hand, found it practical to proportion concrete mixtures of adequate
workability and a water-cement ratio of 0.6 by weight with aggregates of which SO to 90 per cent
were flat or elongated particles or both. He, therefore, concludes that specification limits on particle
shape are unduly restrictive, and notes that "there is good reason for preferring 'poor shape'
whenever the controlling features are bond between aggregate particles and paste, and the modulus
of rupture."
In another study, Mercer [80] showed that compressive strength can be materially reduced when flat
particles of coarse aggregate are present in large numbers and oriented vertically. Tests of cylinders
containing hard-placed flat particles that were then grouted gave the following results:
Condition Compressive strength, 7 days, psi Unit weight, lb/ft3

Stone vertical 1330 159.0
Stone horizontal 2235 159.5
Mortar grout 2255 136.0
Normal arrangement 2260 160.0
Studies by Lane and Carlson [81] of the movement of gravel bed material incanals developed data
on the relation of sphericity of gravel particles to movement that may be applicable to the effects of
sphericity on workability of concrete. The effect of sphericity on susceptibility to movement by
flowing water as contact load was found to be:

Daneshlink.com
daneshlink.com
Relative weights of particles of equal susceptibility to

Sphericity range
movement
0.5 to 0.6 0.63
0.6 to 0.7 0.67
0.7 to 0.8 1.25
0.8 to 0.9 1.67
Elongated particles have also been reported [5] to have had objectionable effects in pavement
concrete by interfering with finishing operations; such effects were not found in the studies
discussed by Walker [7 6].
Gilkey [82] reported tests in which concrete was made with broken scraps of plate glass crushed to
pass a 1%-in. sieve. Concrete made with 1 /8-in. glass had lower square area were embedded in
compressive strength than that made with %-in. glass, which, in turn, had lower strength than that
made with normal aggregate. He concluded that flat pieces and are a source of weakness because
they are loaded as beams. In other fields their effects may be of more significance, for example, a
limit of 10 per cent was placed on flat and elongated pieces in specifications for trickling filter media
[83]. This was based on ratios of 1.5:1 for L/W and W/Tand, hence, corresponds to the
classification of Zingg [39] (see Fig. 1). Lees [84] reviewed the parameters of particle shape and the
methods of measurement. He concluded that for some mixtures containing asphalt as a binder a
flaky aggregate has superior performance.
Roundness
--``,```,`,`,`,,,,`,`````,`,,`,,-`-`,,`,,`,`,,`---
Almost all studies in which roundness was a variable have compared the performance of "rounded"
gravel with "angular" crushed stone without indicating the degree to which roundness varied or the
degree to which sphericity, surface texture, and other properties also varied. Well-rounded particles
might be expected to require less cement paste for equal workability than angular particles of equal
sphericity and similar surface texture. Shergold [13] showed that decreasing roundness or increasing
angularity directly affects the percentage of voids in aggregate; this in turn affects workability or
mixture proportions in concrete.
Blanks [45] reported experience on two similar projects where sand particles were in one case "well-
rounded" and in the other "harsh and angular." He states:
For concrete with 1%-in. max size and a water-cement ratio of 0.50, by weight, using the angular
material, 300 lb of water and 6.39 sacks of cement per cubic yard of concrete were required. Using
the rounded aggregate, the same factors were 220 lb of water and 4.69 sacks of cement per cubic
yard of concrete ... The latter concrete, in addition to being of higher quality by reason of its lower
water requirement saved 1. 7 sacks of cement per cubic yard, which would more than pay for an
elective abrading operation to improve the characteristics of the angular aggregate.
Malhotra [85] using the method proposed by Rex and Peck [54] studied nine natural sands and seven
crushed sands; the time indexes range from 1.05 for Leighton Buzzard sand to 1.57 for a fine
aggregate produced by crushing a coarse granite. He found a significant relation between flow of
mortar and time index; sand with a higher time index had a higher water requirement.
Shergold [13] provided a procedure by which "angularity number" may be determined and notes
that it was developed for use in connection with the British Standard for Concrete Aggregates (B.S.

Daneshlink.com
daneshlink.com
882:1944, Appendix 11) which distinguishes "rounded" from "angular" but gives no method of
defining or measuring roundness. Knight and Knight [71] noted that, in the case of sedimentary
rocks, angularity indicates high quality, whereas rounded grains usually indicate a relatively low
structural strength.
Surface Texture
As surface smoothness increases, contact area decreases; hence, a highly polished particle will have
less bonding area with the matrix than will a rough particle of the same volume. A smooth particle,
however, will require a thinner layer of paste to lubricate its movement with respect to other
aggregate particles. It will, therefore, permit a tighter packing for equal workability and, hence, will
require a lower paste content than a rough particle of similar roundness and sphericity. Goldbeck
[86] reported tests in which pieces of aggregate shaped to a 1-in.-square area were embedded in
cement briquets and tested for bond strength in tension. The different surface textures and bond
strengths in tension reported were:
Bond strength in tension, psi

Surface texture of 28 days in water, then 28
28 days in water, then 28
aggregate 28 days in water cycles of wetting and
days in air
_dryin_g_
Rough, porous 350 260 235
Rough 240 275 230
Rough 215 300 245
Fairl_y_ rough 250 185 160
Smooth 120 45 -
Smooth, nonchoidal 285 170 45
Very smooth 195 40 25
No specifications are known in which requirements are imposed for surface texture of aggregate
--``,```,`,`,`,,,,`,`````,`,,`,,-`-`,,`,,`,`,,`---
particles. Blanks [70] stated:
Salients and depressions on the particles, particularly when the side of these roughnesses are almost
perpendicular to the general surface, assist the adherence of the paste to the aggregate. Undulatory
roughness is less helpful and may even be harmful to bond as the mortar changes in volume.
Rugosity is probably less significant than physical penetration of cement into the aggregate.
The beneficial effects of rough texture of aggregate on flexural strength were shown by some
unpublished data developed by the Corps of Engineers for concrete made with a 5.5-bag-per-cubic-
yard cement factor and a 1V2-in.-max size as follows:
Percent of particles Water-cement ratio, Strength, 28 days, psi

Smooth Rough gal/bag flexural compressive
100 0 6.0 S95 4850
so so 6.4 645 4SOO
0 100 6.7 670 4125
Backstrom et al. [87] reported studies in which eleven fine aggregates were divided into two groups:
four having smooth surfaces and seven having rough surfaces based on examination using a
microscope. Each sand was prepared to a given grading and used in concrete of a given water-
cement ratio and comparable concentration of air-entraining admixture. The average values for

65
Daneshlink.com
daneshlink.com
surface area of voids were 742 in.- 1 for smooth and 1037 in.- 1 for rough fme aggregates, and the
average void spacing factors were 0.0065 and 0.0045 in., respectively.
Coatings
Coatings that are removed from aggregate particles during the mixing of the concrete have only the
effect of altering the grading by adding the coating material to the mixture. Coatings that remain on
the particles after the concrete has been placed affect the properties of the concrete exactly as if the
materials of which they are composed were present other than as coatings. An opal coating or a
gypsum coating may be undesirable because of its composition, but, if so, it is only incidental that it
is a coating. Specifications for concrete aggregates exist in which "adherent coatings," "asphalt
coatings," and "incrustations" or merely "coatings" are the basis for regarding the particles
possessing them as deleterious or objectionable.
Lang [4] noted that adherent clayey coatings may interfere with bond, that nonadherent dust
coatings will increase fmes and water demand, and that little is known of the effects of cemented
coatings. Goldbeck [88] reported that each 1 per cent of dust, probably as a consequence of its
effect on water demand, will cause a 1 to 1Vz per cent decrease in flexural strength and a 0 to 2 per
cent decrease in compressive strength of concrete.
SPECIFICATIONS
ASTM Specifications for Concrete Aggregates (C 33) are silent regarding the properties discussed in
this paper. The Federal Specifications [89] are similarly silent. The 19 53 Corps of Engineers
specifications for civil works [90,91] provided for an optional limit of 25 per cent on flat and
elongated particles in coarse aggregate, as determined by the Corps of Engineers method [31], and
instructed [92] that this requirement should be included in the specifications for a project when
preliminary work indicated the need for so doing. In 1963 [91,92], however, this requirement had
become mandatory for all work of sufficient importance to require use of the standard
specifications. None of these specifications contain any specific requirements regarding surface
texture and coatings. It, thus, appears that no cognizance is taken in specifications for general use in
the United States of any of the properties of aggregates with which this paper is concerned except by
the Corps of Engineers in its specification for concrete for civil works where flat and elongated
particles, as defined by it, are limited to 25 per cent. I know of no case in which this requirement has
caused rejection of a source of aggregates, and only one case in which the requirement caused a
producer to modify his crushing procedures.
The British Standard for Single-Sized Gravel Aggregates for Roads [93] requires that the flakiness
index not exceed 40 for 1 Vz-in. nominal size aggregate and 35 for 1, %, and Vz-in. nominal size
aggregates. The determination is made as described by Markwick [10]. The British Standard for
Concrete Aggregates from Natural Sources [94] requires that the aggregates be "free from adherent
coatings such as clay" but does not require compliance with any specific limits relating to shape or
surface texture. It also requires that the supplier shall provide, when requested by the purchaser, data
on shape and surface texture as determined according to clause 10 of British Standard 812 [49]
which are the methods described by Markwick [10].
--``,```,`,`,`,,,,`,`````,`,,`,,-`-`,,`,,`,`,,`---

66
Daneshlink.com
daneshlink.com
CONCLUDING STATEMENT
Engineers who are required to evaluate the concrete-making properties of aggregates or to discuss
the factors responsible for variations in the behavior of concrete are generally aware that variations
in aggregate particle shape, surface texture, and coatings may produce effects of significance to these
considerations. They have not been provided with adequate information with which to convert their
general awareness of the potential significance of these variations into useful tools which they can
--``,```,`,`,`,,,,`,`````,`,,`,,-`-`,,`,,`,`,,`---
employ in making engineering judgments. To provide such information, it will be necessary for
scientists, engineers, and technologists concerned with these questions to: (1) develop adequate
definitions and concepts of the properties, (2) develop suitable techniques for determining the
degree to which aggregates differ with respect to them, and (3) determine the effects of such
differences on the behavior of concrete. The results of such work may justify the use of aggregates
that are now regarded as unsuitable, and may cause the establishment of new requirements for
aggregates. In any case, the results of this such work will contribute knowledge to an area in which
there is now very little.
REFERENCES
1) H.J. Gilkey, "Size, Shape, Surface Texture, and Gradings of Aggregates," Report on Significance
of Tests of Concrete and Concrete Aggregates, ASTM STP 22A, American Society of Testing and
Materials, 2nd edition, 1943, p. 92.
2) Fred Hubbard, "Specific Gravity, Unite Weight, and Voids in Aggregates," Report on
Significance of Tests of Concrete and Concrete Aggregates, ASTM STP 22A, American Society of
Testing and Materials, 2nd edition, 1943, p. 108.
3) H. F. Kriege, "Mineral Composition of Aggregates," Report ofSignificance ofTests of Concrete and

Concrete Aggregates, ASTM STP 22A, 2nd edition, 1943, p. 152.
4) F. C. Lang, "Deleterious Substances," Report on Significance ofTests of Concrete and Concrete

Aggregates,ASTM STP 22A, American Society ofTesting and Materials, 2nd edition, 1943, p.
138.
5) C. W. Allen, "Influence of Mineral Aggregates on the Strength and Durability of Concrete,"
Mineral Aggregates, ASTM STP 83, American Society of Testing and Materials, 1948, p. 152.
6) Roger Rhoades and R. C. Mielenz, "Petrographic and Mineralogic Characteristics of

Aggregates," Mineral Aggregates, ASTM STP 83, American Society of Testing and Materials,
1948, p. 20.
7) N. C. Rockwood, "Production and Manufacture of Fine and Coarse Aggregates," Mineral

Aggregates,ASTM STP 83, American Society of Testing and Materials, 1948, p. 49.
9) D. 0. Woolf, "Needed Research," MineraiAggregates,ASTM SIP 83, American Society of

Testing and Materials, 1948, p. 221.
10) A. H. D. Markwick, "The Shape of Road Aggregate and Its Measurement," Road Research
Bulleting No. 2, Department of Scientific and Industrial Research, London, 1936.

67
Daneshlink.com
daneshlink.com
11) A. H. D. Markwick, "The Shape of Crushed Stone and Gravel and Its Measurement,"
Journal, Society Chemical Industry (Chemistry and Industry), Vol. 56, No.9, 1937, pp. 206-
213; RoadAbstracts, No.4, No. 114, 1937.
12) F. A. Shergold and J. R. Manning, "The Assessment of Shape in Roadmaking Aggregates,"

Roads and Road Construction, Vol. 31 (361 ), Jan., 19 53, pp. 4-7.
13) F. A. Shergold, "The Percentage Voids in Compacted Gravel as a Measure oflts

Angularity," Magai:fne of Concrete Research, No. 13, Aug., 1953, pp. 3-10.
14) F. Schiel, "Die Komform der Betonzuschlagstoffe und ihre Prufung" (The Shape of
Concrete Aggregate and Its Testing"), Betonstrasse, Vol. 16, Berlin, 1941, pp. 181-186; Road
Abstracts, Vol. X, No.3, 1943;Joumal, American Concrete Institute, Proceedings, Vol. 39, 1943,
p. 604.
15) F. Schiel, "Bestimmung der Komform durch Siebung (Determining the Shape of Aggregate
Gains by Sifting)," translated by Howard Houk, U.S> Group Control Council for Germany,
Field Information Agency, Technical Final Report No. 1112, 1947, (microfilm).
16) F. Schiel, "Bestimmung der Komform," Abhandlungen uber Bodenmechanik und Grundbau, Erich
Schmidt Verlag, Berlin, 1948, pp. 63-65.
17) K. Walz, "Determination of the Particle Shape of Aggregates," Betonstrasse, Vol. 11, No.2,
Berlin, 1936, pp. 27-32.
18) F. Schulz, "The Problem ofParticle Shape Determination," Strassen undTid!Jau, Vol. 6, No.
11, 1952, pp. 340-345; Road Abstracts, Vol. XX, No.8, Aug., 1953.
19) F. Schulz and G. Stelzer, "An Accurate Rapid Method for the Determination of Particle
Shape," Strasse undAutobahn, Vol. 5, No.2, 1954, pp. 48-53; Road Abstracts, No. 755, Oct.
1954.
20) Addendum to 19), Strasse unAutobahn, Vol5, No.6, 1954, p. 210.

--``,```,`,`,`,,,,`,`````,`,,`,,-`-`,,`,,`,`,,`---
21) K. Schulze, "Rapid Method for Particle Shape Determination and Suggestions for Definition
of Ordinary and High-Quality Chippings," Strasse undAutobahn, Vol. 4, No.8, 1953, pp. 253-
257; RoadAbstracts, Vol. XXI, No.4, Apr., 1954.
22) W. Pickel and G. Rothfuchs, "Bewertung der Komform von Edelsplitt," Bitumen, Vol. 8,
Berline, 1938, pp. 88-91, 116-118; Strassenbau, Vol29, 1938, pp. 273-277.
23) W. Pickel, "Die Bestimmung und Dewertung der Komform von Edelsplitt," Betonstrasse,
Berlin, May, 1937, p. 104.
24) Forschungsgesellschaft fur das Strassenwesen E. V., ''Vorlaufiges Merkblatt fUr die
Beurteilung der Komform mit der Komform-Schlieblehre," Cologne, 1960, (Provisional
Specification for the Determination of Aggregate Particle Shape with the Caliper-gage).

Daneshlink.com
daneshlink.com
25) N. Von Matern and A. Hjelmer, "Forsok med Pagrus (Tests with Chippings)," Bulletin 65,
Statens Vaginstitut, Stockholm, 1943, (English summary, pp. 56-60).
26) R. Feret, "Sur la Forme et l'etat de Surface des Elements Inertes des Betons," Annates de
l'Institut Technique, No. 2, Mar.-Apr., 1937, Paris, p. 52.
27) R. Feret, "Nouvelles Recherches sur la Forme des Elementes Inertes des Betons," Annals de
l'Institut Technique, No.3, May-June, 1938, Paris, p. 31.
28) A. T. Goldbeck, "Stone Sand for Use as Fine Aggregate," Concrete, Vol. 59, 1951, pp. 28-32.
29) Eugene Y. Huang, "A Test for Evaluating the Geometric Characteristics of Coarse
Aggregate Particles," Proceedings, American Society for Testing and Materials, Vol. 62, 1962,
pp. 1223-1242.
--``,```,`,`,`,,,,`,`````,`,,`,,-`-`,,`,,`,`,,`---
30) Handbook for Concrete and Cement, Corps of Engineers, Central Concrete Laboratory, Mt.
Vernon, N.Y., 1942, p. 95.
31) "Method of Test for Flat and Elongated Particles in Coarse Aggregate," Designation CRD-C
119-48, -52, -53, Handbook for Concrete and Cement, U. S. Army Corps of Engineers, Waterways
Experiment Station, Vicksburg, Miss., 1949, 1953, 1953.
32) ''Method of Test for Flat and Elongated Particles in Fine Aggregate," Designation CRC-C
120-55, Handbook for Concrete and Cement, U.S. Army Corps of Engineers, Waterways
Experiment Station, Vicksburg, Miss., 1955.
33) C. K. Wentworth, "A Laboratory and Field Study of Cobble Abrasion," Journal of Geology,
Vol. 27, 1919, p. 507.
34) C. K. Wentworth, "A Method of Measuring and Plotting the Shapes of Pebbles," Bulletin,
U.S. Geological Survey, No. 730, 1922, pp. 91-102; Professional Paper No. 131, U.S.
Geological Survey, 1923.
35) C. K. Wentworth, "The Shapes of Rock Particles: A Discussion," Journal rifGeo!ogy, Vol. 41,
No.3, Apr.-May, 1993, pp. 306-309.
36) Hakon Wadell, "Volume, Shape, and Roundness or Rock Particles," Journal rif Geology, Vol.
41, No.3, Apr.-May, 1993, pp. 306-309.
37) Hakon Wadell, "Sphericity and Roundness of Rock Particles," Journal of Geology, Vol. 41,
1933, pp. 310-331.
38) Hakon Wadell, "Volume, Shape, and Roundess of Quartz Particles," Journal rifGeology, Vol.
43, 1935, pp. 250-280.
39) Th. Zingg, "Beitrag zur Schotteranalyze," Schweizerische mineralogische undpetrographische

Mitteilungen, Vol. 15,1935, pp. 39-140.

69
Daneshlink.com
daneshlink.com
40) W. C. Krumbein and F. J. Pettijohn, Manual ofSedimentary Petrograpi?J, Appleton-Century Co.,

New York, N.Y., 1938, p. 145.
41) W. C. Krumbein and L. L. Sloss, Stratigrapf?y and Sedimentation, W. H. Freeman & Co., San
Francisco, Calif., 1951.
42) W. C. Krumbein and L.L. Sloss, Stratigrapf?y and Sedimentation, W. H. Freeman & Co., San
Francisco, Calif., 1951.
43) F. J. Pettijohn, "Sedimentary Rocks," Harper & Brothers, New York, N.Y., 1949.
44) Edmund D. Sneed and Robert L. Folk, "Pebbles in the Lower Colorado River, Texas. A
Study in Particle Morphogenesis," Journal of Geology, Vol. 66, No.2, Mar., 1958, pp. 114-150.
45) R.F. Blanks, "Good Concrete Depends on Good Aggregate," Civil Engineering, Vol. 122, No.
9, Sept. 1952, pp. 651-655.
46) W. H. Coghill, 0. W. Holmes, and A. B. Campbell, "Determination of Flakiness of Ores,"

Rtport of Investigation No. 2899, U.S. Bureau of Mines, 1928.
47) D. A. Stewart, The Design and Placing of High Qualiry Concrete, E.&F.N. Spon, Ltd., London,
1951.
48) Niels M. Plum, "Concrete Manual," Bulletin No. 39, Christiani and Nielsen, Copenhagen,
1944. (In English, see pp. 12, 179, 216.)
49) "Sampling and Testing of Mineral Aggregates, Sands and Fillers," British Standard 812, British
Standards Institute, 1951; "Method for Sampling and Testing of Mineral Aggregates, Sand,
and Fillers," British Standard 812, British Standard Institute, 1960.
50) G. Lees, "The Measurement of Particle Elongation and Flakiness: A Critical Discussion of
British Standard and Other Methods," Maga;dne ofConcrete Research, Vol. 16, No. 49, Dec.,
1964,pp.225-230.
51) "Standard Specifications for Coarse Aggregate for Pordand Cement Concrete (M 80-51),"
AASHO Specifications, American Association of State Highway Officials, Part I, 1955, p. 74.
52) Ottokar Stern, "Mittelbare Bewertung der Kornform von Zuschlagen," Zement, No. 43-43,
Berlin, Oct., 1937, pp. 693-696,706-711.
53) Hakon Wadell, "Shape Determinations of Large Sedimental Rock-Fragments," Pan-American

Geologist, Vol. 61, 1935, p. 187.
54) H. M. Rex and R. A. Peck, "A Laboratory Test to Evaluate the Shape and Surface Texture of
Fine Aggregate Particles," Public Roads, Vol. 29, No. 5, Dec., 1956, pp. 118-120.
55) H. Heywood, Proceedings, Institute Mechanical Engineers, Vol. 125, 1933, p. 383.
--``,```,`,`,`,,,,`,`````,`,,`,,-`-`,,`,,`,`,,`---

70
Daneshlink.com
daneshlink.com
56) W. Davies and W. J. Rees,Journal, Iron and Steel Industry, Vol. 40, No.2, 1944, pp. 19-47.
57) R. H. S. Robertson and B. Emocli, Nature, Vol. 152, p. 539.
58) P. C. Carman, Transactions, Institute Chemical Engineers, Vol. 15, 1937, p. 150;Journal,
Society Chemical Industry, Vol. 57, 1938, p. 225; Vol. 58, 1939, p. 1.
59) F.M. Lea and R.W. Nurse, journal, Society Chemical Industry, Vol. 58, 1939, p. 227.
60) A. G. Loudon, "Computation ofPermeability form Simple Soil Tests," Geotechnique, Vol. 3,
No.4, Dec., 1953, pp. 165-183.
61) T.C. Powers, "Topics in Concrete Technology: 1. Geometric Properties of Particles and
Aggregates," Journal of the PCA Research and Development Laboratories, Vol. 6, No. 1,Jan., 1964,
pp. 2-15.
62) R.D. Mackey, "The Measurement of Particle Shape," Civil Engineering and Public Works Review,
Vol. 60, No. 703, Feb., 1965, pp. 211-214.
63) F.E. Jones, "The Physical Structure of Cement Products and Its Effect on Durability,"
Proceedings, 3rd International Symposium on the Chemistry of Cement, London, 1952, 1954,
p. 368.
64) H.K. Herschmann, Technical News Bulletin, National Bureau Standards, Vol. 31, No. 3, Mar.
29, 1947.
65) The Instrument Manual, United Trade Press, London, 1949.
66) J. A. Broadston, "Surface Finish Measurement Instrumentation," Instruments, Apr., and June,
1949.
67) W. Mikelston, "Surface Roughness Measurements of Inspection," Co!iference ofInstrument

Society ofAmerica, Pitssburgh, Pa., 1946.
68) R.E. Reason et al., "Conference on Surface Finish," Proceedings, Institute Mechanical
Engineers, Vol. 15, No. 10, 1945.
69) Fritz Kramrisch, "Zur Rauigkeitsbestimmung von Gesteinsbruchfliichen," Geologie und

Baurwesen, Vol. 2, 1935, pp. 35-39.
70) R.F. Blanks, "Modern Concepts Applied to Concrete Aggregate," Transactions, American
Society of Civil Engineers, Vol. 115, 1950, p. 403.
71) B.H. Knight and R.G. Knight, Road Aggregates, Their Uses and Testing, Edward Arnold and Co.,
London, 2nd edition, 1948.
--``,```,`,`,`,,,,`,`````,`,,`,,-`-`,,`,,`,`,,`---

Daneshlink.com
daneshlink.com
72) M.R. Smith and G.M. Kidd, "Concrete Technology and Aggregate Production for St.
Lawrence Seaway," Journal, American Concrete Institute, Proceedings, Vol. 56, Nov., 1959, pp.
361-376.
73) M. Fujiii, Effects ofForms of Particles Upon the Technical Properties ofAggregates, Research Office of
Public Works, Toyko, 1935.
74) A. T. Goldbeck, "An Investigation of the Effect of Flat and Elongated Pieces in Crushed
Stone Used as an Aggregate in Concrete Pavements," Crushed Stone Journal, Vol. VI, No.5,
May, 1930, pp. 3-11; Rock Products, Vol. 33, Dec. 6, 1930, p. 53.
75) F.C. Lange, "Effect of Shape and Character of Coarse Aggregate on the Strength of
Concrete," Concrete, Vol. 32, No. 3, 1928, pp. 37-38.
76) Stanton Walker, "Effect of Flat and Elongated Particles in Mineral Aggregates," Proceedings,
Highway Research Board, Vol. 10, 1931, pp. 93-100.
77) R. Pramstig, "Manufacturing Concrete Aggregates with a Good Particle Shape," Transactions,
--``,```,`,`,`,,,,`,`````,`,,`,,-`-`,,`,,`,`,,`---
7th International Congress on Large Dams, Paris, 1962, pp. 425-430.
78) C.H. Scholer, "Durability of Concrete," Proceedings, Highway Research Board, Vol. 10, 1931,
p.132.
79) L.B. Mercer, "Aggregate Particle Shape Determination," Pit and Quarry, Vol. 44, No. 11,
June, 1952, pp. 111-112.
80) L.B. Mercer, "Effect of Particle Orientation Upon Compressive Strength Results," Pit and
Quarry, Vol. 45, No.3, Sept., 1952, p. 212.
81) E. W. Lane and E.J. Carlson, "Some Observations on the Effect of Particle Shape on the
Movement of Coarse Sediments," Transactions, American Geophysical Union, Vol. 35, No.
3, June, 1954, pp. 453-462.
82) H. J. Gilkey, "Coarse Aggregate in Concrete as a Field for Needed Research," Journal,
American Concrete Institute, Proceedings, Vol. 23, 1927, pp. 363-414.
83) "Federal Specification for Crushed-Stone, Crushed Slag, and Gravel; (for) Sewage-Trickling-
Filter Media," SS-C-744a, Federal Standard Stock Catalog, Dec., 1949.
84) G. Lees, "The Measurement of Particle Shape and Its Influence on Engineering Materials,"
Journal of the British Granite and Whinstone Federation, Vol. 4, No.2, Autumn, 1964, pp. 17-38.
85) V.M. Malhotra, "Correlation Between Particle Shape and Surface Texture of Fine Aggregates
and Their Water Requirement," Materials Research and Standards, American Society of Testing
and Materials, Vol. 4, Dec., 1964, pp. 656-658.
86) A. T. Goldbeck, Ct7tshed Stone Jounral, Vol. 14, No.5, 1939, p. 8.

Daneshlink.com
daneshlink.com
87) J.E. Backstrom; R.W. Burrows; R.C. Mielenz; and V.E. Woldkodoff, "Origin, Evolution,a nd
Effects of the Air Void System in Concrete. Part 3. Influence of Water-Cement Ratio and
Compaction," Journal, American Concrete Institute, Proceedings, Vol. 55, Sept., 1958, pp. 359-
375.
--``,```,`,`,`,,,,`,`````,`,,`,,-`-`,,`,,`,`,,`---
88) A.T. Goldbeck, "A Digest of a Report on Effect of Stone Dust on the Properties of
Concrete," Proceedings, American Society of Testing and Materials, Vol. 29, Part 1, 1929, p.
301.
89) "Federal Specification for Aggregate; (for) Portland-Cement Concrete," SS-A-281 b, Federal
Standard Stock Catalog,July 8,1954, withAmendment-1,Jan. 25, 1957.
90) Revised Guide Specifications for Concrete for Civil Works Construction, Corps of Engineers, Office,
Chief of Engineers, Washington D.C., 1946.
91) Concrete Guide Specifications, Corps of Engineers, Office, Chief of Engineers, Washington D.C.,
1953. (Par. -06 d(2), "Particle Shape"); also "Standard Guide Specifications for Concrete,"
CE 1401.01, Aug., 1963.
92) "Standard Practice for Concrete," Corps of Engineers, Office, Chief of Engineers,
Washington, D.C., 1953. (Par. 3-03 d(3), "Particle Shape"); also EM 1110-2-2000, Dec. 15,
1963.
93) "Single-Sized Gravel Aggregates for Roads," British Standard 1984, British Standard Institute,
1953.
94) "Concrete Aggregates from Natural Sources," British Standard 882:1201, British Standards
Institute, 1954.
SUPPLEMENTARY BIBLIOGRAPHY
95) L.H. Banning and F.D. Lamb, "A Vibrating Screen Surface for the Removal of Flat and
Elongated Pieces from Crushed Stone," Repott ofInvestigation No. 3 781, U.S. Bureau of Mines,
Sept. 1944.
96) R.F. Blanks, Proceedings, American Society of Civil Engineers, Vol. 75, No.4, 1949, pp. 441-
468.
97) "Single-Sized Roadstone and Chippings," British Standard 63, British Standard Institute, 1951;
RoadAbstracts, Vol. XX, No.9, No. 800, Sept., 1935.
98) R.G.H. Clements, "Particle Shape and Surface Characteristics of Aggregates," Journal, Society
Chemical Industry (Chemistry and Industry), Vol. 56, No. 11, 1937, pp. 248-250.
99) E.P. Cox, "A Method of Assigning Numerical and Percentage Values to the Degree of
Roundness of Sand Grains," Journal ofPaleontology, Vol. 1, Dec., 1927, pp. 179-183.

73
Daneshlink.com
daneshlink.com
100) W.M. Dunagan, "The Application of Some of the Newer Concepts to the Design of
Concrete Mixes," Proceedings, American Concrete Institute, Vol. 36, 1940, pp. 649-684.
101) W.M. Dunagan, "Aggregate Sorting Machines," Engineering News-Record, Vol. 124, No. 11,
1940, pp. 67-70.
102) R. Durand, "Differentiates Definitions Techniques de la Forme des Galets" (Different

Technical Definitions of Pebble Shape), La Houille Blanche, No. Special "A," 1951, pp. 253-
262.
103) A.T. Goldbeck, "A Study of the Effect of Shape of Particle on the Mortar-Making
Properties of Stone Sand," Crushed Stone Journal, Vol. 13, 1938, pp. 3-6.
104) Moreland Herrin and W.H. Goetz, "Effect of Aggregate Shape on Stability of Bituminous
Mixtures," Proceedings, Highway Research Board, Vol. 33, 1954, p. 293.
105) H. Heywood, "Numerical Definitions of Particle Size and Shape," Chemistry and Industry, Vol.
56, No.7, Feb. 13, 1937, pp. 149-154.
106) J. Hirschwald et al., "Untersuchungen and kleinschlagdecken behufs gewinnung einer

grundlage fur die prufung der naturlichen gesteine auf ihre verwenbarkeit als
strassenbaumateriel," Bautechnische Gesteinsuntersuchung, Vol. 4, No.7, 1918, pp. 39-53.
107) D.L. Inman, "Areal and Seasonal Variations in Beach and Nearshore Sediments at La Jolla,
California," Technical Memorandum No. 39, U.S. Beach Erosion Board, Mar., 1953.
108) A. Kleinlogel, Influences on Concrete, Frederick Ungar, New York, N.Y., 1951. Translated from
1941 edition.
109) H.B. Milner, Sedimentary Petrograply, Thomas Murphy and Co., London, 1940.
--``,```,`,`,`,,,,`,`````,`,,`,,-`-`,,`,,`,`,,`---
110) Z. Schiel, "Die Komform der Zuschlagstoffe in Strassenbau," Forschungsarbeiten aus dem
Strassenwesen, Volk und Reich Verlag, Berlin, Band 29, 1941, p. 75.
111) A.C. Tester, "The Measurement of the Shape of Rock Particles," Journal of Sedimentary
Petrolo!!Ji, Vol. 1, No.1, May, 1931, pp. 3-11.
112) A.C. Tester and H.X. Bay, "The Shapometer: A Device for Measuring the Shapes of
Pebbles," Science, Vol. 73, No. 1899, May 22, 1931, pp. 565-566.
113) D. Thoenes, "Straatzand," Wegen, Vol. 23, No.2, Feb., 1949, pp. 25-31.
114) Stanton Walker and C.E. Proudley, "Effect of Flat Particles on Concrete Making Properties
of Gravel," Proceedings, American Society of Testing and Materials, Vol. 29, 1929, p. 322;
National Sand and Gravel Bulletin, July, 1929.

74
Daneshlink.com
daneshlink.com
11 5) "Method of Petrographic Examination of Aggregates for Concrete," Designation CRD-C

127-54, Handbook for Concrete and Cement, U.S. Army Engineers, Waterway Experiment
Station, Vicksburg, Miss., 1954.
116) Bryant Mather, "Shape, Surface Texture, and Coatings," Signijicance ofTests and Properties on
Concrete and Concrete Aggregates, ASTM SIP 169, American Society of Testing and Materials,
1955, pp. 284-296.
--``,```,`,`,`,,,,`,`````,`,,`,,-`-`,,`,,`,`,,`---

75
Daneshlink.com
daneshlink.com
WtT
5.0 4.0 3.0 20 1.5 1.0
.......,...,-r-....."""1'1'---r---,-----., 1.0
1.5
L/W
b/a 2.0
0.
3.0
4.0
5.0
--``,```,`,`,`,,,,`,`````,`,,`,,-`-`,,`,,`,`,,`---
0
0
c/b
Fig. 1 - Relation of sphericity (curoes) calculatedfrom (b/aJ = (sphericity-3) / (c/b), Zingg Classes (I-IV), andASTM
C 125 Classes (W /T, L/W). After Krumbein [41, Fig. 4]. Stippled area indicates particles classified as flat, elongated, or
both ry the Corps ofEngineers [31J using W /T and L/W = 3 as criteria.
0.8.5
61;,a
0.80
D Gravel
~
161a • Crushed Stone
~ 0.75
,q
- 0
l.t.2a (I
lila
0
v
-~ II>
.c: 0
C!.
VI 0.70 161b
-"' 0 0 0
~
178.1
~
14.1b
0
•
e11sa
..
:::;;
0.65 61b
0 0 0
l~C
2S'j')b
... •
0
•
0.60
61C
• •
61bb
(I 0 27Bb ~
• r~b
. , 1&::
0.55
3.0 5.0 7.0 9.0 11.0 13.0 15.0 17.0 19.0
Particle Index
Fig. 2 - Relation between particle index and Wadeli's sphericity value of typical aggregate samples after [29].

76
Daneshlink.com
daneshlink.com
sin a= 1:3
t:J. = 3.1
Fig. 3 - Caliper, or gage, as developed f:?J Schulze [24].
--``,```,`,`,`,,,,`,`````,`,,`,,-`-`,,`,,`,`,,`---

Daneshlink.com
daneshlink.com
(t)
a) Test for elongation- Rightjaw of caliper is set equal to length L ofparticle; liftjaw is then automaticallY set toLl 3; hence,
it is "elongated" qy the Corps ofEngineers criterion.
b) Test for flatness- Rightjaw of caliper is set equal to width W ofparticle; liftjaw is then set to WI 3. Particle is tested to see
--``,```,`,`,`,,,,`,`````,`,,`,,-`-`,,`,,`,`,,`---
if its thickness T is less orgreater than WI 3; T for this particle is <WI 3; hence, it is 'Jiat" f?y the Corps ofEngineers
criterion.
c) Test for elongation- Test is similar to a) but particle is not elongated.
Fig. 4 - Corps ofEngineers caliper.

78
Daneshlink.com
daneshlink.com
PART IV- DURABILITY OF CONCRETE

--``,```,`,`,`,,,,`,`````,`,,`,,-`-`,,`,,`,`,,`---
Part IV deals in a broader sense with the durability issues of concrete. Durability is the ability of
concrete to resist weathering action, chemical attack, abrasion, and other conditions of service. The
durability aspects of concrete are becoming increasingly important as engineers and material
scientists recognize its importance in performance-based design philosophy.
Katharine and Bryant Mather were extremely interested in durability issues of concrete. They wrote
a number of papers dealing with durability, its importance for concrete as a material and its
implications on structural concrete. A total of seven notable papers of Katharine and Bryant were
selected to be included in this section.
The paper "Reflections on Concrete Durability and on International Conferences on Concrete

Durability," published in 1987, provides readers with the insight ofi<atharine and Bryant's
understanding of the subject. In this paper they concluded that concrete need not deteriorate and
that concrete can be made as durable as desired but in doing so, it should be incumbent on the
materials engineer not to increase the cost of the material more than is necessary. In discussing
factors that influence the deterioration of concrete and the manner in which cracking of concrete
may be induced by environmental effects, they also emphasized the factors that may cause
deterioration and cracking as a consequence of internal expansion in the concrete.
This part is further divided into three sections:

• Resistance to freezing and thawing;
• Resistance to alkali-silica reaction (ASR); and
• Resistance to sulfate attack.
Under the heading of Resistance to Freezing and Thawing, two papers were selected for this
publication. In his paper "How to Make Concrete that Will Be Immune to the Effects of Freezing
and Thawing," published in 1990, Bryant discussed the recipe for concrete that will be immune to
the effects of freezing and thawing: sound aggregate; proper air-void system, and a compressive
strength of about 4,000 psi. He extensively evaluated this distress and provided guidelines for
making concrete immune from the effects of freezing and thawing. One of the most important
factors for making concrete immune from freezing and thawing is the entrainment of air voids. In
his paper "Crystal Growth in Entrained-Air Voids," published in 2001, Bryant discussed how non-
air-entrained concrete that can get saturated and be subjected to cycles of freezing and thawing can
exhibit cracking and disintegration. He stated that sometimes crystals are found in entrained air
voids of concrete that have been "left out in the rain" or stored while damp. Bryant concluded that
such crystals are not expected to seriously compromise the resistance to freezing and thawing of
properly air-entrained concrete.
Under the heading of Resistance to Alkali-Aggregate and Alkali-Silica Reactions, two of his classic
papers describing the extensive studies in relation to the Tuscaloosa Lock and the study of Sulfate
Attack on Hydraulic Cement have been reproduced in this publication. The paper "Cracking of
Concrete in the Tuscaloosa Lock" has been recognized as one of the milestone papers dealing with
the issues of ASR. In this paper that was published in 1952 after a comprehensive investigation, he

79
Daneshlink.com
daneshlink.com
concluded that the alkali-silica reaction was the major cause of the cracking and not the sulfate
attack. He described that calcium sulfoaluminate indicative of sulfate attack was also found but was
not a significant factor. Katharine Mather performed petrographic work for this investigation.
Together, Katharine and Bryant evaluated alkali-silica reaction and showed how to identify alkali-
silica reaction. Bryant also provided guidelines on how to avoid excessive expansion due to alkali-
silica reaction in "How to Make Concrete That Will Not Suffer Deleterious Alkali-Silica Reaction,"
published in 1999.
Under the heading Resistance to Sulfate Attack, two papers were selected. The first paper entitled
"Sulfate Soundness, Sulfate Attack, and Expansive Cement in Concrete," published in 1969, deals
with expansion related to sulfate soundness, sulfate attack, and expensive cement. Bryant again
addressed sulfate attack in 2000 using the same concluding remark of his paper of 1969. In his paper
"Sulfate Attack on Hydraulic-Cement Concrete," Bryant stressed the need for continuous study into
sulfate attack on concrete.
--``,```,`,`,`,,,,`,`````,`,,`,,-`-`,,`,,`,`,,`---

80
Daneshlink.com
daneshlink.com
Reprinted from Concrete Durability: Katharine and Bryant Mather International Coiference, ACI SP-1 00, V. I and II, 1987,
Detroit, MI, 2179 pp., ©American Concrete Institute.
REFLECTIONS ON CONCRETE DURABILITY AND ON

INTERNATIONAL CONFERENCES ON CONCRETE DURABILITY
--``,```,`,`,`,,,,`,`````,`,,`,,-`-`,,`,,`,`,,`---
by K. and B. Mather
SYNOPSIS
The Organizing Committee for this conference requested that we make some remarks and prepare
some comments for inclusion in the Proceedings. These words are intended as a basis for the former
and to serve as the latter. We observe that: Concrete is international; it is made locally; it has infinite
variability; it can be made to be very uniform; and it can be made to last as long as you want it to. An
unsolved problem is assessment of the nature and severity of the exposure, so that requirements can
be graduated according to severity. Only when this problem is solved will we be able to stamp out
specification overkill.
The Organizing Committee also suggested we furnish biographical data and bibliographic data. The
standard material accompanying this publication should suffice for the former. An appendix of
selected items is provided for the latter purpose.
Keywords: concrete durability; concretes; exposure; specifications; variability.
ACI Honorary Member Katharine Mather, a resident of Clinton, Mississippi, retired from the
Waterways Experiment Station, Vicksburg, Mississippi, in 1982 after 40 years of service with the
U.S. Army Corps of Engineers. Mrs. Mather received the Anderson Award in 1982, and the Wason
Medal in 1955. She has served on the ACI Board of Direction, and is a member of numerous
technical committees. She is also an honorary member of ASTM and is a fellow of the Mineralogical
Society of America.
Bryant Mather is Chief, Structures Laboratory, Waterways Experiment Station, U.S. Army Corps of
Engineers, Vicksburg, Miss., and has been with the Corps since 1941. He is an Honorary Member
and a past president of ACI and is a member of several technical committees. He is currently
Chairman of Committee 116, Notation and Nomenclature. His awards and honors include ACI's
Henry C. Turner Medal in recognition of "his many years in improving the quality of concrete work
... and for his vigorous personal efforts both inside and outside of that assignment."
INTRODUCTION
During the 44th Annual Meeting of the Highway Research Board (HRB) in 1965, Committee MC-
B6 on Basic Research Pertaining to Portland Cement, of which Bill Dolch was Chairman and
Katharine Mather was a member, put on the T. C. Powers Symposium on Structure of Portland
Cement Paste and Concrete, under the cognizance of the HRB Concrete Division of which Bryant
Mather was chairman. Special Report 90, the proceedings of that meeting, is a landmark that all
serious workers should have access to and know.

81
Daneshlink.com
daneshlink.com
The idea that anyone would put on a "Mather Symposium," in some way analogous to the Powers
Symposium, was one that never occurred to either of us and one we found incredible when John
Scanlon told us he was going to do it unless we seriously objected. As can be seen, we didn't. I might
point out that we have here a "Conference," not a "Symposium." There is a story, once published in
the ASTM Bulletin, about how a senior research scientist of a large corporation visited the Chief
Executive Officer to request money to support a symposium and the CEO said ''What is a
symposium?" Whereupon they turned to a dictionary and found, to the dismay of the scientist:
"Symposium, fr. Gr. 'symposion' a drinking party..." At that point the CEO said "I approve, nobody
needs a drinking party more than you research types." We expect that those of you who desire to
participate in a drinking party will not leave Atlanta unsatisfied!
--``,```,`,`,`,,,,`,`````,`,,`,,-`-`,,`,,`,`,,`---
The Organizing Committee has asked that we prepare some comments to go in the Proceedings and to
serve as the basis of remarks that we are asked to make on the fmal day of the conference. In view
of all that the Organizing Committee has put into making the conference a successful contribution
to "durable" concrete and to all of us who have said, in one way or another, that "concrete need not
deteriorate," it is with gratitude that we try to comply with its request.
This Conference has again demonstrated a number of fundamental truths and relationships, most of
which are quite well known to most of those who participated- especially those more than 40
years old. Among them are the following:
Concrete is international and mqy well be or become intet;galactic. Concrete is made everywhere that people
build. Concrete is of interest to those who build regardless of their language, their national heritage,
their climate, or their cultural heritage. Concrete is like air to breath, water to drink, earth to grow
plants in; it provides for all what Dick Gaynor's 1976 National Ready Mixed Concrete Association's
U.S. Bicentennial bumper sticker said: "Concrete-the Foundation of America" and which can be
expanded to "Concrete-the Foundation of Civilization."
Concrete is made locai!J. The mixing water, with negligible exceptions (perhaps when concrete is made
on the moon); the aggregates usually; the cement often, is made from one or more locally available
natural raw materials, in various degrees of processing.
Concrete has itifinite variabili!J. The varieties of concrete that are intentionally produced are more
numerous than the known varieties of natural sedimentary rocks. Concretes have been-and will
be-required to be produced having unit masses from as low as 160 kg/ m 3 (1 0 lb mass per fe) to as
high as 4800 kg/ m 3 (300 lb mass per ft\ that have unconfined compressive strengths from as low as
0.35 MPa (50 psi) to as high as 275 MPa (40,000 psi); that are of any color-white, black, pink,
turquoise, chartreuse, or as one Corps of Engineers' officer demanded: "sand tan;" we await the
demand for "olive drab."
Concrete can be made to be very unifOrm. ACI Committee 214 was organized in 1946. It told concrete
makers and users how to use strength tests to measure uniformity of concrete as produced. It is still
doing it. In 1979, ASTM Standard Method C 917 for Evaluation of Cement Strength Uniformity
was adopted. For many years ASTM Specification C 33 has put rather restrictive limits on
uniformity of grading of fine aggregate in terms of fmeness modulus. There are uniformity
requirements on pozzolans given in ASTM Specification C 618.
82
Daneshlink.com
daneshlink.com
Concrete can be made to last as long asyou want it to. Even as some relatively fragile sedimentary rocks
exist that have survived some hundreds of millions of years, so there are concretes that, if they
remain in relatively mild environments, could last forever, or at least as close to "forever" as we can
measure. The duration of the life of a concrete with its relevant properties not sufficiendy degraded
so as to impair or jeopardize its ability to serve its intended purpose is not usually a function of the
absolute level of any universally relevant property of the concrete, such as slump, unit mass,
strength, modulus of elasticity, or color, but is almost always a function of the interaction of some
property or properties, the required levels of which depend on the severity of the environment to
which the concrete will be exposed in service. One need not worry much about resistance to
freezing and thawing in the tropics, except in facilities using refrigeration; or at the poles, except in
facilities using intermittent heating. One need not worry about acid attack, abrasion attack, sulfate
attack, attack by burrowing mollusks, or any of the myriad hazards described by Kleinlogel, Biczok,
or others, except where they exist and impact the concrete of interest. And where these potentially
destructive influences can be and are brought to bear on the concrete, if their action and its intensity
and duration are properly anticipated, the concrete can be rendered adequately resistant so as to
merit the designation "durable," even when as demanded of us by the Nuclear Regulatory
Commission, "durable" means not significandy degraded for many centuries or in some cases
hundreds of centuries.
The message we have decided to leave with you at this time is (a) that concrete need not deteriorate
and (b) that knowledge exists by the use of which concrete can be caused to exist in any of the
environments of service for which it may be reasonably proposed with tolerable degradation during
its intended service life, if precautions to limit degradation appropriate to the environmental severity
are taken.
The Unsolved Problem. The important matter is to know how properly to assess the nature and severity
of the environment of service; to get a reasonable estimate of intended service life; to get a
reasonable assessment of how much degradation is tolerable structurally and, when relevant,
aesthetically. One must then, from these and known relationships of environmental influences and
concrete composition and constitution, have procedures and criteria by which to select materials,
proportions, and construction practices that will, with reasonable probability of success and
compliance with the owner's budget for the work, give concrete that will not deteriorate.
Most of the raw material of knowledge to do this exists, some in relatively quite refined states. But
the manner in which this knowledge is used in an attempt to make it an effective tool for makers
and users of concrete is primitive. Today we talk about reactive vs. non-reactive aggregate and high-
alkali vs. not high-alkali cement as if all structures, in all environments, involving all ambient
temperatures and temperature ranges, in all degrees of moistness from "it rains on the average every
12 years" to the tidal zone in Mombasa, as if all the precautions and restrictions must be taken at
100 % of the degree we ever need anywhere, everywhere we need to do anything at all. There is no
relaxation of the 0.60% Na20 limit on cement as a function of environmental severity, reactivity of
aggregate, or anticipated level of permeability of concrete. An aggregate is either reactive-or not; if
not, nothing need be done, no precautions taken; if "reactive," then the precautions must be taken
without reference to degree of reactivity or degree of severity of exposure.
Today we talk about need for resistance to freezing and thawing. If the concrete will ever likely
freeze when any of the spaces in it that can hold freezable water are more than 91 %full, we need
air-entrainment, sound aggregate, and protection until moderate maturity is achieved. And, "air-
--``,```,`,`,`,,,,`,`````,`,,`,,-`-`,,`,,`,`,,`---

83
Daneshlink.com
daneshlink.com
entrainment" means an air content that will give an air-void system with an L, a bubble spacing
factor, less than 0.2 mm (0.008 in.) regardless of whether we have one or one hundred cycles of
freezing and thawing per year and regardless of whether the rate of freezing is fast or slow or
whether the minimum temperature achieved 10 mm below the exposed concrete surface is minus
5 Cor minus 50 C. This topic was elegantly explored by Treval Powers years ago and was brought
back to us this week by Bob Philleo.
We have decided to refer to these sorts of approaches as "specification overkill." It involves, much
of the time, paying more than needs to be paid to buy a level of protection that, in many cases, can't
ever be needed and from which no good can ever be derived. "Ever" in the preceding sentence
--``,```,`,`,`,,,,`,`````,`,,`,,-`-`,,`,,`,`,,`---
means not more often than once in a million years. Today we have computers that can be run by
most bright junior high school scientists that could be used to take the knowledge that does exist
and derive from it reasonable criteria for matching degree of precautions needed or degree of
restriction required of the contractor in his exercise of choice among materials, proportions, and
practices; so that only limited precautions and light restrictions are called for when the nature and
intensity of the anticipated environmental hazards and insults the concrete must resist are slight or
of short duration in time. We can then reserve the levels of precautions we now take for all levels of
severity for those cases where the hazard is really very severe and protection is needed for a very
long time. We might well, by so limiting these requirements in their applicability, be able to increase
their degree of protection so as further to minimize the probability that they will be less effective
than needed.
Our challenge to all of you therefore is: Since concrete need not deteriorate, ensure that it doesn't,
but do so without raising the cost more than necessary; stamp out specification overkill.

84
Daneshlink.com
daneshlink.com
APPENDIX
We have selected from the approximately 500 titles in our respective and joint lists of publications
the following for the reasons stated.
Mather, B., 1945, "Effect of Type of Test Specimen on Apparent Compressive Strength of
Concrete," Proceedings, ASTM, V. 45, pp. 802-812. Results of tests of 2375 specimens, drilled cores,
molded cylinders, modified cubes suggested that factors relating specimen types varied with strength
level.
Mather, K., 1947, "Relation of Absorption and Sulfate Test Results on Concrete Sands," American
Sociery for Testing and Materials Bulletin, No. 144, Jan., pp. 26-31. Test data on samples of 409 sands
showed a high correlation of the performance test value and the property test value but this cannot
be used as an acceptance criterion for a specific material.
Mather, B. 1948, "Petrographic Identification of Reactive Constituents in Concrete Aggregates,"

Proceedings, ASTM, V. 48, pp. 1120-1125. A description of criteria for recognizing opal, chalcedony,
tridymite, cristobalite, acid and intermediate volcanic glass in samples of aggregate.
Mather, K. and Mather, B., 1950, "Methods of Petrographic Examination of Aggregates for
Concrete," Proceedings, ASTM, V. 50, pp. 1288-1313. Also 1950 Preprint, No. 76, 25 pp. The
description of the techniques and procedures that became ASTM C 295-54.
Mather, B., 1952, "Cracking of Concrete in the Tuscaloosa Lock," Proceedings, Highwqy Res. Bd., V. 31,
pp. 218-233. A report on an extensive investigation primarily by Katharine Mather of what turned
out to be the Corps of Engineers' first major example of alkali-silica reaction in a structure in
serv1ce.
--``,```,`,`,`,,,,`,`````,`,,`,,-`-`,,`,,`,`,,`---
Mather, K., 1953, "Applications of Light Microscopy in Concrete Research," Symposium on Light
Microscopy, ASTM Special Technical Publication No. 143, pp. 51-70. A review of the development,
since 1882, and state-of-the-art including some previously unpublished results of research on mass
concrete, selected for the ASTM Sanford E. Thompson Award for 1953.
Kennedy, T. B., and Mather, K., 1953, "Correlation Between Laboratory Accelerated Freezing and
Thawing and Weathering at Treat Island, Maine," Journal, American Concrete Institute, Proceedings, Title
50-9, V. 50, pp. 141-172. Also WES Bulletin, 1954. Authors' closure to discussion, ibid (December
19 54), pp 172-15 to 172-20. A major attempt to relate laboratory test data to field test data to
performance in service. Selected for the ACI Wason Medal for Materials Research for 1955.
Mather, B., 1957, "Laboratory Tests of Portland Blast-Furnace Slag Cements," Journal, ACI,
proceedings, V. 54, pp. 205-232. A report on a study of samples of the product at all six of the plants
making portland blast-furnace slag cement in the U.S.A. in 1952.
Mather, B., and Mather, B., 1958, The Butterflies rifMississippi, Tulane Studies in Zoology, V. 6, No.2,
New Orleans, La., pp 63-109, six figures. Supplements 1, 2, 3 have been published in the Jour. Lepid.

85
Daneshlink.com
daneshlink.com
Soc., 13:71-72, 1959; 30:197-200, 1976; and 39:134-138, 1985. Recorded occurrence in Mississippi of
150 species of butterflies; increased from 83 to 122 to 144 to 150.
Pepper, L., and Mather, B., 19 59, "Effectiveness of Mineral Admixtures in Preventing Excessive
Expansion of Concrete Due to Alkali-Aggregate Reaction," Proceedings, ASTM, V. 59, pp. 1178-1203.
Based on the Waterways Experiment Station Technical Report 6-841 (1958) showed that the
amount of pozzolan or slag required to be introduced into a high-alkali cement- reactive aggregate
--``,```,`,`,`,,,,`,`````,`,,`,,-`-`,,`,,`,`,,`---
mortar mixture to reduce expansion by 75% ranged from 10 to 45% and varied as ftneness, silica
solubility, and silica content changed. Selected for the 1961 ASTM Sanford E. Thompson Award.
Mather, K., 1965, High-Strength, High-Density Concrete, WES MP 6-698, 10 pp.; also Proceedings,
Journal of the American Concrete Institute, V. 62, No.8, pp 951-962; reprinted on pp. 1587-1596 of ACI
SP-34, 1972, Concrete for Nuclear Reactors, Concrete having unit weights of 225-235 lb/ ft3 made using
each of two sorts of iron ore (magnetite and ilmenite) as high-density aggregate and having water-
cement ratios of 0.30 and 0.35 by mass developed strengths of 10,000 to 12,000 psi. Concrete can be
simultaneously high-strength and high density.
Mather, B., 1968, "Stronger Concrete," in A Symposium on Concrete Strength, Highway Research Record
No. 210, NAS-NAE-NRC Publication 1553, pp. 1-28. A state-of-the-art review based on one
prepared for the Air Force listing reports of concrete having 10,000 psi strength and higher and
factors affecting production of such concrete.
Mather, B., 1968, "Field and Laboratory Studies of the Sulfate Resistance of Concrete," in
Peifowance of Concrete - Resistance of Concrete to Sulphate and Other Environmental conditions -A Symposium in
Honour ofThorbet;gur Thoroaldson, Canadian Building Series, No.2, University of Toronto Press, pp.
66-76. A review of the chemical reactions involved in sulfate attack and an attempt to relate results
of chemical analysis and calculated phase composition of cements to fteld and laboratory test results.
Mather, B., 1969, "Deterioration of Concrete in Eisenhower Lock," Trans., Ninth International
Congress on Lat;ge Dams, V. VI, pp. 427-429, Paris. A report to the international dam building
community that for as severe an exposure as the St. Lawrence Seaway having very sound aggregate
and a proper air-void system is not enough; it is also necessary that a degree of maturity be achieved
before concrete can, with impunity, be allowed to freeze while critically saturated.
Mather, B., 1969, "Sulfate Soundness, Sulfate Attack, and Expansive Cement in Concrete," in
Proceedings, RII.EM International Symposium on Concrete Durabiliry, Technical University, Prague,
Czechoslovakia, pp. C-209 to C-220. A discussion of relations and differences among the
phenomenology of the "sulfate" test to predict frost effects; sulfate attack on concrete; and the use
of expansive cement. The concept of "crystal growth pressure" is questioned.
Mather, B., 1969, Research on Concrete, Stanton Walker Lecture Series on the Materials Sciences,
Lecture No.7, University of Maryland (National Sand and Gravel Assn., National Ready Mixed
Concrete Association, Washington, D.C.), 15 pp. A review of topics for and beneftts of concrete
research in which it was concluded that "As humanity tries to cope with the exponential increase in
population, pavements, and pollution, research on concrete has an increasingly important role to
play in providing the knowledge that will permit society to provide pavements-and other concrete
structures-for people and reduce pollution."

86
Daneshlink.com
daneshlink.com
Mather, B., 1970, "Portland Cement Concrete, Research, Testing and Performance," ASTM
Journal ofMaterials, V. 5, No.4, pp. 830-841. The 1970 ASTM Edgar Marburg Lecture. It included: "I
believe that the next two major areas for development, innovation, and improvement in
specifications for pordand cement will be to provide assurance (1) of greater uniformity of behavior
and (2) that the cement will perform more uniformly in concrete with the selected kinds of
admixtures with which it is used."
Mather, B., 1974., "Use of Concrete of Low Pordand Cement Content in Combination with
Pozzolans and Other Admixtures in Construction of Concrete Dams," Journal, American Concrete
Institute, Proceedings, V. 71, pp. 589-599, and Author's Closure, Journal, American Concrete Institute,
Proceedings, V. 72, pp. 296-297,June 1975. "Mass concrete having appropriate levels of relevant
properties for the construction of dams may be produced with pordand-cement contents as low as
60 lb/yd3 (36 kg/m3) and with a water content ofless than 80 lb/yd3 (48 kg/m3).
Mather, B., 1976, "How Soon is Soon Enough?" Journal, American Concrete Institute, Proceedings, V. 73,
No.3, pp. 147-150. The time to be sure that what is in a batch of concrete is correct is when it is
hatched, not later; the time to be reassured is when the batch is dumped and tested for slump and air
content; the final check is when the accelerated cured specimen is tested at an age of 24 h.
Mather, K., 1978, "Petrographic Examination" in Significance of Tests and Properties of Concrete and
Concrete-Making Materials, ASTM STP 169-B, pp. 132-145. This revision of the state-of-the-art article
that was In STP 169, 1955, and 169-A, 1966, deals with concrete and closes with the remark by St.
Paul as quoted in Episde to the Hebrews, Chapter 11, Verse 3: "Things which are seen"-concrete
and mortar-"were not made of things which do appear"-to the eye and to the light microscope.
This paper, together with the others in the volume, shared the ASTM Dudley Medal.
Mather, B., 1979, "Concrete Need Not Deteriorate," Concrete International, ACI, V.l, No.9, pp. 32-
37.
Mather, B. 1980, "Use Less Cement," Concrete International, V. 2, No. 10, pp. 22-24. How to save
energy and make better concrete provided the concrete has the needed levels of relevant properties.
Mather, K., 1981, "Condition of Concrete in Martin Dam After 50 Years of Service," Cement,
Concrete, and Aggregates, ASTM, V. 3, No.1, pp. 53-63. A case history of concrete in a structure with
extensive evidence of alkali-silica reaction but which appears to be ready to serve its owner
successfully for another half century.
Mather, K., 1981, "Factors Affecting Sulfate Resistance of Mortars," Proceedings of the 7th International
Congress on the Chemistry of Cement, Paris, V. IV, pp. 580-585; also WES MP SL-80-17, 18 pp. (CTIAC
Report No. 44.) Summarizes recent work showing that some pozzolans are very effective in
--``,```,`,`,`,,,,`,`````,`,,`,,-`-`,,`,,`,`,,`---
preventing expansions of mortar made with sulfate susceptible cements, others are not. The ones
found effective were silica fume and volcanic glass. The ones found less effective or, in some cases,
harmful were fly ashes. A tendency for lignite fly ash to be worse was noted.
Mather, K., 1985, "Preservation Technology: Evaluating Concrete in Structures," Concrete

International, V. 7, No. 10, pp. 33-41. This paper prepared and presented at the ACI Convention in
Boston in 1975 was intended primarily to share concrete technology with architectural and historic
preservation technologists.

87
Daneshlink.com
daneshlink.com
Mather, B., 1986, "Selecting Relevant Levels of Quality," Concrete International, V. 8, No. 3, American
Concrete Institute, pp. 30-36. Presented at the Abdun-Nur/Cordon symposium on Quality
Concrete in Construction, the intent was to suggest that achieving a given level of quality in concrete
in construction is not enough. Also needed are relevant and appropriate criteria for selecting the
levels of quality to be achieved.
--``,```,`,`,`,,,,`,`````,`,,`,,-`-`,,`,,`,`,,`---

88
Daneshlink.com
daneshlink.com
Reprinted from Paul Klieger Symposium on Performance of Concrete, ACI SP-122, 1990, Detroit, MI, 499 pp., ©American
Concrete Institute.
HOW TO MAKE CONCRETE THAT WILL BE IMMUNE TO THE

EFFECTS OF FREEZING AND THAWING
By Bryant Mather
--``,```,`,`,`,,,,`,`````,`,,`,,-`-`,,`,,`,`,,`---
SYNOPSIS
Concrete will be immune to the effects of freezing and thawing if (1) it is not in an environment
where freezing and thawing take place so as to cause freezable water in the concrete to freeze, (2)
when freezing takes place there are no pores in the concrete large enough to hold freezable water
(i.e., no capillary cavities), (3) during freezing of freezable water, the pores containing freezable water
are never more than 91 percent filled, i.e., not critically saturated, (4) during freezing of freezable
water the pores containing freezable water are more than 91 percent full, the paste has an air-void
system with an air bubble located not more than 0.2 mm (0.008 in.) from anywhere (L ~ 0.2 mm),
sound aggregate, and moderate maturity. Sound aggregate is aggregate that does not contain
significant amounts of accessible capillary pore space that is likely to be critically saturated when
freezing occurs. The way to establish that such is the case, is to subject properly air-entrained,
properly mature concrete, made with the aggregate in question, to an appropriate laboratory
freezing-and-thawing test such as ASTM C 666 Procedure A. Moderate maturity means that the
originally mixing water-filled space has been reduced by cement hydration so that the remaining
capillary porosity that can hold freezable water is a small enough fractional volume of the paste so
that the expansion of the water on freezing can be accommodated by the air-void system. Such
maturity was shown by Klieger in 1956 to have been attained when the compressive strength reaches
about 4,000 psi.
Keywords: age; aggregates; air-entrained concretes; air entrainment; capillarity; compressive strength;
concretes; freeze-thaw porosity; saturation; soundness; voids; water
Bryant Mather is Chief, Structures Laboratory, Waterways Experiment Station, U.S. Army Corps of
Engineers, Vicksburg, Miss., and has been with the Corps since 1941. He is an Honorary Member
and a past president of ACI and is a member of several technical committees. He is currendy
chairman of Committee 116, Terminology and Notation.
INTRODUCTION
Most of the concrete that has ever been made in the world was made and used in environmental
circumstances in which the issue of whether or not it would be adversely affected by freezing and
thawing was irrelevant since the environmental exposure either did not involve freezing and thawing
or it could be assumed with confidence that, when the concrete froze, it would not be critically
saturated. It is not necessary to know how to make concrete that will be immune to the effects of
freezing and thawing if the concrete that one is interested in making will never be exposed to
freezing and thawing when in a critically saturated condition.

89
Daneshlink.com
daneshlink.com
Concrete that is used in environmental situations in which it is exposed to freezing and thawing
while critically saturated and which is unaffected by such exposure has been made for probably as
long as concrete has been made. Such inadvertently or unintentionally frost-resistant concrete has
been made using aggregates that were frost resistant, with cement paste that was air entrained, or
was concrete that, in spite of becoming critically saturated, had capillary porosity sufficiently low that
the 9 percent expansion of the freezable water could be accommodated by the elastic and creep
strain capacity of the concrete without rupture.
So far as I can tell from such literature as I have examined, up until about 50 or 60 years ago most
people concerned with resistance of concrete to freezing and thawing simply used the rule of trying
to figure out what materials and proportions and construction practices had been used previously
with success in a comparable exposure and attempted to duplicate or at least simulate those same
materials and proportions and practices in the new concrete; without understanding, in any
particular detail, why those aggregates, a cement of that composition, and a concrete produced in the
way a particular concrete was produced, yielded "durable" concrete.
This symposium honors Paul Klieger. With that in mind, I looked at the index of the first 227
Bulletins of the Research Department of Portland Cement Association Research and Development
Laboratories, published in July 1969, and I observed that 18 Bulletins are listed in the author index
following the entry "Klieger, Paul." I then looked at the titles and abstracts of each of these,
beginning with Bulletin 24 in 1949, which deals with the effect of entrained air on a number of
properties of concrete including freezing-and-thawing resistance, all the way down to Bulletin 218 in
1967 which deals with laboratory studies of blended cements for their effects on properties of
concrete including freezing and thawing. So far as I could tell, 15 of the 18 Bulletins listed were
concerned wholly or in part with freezing and thawing. As I shall emphasize later, it was Klieger's
1956 Highway Research Bulletin No. 150 paper that became PCA Research Bulletin 82 (Klieger
1956) that was called "Curing Requirements for Scale Resistance of Concrete" that contributed for
the first time, at least that I ever took in, the third critical element to the recipe on how to make
concrete that will be immune to the effects of freezing and thawing.
In the development of this topic, I shall state the conclusion first and then discuss some of the
backup and derivation.
RECIPE FOR CONCRETE THAT WILL BE IMMUNE TO THE EFFECTS OF FREEZING AND THAWING
There are two ways of stating the recipe, depending upon how much information the audience being
addressed is presumed to be qualified to assimilate. If the audience is not presumed to be able to
assimilate much information and should not be given the opportunity to exercise much judgment,
then the recipe might properly take the form of saying "Follow the guidance and specification
requirements stated in appropriate applicable guides and specifications." What this will then come
down to is that the recipe will require that the cementitious materials, the aggregates, the mixing
water, and the admixtures will be required to be tested to determine compliance with the appropriate
options of the applicable ASTM or other governing standards, that the mixtures shall be
proportioned according to the applicable portions of theACI Manual of Concrete Practice (ACI 211.1)
and the materials will be handled, hatched, conveyed, mixed, placed, compacted, finished, and cured,
all also in accordance with applicable documents in the ACI Manual of Concrete Practice. One of the
critical features of this recipe will be that someone must decide whether or not it is practical to have
--``,```,`,`,`,,,,`,`````,`,,`,,-`-`,,`,,`,`,,`---

90
Daneshlink.com
daneshlink.com
job-site testing for air content of fresh concrete. If so, under normal conditions, the cementitious
materials used will be nonair-entraining and the air-void system in the cement paste in the concrete
will be obtained by the use of an air-entraining admixture, the appropriateness of the dosage of
which will be confirmed by job-site testing for air content of fresh concrete. If such tests are not
practical and air-entraining cement is available, then air- entraining cement meeting ASTM C 150 for
Type IA, Type IIA, Type IliA, or ASTM C 595 for Type I (Sl\1)-A, Type IS-A, Type I (Pl\1)-A or
Type IP-A, Type IS-A (MS) or Type IP-A (MS) will be specified. The aggregates will be required to
meet ASTM C 33 which will cause only frost-resistant aggregate to be used and the mixing water
and the admixtures will be in accordance with appropriate specifications to ensure that they
contribute what they need to contribute and do not interfere with the achievement of appropriate
levels of relevant properties produced by the interaction of everything else that is going on.
Under the second alternative, where the audience is perceived to be qualified to understand technical
considerations and exercise some level of judgment, the recipe for concrete that will be immune to
the effects of freezing and thawing may take a somewhat different form. Instead of discussing
whether one shall use air-entraining cement or nonair-entraining cement and an air-entraining
admixture, the recipe may instead require that the paste portion of the concrete shall contain an air-
void system characterized by a bubble-spacing factor ( [ ) not greater than 0.008 in. (0.2 mm) (ACI
201, 212); which is to say simply that regardless of how it is achieved, the cement paste in the
hardened concrete shall contain an air bubble not farther away than 0.008 in. (0.2 mm) from
anywhere.
Two comments are appropriate at this point. First, this requirement will, under most conditions, be
met if air-entraining cement meeting the relevant ASTM cement specifications or an air-entraining
admixture meeting ASTM C 260 is used at a dosage sufficient to meet the air content of fresh
concrete requirements stated in the documents prepared by ACI Committees 201 and 212 in the
ACI ManuaL The second comment is that the 0.008 in. (0.2 mm) spacing factor is the maximum for
concrete of a particular level of permeability through which the freezing isotherm is moving at a
given rate as a function of the thermal properties of the concrete and the temperature difference
between the near surface concrete and the surroundings. What this means is that, assuming the
hydraulic pressure mechanism for frost damage, if the depth to which the temperature drops and the
speed with which it drops is great enough or the permeability of the concrete is low enough, then
the bubbles may need to be closer together if they are to serve their intended function. If the
thermal gradient is steeper, the freezing isotherm will move into the concrete more rapidly and if the
concrete is less permeable the excess water will move through it more slowly. One further
consideration is that most processes by which the permeability of concrete can be materially reduced
are also processes by which its porosity is concomitantly significantly reduced. Therefore, all other
things being equal, if it is harder for the water to get through the concrete there will also usually be
substantially less water going through it as a function of whatever mechanism is presumed to be
causing water to move or need to move in connection with the phenomenon of freezing. Finally
there is the point that the liquid in a capillary pore is a solution of given concentration before any of
it freezes. When some of it freezes the solution concentration of the rest increases and it has a lower
freezing point thus it will only freeze when a lower isotherm comes by.
The issue of aggregates may very well be dealt with, not simply by requiring with compliance of
ASTM C 33, but rather by requiring that the aggregates under consideration be combined into a
concrete mixture using either a standard air-entrained cement paste or the cement paste such as that
--``,```,`,`,`,,,,`,`````,`,,`,,-`-`,,`,,`,`,,`---

91
Daneshlink.com
daneshlink.com
likely to be used in the construction under consideration. Once the concrete has been proportioned,
mixed, formed into specimens, finished, and cured, any of several testing procedures may be
employed. Perhaps the most widely used is ASTM C 666 Procedure A, which involves
subjecting prismatic specimens to laboratory freezing and thawing while the specimen is immersed
in water or ice and determining, at selected intervals, the degree to which such treatment has
affected some measurable property of the concrete. The most widely used property is fundamental
resonant frequency of vibration from which dynamic Young's modulus of elasticity may be
calculated. However, alternatively or additionally, the specimens may be tested for pulse velocity,
length change, or change in mass, or they may be sacrificed periodically to measure flexural strength
or tensile strength or compressive strength in comparison with companion specimens moist cured
to the same age. In addition to ASTM C 666, there is also ASTM C 671, the method for critical
dilation of concrete specimens and its companion ASTM C 682, the practice for evaluation of frost
resistance of coarse aggregate in air-entrained concrete by critical dilation. This pair of methods,
derived from recommendations originated by T. C. Powers (1955) and used very effectively by
--``,```,`,`,`,,,,`,`````,`,,`,,-`-`,,`,,`,`,,`---
Bailey Tremper in the California DOT (Tremper and Spellman 1961), and later advocated strongly
by Alan D. Buck (Buck 1976) at the Waterways Experiment Station, permits the measurement of the
time required for a specimen of given initial moisture condition to achieve both critical saturation
and, consequendy, susceptibility to damage by freezing and thawing if the concrete is intrinsically
not frost resistant either by reason of containing nonfrost-resistant aggregate or a paste of
inadequate air-void system or a fractional volume of freezable water greater than can be
accommodated by the air-void system which is a consequence of inadequate maturity at the time of
freezing. Finally, there is the procedure ASTM C 672 for scaling resistance of surfaces exposed to
deicing chemicals, by the use of which one can measure this specific manifestation of lack of
immunity to freezing and thawing. ASTM C 672 is a procedure to the development and use of
which Paul Klieger made probably greater contributions than were made by anyone else (Klieger
1955, 1956; Klieger and Perenchio 1963; Verbeck and Klieger 1956). The mechanism by which a
particle of porous unsound aggregate generates internal force sufficient to produce a popout if
frozen while critically saturated has been elegandy described by Bache and Isen (1968). The
procedure for recognizing such particles has been most thoroughly studied by Larson and Cady
(1969).
Finally, after one has dealt with the air-void system in the paste and the quality of the aggregate,
there is the third critical requirement regarding concrete that is to be immune to frost action and that
is maturity and it is this requirement that we owe primarily to Paul. In his 19 56 Highway Research
Board paper, which became PCA Bulletin 82 (Klieger 1956), one of his conclusions was "the
development of a certain level of strength has merit, as an index to the amount of curing required
for air-entrained concrete before permitting the use of deicers." What he found was that for a
number of different concretes made with Types I, II, and III cement and, in the case of the Type I
and Type II with and without 2 percent calcium chloride, cured at 73, 40, and 25° F, that it took
anywhere from 7 days to more than 60 days of moist curing to develop satisfactory resistance to the
form of freezing and thawing damage called "deicer scaling." He then reported the compressive
strengths of these concretes and they ranged in general from somewhat over 3,000 to somewhat
over 4,000 psi. To be on the safe side, I have recommended that one not allow repeated cycles of
freezing and thawing of critically saturated concrete to occur prior to the concrete having achieved a
strength of 4,000 psi. As I mentioned, Paul said this in 1956. It took awhile for some of us,
especially me, to take it in.

92
Daneshlink.com
daneshlink.com
In 1962 I presented a paper at the Highway Research Board on effects of duration of moist curing
on freezing and thawing of concrete as measured by ASTM C 666 Procedure A (Mather 1962). In it
results were given of quite a lot of tests on quite a number of different concretes made with and
without entrained air at water-cement ratios of 0.5 and 0.8 by mass, subjected to freezing and
thawing after 14 and 180 days of moist curing at 73° F. It was clear from these data that nonair-
entrained concrete, even at 0.5 water-cement ratio, cured for 180 days, which developed strengths of
up to 8,100 ps~ never got a durability factor greater than 10. On the other hand, air-entrained
concrete of 0.8 water-cement ratio cured only 14 days with compressive strengths under 4,000 psi
showed durability factors rarely as high as 60. Lower water-cement ratio (0.5) concrete showed
durability factors in excess of 60 even at 14 days because the volume of originally water-filled space
was initially much less than in the 0.8 w/ c concrete and it was sufficiently reduced by cement
hydration so that when that space was filled with freezable water its fractional volume was not
greater than that which could be accommodated by the air-void system. Since these data are not
readily accessible I present them here. The concretes were made using five portland cements, with
and without an air-entraining admixture used to give an air content of 6.0 ± 0.5 percent. The
aggregate was 3/4 in. nominal maximum size crushed limestone. Mixtures were made at 0. 5 and 0. 8
w/ c by mass. Three rounds were made using a batch on each of three days for each mixture.
Figure 1 (table 1) compares 20 concretes, 10 made at 0.5 water-cement ratio, 10 made at 0.8 water-
cement ratio; 10 with entrained air to give an air content of 6.0 ± 0.5 percent and 10 nonair-
entrained; made using the five portland cements; two Type I cements (one of high- and one of low-
alkali content), and one each of Types II, III, and IV. As might have been expected, the Type III
cement concrete showed the least, and the Type IV cement concrete the greatest increase in strength
between 28 and 180 days. The nonair-entrained concretes all showed very little resistance to freezing
and thawing; but an apparent trend toward a slight increase in durability factor with increasing age
and increasing strength is suggested for those concretes with strengths of 7,000 psi or more. This
increase only brings the durability factor up to 10.
The Type III cement concrete showed a reduction of durability factor between 14 and 180 days at
both water-cement ratios, and the Type I (low-alkali) cement concrete showed a reduction at the 0.8
water-cement ratio.
Figure 2 (see table 2) shows the behavior of concretes of both water-cement ratios, made with Type
II cement, as progressive replacement of the cement, by solid volume, with fly ash is made in three
stages. For both water-cement ratios at both ages there is a progressive reduction of both strength
and durability factor as the proportion of cement replaced by fly ash increases.
Figure 3 (table 2) shows similar data for ground granulated iron blast-furnace slag.
Figure 4 (table 3) shows the behavior of concretes of both water-cement ratios made with Type IV
cement, as the cement is replaced by selected percentages by solid volume of each of six materials.
All concretes made with Type IV cement show an increase in both strength and durability factor
with age.
Figure 5 (table 3) shows similar relations for concretes made with the Type I (low-alkali) cement. At
the 0.5 water-cement ratio all show an increase in strength and durability factor with age. At the 0.8
water-cement ratio the concrete made with Type I (low-alkali) cement, as noted in Fig. 1, shows a
--``,```,`,`,`,,,,`,`````,`,,`,,-`-`,,`,,`,`,,`---

93
Daneshlink.com
daneshlink.com
decrease :in durability factor with age; such a decrease was also shown when this cement was
replaced either by 12 percent uncalc:ined diatomite, 30 percent calc:ined shale, 35 percent pumicite,
or 50 percent slag; with 45 percent fly ash replacement, there was essentially no change :in durability
factor with age, and with 35 percent natural cement the durability factor showed a sharp :increase
with age.
Figure 6 (table 4) compares data on 26 mixtures, all made with Type 11 cement, to which no
pozzolan or special cement was added; 13 at the 0.5, and 13 at the 0.8 water-cement ratio. One
mixture at each water-cement ratio is that shown :in Fig. 1, and conta:ins entra:ined air but no other
chemical admixtures; all the other mixtures conta:in other chemical admixtures, eight each conta:in
calcium lignosulfonate, calcium chloride, and triethanolam:ine.
Figure 7 shows on one graph all the data on air-entra:ined given :in Fig. 1 through 6. Two heavy lines
have been drawn :indicating an envelope with:in which essentially all the data fall. It is suggested,
therefore, that for concrete made us:ing the aggregate under study, the compressive strength
developed is related to the durability factor with:in these limits regardless of the age of the concrete
(14 to 180 days for durability factor; 28 to 180 days for strength), the nom:inal water-cement ratio
(0.5 to 0.8), the presence of chemical or m:ineral admixture; or :in essence what Klieger showed :in
1956, that if you want frost resistance concrete-for example, concrete with a durability factor over
50-you need air-entra:ined concrete which has no higher a fractional volume of freezable water
than that of concrete hav:ing matured so as to develop a compressive strength of about 4,000 psi.
One can do some :interesting arithmetic if one were to assume a value of the fractional volume of
--``,```,`,`,`,,,,`,`````,`,,`,,-`-`,,`,,`,`,,`---
freezable water, consider the 9 percent expansion of this volume, when the water is converted to ice,
and the volume of entra:ined air required to take up that much :increase :in volume of the freezable
water. If one were to assume that there is slighdy more than 10 percent freezable water :in a given
concrete, it would follow that there is 1 percent or so of :increase in volume and ideally only 1
percent by volume of entra:ined air would accommodate this if the air were used with maximum
efficiency and were nearly completely compressed. With a 2 percent air content, the air content of
each air void would only need to be compressed to one-half of its original volume and the volume
:increase would be accommodated. However, in fact, one needs 9 percent fractional volume of air :in
the mortar fraction of concrete :in order to achieve successful protection aga:inst frost action. This
was first emphasized by Klieger in his paper that became PCA Bulletin 77 (Klieger 1956A). The
conventional assumption is that this is because a lot of the air bubbles are a lot bigger than they need
to be. If they were all of the optimum distribution, the total air content would be quite a lot less and
the penalty :in strength of 5 percent per 1 percent of air could be proportionally reduced.
WHY THE RECIPE WORKS
In the preced:ing portions of this paper, I have discussed different forms of the recipe for mak:ing
concrete that will be immune to frost action as a function of the sophistication of the audience to
which the recipe is be:ing presented. There is, however, a third audience that not only wants the
recipe presented :in a form that permits the exercise of judgment through the possession of
understand:ing, but also wants technical and scientific explanations of how and why all these th:ings
work. The principal contributors to the understand:ing of how air entrainment works :in do:ing good
th:ings for the resistance of concrete to freezing and thawing have been T. C. Powers (1949, 1954),
Richard Helmuth (Powers and Helmuth 1953), and Bob Philleo (1986). Contributions by each of

94
Daneshlink.com
daneshlink.com
them are cited in the list of references. The principal contributors to the understanding of the role
played by aggregate properties include also T. C. Powers (1955) but especially George Verbeek
(y erbeek and Landgren 1960). Indeed, the subject of resistance of natural aggregate materials, sand,
gravel, and crushed stone, to freezing and thawing has a very long history going back at least to 1818
when Brard in France (DeThury 1829) developed the prototype of what is now ASTM C 88, the
standard test for soundness of aggregate by the use of sodium sulfate or magnesium sulfate. This
widely misunderstood and generally reviled procedure has been used to assess the durability of
building stone and was spoken of quite favorably by Schaffer (1932). It is not always realized that it
was always the intent of this test to simulate the expansion of water being converted to ice on
freezing by the use of the mechanism of precipitating hydrated sulfate solids in pores in the rock,
dehydrating or partially dehydrating these sulfates by oven drying and causing the internal expansion
that simulates ice formation by the rehydration of the dehydrated or partially dehydrated sulfates
when the specimen is re-immersed in the sulfate solution. ACI Committee 201 has, in the last few
years, given attention to the relationship of this mechanism and the deterioration of concrete
associated with the movement of sulfate solutions into by capillary action, the subsequent
precipitation of salt by dehydration and damage due to rehydration. In this case, the test that was to
simulate freezing and thawing also reproduces a completely unrelated phenomenon which may cause
destruction of concrete by internal expansion in regions where freezing cannot occur.
CONCLUSION
Concrete will be immune to the effects of freezing and thawing even if critically saturated with water
if it is made using sound aggregate, has a proper air-void system, and has matured so as to have
developed a compressive strength of about 4,000 psi.
REFERENCES
American Society for Testing and Materials Standards
c 33 Specification for Concrete Aggregates

c 88 Test Method for Soundness of Aggregates by Use of Sodium Sulfate or Magnesium Sulfate
c 150 Specification for Pordand Cement
c 260 Specification for Air-Entraining Admixtures for Concrete
c 457 Practice for Microscopical Determination of Air-Void Content and Parameters of the Air-
Void System in Hardened Concrete
c 595 Specification for Blended Hydraulic Cements
c 666 Test Method for Resistance of Concrete to Rapid Freezing and Thawing
c 671 Test Method for Critical Dilation of Concrete Specimens Subjected to Freezing
--``,```,`,`,`,,,,`,`````,`,,`,,-`-`,,`,,`,`,,`---
C672 Test Method for Scaling Resistance of Concrete Surfaces Exposed to Deicing Chemicals
c 682 Practice for Evaluation of Frost Resistance of Coarse Aggregates in Air-Entrained Concrete
by Critical Dilation Procedures
American Concrete Institute, Manual of Concrete Practice
ACI 201.2R-77 Guide to Durable Concrete

ACI 211.1-81 Practice for Selecting Proportions for Normal, Heavyweight, and Mass Concrete
ACI 212.1R-81 Admixtures for Concrete

95
Daneshlink.com
daneshlink.com
Bache, Hans Henrick, and I sen, Jens Christian, 1968, "Modal Determination of Concrete Resistance
to Popout Formation," Journal if the American Concrete Institute, Proceedings, V. 65, pp. 445- 450.
Buck, A.D., 1976, Investigation ofFrost Resistance ifMortar and Concrete, TR C-76-4, Waterways
Experiment Station, Vicksburg, Miss.
DeThury, Hericart,. 1829, "On the Method Proposed by Mr. Brard for the Immediate Detection of
Stones Unable to Resist the Action of Frost," Annales de Chimie et de Pf?ysique, V. 38, pp. 160-192.
Gustaferro, A. H., N. Greening, and Paul Klieger, 1966, "Expansive Concrete-Laboratory Tests of
Freeze-Thaw and Surface Scaling Resistance," Journal if the PCA Research and Development Laboratories,
V. 8 No.1, pp. 10-36 (PCA Res Dept Bull190).
Klieger, Paul, 1949, "Effect of Entrained Air on Concretes Made with So-Called 'Sand-Gravel'
Aggregates," Journal if the American Concrete Institute, Proceedings, V. 45, pp. 149-163 (PCA Res Dept
Bull24).
Klieger, Paul, 19 52, "Studies of the Effect of Entrained Air on the Strength and Durability of
Concretes Made with Various Maximum Sizes of Aggregate," Proceedings if the Highwqy Research Board,
31, pp. 177-201, (PCA Res Dept Bull40).
--``,```,`,`,`,,,,`,`````,`,,`,,-`-`,,`,,`,`,,`---
Klieger, Paul, 1955, "The Effect of Atmospheric Conditions During the Bleeding Period and Time
of Finishing on the Scale Resistance of Concrete," Journal if the American Concrete Institute,. Proceedings,
V. 52, pp. 309-326 (PCA Res Dept Bull 72).
Klieger, Paul, 1955A, "Notes on the Effect of Maximum Size of Aggregate on Tests for Frost
Resistance," Transactions Fifth Congress on Large Dams, Paris, 4, pp. 335-338, International Commission
on Large Dams of the World Power Conference (PCA Res Dept Bull 79, Part 2).
Klieger, Paul, 1956, "Curing Requirements for Scale Resistance of Concrete," Highwqy Research Board
Bulletin 150, pp. 18-31 (PCA Res Dept Bull82).
Klieger, Paul, 1956A, "Further Studies on the Effect of Entrained Air on Strength and Durability of
Concrete with Various Sizes of Aggregates," Highwqy Research Board Bulletin 128, pp. 1-19 (PCA Res
Dept Bull 77).
Klieger, Paul, 1957, "Long-Time Study of Cement Performance in Concrete-Chapter 10. Progress
Report on Strength and Elastic Properties of Concrete," Journal if the American Concrete Institute,
Proceedings, V. 54, pp. 481-504 (PCA Res Dept Bull89).
Klieger, Paul, 1957A, "Early High-Strength Concrete for Prestressing," Proceedings, World Conference on
Prestressed Concrete, San Francisco, A5-1 to A5-14 (PCA Res Dept Bull91).
Klieger, Paul, 1960, "Some Aspects of Durability and Volume Change of Concrete for Prestressing,"
Journal if the PCA Research and Development Laboratories, V. 2, No.3, pp. 2-12 (PCA Res Dept Bull118).

96
Daneshlink.com
daneshlink.com
Klieger, Paul, 1966, "Air-Entraining Admixtures," Significance ofTests and Properties of Concrete and
Concrete-Making Materials, ASTM Special Technical Publication No. 169A, pp. 530-542 (PCA Res Dept
Bull199).
Klieger, Paul, and J. A. Hanson, 1961, "Freezing and Thawing Tests of Lightweight Aggregate
Concrete," Journal of the American Concrete Institute, Proceedings, V. 57, pp. 779-796 (PCA Res Dept Bull
121).
Klieger, Paul, and Albert W. Isberner, 1967, "Laboratory Studies of Blended Cements-Pordand
Blast-Furnace Slag Cements," Journal of the PCA Research and Development Laboratories, V. 9, No. 3, pp.
2-22 (PCA Res Dept Bull218).
Klieger, Paul, and William Perenchio, 1963,. "Silicone Influence on Concrete Resistance to Freeze-
Thaw and De-Icer Damage," Highwqy Research Record, No. 18, pp. 33-47 (PCA Res Dept Bull169).
Larson. T. D., and P. D. Cady, 1969, Identijication ofFrost-Susceptible Particles in Concrete Aggregates,
National Cooperative Highway Research Program Report 66, Highway Research Board, Research
Council, Washington, D.C., 62 pp.
Mather, Bryant, 1962, Effect ofDuration ofMoist Curing on the Relative Durability of Concrete in Freeifng and
Thawing, USAEWES Misc. Paper 6-531,9 pp.
Mielenz, R. C.; V. E. Wolkodoff;J. E. Backstrom; and H. L. Flack, 1959, "'Origin, Evolution, and
Effects of the Air Void System in Concrete. Part 1-Entrained Air in Unhardened Concrete,"
Journal ofAmerican Concrete Institute, Proceedings, V. 30, No.1, pp 95-121.
Philleo, Robert E, 1986, Freeifng and Thawing Resistance of High-Strength Concrete, National Cooperative
Highway Research Program Synthesis of Highway Practice 129, 31 pp.
Powers, T. C., 1949, "The Air Requirement of Frost-Resistant Concrete," Proceedings of the Highwqy
Research Board, V. 29, Highway Research Board, National Research Council, Washington, D.C., pp
184-211.
--``,```,`,`,`,,,,`,`````,`,,`,,-`-`,,`,,`,`,,`---
Powers, T. C., 1954, 'CVoid Spacing as a Basis for Producing Air-Entrained Concrete," journal of
American Concrete Institute, Proceedings, V. 25, No.9, pp 741-760.
Powers, T. C., 1955, "Basic Considerations Pertaining to Freezing and Thawing Tests," Proceedings,
ASTM, V. 55, pp 1132-1155.
Powers, T. C., and R. A. Helmuth, 1953, "Theory of Volume Changes in Hardened Pordand
Cement Paste During Freezing," Proceedings of the Highwqy Research Board, V. 32, Highway Research
Board, National Research Council, Washington, D.C., pp 285-297.
Schaffer, R. J., 1932, The Weathering ofNatural Building Stones, Building Research Special Report No.
18, Department of Scientific and Industrial Research, London, 149 pp.
Tremper, B., and D. L. Spellman, 1961, "Tests for Freeze-Thaw Durability of Concrete Aggregates,"
Highwqy Research Board Bulletin 305, National Research Council, Washington, D.C.
97
Daneshlink.com
daneshlink.com
Verbeek, George, and Paul Klieger, 1956, "Studies of 'Salt' Scaling of Concrete," Highwqy Research
Board Bulletin 150, pp. 1-13 (PCA Res Dept Bull 83).
Verbeek, George, and Paul Klieger, 1958, "Calorimeter-Strain Apparatus for Study of Freezing and
Thawing Concrete," Highwqy Research Board Bulletin 176, pp. 9-22 (PCA Res Dept Bull 95).
Verbeek, George, and Robert Landgren, 1960, "Influence of Physical Characteristics of Aggregates
on Frost Resistance of Concrete," Proceedings, ASTM, V. 60, pp. 1063-1079 (PCA Res Dept Bull
126).
--``,```,`,`,`,,,,`,`````,`,,`,,-`-`,,`,,`,`,,`---

98
Daneshlink.com
daneshlink.com
Legend .. 1-L _,. ...JJ!

;ov
eDFE]41Strength2s \~ ~ I I /
80
oDFE 18of5trengthJso .. ~()~
:s
t .!1~ .,. ~jc:~
"'
u.. 60
~ .,I>- c- C.m'""
~ T
:0
r:
' _,, Cl~'0 56' (~ ~ 1 Low-A Ika li
:::J
0 40
20
4 ,. _
...:
~ -tl\ - ~\
"
II High-Alkali
IV
t.• ' Nonair-Entrained 1 ...

0 1000
- ---
2000
w!c 0.8
3000
- j
4000
Compressive Strenth,PSI
5000
w/c -o.s
6000 7000
~
8000
Figure 1 - Comparison of durability factors and compressive strengths of non-air-entrained and air-entrained
concretes.
Table 1-Strength, Age, Durabili~ Data for Five Cements
--``,```,`,`,`,,,,`,`````,`,,`,,-`-`,,`,,`,`,,`---
w/c =0.5 w/c =0.8
Cement
Compressive DFE Compressive DFE Compressive DFE Compressive DFE
strength, 14 strength, 180 strength, 14 strength, 180
psi, 28 days days psi, 180 days day_s psi, 28 days days psi, 180 days days
6±0.5%Air
II 4440 90 6180 92 1990 49 2770 58
I-H 3350 79 4710 88 1630 17 2070 40
I-L 4840 91 5740 92 2520 59 2830 50
III 6110 93 6610 91 2990 67 3220 64
IV 3760 67 6320 89 1420 29 2880 63
Non-air-entrained
II 5620 3 7710 5 2670 2 3860 2
I-H 4340 2 5850 3 1900 2 2400 2
I-L 6610 5 7430 9 2800 2 3620 1
III 6750 5 8120 10 3550 3 3960 3
IV 4070 2 6940 7 1840 2 3790 1

99
Daneshlink.com
daneshlink.com
.....
g w/c= 0.5
~ 60r---~--~--~~~~~~;---~----+----i
-~
...0
~ 40~--~--~~~~--~----~--~----+---~
D
Numbers indiciate percentage

of cement replaced by fly ash,
by, solid • ume
0 1000 2000 3000 4000 5000 6000 7000 8000

Compressive Strength. psi
Figure 2 - Effect of durability factors and compressive strengths of replacing Type II cement withf!y ash I.
11ty Data for Tlype IIC ement, Two M"mera lAd m1xtures
Tabl e 2- Streng1th , A'ge, Dura bT .
% w/c= 0.5 w/c= 0.8
Admixture
Cement Compressive DFE Compressive DFE Compressive DFE Compressive DFE
replaced strength, 14 strength, 180 strength, 14 strength, 180
by Vol. psi, 28 days days psi, 180 days days psi, 28 days days psi, 180 days days
Fly ash I 30 3510 66 5540 85 1180 23 2200 22
Fly ash I 45 2810 51 5000 72 1130 12 2150 14
Fly ash I 60 2060 35 3590 49 770 5 1650 3
Slag I 30 3890 83 5230 91 1840 29 2500 36
Slag I 50 3030 72 4920 83 1790 18 2500 19
Slag I 70 2370 53 3610 39 1320 3 1860 4
100
~·
0
-
80 • """'" I "..
.....
.8v .r
,~ ~ / w/C=O.S
)' 1
60
~
.~
...0
~~ ~
!' •
J
'
/ --``,```,`,`,`,,,,`,`````,`,,`,,-`-`,,`,,`,`,,`---
1:2 40
~
I
+
:J
D "
I
~ o;o lw/c=0.8
20
~~
I '
0 ~
0 1000 20(10 3000 4000 5000 6000 7000 8000
Compressive Strength, psi
Figure 3 - Effect on durability factors and compressive strengths of replacing Type II cement with slag I.

100
Daneshlink.com
daneshlink.com
100
80
....0
:tl
LL
:;....
60
.t=
.J::J
r:! 5= Slag
:::l 40
0 F =Fly Ash
N=Natural Cement
P=Pumicite
20 C = Calcined Shale
Qh---~--~~--~--~~--~--~----~--~
0 1000 2000 3000 4000 5000 6000 7000 8000
Compressive Strength psi
Figure 4 - Effect on durability factors and compressive strengths of replacing Tjpe IV cement with six materials.
Table 3-Strength, Age, Durability Data for Two Cements

Six Mineral Admixtures
% w/c==0.5 w/c == 0.8
Admixture
Cement Compressive DFE Compressive DFE Compressive DFE Compressive DFE
replaced strength, 14 strength, 180 strength, 14 strength, 180
by Vol. psi, 28 days days psi, 180 days days psi, 28 days days psi, 180 days days
Type IV Cement
Slag I 50 2870 62 4850 76 1500 6 2540 13
Nat I 35 3830 61 6580 89 1490 18 2760 43
Fly ash I 45 2140 36 4640 67 870 2 2310 6
PumF 35 2260 61 5280 83 880 13 2430 21
Cal shM 30 2890 56 5430 91 1250 21 2810 43
Unc diat 12 3840 66 6260 94 1420 29 2780 64
Type I (low-alkali Cement
Slag I 50 3810 66 5290 85 1690 15 2370 12
Nat I 35 4750 85 6180 91 1840 31 2540 55
Fly ash I 45 3080 58 4990 77 1190 9 2170 10
PumF 35 3390 72 5380 84 1300 34 2160 19
Cal shM 30 4190 76 5240 90 1960 55 2170 40
Unc diat 12 4910 89 6060 94 2340 63 2170 60
--``,```,`,`,`,,,,`,`````,`,,`,,-`-`,,`,,`,`,,`---

101
Daneshlink.com
daneshlink.com
100
ot!.'!J.
~
80 ~ rJiiN35
~w
..... w/c= 0.5
.8
lO .!:!..;_)
u... 60
~
!""
.q c.•1::;>
.J:J
~
:::J 40
0
20
~:s
w/c=0.8
~
._.F45
0
0 1000 2000 3000 4000 5000 6000 7000 8000
Compressive Strength.. psi
Figure 5 - Effect on durabili!J factors and compressive strengths rif replacing Ijpe I (/ow-alkali) cement with six materials.
60~~~~~~~~·
~(CLS OOOll
Q.= Casting and Curing
II Tempe!Qture
~(CaC12 (73))
40~~~--~~--~
.!. (Gr£:12 (40))
l,(CL5(7J)) (ClS (lOOlJ.Q.
wic =o.a Lega1d
20 1-----11-----.-
=
ClS Caldum lignosu tfonate
CaCI2 =Calcium Chloride
Tri=lh =Triethanolamine
OL---L-~L-~--~~~==~~~~~
0 1000 2000 3 000 4000 5 000 6000 7000 8000

Figure 6 - Durabili!J factors and compressive strengths made with Ijpe II cement with and without chemical admixtures.
--``,```,`,`,`,,,,`,`````,`,,`,,-`-`,,`,,`,`,,`---

102
Daneshlink.com
daneshlink.com
Table 4-Strength, Age, Durability Data for Type II Cement

Three Chemical Admixtures
w/c =0.5 w/c =0.8
Admixture
Compressive DFE Compressive DFE Compressive DFE Compressive DFE
strength, psi 14 strength, psi 180 strength, psi 14 strength, psi 180
--``,```,`,`,`,,,,`,`````,`,,`,,-`-`,,`,,`,`,,`---
28 days days 180 days days 28 days days 180 days days
None 4440 90 6180 92 1990 49 2770 58
CLS 73* 4860 9 6700 92 1930 36 2690 50
CLS 73"* 5310 92 6140 89 2230 59 2910 52
CLS 100* 4940 91 6650 91 2010 51 2720 72
CLS 100** 5290 88 5910 79 2700 54 I 3010 33
CaCh 73* 4810 89 6420 91 2080 48 2900 60
CaClz 73** 5300 91 6110 88 2350 55 3120 44
CaCh40* 5340 91 7180 90 2200 43 3060 73
CaCb 40** 4790 97 6730 89 1900 59 3290 75
Trieth 73* 5110 89 6190 91 2530 48 3190 61
Trieth 73** 4760 90 5610 83 2640 50 3220 45
Trieth 40* 5300 90 6580 89 2400 54 3250 72
Trieth 40** 4670 98 6510 89 2000 68 3590 81
* Castmg temperature
** Casting-and-curing temperature
100
TypeiiWIIhTrie.th
80 Cast ard Cured
at40%
...
9u
60
.2.
.~ X DfE 1iStreng th2B
:.0 ,_.,._-1---+-tO DFEu11:fStrength 1110
tl1 40
:::1
0 --Concrete With .Admixtull!
- ConueteWithout
20 Admi>:ture
0 1000 2000 3000 4000 5000 6000 7000 8000

Figure 7 - Summary o/data for air-entrained concretes from Figs. 1 - 6.

103
Daneshlink.com
daneshlink.com
--``,```,`,`,`,,,,`,`````,`,,`,,-`-`,,`,,`,`,,`---

Daneshlink.com
daneshlink.com
Reprinted from Concrete International, V. 23, No. 5, 2001, Farmington Hills, MI, pp. 35-36, ©American Concrete Institute.
CRYSTAL GROWTH IN ENTRAINED-AIR VOIDS
by Bryant Mather
Crystals are found in some entrained-air voids of all concrete that has been "left out in the rain" or
stored while damp. These crystals can be observed through petrographic examination of thin or
ground concrete sections at normal magnification. If the concrete was made using a cement with a
high tricalcium aluminate (C A) content (over 8%), it is likely that areas in it will be covered with
3
ettringite crystals - secondary ettringite.
Air entrainment is most often used for frost resistance. For successful performance, the active
entrained-air voids should be empty, or nearly so, of pore fluid when the fluid in a capillary cavity in
the vicinity of the void is cooled to its freezing temperature. Such freezing causes hydraulic pressure
on the unfrozen part of the fluid because the other part of the fluid has frozen, which is now a more
concentrated solution with a lower freezing point. The fluid under hydraulic pressure now has time
to diffuse through the paste for no more than 0.2 mm (8 mils) and compress the air in the nearby
entrained-air void and relieve the stress in the system. Upon thawing, the pore fluid is forced by the
compression of the air in the air void back to the capillary cavity from which it came. This must be
the case because air entrainment works, and it could not do so unless the functioning air voids are
empty of fluid or nearly so when the next freezing takes place -which in many laboratory freezing-
and-thawing systems is one hour after thawing.
--``,```,`,`,`,,,,`,`````,`,,`,,-`-`,,`,,`,`,,`---
For any significant amount of crystal growth to take place in an entrained-air void, pore fluid must
remain in the void when the concrete is thawed or get in when no freezing and thawing is involved
(that is, in the summer). Such fluid cannot be present in a functioning uncompromised air void
when the concrete is thawed. Hence, it is only possible for crystals to grow in any quantity in a non-
functioning, compromised air void, presumably one intersected by a microcrack or other damage.
Even if there were mechanisms by which uncompromised air voids could contain substantial
amounts of pore fluid that could yield crystals precipitated in the void, the probability is high that
such a void could still function in freezing and thawing because the crystals that form are, for the
most part, acircular. When examined at low magnification, the void appears "full," but it will have
nearly half its volume still consisting of air, like a box of steel fibers packed as tight as practical
having a mass of less than half that of a solid block of steel of the same size.
In the absence of other damage to the concrete, destroying the effectiveness of a substantial number
of air voids is unlikely to seriously compromise the concrete's frost resistance because the quantity
of freezable water in the concrete is not likely to exceed 10% by volume of the concrete. Properly
air-entrained concrete has about 9% air in the mortar fraction; water freezing to ice expands 9%.
Therefore, there is about 10 times more air volume in the concrete than needed to accommodate the
volume increase of the freezing of the freezable water. Hence, even if one deducts the larger voids
that contribute to the air content of the concrete as measured, there still is a lot more air than
needed. If some of the entrained-air voids became compromised, it likely would not destroy the
frost resistance of the concrete. With regard to the mechanisms by which an entrained-air void may

105
Daneshlink.com
daneshlink.com
be compromised so fluid remains in it when the concrete is thawed, one might consider some of the
implications of the high-pressure test for air content of hardened concrete. If one took an intact 152
x 305 mm (6 x 12 in.) cylinder with minimal damage and tried to perform that test, it didn't work.
After 30 days, the reading was likely to have a small percent of air content when the concrete had
about 7% air. Then the test procedure was revised. The concrete cylinder was oven dried (at
moderately warm temperatures) for 24 hours or so; then the water entered the concrete all the way
to the center and a valid value for air content was obtained in 30 minutes. This suggests that if a
drastic change is produced by what was regarded by the proponents of the test as mild treatment, it
seems likely that even a milder thermal or moisture content change could produce some degree of
microfracturing. This microfracturing can be sufficient for some entrained-air voids to be
compromised by being transected by a microcrack and hence, be capable of retaining (and gaining)
pore fluid when the concrete is thawed. Ultimately, the concrete can have ettringite-"filled"
entrained-air voids, even if there is little or no evidence that the concrete was mishandled at any
point in its history.
Further reading
1. Bache, H. H.; Idom, G. M.; Neppes-Christensen, P.; and Nielsen,]., 1965, "Morphology of
Calcium Hydroxide in Cement Paste," Symposium on Structure of Portland Cement Paste and
Concrete, Special Report 90, Highway Research Board, NAS-HRC, Washington, D.C., pp. 154-
174.
2. Bruere, G. M., 1961, "The Relative Importance ofVarious Physical and Chemical Factors in
Bubble Characteristics in Cement Pastes," Australian Journal ofApplied Science, V. 12, No. 1,
pp. 78-86.
3. Helmuth, R. A., 1961, "Dimensional Changes in Hardened Portland Cement Pastes Caused
by Temperature Changes," Proceedings, Highway Research Board, NAS-HRC, V. 40, pp. 315-
336.
4. Idorn, G. M., 1959, "Microscopiske glimt afbetonteknikkens historie," Beton-Teknik, No.4,
pp. 119.
5. Zelie, J.; Kristulovic, R.; Tkalcec, E.; and Krolo, P., 1999, "Durability of the Hydrated
Limestone-Silica Fume Mortars Under Sulfate Attack," Cement and Concrete Research, V. 20, pp.
819-826.
6. Lawrence, C. D., 1990, "Sulfate Attack on Concrete," Magaifne of Concrete Research, V. 42, pp.
249-264.
7. Mather, B., 2000, "Discussion of Zelie et al.," (1999), Cement and Concrete Research, V. 30, p.
331.
8. Moser, B., and Stark,J., 1999, "ESEM-FEG: A New Scanning Electron Microscope for
Building Materials Research," ZKG International, V. 4, pp. 212-221.
9. Stark,J., and Bollmann, K., 1998, "Ettringite-Cause of Damage, Damage Intensifier, or
Uninvolved Third Party," ZKG International, V. 5, pp. 280-291.
--``,```,`,`,`,,,,`,`````,`,,`,,-`-`,,`,,`,`,,`---

106
Daneshlink.com
daneshlink.com
Reprinted, with permission, from Proceedings, Highwqy Research Board, V. 31, ©Transportation Research Board, National
Research Council, Washington, D.C., 1952.
CRACKING OF CONCRETE IN THE TUSCALOOSA LOCK
Bryant Mather1
SYNOPSIS
The Tuscaloosa Lock was constructed on the Warrior River near Tuscaloosa, Alabama, between
1937 and 1939. It was noted in 1947 that cracks had developed in the lock walls. A board of
consultants was appointed, and the members of the board examined the structure and
recommended that the Concrete Research Division, Waterways Experiment Station, conduct a study
to determine the cause of the cracking. The data developed by this study are summarized and
discussed in this paper. It was concluded that the cracking resulted from a chemical reaction
between the alkalies in the cement and unstable silica in the aggregate. It is believed that this is the
flrst published account of cracking of concrete through cement-aggregate reaction in which
chalcedonic chart is the major control responsible for the deleterious reaction.
Between October 193 7 and September 1939 the Mobile District of the Corps of Engineers
supervised the placement of concrete in the Tuscaloosa Lock and Dam on the Warrior River near
Tuscaloosa, Alabama. The cement came from two mills, the aggregates were natural sand and gravel,
and the mixing water came from the Tuscaloosa city water supply.
Eight years after the placement of concrete was completed, it was noted that cracks had developed
in the lock walls. A board of consultants was appointed to examine the situation and make
recommendations. The board met at Tuscaloosa in November 1947, reviewed the available data,
examined the structure, and arrived at two conclusions: (1) cracking is more advanced in concrete
made with Cement A than in that made with Cement B and (2) additional data on the condition of
the structure should be obtained by drilling cores and by making a detailed crack survey.
A series of cores were drilled including a 36-in.-diameter core from monolith No.5 in the lock wall,
a monolith that showed rather extensive cracking it the surface and which was so situated that
internal movement might interfere with the proper functioning of the lock gates. The board met
again in January 1948, examined the additional data, the cores, especially the 36-in. core hole, and
concluded that: (1) the cracking in monolith No.5 is not as serious as it had appeared from the
surface and (2) an investigation should be made by the Waterways Experiment Station to determine
the causes of the cracking.
LABORATORY INVESTIGATION
--``,```,`,`,`,,,,`,`````,`,,`,,-`-`,,`,,`,`,,`---
It was suspected at the beginning of the laboratory investigation that the cracking in this structure
could have been caused by expansion resulting from chemical reaction between the aggregate and
the alkalies in the cement. This hypothesis was based on the following considerations:
1
Engineer, Concrete Research Division, Waterways experiment Station, Corps of Engineers, U.S. Army,Jackson,
Mississippi

107
Daneshlink.com
daneshlink.com
(1) Several engineers who examined the structure called attention to a similarity in pattern between
the cracking manifested here and that in certain structures in other parts of the country where
the alkali-aggregate reaction is regarded as the cause of the cracking.
(2) There were fairly clear-cut evidences of internal expansion of the concrete.
(3) The cores and the 36-in. core hole appeared to contain deposits of secondary material,
superficially similar to the reaction products of the alkali-aggregate reaction.
(4) The natural gravels of the Atlantic and Gulf Coastal Plains were known to contain chalcedonic
chert, an aggregate constituent that had been shown to be potentially capable of deleterious
chemical reactivity with the alkalies in portland cement.
(5) Certain portland cements produced in the area in which Cement A was made were reported to
--``,```,`,`,`,,,,`,`````,`,,`,,-`-`,,`,,`,`,,`---
have been typically of rather high alkali content during the period when cement from this area
was being furnished for use in the project.
In opposition to the immediate adoption of this hypothesis as the sole explanation of the cracking
were the following considerations:
(1) It was possible that the cracking was at least in part due to other causes, such as leaching of lime
from the concrete by acidic river water or expansion due to reaction between sulfates in the river
water and aluminates in the cement.
(2) At the time the study was started, no defmite occurrence of concrete cracking due to alkali-
aggregate reaction had been described from the South Atlantic or Gulf Coastal region in spite of
very widespread use of cements and aggregates generally similar to those used in the Tuscaloosa
project.
(3) No instance of concrete cracking due to alkali-aggregate reaction had been described from any
locality in which chalcedonic chert was the major aggregate constituent involved.
The laboratory investigation was therefore planned to develop as much detailed data as possible that
could conf1m1 or disprove the hypothesis that cracking was solely or primarily a result of alkali-
aggregate reaction. Two sections of the 36-in. core and sections of 4 %-in. core from three other
monoliths were sent to the laboratory for study. The 36-in. core sections were 2ft. and 5 ft. in
length and a total of 91 ft. of 4 %-in. core was available for study. The 36-in. cores and a majority of
the smaller cones came from concrete made with Cement A. An initial examination of the cores
revealed that all of them showed a development of calcium sulfoaluminate and of gel both as
exudations on the drilled surfaces and as fillings or partial fillings in cracks and voids; and most of
the cores showed cracks visible to the naked eye.
CONDITION OF THE CORES
All of the drilled surfaces of the cores showed white films or mounds of gel on areas of mortar and
on particles of chert coarse aggregate but not on coarse aggregate particles of other types. Frequently
the gel entirely inclosed sawduse particles; mounds of gel rose up as much 1/8 in. above the core
surface (Fig.1). These facts make it plain that some of the gel deposition took place after the cores
were drilled. All of the drilled surfaces cut air pockets and underside voids containing white or clear
exudates with outer surfaces continuous with those of the adjacent paste and mortar and,
accordingly, older than the drilling of the core. A number of the drilled surfaces show cracks visible
to the naked eye. A few show cracks filled or partly filled with exudate (Fig. 2). The smoother
2
The core specimens were packed in sawdust when received at the laboratory.

108
Daneshlink.com
daneshlink.com
surfaces produced by sawing with a diamond-edged blade reveal more cracks than appear on the
outside of the core. When the concrete was broken open and the surfaces of the largest cracks
examined, the surfaces were found to be covered with desiccated gel, which filled the crack and
saturated the adjoining mortar. Some of the drilled surfaces intersected empty pebble sockets where
poorly bonded aggregate was lost during drilling. Some of the surfaces intersected coarse aggregate
--``,```,`,`,`,,,,`,`````,`,,`,,-`-`,,`,,`,`,,`---
particles containing cracks visible to the naked eye. Most of the cracked particles were chert; a few
were quartzite or sandstone. Many of the cores were fairly dense and free of large voids on the
drilled and sawed surfaces. In those cores the concrete near the outside of the structure appeared
generally more dense than the concrete in the interior. Other cores were noticeably less dense, and
in certain cases were found to contain nUillerous large irregular voids and, in one case, showed
conspicuous honeycomb (Figs. 3 and 4). In all of the concrete, macroscopic voids were more
common near coarse aggregates particles than away from them. Many of the pebbles had relatively
narrow, but extensive, underside voids.
The concrete containing Cement A showed more signs of reaction than that containing Cement B,
but the concrete containing Cement B contained cracks in the upper section and gel throughout the
core.
EXUDATES AND DEPOSITS IN THE CONCRETE
Examination of drilled, sawed, and broken surfaces of the concrete showed that several types of
deposits were present in voids and cracks and on aggregate particles.
Gel-Gel was the most abundant and conspicuous deposit in and on the concrete. Most of the
growth of gel took place before the cores were drilled. The late gel usually looked as if it had been a
less viscous material than the gel formed in the structure. It formed flat sheets and low mounds, was
usually quite translucent and bluish white, and had developed shrinkage cracks. It is similar in
general appearance to the majority of the gel found on expanded mortar bars made for the alkali-
reaction test when examined soon after their final removal from storage over water. As the gel dried,
it became white and cracked more. The three samples of late gel examined in immersion media were
all isotropic with indices of refraction in the range between 1.455 and 1.470. Two varieties of gel,
older than the drilling of the cores, formed partial and complete fillings of voids and cracks in
mortar, and pores and cracks in coarse aggregate. The less abundant variety was clear to translucent,
rubbery to brittle, and usually formed the outer shell of the lining if both varieties were present in
one void. The more abundant kind was dull opaque white, rubbery to powdery to brittle, and usually
formed the core if both varieties were present. Sometimes concentric shells of the opaque kind
differed in color and appearance in reflected light, and in structure and index of refraction in
transmitted light. Occasionally, radial or concentric shrinkage cracks in a filling had been filled by
later gel of higher index. Two examples were found where translucent gel formed the interior and
opaque gel the periphery of the filling. Many voids were entirely filled with gel, or evidently were
once entirely filled before shrinkage of the filling. However, most of the void space in the concrete
was still empty, even in areas close to gel-filled cracks. Many voids contained no gel, or only a thin
lining of clear brittle gel. Gel was found in all sections of all the cores, regardless of type of cement
or distance from the outer surface of the structure. The most abundant gel, and the greatest amount
of gel-permeated and gel-whitened paste, was found near cracks in the upper or outer sections. In
the longest core, from monolith No.5, gel and cracking decreased from top to bottom in the core,
but gel was present in the bottom section.

109
Daneshlink.com
daneshlink.com
Forty samples of gel, older than the drilling of the cores, and taken from cores representing four
monoliths, were examined using the petrographic microscope and their indices of refraction
determined. Only 2 of the 40 samples were entirely isotropic. There were two varieties of anisotropic
gel. The most common was very finely grained under crossed nicols, and had a low aggregate
"pepper and salt" birefringence resembling chert. Some of it was isotropic on thin edges. The range
observed in indices of refraction was from 1.478 to 1.511. It is believed that this variety contained
small birefringent crystalline inclusions, but no distinctive shapes or orientations could be made out.
Calcium hydroxide inclusions of the size of the birefringent units would have the birefringence
found. Some of the gel with pepper-and-salt inclusions also contained irregular or rhombic
inclusions with indices of refraction and birefringence agreeing with calcium carbonate. Calcium
carbonate was the only inclusion in some of the gel samples.
The second kind of anisotropic gel showed aggregate polarization, low birefringence, and overall
wavy extinction with a misshapen black cross. The indices of refraction ranged from 1.480 to 1.502.
This variety was regarded as anisotropic due to strain, the strain probably resulting from drying
shrinkage in the gel but possibly due to volume changes in the concrete which originally surrounded
it. Indices of refraction of all43 samples are given below:
Index of Refraction Number of Samples
Sodium light between

Pre-drilling Post-drilling
limits shown
Above 1.520 0
1.515-1.520 1
1.510-1.515 0
1.505-1.510 5
1.500-1.505 1
1.495-1.500 2
1.490-1.495 1
1.485-1.490 6 (one isotropic)
1.480-1.485 6
1.475-1.480 12 (one isotropic)
1.470-1.475 2
1.465-1.470 2
1.460-1.465 1
1.455-1.460 1 {l}lsotropk
Below 1.455 0
40 3
Parsons and Insley (21) reported a range of indices of refraction of from 1.47 to 1.52 in gel from a
specimen where the aggregate was opal. McConnell, Mielenz, Holland, and Greene (17) reported a
--``,```,`,`,`,,,,`,`````,`,,`,,-`-`,,`,,`,`,,`---
range from about 1.46 to about 1.49. These values all seem reasonable considering variations in
cement, aggregate, and water content at time of examination. Not all of the observations agree with
the literature, at least one instance was found where clear isotropic gel was higher in index than
opaque anistropic gel from the same pocket. It is believed that all the observations can be reconciled
if it is assumed that differences in cement, aggregate, mix design, placement, exposure, and timing of
the access of unwanted water influenced the amount and composition of reaction products in

110
Daneshlink.com
daneshlink.com
different field concretes or test specimens which underwent alkali reaction. Two samples of the gel
were subjected to chemical analysis. Sample 1 consisted of material scraped from the wall of the 36-
in. core hole in Monolith 5 on 16-19 June 1948. Sample 2 was collected from pockets in 4%-in.
cores from Monoliths 20 and 60 after the concrete had dried in laboratory air. The results of these
analyses are as follows:
Sample 1 Sample 2
Percent Percent
Moisture loss at 1OS C 34.6 9.81
Composition calculated on dry basis:

Si02 49.82 61.73
CaO 21.11 12.28
Ab03 1.15 2.17
Fe203 0.81 0.47
so3 0.16 0.00
Insoluble residue 5.58 8.1
Loss on ignition - 14.84
Calculated calcium sulfoaluminate 0.47 0.0

McConnell, Mielenz, Holland, and Greene (17) reported that the S102 content of samples of gel
ranges from SO to 85 percent (moisture-free basis).
While cracks and gel were most developed in the upper and outer sections of the core, all sections of
all cores contained gel, and gel grew on the core surfaces after they were drilled and stored in a
damp condition. The late gel grew more abundantly on the middle and lower sections of the cores.
This fact suggests that while growth in the structure had only gone far enough to crack the concrete
near the surface, all the necessary ingredients for growth were present in the deeper concrete and
that the production of gel in this concrete was accelerated when additional moisture was provided.
As an example of the speed with which reaction can develop under favorable conditions, a thin-
section blank from a core from monolith 5 was impregnated with resin, ground smooth on one-face,
and then exposed to warm, humid laboratory air for approximately 36 hr. When the blank was
examined, it was found that three gel pockets and one chalcedony particle on the ground surface had
--``,```,`,`,`,,,,`,`````,`,,`,,-`-`,,`,,`,`,,`---
taken up water from the air and surface had taken up water from the air and swelled up perceptibly
above the general level of the ground surface. Figure 5 is a photograph of the specimen at a
' magnification of about 3X, taken about August 23, 1949. On August 24, 1949, the blank was stored
in a closed container over water, and left until October 17, 1949. It was then reexamined and the
photograph reproduced as Figure 6 was made at the same magnification.
Certain comments on the rapid and obvious development of the reaction products in this specimen
suggest themselves. In the first place, the specimen was heated well above 100 F (probably to at least
200 F) in the process of impregnating it with resin. During that heating it was subjected to a more
abrupt thermal change than it had ever undergone before. The cement paste lost some water which
it would not have lost in laboratory storage at ambient temperatures. Probably the thermal shock
and partial dehydration suddenly opened up the general structure and, particularly, the mortar-coarse
aggregate boundaries to a degree unusual in concrete undergoing mild weathering. Thus the
specimen may have been rendered unusually susceptible to a later addition of moisture. In the
second place, while the resin apparently did not thoroughly penetrate the interior, it formed a
coating around all the surfaces, including that from which it was later removed by grinding. All the

111
Daneshlink.com
daneshlink.com
manifestations of activity were probably therefore concentrated on the one ground surface. This
specimen is considered in detail because other pieces of Tuscaloosa concrete have been stored in the
same closed container since July 1948 without developing any visible changes, except fairly minor
increases in gel deposits. Therefore it is not suggested that deterioration of the concrete in
Tuscaloosa lock may be expected to proceed at the rapid rate suggested by the two photographs and
dates. The photographs do tend to confirm the belief that the reactive potential in the concrete has
not been exhausted.
Calcium Carbonate- Carbonation of the cement paste adjoining cracks was found in thin sections of
concrete from depths up to 20 ft. from the exterior. A small amount of carbonation was found in
one thin section from a core from a depth of about 39 ft. in Monolith 5. It is believed that this
carbonation took place after the core was drilled. The amount and extent of carbonation did not
appear to be unusual, and are not regarded as significant, except as indicating that concrete as deep
as 20 ft. in the structure was accessible to air.
Calcium Su!foaluminate- Calcium sulfoaluminate was found in voids in every section of every core,
regardless of distance from the surface of the structure. It was most abundant in each core near the
outer or upper surface of the structure. In the top section of core from Monolith 3, which was
represented by a series of chunks and fragments, sulfoaluminate was very abundant in voids and in
pebble sockets. In the cracked outer section of core from Monolith 60, about 8 in. from the
surfaces, mats, rosettes, and spherulites of sulfoaluminate filled voids, lined pebblesockets, and
coated sand grains. With increasing depths, the sulfoaluminate was found in tufts, rosettes, and
linings in the voids but not as linings of pebble sockets.
Calcium Hydroxide- Core from depths 38.4 to 40.2 ft in Monolith 5 contained well-developed tufts
and rosettes of sulfoaluminate in voids. In some of the voids, clear colorless plates of calcium
hydroxide were associated with sulfoaluminate. Calcium hydroxide is less abundant than the
sulfoaluminate but is not uncommon. It was the first example of calcium hydroxide in crystals in
voids found in field concrete examined by this laboratory. Subsequent observations indicated that
the growth of calcium hydroxide crystals in voids is a fairly common occurrence in the interior of
structures. The calcium hydroxide in voids was found only in the lower section of the core, where
the alkali-aggregate reaction was least advanced. Since accessible calcium hydroxide is fairly easily
converted to calcium carbonate by moderately dry air containing carbon dioxide (13), the persistence
of the unaltered crystals indicates that this concrete never dried out enough to allow the crystals to
be exposed to relatively dry air containing C02.
AGGREGATE IN CONCRETE FROM STRUCTURE

--``,```,`,`,`,,,,`,`````,`,,`,,-`-`,,`,,`,`,,`---
Identification of over 5,000 coarse-aggregate particles intersected by sawed surfaces (fable 1)

indicated that 42 percent of the coarse aggregate was chert and 58 percent consisted of other types
of material, principally vein quartz, quartzite, and sandstone, with very small quantities of granitic
gneiss and ochre or limonite. Sixty-eight chert particles which were associated with gel pockets were
examined in immersion media; 48 of these particles (71 percent of those examined) contained or
consisted of chalcedony. The index of refraction of 40 of the chalcedony particles was determined
(Fig. 7); the average of the 40 indices is 1.5365 with an observed range from 1.5240 to 1.5420. Since
the chert particles were deliberately selected from those which had adjoining gel pockets, it is
believed that the calculated percentage of chalcedonic chert, 71, is a maximum. If it is assumed that
112
Daneshlink.com
daneshlink.com
from 50 to 70 percent of the chert was chalcedonic, the calculated chalcedony content of the coarse
aggregate ranges from 19 to 30 percent. Using the determined average value of the index of
refraction of the chalcedony and curves given by Donnay (7), the opal content of the chalcedony can
be estimated as between 5.5 and 17.5 percent and, hence, the opal content of the total coarse
aggregate as between 1 and 5 percent.
The silica solubility and reduction in alkalinity were determined on a composite sample taken from
=
particles determined as chalcedony. Duplicate tests gave 5, 477 and R, 46 (5/Rc, 10.4). = =
The fine aggregate was natural sand composed principally of quartz, with some feldspar and a small
amount of other minerals, and chert which was particularly conspicuous in the larger sizes.
According to information provided by the Mobile District, Corps of Engineers, difficulty was
encountered in producing sand with the required amount of fines. The amount of fine sand in the
mortar was relatively small, and consequendy, there are larger areas of cement paste in the mortar
than there usually are in the mortar of concrete made with sand graded in accordance with the
current Corps of Engineers specifications.
Relation of Chert and Gel Pockets - The gel which grew on the cores after they were stored in sawdust
was located on chert particles or on mortar and not on coarse aggregate particles of other types. The
gel intersected by the sawed surfaces adjoined coarse aggregate of all the types present or occupied
cracks or voids in the mortar. Although particles of all the types of coarse aggregate could be found
with adjacent gel, gel within coarse aggregate was confined almost entirely to chert particles. The
exceptions were some highly fractured quartzite pebbles. The association of gel with chert appeared
to be more common than the association of gel with any of the other types of coarse aggregate
(Figs. 8 and 9). To test this indication, macroscopic counts were made of sawed surfaces to
determine the number of chert particles with and without associated gel, the number of other coarse
aggregate particles with and without associated gel, and the number of gel pockets or linings with no
visible association with a coarse aggregate particle (fable 2). Counts were made on 57 sawed
surfaces; 580 gel pockets were found, of which 110 were apparently isolated, 281 adjoined chert
particles, and 189 adjoined coarse aggregate particles value of other types; 5,518 coarse aggregate
particles were found, of which 42 percent were chert and 58 percent were not chert. Gel pockets
adjoined 12 percent of the chert particles found and only 6 percent of the non-chert particles. These
figures leave out of account the possible connections at depth of the apparendy isolated pockets.
--``,```,`,`,`,,,,`,`````,`,,`,,-`-`,,`,,`,`,,`---
The data in Table 2 have been used to calculate whether the observed distribution of gel pockets
with respect to chalcedony and other types of aggregate is a random one or whether the distribution
indicates an association between gel and chert greater than that likely to arise by chance (Fisher (8)
pp. 85-99). These calculations were made for the totals of all the surfaces counted. The probability
that the association found would occur by chance is less than 1 in 1,000. A study of the detailed data
from which Table 2 was prepared also indicated that the percentage of gel pockets in the core
containing Cement B, was lower than in any of the cores containing Cement A.
Chalcedonic chert was the only constituent found in the coarse or flne aggregate known to be
capable of deleterious reaction with the minor alkalies of portland cement. Such chert was a much
more important constituent of the coarse aggregate than of the fine aggregate.
A sample of gravel taken in 1948 from a commercial pit in the area from which the aggregate used in
the structure was reported to have been obtained was examined in the sizes from 1 in. to No.4 and
had more than 50 percent quartz in all sizes smaller than 1 in. and 46 percent quartz in the fraction

113
Daneshlink.com
daneshlink.com
retained on the 1-in. sieve. Chert was present as follows: 39 percent on the 1-in., 31 percent on%-
in., 35 percent on Vz-in., 22 percent on 3/8-in., and 36 percent on the No.4 sieves. The remainder
of the sample consisted of quartzite, sandstone, and granite. Chalcedony was found in one or two of
the chert particles; most of the particles which were powdered and examined were found to contain
no material with an index of refraction lower than that of quartz. Physical tests on sand and gravel
samples from this deposit are summarized below:
Bulk specific gravity, saturated surface dry: sand = 2.65, gravel = 2.62
Absorption, percent: sand = 0.4, gravel = 0. 7
Percentage of particles lighter than 2.40 after 5 hr. boiling: gravel = 2.1
Loss after 5 cycles of magnesium sulfate soundness test, percent: sand = 4. 7, gravel = 4.0
Linear coefficient of thermal expansion of sand mortar x 106 per deg. F. = 7.4
Durability factor of concrete after 300 cycles of accelerated freezing and thawing 46 =
It is therefore suggested that the coarse aggregate used in the construction of the Tuscaloosa Lock
contained a materially greater proportion of chalcedonic chert than the sample examined in 1948.
The available information on these aggregates, as set forth above, would suggest that by all usual
criteria of evaluation they were of as high or higher quality than the majority of the aggregates
generally accepted for use and used without serious trouble throughout the coastal portions of the
southern United States. Indeed the quite disturbing prospect is raised by the behavior of the
Tuscaloosa concrete that it might possibly have experienced less trouble if the aggregate had been of
poorer quality, either more porous or with a larger proportion of reactive constituents.
RIVER WATER
The presence of calcium sulfoalurninate as an abundant secondary material in the concrete and the
nature of the deterioration in general suggested that it was desirable to obtain information on the
river water to help to confirm or rule out the probability of sulfate attack on the concrete as a major
or primary cause of deterioration. Results of chemical tests on a sample of water &om the Warrior
River, taken in 1948 at Lock 15 upstream from Tuscaloosa, as reported by the South Atlantic
Division Materials Testing and Laboratory, Corps of Engineers, are as follows:
pH .................................... 6.2
ppm.
Total solids ........................... 35
Calcium as Ca ........................ 0.4
Magnesium as Mg ................... 0.5
Iron as Fe ............................ 0.3
Silica as Si02 •..•••.•••••••••••...•.•• 9
Sulfates as S04 •.•••••••..•••••••••••• 0
Free carbon dioxide as C02 ••.•..• .4
Total hardness as CaC0 3 .•••••••••• 3
Carbonate hardness as CaC03" ..... 0
Stewart J. Lloyd, assistant state geologist, Geological Survey of Alabama, states that an analysis of a
--``,```,`,`,`,,,,`,`````,`,,`,,-`-`,,`,,`,`,,`---
sample of water from the Warrior River above Tuscaloosa showed 10 parts per million of the S0 4
radical. This is not regarded as an excessive amount, since it is reported that the city water supply of
Birmingham, Alabama, on analysis in 1932, showed 61 ppm ofS0 4•
114
Daneshlink.com
daneshlink.com
The cement used in the Tuscaloosa concrete was required to contain not more than 8 percent of
calculated tricalcium aluminate. Cement A was found to have from 6.33 to 7.42 percent and averaged
6.89 percent. Cement B had from 6.50 to 6.96 percent and averaged 6.70 percent.
AcidifY ofRiver Water- It has been reported by Steele (23) that the average CaC03 content of the
water in the Warrior River at the Tuscaloosa Lock is 24 ppm. and the average pH is 5.9. These data
were discussed as follows (23):
It would not be surprising if a reader of this article should point to Tuscaloosa Lock and Dam in
Alabama and say that the average pH value noted for that dam ... indicated the possibility of
corrosive attacks ... Possibly there may be a measure of truth in that statement since the concrete in
Tuscaloosa Lock and Dam, after 12 years, shows rather serious deterioration. This deterioration,
however, is primarily due to the so-called alkali-aggregate reaction ... It is possible that slightly acid
waters entering the cracks and crevices in this structure have bad a limited effect, but it appears
doubtful whether the effect of acid waters would have been noticeable had not the expansive growth
and extensive cracking produced by alkali reaction made this possible -in other words, there might
have been some effect on the surface produced by the mildly acid waters, but it would have been
confined to a thin surface layer of concrete.
These data suggest than any sulfate attack or acid leaching effects on the Tuscaloosa concrete are of
very minor and probably insignificant importance.
CHERT AS A CHEMICALLY DELETERIOUS CONSTITUENT
Descriptions of chert and chalcedony have been published by the American Society for Testing
Materials (2, 3). Chalcedony is described as follows:
Chalcedony has been considered both as a distinct mineral and as a variety of quartz. It is now
generally believed to be composed of a submicroscopic mixture of fibrous quartz with a smaller but
variable amount of opal. The properties of chalcedony are intermediate between those of opal and
quartz, from which it can be distinguished only by laboratory tests. It frequently occurs as a
constituent of the rock chert and is reactive with the alkalies in portland cement.
Chert is described as: a fine-grained rock which is characterized by hardness, conchoidal fracture in
dense varieties, the fracture becoming splintery in porous varieties, and a variety of colors ... Chert is
composed of silica in the form of chalcedony cryptocrystalline quartz, or opal, or combinations of
--``,```,`,`,`,,,,`,`````,`,,`,,-`-`,,`,,`,`,,`---
any of these three. The determination of which form or forms of silica are present requires careful
determination of optical properties, absolute specific gravity, or both.
The writer (14, 15, 16), following Donnay (7) and others, assumes that chalcedony consists of
microfibrous quartz with interstitial opal, that the observed chemical reactivity between certain
cherts and the alkalies in portland cement is actually a reaction involving the opal portion of the
chalcedony component in the chert, and that in the absence of opal as such or of chalcedony no
reaction should be expected. Attention, however was called (15) to the indications obtained by
Alderman, Gaskin, Jones, and Vivian (1) that cryptocrystalline quartz itself may be reactive. Studies
conducted at Pennsylvania State College have been reported recently (9, 10) in which chalcedony has
been found to have features suggesting that its properties and those of chalcedonic chert "are

115
Daneshlink.com
daneshlink.com
adequately explained by the hypothesis that it consists of fine-grained quartz plus a variable quantity
of free water in the fonn of distinct bubbles; there is no evidence of admixed opal" (10).
About all that can be said at this time there fore is that fonns of silica consisting of or containing
opal, fonns having the properties of chalcedony, and possibly certain fonns having the properties of
microcrystalline quartz are capable of reacting chemically with the alkalies in pordand cement. The
study reported in this paper serves to document a case in which it is apparendy those aggregate
particles consisting of or containing silica having the properties of chalcedony that participated in a
chemical reaction with the alkalies from the cement with the result that the concrete suffered
deterioration. The same caution which was previously suggested (15) is believed still necessary in
extrapolating from these data to other concrete containing similar constituents: "The possibility that
the expected reaction will not occur or that it may be forestalled ... should always be borne in mind
by the petrographer when called on to predict what will happen if a certain aggregate is used, or by
the engineer who evaluates such predictions."
The ability of chalcedony to be a deleteriously chemically reactive constituent in concrete aggregate

has been the subject of considerable discussion. Rhoades (22) states: "Typical chalcedony perhaps is
no more to be suspected of potential reactivity than quartz. However, authoritative conjecture
surrounds the possibility that some chalcedony verges toward an opaline character ... " Kelly,
Schuman, and Hornibrook (12) state: "the literature has thus far not considered chalcedony a likely
source of concrete expansion resulting from alkaline reactivity." On the other hand, chalcedony has
been listed as "deleteriously reactive with high-alkali cement" by Mielenz (19); McConnell, Mielenz,
Holland and Greene (17,18); Blanks (6), and in other papers prepared at the Bureau of Reclamation.
--``,```,`,`,`,,,,`,`````,`,,`,,-`-`,,`,,`,`,,`---
It is believed that the present paper represents the flrst published description of deterioration of
concrete through cement-aggregate reaction in which chalcedony is the major aggregate constituent
responsible for the deleterious reaction. McConnell, Mielenz Holland, and Greene (18) indicated
that chalcedony was a minor control in seven previous examples of such deterioration but that in all
of these cases glassy volcanic rocks comprised the major control. These seven cases are: Parker
Dam, Gene Wash Dam, Copper Basin Dam, Stewart Mountain Dam, American Falls Dam, Owyhee
Dam, and street pavements at Kimball, Nebraska. In addition, it may be noted that while phyllite is
cited (18) as the only control for the deterioration of the concrete in the structures at Buck Dam,
Virginia, petrographic work reported by Kelly, Schuman, and Hornibrook (12) indicates that
chalcedony is a major constituent of the phyllite in question.
SUMMARY AND CONCLUSIONS
All of the cores, regardless of brand of cement, contain deposits of gel in voids, cracks, and
aggregate particles in all sections, regardless of depth from the upper or outer surface of the
structure. The widest fresh cracks are coated with gel and the adjoining paste is soaked with gel. In
all of the concrete there is more gel than sulfoaluminate; more voids are filled with gel than with
sulfoaluminate. Sulfoaluminate is most developed in the most highly leached and cracked concrete.
It is believed that the gel produced by alkali-aggregate reaction is the major cause of the cracking of
the upper and outer portions of the cores.
A study of aggregate particles from the concrete revealed that approximately 70 percent of the chert
pebbles consisted, at least in part, of chalcedony. Chalcedony is one of the materials capable of

116
Daneshlink.com
daneshlink.com
participating in a deleterious chemical reaction in concrete. The study of the aggregate did not reveal
the presence of any other constituent regarded as capable of participating in such a reaction. The
study of the concrete specimens confirmed the indications developed from the examination of the
structure that the cracking was largely confined to near-surface zones and was more pronounced in
those portions of the structure in which it was reported that Cement A was used. Although specific
data on the alkali contents of the cements used in this project are not available, it is regarded as
probable that Cement A contained a larger percentage of alkalies than did Cement B.
All of the concrete cores were characterized by the presence of deposits of crystals of calcium
sulfoaluminate. Calcium sulfoaluminate is the normal product of the reaction between the tricalcium
aluminate in the cement, and the gypsum which is interground with the cement for the purpose of
controlling time of set. In normal, nondeteriorated concrete the calcium sulfoaluminate is widely
distributed in the cement paste and does not appear as deposits of crystals. Concrete which has
suffered deterioration from any cause frequently exhibits crystalline deposits of this material. Unless
it can be shown that additional sulfate has been provided from an external source for further
reaction with the aluminate portion of the cement, such deposits do not indicate that the concrete
containing them has undergone any deterioration due to sulfate attack. In the case of the Tuscaloosa
concrete, since the available information does not indicate an additional source of sulfate and since
the evidence of deterioration due to alkali-chalcedony reaction is thoroughly established, it is not
regarded as likely that there was a significant sulfate-attack factor involved in this occurrence. The
possibility of attack on the concrete by the slightly acid river water is regarded as having, at most, an
insignificant effect.
ACKNOWLEDGMENTS
This paper is based in large part on a report prepared for the District Engineer, Mobile District,
Corps of Engineers, 25 August 1949. The petrographic work was done by Katharine Mather,
assisted by Nelson B. Dodge and 0. G. Alessio; the chemical tests were made by J. C. Woosley and
L. M. Pepper. The entire investigation was directed by Herbert K. Cook.
--``,```,`,`,`,,,,`,`````,`,,`,,-`-`,,`,,`,`,,`---
REFERENCES
1) Alderman, A. R., Gaskin, A. ].,Jones, R. H., and Vivian, H. E. Australian Aggregates and
Cements, Part I of: "Studies in Cement-Aggregate Reaction." Australian Council for Scientific and
Industrial Research, Bulletin No. 229 (1947).
2) American Society for Testing Materials. Report of Committee C-9 on Concrete and Concrete
Aggregates. Proceedings, ASTM, Vol. 49, p. 292 (1949).
3) American Society for Testing Materials. "Descriptions of Constituents of Natural Mineral

Aggregates." ASTM Bulletin No. 171, pp. 32-34,Jan. (1951).
4) Bean, Leonard and Tregoning, J. J. "Reactivity of Aggregate Constituents in Alkaline Solution."

Journal ACI, Proceedings Vol. 41, pp. 37-51, Sept. (1944).

117
Daneshlink.com
daneshlink.com
5) Blanks, Robert F. Notes on the "Effect of Alkalies in Portland Cement on the Durability of
Concrete," Appendix to the Report of Committee C-1 on Cement. Proceedings, ASTM, Vol. 43, p.
203 (1943).
6) Blanks, Robert F. Modern Concepts Applied to Concrete Aggregate, Trans. (1950). ASCE, Vol.
115, pp. 403-431, Proceedings Apr. (1949).
7) Donnay,J. D. H. La Birefringence de forme dans la caleedoine. Annales de la Societe Geologique

de Belgique, Vol. LIX, pp. 289-302, July (1936).
8) Fisher, R. A. Statistical Methods for Research Workers. G. E. Stechert, New York (1946).
9)Folk, Robert L. Petrology of Lower Ordovician Cherts in Central Pennsylvania. Bulletin, Geol.
Soc. Am., Vol. 60, No. 12, Part 2, p. 1960, Dec. (1949).
10) Folk, Robert L., and Weaver, Charles Edward. Surface Features of Chert as Studied by the
Electron Microscope. Program, 1950 Annual Meetings, Geol. Soc. Amer. pp. 33-34, Nov. (1950).
11) Insley, Herbert. Discussion of: Petrographic Identification of Reactive Constituents in Concrete
Aggregates, by Bryant Mather. Proceedings, ASTM, Vol. 48, pp. 1126-1127 (1948).
12) Kelly, T. M., Schuman, L., and Hornibrook, F. B. "A Study of Alkali-Aggregate Reactivity by
Means ofMortar-Bar Expansions." JoumalACI, Proceedings, Vol. 45, pp. 57-80, Sept. (1948).
13) LEA, F. M. AND DESCH, C. H. "The Chemistry of Cement and Concrete." Longmans, Green;
London, (1935).
14) Mather, Bryant. Discussion of: Re activity of Aggregate Constituents in Alkaline Solutions, by
Leonard Bean and J. J. Tregoning. Journal ACI, Proceedings, Vol. 41, pp. 61-1 to 51-4, Nov. (1945).
15) Mather, Bryant. "Petrographic Identification of Reactive Constituents in Concrete Aggregates."

Proceedings, ASTM, Vol. 48, pp. 1120-1125 (1948),
16) Mather, Bryant. Discussion of: A Study of Alkali-Aggregate Reactivity by Means of Mortar Bar
Expansions, by T. M. Kelly, L. Schuman, and F. B. Hornibrook. Journal ACI, Proceedings, Vol. 45,
pp. 80-1 to 80-3,June (1949).
17) McConnell, Duncan, Mielenz, R. C., Holland, W. Y., and Greene, K. T. Cement-Aggregate
Reaction in Concrete. Journal ACI, Proceedings, Vol. 44, pp. 93-128, Oct. (1947).
18) McConell, Duncan, Mielenz, R. C., Holland, W. Y., and Greene, R. T. Petrology of Concrete
Affected by Cement-Aggregate Reaction. Geol. Soc. Amer., Engineering Geology Volume, pp. 225-
250, Nov. (1950).
19) Mielenz, Richard C. Petrographic Examination of Concrete Aggregates. Bulletin, Geol. Soc.
Amer. Vol. 57, pp. 309-318, Apr. (1946).

118 --``,```,`,`,`,,,,`,`````,`,,`,,-`-`,,`,,`,`,,`---
Daneshlink.com
daneshlink.com
20) Parsons, W. H., and Insley, Herbert. Alkali Etching Tests on Concrete Aggregates. Journal ACI,
Proceedings, Vol. 40, pp. 229-243,Jan. (1944).
21) Parsons, W. H., and Insley, Herbert. Aggregate Reaction With Cement Alkalies. Journal ACI,
Proceedings, Vol. 44, pp. 625-632, April (1948).
22) Rhoades, Roger. Discussion of: California Experience with the Expansion of Concrete Through
Reaction Between Cement and Aggregate, by Thomas E. Stanton, 0. J. Porter, L. C. Meder, and
Allen Nicol. Journal ACI, Proceedings, Vol. 38, pp. 236-7 to 23&-I I, Nov. (1942).
23) Steele, Byran W. Comments on I COLD Paper R. 25, prepared for the discussion during the
New Delhi, India, meeting in Jan. 1951. (1950).
--``,```,`,`,`,,,,`,`````,`,,`,,-`-`,,`,,`,`,,`---
119
Daneshlink.com
daneshlink.com
Fig. 1 - Flecks rifgel with included sawdust (A) and mound rifgel (B), which grew on core suiface cifter drilling (slight!J
enlarged). Core 5-1. depth 7.5- 9.0.
--``,```,`,`,`,,,,`,`````,`,,`,,-`-`,,`,,`,`,,`---
Fig. 2 -Open crack. (A-A). Partiai!J filled with gel (arrows). The crack reaches its greatest width in the chert pebble near the
center rif the photograph (slight!J reduced). Core 60-1, 7 in. from formed surface.

120
Daneshlink.com
daneshlink.com
F{g. 3 - Concrete with numerous lar;ge voids, depth about 33ft, core 5-1. The surface is typical rifthe core from 14.5 ft to 39.6
ft. Direction rifplacement towards bottom rifphotograph. The voids are more frequent ac§acent to coarse aggregate particles.
F{g. 4 - Honrycombed concrete from core 5-1, depth 13-14 ft. Direction rifplacement towards bottom rifphotograph. Anvws
indicate gel pockets; the lar;ge voids in the lower section are gel-lined.

--``,```,`,`,`,,,,`,`````,`,,`,,-`-`,,`,,`,`,,`---
121
Daneshlink.com
daneshlink.com
--``,```,`,`,`,,,,`,`````,`,,`,,-`-`,,`,,`,`,,`---
Fig. 5 -Thin section blank impregnated with resin, ground, exposed to warm humid airfor about 35 hr. Magnification about
3x. Photomicrographed August 23, 1949. Two large gelpockets (upper right and upper center) had taken up moisture and
swelled. The chalcedonic pebble (lower-left center) had taken up moisture; there was a wet spot on a sand grain at the lower left.
122
Daneshlink.com
daneshlink.com
F~. 6 - Same specimen cifter storage over water at room temperature between August 24, 1949, and October 18, 1949. The
gelpockets shown in the previous photograph have enlar;ged; ma'!Y more have appeared; cracks which were not obvious bifore are
now traced in reaction product.
Table 1 -Composition of Coarse Aggregate in Tuscaloosa Lock, as Determined by Counts of

Particles Intersected on Sawed Surfaces
Number of coarse aggregate particles,
Monolith and hole no. Sawed surfaces counted percent•
Chertb Other"
3-1 4 41 59
5-1 33 42 58
20-1 9 42 58
60-1 3 37 63
60-2 3 38 62
60-3 2 32 68
60-4 3 42 58
Average of all coresd - 42 58
•Calculated as a percentage of the total number of coarse aggregate parucles tntersected on the sawed surface.
hChert determined by macroscopic examination, without any distinction made between chert consisting of quartz
and chert consisting of chalcedony.
c'fhe coarse aggregate constituents other than chert are quartz, quartzite, sandstone, with a few particles of granite
gneiss.
dArithmetic averages, based on a total of 5518 coarse aggregate particles.
--``,```,`,`,`,,,,`,`````,`,,`,,-`-`,,`,,`,`,,`---

123
Daneshlink.com
daneshlink.com
TOTAL. NliM9E:A t:XAMINED =40

AVERAGE • I, !:130$
OBSE:RVE.O RANGE,. 1.!1240- 1.!>4ZO
14
--``,```,`,`,`,,,,`,`````,`,,`,,-`-`,,`,,`,`,,`---
0
1.6240 1.5260
I
L»l!:l 1.5J45 1.5370 1.:>400 1,5420
1.5415
IND£)( 01" RE:f'RACTlON
Fig. 7 - Distribution rif chalcedonic chert coarse-aggregate particles selectedfrom concrete from Tuscaloosa Lock, with respect to
index rif refraction.
Fig. 8 - Sawed surface rif core cut longitudinal!J. The direction rifplacement is toward the top rif the photograph. A-A indicated
the ends rif a crack .rystem which can be traced across the core, passing through two chert pebbles (1 and 2). The white chert pebble
(1) has an outer zone very firmlY bonded to the matrix and the center rif the pebble separated 1?J cracks from the outer zone. The
chert pebbles indicated 1?J arrows (2 through 7) show the wide cracks common in chert particles in this concrete. In pebble (2) the
gel filling ofparts rif the crack rystem can be seen at the left and right. G indicates gel pockets.

124
Daneshlink.com
daneshlink.com
--``,```,`,`,`,,,,`,`````,`,,`,,-`-`,,`,,`,`,,`---
Fig. 9 - Part of area shown in Figure 8 magnified about 3 times. Arrows at mat;gin mark ends of crack. 1, 2, 3 are chert
pebbles with well-developed wide cracks; those in 3 are locai!J geljil/ed. 4, 5 are crackedpebbles; the cracks are natTower than
those in 1, 2, 3.
Table 2 - Distribution of Gel Pockets on Sawed Surfaces of Concrete from Tuscaloosa Lock
Numbers intersected on sawed surfaces, as percent
Monolith and hole Coarse aggregate not Coarse aggregate associated

number associated with gel with gel
Isolated gel pockets
Chert Other Chert Other
3-1 36 56 5 2 1
5-1 36 56 5 2 1
20-1 38 48 5 5 3
60-1 31 57 4 3 2
60-2 32 56 5 5 3
60-3 19 46 9 13 13
60-4 34 48 6 7 4
Average, all 7 cores 36 53 5 4 2
Numerical Totals 2032 3016 281 189 110

125
Daneshlink.com
daneshlink.com
--``,```,`,`,`,,,,`,`````,`,,`,,-`-`,,`,,`,`,,`---

Daneshlink.com
daneshlink.com
Reprinted, with permission, from Cement and Concrete Research, V 29, 1999, pp. 1277-1280, ©Elsevier, London, England.
HOW TO MAKE CONCRETE THAT WILL NOT SUFFER DELETERIOUS

ALKALI-SILICA REACTION
Bryant Mather'
ABSTRACT
Excessive expansion of concrete due to alkali-silica reaction will not occur if any one of the
following circumstances exists: (1) the aggregate is insufficiently reactive; (2) the pH of the pore fluid
is not too high; (3) the amount of reaction product formed is not sufficiently large or not sufficiently
expansive so that its expansion can cause damage; or (4) there is not enough available water to cause
the reaction to progress so as to develop the expansive product and to be available for imbibition by
the product so as to cause it to swell and disrupt the concrete. Put the other way around, excessive
expansion can occur only if there is enough potentially expansive alkali-silica reaction product and
water so that, as the product takes up the water and swells, the concrete expands excessively. Hence,
what is needed to avoid excessive expansion is to be able to predict the probable reactivity of the
available aggregates, the probable mechanisms by which the pH of the pore fluid might get well
above 13, and, when justified, select and implement appropriate precautions against the undesirable
consequences of these eventualities. Published by Elsevier Science Ltd. All rights reserved.
Keywords
Aggregate; Alkali-aggregate reaction; Expansion; Petrography; Pozzolan
INTRODUCTION
In 1979, ACI published a paper I wrote called "Concrete Need Not Deteriorate," which included
the essence of remarks I have made at quite a number of different places over the past 40 or more
years [1]. It says, in effect, that concrete will provide the sort of performance desired, provided only
that the specifications are correct and that they are followed. We know the relevant levels of
--``,```,`,`,`,,,,`,`````,`,,`,,-`-`,,`,,`,`,,`---
properties needed by concrete to resist any reasonably anticipated set of environmental conditions.
We know what sorts of quality-control procedures the contractor should use and what sorts of
quality-assurance activities the owner should use to see to it that the standards of good practice in
executing the work are followed.
Later I wrote a paper on how to make concrete that will be immune to the effects of freezing and
thawing [2]. Still later I discussed at the Transportation Research Board what one needs to do to
avoid premature deterioration and to obtain the desired performance of concrete in transportation
[3]. Damage to concrete by freezing and thawing and damage caused by excessive expansion due to
alkali-silica reaction (ASR) were the two principal issues relating to concrete durability that were
selected for detailed research in the Strategic Highway Research Program. It seems appropriate now
* Corresponding author. Tel.: 601-634-3264; fax: 601-634-3242

E-mail address: [email protected] (B. Mather), Structures Laboratory, U.S. Army Engineer Waterways Experiment
Station, 3909 Halls Ferr.v Road, Vicksburg, MS 39180-6199. USA
Received 2 September 1998; accepted 12 November 1998

127
Daneshlink.com
daneshlink.com
to try to do some technology transfer about preventing damage to concrete resulting from ASR.
Tom Stanton, of the California Department of Transportation, discovered ASR [4]. I went to work
for the Concrete Laboratory of the Corps of Engineers in August 1941. Word of his discovery had
already reached that laboratory, and one of the first bits of work I did was to look at his "siliceous
magnesian limestone," which was described as the offending ingredient of the concrete that had
deteriorated. At that point, the Corps of Engineers was quite concerned because, to us, "magnesian
limestone" meant dolomite, and dolomite as crushed stone was very widely used as aggregate in
concrete in Corps of Engineers structures. Our position had been that, whereas most limestones
sold as aggregate were satisfactory, substantially all dolomites were. It turned out that the "siliceous
magnesian limestone" that caused the trouble in California is a very unusual rock that contains a lot
of opal. As far as I know, no other dolomite in the world contains enough opal to cause trouble due
toASR.
ASSESSING AGGREGATE REACTIVITY
Normally, the first issue confronting a person contemplating the use of concrete with regard to ASR
is: "Are there some reactive aggregates out there among those that might get used in the work?"
Hence, one usually thinks first of the procedures available for answering the question: "Is the
material in this aggregate source potentially deleteriously reactive with alkalies?"
ASTM C 33-97 [5], the "Standard Specification for Concrete Aggregates," requires in Section 7.3
that: "Fine aggregate for use in concrete that will be subject to wetting, extended exposure to humid
atmosphere, or contact with moist ground shall not contain any materials that are deleteriously
reactive with the alkalies in the cement in an amount sufficient to cause excessive expansion of
mortar or concrete, except that if such materials are present in injurious amounts, the fine aggregate
may be used with a cement containing less than 0.60% alkalies calculated as sodium oxide equivalent
(Na2 0 + 0.658 K 2 0) or with the addition of a material that has been shown to prevent harmful
expansion due to the alkali-aggregate reaction, and in Section 11.2: "Coarse aggregate for use in
concrete that will be subject to wetting, extended exposure to humid atmosphere, or contact with
moist ground shall not contain any materials that are deleteriously reactive with the alkalies in the
cement in an amount sufficient to cause excessive expansion of mortar or concrete except that if
such materials are present in injurious amounts, the coarse aggregate may be used with a cement
containing less than 0.60% alkalies calculated as sodium oxide equivalent (Na2 0 + 0.658 K 20) or
--``,```,`,`,`,,,,`,`````,`,,`,,-`-`,,`,,`,`,,`---
with the addition of a material that has been shown to prevent harmful expansion due to the alkali-
aggregate reaction."
An appendix to C 33 discusses the use of C 295, petrographic examination; C 289, the quick
chemical test; C 227, the mortar-bar test; C 342, the Conrow test; and C 586, the rock cylinder
method. In each case, some advice is given for interpreting the results in terms of the potential for
excessive expansion due to ASR. Action to revise this appendix to cite more recently developed test
methods is currently pending in ASTM. Reference to ASTM C 1260, which was approved in 1994,
will be included. This procedure permits detection in 16 days of the potential of an aggregate for
deleterious ASR and hence the need for additional evaluation.
The U.S. Army Corps of Engineers' Manual [6] includes guidance to field offices in preparing
specifications for concrete. It is assumed that a survey of potential aggregate sources will be made
and data on samples from them will be obtained. The project specifications will contain a list of the

128
Daneshlink.com
daneshlink.com
sources that are regarded as containing adequate amounts of material of satisfactory quality to
provide aggregate for the project. The evaluation of reactivity will include examination and tests by
C 227, C 289, and C 586. The manual states: "If aggregates containing reactive constituents are to be
used, the problem then becomes one of identifying the reactive constituents and determining the
conditions under which the available aggregates may be used. Where the reactive constituents can be
positively determined to occur in such small proportions as to be innocuous, an aggregate, if
otherwise of suitable quality, may be used without special precautions. Where the reactive
constituents occur in such proportions that they are potentially deleteriously reactive, it will be
necessary to use low-alkali cement or an effective pozzolan, or both, with the aggregate. If the
requirement of low-alkali cement would impose serious difficulties of cement procurement or
excessive increase in cost, consideration should be given to the use of pordand blast-furnace slag
cement (a blend of portland cement and slag), pordand-pozzolan cement, cement with GGBF slag,
or cement with a pozzolan, or both, that will prevent excessive reaction even when high-alkali
cement is used. When consideration is given to the use of any of these materials in lieu of low-alkali
cement, mortar-bar tests should be conducted to verify that the potentially deleterious expansion
will be reduced to meet the criteria in Appendix D orE in [6]. If petrographic examination
determines the presence of potentially reactive materials in excess of the limits given in Appendix D,
mortar-bar tests, ASTM C 227 shall be performed. If the petrographic examination determines the
presence of strained quartz in excess of the limit given in Appendix D, the high-temperature mortar-
bar test shall be performed."
PRECAUTIONS TO TAKE TO AVOID EXCESSIVE EXPANSION DUE TO ASR
As noted in the Corps of Engineers' Manual, in some cases, no specific precautions are needed to
preclude excessive expansion due to ASR. These include the following:
1. The aggregates have been shown not to be potentially deleteriously reactive, i.e., innocuous with
respect to ASR.
2. The cementitious medium to be used has been shown to be of sufficiendy low alkali content and
the possibility of alkali from any other source is sufficiendy remote to be disregarded; hence, it
can be confidendy expected that the pH of the pore fluid will not get too high.
3. The cementitious medium to be used includes a sufficient amount of satisfactory ground slag or
pozzolan, so excessive expansion will not occur with this composition, which was selected for
reasons of temperature control or economy or both, unrelated to ASR.
4. The concrete to be produced will all be in an environment in which it will rapidly get dry
(internal relative humidity well below 80%) and remain so during its service life.
In some cases where the aggregate is classed as potentially deleteriously reactive, the level of such
reactivity is sufficiendy low so that the only precaution that is needed is to specify that the cement
be "low alkali." In others, one should preselect one or more aggregate sources that have a good
service record when used in similar work with the local cement or cements. In still others, there
should be a study of other constituents that might be included in the concrete such as ground
granulated blast-furnace slag or a pozzolan. Natural pozzolans (volcanic ash, diatomaceous earth,
etc.), fly ash, and silica fume have all been shown to be able to prevent excessive expansion of
potentially deleteriously reactive cement-aggregate combinations.
--``,```,`,`,`,,,,`,`````,`,,`,,-`-`,,`,,`,`,,`---

129
Daneshlink.com
daneshlink.com
PROCESS AND MECHANISM
While we talk about ASR, in every case of damage to concrete in service I know of caused by it, it
was the presence of alkali (sodium ion or potassium ion or both in solution) that caused the pH of
the pore fluid to rise to levels where it became aggressive to the thermodynamically metastable silica
in the aggregate. It is not the alkali that does the reacting; it is the hydroxyl ions. Vivian [7]
demonstrated that: "The hydroxyl ion must be considered to be the reacting ion." He based this
conclusion on the observation that tetramethyl ammonium hydroxide produced a similar reaction
product to that formed from sodium hydroxide. He found that a mortar bar made with low-alkali
cement to which sufficient tetramethylammonium hydroxide had been added to provide 0.85%
hydroxyl ion by mass of the cement expanded 0.85% at 10 months. A control with low-alkali cement
did not expand.
The alkalies that end up in concrete, from the cement or elsewhere, are almost never NaOH or
KOH unless we are doing, research in the laboratory; they are often potassium sulfate from the
cement, or sodium chloride from sea water, etc. Solutions of such salts are pH neutral, not above the
value of 12.6 of a saturated calcium hydroxide solution. The reason alkalies get a chance to mobilize
hydroxide ions, raise the pH, and allow the pore fluid to become aggressive to the reactive
constituents of aggregate when such are present is that the chloride or sulfate ions react and
precipitate out of the solution, usually to form calcium sulfoaluminate or calcium chloroaluminate or
gypsum. I suggest that some attention might be given to the use of a cement with either litde or no
aluminate as a means of reducing the likelihood of an undesirably high pH developing in the pore
fluid in concrete. "Zero C3A" cements exist.
REMEDIATION
Suppose someday you get the news that your structure (or one of your structures) that you had been
counting on for another half century or more of service has been declared to contain ASR product,
or worse, found to have typical pattern of cracking and, on investigation, found to have both ASR
product and remaining unreacted reactive aggregate constituents. What are the chances that
deterioration will progress to an unacceptable level? This is not an easy question to answer. As one
philosopher put it, "Making predictions isn't easy, especially about the future." What we did years
ago was to store cores with length-change measuring inserts in the fog room [8]. What was found
was that a core from a dry area showed twice as much expansion as any of the others, all of which
came from areas that were wet or damp. This suggested that there was a limited amount of
expansion to be expected, much of which had already taken place in the wet or damp portions of
the structure.
Tuscaloosa Lock in Alabama, the Corps of Engineers' best-studied ASR-affected structure [9,10], is
a classic example of how a structure that when first examined was declared to be "so badly cracked
that the natural impression would be that it is about to collapse" [ 11] and 30 years later [1 0] was be
found to be "of generally good quality."
Another structure, also in Alabama, from which cores were studied in detail at our laboratory is
Martin Dam [12]. From that study, it was concluded that "the condition of the concrete ... can be
considered acceptable" after 50 years of service. ASR evidence was found in all the concrete cores
examined. The reactive constituent was metamorphosed quartz and quartzite that had undulatory
--``,```,`,`,`,,,,`,`````,`,,`,,-`-`,,`,,`,`,,`---
130
Daneshlink.com
daneshlink.com
extinction. The reaction product was typical of ASR. The concrete had less reaction product than
seen in other structures still in good condition after 10 or 24 years of service. One photomicrograph
of Martin Dam concrete was so illustrative of ASR phenomena that it was used as the frontispiece in
the Proceedings of the 1978 Fourth International Conference on the Effects of Alkalies in Cement
and Concrete [13].
A somewhat similar story is that of Stewart Mountain Dam in Arizona, where major expansions
were observed, then they ceased, and 30 years later the concrete was in as good as or better
condition than it had been previously.
ACI Committee 221 on Aggregates has prepared an extensive report on ASR in which remediation
will be addressed.
CONCLUSIONS
Procedures are available to reach a probably accurate answer to the question: "Is the aggregate from
this source reactive?" In some cases, it is not necessary to take any particular precautions to avoid
excessive expansion due to ASR even if the answer is "yes." These cases include (1) when there is a
"nonreactive" aggregate available at comparable cost; (2) when the cement is low alkali and that is all
that is needed; and (3) when the intent is to use ground slag or a pozzolan in adequate amounts for
other reasons. However, if specific precautions are needed, they can be taken. Pozzolans and ground
slag are widely available.
It is better to take the steps to avoid excessive expansion before the structure is built, but even if this
is not successfully accomplished and severe cracking due to internal expansion caused by ASR
occurs, one does not necessarily need to "run for the hills." Tuscaloosa Lock, Stewart Mountain
Dam, and others with severe internal expansion have gone on for 30 or more years giving adequate
service despite having manifested a lot of internal expansion.
Meanwhile, all of us who are interested in ASR should keep up with the state of the art. In 1994
ASTM published a new edition of STP 169 with excellent papers by Stark [14] and Ozol [15]. As
previously noted, ACI 221 on Aggregates has completed a review. The 10th International
Conference was held in Melbourne, Australia, in 1996, and the proceedings are available [16]. There
is a wealth of information in the compiled proceedings of the preceding conferences.
POSTSCRIPT
A reviewer of this manuscript suggested that I should also address the question of why is it that,
since the principles of ASR have been known for more than half a century, examples of distress due
to ASR are still being found in structures much less than 50 years old and how current actual
practice agrees with the concepts I outlined to avoid excessive expansion due to ASR.
There are many situations in which it was assumed none of the aggregate sources in a region were
potentially deleteriously alkali reactive, because no distress had ever been observed that was proven
to have been caused by ASR. Then there was an abrupt change upward from about 0.4% Na20 0 , to
about 1.2% Na20e in the alkali content of the available commercial cement. In one case there was
severe distress from the use of crushed stone containing substantial amounts of opal, the presence
--``,```,`,`,`,,,,`,`````,`,,`,,-`-`,,`,,`,`,,`---

131
Daneshlink.com
daneshlink.com
of which was well known to geologists in the area but not to any concrete people who were
prepared to realize its significance.
There have been strong efforts to dissuade users from specifying low-alkali cement "when it is not
needed," because in many locations it is difficult or impossible for some or all the cement producers
to make it. The methods for establishing the fact of reactivity or nonreactivity of materials from
aggregate sources without a service record have often been somewhat capricious, and the
procedures for evaluating the effectiveness and needed dosage of a mitigating material- slag,
pozzolan, lithium salts - are likewise less precise and straightforward than would be preferred.
In my view, current practice in the Corps of Engineers is more effective than most and close to
being as good as the state of the art allows. I believe that if all constructions were carried out with
the approach outlined in the Corps Manual, there would be few, if any, structures showing distress
due toASR.
REFERENCES
1) B. Mather, Concrete need not deteriorate, Concrete Intemational1 9 (1979) 32-37.
2) B. Mather, How to make concrete that will be immune to the effects of freezing and thawing. in:
D. Whiting (Ed.), Paul Klieger Symposium on Performance of Concrete, Special Publication 122,
American Concrete Institute, Detroit. Mi. 1990, pp. 1-18.
3) B. Mather. Concrete in transportation: Desired performance and specifications, Transportation

Research Record No. 1382. Transportation Res. Bd. NAS-NRC, Washington. DC, 1993, pp. 5-1 0.
4) T.E. Stanton. Expansion of concrete through reaction between cement and aggregate,
Transactions. American Society of Civil Engineers, 107 (1940) 54.
5) American Society for Testing and Materials (ASTM). Standard specification for concrete
aggregate. Designation: C 33-97. 1998 Annual Book of ASTM Standards, West Conshohocken, P A,
04.02 (1998) 10-16.
6) U.S. Army Corps of Engineers, Standard Practice for Concrete for Civil Works Structures,
Engineer Manual II 10-2-2000, including Change 1, 31 July 94; Washington, DC, 20314-1000,
reprinted by the American Society of Civil Engineers as Technical Engineering and Design Guides
as Adapted from the U.S. Army Corps of Engineers No. 8, 122 pp, 1994, ASCE Press, New York.
7) H.E. Vivian, The effect of hydroxyl ions on the reaction of opal, Aust J Appli Sci 2 (1) (1951)
108-113.
8) W.I. Luke, Alkali-carbonate reaction in concrete from Chickamauga Dam Powerhouse,

Proceedings, ASTM, Philadelphia. P A, 64 (1964) 887-902.
9) B. Mather, Cracking of concrete in the Tuscaloosa Lock, Proceedings, Highway Research Board.
Washington. DC, 31 (1952) 218-233.
--``,```,`,`,`,,,,`,`````,`,,`,,-`-`,,`,,`,`,,`---

132
Daneshlink.com
daneshlink.com
10) J.E. McDonald, R.L. Campbell, An investigation of concrete condition, William Bacon Oliver
Lock and Spillway, Miscellaneous Paper C-77-5, U.S. Army Engineer Waterways Experiment
Station, Vicksburg MS, 1977.
11) R.W. Carlson, Wave velocity apparatus and its place in nondestructive testing of Concrete in
Place, Memorandum for Office, Chief of Engineers, November, 1948.
12) K. Mather, Condition of concrete in Martin Dam after 50 years of service, Cement, Concrete,
and Aggregates, ASTM, Philadelphia, PA, 3 (1) (1981) 53-62.
13) S. Diamond (Chairman), Proceedings of the Fourth International Conference on the Effects of
Alkalies in Cement and Concrete, School of Civil Engineering Publication CE-MAT-1-78, Purdue
University, W. Lafayette, IN, 1978.
14) D. Stark, Alkali-silica reactions in concrete, in: P. Klieger, J.P. Lamond (Eds.), Significance of
Tests and Properties of Concrete and Concrete-Making Materials, ASTM STP 169C, ASTM,
Philadelphia, PA 1994, pp.365-371.
15) M.A. Ozol, Alkali-carbonate rock reaction, in: P. Klieger, J.P. Lamond (Eds.), Significance of
Tests and Properties of Concrete and Concrete-Making Materials, ASTM STP 169C, ASTM,
Philadelphia, PA 1994, pp.372-387.
16) A, Shayan, Alkali-aggregate reaction in concrete, in: Proceedings of the 1Oth International
Conference on Alkali-Aggregate Reaction in Concrete, Melbourne, Australia, 1996.
--``,```,`,`,`,,,,`,`````,`,,`,,-`-`,,`,,`,`,,`---
133
Daneshlink.com
daneshlink.com
--``,```,`,`,`,,,,`,`````,`,,`,,-`-`,,`,,`,`,,`---

Daneshlink.com
daneshlink.com
Reprinted, with permission, from International Symposium: Durabi!iry of Concrete- 1969, ©RILEM, Bagneux, France.
SULFATE SOUNDNESS, SULFATE ATTACK, AND EXPANSIVE

CEMENT IN CONCRETE
by Bryant Mather
Concrete Division, U. S. Army Engineer Waterways Experiment Station, Corps of Engineer, Jackson,
Mississippi (USA)
SUMMARY
The phenomena related to the formation of hydrated sulfates in concrete, or in aggregates, cement
pastes, or mortars, have been investigated for many years for a variety of purposes. The cyclic
immersion of aggregate particles in solutions of sodium or magnesium sulfate, followed by drying, is
the basis of one of the oldest procedures employed to develop data purported to relate to aggregate
"soundness." The storage of mortar specimens in sulfate solutions is the basis of many tests for
sulfate resistance of cements. Sulfate-resistance testing procedures in which the mortar is mixed with
added sulfate and the specimens are stored in water are in widespread use. These latter procedures
are similar to procedures employed in studies of expansive cements.
Although the physical and chemical effects of the phenomena associated with the formation of
hydrated sulfates in concrete, aggregate, paste, and mortar have long been studied, many
fundamental aspects of these phenomena are still poorly understood. The relation of the physical
properties of the material surrounding the locus at which the hydrated sulfate is produced to the
amount of expansion that results, the degree to which the expansion can be reduced by restraint, the
force required to prevent expansion or to prevent the reaction from taking place, the consequences
of the restraining force being uniaxial, biaxial, or triaxial rather than hydrostatic are all matters that
appear to fall into the category of those about which less is known than would be desired. Today,
when tools for study are vasdy improved, when our knowledge of the composition and constitution
of the reacting substances has been gready increased, and when we can see the merging of several
practical areas of application of the knowledge to be obtained, appears to be the time when these
gaps in knowledge may be expected to be filled. I hope that this paper will indicate to workers
concerned with chemical attack on concrete, durability of aggregates, and the study of expansive
cements that these areas of study are merging and that progress in each will contribute to progress in
all if the experiments are carefully planned in the light of available knowledge.
The phenomena related to the formation of hydrated sulfates in concrete, or in aggregates, cement
pastes, or mortars, have been investigated for many years for a variety of purposes. The cyclic
immersion of aggregate particles in solutions of sodium or magnesium sulfate, followed by drying, is
the basis of one of the oldest procedures employed to develop data purported to relate to aggregate
"soundness." The storage of mortar specimens in sulfate solutions is the basis of many tests for
sulfate resistance of cements. Sulfate-resistance testing procedures in which sulfate is added to the
mortar and the mortar specimens are stored in widespread use. These latter procedures are similar to
procedures employed in studies of expansive cements.
Mather1, in discussing factors that influence the deterioration of concrete and in discussing2 the
manner in which cracking of concrete may be induced by environmental effects, emphasized the
--``,```,`,`,`,,,,`,`````,`,,`,,-`-`,,`,,`,`,,`---

135
Daneshlink.com
daneshlink.com
factors that may cause deterioration and cracking as a consequence of internal expansion in the
concrete. Among the mechanisms discussed were hydration of calcium oxide (CaO) or magnesium
oxide (MgO), corrosion of corrodible metal., freezing of water, alkali-silica reaction, alkali-carbonate
rock reaction, and sulfate attack. In a review3 on the effects of seawater on concrete and a later
review4 of studies of sulfate resistance, the reactions between externally supplied magnesium sulfate
and the constituents of the cement were examined and the volumetric effects as given by Hansen
and Offutt5 were described. The reaction of tricalcium aluminate with gypsum and water to
form ettringite was given as follows:
3CaO•Al2 0 3+ 3CaS04•2H20 + 25H2 0~

88.8 222.3 450
3CaO•Al2 0 3•3CaS04•31H20
714.7
This reaction involves a net reduction in volume, but the relative volume of anhydrous C3A
(tricalcium aluminate) to the volume of ettringite formed from it are as 1:8. The observed
expansions in sulfate reaction and sulfate-reaction experiments cannot be correlated precisely with
an eight-fold increase in volume of the C3A. Indeed cements of the same assumed C3A content do
not yield test specimens of the same increase in length when exposed to sulfate in excess and time
for the reaction to go to completion. Mather4 presented contrasting results obtained from tests of
two cements.
Cement
13 15
Calculated C3A, % 10.3 12.5
Calculated C3S (tricalcium silicate),% 56.9 67.7
3 days 1800 3315
Compressive strength, psi 7 days 2465 4650
28 days 3875 6140
Heat of hydration, cal/ g, 28 days 85 116
Fineness, air permeability, cm2 / g 3485 3340
Composition by microscope, C3A, % 1.6 8.6
dark prismatic,% 1.9 1.2
"glass," 0/o 8.2 0.3
Total 11.7 10.1
Expansion, %, 1 yr 0.408 0.132 (0.143)
--``,```,`,`,`,,,,`,`````,`,,`,,-`-`,,`,,`,`,,`---
The time versus length-change curve for cement 13 showed a slow but increasing rate, while that for
cement 15 showed a rapid early rate, reaching a maximum increase in length by 56 days and litde
further change to an age of one year. The maturity of pastes made with cement 15, as measured by
heat evolution or by strength development, was as great at 7 or 28 days as was that with cement 13
at one year. The C3A in cement 13 was predominandy present as the material classified by
microscope examination as glass, while that in cement 15 was present as material recognized as C 3A
crystals. The specimens made with cement 15 expanded rapidly and matured rapidly; those made
with cement 13 expanded and matured slowly.
All of the procedures that have been mentioned involve measuring the effects of internal expansions
resulting from chemical processes that lead to the formation of hydrated sulfates. When an

136
Daneshlink.com
daneshlink.com
expansion occurs, it is because the hydrated sulfates that are formed, either by precipitation from
solution or by reactions in place, occupy a larger volume than was available and hence exert stress
on their surroundings. The stress so exerted produces elastic and inelastic strains of magnitudes and
proportions depending on the level of stress, the mechanical properties of the system, and the
geometry of the specimen. The relevant mechanical properties include modulus of elasticity, creep,
porosity, and strength. All of the hydrated sulfates of interest are crystalline to some degree;
consequendy, their formation may be referred to as crystallization, and an increase in the amount
that is present may be referred to as crystal growth. The stresses accompanying such formation have
been referred to as resulting from crystal growth even as the stresses accompanying the freezing
water have been referred to as resulting from ice crystal growth. It is here suggested that the facts
that the substances are crystalline and that an increase in the amount present involves crystal growth
are of minor significance to an understanding of the mechanisms by which such formation produces
--``,```,`,`,`,,,,`,`````,`,,`,,-`-`,,`,,`,`,,`---
stresses which induce the strains that are observed as expansions and cracking.
If the space occupied ill an aggregate particle or in a sample of mortar, cement paste, or concrete
by a quantity of hydrated sulfate after equilibrium of strain had been occupied by a liquid or gas that
had exerted the same stress under the same conditions of confmement, the fmal volumetric relations
of the system in which it existed would be the same. The relative phenomenology of the sulfate
soundness test of aggregates, the various sulfate susceptibility tests for pastes, mortar, and concretes,
the various expansivity tests for expansive cements, tests for resistance to freezing and thawing, tests
for the potential expansivity caused by chemical reactions of alkalis and aggregates, tests for
oxidation or hydration of aggregate or cement constituents or embedded corrodible metals does not
primarily depend on whether the products formed are crystalline or noncrystalline. The relevant
phenomenology concerns the specific volume of products formed, the amount of products formed,
the compressibility of the products formed, the restraint imposed by the surroundings, and the rate
of formation of the products.
The best known phenomenon by which an internal expansion can be produced in an aggregate.
particle or in a sample of cement paste, mortar, or concrete is that involving the freezing of water to
produce ice. Ice occupies about 9% more volume than the water that was frozen in its formation. It
has been shown that if liquid water completely fills the space in which it exists, is completely
restrained from increasing in volume, and the temperature of the water can be lowered as far as may
be desired, a restraint by the surroundings of 31,000-psi hydrostatic force will be imposed on the
water to keep it from undergoing the change into ice. These data also indicate that the maximum
internal force that can be exerted by water trying to change into ice as the temperature is reduced is
31,000 psi. This is a stress that, if exerted, will prevent ice crystals from forming; it is not the force
exerted on the surrounding by the growth of the ice crystals.
In the case of one of the better studied hydrated sulfates of interest, ettringite, calcium aluminum
sulfate hydrate (C 6 AS3 H 32 ) it has been observed that, if one takes the anhydrous analogue of this
material (C 4 A 3S) plus calcium sulfate and calcium oxide and packs it tighdy in a container capable
of providing a very high level of volumetric restraint and then allows it to hydrate in place, the
strains observed in the container can reach levels indicative of stresses of the order of 50,000 psi.
All of the systems of interest to students and users of concrete have "elastic" properties, inelastic
properties, and tensile strength. In most such systems the magnitude of the modulus of elasticity,
ability to creep, or tensile strength varies with time. Consequendy, in most such systems the

137
Daneshlink.com
daneshlink.com
volumetric response of a sample of such a system to the development within it, at a given time in its
history, of a unit of product, the formation of which exerts a given stress on its surroundings, will be
different from the response exerted at any other time in its history, regardless of whether the internal
stress of given magnitude results from one or another of a large number of mechanisms. Similarily,
the response of the system to the generation of such an internal stress, at a given time, will vary
depending on the locus of this stress relative to the geometry of the sample of the system that is
under study and hence the restraint supplied by the material of the system-at the locus at which the
stress develops. If the sample of the system is a sphere and the locus of the formation of the
product that, when formed, occupies a larger volume than is available is at the center of the sphere,
the observed expansion of the sample will be a uniform increase in all diameters of the sphere,
provided the magnitude is within the amount that can be accommodated by elastic and creep strains.
If the locus is sufficiently near the surface of the sample, regardless of the geometry of the sample,
the local internal expansion will, without any greater force being involved, create a popout and
crater.
Bache and Isen6 have reported the results of experiments involving the production of internal
hydrostatic pressures of known magnitude at different depths from the surface of samples of
mortars having different mechanical properties. Their-results show that, as the locus of internally
generated hydrostatic stress is placed deeper beneath the surface of the samples, the higher the stress
must be in order for a popout and crater to form. The relation that they found was as follows:
p h
-~10-
d at
where P = hydrostatic pressure required to form a popout
at= tensile strength of the system
h = distance from the surface to the locus of stress
d = diameter of the sphere exerting the hydrostatic stress
Aroni et al. 7 note that it wag Michaelis 8 who first attributed the disruption of concrete, when
attacked by aggressive waters containing sulfates, to the reaction between C3A and sulfates to form
ettringite ( C 6 AS3 H 32 ). They note that the useful feature of this reaction to the application of
expanding cement to chemical prestressing is the high energy of the expansion; i.e., its ability to
perform work and expand under restraint. They discuss the various papers that have attributed
internal expansion to the monosulfate (C 4 ASH 12 ) rather than, or in addition to the trisulfate
(ettringite) and conclude that the weight of the evidence favors associating it with the latter. They
note that expansive cements, based on the work of Klein, derive their expansion from the hydration
of C 4 A 3S to form C 6 AS3 H 32 •
Aroni et al. 7 also note that there is little in the literature on the basic phenomena of crystal growth
under pressure. As did Schaffer,9 they refer to the work of Beeker and Day, 10 but do not appear to
have rejected the hypothesis. They refer to the conclusions of Hansen11 end Taber12 and ask what
the applicability of these conclusions is to expanding concrete. They consider two cylinders of
expanding concrete of equal expansive potential, one free to expand in all directions, the other re-
strained longitudinally. They say one would expect an "inverse" relation between longitudinal and
transverse strain-a longitudinal shortening should, by the Poisson effect, create a transverse
elongation. They quote Klein et al., 13 who referred to evidence that in small specimens with
symmetrical restraint, the expansive movements in all directions are substantially equal per unit
--``,```,`,`,`,,,,`,`````,`,,`,,-`-`,,`,,`,`,,`---
138
Daneshlink.com
daneshlink.com
length, even under uniaxial restraint, where the cross section normal to the axis of restraint is under
"adequate" compression. Aroni et al cite a number of instances that indicate a direction relation
including comparisons of specimens with uniaxial Oongitudinal) versus triaxial restraint, in which
those with only uniaxial restraint showed an increase in longitudinal expansion of up to 25%.
Hansen11 reviewed the literature relating to the phenomenon of crystal growth as a source of
expansion in pordand-cement concrete. He begins his review with the statement that "portland-
cement pastes expand during setting and hardening, whereas they would be expected to contract
because the volumes of the hydrated phases are less than the sum of the volumes of the water and
the reactants forming these phases." It is interesting that he states expansion to be the volume
change that occurs on setting and hardening, because other writers imply, quite as blundy, that there
is a contraction or shrinkage on setting. He states that the expansion has been explained on the basis
that the hydrated compounds form by direct reaction with water from the unhydrated solids. As do
others, he refers to the work of Becker and Day10 and attributes the observed ability of a growing
crystal to lift a weight as a consequence of the existence of a film of solution beneath the growing
crystal, the concentration differences in the solution, and the relation of solubility of the crystal to
the pressure on it. He cites the conclusions ofTaber, 12 both with regard to these circumstances and
with regard to the relations of crystals growing in porous media having a range of pore sizes. Taber
showed that crystals growing in larger pores will receive material from the solution in the smaller
--``,```,`,`,`,,,,`,`````,`,,`,,-`-`,,`,,`,`,,`---
pores and grow at the interface between liquid and crystal. Hansen reviews the circumstances
relating to the growth of a group of crystals nucleating from a saturated solution and concludes that,
if none of the crystals are under external pressure, there could develop a single crystal without
pressure being exerted. He concludes that theories that attribute expansions in concrete to crystals
growing from solution against one another do not appear to be realistic. He describes two
conditions under which a crystal may grow and exert pressure on materials opposing its growth.
One is the condition in which a weighted crystal rests on the floor of a container on a film of
solution that is supersaturated with respect to the bottom face of the crystal. The other is the
condition in which a single crystal fills a container that opens into another container filled with a
solution supersaturated with respect to the film of solution supporting the crystal. He mentions last
the rehydration of a partially dehydrated solid exposed to water vapor, noting that in this way,
crystals can grow without the presence of liquid water.
Schaffer9 wrote, "Crystallization tests, in which salts are allowed to crystallize in the pores under
controlled conditions, furnish a useful method of investigating the properties of building materials.
Tests of this kind originated with Brard (Hericart de Thury14), who used a saturated solution of
sodium sulfate ... The effect of the crystallization of soluble salts in a particular material is due to
combination of circumstances which includes, among others, the size and distribution of the pore
spaces and the mechanical strength of the material. .. When an anhydrous salt changes to the
hydrated form with increase in volume, the bulk expansion is an adequate explanation of the origin
of the disrupting forces. For instance, sodium sulfate can exist in a hydrated form
(Na 2S0 4 •10H 2 0) which has a specific gravity of 1.49 and also in an anhydrous form (Na2S04)
with a specific gravity of 2.66. Obenniller15 has shown that the changes in temperature and relative
humidity which occur in the atmosphere are sufficient to cause repeated changes in the state of
hydration of sodium sulfate exposed to it.
''When a salt does not undergo changes in phase with consequent changes in volume, any thrust
developed by its growth might be attributed to axial growth of the crystals ... There is considerable
doubt as to whether or not the axial growth is actually associated with a force. Correns 16 concluded

139
Daneshlink.com
daneshlink.com
that the so-called crystallization forces, which cannot be explained by volume increase due to a
change in phase, may be explained on the basis that supersaturated solution is drawn into capillary
space between the crystals and the adjacent matter by surface tension forces."
Wellman and Wilson17 wrote: "The fragmentation of rocks by the crystallization of salts ... is
important in a restricted range of environments and produces distinctive topographic forms ...The
chemical free energy of a given mass of solid increases with its surface area. Therefore, any system
tends to reduce the area of its interfaces to a minimum, and in a system containing crystals in
equilibrium with a saturated solution larger crystals will grow at the expense of the smaller. It is less
obvious why, when the larger crystals entirely fill a pore space, they continue to grow against the
constraint imposed by the walls of the pore, expand the pore and fragment the rock." They cite
Lewis and Randall18 for the equation for the work required to be done during crystal growth at one
face of a crystal as equal to (P 1 - P jdV, where P 1 is the pressure in the liquid, P. the pressure in the
solid, and dV the increase in volume. They continue, still citing Lewis and Randall, "This must equal
the work done in extending the surface, which is equal to crdA, where cr is the interfacial tension
between the crystal face and its saturated solution and dA is the increment of volume. Since cr is
independent of V
dA
P1 - P = a -
s dV
Consider crystallization in a porous solid with large pores and small pores, both filled with the
saturated salt solution; let water evaporate and escape from the system, or let crystallization be
induced by a temperature change. First, the larger crystals in the large pores will grow at the expense
of smaller crystals in the small pores. Let the process continue until salt crystals completely fill the
=
large pores. Now, since P.- P 1 cr (dA/dV), for the crystals to grow down the capillary pores
would gready increase the area of the crystal, but only slighdy increase its volume. The crystal will
therefore grow in the large pore until the pressure builds up to such an extent that either mechanical.
fracture occurs or (P.- P 1) / cr becomes greater than the necessary dA/dV to make the crystal grow
down the capillary pore. Thus (for a given crystal and therefore for a given cr) whether or not
fracture occurs depends on smallness of the small pores and the value of cr compared with the
mechanical strength of the porous material. Hence a large rock pore will be enlarged provided that
the surface tension of the salt times the dA/ dV of the micropores is greater than the mechanical
strength of the rock. Thus for rocks of equal mechanical strength, those with large pores separated
from each other by microporous regions will be most liable to salt weathering...Although at present
salt weathering is destroying buildings, in the past it provided man with useful shelters.
CONCLUSIONS
From the matters considered in the preceding discussion, it is apparent to the author that, although
the physical and chemical effects of the phenomena associated with the formation of hydrated
sulfates in concrete, aggregate, paste, and mortar have long been studied, many fundamental aspects
of these phenomena are still poorly understood. The relation of the physical properties of the
material surrounding the locus at which the hydrated sulfate is produced to the amount of expansion
that results, the degree to which the expansion can be reduced by restraint, the force required to
prevent expansion or to prevent the reaction from taking place, the consequences of the restraining
force being uniaxial, biaxial, or triaxial rather than hydrostatic are all matters that appear to fall into
the category of those about which less is known than would be desired. Today, when tools for study
--``,```,`,`,`,,,,`,`````,`,,`,,-`-`,,`,,`,`,,`---

140
Daneshlink.com
daneshlink.com
are vastly improved, when our knowledge of the composition and constitution of the reacting
substances has been greatly increased, and when we can see the merging of several practical areas of
application of the knowledge to be obtained, appears to be the time when these gaps in knowledge
may be expected to be filled. I hope that this paper will indicate to workers concerned with chemical
attack on concrete, durability of aggregates, and the study of expansive cements that these areas of
study are merging and that progress in each will contribute to progress in all if the experiments are
carefully planned in the light of available knowledge.
REFERENCES
1) Mather, B.,1967, "Factors Which Influence the Deterioration of Concrete in Dams, and
Measures for Prevention of Deterioration," Transactions, Ninth International Congress on Lat;ge
Dams, Istanbul (Paris), V. 4, Q. 34, R. 24, pp. 419-433.
2) Mather, B., 1968, "Cracking Induced by Environmental Effects," Causes, Mechanism, and
Control of Cracking in Concrete, ACI SP-20, American Concrete Institute, Detroit, Mich., pp.
67-72.
3) Mather, B., 1966, "Effects of Seawater on Concrete," Symposium on Effects ofAggressive Fluids
on Concrete, Highway Research Record No. 113, pp. 33-42.
4) Mather, B., 1968a, "Field and laboratory Studies of the-Sulfate Resistance of Concrete,"
Peiformance of Concrete-Resistance of Concrete to Sulphate and Other Environmental Conditions-A
Symposium in Honor ofThorbergur Thorvaldson, Canadian Building Series, No. 2, University of
Toronto Press, pp. 66-76.
5) Hansen, W. C., and Offutt, J. S., 1962, Gypsum and Anf?ydrite in Portland Cement, U.S. Gypsum
Company, Chicago, 63 pp.
6) Bache, H.H., and I sen, J .C., 1968, "Modal Determination of Concrete Resistance to Popout
Formation," Journal of the American Concrete Institute, Proceedings, V. 65, June, pp. 445-450.
7) Aroni, S.; Polivka, M.; and Bresler, B., 1966, "Expansive Cements and Expanding Concrete,"
Report No. 66-7, Structures and Materials Research, Department of Civil Engineering,
University of California, Berkeley, Calif., 74 pp.
--``,```,`,`,`,,,,`,`````,`,,`,,-`-`,,`,,`,`,,`---
8) Michaelis, W., 1892, (tide not available). Tonindustrie-Zeitung, V. 16, p. 105.
9) Schaffer, R. J., 1932, "The Weathering of Natural Building Stones," Building Research Special
Report No. 18, Department of Scientific and Industrial Research, London, 149 pp.
10) Becker, G. P., and Day, A. L., 1905, "The Linear Force of Growing Crystals," Washington
Academy of Science, V. 7, p. 283.
11) Hansen, W. C., 1963, "Crystal Growth as a Source of Expansion in Portland-Cement

Concrete," Proceedings, American Sociery for Testing and Materials, V. 63, pp. 932-945.

141
Daneshlink.com
daneshlink.com
12) Taber, S., 1916, "The Growth of Crystals Under External Pressure," American Journal of
Science, V. XLI, p. 532.
13) Klein, A.; Karby, T.; and Polivka, M., 1961, "Properties of an Expansive Cement for
Chemical Prestressing," Proceedings, American Concrete Institute, V. 58, pp. 59-872.
14) DeThury, H., 1829, "On the Method Proposed by Mr. Brard for the Immediate Detection
of Stones Unable to Resist the Action of Frost," Annales de Chimie et de Prysique, V. 38, pp.
160-192.
15) Obermiller, J., 1927, "The Dehydration of Sodium Sulfate at Ordinary Temperatures," Zeit.
Angew., Chem., V. 40, No. 26, p. 755.
16) Correns, C. W., 1926, "Explanation of the So-Called Force of Crystallization," Sitzunggberichte
der preussichen Akademic der Wissenschqften, pp. 10-11, 81-88.
17) Wellman, H.W., and Wilson, A. T., 1965, "Salt Weathering, a Neglected Geological Erosive
Agent in Coastal and Arid Environments," Nature, V. 205, 13 Mar., No. 4976, pp. 1097-
1098.
18) Lewis, G. N., and Randall, M., 1961, Thermocfynamics, Second Edition, McGraw-Hill, New
York, N.Y.
--``,```,`,`,`,,,,`,`````,`,,`,,-`-`,,`,,`,`,,`---
142
Daneshlink.com
daneshlink.com
Reprinted from Fifth G4NMET/ACI International Conference Supplementary Papers, 2000, ©American Concrete Institute.
SULFATE ATTACK ON HYDRAULIC-CEMENT CONCRETE.
by Bryant Mather"
--``,```,`,`,`,,,,`,`````,`,,`,,-`-`,,`,,`,`,,`---
SYNOPSIS
Concrete has suffered damage as a consequence of the reaction, in the concrete, of sulfate ion and
the tticalcium-aluminate (C3A) constituent of portand cement or hydration products of C 3A to form
ettringite. This damage, which was first reported in 1892, can be avoided if the concrete is protected
from potentially harmful amounts of sulfate ion or potentially harmful amounts of C 3A. Other
physical and physical-chemical phenomena that have had or could have had or might have done
damage to concrete have been described in the literature.
Keywords
Concrete; sulfate attack; tricalcium aluminate; ettringite.
INTRODUCTION
In 1969 (M:ather 1969) at the request of Dr. Wolfgang Czernin, I prepared and presented at Dr.
Valenta's meeting in Prague, a summary of what I thought I knew about sulfates and concrete. It
included material previously summarized in Mather 1966, 196 7, 1968, and 1968a, as well as by many
others in many other places. Since then, much has been done in many places, but it is unclear to me
whether or not most of the questions to which I did not have good answers in 1969 are issues that
have as yet been adequately resolved; indeed, in my current speculations, it seems that there are
more new unanswered questions than there are answers to questions, old or new.
In Mather (1966) on the effects of seawater on concrete and in Mather (1968a) on studies of sulfate
resistance, the reactions between externally supplied magnesium sulfate and the constituents of the
cement were examined and the volumetric effects as given by Hansen and Offutt (1962) were
described. The reaction of tricalcium aluminate with gypsum and water to form ettringite was given
as follows:
3CaO•Al2 0 3 + 3CaSO 4•2H 2 0+25H 2 0~
88.8 222.3 450
3CaO•Al20 3·3CaSO4•31H20
714.7
This reaction involves a net reduction in volume if one assumes that the water is already taking up
space in the system, but the relative volumes of anhydrous tticalcium aluminate (C3A) to that of the
ettringite formed from it are as 1:8. The observed expansions in sulfate reaction and sulfate-reaction
expansions cannot be correlated precisely with an eightfold increase in volume of the C 3A. Indeed,
cements of the same assumed C3A content do not yield test specimens of the same increase in
length when exposed to sulfate in excess and time for the reaction to go to completion. Mather
• Prepared at the request of Professor Kumar Mehta for presentation in Barcelona, Spain, June 1999.
•• Former Director, Structures Laboratory, U.S. Army Engineer Research and Development Center, 3909 Halls
Ferry Road, Vicksburg, MS 39180-6199, USA.

143
Daneshlink.com
daneshlink.com
(1968a) presented contrasting results obtained from tests of two cements from the PCA Long-Time
Study group. The data are shown in the following table.
Cement
No.13 No.15
Calculated C3A, % 10.3 12.5
Calculated C3S (tricalcium silicate),% 56.9 67.7
3days 1800 3315
Compressive strength,
7 days 2465 4650
ps1
28 days 3875 6140
Heat of hydration, cal/g, 28 days 85 116
Fineness, air permeability, cm2 / g 3485 3340
Composition by microscope, C3A, % 1.6 8.6
Composition by microscope, dark prismatic, % 1.9 1.2
Composition by microscope, "glass," % 8.2 0.3
Composition by microscope, Total 11.7 10.1
0.132
Expansion, %, 1 yr 0.408
(0.143)*
*Rerun.
The time versus length-change curve for cement 13 showed a slow but steadily increasing rate, while
that for cement 15 showed a rapid early rate, reaching a maximum increase in length by 56 days and
little further change to an age of one year. The maturity of pastes made with cement 15, as measured
by heat evolution or by strength development, was as great at 7 or 28 days as was that with cement
13 at one year. The C 3A in cement 13 was predominantly present as the material classified by
microscope examination as glass, while that in cement 15 was present as material recognized as C 3A
crystals. The specimens made with cement 15 expanded rapidly and matured rapidly; those made
with cement 13 expanded and matured slowly.
The procedures used involve measuring the length changes caused by internal expansions resulting
from chemical processes that lead to the formation of hydrated sulfates. When an expansion occurs,
it is because the hydrated sulfates that are formed occupy a larger volume than was available and
hence exert stress on their surroundings. The stress so exerted produces elastic and inelastic strains
of magnitudes depending on the level of stress, the mechanical properties of the system, and the
geometry of the specimen. The relevant mechanical properties include modulus of elasticity, creep,
porosity, and strength. All of the hydrated sulfates of interest are crystalline to some degree;
consequently, their formation may be referred to as crystallization, and an increase in the amount
that is present may be referred to as crystal growth.' The stresses accompanying such formation have
been referred to as resulting from crystal growth even as the stresses accompanying the freezing of
--``,```,`,`,`,,,,`,`````,`,,`,,-`-`,,`,,`,`,,`---
water have been referred to as resulting from ice crystal growth. I suggest that the fact that the
substances are crystalline and that an increase in the amount present involves crystal growth are of
minor significance to an understanding of the mechanisms by which such formation produces
stresses which induce the strains that are observed as expansions and cracking.
If the space occupied in an aggregate particle or in a sample of cement paste, mortar, or concrete by
a quantity of a hydrated sulfate after equilibration of strain were occupied by a liquid or gas that had
"Crystallization" refers to formation of crystals from a solution or from a melt; "crystal growth" refers to such
crystals increasing in size. I am not sure this distinction is useful for the phenomena discussed in this paper.

144
Daneshlink.com
daneshlink.com
exerted the same stress under the same conditions of confinement, the final volumetric relations of
the system in which it existed would be the same. The relevant phenomenology of the sulfate-
soundness test of aggregates, the various sulfate susceptibility tests for pastes, mortars, and
concretes, the various expansivity tests for expansive cements, tests for resistance to freezing and
thawing, tests for the potential expansivity caused by chemical reactions of alkalies and aggregates,
tests for oxidation or hydration of aggregate or cement constituents or embedded corrodible metals
does not primarily depend on whether the products formed are crystalline or noncrystalline. The
relevant phenomenology concerns the specific volume of products formed, the amount of products
formed, the compressibility of the products formed, the restraint imposed by the surroundings, the
rate of formation of the products, and the thermodynamics of the reactions.
The best known phenomenon by which an internal expansion can be produced in an aggregate
particle or in a sample of cement paste, mortar, or concrete is that involving the freezing of water to
produce ice. Ice occupies about 9% more volume than the water that was frozen in its formation. It
has been shown that if liquid water completely fills the space in which it exists, is completely
restrained from increasing in volume, and the temperature of the water can be lowered as far as may
be desired, a restraint by the surroundings of 31,000-psi hydrostatic force will be imposed on the
water to keep it from undergoing the change into ice. These data also indicate that the maximum
internal force that can be exerted by water trying to change into ice as the temperature is reduced is
31,000 psi. This is a stress that, if exerted, will prevent ice crystals from forming; it is not the force
--``,```,`,`,`,,,,`,`````,`,,`,,-`-`,,`,,`,`,,`---
exerted on the surroundings by the growth of the ice crystals.
In the case of one of the better studied hydrated sulfates of interest, ettringite, calcium aluminum
sulfate hydrate (C 6AS 3 H 3 ~, it has been observed that, if one takes the anhydrous analogue of this
material (C4A 3S) plus calcium sulfate and calcium oxide and packs it tighdy in a container capable of
providing a very high level of volumetric restraint and then allows it to hydrate in place, the strains
observed in the container can reach levels indicative of stresses of the order of 50,000 psi. The value
for this stress has been computed from measured strains in an experiment (35,000 psi) (Mehta and
Klein 1966) and been calculated from thermodynamic free-energy considerations (49,000 psi) (ACI
223 1971).
All of the systems of interest to students and users of concrete have elastic properties, inelastic
properties, and tensile strength. In most such systems the magnitude of the modulus of elasticity,
ability to creep, and tensile strength varies with time. Consequendy, in most such systems the
volumetric response of a sample of such a system to the development within it, at a given time in its
history, of a unit of product, the formation of which exerts a given stress on its surroundings, will be
different from the response exerted at any other time in its history, regardless of whether the internal
stress of given magnitude results from one or another of a large number of mechanisms. Similarly,
the response of the system to the generation of such an internal stress, at a given time, will vary
depending on the locus of this stress relative to the geometry of the sample of the system that is
under study and hence the restraint supplied by the material of the system at the locus at which the
stress develops. If the sample of the system is a sphere and the locus of the formation of the
product that, when formed, occupies a larger volume than is available is at the center of the sphere,
the observed expansion of the sample will be a uniform increase in all diameters of the sphere,
provided the magnitude is within the amount that can be accommodated by elastic and creep strains.
If the locus is sufficiendy near the surface of the sample, regardless of the geometry of the sample,

145
Daneshlink.com
daneshlink.com
the local internal expansion will, without any greater force being involved, create a popout and
crater.
Bache and lsen (1968) have reported the results of experiments involving the production of internal
hydrostatic pressures of known magnitudes at points at different depths from the surface of samples
of mortars having different mechanical properties. Their results show that, as the locus of internally
generated hydrostatic stress is placed deeper beneath the surface of the samples, the higher must be
the stress in order for a popout and crater to form. The relation that they found was as follows:
_!_~to.!:
O't d
where P = hydrostatic pressure required to form a popout
0'1 = tensile strength of the system
h = distance from the surface to the locus of stress
d = diameter of the sphere exerting the hydrostatic stress
Aroni et al. (1966) noted that it was Michaelis (1892) who first attributed the disruption of concrete,
when attacked by aggressive waters containing sulfates, to the reaction between C3A and sulfates to
form ettringite (C 6AS 3H 32). They note that the useful feature of this reaction to the application of
expanding cement to chemical prestressing is the high energy of the expansion; i.e., its ability to
perform work and expand under restraint. They discuss the various papers that have attributed
internal expansion to the monosulfate (C 4ASHd rather than, or in addition to, the trisulfate
(ettringite) and conclude that the weight of the evidence favors associating it with the latter. They
note that expansive cements, based on the work of Klein, derive their expansion from the hydration
of C4A 3S to form (C6AS 3H 3z).
Aroni et al. (1966) also noted that there is litde in the literature on the basic phenomena of crystal
growth under pressure. As did Schaffer (1932), they refer to the work of Becker and Day (1905), but
do not appear to have rejected their hypothesis. They refer to the conclusions of Hansen (1963) and
Taber (1916) and ask what the applicability of these conclusions is to expanding concrete. They
consider two cylinders of expanding concrete of equal expansive potential, one free to expand in all
directions, the other restrained longitudinally. They say one would expect an "inverse" relation
between longitudinal and transverse strain-a longitudinal shortening should, by the Poisson effect,
create a transverse elongation. They quote Klein et al. (1961), who referred to evidence that in small
specimens with symmetrical restraint, the expansive movements in all directions are substantially
equal per unit length, even under uniaxial restraint, where the cross section normal to the axis of
restraint is under adequate compression. Aroni et al. cite a number of instances that indicate a direct
relation including comparisons of specimens with uniaxial Oongitudinal) versus triaxial restraint, in
which those with only uniaxial restraint showed an increase in longitudinal expansion of up to 25%.
Hansen (1963) reviewed the literature relating to the phenomenon of crystal growth as a source of
expansion in pordand-cement concrete. He began his review with the statement that "pordand-
cement pastes expand during setting and hardening, whereas they would be expected to contract
because the volumes of the hydrated phases are less than the sum of the volumes of the water and
the reactants forming these phases." He states expansion to be the volume change that occurs on
setting and hardening, since other writers imply, quite as blundy, that there is a contraction or
shrinkage on setting. He states that the expansion has been explained on the basis that the hydrated
compounds form by direct reaction with water from the unhydrated solids. As do others, he refers

146 --``,```,`,`,`,,,,`,`````,`,,`,,-`-`,,`,,`,`,,`---
Daneshlink.com
daneshlink.com
to the work of Becker and Day (1905) and attributes the observed ability of a growing crystal to lift a
weight as a consequence of the existence of a f:tlm of solution beneath the growing crystal, the
concentration differences in the solution, and the relation of solubility of the crystal to the pressure
on it. He cites the conclusions of Taber (1916), both with regard to these circumstances and with
regard to the relations of crystals growing in porous media having a range of pore sizes. Taber
showed that crystals growing in larger pores will receive material from the solution in the smaller
pores and grow at the interface between liquid and crystal. Hansen reviews the circumstances
relating to the growth of a group of crystals nucleating from a saturated solution and concludes that,
if none of the crystals are under external pressure, there could develop a single crystal without
pressure being exerted. He concludes that theories that attribute expansions in concrete to crystals
growing from solution against one another do not appear to be realistic. He describes two
conditions under which a crystal may grow and exert pressure on materials opposing its growth.
One is the condition in which a weighted crystal rests on the floor of a container on a film of
solution that is supersaturated with respect to the bottom face of the crystal. The other is the
condition in which a single crystal fills a container that opens into another container filled with a
solution supersaturated with respect to the film of solution supporting the crystal.
Schaffer (1932) wrote, "Crystallisation tests, in which salts are allowed to crystallise in the pores
under controlled conditions, furnish a useful method of investigating the properties of building
materials. Tests of this kind originated with Brard (Hericart de Thury 1829), who used a saturated
solution of sodium sulphate ... the effect of the crystallisation of soluble salts in a particular material
is due to a combination of circumstances which includes, among others, the size and distribution of
the pore spaces and the mechanical strength of the material.... When an anhydrous salt changes to
the hydrated form with increase in volume, the bulk expansion is an adequate explanation of the
origin of the disrupting forces. For instance, ... sodium sulphate can exist in a hydrated form
(Na2 S0 4•10H2 0) which has a specific gravity of 1.49 and also in an anhydrous form (Na2S04) with a
specific gravity of 2.66."
Wellman and Wilson (1965) wrote: "The fragmentation of rocks by the crystallization of salts ... is
important in a restricted range of environments and produces distinctive topographic forms .... The
chemical free energy of a given mass of solid increases with its surface area. Therefore, any system
tends to reduce the area of its interfaces to a minimum, and in a system containing crystals in
--``,```,`,`,`,,,,`,`````,`,,`,,-`-`,,`,,`,`,,`---
equilibrium with a saturated solution larger crystals will grow at the expense of the smaller. It is less
obvious why, when the larger crystals entirely fill a pore space, they continue to grow against the
constraint imposed by the walls of the pore, expand the pore and fragment the rock."
Most of the foregoing text is only lightly revised from Mather (1969). Since it was Czernin who
invited me to the meeting in Prague in 1969, I was happy to obtain a copy of the English translation
of his book (Czernin 1980). In the light of current interest in so-called "delayed ettringite formation
(DEF)," I think it of interest that in 1980 he made the following statement (p 125): "Quite
irrespective of whether too much gypsum is incorporated in the cement during grinding, whether
the gypsum is contained in the aggregates or whether sulphate-bearing water penetrates the
hardened cement paste, in every case the final damage to concrete is due to the formation of
voluminous water-rich calcium sulphoaluminates."
I have read a fair amount of the literature on DEF and have thus far been unable to establish criteria
for recognizing it as anything new and different --except to those who have not dealt with sulfate
attack in any detail previously. The deposition of secondary ettringite in cracks and voids in concrete
147
Daneshlink.com
daneshlink.com
--``,```,`,`,`,,,,`,`````,`,,`,,-`-`,,`,,`,`,,`---
has nothing to do with sulfate attack except that in some cases some of the ettringite that is
dissolved may be ettringite that resulted from sulfate attack. Generally, the ettringite filling such
cracks and voids is primary ettringite produced intentionally and desirably as part of the early
hydration reactions of cement, and its subsequent dissolution, transport, and precipitation in cracks
and voids in concrete is completely harmless. There has been informal discussion of ettringite-filled
entrained-air voids being indicative of how "sulfate attack" can damage concrete by damaging the
air-void system the concrete needs to retain its frost resistance. Obviously ettringite-filled entrained-
air voids are simply secondary ettringite deposition in air voids that previously were compromised by
some penetration by microcracking. No pore fluid can be present in an uncompromised entrained-
air void in concrete that is thawed. All functioning entrained-air voids must empty on thawing so
they will be ready to function on the next freezing cycle-which, in our freezing-and-thawing testing
facility, is an hour from the time of thawing. If there is no pore fluid in an air void when the
concrete is thawed, nothing can deposit itself as a secondary product upon evaporation of the pore
fluid.
There is discussion in 1999 of the precautions that need to be taken to protect concrete against
sulfate attack when gypsum or other forms of calcium sulfate (anhydrite, hemihydrate) are in soil in
contact with the concrete. In this connection, Czernin (1980) said:
''Waters containing gypsum, mainly ground-waters in gypsum-containing soils, occur

comparatively frequently. The range of permissible gypsum concentrations depends
on whether the aggressive water is still, running or acts under a hydrostatic pressure
on the concrete. Special methods of protecting concrete against corrosion may
become necessary if a still water of this kind contains more than 400 mg S04 /litre,
while a running water could prove dangerous while containing not more than 200
mg S004 /litre.
"Sulphate attack on concrete is, however, not limited to gypsum only; every sulphate
of sufficient solubility in water will, when in contact with calcium hydroxide released
by set cement paste, react to give initially gypsum and finally calcium
sulphoaluminate.
"Other reactions can also lead to sulphate corrosion of concrete. Hydrogen sulphide,
for example, which is a weakly acidic gas produced in decaying organic matter, can
be oxidized by atmospheric oxygen to sulphuric acid. Sulphate attack of this kind is
sometimes observed in concrete sewers or similar effluent carriers."
As noted in the foregoing quotation from Czernin, the obvious initial reaction product of sulfate ion
entering concrete is gypsum formed by the reaction with the calcium ion with which the pore fluid
of concrete is saturated. Since this is a through-solution reaction (similar to the hydration of
portland cement), it causes no internal expansion; i.e., when there is no more initially mixing-water-
filled space to deposit gypsum in, no more gypsum forms; just as cement hydration ceases when
there is no more initially mixing-water-filled space in which to deposit cement hydration products.
Recent work at Menashi Cohen's laboratory at Purdue has suggested that, in sulfate attack, quite a
lot of gypsum forms without any related expansion, but, under some conditions, later gypsum
formation may be associated with expansion (fian 1999).

148
Daneshlink.com
daneshlink.com
Two other bits that I did not know in 1969 are: (1) the Mg or Na sulfate dehydration-rehydration
procedure for damaging porous brittle substances developed by Brard and used in ASTM C 88 also
happens in nature. Binda and Baronio (1987) showed that in Milan conditions for changing between
Na2S04 and Na2S0 4•10H20 occur in every month of the year at the rate of as few as 4 to as many as
24 per month. Also, in 1969 I did not know that salt weathering was, in fact, chemical, not physical.
Young (1987) concluded that salt weathering, previously attributed to crystallization pressure or salt
hydration, was caused by enhancement of the rate of silica solution by salts in solution, especially
sodium chloride. In 1969, I was mystified at the concept that sodium chloride (NaCl) solution in
pores of concrete (or rock) could do damage to the solid by losing its water on evaporation and
going back into solution on rewetting since there are no hydrates of NaCl and, hence, the
mechanism could not be the dehydration-rehydration mechanism of the Brard Test (ASTM C 88)
which, as previously noted, explained the widespread misconception that "crystal growth pressure"
was what went on there. I was thus very happy to read Young (1987) that did away with the need for
a physical explanation of this anomaly.
My most recent contribution is Hime and Mather (1999). In that paper, my coauthor, a chemist,
reviews 10 relevant equations including those relating to the hydration and dehydration of sodium
sulfate and magnesium sulfate. We conclude that more research and analysis is needed, a conclusion
that echoes that by Cohen and Mather (1991).
CONCLUSIONS
From the matters considered in the preceding discussion, it is apparent to me that, although the
physical and chemical effects of the phenomena associated with the formation of hydrated sulfates
in concrete, aggregate, paste, and mortar have long been studied, many fundamental aspects of these
phenomena are still poorly understood. The relation of the physical properties of the material
surrounding the locus at which the hydrated sulfate is produced to the amount of expansion that
results, the degree to which the expansion can be reduced by restraint, the force required to prevent
expansion or to prevent the reaction from taking place, the consequences of the restraining force
being uniaxial, biaxial, or triaxial rather than hydrostatic are all matters that appear to fall into the
category of those about which less is known than would be desired. Today, when tools for study are
vastly improved, when our knowledge of the composition and constitution of the reacting
substances has been greatly increased, and when we can see the merging of several practical areas of
application of the knowledge to be obtained, appears to be the time when these gaps in knowledge
may be expected to be filled. I hope that this paper will indicate to workers concerned with chemical
attack on concrete, durability of aggregates, and the study of expansive cements that these areas of
study are merging and that progress in each will contribute to progress in all if the experiments are
carefully planned in the light of available knowledge. These concluding remarks are unchanged from
Mather (1969).
REFERENCES
ACI Committee 223, 1971, Committee Closure to Discussion by Bryant Mather and P. K. Mehta of
"Expansive Cement Concretes-Present State of Knowledge," ACI Journal, Proceedings, V. 68, pp.
294-296.
--``,```,`,`,`,,,,`,`````,`,,`,,-`-`,,`,,`,`,,`---

149
Daneshlink.com
daneshlink.com
Aroni, S.; Polivka, N.; and Bresler, B., 1966, Expansive Cements and expanding Concrete, Report No. 66-
7, Structures and Materials Research, Department of Civil Engineering, University of California,
Berkeley, Calif., 74 pp.
Bache, H. H., and lsen, J. C., 1968, "Modal Determination of Concrete Resistance to Popout
Formation," Journal of the American Concrete Institute, Proceedings, V. 65, June, pp. 445-450.
Becker, G. F., and Day, A. L., 1905, "The Linear Force of Growing Crystals," Washington Academy
of Science, V. 7, p. 283.
Binda, L., and Baronio, G., 1987, "Mechanisms of Masonry Decay Due to Salt Crystallization,"
Durability ifBuildingMaterials, V. 4, pp. 227-240, Elsevier, Amsterdam.
Cohen, M.D., and Mather, B., 1991, "Sulfate Attack on Concrete-Research Needs," ACI Materials
Journal, V. 88, No.1, pp. 62-69.
Czernin, W., 1980, Cement Chemistry and PI?Jsics for Civil Engineers, 2nd English Edition, Bauverlag
GMBH, Wiesbaden, 196 pp.
De Thury, H., 1829, "On the Method Proposed by Mr. Brard for the Immediate Detection of
--``,```,`,`,`,,,,`,`````,`,,`,,-`-`,,`,,`,`,,`---
Stones Unable to Resist the Action of Frost," Annales de Chimie et de Pf?ysique, V. 38, pp. 160-192.
Hansen, W. C., 1963, "Crystal Growth as a Source of Expansion in Portland-Cement Concrete,"
Proceedings, American Society for Testing and Materials, V. 63, pp. 932-945.
Hansen, W. C., and Offutt,). S., 1962, Gypsum andAnf?ydrite in Portland Cement, U.S. Gypsum
Company, Chicago, Ill., 63 pp.
Hime, W. G., and Mather, B., 1999, "'Sulfate Attack' or Is It?," Cement and Concrete Research, V. 29,
pp. 789-791.
Klein, A.; Karby, T.; and Polivka, M., 1961, "Properties of an Expansive Cement for Chemical
Prestressing," Proceedings, American Concrete Institute, V. 58, pp. 59-82.
Mather, B., 1966, "Effects of Seawater on Concrete," Symposium on Effects ofAggressive Fluids on
Concrete, Highway Research Record No. 113, pp. 33-42.
Mather, B., 1967, "Factors Which Influence the Deterioration of Concrete in Dams and Measures
for Prevention of Deterioration," Transactions, Ninth International Congress on Lat;ge Dams, Istanbul
(Paris), V. 4, Q.34, R.24, pp. 419-433.
Mather, B., 1968, "Cracking Induced by Environmental Effects," Causes, Mechanism, and Control of
Cracking in Concrete,ACI SP-20, pp. 67-72,
Mather, B., 1968a, "Field and Laboratory Studies of the Sulfate Resistance of Concrete," Performance
of Concrete-Resistance of Concrete to Sulphate and 0 ther Environmental Conditions-A Symposium in Honor if
Thorbet;gurThoroaldson, Canadian Building Series No.2, University ofToronto Press, pp. 66-76.
150
Daneshlink.com
daneshlink.com
Mather, B., 1969, "Sulfate Soundness, Sulfate Attack, and Expansive Cement in Concrete," in
Proceedings, RILEM International Symposium on Concrete Durabiliry, Technical University, Prague,
Czechoslovakia, pp. C-209 to C-220.
Mehta, P. K., and Klein, A., 1966, "Investigations on the Hydration Products in the System
4CaO·S03·3Al20•S0 3•CaS04•H2 0" in Symposium on Structure of Portland Cement Paste and Concrete,
Highway Research Board, Special Report 90, pp. 328-352.
Michaelis, W., 1892, (fide not available), Tonindustrie-Zeritung, V.16, p. 105.
Schaffer, R. J., 1932, The Weathering ofNatural Building Stones, Building Research Special Report No.
18, Department of Scientific and Industrial Research, London, 149 pp.
Taber, S., 1916, "The Growth of Crystals Under External Pressure," American Journal of Science, V.
XLI, p. 532.
Tian, B., 1999, "Damage to Concrete During Sulfate Attack Caused by Gypsum Formation," Cement
and Concrete Research, V. 29
Wellman, H. W., and Wilson, A. T., 1965, "Salt Weathering, A Neglected Geological Erosive Agent
in Coastal and Arid Environments," Nature, V. 205, 13 Mar., No. 4976, pp. 1097-1098.
Young, A. R. M .. 1987, "Salt as an Agent in the Development of Cavernous Weathering," Geology, V.

15, pp. 962-966.
--``,```,`,`,`,,,,`,`````,`,,`,,-`-`,,`,,`,`,,`---
151
Daneshlink.com
daneshlink.com
--``,```,`,`,`,,,,`,`````,`,,`,,-`-`,,`,,`,`,,`---

Daneshlink.com
daneshlink.com
PART V- PETROGRAPHIC EXAMINATION OF AGGREGATES AND

CONCRETE
Petrographic examination of aggregates and concrete provides the possibility of a much more
detailed assessment of concrete quality than can be obtained by any other method. It involves
preparing very thin sections from concrete samples usually cored from existing structures.
Petrographic examination can provide information on composition of concrete, including mixture

proportions, free water, and water-cementitious material ratios (w/ em). It can provide other
information, including the presence and position of reinforcement, the extent to which this
reinforcement is corroded, the nature of the external surface of concrete, the feature and distinction
of macro- and microcracks, the distribution and size range, type of coarse and fine aggregates, and
the type and condition of the cement paste. It can also detect evidence of deleterious products
affecting the concrete.
The Mathers published a number of papers dealing with petrographic studies. They described the
merits of petrographic examination of concrete aggregate materials and the results of such
examinations. In their 1950 paper "Method of Petrographic Examination of Aggregates for
Concrete," the Mathers provided detailed information that they hoped would help clarify the
concept of what is meant by petrographic examination of aggregates and contributed to a greater
uniformity in making such examinations.
Their comprehensive description encompassing sampling, purposes of the examination,

classification of materials present in samples of aggregate, apparatus, and the reporting of the
findings with examples is clearly presented in the above paper.
Katharine Mather did extensive research work in the application of the microscope to the
investigation of portland cement and other cementitious or pozzolanic materials used as a
supplement, or partial substitutes, for pordand cement. These research efforts are discussed in
"Applications of Light Microscopy in Concrete Research," published in 1953.
--``,```,`,`,`,,,,`,`````,`,,`,,-`-`,,`,,`,`,,`---
Even though Katharine made many contributions with her research, she concluded that much
remained to be done in terms of observation and description of microstructure, relation of
microstructure to larger structures, and detailed interpretation of the observed phenomena.

153
Daneshlink.com
daneshlink.com
--``,```,`,`,`,,,,`,`````,`,,`,,-`-`,,`,,`,`,,`---

Daneshlink.com
daneshlink.com
Reprinted, with permission, from Proceedings, ASTM V 50, ©ASTI\1 International, 1950, 100 Barr Harbor Drive, West
Conshohocken, PA 19428.
METHOD OF PETROGRAPHIC EXAMINATION OF AGGREGATES FOR

12
CONCRETE '
Katharine Mather3 and Bryant Mather3
SYNOPSIS
A detailed description of the method for petrographic examination of aggregates for concrete as it
has been developed for use by the Concrete Laboratories of the Corps of Engineers is presented.
The differences in procedure which depend upon the nature of the sample submitted and the
purposes of the examination are described. Suggestions are made concerning the features to be
delineated, the amount of material to be examined, and the organization and presentation of the
results of the examination. It is hoped that the information presented will serve to clarify the
--``,```,`,`,`,,,,`,`````,`,,`,,-`-`,,`,,`,`,,`---
concept of what is meant by petrographic examination of aggregates and, perhaps, ultimately,

contribute to greater uniformity in the making of such examinations.
There does not appear to have been a previous presentation of a outline of the method by which a
qualified petrographer may apply the knowledge and techniques of geology to the examination of
various types and conditions of material that are considered for use as aggregates for concrete. There
are an increasing number of engineering organizations that are undertaking the application of
petrographic methods to the study of concrete aggregate, an increasing number of petrographers
who are being confronted with the need to examine concrete aggregates, and an increasing number
of engineers who are called upon to evaluate petrographic data. It is therefore felt desirable to
present a detailed statement of the method for petrographic examination of aggregates for concrete
as it has been developed for use by the seven concrete laboratories of the Corps of Engineers.
In a previous paper4 the use of petrographic methods for the identification of reactive constituents
in concrete aggregate was discussed. The merits of petrographic examination of concrete aggregate
materials and the results of such examinations have been dealt with in a number of papers, many of
which are included in this bibliography.
The method as set forth here represents the most recent product of work begun more than a decade
ago by Elliot P. Rexford, the first concrete laboratory petrographer of the Corps of Engineers,
whose last paper5 gives a valuable summation of his long experience. To him and to the other
petrographers with whom the authors have been associated, appreciation is expressed for much
1
Presented at the Fifty-third Annual Meeting of the Society,June 26-30, 1950.
2 Approved for presentation and publication by the Office, Chief of Engineers, U.S. Army.
3 Head, Petrography Section and Chief, Special Investigations Branch, respectively, Concrete Research Division,
Waterways Experiment Station, Jackson, Miss.
4 B. Mather, "Petrographic Identification of Reactive Constituents in Concrete Aggregates," Proceedings, American Society
for Testing and Materials, V. 48, pp. 1120-1127 (1948).

5
E. P. Rexford, "Some Factors in the Selection and Testing of Concrete Aggregates for Large Structures," Transactions,
American Institute Mining and Metallurgical Engineers, V. 187, No. 3, pp. 39 5-405 (19 50). (Technical Publication 2837
H).

155
Daneshlink.com
daneshlink.com
assistance; they include: W. L Donn, Rosalind Ravitch, George Townsend, Louis Moyd, Pauline
Moyd,Jeanne Weiser, 0. G. Alessio, N. B. Dodge, R. A. Dunkerley, and R. V. Tye.
This method has been editorially modified from that published as CRD-C 127-50 in the Handbook
for Concrete and Cement issued by the Waterways Experiment Station.
EXAMINATION AND CLASSIFICATION
This method outlines the procedures to be employed in the petrographic examination of samples
representative of materials proposed for use as aggregates in concrete. The scope of such an
examination may be limited as in the case of preliminary examinations, or complete, as in the case of
final examinations.
Preliminary Examination
The preliminary examination is designed to assist the field geologist in judging which of apparently
suitable aggregate sources contain material of sufficiently promising quality to warrant further and
more complete investigation. Such preliminary examinations should be used as a part of the process
of elimination in the survey of economically available aggregate sources when other evidence is not
available as a basis for sound initial judgment of quality. The value of a preliminary petrographic
examination will be dependent largely upon the adequacy of the samples provided, upon the data on
the deposit (amount of material, predictable difficulties in working, etc.) which are provided with the
samples, and on the proposed use of the material. The results of such an examination should not be
used alone as a basis for final acceptance of a material.
Complete Examination
The complete petrographic examination is designed to constitute an integral part of the
comprehensive study of the suitability of an aggregate material for the use for which it is proposed.
The value of such an examination will be greatly influenced by the adequacy of the procedures used
in obtaining the samples which are submitted for examination.
The specific procedures employed in petrographic examination of any sample of aggregate material
depend to a large extent both on the purpose of the examination and on the nature of the sample
submitted. The purposes to be served by petrographic examination are as follows:
Properties
The delineation of all physical and chemical properties that may be observed by petrographic means
--``,```,`,`,`,,,,`,`````,`,,`,,-`-`,,`,,`,`,,`---
which bear on the quality of the materials for their intended use.
Classification
The description and classification of materials present in samples of aggregate will logically include
identification of the rocks in the sample. Identification is usually a necessary step towards
recognition of the properties which may be expected to influence the behavior of the materials in
their intended use. It is not an end in itself. Rock of any type may perform well or poorly as concrete
aggregate depending on the physical conditions of the aggregate, and on its physical and chemical
properties as compared to the matrix in which it may be placed.
The data developed by a petrographic examination can provide a basis for predicting quality, provide
an explanation of the results obtained from performance tests, and give information on factors not
156
Daneshlink.com
daneshlink.com
at present capable of being investigated by performance tests. The usefulness of any petrographic
examination to the agency requesting it will depend largely on three factors:
1) The representativeness of the samples provided.
2) The information accompanying the samples, which should include the location, type of
sampling, extent of deposit (if known), name of geologic formation (if known).
3) The petrographer's degree of understanding of the information needed by the agency
requesting the examination. It should be the responsibility of the agency requesting the
examination to make its needs clearly and specifically known to the laboratory. It should be
the responsibility of the petrographer to understand and make use of the information
available to him. This information will frequently include the results of elementary tests on
the sample, chemical tests of potential reactivity, and freezing and thawing of concrete
containing the aggregates under test.
APPARATUS AND SUPPLIES
Specimen Preparation
The apparatus and Supplies consist of:
1) Diamond saw, preferably with 14-in. blade,
2) Horizontal grinding wheel, preferably 16 in. in diameter,
--``,```,`,`,`,,,,`,`````,`,,`,,-`-`,,`,,`,`,,`---
3) Polishing wheel, preferably 8 to 12 in. in diameter,

4) Abrasives: silicon carbide grit No. 100, No. 240, FFF; emery such as American Optical
Co. M-303, and M-304,
5) Prospector's pick,
6) Microscope slides, clear, noncorrosive, 25 by 45 mm. in size,
7) Canada balsam, neutral, in xylol,
8) Xylol,
9) Glycol phthalate,
10) Small laboratory oven,
11) Plate-glass squares about 12 in. on an edge for thin section grinding,
12) Thin-section holder,
13) Jones rifle sampler with pans,
14) Micro cover glasses, noncorrosive, square 12-18 mm, 25 mm, etc., and Plattner mortar.
Specimen Examination
The apparatus and supplies consist of:
1) Polarizing microscope with mechanical stage; low-, medium-, high-power objectives,
objective centering devices; eyepieces of various powers; full- and quarter-wave
compensators; quartz wedge;
2) Microscopic lamps;
3) Stereoscopic microscope with paired objectives and paired oculars to give final
magnifications from about 6 X to about 60 X;
4) Magnet, preferably alnico, or an electromagnet;
5) Needleholder and points;
6) Dropping bottle, 60 ml. capacity;
7) Petri culture dishes;
8) Smooth straight-pointed forceps;
9) Lens paper;
157
Daneshlink.com
daneshlink.com
10) Immersion media, n = 1.410 ton= 1.785 in steps of0.005;

11) Abbe refractomer;
12) Sodium arc lamp;
13) Zirconium arc lamp;
14) Wratten light fllters;
15) Counter;
16) Photomicrographic camera and accessories
Examples of the above-listed items are shown in Figs. 1 to 8.
Note-The apparatus and supplies listed above and illustrated in Figs. 1 to 8 comprise a
recommended selection. It is not intended that the specific items listed and illustrated must
necessarily be provided. It is however regarded as essential that an adequate selection be provided to
enable the use of all of the procedures described in this method for the examination of the particular
types of samples that are to be studied.
Samples
Samples for petrographic examination should be taken by or under the direct supervisions of a
geologist familiar with the requirements of aggregates for concrete and in general following the
requirements of A.S.T.M. Methods of Sampling Stone Slag, Gravel, Sand, and Stone Block for Use
as Highway Materials (D 75-48). 6 The exact location from which the sample was taken, the geology
of the site, and other pertinent data should be submitted with the sample.
Undeveloped quarries should be samples by means of cores, drilled through the entire depth expected
to be productive of sound material. Drilling of such cores should be in a direction which is
essentially normal to the dominant structural feature of the rock. Massive material may be sampled
by "NX" cores. Thinly bedded or complex material should be represented by cores not less than 4
in. in diameter. There should be an adequate number of cores to cover the limits of the deposit
proposed for the work under consideration. The entire footage of the recovered core should be
included in the sample and accurate data given as to elevations, depths, and core losses.
--``,```,`,`,`,,,,`,`````,`,,`,,-`-`,,`,,`,`,,`---
Operating quarries and operating sand and gravel deposits in which stock piles of the material produced are
available, should be represented by not less than 100 lb. or 300 pieces, whichever is larger, of each
size of material to be examined. Samples from stock piles should be composed of representative
portions of larger samples collected with due consideration given to segregation in the piles.
Exposedfaces of non-producing quarries, where stock piles of processed material are not available, should
be represented by not less than 4lb. from each distinctive stratum or bed with no piece weighing
less than 1 lb., or by drilled core as described in the first paragraph of this section.
Undeveloped sand and gravel deposits should be samples by means of test pits dug to the anticipated
depth of future economic production. Samples should consist of not less than the following
quantities of material which are representative of the deposit:
6
1949 Book of A.S.T.M. Standards, Part 3, pp. 758.
158
Daneshlink.com
daneshlink.com
Sieve Size Quantity

Lb. pieces
6 to 12 in. ... a
3 to 6 in. ... 3ooa

11/2to3in. 400
3/4 to 11/2 in. 200
No.4 to 3/4 in. 100
Sand 50
a Not less than one p1ece from each apparent type of rock.
The amount of material actually studied in the petrographic examination will be determined by the
nature of the examination to be made and the nature of the material to be examined, as discussed in
the next section.
TYPES OF MATERIAL
As previously noted, the specific procedures used in any petrographic examination depend to a large
extent on both the purpose of the examination and on the nature of the sample submitted. The
procedures to be selected will be those that will give adequate information on the properties and
classification of the materials contained in the sample examined. When the intended use of the
aggregate is known, the specific methods used in a petrographic examination will depend on the type
of sample and the complexity of the sample. The samples considered in this test method may be
classified into two groups.
Natural gravel and natural sand

This group includes those materials in which the particle shape and particle surface of the finished
graded aggregate result chiefly from natural processes.
Drilled core, ledge rock, crushed stone, and manufactured sand

This group includes those materials in which the particle shape and particle surface of the finished
graded aggregate result chiefly from the manufacturing process.
Methods for the examination of samples of the first and second groups will be discussed in more
detail. Many of the procedures described for the examination of materials of one group will be
--``,```,`,`,`,,,,`,`````,`,,`,,-`-`,,`,,`,`,,`---
applicable to some samples in the other group.
PROCEDURE FOR PETROGRAPHIC EXAMINATION OF NATURAL GRAVEL AND SAND
Samples
Samples of gravel and natural sand for petrographic examination should be sieved according to
A.S.T.M. Method of Test for Sieve Analysis of Fine and Coarse Aggregates (C 136-46) 7 to provide
samples of each sieve size. In the case of sands an additional portion should then be sieved, without
being washed and test according to A.S.T.M. Method of Test for Amount of Material Finer Than
No. 200 Sieve in Aggregates (C 117-49)8 to provide a sample of the material passing the No. 200
7 1949 Book of A.S.T.M. Standards, Part 3, p. 763.

B 1949 Book of ASTM Standards, Part 3, p. 748.

159
Daneshlink.com
daneshlink.com
sieve. The results of the sieve analysis of each sample made according to A.S.T.M. Method of Test
for Sieve Analysis of Fine and Coarse Aggregate (C 136-46) 6 should be provided to the petrographer
making the examination and used in calculating results of the petrographic examination. Each sieve
--``,```,`,`,`,,,,`,`````,`,,`,,-`-`,,`,,`,`,,`---
fraction should be examined separately, starting with the largest size available. Rocks are more easily
recognized in larger pieces; the breakdown of a heterogeneous type present in the larger sizes may
have provided particles of several apparently different types in the smaller sizes; some important and
easily confused types may be recognizable using the stereoscopic microscope if they are first
recognized and separated in the larger sizes, but may require examination using the petrographic
microscope if they are first encountered in the smaller sizes. The number of particles of each sieve
fraction examined will be ftxed by the required precision of determination of the less abundant
constituents. 9 It is believed that at least 300 particles of each sieve fraction should be identified and
counted in order to obtain reliable results. Precise determinations of small quantities of an important
constituent will require counts of larger samples.
Examination of natural gravel

Coatings-The pebbles should be examined to establish whether exterior coatings are present. If the
are, it should be determined whether the coatings consist of materials likely to be deleterious in
concrete (opal, gypsum, easily soluble salts, organic matter). It should also be determined how firmly
the coatings are bonded to the pebbles.
Rock types
The sieve fraction should be sorted into rock types by visual examination. If all or most of the
groups present are types easily identifiable in hand specimen by examination of a natural or broken
surface, and by scratch and acid tests, no further identification may be needed. Fine-grained rocks
which cannot be identified megascopically or which may consist of or contain constituents known to
be deleterious in concrete should be checked by examination with the stereoscopic microscope. If
they cannot be identified by that means, they should be examined using the petrographic
microscope. The amount of work done in identifying fmc-grained rocks should be adapted to the
information needed about the particular sample. Careful examination of one size of a sample, or
information from previous examination of samples from the same source, will usually make
apparent the amount of additional detailed microscopic work required to obtain information
adequate for the purpose.
Condition
The separated groups belonging to each rock type should be examined to determine whether a
further separation by physical condition is necessary. If all of the particles of a rock type are in a
comparable condition, that fact is noted. More frequently particles in several degrees of weathering
will be found in a group. They should be sorted into categories based on condition and on the
expectation of comparable behavior in concrete. The types of categories intended are: (a) fresh
dense; (b) moderately weathered; (c) very weathered: or (a) dense; (b) porous (or porous and friable).
9
Assuming that the field sampling and laboratory sampling procedures are accurate and reliable, the number of particles
examined, identified, and counted in each sieve fraction will depend on the required accuracy of the estimate of
constituents present in small quantities. The numbers given in this method are minimal. They are based on experience
and on statistical considerations discussed in the following two references:
G. G. Simpson and A. Rowe, 1939, "Quantitative Zoology," McGraw-Hill Co., Inc., New York, N.Y., pp. 182-
185.
A.L. Dryden, Jr., 1931, "Accuracy in Percentage Representation of Heavy Mineral Frequencies," Proceedings,
Natural Academy of Sciences U.S., V. 17, No.5, May, pp. 233-238.
160
Daneshlink.com
daneshlink.com
It is usually impractical to recognize more than three conditions per rock type, and one or two may
be sufficient. An important constituent present in large quantities may sometimes require separation
into four groups by condition. The conspicuous example is chert when it is the major constituent of
a gravel sample. It may be present as dense, unfractured chert; as vuggy chert; as porous chert; and
as dense but highly fractured chert. The determination of which of these four conditions
characterizes a particle may be expected to have an important influence on prediction of the
behavior of the particle in concrete.
Sequence of Examination of Sizes

Examination of the largest size usually indicates the major groups of constituents in a gravel and the
principal considerations which are likely to affect the suitability of the aggregate for the intended use.
The other sieve fractions should then be examined in order of diminishing size, separated into rock
types, and counted. As new groups appear, they should be identified and counted. Changes in the
condition of each group from sieve fraction to sieve fraction should be recorded. Ordinarily it will
not be necessary to examine in detail the material passing the No. 4 sieve in a gravel sample, unless it
amounts to more than 5 percent of the nominal No. 4 to 3/4-in. size range, or unless inspection
suggests that it contains constituents which may be expected to have an important effect on the
suitability of the aggregate for the intended use.
Record
Notes should be taken during the examination. Each rock-type should be described; the relevant
features may include:
1) Particle shape,
2) Particle surface,
3) Grain size,
4) Texture and structure, including observations of pore space, packing of grains,
cementation of grains,
5) Color,
6) Mineral composition,
7) Significant heterogeneities,
8) General physical condition of the rock-type in the sample, and
9) Presence of constituents known to cause deleterious chemical reaction in concrete.
Particle counts should be recorded so that tables can be made for inclusion in the report. When the
examination has been completed, the notes should contain enough information to permit the
preparation of tables and descriptions. Tables should be prepared showing the composition and
condition of the samples by sieve fractions, and the weighted average composition, based on the
gradation of the sample as received and on the distribution of constituents by sieve fractions.
Descriptions of constituent groups should be prepared containing the relevant features among those
listed in the preceding paragraph.
Examination of natural sand

The procedure for the examination of natural sand is similar to that for the examination of gravel,
with the modifications necessitated by the differences in particle size.
--``,```,`,`,`,,,,`,`````,`,,`,,-`-`,,`,,`,`,,`---

161
Daneshlink.com
daneshlink.com
Sizes coarser than No. 30

Each sieve fraction present that is coarser than the No. 30 sieve should be reduced in the Jones riffle
until a split or splits containing at least 300 particles are obtained. The reduced sample of each sieve
fraction should be examined, and its constituents identified, and counted using the stereoscopic
microscope. It is convenient to spread out the sample in a flat-bottomed glass dish such as a Petri
dish and manipulate the grains with a forceps and dissecting needle. The identification of grains in
--``,```,`,`,`,,,,`,`````,`,,`,,-`-`,,`,,`,`,,`---
the coarser sand sizes is often easier when the grains are just submerged in water. The submergence
lessens reflection from the outer surfaces and may show diagnostic features that cannot be seen
when the grains are dry. There are exceptions to this generalization. Where identification is difficult,
the examination includes examination of the natural surface (dry and wet), examination of a broken
surface (dry and wet), and scratch and acid tests. Only after all of these steps have been taken and
the grain is still unidentified should the petrographer resort to the petrographic microscope. Grains
which cannot be identified using the stereoscopic microscope or which are suspected of consisting
of or containing substances known to react deleteriously in concrete should be set aside to be
examined with the petrographic microscope. If the question of reaction with the minor alkalies of
portland cement is important in the examination of the sample, certain additions to the procedure
are indicated. If the coarser sand sizes contain fine-grained possibly glassy igneous rocks, several
typical particles of each variety of such rocks should be selected for more thorough examination.
The petrographer should determine the presence or absence of glass by crushing typical grains and
examining them in immersion media using the petrographic microscope. In difficult or especially
important cases, it may be necessary to break suspected grains, make immersion mounts of part of
the grain and a thin section of another part. Where the sand contains chert and the potential
reactivity of the chert is an important consideration, a number of chert particles from the fractions
retained on the No. 30 sieve should be set aside for determinations of the index of refraction.
Sizes finer than No. 30

The sieve fraction finer than the No. 30 sieve should each be reduced in a Jones riffle to about 4 or
5 g. The volume will usually be less than a level teaspoonful. In some gradations the fraction
retained on the No. 100 and No. 200 sieves may be present in such small amounts that reduction is
unnecessary. These splits should be further reduced on a miniature Jones riffle or by coning and
quartering with a spatula on a clean sheet of paper. A representative portion of each reduced split
should be mounted in immersion oil on a clean glass slide and covered with a clean cover glass. No
entirely satisfactory method of reducing a split to about 300 grains is known. The reduced split can
be sampled by spreading it in a thin layer on glass or clean paper, dragging the end of a dissecting
needle moistened in immersion oil through the sample, and transferring the grains that adhere to the
needle to a drop of immersion oil on a clean slide. If this is done carefully, a fairly representative
sample will be obtained. If the dissection needle is magnetized steel, a concentration of magnetic
minerals may result. It is usually necessary to make several mounts of the No. 50 and No. 100 sieve
fractions to obtain at least 300 grains of each. The index of the immersion oil should be selected to
make the identification of the important constituents as easy and as definite as possible. Practice at
the Waterways Experiment Station is to use an immersion oil with an index of or just below the
lower index of quartz (1.544). The slide should be mounted on a petrographic microscope equipped
with a mechanical stage. Several traverses should be made, and each grain that passes under the
cross hair should be identified and counted. Care should be taken to move the slide on the north-
south adjustment between traverses so that no grain should be counted twice. Each sieve fraction
passing the No. 30 and retained on the No. 200 sieve should be examined in this way. Ordinarily,
the material passing the No. 200 sieve is mounted on a slide following the procedure described
162
Daneshlink.com
daneshlink.com
above, examined using the petrographic microscope, and its composition estimated. If an unusually
large amount of this size is present, or if it contains constituents that may be expected to have an
important effect on the suitability of the aggregate for the intended use, it should be counted.
PROCEDURE FOR PETROGRAPHIC EXAMINATION OF DRILLED CORE, LEDGE ROCK, CRUSHED

STONE, AND MANUFACTURED SAND
Drilled core
Each core should be examined and a log prepared showing footage of core recovered, core loss and
location; location and spacing of fractures and parting planes; lithologic type or types; alternation of
types; physical condition and variations in condition; toughness, hardness, coherence; obvious
porosity; grain size, texture, variations in grain size and texture; type or types of breakage; presence
of constituents capable of deleterious reaction in concrete. If the size of the core permits, the
probability that the rock will make aggregate of the required maximum size should be considered. If
the surface of the core being examined is wetted it is usually easier to recognize significant features
and changes in lithology. Most of the information usually required can be obtained by careful visual
examination, scratch and acid tests, and hitting the core with a hammer. In the case of fine-grained
rocks, it may be necessary to examine parts of the core using the stereoscopic microscope, or to
prepare thin sections of selected portions. Some considerations and procedures are more applicable
to particular rock-types than to others. Ordinarily, the layered rocks considered for concrete
aggregate will be limestone, and occasionally metamorphic rocks, such as phyllite or schist. One of
--``,```,`,`,`,,,,`,`````,`,,`,,-`-`,,`,,`,`,,`---
the most important questions arising in the examination of limestone is that of the presence, type,
and distribution of argillaceous impurities. Limestones which contain intercalated thin beds of soft
shale may make suitable sources of aggregate if the shale is so distributed that it does not prevent
manufacture of the required maximum size, and if the shale can be eliminated or reduced in
processing. Where argillaceous impurities are present, it should be determined whether they actually
consist of clay minerals or of other minerals in clay sizes. If they do consist of clay mineral, it should
be established whether the clay minerals include members of the montmorillonite group (swelling
clays). 10 In the examination of fine-grained igneous rocks, particular attention should be directed to
the nature of the groundmass. This examination should include determination of the presence or
absence of opal, chalcedony, natural glass and clay minerals of the montmorillonite group; if any of
these are found the amount should be estimated; if natural glass if found the type should be
determined.
Ledge rock
The procedure in examination should be the same as for core samples to the extent that the spacing
of samples and size of the individual pieces allow. If the sample consists of a relatively large quantity
of broken stone produced by blasting, it is desirable to inspect the whole sample, estimate the
relative abundance of rock types or varieties present, and sample each type before further
processing. Subsequent procedure should the same as given below for crushed stone.
Crushed stone
The procedure for examination of crushed stone should be similar to that for core, except that
necessary quantitative data should be obtained by particle counts of the separated sieve fractions.
10 G.F. Loughlin, 1928, "Usefulness of Petrology in the Selection of Limestone," Rock Products, V. 31, No.6, pp. 50-59.

163
Daneshlink.com
daneshlink.com
Manufactured sand
If no samples of the rock from which the sand was produced are available, the examination
procedure should be similar to that for natural sand, with particular emphasis on the amount and
extent of fracturing developed by the tnilling operations.
CALCULATIONS
The composition of each sieve fraction of a heterogeneous sample and the weighted sample should
be calculated as follows:
1) The composition of each sieve fraction shall be expressed by summing the total number
of particles of that fraction counted, and calculating each constituent in each condition as a
percentage of that total (amount, as number of particles in percent, in each sieve fraction). It
is convenient to calculate and record the percentages to tenths at this stage. An example of
these calculations is given in the upper half of Table I.
2) The percentage by weight of the sieve fraction in the whole sample (individual
percentages retained on consecutive sieves) shall be obtained from the gradation of the
sample as determine by A.S.T.M. Method (C 136-46) 6 (Individual percent retained on sieve
in Table 1).
3) By multiplying the percentage of the constituent in the sieve fraction, determined under
(1) by the percentage of the sieve fraction in the whole sample, obtained in (2), the
percentage in the whole sample of that constituent in that size shall be obtained (weighted
percentage of constituents in each sieve fraction, Table I). It is convenient to calculate and
record these percentages to tenths.
4) By adding the weighted percentages of each constituent in each sieve fraction, the
weighted percentage of each constituent in the whole sample shall be obtained (see under
weighted composition of sample in Table I).
5) A table shall be constructed to show the composition of each sieve fraction and the
weighted composition of the whole sample. Values shall be reported to the nearest whole
number. Constituents amounting to 0.5 % or less of a sieve fraction or of the whole sample
shall be reported as traces. Table II is an example constructed from the data obtained in
Table I. As a convention, the total in each sieve fraction and the total in the whole sample
shall each be 100% without the traces. Difficulties in abiding by this convention can usually
be avoided by grouping minor constituents of litde engineering importance. It is preferable
to tabulate constituents known to react deleteriously in concrete so that their distribution
will be apparent from inspection of the table, even though the amount in the whole sample
or in any fraction is very small.
--``,```,`,`,`,,,,`,`````,`,,`,,-`-`,,`,,`,`,,`---
REPORT
The report of the petrographic examination should consist of two principal parts, the summary and
the detailed report as indicated in Appendix A. The summary report customarily consists of two
paragraphs that summarize, respectively, the essential data needed to identify the sample as to source
and proposed use, and the features of the material as revealed by the examination. The complete
report records the test procedures employed, the data developed on the composition of the material,
and a description of the nature and features of each important constituent of the sample,
164
Daneshlink.com
daneshlink.com
accompanied by such tables and photographs as may be required to present adequately the findings
of the examination.
Since it is improbable that the engineer who must make the ultimate decision as to the suitability of a
material for use as concrete aggregate will be a geologist or petrographer, it is highly undesirable that
the terminology used in the summary report be simplified to the greatest extent compatible with an
adequate presentation of the data. A list of rock and mineral names is given in Appendix B. It is
suggested that insofar as practicable the terms used in the summary reports be those included in this
list. The petrographic report should be a statement of the findings of the examination. When the
sample has been found to possess properties or constituents which are known to have specific
unfavorable effects in concrete those properties or constituents should be described qualitatively
and, to the extent practicable, quantitatively. The unfavorable effects that may be expected to ensue
in concrete should be mentioned also. When appropriate, it should be stated that a given sample was
not found to contain any undesirable features. The report should not, however, contain a general
statement of opinion by the petrographer concerning the quality of the material. The petrographer
should avoid such general statements of opinion because they are subjective and therefore not
properly a part of a report of objective findings, and since they can properly be made only when data
on different materials can be compared in connection with the requirements for a specific use.
--``,```,`,`,`,,,,`,`````,`,,`,,-`-`,,`,,`,`,,`---
165
Daneshlink.com
daneshlink.com
APPENDIX A
PETROGRAPHIC REPORT
XS ummary- D eta1"ld
e
Symbol: Project: McNary Dam I Date report Submitted: I Initials: KM
PORT-3 6088 April 15, 1948
Serial No.: Source: G-1, Berrian Island, Washington Shore S-1; 50% Berrian Island, 50% Jones-
PORT-3 G-1, S-1 Scott Pit, Oregon Shore
Sample
On February 17, 1948, two samples, one consisting of 3 bags of natural gravel, and the other of 3
bags of sand, were received. Information accompanying the samples stated that the source of the
gravel was Berrian Island, Washington Shore, and the sources of the sand were 50 percent Berrian
Island, Washington Shore, and 50 percent Jones-Scott Pit, Oregon Shore. CRD Serial No. PORT-3
G-1 was assigned to the gravel and CRD Serial No. PORT-3 S-1 to the sand.
Summary
Petrographic examination has been made of samples of natural gravel of 1-in. maximum size from
Berrian Island, Washington Shore (PORT-3 G-1) and of sand produced by combining material from
Berrian Island, Washington Shore with that from the Jones-Scott Pit, Oregon Shore (PORT-3 S-1).
The coarse aggregate is tough dark rounded gravel consisting principally of basalt, with andesite,
rhyolite, quartzite, quartz, chert, granite, and gabbro. The fine aggregate is composed of similar
rocks and of the minerals derived from them: basalt, feldspar, quartz, ferromagnesian minerals, ore
and accessory minerals, and clay. About 5 percent of the coarse aggregate consists of rhyolite and
andesite containing acid and intermediate natural glass and ryholite containing tridymite; about 20
percent of the coarse aggregate consists of basalt containing intermediate natural glass, with a minor
amount of basalt containing basic natural glass. Some of the particles of glassy basalt also contain
small amounts of secondary chalcedony and opal. Approximately 27 percent of the fine aggregate
consists of basalt, andesite, and rhyolite containing intermediate or acid glass, particles of natural
glass, and chalcedony. Acid and intermediate natural glass, tridymite, chalcedony, and opal have all
been shown to be reactive with the minor alkalies of portland cement.
--``,```,`,`,`,,,,`,`````,`,,`,,-`-`,,`,,`,`,,`---
166
Daneshlink.com
daneshlink.com
PETROGRAPHIC REPORT
- Summa11 XD etat"1e d
Symbol: Project: MeN ary Dam Date reported submitted: I Initials: KM
PORT-3 6088 April 1S, 1948
Serial No: Source: G-1, Berrian Island, Washington Shore S-1; SO% Berrian Island, SO% Jones-
PORT-3 G-1, S-1 Scott Pit, Oregon Shore
Test Procedure
Representative portions of each sieve fraction of the gravel were examined and sorted
megascopically and with the stereoscopic microscope. Forty-five thin sections were prepared and
studied using the petrographic microscope. A large number of powders were examined using the
petrographic microscope, to determine the mineral composition of particles and to determine the
index of refraction of natural glass detected in the thin sections. Several photographs were taken.
Representative portions of each sieve fraction of the sand were examined and sorted with the help
of the stereoscopic and petrographic microscopes.
Composition
The coarse aggregate is dark rounded natural gravel, composed principally of fine-grained volcanic
rocks with some quartzite, chert, quartz, mylonite, granite, and gabbro. The sand is composed of
fragments of similar rocks and of the minerals derived from them. The most abundant rock-type in
the gravel and the sand is basalt. About half of the basalt pebbles in the gravel contains natural glass;
a few contain chalcedony or opal. Rhyolite and andesite containing glass and rhyolite containing
tridymite amount to approximately 5 percent of the sample. Almost all of the gravel is dense, tough,
and unweathered. The sand is more weathered than the gravel, and many of the basalt particles are
partially coated with clay. The composition of the gravel and the sand is shown in Tables III and IV,
and the constituents will be described later.
Volcanic rocks without glass

A wide variety of types are included here; the pebbles range from acid types such as rhyolite, rhyolite
and quartz porphyry though andesite to basic types, principally basalt.
--``,```,`,`,`,,,,`,`````,`,,`,,-`-`,,`,,`,`,,`---
Rhyolite and rf:yolite potpf:yry - The particles in this group are rounded, usually pale in color, smooth,
and hard. Most are buff, light gray, or white on freshly broken surfaces. Almost all have porphyritic
texture with phenocrysts of quartz, orthoclase, or acid plagioclase in a fine-grained groundmass. In
many, the groundmass consists largely of spherulites of radially crystallized feldspar and quartz; in
others, the groundmass consists of a granophyric intergrowth of feldspar and quartz. All of the
particles in this group are completely crystalline, and almost all are dense and tough.
Andesite and altered andesite- Most of the pebbles in this group are rounded, greenish-gray, fine-
grained, and dense. Some of them are porphyritic, with phenocrysts of intermediate plagioclase or of
augite in a groundmass composed of microlites of plagioclase. The groundmass in some is felted; in
others, the feldspar microlites show parallel orientation and flow structure. Several of the pebbles in
this group are altered andesites, in which the original ferromagnesian minerals have been replaced by
167
Daneshlink.com
daneshlink.com
minerals of the chlorite group. All the particles in this class are completely crystalline, and almost all
are dense and tough.
Basalt- Approximately 20 percent of the gravel consists of rounded pebbles of tough dark green,
gray, or black basalt. The majority of the crystalline basalts studied in thin section are composed of
laths of calcic plagioclase with interstitial pyroxene and ore minerals. A minority contain olivine as
well as pyroxene. Some contain large quantities of interstitial ore minerals which make the
groundmass entirely opaque and raise the specific gravity of the particles.
Volcanic rocks with glass or tridymite

This group includes pebbles of light-colored rhyolite, greenish andesite, and dark, sometimes
vesicular, basalt. All of them contain materials which are known to be reactive with the minor
alkalies of portland cement.
Rhyolite and andesite - The rhyolite and andesite are generally similar to those described in the
paragraph above, but contain interstitial natural glass of intermediate or acid composition. In one
thin section of a rhyolite, sparse phenocrysts of orthoclase and sodic plagioclase were scattered in a
spherulitic groundmass containing small crystals of tridymite.
Basalt-The basalt particles which contain natural glass are gray, dark green, or black. Some contain
phenocrysts which can be seen with the naked eye; others are apparently structureless and break
with conchoidal fracture. Some are vesicular, with the rims of the vesicles bleached; in a few the
vesicles have been filled with hisingerite, chalcedony (Fig. 9), opal, or clay minerals of the
montmorillonite group. The crystalline constituents of these pebbles are the same as those of the
basalt without glass, but the crystals are generally smaller, and the groundmass is made up of glass
(Fig. 10(a)) or contains patches of glass (Fig. 10 (b)). The glass varies in color from pale yellow
through tan and brown to black. It is believed that the dark color is due to minute diffused particles
of opaque ore minerals. Almost all of the natural glass in the basalt pebbles has indices of refraction
--``,```,`,`,`,,,,`,`````,`,,`,,-`-`,,`,,`,`,,`---
below 1.57; in a few particles glass of index above 1.57 was detected.
Miscellaneous
A wide variety of rock and mineral fragments amount to approximately 25 percent of the gravel and
7 percent of the sand. In the coarse aggregate, tough fresh quartzite particles are the most abundant
constituent of this group. Quartz is a consistent minor constituent of the coarse and the fine
aggregate. Chert composed of microcrystalline quartz is present in the gravel. Minor amounts of
chalcedony were found in the sand. Pebbles of mylonite, granite, and gabbro are present in the
coarse aggregate. In the sand, quartzite particles, clay, calcite, ore minerals, and accessory mineral
such as zircon were assigned to this group.

168
Daneshlink.com
daneshlink.com
APPENDIX B
LIST OF ROCK AND MINERAL NAMES FOR USE BY CONCRETE LABORATORY PETROGRAPHERS
Rock Names
Group terms Varietal or minor rock names
1. Igneous rocks 1. Syenite
2. Volcanic rocks 2. Pegmatite
3. Basic igneous rocks 3. Gabbro-diorite
--``,```,`,`,`,,,,`,`````,`,,`,,-`-`,,`,,`,`,,`---
4. Acid igneous rocks 4. Amphibolite
5. Basic volcanic rocks 5. Diabase
6. Acid volcanic rocks 6. Serpentine
7. Glassy volcanic rocks 7. Dacite
8. Porphyry
9. Metamorphic rocks 8. (-)porphyry
10. Sedimentary rocks 9. Glassy(-)
11. Carbonate rocks 10. Gneissic(-)
Major rock types
11. Argillaceous limestone
1. Granite 12. Arenaceous limestone
2. Granodiorite 13. Cherty limestone
3. Diorite 14. Shaly limestone
4. Gabbro 15. Oolitic limestone
5. Basalt 16. Siliceous limestone
6. Rhyolite} F "t 17. Magnesian limestone
{ 7. Andesite e151 e 18. Dolomitic limestone
8. Pumice 19. Lithographic limestone
9. Gneiss 20. Chalk
10. Schist 21. Mudstone
11. Marble 22. Argillite
12. Quartzite 23. Phyllite
24. Argillaceous sandstone
13. Limestone 25. Calcareous sandstone
14. Shale 26. Arkose
15. Slate 27. Arkosic quartzite
16. Siltstone 28. Coal
17. Chert 29. Lignite
18. Sandstone 30. Clay-ironstone
19. Graywacke
20. Conglomerate
Mineral Names •
Group terms
1. Accessory minerals 8. Mica
2. Carbonate minerals 9. Limonite
3. Clay minerals 10. Wad
4. Secondary minerals 11. Amphibole
5. Ferromagnesian minerals 12. Pyroxene
6.Volcanic glass (a rock constituent but not a mineral) 13. Zeolite
7. Feldspar 14. Sulfides
Major minerals (and rock components)
1. Quartz 13. Chlorite
2. Chalcedony (a mixture of a mineral and a mineraloid) 14. Calcite
3. Opal (a mineraloid) 15. Dolomite
169
Daneshlink.com
daneshlink.com
4. Tridymite 16. Pyrite

5. Cristobalite 17. Hematite
Feldspar
6. Orthoclase 18. Magnetite
7. Plagioclase 19. Gypsum
8. Microcline 20. Kaolin
Pyroxene Volcanic Glass (not a mineral)
9. Augite 21. Acid Volcanic Glass
Amphibole
10. Hornblende 22. Basic Volcanic Glass
Mica
11. Muscovite 23. Intermediate Volcanic Glass
12. Biotite
Varietal Terms and Less-Frequent Minerals
1. Gold 20. Socialite
2. Graphite 21. Olivine
3. Zircon 22. Montmorillonite
4. Garnet 23. Biedellite
5. Apatite 24. Halloysite
6. Allanite 25. Serpentine
7. Sphene 26. Alunite
8. Epidote 27. Bytownite
9. Fluorite 28. Andesine
10. Tourmaline 29. Labradorite
11. Galena 30. Albite
12. Pyrrhotite 31. Anorthite
13. Marcasite 32. Perthite
14. Sphalerite 33. Microperthite
--``,```,`,`,`,,,,`,`````,`,,`,,-`-`,,`,,`,`,,`---
15. Chalcopyrite 34. Aragonite

16. Analcite 35. Siderite
17. Natrolite 36. Barite
18. Laumontite 37. Glauconite
19. Nepheline
a The terms given in this list have been used in petrographic reports issued by the Concrete Research Division,
Waterways Experiment Station, or occur in published literature on concrete aggregates.
170
Daneshlink.com
daneshlink.com
BIBLIOGRAPHY
1) Bulletin, Geol. Soc. Am., New York, N.Y.

2) Bulletins, Research Laboratories of the Portland Cement Association, Chicago, Ill.
3) Bulletins, Waterways Experiment Vicksburg, Miss.
4) Chemical Abstracts, American Chemical Society, Columbus, Ohio.
5) Journal, American Concrete Institute, Proceedings, (monthly, September to June), Detroit,
Mich.
6) Journal of Geology, Chicago, Ill.
7) Journal of Sedimentary Petrology, (three numbers a year), Tulsa, Okla.
8) Magaifne of Concrete Research, Cement and Concrete Association, 52 Grosvenor Gardens,
London SW 1, England.
9) Proceedings, American Society of Civil Engineers, New York, N. Y.
10) Proceedings, American Society for Testing and Materials (annual volume), Philadelphia, Pa.
11) Rock Products, Maclean-Hunter Publishing Co., (monthly), Chicago, Ill.
12) The American Mineralogist (bimonthly), Washington D.C.
REFERENCE BOOKS
1) E. M. Chamot and C. W. Mason, "Handbook of Chemical Microscopy," 2 Vol. 2nd Edition:

John Wiley and Sons, Inc., New York, New York, N.Y. (1938).
2) E. S. Dana and W. E. Ford, "Textbook ofMineralology," 4th Edition, John Wiley and Sons,
Inc., New York, N.Y., (1902).
3) F. F. Grout, "Kemp's Handbook of Rocks," 6th Edition, D. Van Nostrand Co., Inc., New
York, N.Y., (1940).
4) F. F. Grout, "Petrography and Petrology," McGraw-Hill Book Co., Inc., New York, N.Y.
(1932).
5) A. Harker, "Metamorphiem," E.P. Dutton and Co., Inc., New York, N.Y., (1932).
6) A. Harker, "Petrology for Students," 7th Edition, London: Cambridge University Press
(MacMillan), (1935).
7) Arthur Holmes, "Petrographic Methods and Calculations," D. Van Nostrand Co., Inc., New
York, N.Y. (1921).
8) Arthur Holmes, "The Nomenclature of Petrology," D. Van Nostrand Co., Inc., New York,
N.Y.
9) A. Johannsen, "Manual of Petrographic Methods," McGraw-Hill Book Co., Inc., New York,
N.Y. (1918).
10) B. H. Knight and R. G.Knight, "Road Aggregates-Their Uses and Testing," 2nd Edition,
London: Edward Arnold and Co., (1948).
11) W. C. Krumbein and F. J. Pettijohn, "Manual of Sedimentary Petrography," Appleton-
Century-Crofts, Inc., New York, (1938).
12) C. K. Leith and W. J. Mead, "Metamorphic Geology," Henry Holt and Co., Inc., New York,
N.Y. (1915).
13) L. C. Martin and B. K. Johnson, "Practical Microscopy," Chemical Publishing Co., New
York,N. Y.
14) H. B. Milner, "Sedimentary Petrography," 3rd Edition, Nordeman Publishing Co., London:
Murby; New York, N.Y. (1940).

171
--``,```,`,`,`,,,,`,`````,`,,`,,-`-`,,`,,`,`,,`---
Daneshlink.com
daneshlink.com
15) F. J. Munoz and H. A. Charipper, "The Microscope and Its Use," Chemical Publishing Co.,
New York, N. Y. (1943).
16) F. J. Pettijohn, "Sedimentary Rocks," Harper and Brothers, New York, N.Y. (1949).
17) A. F. Rogers and P. F. Kerr, "Optical Mineralogy," 2nd Edition. McGraw-Hill Book Co.,
Inc., New York, N. Y (1942).
18) 0. G. Runner, "Geology for Civil Engineers-As Related to Highway Engineering," Gillette
Publishing Co., Chicago, Ill., (1939).
19) S. J. Shand, "Eruptive Rocks," Revised 3rd Edition, London: Murby; John Wiley and Sons,
20) C. P. Shillaber, "Photomicrography on Theory and Practice," John Wiley and Sons, Inc.,
New York, N.Y., (1944).
21) A. F. Taggart, "Handbook of Mineral Dressing, Ores, and Industrial Minerals," 2nd printing,
John Wiley and Sons, Inc., New York, N.Y. (1947).
22) F. G, Tickell, "The Examination of Fragmental Rocks," Stanford University, California:
Stanford University Press, (1948).
23) W. H. Twenhofel "Principles of Sedimentation," McGraw-Hill Book Co., Inc., New York,
N.Y. (1939).
24) G. W. Tyrell, "The Principles of Petrology," 2nd Edition, E.P. Dutton and Co., Inc., New
York, N.Y., (1929).
25) E.E. Wahlstrom, "Igneous Minerals and Rocks," John Wiley and Sons, Inc., New York,
N.Y., (1947).
26) E. E. Wahlstrom, "Optical Crystallography," John Wiley and Sons, Inc., New York, N.Y.,
(1943).
27) A. N. Winchell, Elements of Optical Mineralogy, 3 Vol., 4th Edition, John Wiley and Sons,
Corps of Engineers reports

1) Concrete Conference, 1949, Kansas City, Mo., Mar. 1-3,1949, Office, Chief of Engineers,
pp.1-71,A-1 toA-61.
2) Concrete Durability Conference, Kansas City, Mo., Mar. 31, Apr. 1-2, 1949, Office, Chief of
Engineers, 1949, pp. 1-236.
3) Manufactured Aggregate Conference, Washington D.C., Nov. 6, 1947, Chief, Office of
Engineers, 1948, pp. 1-111, plates I-XXVII.
4) Petrographic Conference, Washington D.C., Oct. 5-8, 1948, Office, Chief of Engineers,
1948, pp. 1-47, 8 reports, bibliography.
--``,```,`,`,`,,,,`,`````,`,,`,,-`-`,,`,,`,`,,`---
U.S. Geological Survey Bulletins (reference books), Washington, D.C.

1) F. W. Clarke, "The Data of Geochemistry," Bulletin 848, (1934).
2) E.S. Larsen and H. Berman, "The Microscopic Determination of Non-opaque Minerals,"
Bulletin 848, (1934).
3) Grace M. Wilmarth, "Lexicon of Geologic Names of the United States," 2 Vols. (1938).
4) M. N. Short, "Microscopic Determination of the Ore Minerals," Bulletin 914, 2nd Edition,
(1940).
5) J. N. Nickles et al., "Bibliography of North American Geology."
Bulletin 746, and 747, 1785-1918.
Bulletin 823, 1918-1928.
Bulletin 937, 1929-1938.
Bulletin 938, 1940-1941.

172
Daneshlink.com
daneshlink.com
Bulletin 949, 1942-1943.

Bulletin 952, 1944-1945.
Selected references
1) A. R. Alderman et al., "Australian Aggregates and Cements," Part I of Studies in Cement-
Aggregate Reaction, Melbourne, Commonwealth of Australia, Council for Scientific and
Industrial Research, Bulletin 229, pp. 7-46, (1947).
2) R. F. Blanks, "Modern Concepts Applied Concrete Aggregate," Proceedings, American Society
of Civil Engineers, pp. 441-468, (1949).
3) R. F. Blanks and H. S. Meissner, "The Expansion Test as a Measure of Alkali-Aggregate
Reaction," Journal, American Concrete Institute, Vol. 17, No. 5, Apr., 1946; Proceedings, Vol.
42, pp. 517-539.
4) R. F. Blanks, et al., "Notes on the 'Effect of Alkalies in Pordand Cement on the Durability
of Concrete,'" Proceedings, American Society of Testing and Materials, Vol. 43, pp. 199-220,
(1943).
5) M. N. Bramlette, "The Monterey Formation of California and the Origin of its Siliceous
Rocks,'' U.S. GeologicaiSurory, PP 212 (1946), esp. pp. 16-17 (57 pp.).
6) Cloyd M. Chapman, "Influence of Aggregates upon Shrinkage of Mortar and Concrete,"
Proceedings, American Society of Testing and Materials, Vol. 24, Part II, p. 767 (1924).
7) Howard A. Coombs, "Expansion of Concrete Due to Reaction Between Andesitic
Aggregate and Cement," American Journal of Science, Vol. 240, Apr., 1942, pp. 288-297.
8) F. K. Fears, "Bibliography on Mineral Aggregates," Highway Research Board, Bibliography
No. 6, pp. 1-89 (1949).
9) E. A. Finney, "Microscopic Investigation of Scaled and Unsealed Concrete," Michigan
Engineering Experiment Station, Bulletin .No. 101, Nov., 1944.
--``,```,`,`,`,,,,`,`````,`,,`,,-`-`,,`,,`,`,,`---
10) W. H. Fry, "Petrographic Methods for Soils Laboratories," Technical Bulletin 344, U.S.
Department of Agriculture, (1933).
11) J. R. Gilbert, "The U.S. Corps of Engineers Approach to More Durable Concrete," Crushed
Stone Journal, Vol. XXIV, June, 1949, pp. 3-10,30.
12) J. H. Griffith, "Physical Properties of Typical American Rocks," Bulletin 131, Iowa
Engineering Experiment Station, 56 pp. (1937).
13) H. A. Kammer and R. W. Carlson, "Investigation of Causes of Delayed Expansion of
Concrete in Buck Hydroelectric Plant," Journal, American Concrete Institute, Vol. 00, No.
00, 1941; Proceedings, Vol. 38, pp. 665-673.
14) T. M. Kelly, et al., "A Study of Alkali-Aggregate Reactivity by Means of Mortar Bar
Expansions," Journal, American Concrete Institute, Vol. 20, No.1, Sept., 1941, Proceedings,
Vol. 45, pp. 57-80.
15) Paul D. Krynine, "The Megascopic Study Field Classification of Sedimentary Rocks," Journal
of Geology, Vol. 56, No. 2, pp. 000, (1948); Technical Paper 130, Pennsylvania State College,
Mineral Industrial Experiment Station.
16) E. C. E. Lord, "Examination and Classification of Rocks for Road Building,'' U.S.
Department of Agriculture, Office of Public Roads, Bulletin 3 7, pp. 1-28 (1911).
17) G. F. Loughlin, "Qualifications of Different Kinds of Natural Stone for Concrete
Aggregate," Proceedings, American Concrete Institute, Vol. 23, pp. 319-351, (1927).
18) G. F. Loughlin, "Usefulness of Petrology in the Selection of Limestone," Rock Products, Vol.
31, No.6, Mar. 17, 1928, pp. 50-59, (15 figs.).
173
Daneshlink.com
daneshlink.com
19) G. F. Loughlin and]. C. Pearson, "An Interesting Case of Dangerous Aggregate: The
Altered Feldspar and Causes of Its Disintegration," Proceedings, American Concrete Institute,
Vol19, pp. 142-153, (1923); Concrete, Vol. 22, No.4, Apr., 1923, pp. 144-147.
20) B. Mather, Discussion of "Reactivity of Aggregate Constituents in Alkaline Solutions" by
Leonard Bean and]. J. Tregoning, Journal, American Concrete Institute, Vo. 17, No.2, Nov.,
1945; Proceedings, Vol. 44, Gune 1948), pp. 1058-1060.
22) B. Mather, "Petrographic Identification of Reactive Constituents in Concrete Aggregates,"
Proceedings, American Society ofTesting and Materials, Vol. 48, pp. 1120-1127, (1948).
23) K. Mather and J. Weiser, "The Petrography of Aggregates for Mass Concrete," Vicksburg:
Waterways Experiment Station Bulletin, 1950.
24) D. McConnell and W. H. Irwin, "Notes on Cement-Aggregate Reaction in Concrete,"
American Mineralogist, Vol. 30, pp. 78-80, (1945).
25) D. McConnell et al., "Cement-Aggregate Reaction in Concrete," Journal, American Concrete
Institute, Vol. 19, No.2, Oct. 1947; Proceedings, Vol. 44, pp. 93-128.
26) F. R. McMillan and G. W. Ward, "Influence of Coarse Aggregate on the Durability of
Concrete," Proceedings, American Society of Testing and Materials, Vol. 29, Part II, pp. 816-
823, (1929).
27) H. S. Meissner, "Cracking in Concrete Due to Expansive Reaction Between Aggregate and
High-Alkali Cement as Evidenced in Parker Dam," Journal, American Concrete Institute,
Vol. 00, No. 00, 1941; Proceedings, Vol. 38, pp. 549-568.
28) R. C. Mielenz, "Petrographic Examination of Concrete Aggregates," Bulletin, Geol. Soc. Am.,
Vol. 57, Apr., 1946, pp. 309-318.
29) R. C. Mielenz, et al., "Chemical Test for Reactivity of Aggregates with Cement Alkalies;
Chemical Processes in Cement-Aggregate Reaction," Journal, American Concrete Institute,
Vol. 19, No.3, Nov., Proceedings, Vol. 44, pp. 193-221.
--``,```,`,`,`,,,,`,`````,`,,`,,-`-`,,`,,`,`,,`---
30) R. C. Mielenz and L. P. Witte, "Tests Used by the Bureau of Reclamation for Identifying
Reactive Concrete Aggregates," Proceedings, American Society of Testing and Materials, Vol.
48, pp. 1071-1107 (1948).
31) W. H. Parsons and H. Insley, "Alkali Etching Test on Concrete Aggregate," Journal,
American Concrete Institute, Vol. 15, No. 3,Jan., 1944; Proceedings, Vol. 40, p. 229.
32) W. H. Parsons and W. H. Johnson, "Factors Affecting the Thermal Expansion of Concrete
Aggregate Materials," Journal, American Concrete Institute, Vol. 15, No.4, April, 1944;
Proceedings, Vol. 40, pp. 457-466.
33) J. C. Pearson, "A Concrete Failure Attributed to Aggregate of Low Thermal Coefficient,"
Journal, American Concrete Institute, Vol. 13, No.1, Sept., 1941; Proceedings, Vol. 38, pp. 29-
36.
34) C. E. Proudley, "Sampling Mineral Aggregates," American Society of Testing and Materials,
pp. 73-87 (1948); (Symposium issued as separate publication SIP No. 8J).
35) Elliot P. Re.xford, "Some Factor in the Selection and Testing of Concrete Aggregates for
Large Structures," Transactions, American Institute Mining and Metallurgical Engineers, Vol.
187, No.3, pp. 395-402 (1950), (Technical Publication 2837 H.).
36) Roger Rhoades and R. C. Mielenz, "Petrography of Concrete Aggregate," Journal, American
Concrete Institute, Vol. 17, No.6, 1946; Proceedings, Vol. 42 (1946), pp. 581- 600 (with
extensive bibliography, 65 items).
37) Roger Rhoades and R. C. Mielenz, "Petrographic and Mineralogic Characteristics of
Aggregates," American Society of Testing Materials, pp. 20-48 (1948). (Symposium issued as
separate publication SIP No. 83).

174
Daneshlink.com
daneshlink.com
38) G. Rittenhouse, "Analytical Methods as Applied in Petrographic Investigations of

Appalachian Basin," U.S. Geological Survey, Circular 22, Mar., 1948.
39) R. Russell, "Mineral Composition of Mississippi River Sands," Bulletin, Geol. Soc. Am., Vol.
48, pp. 1307-1348 (1937).
40) R. J. Schaffer, "The Weathering of Natural Building Stones," London: Special RepottNo. 18,
Building Research, Department of Scientific and Industrial Research, (1932).
41) F. A. Shergold, "A Review of Available Information on the Significance of Roads tone
Tests," London: Department of Scientific and Industrial Research, Road Research
Laboratory, Technical Paper No. 10, (1948).
42) R. B. Sosman, "The Properties of Silica," Monograph No. 3 7, American Chemical Society,
Chemical Catalog Co., 856 pp., (1927).
43) E. J. Spain, Jr., and N. A. Rose, "Geological Study of Gravel Concrete Aggregate of the
Tennessee River," American Institute of Mining and Metallurgical Engineers, Technical
Publication No. 840, 16 pp., (1937).
44) T. E. Stanton, "Expansion of Concrete Through Reaction Between Cement and Aggregate,"
Proceedings, American Society of Civil Engineers, Vol 66, p. 1781, (1940).
45) T. E. Stanton, "Studies to Develop an Accelerated Test Procedure for the Detection of
Adversely Reactive Cement-Aggregate Combinations," Proceedings, American Society of
Testing and Materials, Vol. 43, pp. 875-904, (1943).
46) T. E. Stanton, "Correlation of Laboratory Tests with Field Experiences of Excessive
Concrete Expansion Induced by a Reaction Between the Cement and Aggregate," Proceedings,
American Society of Testing and Materials, Vol. 48, pp. 1057-1066, (1948).
4 7) H. S. Sweet, Physical and Chemical Test of Mineral Aggregates and Their Significance, "
Symposium on Mineral Aggregates, American Society of Testing and Materials, No. 83,
1948, pp. 49-73.
48) H. S. Sweet and K.B. Woods, A Study of Chert as a Deleterious Constituent in Aggregates,
Highway Research Bulletin No.9, Research Series No. 86, The Engineering Experiment
Station, Purdue University, Sept., 1942, 106 pp.
--``,```,`,`,`,,,,`,`````,`,,`,,-`-`,,`,,`,`,,`---
49) W. A. Tart, "Terminology of the Chemical Siliceous Sediments," Exhibit A in Report of the
Committee on Sedimentation, Division of Geology and Geography of the National Research
Council for 1937-38.
50) Karl Terzaghi, "Stress Conditions for the Failure of Saturated Concrete and Rock,"
Proceedings, American Society of Testing and Materials, Vol. 45, pp. 777-792, (1945).
Discussion by Douglas McHenry, D. W. Kessler, K. Terzaghi, pp. 793-801.
51) Bailey Tremper, "Correlation of Laboratory Tests with Field Experience in Alkali-Aggregate
Reaction," Proceedings, American Society ofTesting and Materials, Vol. 48, pp. 1067-1078,
(1948).
52) H. S. Washington, "How Petrography Can Aid the Stone Producer," The Explosives
Engineer, (Oct. 1925).
53) T. F. Willis and M. E. De Reus, "Thermal Volume Change and Elasticity of Aggregates and
Their Effect on Concrete," Proceedings, American Society of Testing and Matserials, Vol. 39,
pp. 919-929 (1939).
54) K. B. Woods, "Distribution of Mineral Aggregates," Symposium on Mineral Aggregates,
American Society of Testing and Materials, pp. 4-19, (with 129 references) (1948).
(Symposium issued as separate publication S. T.P. No. 83).
Miscellaneous
1) Geologic Map of the Unites States, 1/2,500,000, U.S. Geological Survey, 1932.
175
Daneshlink.com
daneshlink.com
2) Geologic Maps of Particular Areas.

3) "Rock-Color Chart," National Research Council, (1948).
Nate-Workers concerned with any particular area will of course obtain and use specific works on
--``,```,`,`,`,,,,`,`````,`,,`,,-`-`,,`,,`,`,,`---
the geology of that area; such publications can be located by reference to the State Geological
Survey, State Engineering Experiment Station, Bibliography of North American Geology, etc.

176
Daneshlink.com
daneshlink.com
Figure 1 -
. /~.
Equipment for examination of samples at low magnifications.
--``,```,`,`,`,,,,`,`````,`,,`,,-`-`,,`,,`,`,,`---
Figure 3 - Horizontal lap for grinding and smoothing sawed suifaces of rock.
177
Daneshlink.com
daneshlink.com
, Figure 4 - Grinding thin section blank on horizontal lap.

--``,```,`,`,`,,,,`,`````,`,,`,,-`-`,,`,,`,`,,`---
'
Figure 5 - Final grinding of thin section on glass plate.
Figure 6 - Petrographic microscope for examination of thin sections orpowder mounts, index, media, counter.

178
Daneshlink.com
daneshlink.com
Figure 7 - Sodium arc lamp and abbe refractometerfor calibrating index media.
--``,```,`,`,`,,,,`,`````,`,,`,,-`-`,,`,,`,`,,`---
Figure 8 - Photomicrographic camera in use overpetrographic microscope.

179
Daneshlink.com
daneshlink.com
Tabl e 1-CaIcu Iat1on . Ie counts

. of resu ts of part1c
Composition of fractions retained on sieves shown below
3/4-in. 1/2-in. 3/8-in. No.4
Constituents
Number of Number of Number of 0 Number of
% 0
/o /o 0
/o
particles particles particles particles
AI 250 50.0 200 40.0 150 30.0 50 10.0
A2 50 10.0 100 20.0 125 25.0 100 20.0
A3 10 2.0 50 10.0 75 15.0 100 20.0
Bl 107 21.4 70 14.0 62 12.4 32 6.4
B2 76 15.2 53 10.6 19 3.8 87 17.4
B3 - - 20 4.0 43 8.6 96 19.2
c1 5 1.0 5 1.0 20 4.0 20 4.0
c2 2 0.4 2 0.4 6 1.2 10 2.0
c3 - - - - - - 5 1.0
Totals 500 100 500 100 500 100 500 100
Individual %
retained on 17.4 32.6 29.5 20.5
sieve
Weighted percentages of constituents in each sieve fraction
Weighted
3/4-in. 1/2-in. 3/8-in. No.4 composition of
sample
AI
A2
8.7
1.7
13.0
6.5
8.9
7.4
2.1
4.1 32.7}
19.7 64.6 (total A)
A3 0.4 3.3 4.4 4.1 12.2
B1
B2
3.7
2.6
4.6
3.5
3.7
1.1
1.3
3.6 13.3}
10.8 31.8 (total B)
B3 - 1.3 2.5 3.9 7.7
--``,```,`,`,`,,,,`,`````,`,,`,,-`-`,,`,,`,`,,`---
c1
c2
0.2
0.1
0.3
0.1
1.2
0.3
0.8
0.4 2.5}
0.9 3.6 (total C)
c3 - - - 0.2 0.2
Total in sieve
17.4 32.6 29.5 20.5
fraction
Total in sample, condition 1 48.5

180
Daneshlink.com
daneshlink.com
Table 11-Composition and Condition of an Aggregate Sample

(Table Constructed from Calculations Shown in Table I)
Amount, as number of particles in percent
In fractions retained on sieves
In whole sample b
Constituents shown below a
3/4 3/8 Totals
1/2in. No.4 Condition 1 Condition 2 Condition3
in. in.
A 62 70 70 50 33 20 12 65
B 37 29 25 43 13 11 8 32
c 1 1 5 7 2 1 trace 3
Total 100 100 100 100 - - - 100
Weighted average, condition 1 48 - - -
Weighted average, condition 2 - 32 - -
.
Weighted average, condition 3
Based on count of 500 parncles m each sieve fracnon .
- - 20 -
h Based on gradation of the sample as received, and on the distribution of constituents by sieve fractions shown at the left
above. (If the petrographic report forms part of a complete investigation of the sample, including a report of the
gradation, the gradation need not be shown. If the petrographic report is to be submitted alone, the gradation of the
sample should be included with it).
Table Ill-Composition of Gravel from Berrian Island, Washington Shore

(PORT·3 G·1)
Constituents Amount, as number of particles, in percent
Volcanic rocks without glass:
50
@_asalt, andesite, rhyolite and rhyolite porphyry)
Volcanic rocks with glass or tridymite:
20
Basalt containing glass
--``,```,`,`,`,,,,`,`````,`,,`,,-`-`,,`,,`,`,,`---
Andesite and rhyolite with glass, rhyolite with
5
tridymite
:Miscellaneous:
(Granite, gabbro, quartzite, myolinite, quartz, 25
chert)
Total 100
Table IV-Composition of Sand Proposed for Use in McNarv Dam (PORT-3 G-1)
Amount as number of particles, in percent a
In fraction retained on sieves shown
Constituents below: In whole
No. No. No. No. No. No. sandb
8 16 30 50 100 200
Volcanic rocks with glass; glass; chalcedony 39 37 32 23 8 3 27
Volcanic rocks without glass 50 44 47 11 11 10 32
Feldspar 4 5 8 47 47 56 23
Quartz 6 7 6 11 6 4 7
Ferromagnesian mineral (Pyroxene, amphibole, olivine,
- 1 2 1 15 15 4
mica)
:Miscellaneous (Sediments; clay; ore and accessory
1 6 5 7 13 12 7
minerals)
Total 100 100 100 100 100 100 100
a Based on the examination of 500 particles In each s1eve fraction.
b Based on grading of the sand as received, and on the distribution of constituents by sieve fractions shown the left
above.

181
Daneshlink.com
daneshlink.com
Figure 9 - Basalt pebble from Berrian Island, Washington shore (Port-3 G-1)
Photomicrograph ofglasry vesicular basalt (X 450, crossed nicols), showing vesicle .filled with chalcedof!Y at center. Much ofthe
dark grqy groundmass is somewhat devitrified naturalglass.
--``,```,`,`,`,,,,`,`````,`,,`,,-`-`,,`,,`,`,,`---

182
Daneshlink.com
daneshlink.com
Figure 1O(a)- Photomicrograph ofglasry basalt (X 200, plane polarized light), constituents indicated f?y letters: A = augite,
F =plagioclase feldspar, G =glass. The black pofygons are ore minerals. Almost all ofthe medium to dark grcry groundmass is
intermediate naturalglass.
Figure 10 (b)- Photomicrograph ofglasry basalt (X 200, plane polarized light). The constituents are the same as those shown
above, but the crystals are lm:ger and less glass is present.
Figures 10(a) and (b)- Basalt pebbles from Berrian Island, Washington Shore (Port-3 G-1).
--``,```,`,`,`,,,,`,`````,`,,`,,-`-`,,`,,`,`,,`---

183
Daneshlink.com
daneshlink.com
--``,```,`,`,`,,,,`,`````,`,,`,,-`-`,,`,,`,`,,`---

Daneshlink.com
daneshlink.com
Reprinted, with pennission, from Proceedings, ASTM STR 149, ©ASTM International, 1953, 100 Barr Harbor Drive,
West Conshohocken, PA 19428.
APPLICATIONS OF LIGHT MICROSCOPY IN CONCRETE RESEARCH
By Katharine Mather'
SYNOPSIS
The applications of light microscopy in concrete research are outgrowths of its applications in
petrology, mineralogy, and chemistry, but there is more quantitative emphasis in its use in concrete
research than is common in petrology. Metallographic and combined petrographic and
metallographic techniques are used to study pordand cement. Normal petrographic methods are
used to study aggregates and cement-aggregate reactions. Air content and bubble spacing in concrete
are investigated by linear traverse and point-count techniques. The use of light microscopy in
making comparative studies of the microstructure of concrete is described.
SUBJECTS FOR MICROSCOPY IN CONCRETE RESEARCH
The major ingredients of concrete are the cement, the aggregates, and the mixing water with which
the cement reacts by hydrolysis and hydration to form the hardened cement paste which envelops
and bonds the aggregates together. Microscopy is used in concrete research in the investigation of
pordand cement and other cementitious or pozzolanic materials used as supplements or partial
substitutes for pordand cement, in the investigation of natural and artificial aggregates, and in the
investigation of hydrated cements, mortars, and concretes. The polarizing microscope was the first
type to be used; at present, polarizing, metallographic, and stereoscopic microscopes are all
employed.
In the general field of concrete research the first extensive use of microscopy was made in the
investigation of pordand cement, a specialized area in high-temperature physical chemistry, that is
closely related to experimental petrology, and to studies of ceramics, refractories, slags, and
aluminous cements. The polarizing microscope was the standard tool used in the identification of
phases and compounds formed in such high-temperature phase research. It is a most important
tool, particularly before X-ray methods became available as alternate techniques, for the
identification of specific crystalline compounds. It was usually possible to confirm the results of
microscopy by some other chemical or physical test. This situation tended apparendy to make the
microscopists cautious and specific in their interpretations; aware that their results needed to be
related to other tests of the same materials; concerned with confirming what they saw by other
means; and interested in technique and in quantitative methods.
The use of microscopy in the examination of natural aggregates developed later and might be
regarded as an offshoot of descriptive petrography and related to the petrography of building stone
and roadstone. In this field, the microscopist's results have been less subject to external
confirmation, interpretations have been less cautious and more generalized, and interest in
• Chief, Petrography Section, Concrete Research Division, Waterways Experiment Station, Corps of Engineers, U.S.
Army, Jackson, :Miss.
185
--``,```,`,`,`,,,,`,`````,`,,`,,-`-`,,`,,`,`,,`---

Daneshlink.com
daneshlink.com
quantitative methods, specific information, and refinements in technique have been more limited
than in the microscopy which is more closely connected with high temperature plus physical
chemistry.
The microscopy of the individual constituents of portland cement, their hydration products, and
ground cement developed along with the microscopy of unground portland-cement clinker. There
were difficulties in obtaining crystals of convenient size, and even greater difficulties in the
preparation of thin sections of neat cement, mortar, and concrete, because the hydration products
may become altered during specimen preparation. The first great problem to be overcome is the
problem of preparation, since no satisfactory interpretation can be produced without assurance that
the sample examined correctly represents the mortar or concrete and not some alteration product
developed during preparation.
Up to 1940 it was generally assumed that aggregates were normally chemically inert toward cement
paste. Most of the problems of concrete which had been regarded as connected with aggregate were
blamed on properties of the aggregates such as softness or high absorption. But in 1940 reports of
what was termed cement-aggregate reaction or alkali-aggregate reaction began to accumulate, and it
presently became clear that some aggregates were not inert under some circumstances.
A review of microscopy in concrete research begins logically and historically with microscopy
applied to the investigation of portland cement. Attention to the microscopy of aggregates and
concretes developed later and at about the same time. The microscopy of aggregates developed
rapidly in some respects, but except for the petrography of alkali-aggregate reaction, the microscopy
of concretes was largely neglected. It seems reasonable to consider aggregates after cement, to
discuss alkali-aggregate reaction as a separate topic, and to leave for the last the most complex and
neglected problem, the microscopy of concretes.
PORTLAND CEMENT
Le Chatelier's investigations of portland cement began to appear in 1882 (1); the date may be taken
as a starting point in the use of the microscope in portland cement technology. Bogue (2)
summarized progress from that time to the early 1930's in a way that indicates the importance of
microscopy in the development of research in portland cement:
"There cannot be said to have been any precise knowledge of the

constitution of clinker prior to the microscopic examinations of Le Chatelier
and Tomebohm (3). These examinations were substantially in agreement
with respect to the classification of phases, denoted by T omebohm, alite,
belite, celite,felite and an isotropic residue, but the chemical nature of the phases
has remained a matter of controversy up to recent years ..... .
"Alite was considered to be 3CaO·Si02 by Le Chatelier in 1884 and this view

has been confirmed by the investigations of Rankin and Wright (4) who
demonstrated the existence of the compound, Bates and Klein (5) who
showed its presence in clinker, Hansen and Brownmiller (6) who identified
its presence in clinker by X-ray, Guttmann and Gille (7) who separated the
alite from clinker, and many others. But in 1931, conceding the evidence of
186
--``,```,`,`,`,,,,`,`````,`,,`,,-`-`,,`,,`,`,,`---

Daneshlink.com
daneshlink.com
the X-ray means, Guttmann (8) closed the book with the proclamation 'At
last the fifty year old question ''What is Alite" is answered.' A year later
Janecke (9) withdrew his entry, 8CaO·Ah03"2Si02, and a half century after
the original announcement of Le Chatelier, the alite of clinker is generally
acknowledged to consist 'essentially' of tricalcium silicate.*
"Modem progress is the result chiefly of advances in the technique of

systematic study. Little could have been accomplished without the
generalizations provided by an application of the phase rule within the
various systems encountered in the complex of cement clinker. But the
successful application requires tools for the positive identification of the
phases that are produced by the heat treatments. The microscope was used
by Le Chatelier and Tornebohm, but improvements and new developments
in its use have opened up vastly greater possibilities in the resolution and
identification of the clinker constituents. The X-ray was first applied in
cement research in 1927 (1 0), but already it has become indispensable in the
recognition of the cement compounds and in the still more basic
understanding of the atomic rearrangements that occur in the cement kiln.''
The techniques employed in the classic investigations of the constitution of cement were immersion
and thin-section methods. While immersion mounts were most useful for the identification of
constituents, and thin sections for the display of textures and structures, neither was particularly
suited for the determination of quantitative relations among clinker constituents.
In 1934, Tavasci (11) demonstrated that the constituents could be distinguished in properly prepared
etched polished sections examined under vertical illumination. Insley (12) extended the application
of this metallographic technique, improving the etching methods, and developing polished thin
sections (13) which permitted accurate correlation of observations made in transmitted and reflected
light. He also adapted the linear traverse technique, using the Wentworth integrating stage (14)to
make quantitative estimates of the constituents in polished sections (15). Micrometric procedures
are difficult to apply to thin sections of clinker because the thickness of the section is greater than
the grain diameters of some of the constituents. Due to the resulting overlap, it is particularly hard
to distinguish the interstitial constituent which form the matrix between the cement silicates. On
polished sections the boundaries are distinct and measurements can be made and reproduced more
precisely.
Quantitative determinations of constituents both in laboratory and commercial clinker have been
made in a number of laboratories in this country and abroad. The original interest at the Bureau of
Standards was connected with studies aimed at the production of low-heat cement (12). The
Portland Cement Association Fellowship used this approach to determine the effects of heat
treatment and cooling rate on the structure of clinker (16). The Corps of Engineers included
micrometric analyses of clinkers among the procedures in a research program intended to develop
an improved specification for cement (17). Users of the method have tried persistently to improve it
*X-ray investigations recently published indicate some disagreements about crystal structure. See H.
O'Daniel and E. Hellner, "Structure of 3CaO·SiOz," Neues Jahrb, Mineral. Geol.,A, pp. 108-111 (1950);]. W. Jeffrey,
"Crystal Structure ofTricalciurn Silicate," Acta Crystallographica, V. 5, pp. 26-35 (1952); and "The Tricalciurn Silicate
Phase," paper No. 2, Third International Symposium on Chemistry of Cement, London, 1952.
--``,```,`,`,`,,,,`,`````,`,,`,,-`-`,,`,,`,`,,`---

187
Daneshlink.com
daneshlink.com
(16, 2), particularly by making variation in etching reagents and techniques with the aim of obtaining
a clearer discrimination among the interstitial constituents. When the purpose is to determine the
variation in composition with clinker-pebble size, whole pebbles are mounted and a cross-section
polished and etched. If the purpose is to determine composition micrometrically, samples of the
whole clinker, crushed to pass No.8 sieve and be retained on No. 16, are quartered, cast in Bakelite
plugs, and a section of plug is polished and etched.
A recent publication by Brown (18) contains an interesting account of modifications in technique,

observations of habit and association of constituents in clinkers, further discussion of the influence
of cooling conditions on structure and texture, and comparison of micrometric results with chemical
and physical tests. Nine entities could be recognized and measured in one integration on a
Wentworth stage with the arm extended to carry nine micrometer screws. The constituents*
measured were C 3S, C 2S, C3A, C~F, free CaO, free MgO, dark prismatic interstitial material, glass,
and undifferentiated. Dark prismatic is interpreted as essentially C3A crystallized under particular
conditions. The most satisfactory etching technique used distilled water at pH 6.8 to 7.0 applied by
pouring it on a velveteen-covered lap just ahead of the face of the specimen. The etched surfaces
produced were free of the dark f1lm often formed when the specimen is dipped in the etching
reagent, and permitted separation and measurement of eight of the phases in one integration instead
of the two or three necessary when several etches are used.
--``,```,`,`,`,,,,`,`````,`,,`,,-`-`,,`,,`,`,,`---
AGGREGATE
The description and classification of rocks for use as concrete aggregate is the petrography of
aggregates. The methods employed should depend on the complexity of the sample and the
economic and engineering importance of the intended use; thus the amount and kind of microcopy
in a petrographic examination should be adjusted to the information needed. Proper field and
laboratory sampling, sufficient sample size, and comparable calculation of results from sample to
sample are vital if petrographic examinations of aggregates are to yield satisfactory information
proportionate to the expense incurred in making the examination. While microscopy of some sort is
most important in almost every petrographic approach to aggregates, many other aspects must be
considered (19).
The petrographic investigation of building stone and roads tone has produced information still useful
in the interpretation of some aspects of the behavior of concrete aggregates. Lovegrove's
experiments (20) indicated that loss by attrition in the Deval machine and resistance to fracture by
impact were related to texture, mineral composition, and extent of alternation of the rocks tested.
Dense unaltered igneous rocks with microgranitic, micrographic, and ophitic textures, which are all
fine-grained and highly interlocked, showed the least loss by attrition; the greatest resistance to
fracture by impact was found in quartz-rich rocks with fine grained textures. Lord (21) reported that
deep-seated igneous rocks with granitic granular textures were harder but less tough than their fine-
grained near-surface equivalents. The rocks were classified by examining thin sections. Mineral
composition of medium-grained rocks was determined using a reticule divided into 100 squares
mounted in the ocular, counting the number of squares subtended by each mineral in the field at 55
X, and averaging 20 such determinations in one section. Fine-grained rocks were examined the
same way, using an objective of higher power to give a final magnification of 105 X. Rocks of
*The common abbreviated symbols for cement constituents use C = CaO, S = SiOz, A =Ah03, F =Fe20 3•
188
Daneshlink.com
daneshlink.com
variable or coarse grain size were examined at 30 X with an ocular grid of 25 squares; two or more
thin sections were used in each determination. The Wentworth stage (14) or the Chayes point-count
stage (22) now permit similar and probably superior estimates of mineral composition to be made in
less time with less fatigue.
The first record of petrographic description and classification of concrete aggregate located by the
author is C.W. Tomlinson's 1915 account (23) of a most interesting method that he developed under
the direction of M.O.Withey and A.M. Winchell for use in Withey's survey of the concrete aggregates
of Wisconsin. The procedure consisted of separating a sample of sand weighing about 300 g on
four sieves, separating each sieve-fraction into four specific-gravity classes by heavy liquids,
examining and counting each class in each size-range using a hand-lens or microscope, and
computing a weighted average composition for the sample from the weights and counted
compositions of the separates. One of the reasons given for separating the sand into sieve-fraction
is that "the resulting information is of value, since the composition of sand differs in the several
sizes according to definite principles"(23)- an early statement concerning an empirical law still
overlooked by many sedimentary petrologists but likely to be impressed on anyone who makes a
long series of petrographic examination of natural sand.
G.F. Loughlin (24) made one of the earliest investigations of a failure of concrete ascribed to the
properties of constituents in the aggregate. The concrete was cast stone made with aggregate
manufactured from altered anorthosite. Loughlin inspected the deteriorated concrete in place, visited
the quarry, examined thin sections of the aggregate and thin sections of unaffected and deteriorated
cast stone made from it. Considerable portions of the quarry sample and the sand in the
deteriorated concrete had altered to a zeolite, laumontite, which characteristically loses water and
falls to pieces when it is exposed to air. The sand in the unaffected concrete was relatively free of
that kind of alteration.
Loughlin next reviewed the qualifications of different kinds of natural stone for concrete aggregate
(25), writing from the viewpoint "of the geologist interested in the correct use of stone but
inexperienced in the mixing and handling of concrete." Perhaps his most valuable specific emphasis
was on the potentialities for trouble from day distributed in aggregate so as to be accessible to water.
Almost all of the rest of Loughlin's review seems to be based on the literature dealing with building
stone and roadstone and on considerations of the behavior of rocks exposed to natural weathering.
There seemed to be no recognition of the importance of particle surface or of the differences in
bond which particle surface can cause. The solubility of calcite was considered without mention that
the cement matrix of concrete may contain more easily soluble materials. But these omissions were
almost inevitable considering the stated viewpoint, and the paper probably represented as much of a
contribution as could be made without benefit of detailed examinations of concrete and without the
realization that performance of a particular aggregate in a particular concrete is a question not of the
properties of the aggregate a such, but of the interrelated properties of the aggregate and the matrix
in the circumstances of the exposure.
The interest aroused by this review caused several engineers to send Mr. Loughlin samples of
limestones with poor service records or suspicious test results from Pennsylvania, Illinois, Iowa, and
Kansas. The results of the examination of these samples became the subject of a paper published
the next year (26). Service records and determinations of absorption and loss in the sodium sulfate
test were available for most of the samples. Hand specimens and thin sections of the rocks were
examined and the amount of material insoluble in hydrochloric acid was determined. The mineral
--``,```,`,`,`,,,,`,`````,`,,`,,-`-`,,`,,`,`,,`---
189
Daneshlink.com
daneshlink.com
composition of the insoluble residues was investigated microscopically by C.S. Ross of the United
States Geological Survey. All of the limestones contained montmorillonitic clays. Variations in
amount of clay, size and distribution of clay pellets, absorption, and distribution of permeable
bedding planes and fractures appeared to explain the differences between beds in the same quarry
and from quarry to quarry. Small differences in amounts and arrangement of clay accounted for
differences in test results and in service. It is interesting to recall that this well-documented account
of troubles due to swelling clays in limestones was based on diagnosis made with the light
microscope, supplemented in one instance by chemical analysis.
Wuerpel and Rexford (27) investigated the correlation of absorption, bulk specific gravity, and loss
in the magnesium sulfate test of the chert in 12 samples of gravel. All of the chert examined was
composed of crytocrystalline quartz rather than chalcedony or opal. They found that loss in the
magnesium sulfate test increased with absorption and with decrease in bulk specific gravity. Using
thin-section, polished-section, and powder-mount techniques, they concluded that as bulk specific
gravity decreased, microscopically visible pores increased except in some porous cherts which had
some of their pores filled with hydrous iron oxide which increased the bulk specific gravity. But
these filled pores did not decrease the absorption or loss in magnesium sulfate by a comparable
amount.
The most extensive effort in recent years to relate mineral composition, texture, and amount and
kind of alteration of rocks to behavior in service and tests was made by the Knights (28) in their
book, "Road Aggregates." While their major emphasis is on aggregates for bituminous mixtures and
stone for sett paving, many of the relations between textures and alterations and physical behavior
are illuminating when applied to the interpretation of freezing-and-thawing test results.
ALKALI-AGGREGATE REACTION
In 1940, Stanton (29) reported instances of failure by expansion and cracking in concrete pavements
in California which had been observed from 1923 on, at first near the coast from Monterey to Los
Angeles County and later near the San Joaquin River. The failure of a section of pavement north of
Bradley, California, in 1938 was correlated with the use of sand from one source and cement
containing over 1 percent total alkali calculated as Na20. The laboratories of the California Division
of Highways made extensive engineering tests of the sand and of other California aggregates, and
developed a test procedure to determine reactivity. The procedure consisted of making mortar bars
of the aggregate and a high-alkali cement, storing the bars in a sealed container over water, and
measuring changes in their length at intervals up to one year.
The first rock demonstrated to cause expansion in this test was described as "siliceous magnesian
limestone" (30) or opaline dolomitic limestone from the Monterey group of California. Its mineral
composition was determined microscopically, and samples of each mineral were made up in mortar
bars for test by the method outlined above. Only the bars containing opal expanded (31). Further
microscopic examinations of other California sands demonstrated that all except one of the sands
which had been used in concrete that had shown this type of expansion and cracking contain opal in
small amounts, as opaline chert, opaline shale, or opalized wood.
Reports of similar failures accumulated soon after Stanton's paper appeared. Meinser (32) described
the cracking, expansion, and exudates found at Parker Dam two years after its completion in 1938.
--``,```,`,`,`,,,,`,`````,`,,`,,-`-`,,`,,`,`,,`---
190
Daneshlink.com
daneshlink.com
Careful examination of cores drilled from the dam indicated that the gel exudates were usually
associated with andesitic pebbles. Some andesite and rhyolite pebbles were found with dark outer
shells regarded as differing from those produced by weathering, in that the matrix of the rock in the
--``,```,`,`,`,,,,`,`````,`,,`,,-`-`,,`,,`,`,,`---
shells was soft but feldspar and mica phenocrysts were hard and fresh. A small amount of
secondary opal was found in some of the andesite but not in all of it. The aggregate had been tested
by the usual methods and by petrographic examination and had been regarded as satisfactory
material. The reactive ingredient in the andesite had not been identified; feldspar was suspected. In
discussion, Insley (32) pointed out that amorphous silica, as opal, had been demonstrated as reactive
in California, and andesite was a rock which often contained amorphous silica as volcanic glass.
Additional instances of expansion and cracking diagnosed a alkali reaction were recorded from
Virginia (33), the State of Washington (34), Nebraska, and Idaho (30). Stanton, Porter, Meder, and
Nicol(30) established, by examination of concrete and thin sections, that reaction had affected
andesites from Parker Dam, Friant Dam, a traffic circle at Fresno, Calif., and bridges in Washington
state; the common factor appeared to be the glassy groundmass of all of the andesites. They
synthesized andesite glass and found that mortar bars made with high-alkali cement and the
synthetic glass expanded and cracked. Following up indications Stanton had found, Coombs (35)
investigated the possibility of detecting aggregate which had reacted by means of ultraviolet light and
staining thin sections. Reacted concrete showed more fluorescent rims than unreacted, but the
results were not entirely satisfactory. Interpretation of the stained areas in sections was complicated
by the presence of clay minerals and hydrous iron oxides which absorbed the dyes, and it was
uncertain whether the dye absorption had anything to do with reactivity. Hornibrook, Insley, and
Shuman (36) found that tridymite and the natural glasses, obsidian and pitchstone, also produced
expansion of mortar bars.
Parsons and Insley (37), in 1944, reported investigations of the effects of alkaline solutions on
polished surface of rocks and mineral known to have reacted with alkalies in cement or suspected of
possible reactivity. They suggested a test for potential reactivity which consisted of immersing
polished specimens of rocks and minerals from the aggregate in a 10 percent NaOH solution
maintained at SO C for 18 to 24 hr; etching produced on minerals or rock except calcite or dolomite
would be an indication of potential reactivity. The test was not widely applied.
The identification of reactive materials based on tests in mortar bar and on diagnose of field
concrete established a category of materials which had been shown to be reactive. The list now
includes opal, chalcedony, tridymite, cristobalite, acid and intermediate volcanic glass, pyrex ( 38),
and silica gel (39). If chalcedony is assumed to consist of a submicroscopic mixture of fibrous low-
temperature quartz and interstitial opal (40), all the substances listed above may be described as
consisting of or containing "highly siliceous materials which are thermodynamically metastable at
ordinary temperatures." (40) That description does not appear to cover the hydromica (38) that is a
suggested reactive constituent of the Buck Dam phyllite. Very fine-grained quartz (38) had also
been suggested as the reactive constituent at Buck Dam, and its status appears uncertain (40) in
terms of reaction at normal temperature.
Hansen (39) advanced a hypothesis to explain the alkali-aggregate reaction, based on

experimentation, chemical considerations, and examination of mortars and concretes at low
magnification. It may be summarized as follows: Si02 as opal or as pure silica gel would react with
NaOH or KOH in the solution phase in the cement paste, presumably associating as complex

191
Daneshlink.com
daneshlink.com
silicate ions, assumed to be incapable of diffusing through the paste. The cement paste would then
act as a semipermeable membrane. It would allow water and the ions and molecules of the alkali
hydroxides to pass through it without allowing passage of the silicate ions and would thus form an
osmotic pressure cell. The alkali silicate formed on a silica particle would draw solution from the
pate and produce a pocket of liquid which would exert hydrostatic pressure. The silicates would
occupy greater volume than the original particle and thus increase pressure against the surrounding
paste. An additional reaction between alkali-silicate solution and Ca(OH)2 might occur to form
calcium silicate and regenerate the alkali hydroxides for further reactions. The transparent gel-like
product of reaction was regarded as the alkali silicate and the white opaque or porcelanous product
as the calcium silicate. The difference in behavior between Ca(OH)2 and alkali hydroxides with silica
was believed to be due to lower solubility of calcium silicates than of alkali silicates.
McConnell, Mielenz, Holland and Green (41) provided an interesting summary of criteria for the
recognition of alkali-reaction, with additional information on reaction products in concrete. They
reviewed their experience with many proposed tests for reactivity and gave a large accumulation of
results obtained in mortar-bar tests. The primary criterion cited as proof of alkali reaction was the
presence of gel reaction product in hand specimens or thin sections of the concrete. The gel ranged
from colorless and transparent to opaque and white, and in consistency from thin and watery to
rubbery to hard and brittle. It was found in voids or cracks, as rims around aggregate particles, or in
pore space in aggregate particles. The indices of refraction of the gel varied with moisture content,
ranging in desiccated gels from about 1.46 to 1.49. The second criterion given was the presence of
reaction rims around aggregate particles. To be applicable, it must be demonstrated that the rims
formed after the incorporation of the aggregate in the concrete and that the aggregate was a type
capable of reaction. The rims appeared dark in reflected light but more translucent in thin sections
than the interiors of the aggregate particles. The precautions necessary in applying this criterion
were clearly described.
Some differences of opinion about the location of the osmotic pressure cells hypothesized by
Hansen have been recorded. The petrographers of the Bureau of Reclamation (41,42) have
considered the sockets around reactive particles or the outer shells of the particles as the location of
the semipermeable membrane and have apparently regarded the pressures as caused by increase in
volume of the reacting aggregate particles. As supporting evidence, they have published a number
of photomicrographs of concrete in which the majority of the fractures appear to originate at
aggregate particles and die out in the matrix. Parsons and Insley (42) studied the movement of gels
formed by reaction between opal and high-alkali cement in pastes and mortars by microscopic
examination of broken surfaces and thin sections. They observed that, in their specimen, the gel
--``,```,`,`,`,,,,`,`````,`,,`,,-`-`,,`,,`,`,,`---
which formed around the opal grains moved into surrounding mortar and filled available voids;
cracks subsequently developed, apparently originating at the voids and dying out in the matrix. The
author has observed some cracks which appear to start from aggregate particles, some which appear
to start from voids, and many which did not appear to be related to anything in the plane of the
surface or section being examined.
One type of crack which appeared frequently in chert particles in the concrete ofTucaloosa Lock
(43) was later found in less marked degree in chert particles in mortar bars; it appears to be a useful
indicator of chert particles in which reaction has begun and does not appear to be a normal feature
of the Gulf coast chert gravel before they are incorporated in concrete. In the Tucaloosa cores,
many of the chert particles on sawed or drilled surfaces showed one crack, of greatest width in the
192
Daneshlink.com
daneshlink.com
interior of the particle, often dying away before the outside of the pebble or sand grain was reached,
but sometimes passing the boundary to dwindle to nothing in the mortar. Several stages were
observed and illustrated (43), including empty fractures confined to the aggregate, empty and pardy
filled fractures passing into the mortar, and some crack systems passing through several chert
pebble, wider in the pebbles and narrower in the mortar. The bond of the surfaces of such pebbles
to the mortar is usually extraordinarily good, much better than is to be expected of smooth polished
chert, so good that the concrete will frequendy break through rather than around the pebble. The
shape of the cracks, their usual location inside the pebbles, and the unusually good bond, tempt one
to imagine that they could have been formed by a tensile stress moving the whole exterior of the
pebble out away from the center.
Although an enormous amount of work has been done since 1940 in the attempt to control alkali-
aggregate reaction, to identify and characterize reactive constituents, and to explain the chemistry
and mechanics of the reaction, a great deal remains to be done. The evidence obtained by
microscopic examination of reacted concrete has been useful but is in its nature indirect; frequendy
it has been ambiguous, particularly regarding the mechanics of the expansion and cracking.
Fear of the possibility of alkali reaction has been an active influence in increasing the use of
petrographic examination in acceptance testing of concrete aggregate. The same fear has increased
the care and detail with which the examinations are conducted. The particular value of petrography
in this respect is that if an adequate sample is studied with proper care by a competent petrographer,
it can indicate definitely whether reactive constituents are present, and if found, their nature,
amount, and distribution with respect to particle size.
CONCRETE
Microstructure
Any microscopist working on cement hydrates, mortar, and concrete soon discovers that the
literature is widely scattered, hard to locate, and apparendy scanty. Much information of value to the
microscopist is found in publications with tides which do not so indicate. Many of the publications
of the Pordand Cement Association Fellowship and the Bureau of Standards may be overlooked on
that account.
In 1915 Rankin and Wright's paper (4) on the system CaO·Al2 0 3 ·Si02, Tomlinson's method (23) for
mineralogical analyses of sand, and two ofN. C. Johnson's papers (44) on the microstructure of
concrete appeared. Johnson had accumulated many specimens of concrete exposed to fresh and salt
water and had examined polished surfaces at magnifications up to 160 X. Apparendy he was not
interested in the composition or structure of pordand cement or in the optical identification of
hydration or alteration products. As a result of this disregard for the necessary basis of interpretation
he misinterpreted some of the phenomena he illustrated; some of the misinterpretations were
discussed at the time by Messrs. Bates and Spackman.
Some of the background needed was already available (45). Considerable additional information on
the nature and optical properties of the hydration products of pure compounds in cement (5, 46, 4 7)
and on crystalline substances common in hydrated cement and concrete (48, 49) became available in
American journals before 1935. Brown and Carlson's studies (50) of hydrated cements in thin
section are of great value because they describe and illustrate a wide range of variations in texture of
--``,```,`,`,`,,,,`,`````,`,,`,,-`-`,,`,,`,`,,`---

193
Daneshlink.com
daneshlink.com
hydrates, variations related to water-cement ratio, fineness, curing conditions, and admixtures, with
information about tensile or compressive strength of several of the groups and observations of the
behavior of the specimens during preparation. These photomicrographs and descriptions permit
comparisons of textural differences in specimens of recorded composition and age, cured under
laboratory conditions. Additional information on the effects of increased curing time on texture was
provided by Ward (51). While several interesting and valuable microscopic and chemical
investigations of deteriorated concrete that had been subjected to chemical and physical attack or
chemical attack (52, 53, 54) have appeared, the work by Brown and Carlson and the work by Ward
seem to provide the only published basis for extrapolation about undeteriorated hydrated cement or
mortar in concrete in thin sections.
It has seemed to the author that the texture and microstructure of concrete examined in thin section
should provide information as useful in the interpretation of the history of the concrete as is the
information provided by microscopic petrography in the interpretation of rocks - i f the
considerable problems of preparation of thin sections could be overcome, if a wide enough range of
comparative material could be assembled, and if the interpretation was made cautiously and checked
against all the other kinds of information that might be obtained about the concrete.
Thin sections of satisfactory quality can be prepared by cooking the blank in a synthetic resin until
the surface to be ground is impregnated, grinding the surface with progressively flner abrasives in
some fluid other that water (kerosene or isopropyl alcohol, depending on the solubility of the resin),
mounting the ground surface on a slide, allowing the mounted blank to cool slowly so that thermal
shock to the resin is avoided, and grinding the mounted bland to satisfactory thickness. Thickness
can be judged by the birefringence of the calcium hydroxide, which should show a maximum
interference color about fust-order yellow between crossed nicols. If calcium hydroxide is absent (as
it may be with thoroughly hydrated concretes containing an efficient pozzolan or in advanced cases
of alkali reaction or leaching), unhydrated C3S should show dark gray, or quartz in the aggregate,
pale to medium gray interference colors.
Such thin sections, of the order of 10 to 20 J..l thick, are desirable because the unhydrated cement
grains, minute aggregate particles, calcium hydroxide, and alteration products have a grain size so
small that they overlap and interfere in a thicker section, and the gel ground mass is often fairly dark
and opaque even in a section of less than standard thickness.
--``,```,`,`,`,,,,`,`````,`,,`,,-`-`,,`,,`,`,,`---
The constituents which can be recognized in thin sections of undeteriorated pordand-cement

concrete are aggregate particles, unhydrated cement, calcium hydroxide, and the amorphous gel
groundmass. The most common secondary alteration product is calcium carbonate which, in
concrete which has been exposed to air but not to excessive leaching or chemical stack, is usually
confined to thin zone near the outer surface of the structure and thin zones along cracks. High-
sulfate calcium sulfoaluminate is another common alteration product which may be found in small
quantities in air voids in normal concrete several years old, particularly if the concrete was made with
Type I or Type III pordand cement. A thin section of dense adequately consolidated concrete, under
10 years old, taken from a location inside the structure and away from cracks, will probably contain
only aggregate, unhydrated cement, calcium hydroxide, and gel groundmass - a restricted group of
constituents. The variations which can be observed from concrete to concrete are variations in
amount of each constituent, in size of constituents, particularly hydroxide crystals and crystal groups
and gel areas, and variations in arrangement of constituents. The close attention paid to the

194
Daneshlink.com
daneshlink.com
arrangement, crystal size, and abundance of calciutn hydroxide is because it is crystalline, anisotropic,
recognizable, and presutnably capable of yielding information indirectly about the state of the gel
and whether the concrete has been subject to leaching and chemical attack.
The four pairs of photomicrographs are reproduced as examples of the difference in texture and in
size of calciutn hydroxided crystals and gel areas which may be found among undeteriorated mass
concretes. They were selected as representative after comparisons of five sections &om each of four
cores, each core taken from exterior concrete of a different dam. Photographic data and information
about the concretes is shown in Table I. The concrete of cores A and B (Figs. 1 and 2) contained
limestone coarse and fine aggregates, while C and D (Figs. 3 and 4) contained granodiorite and
diorite coarse and fine aggregates, respectively. One conspicuous difference between the two sets of
pairs is the difference in crystal size of the calciutn hydroxide between the concretes with limestone
aggregate and the concretes with siliceious aggregate. Those with limestone aggregate have relatively
small hydroxide crystals and those with siliceous aggregate have relatively large ones. This general
relation appears to obtain in the comparisons made so far between undeteriorated concretes with
limestone aggregate and concretes with siliceious aggregate; it exists between concretes of the same
age and cement factor made with the same cement and what is intended to be the same gradation of
fme aggregate. Another difference which appears to be consistent between the concretes with
limestone aggregate and the concretes with siliceous aggregate is a difference in amount and particle
size distribution of material finer than the No. 200 sieve in the aggregate. The reported amounts of
aggregate finer that the No. 200 sieve were larger in A and B than inC and D; the concretes
preswnably contain what is reported plus some additional amount which is either that adhering to
coarse aggregate and not allowed for or that produced between the point where samples were taken
for sieve analysis and the final location of the concrete, or both.
The additional unreported fines can be expected to amount to more in limestone aggregates than in
siliceous aggregates, because of the differences in hardness between the minerals in the rocks. In the
crushing of brittle crystalline solids, the amount of new surface formed per unit of energy input
increases as the hardness on Moh's scale decreases (55). Calcite has a hardness of 3; quartz has a
hardness of 7; the range of hardness in the feldspars is from 6 to 6.5. The same expenditure of
energy yields about 1.8 times as much new surface on calcite as on quartz and feldspar (55).
Assuming that the limestones and the granodiorite and diorite were all similar in original grain size
and interlock and were subjected to the same amount of accidental impact during hatching and
mixing, the amount of new surface produced on the limestones should be about 1.8 times that
produced on the diorite and granodiorite, and the fmes should be proportional. As a matter of fact,
the limesotnes were of finer grain; the degree of interlock was hard to compare between limestones
and the diorite and granodiorite. Comparing these sections and all the other fairly comparable sets
that could be obtained, it did appear that material fmer than No. 200 was likely to be of different
particle size distribution, generally smaller in the limestone aggregates but varying from one
limestone to another.
The difference in crystal size of the hydroxide and the difference in abundance and size of the
aggregate passing the No. 200 sieve implies another point that the photomicrographs illustrate-
there are larger areas of gel uninterrupted by crystalline particles in the concrete with siliceous
aggregates. This last point seems to be correlated with the tendency to crack in the cores &om which
the sections were made; when sawed surfaces of the concretes were examined at 10 X after several
--``,```,`,`,`,,,,`,`````,`,,`,,-`-`,,`,,`,`,,`---
195
Daneshlink.com
daneshlink.com
months exposure to air, B had more cracks per 100 sq in. of surface than A, and A and B both had
more than either C or D.
Air Content
As the practice of entraining air in concrete has become more widespread particularly in the last ten
years, there has been increased interest in means of determining air content of plastic and hardened
concrete. Verbeck (56) described a method in which polished specimens with pigment rubbed into
the voids were mounted on a microscope equipped with a camera Iucida. Magnification in the
microscope was 50 X to 90 X; magnification in the camera-lucida image was greater and was
determined in order to calculate the average void area. The camera-lucida image of a field of known
dimensions was traced, using separate colors to distinguish voids from aggregate particles.
Planimeter measurements were made of the total area of air voids and of aggregate particles, and the
number of air voids was counted. These measurements, with the dimensions of the traced area and
the fmal magnification in the tracing, were used to calculate air content, average void area, and void
concentration. The measurement of four fields per section and two sections per specimen was
recommended to obtain representative results. The results obtained were in good agreement with
determinations of the air content of the plastic concrete.
In discussion of Verbeck's paper, Rexford (56) briefly described four modifications of the linear
traverse technique: one using 5 X to 10 X magnification of a ground surface on a fairly large
specimen such as a 6-in. diameter core or cylinder; another using photographic enlargement of a
fairly large specimen, with traverses measured on the print; the third using photomicrographs of thin
or polished sections taken at a magnification to make the smallest voids visible, measuring traverses
on the print; and the fourth using thin sections and a Wentworth integrating stage.
Brown and Pierson (57) have described a large motor-driven integrating stage working on the
principle of the recording micrometer stage with two motions, and a carriage accommodating
specimens up to 6 by 10 in. The specimens used are sawed ground slices of concrete. The traverses
are measured during observation of the specimen with a stereoscopic microscope at magnification
of 30 X to 40 X in oblique illumination. One micrometer is used to measure air and the other to
measure coarse aggregate and mortar. Brown and Pierson found that with this equipment,
measurement of a traverse length of 100 in. on a 6 by 8-in. specimen of concrete with aggregate of
1 Yz in. maximum size would determine the air content of the sample with a standard error of 0.4
percent. If a counter is added to record the number of voids per traverse, it is possible to calculate
the bubble spacing factor, which Powers (58) has shown to be important to the frost resistance of
concrete. A similar integrating stage has been constructed and used by the Missouri State
Highwaydepartment (57).
An alternate approach of comparable precision uses a point-counter stage of the type described by
Chayes (22, 57). In the linear traverse method, the lengths of intercept of each constituent are
measured on equally spaced traverses and the lengths for each constituent accumulated. In the
point-count method, each traverse is broken down to a series of equally spaced points. The
constituent found at each point in the grid is identified and recorded on a tabulating counter with
one key assigned to each constituent. The accumulated totals show the number of times each
constituent was encountered at an intersection of the grid. The principle of the point-count can be
used in the determination of proportions of constituents in any heterogeneous solid; it is much
quicker and less fatiguing to the operator than a micrometer stage, and fairly extensive tests of its
precision on thin sections of granites have already been made (59, 60). The petrographers of the
--``,```,`,`,`,,,,`,`````,`,,`,,-`-`,,`,,`,`,,`---

196
Daneshlink.com
daneshlink.com
Waterways Experiment Station (57) and Mr. Willis of the Missouri State Highway department (61)
have found that determinations of air content of concrete by point count and linear traverse are
comparable and comparably precise. So far it has not been adapted to determinations of bubble
spacmg.
CONCLUSION
--``,```,`,`,`,,,,`,`````,`,,`,,-`-`,,`,,`,`,,`---
From the days of Le Chatelier to the present, a number of ingenious and enterprising students have
contributed ideas and information to the microscopy of concrete and its constituents. Much remains
to be done in terms of observation and description of microstructure, relation of microstructure to
larger structures, and detailed interpretation of the phenomena observed.
REFERENCES
1) H. Le Chatelier, Comptes rendus hebdomadaired des seances de 1'academie des sciences,

Vol. 94, p. 13 (1882).
2) R.H. Bogue, "Constitution of Portland Cement Clinker," Symposium on the Chemistry of

Cements, Stockholm, Sweden, P. 59 (1939). Discussion by T.W. Parker, p. 113.
3) A.E. Tomebohm, Tonindustrie-Zeitung, Vol. 21, p. 1148 (1897); Baumaterielenkunde, Vol

6, p. 142 (1910-11); Zement, Vol. 4, p. 287 (1903).
4) G. A. Rankin and F.E. Wright, "The Ternary System Ca0Al20 3·Si02," Am. Journal Science
(4) Vol 39, p. 1 (1915).
5) P.H. Bates and A.A. Klein, "Properties of the Calcium Silicates and the Calcium Aluminate
Occurring in Normal Portland Cement," Technologic Paper No. 78, Nat. Bureau Standards
(1917).
6) W.C. Hansen, "Further Studies on Portland Cement Compounds by the X-Ray Diffraction
Method," Journal Am. Ceramic Soc., Vol 11, p. 68 (1928).
7) A. Guttmann and F. Gille, Zement, Vol 16, p. 921 (1927); Vol. 17, p. 296 (1928).
8) A. Guttmann and F. Gille, "Closing Remarks on Alike Problem," Zement, Vol. 20, p. 144
(1931).
9) E. Janecke and R. Brill, "The Question of the Existence of the Tricalcium Silicates,"
Zement, Vol. 21, p. 380 (1932).
10) E.A. Harrington, "X-Ray Diffraction Measurements of Some of the Pure Compounds of
Portland Cement," American Journal Science, (5), Vol. 13, p. 467 (1927).
11) B. Tavasci, "Richerche Sulla Costituzione del Clinker di Cemento Portland," Giomale
Chimica Industriale, ed. Applicata, Vol 16, p. 538 (1934).
197
Daneshlink.com
daneshlink.com
12) H. Insley, "Structural Characteristics of Some Constituents ofPordand Cement Clinker,"

Journal of Research, National Bureau of Standards, Vol. 17, p. 353, 1936 (RP 917).
13) H. Insley and H.F. McMurdie, "Minor Constituents in Pordand Cement Clinker," Journal of
Research, National Bureau Standards, Vol. 20, p. 173, 1938 (RP 1074).
14) C.K. Wentworth, "An Improved Recording Micrometer for Rock Analysis," Journal of
Geology, Vol. 31, p. 228 (1923).
15) H. Insley; E.P. Flint; E.S. Newman; and J.A. Swenson, "Relation of Compositions and Heats
of Solution ofPordand Cement Clinker," Journal of Research, National Bureau Standards,
Vol. 21, p. 355, 1938 (RP 1135).
16) G.W. Ward, "Effect of Heat Treatment and Cooling Rate on the Microstructure of Pordand
Cement Clinker," Journal of Research, National Bureau Standards, Vol. 26, p. 49,1941 (RP
1358).
17) Cement Durability Program, War Department Corps of Engineers, U.S. Army, North
Adantic Division, First Interim Report, Mount Vernon, N.Y., June, 1942.
18) L.S. Brown, "Microscopical Study of Clinkers," Long-Time Study of Cement Performance in
Concrete, Bulletin No. 26, Research Laboratory Pordand Cement Association, Proceedings,
American Concrete Institute, Vol. 44, p. 877 (1948).
--``,```,`,`,`,,,,`,`````,`,,`,,-`-`,,`,,`,`,,`---
19) K. Mather and B. Mather, "Method of Petrographic Examination of Aggregates for

Concrete," Proceedings, American Society for Testing and Materials, Vol. 50, p. 1288 (1950).
20) Lovegrove, Surveyor, Vol. 28, No. 721-728 (1905).
21) E.C.E. Lord, "Examination and Classification of Rocks for Road Building," Department of
Agriculture, Office of Public Records, Bulletin No. 37 (1911).
22) F. Chayes, "A Simple Point-Counter for Thin-Section Analysis," American Mineralogist,
Vol. 34, p. 1 (1949).
23) C.W. Tomlinson, "Method of Making Mineralogical Analyses of Sand," Bulletin No. 101,
American Institute Mining and Metallurgical Engineers, p. 947, May, 1915.
24) J.C. Pearson and G.F. Loughlin, "An Interesting Case of Dangerous Aggregate,"
Proceedings, American Concrete Institute, Vol. 19, p. 142 (1923).
25) G.F. Loughlin, "Qualifications of Different Kinds of Natural Stone for Concrete
Aggregate," Proceedings, American Concrete Institute, Vol. 23, p. 319 (1927).
26) G.F. Loughlin, "Usefullness of Petrology in the Selection of Limestone," Rock Products,
Vol. 31, p. 50 (1928).
198
Daneshlink.com
daneshlink.com
27) C.E. Wuerpel and E.P. Rexford, "The Soundness of Chert as Measured by Bulk Specific
Gravity and Absorption," Proceedings, American Society for Testing and Materials, Vol. 40,
p. 1021 (1940).
28) B.H. Knight and R.G. Knight, "Road Aggregates, Their Uses and Testing," Edward Arnold,
London, (1948).
29) T.E. Stanton, "Expansion of Concrete Through Reaction of Cement and Aggregate,"
Proceedings, American Society of Civil Engineers, Vol. 66, p. 1781-1811 (1940).
30) T.E. Stanton; O.J. Porter; L.C. Meder; and A. Nicol, "California Experience with the
Expansion of Concrete Through Reaction Between Cement and Aggregate," Proceedings,
American Concrete Institute, Vol. 38, p. 209 (1942). Discussion by H.E. Meissner, pp. 236-1.
Authors' Closure, p. 236-17.
31) T.E. Stanton, Proceedings, American Society of Civil Engineers, Vol. 67, pp. 1402-1418,
discussion of own paper in Vol. 66, pp. 1781-1811 (1941).
--``,```,`,`,`,,,,`,`````,`,,`,,-`-`,,`,,`,`,,`---
32) H.E. Meissner, "Cracking in Concrete due to Expansive Reaction Between Aggregate and
High-Alkali Cement as Evidenced in Parker Dam," Proceedings, American Concrete
Institute, Vol. 37, p. 549 (1941). Discussion by H. Insley, pp. 568-3.
33) A.H. Kammer and R.W. Carlson, "Investigation of Causes ofDelayed Expansion of
Concrete in Buck Hydroelectric Plant," Proceedings, American Concrete Institute, Vol. 37,
p. 665 (1941).
34) Bailey Tremper, "Evidence in Washington of Deterioration of Concrete Through Reactions

Between Aggregates and High-Alkali Cements," Proceedings, American Concrete Institute,
Vol. 37, p. 673 (1941).
35) H.A. Coombs, "Expansion of Concrete due to Reaction Between Andesitic Aggregate and
Cement," American Journal Science, Vol. 240, p. 288 (1942).
36) F.B. Hornibrook; H. Insley; and L. Shuman, "Notes on the 'Effects of Alkalies in Portland
Cement on the Durability of Concrete,"' Proceedings, American Society for Testing and
Materials, Vol. 43, p. 218 (1940).
37) W.H. Parsons and H. Insley, "Alkali Etching Tests on Concrete Aggregates," Proceedings,
American Concrete Institute, Vol. 40, p. 229 (1947).
38) R.F. Blanks, "Modern Concepts Applied to Concrete Aggregate," Transactions, American
Society of Civil Engineers, Vol. 115, p. 403 (1950). D. McConnell; R.C. Mielenz; W.Y.
Holland; and K.T. Greene, "Petrology of Concrete Affected by Cement-Aggregate
Reaction," Application of Geology to Engineering Practice, Geological Society of America,
1950.
199
Daneshlink.com
daneshlink.com
39) W.C. Hansen, "Studies Relating to the Mechanism by Which the Alkali-Aggregate Reaction
Produces Expansion in Concrete," Proceedings, American Concrete Institute, Vol. 40, p.
213 (1944).
40) B. Mather, "Petrographic Identification of Reactive Constituents in Concrete Aggregate,"

Proceedings, American Society for Testing and Materials, Vol 48, p. 1120 (1948). Discussion
by H. Insley, p. 1129.
41) D. McConnell; R.C. Mielenz; W.Y. Holland; and K.T. Greene, "Cement-Aggregate Reaction
in Concrete," Proceedings, American Concrete Institute, Vol. 44, p. 93 (1947).
42) W.H. Parsons and H. Insley, "Aggregate Reaction with Cement Alkalies," Proceedings,
American Concrete Institute, Vol. 44, p. 625 (1948). Discussion by McConnell et al., p. 632-
1.
43) B. Mather, "Cracking of Concrete in the Tuscaloosa Lock Alabama," Proceedings, Highway
--``,```,`,`,`,,,,`,`````,`,,`,,-`-`,,`,,`,`,,`---
Research Board, Vol. 31, p. 218 (1952).
44) Nathan C. Johnson, "Use of the Microscope in Examining Concrete," Engineering News
Record, Vol. 71, pp. 98, 160, 194, 263,301,320 (1915); "The Microstructure of Concrete,"
Proceedings, American Society for Testing and Materials, Vol. 15, Part II, p. 172 (1915).
45) A.A. Klein and AJ. Phillips, "Hydration ofPordand Cement," Technologic Paper No. 43,
National Bureau of Standards (1915).
46) R.J. Colony, "The Products of Hydration and Hydrolysis of Pordand Cement," Part I,
Bulletin No.4, Civil Engineering Testing Lab, Columbia University, New York, N.Y. (1931).
47) R.H. Bogue and W. Lerch, "Hydration ofPordand Cement Compounds," Ind. and Eng.
Chern., Vol. 26, p. 837 (1934).
48) F.W. Ashton and R. Wilson, "The Preparation and Optical Properties of Calcium Hydroxide
Crystals," American Journal Science, Vol. 209, p. 13 (1927).
49) W. Lerch; F.W. Ashton; and R.H. Bogue, "The Sulfo-aluminates of Calcium," Journal of
Research, National Bureau of Standards, Vol. 2, p. 715 (1929) (RP 54).
SO) L.S. Brown and R.W. Carlson, "Petrographic Studies of Hydrated Cements," Proceedings,
American Society for Testing and Materials, Vol. 36, p. 332 (1936).
51) R.H. Bogue, "The Chemistry of Portland Cement," Reinhold Publishing Corp., New York,
N.Y., pp. 446-454 (1947). (Discussion of unpublished work by G.W. Ward).
52) E.A. Finney, "Microscopic Investigation of Scaled and Unsealed Concrete," Bulletin No.
101, Michigan Engineering Experiment Station (1944).
53) R.D. Terzaghi, "Concrete Deterioration due to Carbonic Acid," Journal, Boston Society
Civil Engineers, Vol. 44, p. 977 (1948).

200
Daneshlink.com
daneshlink.com
54) R.D. Terzaghi, "Concrete Deterioration due to Carbonic Acid," Journal, Boston Society
Civil Engineers, Vol. 36, p. 136 (1949).
55) J.W. Axelson, et al., "Basic Laboratory Studies in the Unit Operation of Crushing,"
Transactions, American Institute Mining and Metallurgical Engineers, Mining Engineering,
Vol. 3, p. 1061 (1951). (Technical Publication 3194B.)
56) G J. Verbeck, "The Camera-Lucidia Method for Measuring Air Voids in Hardened
Concrete," Proceedings, American Concrete Institute, Vol. 43, p. 1025 (1947). Discussion by
E.P. Rexford, pp. 1040-1041.
57) L.S. Brown and C.U. Pierson, "Linear Traverse Technique for Measurement of Air in
Hardened Concrete," Proceedings, American Concrete Institute, Vol. 4 7, p. 117 (19 50).
Discussion byT.F. Willis, p. 124-1; K. Mather, p. 124-2.
58) T.C. Powers, "The Air Requirement of Frost-Resistant Concrete," Proceedings, Highway
Research Board, Vol. 29, p. 184 (1949).
59) F. Chayes, "Modal Analyses of the Granite and Diabase Test Rocks," Part 5, Bulletin 980,
U.S. Geological Survey, pp. 59-68 (1951).
60) F. Chayes and H.W. Fairbairn, "A Test of the Precision of Thin-Section Analysis by Point-
Counter," American Mineralogist, Vol. 36, p. 704 (19 51).
61) T.F. Willis, "Measuring Air in Hardened Concrete," Proceedings, American Concrete
Institute, Vol. 48, p. 901 (1952).
DISCUSSION
A Member--How long does it take to make a particle constituent count by both methods?
Mrs. Katharine Mather (author}--It depends on the number of constituents that you want to deal with.
For instance, if there are four constituents to be counted in a thin section of granite, an experienced
operator using the point-count method ought to be able to cover one thin section, which is an area
of about % sq in., in 20 min. If for some reason you have a more complex system and a count of all
the different constituents is desired, the time will increase somewhat. The time, using linear traverse
techniques, would be two or three times as great.
Mr. A. Borowik*-How can the data of microscopical examination be correlated with those of
strength or durability of concrete?
Mrs. Mather--The factor which seems to make the greatest difference in durability of otherwise
comparable concretes is air content. Probably the air content of the durable and nondurable
concrete should be measured micrometrically.
• Metallurgist, Frankford Arsenal, Philadelphia, Pa. and Associate Professor, Villanova College, Villanova, Pa.
--``,```,`,`,`,,,,`,`````,`,,`,,-`-`,,`,,`,`,,`---
201
Daneshlink.com
daneshlink.com
Strength might be indicated by freedom from microfractures, by denseness of texture, by absence of

alteration and deterioration products, and by degrees of hydration of the cement-how much
unhydrated cement there is left.
Of course, one cannot look at a thin section concrete and deduce the mixture design with accuracy.
Usually some difference can be found between two concretes which will at least suggest an
explanation of the differences in their behavior.
Mr. C. W. Mason**-How much is done in the study of the fractured surface of the concrete?
Mrs. Mather--The fractured surfaces of concrete are extremely useful in a number of way. They are
useful for examination at low magnification to determine gross structure. They are useful when one
is examining deteriorated concrete to locate deterioration products for subsequent examination with
a polarizing microscope. I do not believe that anybody has done much work with high power-say
over 60 X magnification of fractured surfaces.
•• Professor of Chemicall'vficroscopy, Cornell University, Ithaca N.Y.

202 --``,```,`,`,`,,,,`,`````,`,,`,,-`-`,,`,,`,`,,`---
Daneshlink.com
daneshlink.com
Table 1-Data for Photomicrographs

Photographic Data: American Optical Co. 16 mm chromatic objective, 10 X Hyperplane ocular; 40 w Zirconarc lamp at
normal setting, Wratten B Filter; Isopan film; exposure time V2 sec for plane light views, 1 1/2 for crossed nicols.
Magnification 165 X.
Concrete Data·
Dam Water-Cement Actual Cement Fine Aggregate Age of Concrete
Ratio by Factor, bags Aggregate, Finer than no at Examination,
Weight peryd3 percent by 200 Sieve, months
Weight of percent of Fine
Aggregate Agg!"egate
A .......... 0.51 3.9 26 4.8 8
B .......... 0.59 3.8 26 8.1 19
c .......... 0.54 3.9 24 3.2 11
D ......... 0.56 3.8 26 3.2 5
Note-Type II portland cement was used tn all four concretes. The cements came from different mills.
(a) Plane light. (b) Crossed nicols

Fig. !-Mass Concrete Dam A (X/65).
In (a) V marks voids, A aggregate; the abundant small dark irregular grains in the matrix are residual and unf!)ldrated cement.
In (b) the larger calcium f!)ldroxide crystals are marked H. Hydroxide forms narrow borders on the two largest aggregate grains
and an almost complete rim around the right void The most irregular small light spots in the dark gel background are small
f!)ldroxide crystals and the minute light dots, limestone dust.
--``,```,`,`,`,,,,`,`````,`,,`,,-`-`,,`,,`,`,,`---

~03
Daneshlink.com
daneshlink.com
--``,```,`,`,`,,,,`,`````,`,,`,,-`-`,,`,,`,`,,`---
(a) Plane light. (b) Crossed nico/s.

Fig. 2 - Mass Concrete Dam B (X165).
In (a) V marks voids; L, limestone, and the dark rounded grains in the matrix are unf!Jdrated cement. In (b) the larger
f!Jdroxide crystals are marked H, and one large uni!Jdrated cement grain with a f!Jdroxide rim, C The dark gel groundmass is
almost lost, obscured l?J the myriad rif limestone dust particles and some small f!Jdroxide crystals.
(a) Plane light. (b) Crossed nicols.

Fig. 3 - Mass Concrete Dam C (X165).
~regale is marked A in (a) and f!Jdroxide is marked H in (b). The crystals rif ~droxide are large, irregular in shape, and
widefy separated in the dark gel background. Compare with Figs. 1 and 2 for the differences between concrete with siliceous
aggregate and concrete with limestone aggregate.

204
Daneshlink.com
daneshlink.com
--``,```,`,`,`,,,,`,`````,`,,`,,-`-`,,`,,`,`,,`---
(a) Plane light. (b) Crossed nicols.

Fig. 4 - Mass Concrete Dam D (X 165).
Voids are marked V and aggregate A, in (a). The clear hexagonalgrain, C, is tricalcium silicate, two dicalcium silicate grains
are labeled C in (b). The aggregate grain at the middle upper ma'l,in shows cross-hatched as well as lamellar plagioclase
twinning, with crossed nicols. Hydroxide crystals are marked H in (b); the three lafl,est near the center of the field surround or
part!J surround several unfydrated cement grains. The dark gel groundmass is almost free of minute fydroxide crystals and
minute aggregate particles.
205
Daneshlink.com
daneshlink.com
--``,```,`,`,`,,,,`,`````,`,,`,,-`-`,,`,,`,`,,`---

Daneshlink.com
daneshlink.com
PART VI- SPECIFICATIONS FOR USE OF CONCRETE IN

TRANSPORTATION STRUCTURES
Transportation structures are an extremely important component of the civil infrastructure.

Concrete is widely used for transportation structures and the reference specifications for use of
concrete in transportation structures are essential for new construction as well as maintenance of
existing systems.
The Transportation Research Board (TRB) is a division of the National Research Council (NRC),
which serves as an independent advisor to the federal government. TRB activities annually draw
more than 40,000 engineers, scientists, practitioners from public and private sectors, and academia.
Bryant Mather was recognized by the Transportation Research Board (TRB) to be a distinguished
lecturer. In 1993, for his distinguished lecture at TRB, he was introduced with the following
statement:
Bryant Mather is the director of the Structures Laboratory at the U.S. Army Corps ofEngineers
Waterwqys Experiment Station in Vicksbut;g, Mississippi.
An emplqyee of the Corps ofEngineers since 1941, Mr. Mather first worked as a geologist and
later as an engineer. Throughout his career, he has specialized in concrete research. His first
assignment was with the Central Concrete Laboratory at the U.S. Military Academy, West Point,
and at Mount Vernon, New York. He has been qffiliated with the Waterwqys Experiment
Station since 1946.
He received a bachelor's degree in geology from Johns Hopkins Universi!J, where he also did
graduate work in geology. He also did graduate work in economics at American Universi!J. Mr.
Mather received an honorary doctorate from Clarkson Universi!J in 1978.
Throughout his career, Mr. Mather has been active in ASTM, the American Concrete Institute
(ACI), and TRB. He is a member of several ASTM committees and has a total equivalent of
more than 162years of service on nine ACI technical committees. He is one of on!J three
individuals to have seroed as president of both ASTM and ACI.
For TRB, Mr. Mather seroed as Chairman of the Curing Concrete Committee from 1971 to
1974 and Chairman of the Concrete Division from 1963 to 1969. He received the Roy W Crum
Distinguished Service Award in 1966. His colleagues elected him to honorary membership of the
TRB concrete committees, on which he remains active. He also seroedfor matryyears on the
NCHRP SP20-5 Panel on Synthesii]ng Information on Highwqy Projects. Additional!J, he has
been a member of the Cement and Concrete/Structures Advisory Committee ofthe Strategic
Highwqy Research Program since the program's inception in 1984.
As one of the foremost experts on concrete, Mr. Mather has received mai!J government and industry
awards. His awards include the Department of the Army's Meritorious Civilian Service Award
and the Decoration for Exceptional Civilian Service. President Carter made him a charter member
ofthe government's Senior Executive Service in 1979. ASTM has bestowed on Mr. Mather its
--``,```,`,`,`,,,,`,`````,`,,`,,-`-`,,`,,`,`,,`---

207
Daneshlink.com
daneshlink.com
Award ifMerit, the Sanford E. Thompson Award, the William T. Cavanaugh Award, the
Frank E. Richart Award, the Charles B. Dudlry Award, and honorary membership. ACI
presented him with the Henry C. Turner Medal and honorary membership. He is an honorary
fellow rifEngland's Institute if Concrete Technology. In 1992 he was elected to the National
Academy rifEngineering.
Mr. Mather was the Henry M. Shaw Lecturer at North Carolina State University in 1967, the
Stanton Walker Lecturer at the University ifMaryland in 1969, the ASTM Edgar Marbut;g
ucturer in 1970, the IBM Visiting Scholar at Northeastern University in 1985, and a visiting
professor at the Science University ofTofgo, Japan, in 1988. He is author or co-author if more
than 500 technical reports andpapers.
Mr. Mather is recognized internationallY as 'Mr. Concrete" because if his great contributions and
dedicated interest in concrete technology transfer. With more than 50 years if a highfy committed
and most productive careerfor the U.S. Army Corps if Engineers, he continues to be a world-
renowned researcher, lecturer, consultant, and activist in consensus standards-writing ot;ganizations.
The introduction above is reproduced in the publication as it summarizes the breadth of his
professional contributions and accomplishments.
Bryant was keenly aware of the importance of good, clear, and unambiguous specifications and their
impact on construction. For his distinguished lecture at TRB, he focused on transportation
structures.
--``,```,`,`,`,,,,`,`````,`,,`,,-`-`,,`,,`,`,,`---
His paper, "Concrete in Transportation: Desired Performance and Specifications," was selected to
be reproduced in this compilation publication. In this paper, Bryant concluded that one should:
1. Decide what sort and levels of imperfections are tolerable and intolerable,
2. Understand what causes imperfections that are intolerable,
3. Evaluate the environment in which the concrete is to serve to recognize the presence of
influences that must be resisted if the concrete is to perform as desired,
4. Prepare concrete specifications that require appropriate levels of relevant properties of
concrete so that the concrete can resist the deteriorative influences it will encounter in
service, and
5. Ascertain that both the contractor's quality control and the owner's quality assurance
systems work to ensure that the concrete produced is as specified.

208
Daneshlink.com
daneshlink.com
--``,```,`,`,`,,,,`,`````,`,,`,,-`-`,,`,,`,`,,`---
Reprinted, with permission, from Transportation Research Record 1382, 1993, ©Transportation Research Board,
National Research Council, Washington D.C.
CONCRETE IN TRANSPORTATION: DESIRED PERFORMANCE AND

SPECIFICATIONS
by Bryant Mather
INTRODUCTION
This lecture is about concrete - specifically, hydraulic-cement concrete. If one starts with the dry
powder that is hydraulic cement- usually the particular class of hydraulic cement known as
portland cement- and adds water, what results, depending on the amount of water added, is cement
paste or grout. Grout can be poured like gravy. If fine aggregate is added, the result is mortar or
sanded grout. If both fine aggregate and coarse aggregate are added, the result is concrete. As the
Supreme Court of Pennsylvania once wrote in a decision dealing with cement-manufacturing plants,
"cement is to concrete as flour is to fruitcake." My first point is, to get proper concrete, get the
terminology right. There is no such thing as a cement mixer. And sand is not a synonym for fine
aggregate; sand is a class of fine aggregate produced by nature rather than by rock crushers and
grinding mills.
Having dealt briefly with terminology, I will provide the procedure for obtaining concrete that has
the desired performance. It is, simply, to include the relevant requirements in the contract
documents and ensure through proper contractor quality control and proper owner quality
assurance that the requirements of the specifications are followed. In fewer words, order the
concrete that you need and make sure that you get it.
DEVELOPMENT OF HARDENED CONCRETE PROPERTIES
Robert Philleo, who in 1986, as chairman, signed the certificates making Katherine Mather and me
honorary members of the Concrete Section of TRB and all of its committees, also in 1986 described
the fundamental features of concrete. The following is paraphrased from his papers (1,2). For
concrete to stiffen, harden, and develop strength, there must be a chemical reaction between the
constituents of the cement and the mixing water. This reaction causes the anhydrous calcium
silicates in the cement to be converted into calcium silicate hydrate (CSH), the cement gel or
hydrated cement. The critical feature of this reaction is that if the ratio of the volume of water to the
volume of cement is 1.2, then all the water and all the cement can combine and all the originally
mixing water-filled space can be filled with hydration product. A water-cement ratio (w/ c) of 1.2 by
volume is 0.4 by mass. If the w/ cis higher than 0.4, even if all the cement hydrates, there will always
be some residual originally mixing water-filled space that can hold freezable water. If thew/ c
is lower than 0.4, some of the cement will always remain unhydrated but, in theory, all the originally
mixing water- filled space could be filled with CSH. There is a misconception, often stated, that it
takes only the amount of water in a 0.2 w/ c paste to hydrate all of the cement. This is based on the
fact that only 0.2 unit of water by mass chemically combines with cement during hydration.
However, for a given volume of cement to hydrate, there must be an amount of originally mixing
water-filled space equal to 1.2 times the volume of the cement. This is because the hydration
209
Daneshlink.com
daneshlink.com
product has about 30 percent pore space that must be present, and water must be available to fill it.
If the amount of originally mixing water-filled space is less than that provided at a w/ c of 0.4, not all
of the cement can hydrate, even though only half of that water will go into chemical combination.
A lot of modem "high-performance" concrete made at w/ r! s well below 0.4 by mass will not, in
fact, have all of the originally mixing water-filled space filled with hydration product. This is because
the part of the mixing water that ends up in the gel pores undergoes a 10 percent reduction in
volume because the pores are so small and the water is adsorbed. Philleo's 1991 discussion goes into
the implications of the fact that additional water may enter while test specimens of very low w/ c
concrete are curing in the laboratory, but such externally available water will not get very far into
larger masses of field concrete even if it is made available. It is worth noting that even though I am a
well-known advocate of membrane curing for field concrete, I put in the American Concrete
Institute (ACI) standard for curing concrete (ACI 308-81) Section 3.2.2 that the use of liquid
membrane-curing compounds should not be approved "when the concrete has a water-cement ratio
of 0.4 or less."
In these comments there may have been the implication that something is desirable about hydrating
all the cement in a concrete mixture. If thew/cis higher than 0.4, then the more of the cement that
hydrates, the larger the proportion of the originally water-filled space that gets filled, the higher the
strength, and the lower the permeability. Therefore, if the concrete needs all the strength and
reduced permeability that it can get, then the longer it is kept moist, the closer it will come to having
all its cement hydrated and the greater its strength and impermeability will be. There are, however, at
least two possibly undesirable consequences of a concrete's having all of its cement hydrated and
becoming as strong it can become. First, if there is no remaining unhydrated cement, then there can
be no autogenous healing of micro-fractures as water enters and is available to react with unhydrated
cement to accomplish the healing. Second, if the concrete is as strong as it can possibly become, it
will have higher modulus of elasticity, be more brittle, and crack at a lower strain level. Reduced
strain capacity - both elastic strain and creep strain - is not a desirable property in a lot of
concrete. As Philleo put it,
"the desirable magnitude of creep is an issue on which practitioners have not
agreed. Structural engineers find it a nuisance they could easily do
without ... On the other hand, builders of unreinforced mass concrete
structures find creep an indispensable property of concrete.... Creep
redistributes stresses ... permitting highly stressed regions to shed some of
their stresses to low-stressed regions before cracking occurs ... [Mass
concrete] structures could not survive if concrete behaved elastically. (1)"
Those further interested in concrete should read Philleo (1 ,2).
CLASSES OF DETERIORATION
Very little concrete fails to provide the desired level of performance because its mechanical strength
is intrinsically insufficient. Indeed, as noted earlier, one of the undesirable things that can happen to
concrete is that, by getting too strong, it can also get too brittle and have an undesirably low strain
capacity before fracturing. Consequently, the efforts to achieve the desired performance should be
those that avoid the problems generally comprehended under the subject "durability."
--``,```,`,`,`,,,,`,`````,`,,`,,-`-`,,`,,`,`,,`---

210
Daneshlink.com
daneshlink.com
I have spent a great deal of time over a good many years explaining to people that concrete has no
property called durability. Any concrete, no matter how unsuitable for use in many environments,
will be completely durable in some environment. I have argued that any concrete strong enough to
resist the loads to which it will be subjected in service will also be durable in that service, regardless
of all other considerations, if it is allowed to get dry and to stay dry. "Dry" in this context means that
the evaporable water is allowed to and does escape and the internal relative humidity drops below 80
percent, under which conditions there are no longer any chemical reactions that can take place and
there is no water that can freeze. Earthquakes and possibly fire could damage such concrete, of
course, but generally speaking, the resistance to earthquakes comes under the heading "if it is strong
enough" and the damage due to fire is much lower in concrete that is dry than in concrete that is
not.
We may thus explore durability as it relates to concrete in transportation with the assumption that
the concrete that we are talking about is being used in an environment in which rarely, if ever, are
the structures allowed to get dry and stay dry. However, remember that if we have any such
structures, we are in a fortunate position with respect to the steps we need to take to achieve
satisfactory performance.
I suggest classifying the causes of deterioration of concrete into two categories. The first category
includes those causes of damage in which the cause acts on the concrete, causes the damage, and
then ceases to act usually for a very long time or forever. These causes include earthquakes,
tornadoes, hurricanes, fire, and lightning strikes. Once a gasoline truck has burned up on a
pavement, the damage done to that concrete by that fire is the damage that can be discovered by
simply examining the affected structure. Any concrete that has not been damaged by that influence
is not going to be damaged by that event subsequently. Therefore, from the standpoint of
maintenance and repair, if what has been damaged is repaired, that which is undamaged will never
be damaged by whatever brought about the need for the repair.
The second, more difficult class of deteriorative influences includes all of the mechanisms that,
when recognized as having caused damage to some concrete in service, are of such a nature that the
prudent conclusion is that much more of the concrete may be expected to be damaged unless it can
be protected from the processes that caused the initial damage. In this class are those concretes that
are unable to resist freezing and thawing or chemical attack (especially sulfate attack); those
--``,```,`,`,`,,,,`,`````,`,,`,,-`-`,,`,,`,`,,`---
concretes that have been produced of materials that contained the seeds of their own destruction
such as aggregates of inadequate volume stability, unsound cement, and unfortunate combinations
of alkali-reactive aggregate and high-alkali cement or high-alkali solutions from the environment;
and those unprotected reinforcing steel with inadequate cover over the steel and an excess of
substances that promote corrosion in the environment. All such damage may be expected to be
progressive unless there is a way to stop it. This line of reasoning yields the conclusion that a very
important purpose of the investigation of damage, distress, deterioration, and failure is to have a
clear appreciation of the causes. This is essential to establishing whether the causes were permitted
to work because of a defective specification or the failure to follow a proper one, which is relevant
to issues of liability. It is also essential to planning a proper repair or replacement to one's overall
strategy in preparing proper specifications for future work.
The way to obtain satisfactory concrete is to know what to avoid and to take proper steps to avoid
it. The art of knowing what to avoid means matching the intended environment of service with
similar environments in which concrete has performed less well than desired and knowing what
211
Daneshlink.com
daneshlink.com
interrelationship of environmental stress and concrete deterioration produced that less-than-desired

result. There is enough available knowledge, summarized perhaps best in the 1992 report of ACI
Committee 201, that we need not to go back to square one (3). The ACI committee report takes up
the problems of durability of concrete in chapters dealing with freezing and thawing; aggressive
chemical exposure including sulfate attack, acid attack, and carbonation; abrasion; corrosion of
embedded metals and other materials; and chemical reactions of aggregates. The report also includes
two more chapters, which address the repair of concrete and the use of protective barrier systems to
enhance concrete durability, respectively.
I will talk no further about repair practices, but I will say in passing that although protective barrier
systems can function beneficially - especially if by some means one can cause some concrete to
become dry and the barrier can be applied in such a manner as to cause the concrete to stay dry-
then if the concrete is afflicted with a defect that would make it nondurable in service and it can be
caused to become dry and stay dry, it may very well survive for a long time in that environment.
However, most concretes that are vulnerable to damage in service have difficulty getting dry, and if
they do dry, it is bird to keep them dry by sealing them because one cannot seal all six sides: top,
bottom, front, back, left, and right. Some bridge decks appear to be the exception to this rule. In
most other concrete structures in transportation, at least the bottom is in contact with moisture.
Let me now comment briefly on the several classes of phenomena potentially harmful to the
durability of concrete as they are dealt with in the ACI 201 report.
FREEZING AND THAWING
I recendy summarized in a paper what I thought I understood about the resistance of concrete to
freezing and thawing (4). The content of this paper is not in conflict, I think, with what larger ACI
Committee 201 says; however, I approach it from a slighdy different point of view. I point out,
--``,```,`,`,`,,,,`,`````,`,,`,,-`-`,,`,,`,`,,`---
rather obviously, that concrete will be immune to the effects of freezing for several reasons: (a) if it
is not in an environment in which freezing and thawing take place so as to cause water in the
concrete to freeze; (b) if, when freezing takes place, there are no pores in the concrete large enough
to hold freezable water; (c) if any pores can hold freezable water, they are less than 91 percent full at
the time of freezing; or (d) if pores that can hold freezable water are more than 91 percent full, the
cement paste has a proper air-void system, sound aggregate, and moderate maturity. In the extensive
work of Paul Klieger, moderate maturity was discovered to exist if the combination of w/ c and
cement hydration has proceeded so that the concrete has developed a compressive strength of about
30 MPa (4,000 psi) before it is allowed to freeze and thaw in a critically saturated state. Sound
aggregate is the kind of aggregate that when used in concrete containing a proper air- void system in
the paste, which is allowed to get moderately mature before freezing, gives frost-resistant concrete as
can be measured in the laboratory using ASTM C666 Procedure A. A satisfactory air-void system in
the paste is one that is characterized by having an air bubble not more than 0.2 mm (0.008 in.) from
anywhere.
As far as I am aware, there is essentially no place in the world today where concrete is needed in
transportation structures that is likely to be critically saturated when exposed to freezing and thawing
for which sound aggregates cannot be obtained or the concrete be protected against freezing and
thawing while critically saturated until it is moderately mature, and for which circumstances preclude
using an appropriate amount of proper air-entraining admixture so as to produce concrete with a
212
Daneshlink.com
daneshlink.com
satisfactory air-void system. Some of us feared that we might be in a situation in which we could not
produce concrete with a satisfactory air-void system using roller-compacted concrete, which is of
considerable interest especially for paving in military reservations. However, recent work has made it
clear that adequately frost-resistant concrete can be made even at the levels of harshness and dryness
that characterize concrete suitable for roller compaction (5).
AGGRESSIVE CHEMICAL EXPOSURE
Sulfate Attack
The Corps of Engineers' requirements for taking account of sulfate in the environment say that if
the level of water-soluble sulfate in soil or dissolved in fresh water expressed as SO4 is less than 0.1
percent or 150 ppm, respectively, no precautions are necessary (6). If the amount of sulfate is
between 0.1 and 0.2 percent in soil or 150 and 1,500 ppm in water, the attack is considered moderate
and the appropriate precaution is to use Type II moderately sulfate-resisting cement in which the
calculated C3A content is 8 percent or less. If Type II cement is not economically available, an
equivalent degree of sulfate resistance can usually be obtained by using either blended cement made
with pozzolan or granulated blast-furnace slag or by adding a proper amount of ground slag or an
effective pozzolan such as fly ash or silica fume. Finally, if the sulfate content is higher than the
larger values just mentioned, the level of protection should be proportionately greater and will be
achieved by using either Type V highly sulfate-resistant cement in which the C3A content must be
kept to 5 percent or less. If such cement is not economically available, a proper amount of an
appropriate ground slag or pozzolan can be used.
It is interesting to note that much work has indicated that water; if the permeability of the concrete
is reduced, either by adhering to a lower w/ cor by adding ground slag, the amount of C3A in the
cement can be allowed to rise. The Corps of Engineers now permits up to 10 percent calculated C3A
if the w/ c of the concrete is kept below 0.45 and the concrete is permanently submerged in seawater.
Another relevant recent development is a performance test that will properly evaluate the sulfate
resistance of a cementitious material or blend of such materials. This is ASTM C1012, which
involves fabricating mortar bars and storing them in water until they develop a specific degree of
maturity and thereafter immersing them in a standard sulfate solution. The sulfate resistance of the
cementitious material is regarded as adequate if the expansion of the mortar bars remains below 0.1
percent. This limit has been adopted in the new performance-based ASTM specification for blended
hydraulic cements (ASTM C1157).
Acid Attack
Besides sulfate attack, other aggressive chemical exposures involve acid and C0 2 • Concrete in
transportation structures is not likely to be exposed to acid very often or very severely. If I were
making this talk 10 or 15 years ago, I would warn against the hazards to concrete bridge piers that
might result from the establishment of an industry upstream that released strong acids into the
water, but clean water laws make this much less likely today. There are cases in which acid from coal
mine drainage can cause rivers and streams to have very low pH, in which case it would be better if
no transportation structure made out of concrete came into contact with the acid. The Corps of
Engineers was asked once to build a dam where the so-called water in the reservoir was said to have
a pH as low as 2.0, except after a heavy rain when it went up to 2.4. Nobody suggested making
--``,```,`,`,`,,,,`,`````,`,,`,,-`-`,,`,,`,`,,`---

213
Daneshlink.com
daneshlink.com
concrete that would be intrinsically resistant to this exposure; instead, we worked on developing an
organic coating that would keep the acid from coming into contact with the concrete.
Carbonation
The effects of exposure to concrete of C0 2 in the atmosphere and the resulting carbonation of the
concrete appear to be the current scare phenomenon in concrete technology. People have noticed
for a century or more that, after some years of exposure in the first few millimeters below the
surface, the cement paste in hardened concrete typically contains no longer calcium hydroxide but
calcium carbonate. It has been assumed, correctly, that this represents the reaction of CO, from the
air with calcium hydroxide in the cement paste to produce calcium carbonate. The carbonation of
the cement paste in concrete lowers the amount of calcium hydroxide that is available to replenish
the aqueous phase of that concrete with calcium hydroxide in solution to maintain saturation and,
hence, a pH of about 12.6- which is, of course, the normal and desirable circumstance from the
standpoint of preserving embedded steel from corrosion. Much nonsense has been perpetrated
largely by people who oppose the use of ground slag and pozzolans in concrete; they have argued
that such use consumes some of the calcium hydroxide that would otherwise be present and is
needed to keep up the pH and prevent corrosion of steel. The other argument has been that one
ought to use pozzolans for converting the otherwise useless and to some extent undesirable calcium
hydroxide - or, as they would call it, free lime - to beneficial CSH with the concomitant reduction
in the amount of soluble calcium hydroxide that could go into solution, be carried to the surface
through cracks, exit the cracks, and produce ugly efflorescence.
In my judgment, both of these arguments are without practical merit and have little theoretical
justification. It would take a great deal of activity to produce an aqueous phase in concrete of even
modest quality that was not a saturated solution of calcium hydroxide considering the great excess of
calcium in portland cement beyond that needed to make CSH of all the silica in the cement plus all
of the available pozzolanic silica at the normal ratios of cement to pozzolan in high-quality structural
concrete as should be used in transportation structures. To avoid efflorescence, one should provide
appropriate drains so that water does not pass through cracks in walls and bridges and deposit
calcium hydroxide that later carbonates.
Abrasion
The third chapter of the ACI 201 report deals with abrasion. This subject is a very important one in
transportation because probably more concrete in the world is subject to abrasion in transportation
than in any other class of use. When concrete is less resistant to abrasion than desired, sometimes
--``,```,`,`,`,,,,`,`````,`,,`,,-`-`,,`,,`,`,,`---
unsafe surface textures develop that create severe hazards because the pavement is nonresistant to
skidding.
ASTM has developed several tests to evaluate the degree to which concrete is worn away by rubbing
and friction; however, it is important to separate the things that are relevant to one class of structure
from those relevant to another. The wearing away of the top surface layer of a very smooth
industrial floor under vehicular or other traffic and the consequent production of dust that can harm
manufacturing operations is quite different from the wearing away of the texture of a pavement
surface. Similarly, the issue of the relative resistance to wear of the mortar portion of the concrete
and the coarse aggregate portion becomes very important to skid resistance. It has been necessary
and effective to use a special abrasion-resistant coarse aggregate so that if the surface mortar skin is
removed by abrasion, the coarse aggregate will continue to protrude above the base level and
provide a skid-resistant surface.
214
Daneshlink.com
daneshlink.com
The ACI 201 recommendations suggest that almost all the good practices in concrete production
benefit the improvement of abrasion resistance of the surface, especially practices that reduce
segregation and the consequent development of a thicker-than-desired layer of mortar at the surface,
avoiding bleeding that may cause the surface mortar to have a higher-than-intended w/ c and
avoiding finishing procedures that bring up more fines or take place at an undesirable time.
Obviously, one cannot develop an abrasion-resistant concrete surface unless the concrete at that
surface is allowed to develop its potential strength and, hence, abrasion resistance. Thus, a major
factor in producing abrasion-resistant concrete is the quality of the curing.
Section 3.6 in the ACI 201 report deals specifically with wear on concrete resulting from tire chains
and studded snow tires. It says that studded snow tires have caused widespread and serious damage
even to high-quality concrete. It adds that "fortunately the use of studded snow tires has been
declining for a number of years," and it specifically calls attention to NCHRP Synthesis ifHighwqy
Practice 32 (7).
Corrosion of metal
It is my guess that more concrete that has failed to provide satisfactory service for its intended
service life because of premature deterioration from interacting with its environment, especially over
the last couple of decades, has done so because the precautions taken to prevent corrosion of the
reinforcing steel turned out to be inadequate. I have gone on record several times to the effect that
all that is needed to prevent corrosion of the steel is to have a cover of high-quality concrete 50 mm
(2 in.) thick over the steel and appropriate measures to control concrete cracking. This approach
usually works with massive constructions such as bridge piers, dry docks, wharf structures, and off-
shore structures. However, this has not been the case with highway bridge decks and superstructures
and parking structures. People who design bridge decks, or at least who did in the past, often failed
to provide decks thick enough that the required reinforcing steel could be put in place and still have
50 mm (2 in.) of good concrete over it. Thin, strong bridge decks are preferred because they have
less deadweight and are more aesthetically pleasing. We need to take extra precautions to protect
steel from corrosion if we must live with thin reinforced concrete structures in environments of high
potential for materials corrosion; these include highway bridge decks that are deiced with chloride-
based deicing chemicals and parking garages in which slush caught under vehicle fenders melts and
runs out onto the floor as water with a rather high chloride concentration. We should not, however,
--``,```,`,`,`,,,,`,`````,`,,`,,-`-`,,`,,`,`,,`---
in the process give up the effort to make the concrete that we do have concrete of high quality, low
w/ c, and proper curing. What we do is one or both of two things; some recent experiences have
suggested that where the need for protection is great, it is prudent to do both. These two measures
are as follows: (a) use properly prepared, thermally bonded epoxy coatings on the reinforcing steel,
and (b) use corrosion-reducing chemical admixtures such as calcium nitrite. I believe that in nearly
all cases of serious risk of corrosion, either of these solutions by itself will be completely satisfactory
if the potential benefits are realized. However, it has become apparent that there are structures
whose owners believed that they were using properly protective epoxy-coated steel when, in fact, the
steel put into the structure corroded very quickly. I have not heard recently of an owner who
believed that he or she was protected by a chemical admixture and was not, but I am sure I will
someday. Then one should consider the use of lightweight concrete with which, for the same
deadweight, one can have greater cover over the steel. Alternatively is the option of using up to 30
percent less concrete to achieve a 25 percent reduction in deadweight by using stronger concrete,
60-MPa (8,700-psi) concrete, rather than 35-MPa (5,000-psi) concrete and external prestressing.

215
Daneshlink.com
daneshlink.com
There has been much debate on the issue of how much chloride should be in the concrete at the
time it is produced. We are aware now, of course, that in addition to the intentional use of calcium
chloride as an accelerator to assist in producing early strength in concrete, there is chloride to some
degree in other admixtures used for other purposes. Chloride is present in some cementitious
materials, and chloride can be present in sea-dredged aggregates, as are used in several places,
notably England and Japan, and in certain limestones that are used as concrete aggregate in the
United States far from the seacoast. There is also chloride in some mixing water. If one is using
epoxy-coated steel or calcium nitrite, it should not matter how much chloride is in the concrete as
made. If neither of these precautions is taken, it might matter a lot.
The question of whether it is ever prudent to assume that none of the measures that I have
mentioned will be satisfactory and to install cathodic protection for the steel at the time the structure
is built is, I think, still undecided. This may be wise in some cases, but I expect that they are few and
far between.
Chemical reactions of aggregates

Tom Stanton, of the California Department of Transportation (although they didn't call it that then),
discovered alkali-silica reaction in 1940 (8,p.54ff). I went to work for the Concrete Laboratory at the
Corps of Engineers in August 1941. Word of his discovery had already reached that laboratory, and
one of the first bits of work I did was looking at "siliceous magnesian limestone," which was
described as the offending ingredient of concrete that had deteriorated. At that point the Corps of
Engineers was quite concerned because, to us, magnesian limestone meant dolomite, and dolomite
was very widely used in Corps of Engineers structures. Our position had been that most limestones
were satisfactory but substantially all dolomites were. It turned out that the siliceous magnesian
limestone that caused the trouble in California is an almost unique rock that contains a lot of opaline
chert. As far as I know, no other dolomite in the world contains enough opaline chert to cause
trouble.
The Ninth International Conference on Alkali-Aggregate Reaction on Concrete took place in July
1992 in London in the Queen Elizabeth II Conference Center across the street from Westminster
Abbey; several hundred people attended from essentially everywhere in the world. Aggregates that
were not previously regarded as reactive are now shown to be reactive when used with cements of
higher alkali content than they had been used with. Cements in many parts of the world are now of
higher alkali content than they used to be, partly because more of them are made using raw materials
containing relatively large amounts of alkali (such as, for example, all of the cement in Iceland for
which all of the calcareous raw material is seashells dredged from the ocean) and also partly because
of the capture of kiln dust and its insertion into the product. With regard to this last point, I am told
that some farmers now must buy substantial amounts of potash fertilizer because cement plants
upwind are catching the kiln dust that used to go out their stacks and putting it back into the
cement, to the detriment of both the cement and the farmer.
Alkali-silica reaction involves a series of events. A reactive aggregate is one that contains silica, Si02
in a form that is capable of being dissolved by pH solutions substantially higher than the 12.6 pH
--``,```,`,`,`,,,,`,`````,`,,`,,-`-`,,`,,`,`,,`---
that characterizes the saturated calcium hydroxide solution that is normal in concrete. Either the
cement contains enough sodium or potassium so that the pH is raised when the cement hydrates
and these ions go into solution or the alkalies come in from the outside and raise the pH. In any
event, the pore fluid in the concrete gets to be high enough in pH to dissolve the silica and produce

216
Daneshlink.com
daneshlink.com
an alkali-silica gel that has the property of taking up water, swelling, and expanding and disrupting
the concrete.
If the circumstances are such that any problems that might be brought about by alkali from the
environment can be ignored, then, in many cases, all that is necessary is to invoke the specification
option for cement and require the use of low-alkali cement when reactive aggregate must be used.
This requirement means that the cement will not be allowed to contain more than 0.60 percent
alkalies calculated as the percentage Na20 plus 0.658 times the percentage K 2 0. If low-alkali cement
is not available, equivalent protection can usually be provided by using an appropriate amount of an
acceptable ground slag or pozzolan, the acceptability being based on tests by ASTM C441.
There is also an alkali-carbonate rock reaction that has been found in a number of highway
structures, especially some in Illinois, Indiana, Iowa, Michigan, Missouri, New York, South Dakota,
Tennessee, Virginia, and Wisconsin. Not all such reactions are deleterious. Sometimes the reaction
simply takes place, and a small portion inward from the outer surface of an aggregate particle is
altered so that when one slices through the concrete and acid-etches the sawed surface, the aggregate
particle develops two levels of etching. Sometimes the interior is more acid-soluble than the rim, and
sometimes it is less acid-soluble than the rim; in either case the fact that the rim is on crushed stone
particle is evidence that a reaction took place in the concrete.
To avoid harmful alkali-carbonate-rock reactions, use non-reactive aggregate or figure out a

combination with low-alkali cement. The ACI committee report recommends a value less than 0.60
percent Na2 0 equivalent.
CONCLUSION
To obtain the performance that is desired of concrete in transportation, it is necessary to
1. Decide what sorts and levels of imperfections are tolerable and intolerable,
2. Understand what causes imperfections that are intolerable,
3. Evaluate the environment in which the concrete is to serve to recognize the presence of
influences that must be resisted if the concrete is to perform as desired,
4. Prepare concrete specifications that require appropriate levels of relevant properties of
concrete so that the concrete can resist the deteriorative influences it will encounter in
service, and
5. Ascertain that both the contractor's quality control and the owner's quality assurance
--``,```,`,`,`,,,,`,`````,`,,`,,-`-`,,`,,`,`,,`---
systems work to ensure that the concrete produced is as specified.
ACKNOWLEDGMENT
The author gratefully acknowledges permission from the Chief of Engineers to publish this paper.
REFERENCES
1) R. E. Philleo. Concrete Science and Reality. In Materials Science of Concrete II (J. Skalny
and S. Mindess, eds.), American Ceramic Society, Westerville, Ohio, 1991, pp. 1-8.
217
Daneshlink.com
daneshlink.com
2) R. E. Philleo. NCHRP Synthesis of Highway Practice 129: Freezing and Thawing Resistance
of High-Strength Concrete. TRB, National Research Council, Washington, D.C., 1986.
3) Guide to Durable Concrete (ACI 201.2R-92) (Draft). ACI Materials Journal, Vol. 88, No.5,
Sept.-Oct. 1991, pp. 544-582 (to be published in Manual of Concrete Practice, American
Concrete Institute, Detroit, Mich., 1993).
4) B. Mather. How To Make Concrete That Will Be Immune to the Effects of Freezing and
Thawing. In ACI SP-122: Paul Klieger Symposium on Performance of Concrete (D.
Whiting, ed.), American Concrete Institute, Detroit, Mich., 1990, pp. 1-18.
5) S. A. Ragan. The Use of Air Entrainment To Ensure the Frost Resistance of Roller-
Compacted Concrete Pavements. In ACI SP-126: Durability of Concrete, American
Concrete Institute, Detroit, Mich., 1991, pp. 115-130
6) Standard Practice for Concrete. Engineer Manual 11 10-2-2000. Office of the Chief of
Engineers, U.S. Army Corps of Engineers, Washington, D.C., 1985.
7) L. D. Byrd. NCHRP Synthesis of Highway Practice 32: Effects of Studded Tires. TRB,
National Research Council, Washington, D.C., 1975.
8) T. E. Stanton. Expansion of Concrete Through Reaction Between Cement and Aggregate.

Transactions, ASCE, Vol. 107,1940.
--``,```,`,`,`,,,,`,`````,`,,`,,-`-`,,`,,`,`,,`---

718
Daneshlink.com
daneshlink.com
PART VII- RESEARCH AND EMERGING TECHNOLOGIES
Research is an integral part of uncovering knowledge to promote a better understanding of complex

issues involved in all aspects of concrete, including production, placement, design, and construction
with concrete materials. The successful research efforts culminate in developing methodologies for
emerging technologies that are experimented with by the concrete industry. If the initial usage is
--``,```,`,`,`,,,,`,`````,`,,`,,-`-`,,`,,`,`,,`---
found to be successful and economically cost effective, the emerging technology is embraced by the
concrete industry to enhance its competitiveness in the construction market. Research and
development efforts in the concrete industry have been instrumental in advancing better use of
concrete for its intended application and in providing better understanding of issues of maintenance
of existing concrete systems.
Bryant was very interested in the research and development efforts for concrete. In this part, four
very notable contributions of Bryant have been reproduced.
Bryant advocated research and development, as is evident in his paper "Research on Concrete,"
published in 1969. He knew that there was a payoff in research. He stated that in the decade from
19 56 to 1965, the Corps of Engineers realized cost savings in civil works construction, as a result of
applying its research, of more than five times the cost of that research.
In his paper "High-Performance Concrete in the U.S. Army Corps of Engineers," published in 1994,
he summarized research conducted by the U.S. Army Corps of Engineers for two centuries in the
development of high-performance concrete. Bryant concluded that the levels of quality expected
should be those that are relevant to the service needed-stamp out specification overkill.
The earliest work for placing concrete underwater in the United States involved the use of the tremie
method for the construction of Fort Carroll in Baltimore Harbor in the mid-19th century, which
Bryant discussed in his paper "Robert E. Lee and Concrete Innovations," published in 1999.
During this period and afterwards, new technologies were developed or evaluated, including high-
strength, high-performance mass concrete for dams; high-strength high-density, high abrasion-
resistant, washout-resistant, and ultra high-strength concrete with strengths about 30,000 psi.
Bryant was always open to ideas. He listened and captured the benefits of existing materials or new
technologies. In his 2001 paper "Self-Curing Concrete, Why Not?," he explained the benefits of
using lightweight aggregate for internal curing, including in areas difficult to reach to cure. He also
suggested that self-curing can be accomplished by some "enteric" -coated particles or capsules
containing a membrane-forming curing compound- or containing a substance that reacts with
water to do so - that would open if the surface became dry and formed a membrane. Although
such a system is not available yet, it is within the reach of the concrete industry.

?.19
Daneshlink.com
daneshlink.com
--``,```,`,`,`,,,,`,`````,`,,`,,-`-`,,`,,`,`,,`---

Daneshlink.com
daneshlink.com
Reprinted, with permission, from Lecture No. 7, Stanton Walker Lecture Series on Materials Science, ©National Sand and
Gravel Association, 1969, National Ready :Mixed Concrete Association, 900 Spring Street, Silver Spring, MD 20910.
RESEARCH ON CONCRETE
by Bryant Mather
No one could question the appropriateness of "Research on Concrete" as a topic for a Stanton
Walker Lecture on the Materials Sciences. Research, according to Webster's Dictionary, is "critical and
exhaustive investigation or experimentation having as its aim the discovery of new facts and their
correct interpretation; the revision of accepted conclusions, theories, or laws, in the light of newly
discovered facts; or the practical applications of such new or revised conclusions." Dr. Bates noted,
in the flrst of these lectures in 1963,1 that it had recendy been said that "concrete is not a material, it
is a process." However, in 1967, when the American Concrete Institute flnally got around to
publishing an offlcial deflnition of concrete,2 hat deflnition read: "A composite material which
consists essentially of a binding medium within which are embedded particles or fragments of
--``,```,`,`,`,,,,`,`````,`,,`,,-`-`,,`,,`,`,,`---
aggregate; in pordand cement concrete, the binder is a mixture of portland cement and water."
The pamphlets in which these lectures are published include on their covers a montage based on a
graph. In September 1950, in a second-hand bookstore in Baltimore, Maryland, I bought a copy of
Part 11 of the Proceedings of the American Society for Testing Materials for 1919, in which, on page
561, appears Figure 23 of the paper by Stanton Walker on Modulus of Elasticity of Conerete3 -
which begins on page 510 and, with discussion, extends to include page 606 - and contains more
evidence of "critical and exhaustive investigation" on modulus of elasticity of concrete than I believe
has been presented before or since. The curve on Figure 23 of that paper has the equation
Ei = 45,500 S 0 ·58 , as does the one on the cover of these lectures, but in the ASTM paper it is based
on data only from tests at 7 days, 28 days, 6 months, and 1 year; while the one used in the montage
includes data from tests also at 1, 2, 3, and 5 years. These considerations bring me to the flrst
conclusion to be stated in this lecture on research in concrete: There is need today and there always
will be need for critical and exhaustive investigations of concrete of the sort that Stanton Walker was
reportingin 1919.
I sent a copy of the draft of the manuscript for this lecture to Stanton Walker and I believe some of
his comments should be recorded at this point. He wrote: 4
"Thank you for comments on my Modulus of Elasticity paper. That was one of the
most rewarding, frustrating, and educational writings I have done. I would not have
been able to perform it had it not been for my boss - Duff Andrew Abrams. He
was basically a kindly, friendly, and sympathetic man and he craved similar attitudes
toward himself. He was a brilliant researcher - no one can convince me otherwise,
although a few have tried. Also, he was a moody man; at times hypercritical; an
enigmatic man - one respected but not loved. I shall always remember him with
respect and affection in spite of his many faults - displayed most prominendy in
the later years.
"Some history of my writing the paper may be of personal interest to you.

We were making a lot of modulus of elasticity tests as a by-product of

221
Daneshlink.com
daneshlink.com
compression tests made in studies of many things: curing; vibration, jigging

and pressure; fineness of cement; storage of cement; organic impurities;
grading of aggregates; amount of mixing water; etc. - and we interspersed
some tests pointed direcdy at :measuring elastic modulus.
"I was serving my apprenticeship by actually manually performing these tests.

Abrams suggested that I collect and organize the data, which I felt I could do
if I could find the time. He finally ordered me to do it. After the organization
of data got well under way he suggested I use it for an ASTM paper. It was
my first 'big paper' and although I had helped him with some important
papers, I was not ready for it - but I started.
"The whole purpose of this 'history' is as background for a brief statement

concerning my frustration in writing the paper, which also accounted for it
being rewarding and educational. Abrams and I reviewed what I had written
from time to time - a lot of that review being on Saturday afternoon at
Garfield Park while awaiting our time to play golf. He read like lightning -
down the middle. He would select a paragraph, a page, several pages, and say
'Rewrite this!' I would say, 'O.K. -but why, how, in what way?' He would
say, 'You will fmd out; rewrite it.' And I did; and I did!
"I have always wanted to try this on some of the staff responsible to me, but
I have always lacked the guts. They probably would have told me to go to
hell! Needless to say this constitutes no suggestion, or permission, implied or
explicit for modifications or additions to your paper. This is 'classified
information' as you bureaucrats are so fond of saying!"
You will note that this beautiful and important statement of how significant research on concrete
may be accomplished was made with the instruction that it was not to be published. I am sure you
share with me my pleasure in being able to report that, on appeal, the decision was reversed, and
--``,```,`,`,`,,,,`,`````,`,,`,,-`-`,,`,,`,`,,`---
permission was granted for its inclusion here.
It is appropriate in 1969 to consider research in concrete critically because of the climate of society
in which we must plan to function - in which everything is being critically examined - including
concrete and research on concrete. In Science for 9 May 1969,5 Joel Hedgpeth, reviewing the
proceedings of an MIT Conference on Planning, wrote:
"A key contribution 'aimed at architects and urban designers,' is that of

Harold J. Barnett on natural resources. According to this author, there will be
an abundant supply of materials from which to build structures and increase
cities. A great pity; think what a shortage of concrete would do for mankind.
The diversity being thought of here is within cities and aggregations of
people, not in the natural environment; control over the plans may well be
more from the force of numbers of people and the increasing masses of
concrete drawing upon those abundant resources, and we must disagree with
the summary statement that 'nature is not the limitation for either man
himself or his material needs."'

222
Daneshlink.com
daneshlink.com
Hedgpeth ends his review with these words: "But whatever is to be done money cannot be the main
consideration, for time is of the essence when population, pavements, and pollution are increasing
exponentialfy."
Spilhaus,6 writing in The American Scientist in the spring of 1969, reminded us that between now and
the year 2000 we will need a family unit for 2.5 people every 27 seconds just for the increase in the
population.
Siggia/ in the same issue, noted with regard to industrial research that
"Companies are profit oriented. In today's economic profit squeeze, there is
little money for what is known as basic- blue-sky, far-out research. In most
companies, a research man can justify any research program where he can
show that the results will lead to an increase of company profits. It is difficult
to sell basic research to management because the chances of direct dollar
return are small. However, one can usually sell a low-risk applied program
because it is more of a sure thing ... Industry has hurt itself in several ways by
its accent on 'me-too' or 'sure thing' research. The main reason for de-
emphasis on basic research is decreased profits. The main reason for
decreased profits is increased competition. The main reason for increased
competition is too many people making the same product. This has come
about by too much me-too research."
The cement and concrete industries both in the United States and elsewhere are in a worse profit
squeeze than most other industries. John Mundt told the 1969 conventions of the National Sand
and Gravel Association and the National Ready Mixed Concrete Association8 that NRMCA
reporting members in 1967 earned only 3.82% before tax on gross operating investment and that in
the cement industry return on investment had dropped to 5 percent. In Punch9 for 18 June 1969, the
statement of Sir John Reiss, Chairman of The Associated Portland Cement Manufacturers Ltd., in
the United Kingdom, was quoted as follows:
"Our biggest problem is the steady reduction of our profit margins in this
country. The fact is that present profitability is quite insufficient to provide
for the capital sums which will be required for expansion during the next
decade. We sincerely hope that the need for a substantial price increase will
be recognized in the near future. Unless some recognition is soon given to
our difficulties on the home market your Board will have no alternative but
to confine future capital investment to our Overseas Companies."
One of my friends who is deeply concerned about research in concrete said to me privately: "Most
of the concrete research in progress in the world is still pointlessly aimed at acquiring a deeper
recognition of the nature and characteristics of conventional concrete - which can in no way be
altered thereby (i.e., it cannot pay for the research)."
Dr. Bates, in the first of these lectures in 1963,1 suggested that the concrete industries do not spend
as much as one tenth of one percent of their sales income on research - even generously defined
and said, "No other industry of consequence spends so little ... Probably less than one hundredth of
one percent of the concrete sales dollar is spent on research to provide fundamental information." Is
--``,```,`,`,`,,,,`,`````,`,,`,,-`-`,,`,,`,`,,`---

223
Daneshlink.com
daneshlink.com
there a cause and effect relation between this fact and the fact that the concrete industry is at the
bottom of the list on profit as a percentage of investment?
I propose that what concrete research needs is improved understanding- and hence improved
appreciation- by all concerned- of its nature and purpose. At the beginning of this lecture I
defined research and concrete. I now wish to comment on the nature and purpose of research on
concrete and on whom I mean by "all concerned" Those concerned with research on concrete are all
of the members of the human race - society - the public - mankind. As Joel Hedgpeth and
others have realized, perhaps subconsciously, concrete is the symbol, the metaphor, par excellence,
for the actions of humanity in modifying the natural environment in which it finds itself. As he said,
"population, pavements, and pollution are increasing exponentially." Concrete may be the symbol of
mankind's often not-so-intelligent modification of its natural environment up to now, but it is
necessarily the product that man must use more wisely if he is to do better at the further
modification - or remodification - of his environment. As John Mundt said in J anuary8: "Ours is
a product of social significance ... (it) can be instrumental in halting the erosion in the quality of
human existence in our cities by ending substandard housing, overcoming traffic congestion,
eliminating urban blight." To appropriate degrees, the public must learn that, useful though concrete
is today, it is inefficiently and, to varying degrees harmfully, produced and wastefully used. Today we
discard as uneconomical many cement and aggregate raw materials that must and will be used in the
years to come. Very little old concrete is today recycled to be new concrete. Very little of our
civilization's solid wastes are used in making concrete. A letter10 in The Engineering News-Record for 31
July 1969 specifically stated, "research on use of crushed brick and concrete should be undertaken."
I have on several previous occasions expressed a view that the future will involve the use of
--``,```,`,`,`,,,,`,`````,`,,`,,-`-`,,`,,`,`,,`---
computers to perform the arithmetic that will be required to examine large amounts of relevant
quantitative data on materials, mixtures, loads, and on environmental and performance parameters
and thus permit a selection of appropriate levels of relevant aspects of composition and properties
of materials, and proportions of concrete mixtures to match requirements of use - both structural
and environmental. It would thus be possible to use, in any given work, the particular materials and
the specific mixture that would provide the needed level of quality at the lowest cost. I have also
suggested that the greatest need in concrete research was for ways of quantifying the properties of
the environment so that these requirements could be stated in numbers and the numbers made
available to the computer.
I quote a sentence from Dr. Powers' 1965 lecture, 11 "It does not make economic sense to design a
mixture for maximum life under conditions of perpetual wetness and actual freezing stress, when in
fact the structure to be built involves no such situation and will never suffer such stress." I quote
another from Mr. Ketchum's lecture one year later: 12 "If proper production control could be
maintained, a nominal cylinder strength of 2500 psi should be satisfactory for most concrete
structures."
The following year Mr. Philleo 13 described some of the tools that we now have to deal reasonably
with the indeterminacy of concrete strength. Last year Professor N eville14 reviewed aspects of the
nonelastic nature of concrete that, "there is no major discrepancy between design assumptions and
reality ... our structures are safe and serviceable." Note that he did not say that they were efficient or
economical.

224
Daneshlink.com
daneshlink.com
I advocate a program of research and development, carried out in widely separated places, by people
of different backgrounds, all contributing ultimately to a unified understanding of concrete. The
need for this program must be appreciated by the concrete industry, by the cement producers and all
the other materials producers, the contractors, the architects, the civil engineers, the public agencies
concerned with housing, transportation, health, water resources, and recreation - indeed by all
concerned.
The "program of research and development" of which I speak may never be formally outlined and I
hope it will never be administered by an authoritarian body. It may never have a greater degree of
centralized control and administration than that exercised today by ACI Committee 115 on Current
Research over the program of current concrete research in the world described in the compilations
published by ACI for Committee 115.
I do not advocate taking anybody anywhere away from his personally selected pet research project
for which he has obtained funding. The researcher who has his funding in hand and who is using it
in accordance with the terms under which he got it is a free agent and should remain so. If it is
necessary for somebody in each state of the United States, each decade, to discover all over again
that compressive strength tends to go up as water-cement ratio goes down or that entrained air
tends to improve frost resistance, so be it. It is sad but not criminal.
An earlier draft version of this lecture was entided "Atomic Structure and Your Friendly
Neighborhood Ready-Mixed Concrete Producer." The words "atomic structure" in that tide were
intended to refer to the body of scientific knowledge that is coming to be known as materials
science, and to the efforts that are being made to fill in the gaps in this body of knowledge insofar as
it concerns pordand cement, aggregates, and the products that result from the intermingling and
interactions of pordand cement, admixtures, aggregates, and water. The words, ''Your Friendly
Neighborhood Ready-Mixed Concrete Producer," were intended to refer to some of the people who
are responsible for our being here tonight and who are responsible for selling most of the
intermingled mixtures of interacting pordand cement, admixtures, aggregates, and water that are
sold.
Only a few concrete producers are also materials scientists; even fewer materials scientists are
concrete producers. Arguments could be made - but not profitably - about whether it takes more
brains to be a successful concrete producer or a successful materials scientist; but the future of the
concrete industry in particular - and the construction industry in general- depends to a large
degree on the success of the materials scientists in filling the gaps in the body of knowledge relating
to the fundamental nature, constitution, microstructure, fabric, and evolution of the complex
ecosystem known as concrete and the success with which this body of knowledge is interpreted and
used by the intervening layers of applied scientists, engineers, executives, and administrators in the
research, development, testing, and evaluation complex and ultimately taken account of in the
production and use of concrete.
Much has been written about "bridging the gap between theory and practice" or between research
and application. I suggest that the motivation for accomplishing this task must come from those
engaged in practice and application; in the concrete field, from those whose business it is to make a
profit from selling concrete and whose goal is to increase the percentage profit that is made from
the sale of that product.
--``,```,`,`,`,,,,`,`````,`,,`,,-`-`,,`,,`,`,,`---

225
Daneshlink.com
daneshlink.com
Today, many enlightened purchasers of goods and services are looking more than ever before to the
vendors of these products and services for recommendations - based on sound technical evidence
- for improvements not only in the goods and services available for purchase but also for
proposals for changes in the specifications cited by the purchasers in contracting for such purchases.
Today every major construction contract of the Corps of Engineers contains a value-engineering
incentive clause that invites the contractor to come to the contracting officer with proposals by
which changes will be made in the specifications to increase the contractors profit and decrease the
contract price. The successful vendor will in the future - as at present and in the past - be the one
who offers for sale a better product at a lower cost.
Where can a vendor look for ideas on how to produce a better product at lower cost? Ultimately, if
the product is concrete, he must look to the scientist who is working at the business of filling the
gaps in the body of fundamental knowledge of the nature, constitution, microstructure, fabric, and
evolution of cements, admixtures, aggregates, and the products, formed when these are combined
with water. Usually he cannot profitably go direcdy to the physicist, the physical chemist, the
thermodynamicist, the crystallographer, and request advice on how to make better concrete at lower
cost in his plant. Usually he does not have an awareness of where in his operations the opportunities
are for applying the results of the work of these scientists, and consequendy he cannot ask them
relevant questions. Similarly, few thermodynamicists or crystallographers can go to their
neighborhood concrete plant and tell what changes might be made to increase the probability of
producing a better product at lower cost. Hence the producer talks to and listens to the engineers in
his organization, in the NRMCA, the NSGA, the Pordand Cement Association, and those in such
groups as the engineering schools, the ASTM, the ACI, the American Society of Civil Engineers, the
Highway Research Board and so on. And it is to ASTM, HRB, ACI, and the like that the results of
the work of the material scientists are reported for sieving, beneficiation, and study by those who
--``,```,`,`,`,,,,`,`````,`,,`,,-`-`,,`,,`,`,,`---
can and should bridge the gaps between theory and practice, between research and application.
The Concrete Division of the U.S. Army Engineer Waterways Experiment Station, the Research
and Development Laboratories of the PCA, the Joint Research Laboratory of the NSGA-NRMCA,
the Division of Physical Research of the Bureau of Public Roads, the Engineering Laboratories of
U. S. Bureau of Reclamation, the Cement and Concrete Association in England- and also the
Building Research Station and the Road Research Laboratories there - the Concrete Research
Laboratory sponsored by the Aalborg Cement Company in Denmark, and numerous other groups,
including many in universities; notably those at Purdue, Illinois, and Berkeley, to name a few; the
National Research Council of Canada, and so on- contain, to a greater or lesser degree, people
who are materials scientists, whose work is largely in the business of fundamental studies, and, under
the same roof- and occasionally under the same haircut - people who are ready, willing, and able
to talk to the concrete producer about what he really can do to produce a product of improved
customer acceptance by reason of being of better quality and lower cost.
The concrete producer should know what he can do with his plant and his available materials- and
what he cannot do. If he cannot produce concrete 90% of the test specimens of which will break at
loads in excess of some value such as 4000, or 5000, or 6000, or 7000, or 8000, or 9000, or 10,000
psi, he should not enter into a contract to supply such concrete. But if his competitor can do so, at a
profit, then the man who cannot is likely to be in bad trouble. And, if his competitor can, it is
ultimately because his competitor listened to and acted upon what someone told him who had been
listening to and understood what some materials scientist had been up to.

226
Daneshlink.com
daneshlink.com
Concrete can be -and has been - produced so that it has a unit weight, hardened, of as low as 10
lb / cu ft or as high as 300 lb / cu ft; so that it has no measurable permeability to water under a water
pressure of several hundred psi or so that water flows through it as through loose sand; so that it has
an ultimate compressive strength of less than 50 psi or as much as 25,000 psi - and with
comparable extreme ranges of numerous other properties. No way is now known by which some of
these extremes can be attained simultaneously in the same concrete - for example, no way is
known to have very lightweight concrete also be of very high strength- but all these extreme levels
of properties I have mentioned have been requested by engineers for use on actual construction for
which concrete has been or is proposed to be purchased commercially. The man who wants to buy
25,000 psi compressive strength concrete may be willing to pay 5 or 10 more times as many dollars
per cubic yard as the going price at his neighborhood ready-mixed concrete plant - but will there
be a producer who will be able to bid his work? Somebody will.
What should the researchers be doing? They should be doing everything that needs to be done to
learn what happens when cement and water are mixed; why cement paste hardens; why cement
paste changes volume on drying; what happens at the boundary between cement paste and
aggregate; how water moves in concrete; how can cement be made more economically; how can
better cement be made; how can better houses/roads/dams be built- quicker-and at lower cost;
and so on and on and on.
How do the researchers go about doing this? Each researcher decides what his thing is and gets
funding to do it; a process which may include deciding what your thing is after you have gotten the
money to do it. One researcher I know is conducting "a systematic exploration of low-temperature
dynamic-mechanical response transitions in hardened cement paste; (which) manifest themselves as
internal friction peaks and sharp changes in the dynamic modulus." A federal agency proposes to
contract for research "to determine the effect of the composition and physical properties of cement
on the migration of moisture to and from the coarse aggregate particles in concrete." This recalls the
last section in Dr. Powers' 1965 lecture11 entided "A neglected area for research," in which he wrote:
''We need meaningful terms for describing significandy different situations with respect to water
movements to and from concrete, and within concrete. Such terms would evolve from systematic
studies of moisture in concrete under all significandy different ambient conditions ... What is needed
is a program of systematic study, theoretical as well as empirical. With that done, it would be
possible to complete the story of the mechanics of frost action in concrete."
Dr. Mielenz in his 1964lecture15 said: "Research to develop more accurate procedures of analysis of
hardened concrete to establish the original proportioning of cement and water should be
encouraged ... (there is urgent) need for reliable (means of) determination of original water-cement
ratio."
Mr. Philleo in his 1967 lecture 13 said: "One of the most fruitful areas of concrete research ... (is) on
the fundamental nature of the bond between the colloidal particles of the paste and the crystals of
the aggregate particles."
I suggest that all the topics mentioned by all previous lectures as appropriate subjects for research
still are. Indeed I note that D. 0. Woolf in his 1948 paper16 on "Needed Research" on mineral
aggregates began by saying that the statements made by Frank Jackson in 1929 on needed research
still applied.
--``,```,`,`,`,,,,`,`````,`,,`,,-`-`,,`,,`,`,,`---

227
Daneshlink.com
daneshlink.com
I add however a word of caution: researchers should avoid the tendency to believe that merely using
a new piece of gear will solve problems. Gilchrise 7 has recently written: "The current trend
everywhere is to tackle all the old problems with more and more sophisticated equipment and to
cover up continuing failure to solve them with language of ever increasing obscurity."
Where will it lead? It will lead to concretes having properties appropriate to their use, produced and
placed at lower cost; to wider use of concrete; to more efficient use of raw materials; to better
homes for people; to safer transportation facilities; to improved profits for producers.
In the decade from 1956 to 1965 the Corps of Engineers realized savings in costs of civil works
construction as a result of application of results of its research, of more than five times the cost of
t..l,.at research. In the concrete field, the official estimate is that $1.2 million is saved annually by
--``,```,`,`,`,,,,`,`````,`,,`,,-`-`,,`,,`,`,,`---
applying the results of research on pozzolans and other mineral admixtures and $1.4 million is saved
annually by allowing higher working stresses in steel in severe exposures. In recent years the Corps
expenditure for Civil Works research has risen to more than 1/2 of 1% of costs of such work, but
the fraction for civil engineering has dropped from over 60% to only a little over 20% of the total. 18
There is a payoff. It does work. The structural researchers have found ways of designing concrete
structures of remarkable thinness and great stability. We can routinely make both concrete boats and
concrete having a specific gravity less than that of water. Expansive cement concrete is a commercial
reality. It is not necessary for any additional concrete to be damaged by frost action or by sulfate
attack. There is no longer a need to put up with cracks in concrete floor slabs. But today, to obtain
these assurances with a high probability of success, we need to use materials of excessively high
quality and use them wastefully.
As humanity tries to cope with the exponential increase in population, pavements, and pollution,
research on concrete has an increasingly important role to play in providing the knowledge that will
permit society to provide pavements - and other concrete structures - for people and reduce
pollution.
REFERENCES
1) Bates, A. A., "Composites in Construction," Stanton Walker Lecture No. 1, 1963.
2) ACI Committee 116 Report, "Cement and Concrete Terminology," ACI SP- 19, American
Concrete Institute, 1967.
3) Walker, S., "Modulus; of Elasticity of Concrete," Proceedings, American Society for Testing
Materials, V .. 19, 1919, p. 510.
4) Walker, S., personal correspondence, September 2, 1969.
5) Hedgpeth, J., review of an M.I.T. Conference on Planning, Science, May, 1969, p. 668.
6) Spilhaus, A., "Technology, Living Cities, and Human Environment," The American Scientist,
V. 57, No. 1, Spring 1969, p. 24.

228
Daneshlink.com
daneshlink.com
7) Siggia, S., "At the Interface," The American Scientist, V. 57, No.1, Spring, 1969, p. 14A.
8) Mundt, J. C., "Bluepring for a Better Concrete/Cement/Aggregates Industry," presented at

the Annual Conventions ofthe NSGA and NRMCA, New York, New York, Jan., 28, 1969.
9) Reiss,J., Sir quoted in Punch, V. 276, No. 6719,June 18, 1969, P. vi.
10) Teopfer, W.K., a letter appearing in Engineering News-R.tcord,July 31, 1969, p. 5.
11) Powers, T. C., "The Mechanism of Frost Action in Concrete," Stanton Walker Lecture No. 3,
1965.
12) Ketchum, M. S., "Properties of Concrete That Affect the Selection of Structural Systems,"
Stanton Walker Lecture No.4, 1966.
13) Philleo, R. E., "The Strength of Concrete- A Statistical View," Stanton Walker Lecture No.5,
1967.
14) Neville, A.M., "Concrete-A Non-Elastic Material in the Laboratory and in Structures,"
Stanton Walker Lecture No.6, 1968.
15) Mielenz, R. C., "Diagnosing Concrete Failures," Stanton Walker Lecture No.2, 1964.
16) Woolf, D. 0 ., "Needed Research," in "Symposium on Mineral Aggregates," Special Technical

Publication No. 83, American Society for Testing Materials, 1948, p. 221.
17) Gilchrist, J. D., in General Report D "Corrosion of Reinforcement and its Prevention,"
RILEM International Symposium on Durabiliry of Concrete, Prague, Czechoslovakia, Sept., 1969, p.
D-43.
18) Research for Civil Works - A Progress Report, Office, Chief of Engineers, Washington,
D.C. Dec., 1968.
--``,```,`,`,`,,,,`,`````,`,,`,,-`-`,,`,,`,`,,`---

?29
Daneshlink.com
daneshlink.com
--``,```,`,`,`,,,,`,`````,`,,`,,-`-`,,`,,`,`,,`---

Daneshlink.com
daneshlink.com
Reprinted from International Workshop on High Performance Concrete, ACI SP-159, 1994, Farmington Hills, MI, 456 pp.,
©American Concrete Institute.
HIGH-PERFORMANCE CONCRETE IN THE U.S. ARMY CORPS OF

ENGINEERS
By Bryant Mather'
--``,```,`,`,`,,,,`,`````,`,,`,,-`-`,,`,,`,`,,`---
SYNOPSIS
For two centuries the U.S. Army Corps of Engineers (CE) has been developing high-performance
concretes. In the 1840s, various Corps of Engineers' officers, including Robert E. Lee, developed
concretes that could be placed underwater for construction of coastal defense facilities. General Q.
A. Gillmore, whose device for measuring time of setting of cement paste is still used, published a
book in 1863 on hydraulic cement and mortar. In 1871, he published a book on concrete, which
introduced concrete technology from France that was significantly higher-performance than that
then used in the United States. Contemporary development of high-performance concrete began in
1935 at the CE Concrete Laboratory at Eastport, Maine, in support of the Passamaquoddy Tidal
Power project. The objective was to develop concrete able to resist twice daily immersion in sea
water and freezing in the winter when the tide went out. That objective was achieved. In 1970, when
confronted with the problem of severe abrasion-erosion damage in stilling basins below dams, a
solution was found in the development of concretes having strengths greater than 100 MPa. This
was done using silica fume and high-range water-reducing admixtures. Similar and higher-strength
high-performance concretes have also been developed for defense purposes as part of the
protective-structures portion of the U.S. military research and development (R&D) program. When
stronger concrete or concrete that must resist a more severe exposure is needed, the Corps of
Engineers' concrete R&D capability has been able to develop it, and I expect it will continue to be
able to do so.
Keywords
Abrasion resistance; concretes; freeze-thaw durability; high-performance concretes; radiation
shielding; silica fume; tremie concrete
INTRODUCTION
I was asked to review the work of the U. S. Army Corps of Engineers on "High-Performance
Concrete" and comment on future research needs. In 1994 the U.S. Army Corps of Engineers
celebrated its 219th anniversary. I regard any concrete that has properties needed for some objective
that are not possessed by ordinary concrete as high-performance concrete. On this basis, as I
contemplate the history of the U.S. Army Corps of Engineers and especially its role in research and
development in concrete technology, I conclude, that more often than not, the Corps came up with
a high-performance concrete that was able to serve some specific sort of purpose for which ordinary
* ACI Honorary member Bryant Mather is a Past President of ACI and is Director, Structures Laboratory, U.S. Army
Engineer Waterways Experiment Station, Vicksburg, Mississippi. He has published more than 500 papers on concrete
and concrete-making materials during his career of more than 54 years.

231
Daneshlink.com
daneshlink.com
concrete was not adequate, when such was needed to meet some specific need of the Corps either in
its military mission, its civil works mission, or in support of some other agency of the U.S.
Government. There have been many such cases. I will mention some examples.
EARLY DEVELOPMENTS
The earliest experimental work in concrete technology in the USA of which I am aware was that
--``,```,`,`,`,,,,`,`````,`,,`,,-`-`,,`,,`,`,,`---
accomplished by Robert E. Lee involving the use of the tremie method for placing concrete
underwater that he conducted in connection with the construction of Fort Carroll in Baltimore
Harbor in 1849 (Freeman 1934). Subsequent study ofletters in the National Archives (Holland and
Holland 1985) reveals that several Corps of Engineer officers were corresponding about and
conducting studies of concrete placement methods as early as 1845 when J. G. Totten wrote to
George Dutton about concrete placement at Key West, Florida.
Q. A. Gillmore has been described as "the preeminent American authority on cementitious materials
for use in mortar and concrete" (Chamberlin 1983). He wrote a book on limes, hydraulic cements,
and mortars (Gillmore 1863) and invented the Gillmore needle that is still used to measure time-of-
setting of cement (ASTM 1993). His book on concrete (Gillmore 1871) was the first in the USA to
deal with concrete making and was based on his studies in France in 1870. Gillmore was closely
associated with the construction of the first concrete arch built in the USA (Chamberlin 1983).
HIGH-PERFORMANCE MASS CONCRETE FOR DAMS
The Corps of Engineers (1968) reviewed a decade of progress in development of concrete of

improved performance for construction of mass concrete structures. During this period, it was
established that the cement content of concrete could be reduced as much as 30 percent by the use
of more efficient aggregate gradings and entrained air without adversely affecting performance
(Mather 1980). In the decade 1956 to1965, the Corps saved $10 million by use of such concrete
(Corps of Engineers 1968). In the same period, it was established the substantial amounts of ground
granulated blast-furnace slag or pozzolan could be used with concomitant further reduction in
cement content and temperature rise. Concrete mixtures were developed that had adequate
performance characteristics with pordand cement content as low as 60 lb/yd3 (36 kg/m3) and water
contents ofless than 80 lb/yd3 (48 kg/m3) (Mather 1974). This development saved another $10
million. Also in this period, concretes proportioned using ingredients not susceptible to alkali-
aggregate reaction and deleterious internal expansion were developed and used. Use of these saved
$14 million. About the time we really had developed high-or optimum-performance mass
concrete for dams, the Corps' dam construction program essentially ceased, but the technology is
widely used throughout the world.
HIGH-STRENGTH CONCRETE
In 1962, the U.S. Air Force requested the Waterways Experiment Station (WES) to investigate the
feasibility of producing high-strength concrete. It was said the interest related to building
underground aircraft shelters. A review was made in which it was concluded that concretes with
strength up to 18,000 psi (124 MPa) had been made in construction and that strength consistendy
above 10,000 psi (69 MPa) could be done by careful material selection and process control. A
summary was published (Mather 1967).
232
Daneshlink.com
daneshlink.com
In 1994, additional studies of high-strength concrete were undertaken at WES to develop mixtures
for construction of hardened military facilities. For this use, a concrete was needed of very high
strength but also substantial ductility. Typical concretes so far produced in this program have had
compressive strengths at 7, 28, and 56 days of 138, 180, and 170 MPa and to 223 MPa at 56 days
when cured at 80 °C.
HIGH-STRENGTH, HIGH-DENSITY CONCRETE
In 1963, the U.S. Office of Civil Defense requested the WES to investigate whether it was feasible
to make concrete that had both high strength and high density for use in biological shielding from
ionizing radiation. Using conventional water-reducing admixtures and either magnetite or ilmenite
ores as aggregate, concretes of about 230 lb/fe (3700 kg/m3) were made that had strengths over
10,000 psi (69 MPa). The water-cement ratios were 0.30 and 0.35, the nominal maximum size of
aggregate was 1/2 in. (12.5 mm) or 1-1/2 in. (37.5 mm), and the slumps were 1/4 to 3/4 in. (6 to 20
mm). A paper on this was published (Mather 1965, 1972).
HIGH-PERFORMANCE MIXTURES NEEDED IN CONNECMON WITH NUCLEAR WEAPONS TESTING
The United States used a lot of technology in investigating the effects of nuclear weapons. One of
the earliest high-performance concrete requirements relating to this effort that the WES was
required to meet was to produce "rock-matching grout." The requirement was for a composition of
matter that could be produced in the form of a grout that could be pumped down a hole so as to fill
the hole, surround sensitive instruments placed at selected elevations in the hole, harden and
develop specific properties essentially identical to the properties of the rock removed from the hole
at the level of the instruments. The intent was that, so far as the instruments were concerned, it was
desired that they should behave as if no hole had been drilled and they just grew in situ where they
were put. What made the grout so special ("high-performance") was that the primary physical
property of interest was the behavior in hydrodynamic shock-the Hugoniot equation-of-state-
and the desire that the material harden and develop its physical properties so they were at the desired
level at the scheduled date of test but not change very much thereafter in case the test was delayed.
There was a lot of such work, but most of the results are classified. However, a summary was
published (Bendinelli and Burnett 1967).
HIGH-ABRASION-RESISTANT CONCRETE
In 1979, word came from the Headquarters, U.S. Army Corps of Engineers that abrasion-erosion
damage under water to stilling basins below dams was the most urgent concrete problem in the
Corps. Investigations conducted at the WES produced a test method able to evaluate concrete with
respect to its resistance to such exposure and test data based on its use (Liu 1981). The method is
now an ASTM Standard (C 1138). Based on the laboratory data (McDonald 1980; Liu 1980), when
the issue of what to do in the repair of the stilling basin at Kinzua Datn arose, we were ready to
develop the high-performance concrete that was needed, using locally available aggregates plus a
high-range water- reducing admixture and silica fume, as described by Holland et al. (1983, 1984,
1986).
--``,```,`,`,`,,,,`,`````,`,,`,,-`-`,,`,,`,`,,`---

213
Daneshlink.com
daneshlink.com
WASHOUT-RESISTING CONCRETE FOR UNDERWATER PLACEMENT
Related to the problems with abrasion-erosion damage just mentioned, which, as noted, resulted in
development of a concrete of greater resistance to such action, was the problem of effective,
economical repair of structures damaged by abrasion erosion.
Mixtures were proportioned to be suitable for placing underwater and to have high washout and
abrasion-erosion resistance. A combination oflow water-cement ratio (w/ c), high cement contents,
fly ash, and silica fume were used to increase the abrasion-erosion resistance. Antiwashout
admixtures (AWA) were used to enhance the resistance of the concrete to washout. Concretes
suitable for traditional placement can be unsuitable for placement underwater in thin lifts, especially
those having a high w/ c; however, increased cement and flne aggregate contents can be essential to
placements underwater in thin lifts. Concrete mixtures having low w/ care more resistant to
abrasion-erosion and more resistant to washout. The type of high-range water-reducing admixtures
(HRWRA) affects the washout characteristics and abrasion-erosion resistance of the concrete
mixtures. Concrete mixtures can be made more resistant to washout with the addition of the proper
type and amount of AWA. Some AWAs and HRWRAs are incompatible. There is some evidence
that the presence of fly ash in concrete mixtures can improve washout and abrasion-erosion
resistance and that the presence of silica fume in the concrete mixtures can improve washout
resistance (Neeley 1988). Additional work (Neeley et al. 1990) further studied the cohesive and flow
parameters of selected concrete mixtures under different placement situations, estimated the
parameters necessary for successful underwater placement, examined the quality of bonding of the
repair concrete to the existing concrete, and examined the effect of underwater placement upon the
abrasion resistance of the concretes. A washout test was devised and used to determine the relative
amount of cement paste lost when the concrete is exposed to a large volume of water. The test for
abrasion-erosion resistance was used. A point-load tensile test was used to determine the bond
strength of the repair concrete to the existing concrete. The results indicated that cohesive, flowable,
abrasion-resistance concrete that will bond well to existing concrete can be placed underwater by
available methods if proper materials are used and precautions are taken. Antiwashout admixtures
(AW A) that have and silica fume can be used to reduce segregation and dilution with the water.
A WA will reduce washout of flnes in fresh concrete allowed to free-fall through 1 m of water, but
dilution will occur. Thus, this technique is not recommended for repair of underwater concrete. The
degree of workability and washout resistance necessary for a successful concrete placement will
depend upon the type of equipment used in the placement. Concrete containing A WA and water-
reducing admixtures (WRA) placed at the point of use will bond securely to hardened concrete and
sustain only relatively small loss of flnes if the bonding surface is clean. Even so, anchors should
probably be used to ensure that the new concrete remains in place. Fluid concrete with good
cohesion will move laterally for short distances, again with only minimal dilution. Concrete, so
placed, can be worked underwater to a rough flnish. As with any underwater concrete work,
agitation and movement should be minimized. Indications are that abrasion resistance of concrete
placed and finished underwater equals that of concrete placed in air, once the laitance is abraded
from the surface. The thickness of the underwater repair will be limited due to thermal stresses
associated with the high cementitious material content of the concretes. The maximum allowable
thickness of the repair concrete should be determined by thermal-stress considerations before the
concrete placement. Multiple layers of concrete could be used if the repair area is deeper than
thermal conditions will allow to be filled in one layer. The thickness of the layer, or layers, of
abrasion-resistant underwater concrete should be relatively uniform. Void areas that are deep,
--``,```,`,`,`,,,,`,`````,`,,`,,-`-`,,`,,`,`,,`---
234
Daneshlink.com
daneshlink.com
relative to the repair as a whole, should be filled with an underwater concrete having a lower
cementitious material content to an average level of the repair area.
HIGH-PERFORMANCE MIXTURES USED IN NUCLEAR-WASTE ISOLATION
In connection with nuclear-waste isolation, WES has been tasked by the U.S. Department of Energy
to develop a number of different concretes and grouts that have rather unusual performance
requirements. For filling cavities excavated in or boreholes drilled into and through large masses or
beds of rock salt, one needs a mixture that does not contain water that will dissolve salt with which
it comes in contact. We have, therefore, done a great deal of work on salt-saturated concretes and
grouts. In some cases the aggregate in the concrete was crushed rock salt graded in both fine and
coarse aggregate sizes. For these applications, it is also sometimes necessary to incorporate
expansive constituents to minimize the degree to which the concrete or grout undergoes reduction
in volume after setting and hardening. In other cases, it may be necessary to develop mixtures that
have unusually low heat-rise characteristics to minimize reduction in volume on cooling. All these
are high-performance concretes (Wakeley and Poole 1987; Gulick and Wakeley 1989; Wakeley and
Emzen 1992).
HIGH-PERFORMANCE CEMENTS AND CONCRETES
The Corps of Engineers is continuing to keep up with the state-of-the-art of high-performance

concrete and to contribute to it. In April 1994 a report on Pyrament Blended Cement was published
(Husbands et al. 1994) that presented the results of work conducted between 1990 and 1993 in
cooperation with Pyrament/Lone Star Industries, Inc., under the Corps' Construction Productivity
Advancement Program (CPAR). Concrete made with this cement is only used when performance
beyond "normal" is needed. In July 1994 the Engineering News-Record (Anon 1994) reported that the
Corps of Engineers plans, in September, to sign an agreement for a $2.1 million study of Reactive-
Powder Concrete 200 developed in France that "attains 29,000 psi." A specimen of this concrete
was said to have developed 72,500 psi. A similar specimen was furnished to us by the producer.
FUTURE RESEARCH NEEDS
I believe that the subjects that need research to facilitate the development and use of high-
performance concrete are the same subjects that research is needed on to advance concrete
technology as a whole. In June 1994 I was asked to list research needs for ACI Committee 123 on
Research. I listed five, as follows:
(1) How can we make high-strength concrete that is not britde, i.e., has a relatively low
modulus of elasticity, and hence has much greater tensile strain capacity before
cracking?
(2) How can we tell when sufficient measures have been taken to reduce the premature
loss of evaporable water from fresh concrete, i.e., when curing is effective?
(3) What is the optimum size and distribution of air voids in air-entrained concrete;
because concrete seldom has over 10% freezable water and the expansion on
freezing is 9% (9% of 10% = 1%); why wouldn't an air content of 1% be enough if
the air voids were of optimum size and distribution?
--``,```,`,`,`,,,,`,`````,`,,`,,-`-`,,`,,`,`,,`---

2)5
Daneshlink.com
daneshlink.com
(4) When cement paste gains strength, are any chemical bonds created; if so, how does it
happen that when a hardened paste of known strength and porosity is ground up
and recompacted to the same porosity it has the same strength?
(5) Why does slag-cement concrete at a given strength only have 1/10 to 1/100 the
permeability of portland-cement concrete of the same strength?
CONCLUDING REMARKS
These examples of high-performance concrete, both those developed in the Corps of Engineers
over the last two centuries and those under development today, suggest two conclusions. First, if
there is a need for a concrete, mortar, or grout mixture that has a type or level of performance not
possessed by ordinary commercial products available over-the-counter locally, the probability is that
the Corps of Engineers can develop the needed high-performance mixture. Second, the types and
levels of performance that should be required in specifications for concrete for any given class of
use and service should be appropriate to those needed. The levels of quality demanded should be
those that are relevant (Mather 1986). Indeed, it has been suggested that one way for concrete to be
of proper (high enough) performance is to "stamp out specification overkill," i.e., to not require
kinds and levels of performance that are not relevant to the service needed (Mather and Mather
1987).
ACKNOWLEDGEMENT
I appreciate the cooperation of the authorities at the U.S. Army Engineer Waterways Experiment
Station and the Headquarters of the U.S. Army Corps of Engineers that permitted me to prepare
and present this paper for publication. I also thank Dr. Tony C. Liu for helpful suggestions and
review of an earlier draft.
REFERENCES
American Society for Testing and Materials, 1993, "Standard Test Method for Time of Setting of
Hydraulic-Cement Paste by Gillmore Needles," Designation: C 266-89, Annual Book ASTM
Standards, V. 04.01, pp. 189-191, Philadelphia, Penn., also Corps ofEngineers, HandbookforConcrete
and Cement, CRD-C 223, Waterways Experiment Station, Vicksburg, Miss. (revised quarterly).
American Society for Testing and Materials, 1993, "Standard Test Method for Abrasion Resistance
of Concrete (Underwater Method), Designation: C 1138-89," Annual Book ifASTM Standards, V.
04.02, pp. 599-602.
Anon, 1994, "Concrete Attains 29,000 psi," Engineering News-Record, V. 233, No. 28, 11 July 1994, p
13, McGraw-Hill, New York, NY 10020.
Bendinelli, R. A., and. Burnett, W. L, 1967, Grouting in Support of Underground Nuclear Testing,
Waterways Experiment Station MP 6-914, 26 pp.
Chamberlin, W. P., 1983, "The Cleft-Ridge Spain: America's First Concrete Arch," Industrial
Archeology, V. 9, No.1, pp. 29-44.
--``,```,`,`,`,,,,`,`````,`,,`,,-`-`,,`,,`,`,,`---
236
Daneshlink.com
daneshlink.com
Corps of Engineers, 1968, Research for Civil Works, A Progress Report, Civil Works Directorate, Office
of the Chief of Engineers, Corps of Engineers, U. S. Army, Washington, D.C., 216 pp.
Freeman, D. S., 1934, R E. Lee, A Biograpf:y, V. 1, pp. Scribner's, New York, N.Y.
Gillmore, Q. A., 1863, Practical Treatise on Umes, Hydraulic Cement, and Mortars, Professional Paper of
the Corps of Engineers, U. S. Army No. 9.
Gillmore, Q. A., 1871, Practical Treatise on Coignet-Beton and Other Artificial Stone, D. Van Nostrand,
NY, 115 pp.
Gulick, C. W., and Wakeley, L. D., 1989, Reference Properties of Cement-Based Plugging and Sealing Materials
for The Waste Isolation Pilot Plan ( IPP), Waterways Experiment Station, TR SL-89-17, 20 pp.
Vicksburg, Miss.
Holland, T. C., 1983,Abrasion-Erosion Evaluation of Concrete Mixtures for Stilling Basin Repairs, Kinzua
Dam, Penn.rylvania, MP SL-83-16, U.S. Army Engineer Waterways Experiment Station, Vicksburg,
Miss.
Holland, T. C., 1984, Abrasion-Erosion Evaluation of Concrete Mixtures for Stilling Basin Repairs, Kinzua
Dam, Penn.rylvania, Report 2, Supplementary Studies, MP SL-83-16, U.S. Army Engineer Waterways
Experiment Station, Vicksburg, Miss.
Holland, T. C., and Holland, L. L., 1985, Letters ofEngineer Officers Describing Experiments Conducted
Using the Tremie for Placing Concrete Under Water, 1845-1851 (unpublished), Mantua, Ohio.
Holland, T. C.; Krysa, A.; Luther; M.D.; and Liu, T. C., 1986,. "Use of Silica-Fume Concrete to
Repair Abrasion-Erosion Damage in the Kinzua Dam Stilling Basin," Ffy Ash, Silica Fume, Slag, and
Natural Pozzolans in Concrete, Proceedings, Second International Conference, Madrid, Spain, V. M. Malhotra,
ed.,ACI SP-91, V. 2, American Concrete Institute, Detroit, Mich., pp. 841-863.
Husbands, T. B.; Malone, P. G.; and Wakeley, L. D., 1994, Peiformance of Concretes Proportioned with
Pyrament Blended Cement, Construction Productivity Advancement Research (CPAR) Program,
Waterways Experiment Station, CPAR-SL-94-2, Vicksburg, Miss, 39180.
Liu, T. C., 1980, Maintenance and Preservation of Concrete Structures, Report 3, Abrasion-Erosion Resistance of
Concrete, Technical Report C-78-4, U. S. Army Engineer Waterways Experiment Station, Vicksburg,
Miss.
Liu, T. C., 1981, "Abrasion Resistance of Concrete," Journal of the American Concrete Institute,
Proceedings, V. 78, No.5, pp 341-350.
Mather, B., 1974, "Use of Concrete of Low Portland Cement Content in Combination with
Pozzolans and Other Admixtures in Construction of Concrete Dams," Journal of the American Concrete
Institute, Proceedings, V. 71, pp. 589-599; closure Journal of the American Concrete Institute, Proceedings, V.
72, pp. 296-297 (1975).
Mather, B., 1980, "Use Less Cement," Concrete International, V. 2, No. 10, pp. 22-24.
237 --``,```,`,`,`,,,,`,`````,`,,`,,-`-`,,`,,`,`,,`---

Daneshlink.com
daneshlink.com
Mather, B., 1986. "Selecting Relevant Levels of Quality," Concrete International,

V. 8, No. 3, pp 30-36.
Mather, B., 1967, "Stronger Concrete," A Symposium on Concrete Strength, Highway Research Record
No. 210, pp. 1-28 (with 60 references), Highway Res. Bd., Washington, DC.
Mather, K., 1965, "High Strength, High Density Concrete," Journal of the American Concrete Institute,
Proceedings, V. 62, No. 8, pp. 951-962, reprinted in Concrete for Nuclear Reactors, Clyde E. Kesler, ed.,
ACI SP-34, V. III, American Concrete Institute, Detroit, Mich., pp. 1587-1596.
Mather, K., and Mather, B., 1987, "Reflections on Concrete Durability and on International
Conferences on Concrete Durability," Concrete Durability, Katharine and Bryant Mather International
Co!iference,John M. Scanlon, ed.,ACI SP-100, V. 1, American Concrete Institute, Detroit, Mich., pp.
--``,```,`,`,`,,,,`,`````,`,,`,,-`-`,,`,,`,`,,`---
1-9.
McDonald, J. E., 1980, Maintenance and Preservation of Concrete Structures, Report 2, Repair ofErosion
Damaged Structures, Technical Report C-78-4, U.S. Army Engineer Waterways Experiment Station,
Vicksburg, Miss.
Neeley, B. D., 1988, Evaluation of Concrete Mixtures for Use in Unde17JJater Repairs, Technical Report
REMR-CS-18, U.S. Army Engineer Waterways Experiment Station, Vicksburg, Miss., AD A193897,
130 PP·
Neeley, B. D.; Saucier, K. L.; and Thornton, Jr., H. T., 1990, LAboratory Evaluation of Concrete Mixtures
and Techniques for Unde17JJater Repairs, Technical Report REMR-CS-34, U.S. Army Engineer Waterways
Experiment Station, Vicksburg, Miss., AD A231195, 92 pp.
Wakeley, L. D., 1990, "Grouts and Concretes for the Waste Isolation Pilot Project (WIPP)," Scientific
Basis for Nuclear Waste Management XIII, Virginia M. Overs by and Paul W. Brown, Materials Research
Society Symposium Proceedings, V. 176, pp. 45-51, Pittsburgh, Penn.
Wakeley, L. D., and Emzen, J. J ., 1992, Grout for Closure ofthe Demonstration Vault at the U.S. DOE
Hanford Facility, Waterways Experiment Station, TR SL-92-21, 156 pp. Vicksburg, Miss.
Wakeley, L D.; Poole, T. S.; Emzen,J.J., and Neeley, B.D., 1993, "Salt Saturated Mass Concrete
Under Chemical Attack," High Performance Concrete in Severe Environments, American Concrete Institute,
ACI SP-140, Paul Zia, ed., American Concrete Institute, Detroit, Mich., pp. 239-268.
Wakeley, L. D., and Poole, T. S., 1987, Studies of the Geochemical Stability of a Salt-Saturated
Expansive Grout, Waterways Experiment Station MP SL-87-10, 76 pp, Vicksburg, Miss.),
238
Daneshlink.com
daneshlink.com
Reprinted from the ACI Fall Convention, Baltimore, j'yf[), 1999, ©American Concrete Institute.
1
ROBERT E. LEE AND CONCRETE INNOVATIONS
By Bryant Mathe(
Douglas Southall Freeman's authoritative biography of Robert E. Lee has a chapter on the building
ofF ort Carroll in the middle of Baltimore Harbor in 1849-1852. In the spring of that year, Lee
established that there was a stable hard surface 45 ft below low water and began to work on the
construction. These preliminary activities, as recounted by Freeman, included the following: "He
experimented in the laying of concrete under water with a tremie." Lee continued with the work
until August 1852 when he was sent to be Commandant at West Point. By then some concrete had
been placed in Fort Carroll. Lee received information from General Totten on 22 June 1849 on
placing concrete with a tremie. Lee replied on 25 June, "I shall make experiments to test the tremie
preparatory to laying foundations." These experiments are among the earliest bits of concrete
research done in the USA.
INTRODUCTION
I grew up in Baltimore, attended Public Schools 32 and 49, Baltimore City College ('34),Johns
Hopkins University (AB '36), and went on an annual Sunday School picnic by excursion boat to the
Eastern Shore, passing close to Fort Carroll both coming and going. I then married a classmate at
the Hopkins who came from Petersburg, Virginia, and later became an ACI Director and Honorary
member. Her family listened to the news daily on the radio from Richmond as read by Douglas S.
Freeman, whose real interest was Robert E. Lee, so when his four-volume life of Lee came out in
1935, they bought it. Later I read parts of it and came upon the reference to experiments with the
use of a tremie for underwater placement of concrete. Much later Ben Gerwick got in touch with me
to urge that I try to get Captain Terry Holland assigned to the Corps of Engineers Concrete
Laboratory in Vicksburg, Mississippi, when he completed his doctoral work at the University of
California at Berkeley (Holland 1983). It worked. Terry came to Mississippi. His thesis dealt with
placing concrete underwater and he and Ben Gerwick were not aware of the work of Lee and other
Corps of Engineers people on this topic back in the middle of the 19th century. They had assumed
that the 04brd English Dictionary (Oxford University Press 1971) knew what it was talking about
when it included under "tremie" that the first use of the word tremie to describe a concrete
placement device in an American publication appeared in a 1905 issue of The Engineering Record
(1905). The term was also used in an issue of the Journal of the American Concrete Institute in 1910 (Hoff
1910).
I called his attention to Freeman. He and his wife, Laura, a librarian, during vacations went to
Washington and ran down in the National Archives a whole series ofletters about experiments on
and use of tremies for concrete placement by the Corps of Engineers in coastal fortification
1
Presented at the Session on "From Robert E. Lee to the Egyptians" sponsored by ACI Committee 120 on History
of Concrete at the ACI Convention, Baltimore, MD, November 1999.
2
Director, Structures Laboratory, U.S. Army Engineer Research and Development Center, 3909 Halls Ferry Road,
Vicksburg, MS 39180-6199.
--``,```,`,`,`,,,,`,`````,`,,`,,-`-`,,`,,`,`,,`---

239
Daneshlink.com
daneshlink.com
construction in 1839 to 1851. They located letters from General Totten, Chief of Engineers, to Lee
dated 22June 1849 and Lee's reply dated 25 June 1849. They are summarized below:
Totten to Lee. 22 June 1849: Totten sent information to Lee, then working on the
construction of a fort in Baltimore harbor, describing the experiments done by Dutton and
Wright. When describing Wright's results, Totten alluded to his own earlier experience using
the trernie, "They accord with my own results - obtained years ago - partial it is true, but
always to my mind decisive." Lee was directed to begin his own experiments with a trernie
"as soon as practicable." Totten also recommended that Lee fabricate his trernie of "rather
thin boiler iron - taking care that the edges of all the laps turn downwards."
Lee to Totten. 25 June 1849: Lee acknowledged receipt of Totten's directions for testing the
trernie at the work in Baltimore harbor. He wrote, "As soon as I am prepared, I shall make
experiments on this subject, preparatory to laying similar foundations when the experience
derived from the works at Cay West and the Tortugas, will be of much value."
Holland and Holland (1985) reported that Totten may have learned about the trernie in connection
with his publication, in 1838, of Esst!JS on Hydraulic and Common Mortars and on Lime-Burning, which
was a translation from the French.
In 197 6 a history of the Baltimore District, Corps of Engineers, was published. It has a chapter on
Fort Carroll and refers the reader to Freeman for details.
TREMIE
--``,```,`,`,`,,,,`,`````,`,,`,,-`-`,,`,,`,`,,`---
There seems to be some confusion about the definition of trernie. ACI I 16R has three entries for
tremie and one for tremie concrete:
tremie - a pipe or tube through which concrete is deposited under water, having at its
upper end a hopper for filling and a bail for moving the assemblage.
tremie concrete - see concrete, tremie.
tremie seal - the depth to which the discharge end of the trernie pipe is kept embedded in
the fresh concrete that is being placed; a layer of tremie concrete placed in a cofferdam for
the purpose of preventing the intrusion of water when the cofferdam is dewatered.
concrete, tremie - subaqueous concrete placed by means of a tremie.
Stein (1993) gives two definitions, not, to me, well differentiated:

Tremie
1. Pipe or tube through which concrete is deposited under water, having at its upper
end a hopper for filling and a bail for moving the assemblage.
2. Pipe having a funnel-shaped upper end into which concrete is fed for depositing.
Bottom end is normally kept continuously buried in the newly placed concrete and is
raised with the rise of the concrete in the structure. Generally used when placing
concrete under water, or when the location of depositing is too difficult or too far to
reach by normal placing methods.
Thrush (1968) gives one definition and gives the source:

240
Daneshlink.com
daneshlink.com
tremie -A box or frame of wood or metal used for depositing concrete underwater. Its
upper section forms a hopper above water to receive the concrete, and it may be moved
laterally or vertically by any suitable device, as a traveling crane. Standard, 1964.
The Dictionnaire Technique des Bmrages in its 1967 edition in which, for the meeting of I COLD in
Istanbul, a seventh column (fi.irkc,:e) was added to those already included (Franc,:ais, English,
Deutsch, Espafi.ol, Italiano, and Portuges) as No. 5.268 has. Webster:r Third International Dictionary
(Gove 1968) tells us that the word "tremie" is from the French "tremie" = "hopper," which comes
from the Latin "trimodia," a measure containing 3 "modii" and that "modius" is a Roman grain
measure= 0.96 peck; thus, 3 modii would be 3 pecks or about 27 L, based on IEEE/ASTM SI 10-
1997, that replaced ASTM E 380 as the standard for use of the SI that says on p. 29 that the "peck
(US dry)" is equal to about 9 E -03 m 3 or 9liters (actually 8.809768 L).
Tiirkc;e Frans;ais English Deutsch Espafiol Italiano Portuges Nos

Su altinda tuyau (m) pow: tremie Rohr (n) fiir tubo (m) para tubo (m) per tubo (m) para 5.268
betondokme betonnage pipe Unter hor-migonado getto di betonagem
borusu sous Wasserbetonierung (j) bajo el agua calcestruzzo debaixo de
l'eau sotto acqua :igua
The Corps of Engineers specification for tremie concrete work at Wolf Creek Dam (Holland and
Turner 1980) required a tremie apparatus with a funnel-shaped hopper of at least 1/2 yd3 volume. If
my conversion is correct, 1/2 yd3 =about 0.4 m 3 or about 400 L, a long way from 27 L.
FORT CARROLL
Kanarek (1976) in his history of the Corps of Engineers Baltimore District noted that after the War
of 1812 when the British burned Washington, folks on the Mid-Atlantic Coast were in favor of forts
to avoid future such depredations. In 1821 the Board of Engineers anticipated two forts near
Baltimore, one at the extreme end of Sollers Point Flats and the other at Hawkins Point, seven miles
below the city on the Patapsco River. These defenses would force an invader to land at a longer
--``,```,`,`,`,,,,`,`````,`,,`,,-`-`,,`,,`,`,,`---
distance from the city, make his march long and costly, and prevent direct bombardment by water.
Fort McHenry would be retained as a secondary barrier.
In 1847 work began on the fort at Sollers Point Flats, seven miles outside of Baltimore near mid-
channel in the Patapsco River. The original design for the proposed fort, formulated by the Board of
Engineers in 1818, called for a seven-sided island fort without flanking arrangements. The Board
later modified its design to a regular hexagon containing three tiers of guns in casemate and one in
barbette. The work began in August under Major C. A. Ogden who remained in charge until 15
November when Robert E. Lee took over. As a second lieutenant in the Corps of Engineers, Lee
served as assistant engineer during construction ofF orts Monroe and Calhoun from 1829 through
1834. From 1837 to 1841 he was the superintending engineer for the harbor at St. Louis, Missouri,
and from 1841 to 1846 he supervised the construction at Fort Hamilton in New York harbor in
what is now Brooklyn.
For three years and five months Lee supervised the building of the fort at Sollers Point Flats.
Because the fort was to be erected on a shoal, Lee's first task was to examine the character of the
shoal and the strength of the foundation it would provide. He arranged to procure piles that would
be long enough for the platforms of the sea walls and directed the building of a crane to lift stone
and other heavy materials. In his report for 1850, Lee could recount considerable progress.

~41
Daneshlink.com
daneshlink.com
During his last full year in Baltimore, Lee became frustrated in his work. A peak had been reached
early during the fort's construction; making further headway proved arduous. Building a fort on an
artificial island was fraught with dangers. The foundation for the sea wall on Front 1 had been
graded and leveled, and three courses of stone laid. Lee procured enough stone to make the wall 14
ft high. Finally, he supervised the building of a carpenter and blacksmith shop on the site of the fort.
Colonel Lee remained in charge of the building of Fort Carroll until27 May 1852, when General
Totten assigned him to become superintendent of the United States Military Academy at West
Point. His active days as an engineer in the United States Army were over. They ended in Baltimore
working on a tedious project which became stalled because of a lack of funds. Congress was
questioning the fort's practicality. As Douglas Southall Freeman, Lee's biographer, notes, the Fort
Carroll experience "with the slow and troublesome construction of stone forts of the old type
unconsciously operated to make him still more an advocate of earthworks."
Despite many setbacks, some progress could be measured. In the fall of 1855, the seawalls had been
filled with concrete and were complete. Seven hundred twenty-two pieces of cut granite had been
laid in these walls with the aid of the diving bell. In addition, the piling for the foundation of the
scarp wall and piers had been started. Two years later, all the piles for the foundation had been
driven in place and sawed off on Fronts 2, 3, and 4.
By the end of 1858, all the foundation piles had been sawed off and the entire grillage timber laid.
The scarp wall had been raised on all six fronts of the fort to heights varying from 18 to 24 ft. In
addition, foundations for two magazines were now in place. Settling on Front 6 brought a halt to
building of the scarp wall in 1859.
Certainly by 1861 and the opening of the Civil War, Fort Carroll was not a reliable defensive
structure that could be counted upon to protect the city of Baltimore. The engineers did load more
than 3300 tons of rough granite into Front 6 to try to prevent further subsidence. Forty-four guns
had been mounted on the fort's other five fronts.
Captain William P. Craighill became Baltimore District Engineer in 1865. He thought the first
priority should go to completing Fort Carroll. "In its present condition," he reminded the Chief of
Engineers, "litde account should be taken of Fort Carroll as its principal face is nearly to the level of
the water ... " An enemy fleet could pass it easily, engage Fort McHenry, and burn the city at the
same time. He wrote:" ... I think the best way in which to complete Fort Carroll is to make it a
--``,```,`,`,`,,,,`,`````,`,,`,,-`-`,,`,,`,`,,`---
casemated work of three tiers without barbette." He was ignored. Fort Carroll was never completed.
Like Fort McHenry, Fort Carroll continued to stagnate during the three decades after the Civil War.
It remained an unfinished, decaying work standing uselessly out in the middle of the Patapsco River.
Only its lighthouse was of any value. In 1878 the Board of Engineers for Fortifications
recommended a radical modification of the original structure to enable it to hold the heaviest
modem rifled guns, protected by armor thick enough to be impervious to naval bombardment.
According to the new plan, those guns in casemates would be arranged in a single tier behind iron
shields; those in barbette would be placed singly between traverses behind earthen parapets. Craighill
noted in 1882, " ... and it should be commenced without delay." Apparendy few supported his view.
Nothing was done except stopgap repairs to prevent total deterioration.

242
Daneshlink.com
daneshlink.com
--``,```,`,`,`,,,,`,`````,`,,`,,-`-`,,`,,`,`,,`---
As early as 1893, the Board of Engineers on Fortifications outlined an elaborate defense plan for
Maryland's most important commercial port. It recognized Baltimore's extensive growth and,
echoing Craighill's sentiments of earlier years, it called for a line of defense far out in front of Fort
Carroll. Specifically, the Board envisioned construction of new batteries at Fort Carroll, Hawkins
Point, North Point, and Rock Point.
The Army officially abandoned the forts in March 1921.
REFERENCES
ACI, 1997, "Guide for Measuring, Mixing, Transporting, and Placing Concrete," ACI Committee
304 in ACI 304R-89 (Reapproved 1997), ACI Manual of Concrete Practice, Part 2, Farmington
Hills, Mich., 48333, 49 pp.
Commission Internationale des Grands Barrages (ICOLD), 1967, Dictionnaire Technique des Barrages,
ICOLD, Paris, 408 pp.
Engineering News-Record, 1905, "The Substructure of the Cambridge Bridge, Boston," The Engineering
Record, V. 51, No. 2,Jan. 14, pp. 52-53.
Freeman, D. S., 1935, R E. Lee, A Biograply, Volume 1, Charles Scribners' Sons, New York, N.Y., pp
301-318.
Gerwick,Jr., B. C., 1964, "Placement ofTremic Concrete," Symposium on Concrete Construction in

Aqueous Environments, ACI SP-8, American Concrete Institute, Detroit, Mich., pp. 9-20.
Gove, P. B. (editor in chief), 1968, Webster's Third New International Dictionary rif the English Language,
Unabridged, G. & C. Merriam Company, Springfield, Mass, 2662 pp.
Hoff, 0., 1910, "Laying Concrete Under Water- Detroit River Tunnel," Journal if the American
Concrete Institute, V. 6, pp. 180-19 5.
Holland, T. C., 1983, Tremie Concrete for Massive Underwater Construction, Ph.D. diss., University of
California, Berkeley.
Holland, T. C., and Holland, L. L., 1985, Letters if Engineer Officers Describing Experiments Conducted
Using the Tremie for Placing Concrete Under Water, Second Preliminary Report, unpublished, 30 pp.
Holland, T. C., and Turner,]. R., 1980, Construction rifTremie Concrete Cutrif!Wall Wo!fCreek Dam,
Kentucky, Structures Laboratory, U.S. Army Engineer Waterways Experiment Station, Vicksburg,
Miss., 39180, Miscellaneous Paper SL-80-10, 90 pp.
IEEE/ASTM, 1997, Standardfor Use rifthe International System if Units (SI): The Modern Metric System,
Institute of Electrical and Electronic Engineers, New York, N.Y. 100 1 7, and ASTM, West
Conshohocken, Penn., 19428, 66 pp.

'243
Daneshlink.com
daneshlink.com
Kanarek, H., 197 6, The Mid-Atlantic Engineers: A History if the Baltimore District, U. S. Army Cops if
Engineers, 1774-1974, U.S. Government Printing Office, Washington, D.C. 20402, 196 pp.
Oxford University Press, 1971, The Compact Edition if the O:iford English Dictionary, London.
Standard, 1964, Funk & Wagnalls Co. Funk & Wagnalls New Standard Dictionary if the English Language,
New York (Copyright 1963), 2816 pp.
Stein, J. S., 1993, Construction Glossary, An Enryclopedic Reference and Manual, Second Edition, John Wiley
and Sons, Inc., New York, N.Y., 10 1 58, 1137 pp.
Thrush, P. W., 1968, A Dictionary ifMining, Mineral, and Related Terms, Special Publication, U.S.
Bureau of Mines, Department of the Interior, Washington, D.C. 20402, 1269 pp.
--``,```,`,`,`,,,,`,`````,`,,`,,-`-`,,`,,`,`,,`---

244
Daneshlink.com
daneshlink.com
Reprinted from Concrete International, V 23, No. 1, 2001, Farmington Hills, MI, pp. 46-47, copyright American Concrete
Institute.
SELF-CURING CONCRETE,
WHY NOT?
by Bryant Mather
MUSINGS BY MATHER
In a letter in the September 2000 issue of Concrete International, D. Srinivasan asked, ''Will there be a
self-curing concrete?" My answer to this is strongly affirmative for three reasons. First, most of the
concrete that is produced and placed each year all over the world already does self-cure. Some of it
wasn't intended to have anything done to its exterior surface. But finishing did in fact take place, and
yet the concrete's ability to serve its intended purpose had not been significantly reduced.
It was this fact that prompted Russell Fling, when he became ACI President, to ask the Technical
--``,```,`,`,`,,,,`,`````,`,,`,,-`-`,,`,,`,`,,`---
Activities Committee (TAC) to re-establish the Institute's Curing Committee. The charge to the
committee arose from President Fling's experience that a lot of the concrete with which he was
concerned received no treatment under the label of "curing," yet seemed to be no worse for that
having been the case. I was given the job of chairing that committee. The committee's report stated
in its Section 1.2:
"Curing is the maintaining of a satisfactory moisture content and temperature in concrete during its
early stages so that desired properties may develop. Curing is essential in the production of concrete
that will have the desired properties. The strength and durability of concrete will be fully developed
only if it is cured. No action to this end is required, however, when ambient conditions of moisture,
humidity, and temperature are sufficiently favorable to curing. Otherwise, specified curing measures
shall start as soon as required."
Second, most of the concrete in the world is placed in quantities that are of sufficient thickness such
that most of the material will remain in satisfactory conditions of temperature and moisture during
its early stages. This is so regardless of what steps are taken, or not taken, to ensure that such
conditions are maintained in the exterior layer.
Third, there are cases in which concrete has been greatly assisted in moving toward a self-curing
status either inadvertently or deliberately through actions taken in the selection and use of materials.
I think it may have been Rudy V alore who told me the following story about concrete in Dallas,
Tex. Cedric Willson, at Texas Industries, was trying to encourage engineers and contractors to use
structural low-density ("lightweight") aggregate concrete. He arranged for its experimental use in
half of the construction of the second-floor slab of a small motel on a street comer; the other half
was placed using normal-density aggregate.
The low-density aggregates were batch wet. When the time came, in the late afternoon, to apply the
white-pigmented curing compound that would form a protective membrane, the contractor's
representative at the site decided to skip this step since the wind direction was such that, had he

2 Licensee=University of Texas Revised Sub Account/5620001114, User=erur, ert
Daneshlink.com
daneshlink.com
applied it, a lot of effort would have been spent getting membrane material off the cars parked
downwind. The next morning, Cedric went to see how his low-density concrete had turned out and
was surprised and pleased. The normal-density concrete was virtually destroyed by drying shrinkage
cracks. On the other hand, the low-density concrete had practically no cracking - enough absorbed
water had been brought to the concrete, though not included in the water-cement ratio, to
effectively compensate for the contractor's failure to apply the prescribed membrane-forming curing
compound.
We at the Corps of Engineers Concrete Laboratory used this bit of technology when confronted
with one of the few cases I am aware of in which one should supply additional water during curing
to permit desired properties to develop. There are, as noted, few such cases. In most instances,
water needs to be applied not because too little was used in making the concrete, but because some
was lost after deposition in the forms. The two cases I know of when added water is needed are 1)
when expansive cement is used or 2) when the concrete has a water-cementitious materials ratio
lower than 0.4 by mass.
Our problem was using concrete with a highly expansive cement to plug tunnels. The customer
wanted the plug to push against the tunnel in order to partially close the cracks parallel to the
interior of the tunnel that form after the bore is complete. In some cases, the plug was placed a
kilometer back in the tunnel. There was no way one could bring water to the surface concrete in
such a plug except to the face at the end of the plug. Some imported vacuum-saturated low-density
aggregate from California and used it in tunnel plugs.
Finally, there is the experience we had with the use of very rapid-hardening concrete that won
George Hoff and Al Buck the ACI Wason Medal for Research in 1985 for their paper
"Considerations in Prevention of Damage to Concrete Frozen at Early Ages" (1983). They
confirmed that the achievement of a compressive strength of 3.5 MPa (500 psi) was sufficient to
protect concrete from damage by a cycle of freezing and thawing. If one were to use a cement such
as Reg-Set or admixtures that provide equal high-early compressive strength such as 20.7 MPa (3000
psi) in 30 minutes, then there is no need for "curing" to protect the concrete against the damage that
can occur from a cycle of freezing and thawing.
If something needs to be invented, it might be a system by which some "enteric" coated particles or
--``,```,`,`,`,,,,`,`````,`,,`,,-`-`,,`,,`,`,,`---
capsules containing a membrane-forming curing compound - or containing a substance that reacts
with water to do so -is distributed over the surface of a slab in the final stages of finishing. The
particles would open if the surface became dry, and a membrane would form while the concrete was
still water-saturated to its top but had no free water on the surface. Phil Cady has a patent on an
indicator that, when placed on top of a slab, changes color at that critical time when something
should be done to prevent undesirable water loss from the first millimeter of concrete at the surface.
If, at that moment, the curing membrane were to automatically create itself on the surface of the
concrete, we'd have it made.
REFERENCES
Hoff, G., and Buck, A., "Considerations in the Prevention of Damage to Concrete Frozen at Early
Ages," ACIJOURNAL, ProceedingsV. 80, No.5, Sept.- Oct. 1983, pp. 371-376.
Srinivasan, D., ''Will there be a self-curing concrete?" Concrete International, V. 22, No.9, Sept. 2000,
PP· 6-7.

Daneshlink.com
daneshlink.com
APPENDIX-SELECTED BIBLIOGRAPHY OF PUBLISHED WORKS OF

KATHARINE AND BRYANT MATHER
This section presents a selected bibliography of published works of Katharine and Bryant Mather.
Their combined careers in concrete technology spanned over 100 years. Their publications greatly
contributed to the field of concrete, and advanced the understanding and promoted better use of
concrete for civil infrastructure applications.
The list initially compiled by Bryant contained over 800 references to publications of Bryant and
Katharine Mather. Eventually, Bryant narrowed down the list to a more manageable 164 published
works. The bibliography includes a vast array of topics, from testing of fine aggregate to unsolved
problems in predicting concrete behavior, and covers over 50 years of research, beginning in 1945.
The bibliography is arranged chronologically by the year of publication.
--``,```,`,`,`,,,,`,`````,`,,`,,-`-`,,`,,`,`,,`---

247
Daneshlink.com
daneshlink.com
1) Mather, B., 1945, "Determination of Surface Moisture in Fine Aggregate," Engineering News-
Record, V. 134, No. 12, pp. 104-105, New York, N.Y.
2) Mather, B., 1945a, "Investigation of Porous Concrete, Santee Cooper Dam," No. Adanta
Division, CE, Mt. Vernon, N.Y., 31 pp.
3) Mather, B., 1945b, "Effect ofType ofTest Specimen on Apparent Compressive Strength of
Concrete," Procedure, ASTM, V. 45, pp. 802-812, Philadelphia, PA.
--``,```,`,`,`,,,,`,`````,`,,`,,-`-`,,`,,`,`,,`---
4) Mather, B., 1946d, "Limulus Polyphemus -The King Crab," Maryland, A Journal tifNatural
History, V. 16, No.3, pp. 47-53 Gul), The Natural History Society of Maryland, Baltimore,
MD.
5) Mather, B., 1947, "The Testing of Aggregates in Air-Entrained Concrete."
6) Mather, B., 1948, "Chemical Reactivity of Dolerite (Diabase) (44-204)," Journa4 ACI,
Proceedings, V. 44, pp. 1058-1060, Detroit, Mich.
7) Mather, B., 1948, "Petrographic Identification of Reactive Constituents in Concrete

Aggregates" Proceedings, ASTM, V. 48, pp. 1120-1125, Philadelphia, P A. Also "Symposium
on Methods and Procedures Used in Identifying Reactive Materials in Concrete," pp. 66-71,
discussion pp. 72-73.
8) Mather, B., 1949, Distribution No.1 of Supplements Corrections, and Revisions to

Handbook for Concrete and Cement, Vicksburg, MS.
9) Kelly, T.M.; Schuman, L.; and Hornibrook, F.B., 1949, "A Study of Alkali-Aggregate
Reactivity by Means of Mortar Bar Expansions," ACI Journal Proceedings, V. 45, p 80-1 to 80-3
Gun), Detroit, MI.
10) Mather, B., 1950, "Method of Petrographic Examination of Aggregates for Concrete" (with
Katharine Mather), Proceeding,ASTM, V. 50, pp. 1288-1313, Philadelphia, PA.
11) Blanks, R.F.; Cook, H.K.; and Kennedy, T.B., 1950, Discussion of "Modem Concepts
Applied to Concrete Aggregates," American Society of Civil Engineers, V. 115, p 432-434,
New York, NY.
12) Mather, B., 1951, "Tests of Commercially Available Cementitious Materials," WES TM 6-
335, Vicksburg, MS.
13) Mather, B., 1951, "Precision and Accuracy," The Mzlitary Engineer, V. 43, No. 292, p 145,
Alexandria, VA ..
14) Mather, B., 19 51, "Tests of Fine Aggregate for Organic Impurities and Compressive
Strength in Mortars," ASTM Bulletin No. 178, pp. 35-43, Philadelphia, PA.
15) Mather, B., 1951, Distribution No. 6 of Supplements, Corrections, and Revisions to

?48
Daneshlink.com
daneshlink.com
16) Mather, B., 1951, Distribution No.7 of Supplements, Corrections, and Revisions to
19) Brewer, H.W., and Burrows, R.W., 1951, Discussion of "Coarse-Ground Cement Makes
More Durable Concrete," ACI]ournal, V. 47, pp .. 360-6 to 360-8, American Concrete
Institute, Detroit, MI.
20) Mather, B., 1952, "Cracking of Concrete in the Tuscaloosa Lock," Proceedings Highwqy Research
.Board, V. 31, pp. 218-233, Highway Research Board, Washington, DC.
23) Mather, B., 1952, Distribution No. 12-13 of Supplements, Corrections, and Revisions to
24) Mather, B., 19 53, Distribution No. 14 of Supplements, Corrections, and Revisions to
28) Mather, B., 1953, "Concrete's Etymological Offspring," AC1 Journal, Proceedings, Letters (A
Reader), pp. 682-684, (Mar), Detroit, MI.
29) Burnett, G.E., and Spindler, M.R., 1953, Discussion of"Effect ofTime of Application of
Sealing Compound on the Quality of Concrete," ACI Journal, Proceedings, V. 49, pp. 200-1 to
100-4 (Dec), Detroit, MI.
30) Mitchell, L.J., 1954, Discussion of"Dynamic Testing ofMaterials," Proceedings Highwqy
Rcsean:h Board, V. 33, pp. 256-258, Highway Research Board, Washington, DC.
--``,```,`,`,`,,,,`,`````,`,,`,,-`-`,,`,,`,`,,`---

249
Daneshlink.com
daneshlink.com
31) Mather, B., 1954, "Radiosotope Techniques in concrete Research," ACI]ournal, Proceedings,
V. SO, pp. 889-890, Detroit, MI.
32) Mather, B., 1954, "Radiosotope Techniques in Concrete Research," ACI Journal, Proceedings,
letters, pp. 889-890.
33) Mather, B., 1955, "Shape, Surface Texture, and Coatings" (on Aggregates) ASTM SIP 169,
pp. 284-296, American Society For Testing Materials, Philadelphia, PA.
34) Mather, B., 1955, Investigations of Cement-Replacement Materials- Preliminary

Investigations, WES MP 6-123, Report 1, Vicksburg, MS.
35) Mather, B., 1955, Plastic-Glass Fiber Reinforcement for Reinforced and Prestressed
Concrete-Summary oflnformation to 1955, WES TM 6-421, Report 1, Vicksburg, MS.
36) Anderegg, F.O., 1955, Discussion of "Some Volume Changes in Mortars and Concrete," by
ASTM Bulletin No. 210 (Dec), pp. 65, Philadelphia, PA.
37) F eld, J., Discussion of "Structural Success or Failure," Proceedings, American Society Civil
Engineers, V. 81, p 758-5 to 758-6 Gul), New York, NY.
38) Philleo, R.E., 1955, Discussion of "Comparison of Results of Three Methods for
Determining Young's Modulus of Elasticity of Concrete," ACI]ournal, Proceedings, V. 51, p
472-1 to 472-3 (Dec), Detroit, MI.
39) Mather, B., 1956, "Eurema daira daira in Mississippi," Lepid News, V. 10, No.6, pp. 204-
206.
40) Mather, B., 1957, "Tests of Schokbeton Precast Concrete Units in Iceland," WES MP 6-235,
Vicksburg, MS.
41) Mather, B., 1957, "Factors Affecting Durability of Concrete in Coastal Structures," Beach
Erosion Board TM 96.
--``,```,`,`,`,,,,`,`````,`,,`,,-`-`,,`,,`,`,,`---
42) Mather, B., 1957, "Laboratory Tests of Portland Blast-Furnace Slag Cements," ACI]ournal,
Proceedings, V. 54, pp. 205-232, Detroit, MI.
43) Mather, B., 1957, Discussion of "Permeability, Pore Pressure, and Uplift in Gravity Dams,"
Proceedings (Sep 55), p 807-5 to 807-6, Transactions, American Society of Civil Engineers, V.
122, p 603, New York, NY.
44) Mather, B., 1958, "Quality Materials for Highway Construction," Engineering Bulletin, Purdue
University, No. 95, pp. 95-100 (WES MP 6-277), also Purdue University, Rnad School
Proceedings, V. 42, No. 4, pp. 95-99.
45) Mather, B., 1958, Distribution No. 37 of Supplements, Corrections, and Revisions.

750
Daneshlink.com
daneshlink.com
46) Mather, B., 1958, "The Partial Replacement of Portland Cement in Concrete," in ASTM
STP No. 205, pp. 37-73, Philadelphia, PA.
47) Mather, B., 1959, "Effectiveness of Mineral Admixtures in Preventing Excessive Expansion
of Concrete Due to Alkali-Aggregate Reaction (with L. Pepper)," WES MP 6-345,
Vicksburg, MS, andASTM Proceedings, V. 59, pp. 1178-1203.
48) Abdun-Nur, E.A., and Waddell, J.J ., 1959, Discussion of "Control of Concrete Mixes," ACI
Journal, Proceedings, V. 31, No.3, p. 1543.
49) Mather, B., 1960, "Plastic-Glass Fiber Reinforcement for Reinforced and Prestressed
Concrete," Summary oflnformation Available 1955-1959 WES TR 6-421, Report 2,
Vicksburg, MS.
50) Mather, B., 1960, "Portland-Pozzolan Cement Made with Tennessee Valley Authority Fly
Ash," WES TR 6-541 (May), 43 pp., Vicksburg, MS.
51) Mather, B., and Mather, K., 1961, "Application of Modem Laboratory Techniques to
Engineering Petrography in Concrete", Journal, Mississippi, Academy of Science, Proceedings, 1955,
V. VI, pp. 96-100.
52) Mather, B., and Tynes, W.O., 1961, "Investigation of Compressive Strength of Molded
Cylinders and Drilled Cores of Concrete", ACI Journal, Proceedings, V. 57, pp. 767-778.
53) Mather, B., 1962, "Investigation of Partially Compacted Weight of Concrete as a Measure of
Workability, Preliminary Tests" WES TR 6-598, Report 1, Vicksburg, MS.
54) Mather, B., 1962, "Investigation of Partially Compacted Weight of Concrete as a Measure of
Workability, Preliminary Tests" WES TR 6-598, Report 1, Vicksburg, MS.
55) Mather, B., 1962, "Investigation of Concrete in Eisenhower and Snell Locks, St. Lawrence
Seaway- Inspection and Repairs" WES MP 6-493, Report 1, Vicksburg, MS.
56) Mather, B., 1962, "Second Conference on Fundamental Research in Plain Concrete," ACI
Newsletter, Apr. 1962, pp. 31-33.
57) Mather, B., 1963, "High - Compressive- Strength Concrete -A Review of the State of the
Art" AFSWC-TDR-62-56 (also WES MP 6-520).
58) Mather, B., 1963, "Research on Suitability of Locally Available Aggregates for Highway
Construction" WES MP 6-568 (April) 13 pp.; also Doc No. 88, Chap VI, Topic 8 of
Proceedings, 9th Pan American Highway Congress, V. 2, pp. 63 7-652.
59) Mather, B., 1964, "Drying Shrinkage- Second Report," Highwqy Research News, No. 15, pp.
34-38.
60) Mather, B., 1964, "The Strength of Portland-Cement Concrete as Affected by Air, Water,
and Cement Content" WES MP 6-624, Vicksburg, MS.
--``,```,`,`,`,,,,`,`````,`,,`,,-`-`,,`,,`,`,,`---
Copyright American Concrete Institute 2'51

Daneshlink.com
daneshlink.com
61) Mather, B., 1964, Examination of Concrete, U.S. Military Academy, West Point, NY (with
A.D. Buck) WES MP 6-653, Vicksburg, MS.
62) Mather, B., 1965, "Shape, Surface Texture, and Coatings of Aggregate" WES MP 6-710,
Vicksburg, MS.
63) Mather, B., 1965, "Investigation of Cement-Replacement Materials, Compressive Strength

Development of 193 Concrete Mixtures During 10 Years of Moist Curing," WES MP 6-
123, Report 12, Vicksburg, MS.
64) Mather, B., 1965e, "The Strength ofPordand-Cement Concrete as Affected by Air, Water,
and Cement Content," Proceedings, Army Science Conference, 1965, V. II, pp. 227-232.
65) Mather, B., 1965, "Research in Plain Concrete," in U. of Illinois, T &M Report 272, pp. 5-16.
66) Mather, B., 1965, "Concrete from A to Z," WES MP 6-709 (also ACI Journal, Proceedings, V.
62, pp. 513-520).
67) Mather, B., 1965, "Research on Suitability of Locally Available Aggregates for Highway
Construction," Proceedings, 9th Pan-American Highway Congress, V. 2, pp. 637-652 (also WES
MP 6-568). Chapter VI Topic 8, Document No.88, 14 pp., with Spanish summary; U.S.
Department of State, Washington, D.C. (4 #200).
68) Mather, B., 1965, "Partially Compacted Weight of Concrete as a Measure of Workability"
WES MP 6-737, Vicksburg, MS. (cf#244).
69) Mather, B., 1965, "Investigation of Expanding Cements," Summary of Information

Available July 1963, WES TR 6-691, Report 1 Vicksburg, MS.
70) Mather, B., 1966, "Concrete Deterioration," National Sand and Gravel Association
Technical Information Letter No. 245, National Ready Mixed Concrete Association
Technical Information Letter No. 236, 3 Mar., Washington DC, pp. 6.
71) Mather, B., 1966, "Shape, Surface Texture, and Coatings- Concrete Aggregates," ASTM
STP 169A, pp. 415-431.
72) Mather, B., 1966, "Partially Compacted Weight of Concrete as a Measure of Workability,"
Journal ACI, Proc., V. 63, April, pp. 441-450. (cf #228).
73) Mather, B., 1966, "Effects of Seawater on Concrete" Highwqy Research Record No. 113,
Symposium on Effects ofAggressive Fluids on Concrete, HRB, pp. 33-42; NAS-NRC Publication
1335.
74) Mather, B., 1967, "Unsolved Problems in Predicting the Behavior of Concrete" WES MP 6-
890, Vicksburg, MS.
--``,```,`,`,`,,,,`,`````,`,,`,,-`-`,,`,,`,`,,`---

Daneshlink.com
daneshlink.com
75) Mather, B., 1968, "The Behavior of Concrete" Department of Civil Engineering, North
Carolina State University, Raleigh, N.C., 24 pp. (Third Annual Henry M. Shaw Lecture in
Civil Engineering).
76) Mather, B., 1968, "Stronger Concrete," Symposium on Concrete Strength, Highway Research
Record No. 210, NAS-NAE-NRC Publication 1553, pp. 1-28.
77) Mather, B., 1968, "Field and Laboratory Studies of the Sulfate Resistance of Concrete,"
--``,```,`,`,`,,,,`,`````,`,,`,,-`-`,,`,,`,`,,`---
Performance if Concrete- Resistance if Concrete to Sulphate and Other Environmental Conditions-A

Symposium in Honour ifThorbergur Thorvaldson, Canadian Building Series, No. 2, University of
Toronto Press, pp. 66-76.
78) Mather, B., 1968, "Cracking Induced by Environmental Effects," Causes, Mechanism, and
Control if Cracking in Concrete, ACI, SP-20, pp. 67-72, American Concrete Institute, Detroit,
MI.
79) Mather, B., 1968, "Concrete Admixtures, Aggregates and Durability," Highway Research
Record, No. 226, Highway Research Board, NAS-NAE-NRC, Publication 1612.
80) Mather, B., 1968, "Unsolved Problems in Predicting the Behavior of Concrete," Symposium on
Correlation ifMaterials Characteristics with Systems Performance; Air Force Materials Laboratory,
pp. 187-198.
81) Mather, B., 1969, "Deterioration of Concrete in Eisenhower Lock," Trans., Ninth
International Congress on Large Dams, V. VI, pp. 427-429, Paris.
82) Shah, S.P., and Slate, F.O., 1969, Discussion of "Internal Microcracking Mortar-Aggregate
Bond and the Stress-Strain Curve of Concrete," Trans., Ninth International Congress on
Large Dams, V. VI, p. 116, Paris.
83) Ruetz, W., 1969, Discussion of "The Two Different Physical Mechanisms of Creep in
Concrete," Trans., Ninth International Congress on Large Dams, V. VI, p. 162, Paris.
84) Mather, B., 1969 "Pozzolan," Reviews in Engineering Geology, V. II, Geological Society of
American, Boulder, CO, pp. 105-118.
85) Mather, B., 19691, "Foreword to Portland Cement Concrete," Highwqy Research Record, No.
268, Highway Research Board, Washington, D.C., p. V.
86) Mather, B., 1969, "Some Notes on Durability of Prestressed Concrete," Fed Int.
Precontrainte.
87) Mather, B., 1969, "Sulfate Soundness, Sulfate Attack, and Expansive Cement in Concrete,"
Proceedings, RILEM International Symposium on Concrete Durability, Technical University, Prague,
Czechoslovakia, pp. C-209 to C-220 (Sep).

Daneshlink.com
daneshlink.com
88) Mather, B., 1969, "Research on Concrete, Stanton Walker Lecture Series on the Materials
Sciences," Lecture No. 7, University of Maryland (National Sand and Gravel Association,
National Ready Mixed Concrete Association, Washington, D.C.), 15 pp.
89) Mather, B., and Buck, A.D., 1970, "Deterioration of Concrete Sidewalks and Curbs,"
Obseroations ofthe Performance of Concrete in S enJice, Highway Research Board Special Report 106,
pp. 44-52, Vicksburg, MS.
90) Newlon, H.H., 1970, Discussion of"Random Cracking ofBridge Decks Caused by Plastic
Shrinkage," Highwqy Research BoardSpecial Report 106, pp. 60-61, Highway Research Board,
Washington, DC.
91) Kokubu, M., 1970, Discussion of "Fly Ash and Fly Ash Cement," Proceedings, Fifth
International Symposium on Chemistry of Cement, Tokyo, Japan, V. IV, p 113.
92) Mather, B., 1970, "Portland Cement Concrete, Research, Testing and Performance," Journal
ofMaterials, V. 5, No.4, pp. 830-841, Dec. (Marburg Lecture), American Society for Testing
and Materials, Philadelphia, P A.
93) Mather, B., 1970, Discussion of Question B; "Discussion on Filling of Air Voids,"
Proceedings, RILEM Symposium on Durabiliry of Concrete, Prague, Czechoslovakia, V. III, pp.
B189-192.
94) Mather, B., 1970, Discussion of "The Filling-Up of Air- Entrained Voids in Concrete,"
Proceedings, RILEM Symposium on Durabiliry of Concrete, Prague, Czechoslovakia, V. III, pp. 193-
194.
95) Mather, B., 1970, "Admixtures," Rapid Construction of Concrete Dams, ASCE, pp. 75-102 (Oct).
96) Mather, B., 1972, "Some Reflections on Expansive Cement Concrete," Proceedings, Seminar on
Expansive Cement Concrete, University of California, Berkeley, CA.
97) Mather, B., 1972, "Alkali-Carbonate Rock Reaction in Concrete: Developments in

Specification and Control," WES MP C-72-9, 18 pp., Vicksburg, MS.
--``,```,`,`,`,,,,`,`````,`,,`,,-`-`,,`,,`,`,,`---
98) Mather, B., 1972d, "Relating Materials Quality to Materials Performance to Structural
Performance of Concrete," in Performance Concept in Buildings, NBS Special Publications
361, V. 1 Proc. OfRILEM-ASTM-CIB Symposium, pp. 719-724 (also issued as WES MP C-
72-7).
99) Mather, B., 1972, "Economic Aspects of Curing Mass Concrete," Economical Construction of
Concrete Dams, Proceedings of the Engineering Fd. Conference, ASCE, pp. 39-53, (also WES
MP C-72-19).
100) Mehta, P .K., 1973, Discussion of the paper "Mechanism of Expansive Associated with
Ettringite Formation," Cement and Concrete Research, V. 3, No. 5, pp. 651-652 (Sep).

~54
Daneshlink.com
daneshlink.com
101) Mather, B., 1973, "Durability of Concrete," Materials '73, Cement and Concrete Association
(England), pp. 17-30.
102) Mather, B., 1974, ''Water/Cement Ratio," Concrete (London), p. 26 (Apr).
103) Mather, B., 1974, "Concrete Engineering- Present and Future," Concrete Journal, Japan
National Council on Concrete QANACC No. 100), V. 12, No.1, pp. 11-13 (in Japanese and
English) Qan).
104) Mather, B., and Mather, K., 1974, "Damage to Concrete from Reactions Between Aggregate
and Alkalies" Proceedings, International Seminar on Alkali-Silica Reaction, Karlstrup, Denmark
(May).
105) Mather, B., 1974, Cement Specifications, Pordand Cement Association, Chicago, Ill., also
National Ready Mixed Concrete Association, Silver Spring, MD., Tech. Info. Letter No. 306.
106) Mather, B., 1974, LowStrength Concrete in Columns, American Concrete Institute, Adanta, Ga.
(31 Oct).
--``,```,`,`,`,,,,`,`````,`,,`,,-`-`,,`,,`,`,,`---
107) Mather, B., 1975, "Use of Concrete of Low Pordand Cement Content in Combination with
Pozzolans and Other Admixtures in Construction of Concrete Dams," ACI Journal,
Proceedings, V. 71, pp. 589-599 (Dec 74), and Author's Closure, Jour. ACI Proc., V. 72, pp.
296-297 Qun 75).
108) Mather, B., 1975, "Developments in Specification and Control," in Cement-Aggregate Reactions,
Transportation Research Record 525, pp. 38-42, (11 ref) Washington, DC.
109) Mather, B., 1975, "New Concern Over Alkali-Aggregate Reaction," WES MP C-75-3;
Concrete Technology Information Analysis Center (CTIAC) Report No. 17, 23 pp.; also
published by the National Sand and Gravel Association (NSGA), and the National Ready
Mixed Concrete Association (NRMCA), Silver Spring, Md., as NSGA Circular No. 122,
NRMCA Publication No. 149, 29 pp. (Mar).
110) Mather, B., 1976, "How Soon is Soon Enough?" ACI]ournal, Proceedings, V. 73, No.3, pp.
147-150 (Mar).
111) Mather, B., 1976, "Strength Depends on Water-Cement Ratio, Not on Cement Content,"
Concrete Construction, V. 21, No.9 (Sep), pp. 477-478.
112) Diamond, S., 197 6, "Discussion of the Paper 'A Review of Alkali-Silica Reaction and
Expansion Mechanisms Reactive Aggregates,' Cement and Concrete Research, V. 6, pp. 813-814,
Pergamon Press.
113) Mather, B., 1977, "Case Histories of Unsatisfactory and Abnormal Field Performance of
Concrete During Construction," Transportation Research Record No. 651, TRB, pp. 25-29.

Daneshlink.com
daneshlink.com
114) Mikhail, R.S.; Youssef, A.M.; and Shater, M., 1978, "A Discussion of the Paper 'Air
Entrainment in Portland Blast Furnace Slag Cement Pastes: Effects on Strength and Pore
Structure,"' Cement and Concrete Research, V. 8, pp. 257-258, Pergamon Press.
115) Mather, B., 1979,"Concrete Need not Deteriorate," Concrete International, V. 1, No.9 (Sep),
pp. 32-37.
116) Mather, B., 1979, "Tests of High-Range Water-Reducing Admixtures," ACI SP-62,
Proceedings International Symposium on Supeplasticiifrs in Concrete, pp. 157-168.
117) Nixon, PJ.; Collins, R.J.; and Rayment, P.L., 1980, Discussion of Paper "The Concentration
of Alkalies by Moisture Migration in Concrete- A Factor Influencing Alkali-Aggregate
Reaction," Cement and Concrete Research, V. 10, pp. 111-113 Qan).
118) Mather, B., 1980, "Mineral Aggregate For Concrete- Needed Research," Concrete Technology,
pp. 57-110, Energy, Mines and Resources, Ottawa, Ontario, Canada, 796 pp ..
--``,```,`,`,`,,,,`,`````,`,,`,,-`-`,,`,,`,`,,`---
119) Mather, B., 1980, "Use Less Cement," Concrete International, V. 2, No. 10 (Oct), pp. 22-24.
120) Mather, B., 1980, "The Future of Expansive Cement," ACI SP-64, Expansive Cement, pp. 1-7.
121) Mather, B., 1981, "Crushed Concrete as Concrete Aggregate," Adhesion Problems in the
Rerycling of Concrete, P.J. Kreijger, Ed., NATO Conference Series, Materials Science, V. 4,
Plenum Press, NY, pp. 293-302.
122) Mather, B., 1982, ''When Will the Concrete Industry Enter the Twentieth Century," Concrete
International, V. 4, No. 1 Gan), pp. 25-28.
123) Mather, B., 1982a, "Concrete in Sea Water," Concrete International, V. 4, No.3 (Mar), pp. 28-
34.
124) Tuthill, L.W., 1983, Discussion of "Alkali-Silica Reaction- 40 Years Later," Concrete
International, V. 5, No.2 (Feb), pp. 65-67.
125) Mather, B., 1983, "Accelerators and Drying Shrinkage," Concrete Construction Gul), pp. 563-
565.
126) Mather, B., 1984, "New Test for Sulfate Resistance," Concrete Construction (Feb), pp. 244, 246.
127) Cohen, M.D., 1984, Discussion of "Theories of Expansion in Sulfoaluminate-type

Expansive Cements: Schools of Thought," Cement and Concrete Research, V. 14, pp. 603-609
Gul).
128) Mather, B., 1984, "Shallow Waves," Concrete Construction Gul), p 666.
129) Mather, B., 1984, "The Glass in Low-Calcium Fly Ash," Cement and Concrete Research, V. 14,
No.6, pp. 887-890 (Nov).

756
Daneshlink.com
daneshlink.com
130) Mather, B., 1985, "Importance of Condition Assessment," presented at Building Technology
Symposium at National Bureau of Standards, Gaithersburg, Md. (5 Nov), 5 pp.
131) Mather, B., 1986, "All Right to Salt the First Winter," Reader Response, Concrete Construction
(Feb), pp. 231.
132) Mather, B., 1986b, "Selecting Relevant Levels of Quality," Concrete International, V. 8, No. 3
(Mar), pp. 30-36.
133) Mather, B., 1987, ''Voluntary Standardization on the Threshold of Change," ASTM
Cavanaugh Memorial Symposium, Standardization News, ASTM V. 15, No. 3, Part 2, pp. 6-9.
134) Mather, B., and Mather, K., 1987, "Reflections on Concrete Durability and on International
Conferences on Concrete Durability," Concrete Durabiliry, Katharine and Bryant Mather
International Conference, ACI SP-100, pp. 1-9.
135) Hope, B.B., and Ip, A.K.C., 1988, Discussion of "Corrosion of Steel in Concrete Made with
Slag Cement," ACI Materials Journal, V. 84, No.5, p. 473.
136) Mather, B., 1988, "Concrete Alkali-Aggregate Reaction," Proceedings of the th International
Cotiference, P.E. Grattan-Bellew, Ed., Canadian Journal of Civil Engineering, V. 15, p 933.
137) Mather, B., 1989, "How to Make Durable Concrete," The Construction Specifier, V. 42, No.
--``,```,`,`,`,,,,`,`````,`,,`,,-`-`,,`,,`,`,,`---
1, pp. 48-51.
138) Mather, B., 1990, Comment on "Some Concrete Basics," Concrete Repair Bulletin, V. 3, No.1,
p 2, Washington, DC.
139) Mather, B., 1990, "Durability- A Different Answer for Each Exposure," in Goeb, E.O.,
Technical Talk, in Concrete Products, V. 93, No. 6, p 5.
140) Mather, B., 1990, "How to Make Concrete That Will Be Immune to the Effects of Freezing
and Thawing," Paul Klieger Symposium on Petformance of Concrete, ACI SP-122, pp. 1-19.
141) Mather, B.; Cohen, M.D., and Olek,J., 1991, "Silica Fume Improves Expansive-Cement
Concrete," Concrete International, V. 13, No.3, pp. 31-38.
142) Mather, B., and Cohen, M.D., 1991, "Sulfate Attack on Concrete- Research Needs,"
Materials Journal, V. 88, No.1, Qan-Feb), American Concrete Institute, Detroit, MI., pp. 62-
69
143) Mather, B., 1991, "Defective Concrete in Condominium Balconies," Concrete Repair Bulletin,
pp. 4, 12-13 Qan-Feb).
144) Mather, B., 1991, "Aggregates," The Heritage ofEngineering Geolo!!Ji: The First Hundred Years,
Geological Society of America, Centennial Special V. 3, Boulder, CO, pp. 323-332.

257
Daneshlink.com
daneshlink.com
145) Mather, B., 1991, "Selecting Relevant Levels of Quality," Qualiry Assurance in Concrete
Construction, Compilation 16, American Concrete Institute, Detroit, MI, pp. 9-16
146) Mather, B., 1992, "Calcium Chloride in Type V-Cement Concrete," Durability of Concrete, G.
M. Idorn International Symposium,ACI SP-131, pp. 169-178.
147) Rzonca, G.F., et al., 1992, Discussion of "Concrete Deterioration, East Los Angeles County
Area: Case Study," journal ofPetjormance of Constructed Facilities, ASCE, V. 6, No. 1, pp. 66-67
(Feb).
148) Mather, B., 1992, "Alkali-Aggregate Reaction: In the Beginning," Alkali-Silica Reaction:
Fiction, Fact, and Fear, ACI Convention, Washington, DC (Mar).
149) Rollings, R.S., and Rollings, M.P., 1993, Discussion of "Pavement Failures: Oversights,
Omissions, and Wishful Thinking," Journal of Constructed Facilities, ASCE, V. 7, No.1, p 75
(Feb 93).
150) Mather, B., 1993, "Use of Admixtures to Prevent Excessive Expansion of Concrete due to
Alkali-Silica Reaction," Transportation Research Record No. 1382, pp. 93-102, Transportation
Research Board, Washington, DC 20418 (38 ref).
151) Mather, B., 1993, "Concrete in Transportation: Desired Performance and Specifications,"
Transportation Research Record No. 1382, pp. 5-10 (1993 TRB Distinguished Lecture),
Transportation Research Board, Washington, DC.
152) Mather, B., 1995, "The Voluntary Standards Process in the United States," Letter to the
Editor, ASTM Standardization News, V. 23, No. 5, p 10.
153) Sadegzadeh, M.; Page, C.L.; and Vassie, P.R.W., 1995, Discussion of"Effects ofUrea on
Durability of Reinforced Concrete," Magazine of Concrete Research, V. 45, No. 164, 1993, pp.
179-186.
154) Mather, B., 1995, "How to Avoid Excessive Expansion of Concrete Due to Alkali-Aggregate
Reaction," Proceedings, Second International Conference on Alkali-Aggregate Reactions in Ffydroelectric
Plants and Dams, Chattanooga, TN, USCOLD, Denver, CO. , pp. 421-439.
155) Pigeon, M., and Malholtra, V.M., 1996, Discussion of "Frost Resistance of Roller-
Compacted High-Volume Fly Ash Concrete" Journal ofMaterials in Civil Engineering, American
Society of Civil Engineers, V. 7, No. 4, pp. 208-211.
156) Mather, B., 1996, "Ettringite Terminology," Letter to the Editor, Cement and Concrete Research,
V. 26, No. 11, p. 1745.
157) Mather, B., 1996, "High-Performance Concrete in the U.S. Army Corps of Engineers,"
International Workshop on High Petjormance Concrete, Bangkok, Thailand, Nov 94, ACI SP-159,
pp. 323-333, American Concrete Institute, Farmington Hills, MI.
--``,```,`,`,`,,,,`,`````,`,,`,,-`-`,,`,,`,`,,`---

758
Daneshlink.com
daneshlink.com
158) Mather, B., 1996, "Eliminating Ripples Over Rebar," Concrete Construction, V. 41, No. 8, p 635
(Aug).
159) Mather, B., 1997, "Cement and Concrete Standards of the Future," Keynote Address at
Workshop on Cement and Concrete Standards for the Future, National Institute of
Standards and Technology, Gaithersburg, MD, 11 Oct 95, Proceedings, NISTR 5933, pp. 19-
24.
160) Mather, B., 1998, "How to Make Concrete That Will Not Suffer Deleterious Alkali-Silica
Reaction" Proceedings if the Engineering Foundation Coiference on Advances in Cement and Concrete,
Banf, Alberta, Canada.
161) Hirve, W.G., and Mather, B., 1999, "'Sulfate Attack' ofls It?," Cement and Concrete Research, V.
29, pp. 789-791.
162) Mather, B., 1999, "Update-Use of Admixtures to Prevent Alkali-Silica Reaction," ACI
Chicago Session ASR Laboratory Testing and Long-Term Field Peiformance, Mar. 1999.
163) Mather, B., 1999, "How to Make Concrete That Will Not Suffer Deleterious Alkali-Silica
Reaction," Cement and Concrete Research, V. 29, pp. 1277-1280.
164) Mather, B., 1999, Realiifng the Potential if Concrete A Construction Material in Creating with Concrete,
(V. 1 of 11), pp. 1-10. Thomas Telford, London.
--``,```,`,`,`,,,,`,`````,`,,`,,-`-`,,`,,`,`,,`---

259
Daneshlink.com
daneshlink.com
--``,```,`,`,`,,,,`,`````,`,,`,,-`-`,,`,,`,`,,`---

Daneshlink.com
daneshlink.com
--``,```,`,`,`,,,,`,`````,`,,`,,-`-`,,`,,`,`,,`---
"We have, however, been working professionally since 1941 in research, development, testing,
and evaluation in support of concrete construction, often seeking to establish how the art
got into the state it seemed to be in and often trying to suggest how it might get into a better
state by the application of the results of experience either in the field or in the laboratory
either from research or practice."
-Katharine and Bryant Mather

"The History of Concrete Construction in the 20th Century," ,
from Concrete International, October 1981

Daneshlink.com

Aci sp-223-2004

Uploaded by

Copyright:

Available Formats

Aci sp-223-2004

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Aci sp-223-2004

Uploaded by

Copyright:

Available Formats

daneshlink.

Copyright American Concrete Institute

ACI Special Publication

American Concrete Institute@;

Copyright American Concrete Institute

First printing, October 2004

Cover photo: Tuscaloosa Dam

Printed in the United States of America

Editorial production: Shannon B. Banchero

Library of Congress Control Number: 2004114043

Copyright American Concrete Institute

This commemorative publication is the result of contributions from a number of individuals. Ed

Copyright American Concrete Institute

Copyright American Concrete Institute

Part I - Potential of Concrete ........................................................................................................... 9

Copyright American Concrete Institute

PART 1- POTENTIAL OF CONCRETE

Copyright American Concrete Institute

Copyright American Concrete Institute

REALIZING THE POTENTIAL OF CONCRETE AS A CONSTRUCTION

Copyright American Concrete Institute

Copyright American Concrete Institute

Copyright American Concrete Institute

Copyright American Concrete Institute

The Symposium was held on November 4,1971, in Buffalo, New York.

Copyright American Concrete Institute

"The concept of distinguishing acceptable non-conforming concrete from

I am told that a new edition of this may appear in 1999.

Copyright American Concrete Institute

The potential of concrete as a construction material will be better realized when:

Copyright American Concrete Institute

Copyright American Concrete Institute

PART II- CEMENTITIOUS MATERIALS

Cementitious materials are an integral component of concrete. The properties of cementitious

The main conclusions drawn in the above-mentioned paper were:

4. Specifications for evaluating admixtures for effectiveness in preventing excessive

Copyright American Concrete Institute

Copyright American Concrete Institute

EFFECTIVENESS OF MINERAL ADMXTURES IN PREVENTING

by Leonard Pepper and Bryant Mather'

Army Engineer Waterways Experiment Station, Concrete Division, Jackson, Miss.

Copyright American Concrete Institute

Class of Material Material

2. Natural cements Natural I (Nat I)

3. Fly Ashes Fly Ash I (FA I)

4. Natural Volcanic Gases Purnicite (Prun F)

5. Calcined opaline shales Calcined Shale (C Sh M)

6. Uncalcined Diatomite Uncalcined diatomaceous earth(Unc D)

7. Uncalcined Quart Silica Flour

8. Glass Synthetic, pure silica glass (SS glass)

Copyright American Concrete Institute

TESTS AND RESULTS

Test for Reactivity with Sodium Hydroxide (NaOH):

Modified Test for Reactivity with NaOH:

Copyright American Concrete Institute

Reduction in 14 days 0.154 RoH + 38.2 0.84 12.4 238.6

percent reduction in mortar expansion.

Copyright American Concrete Institute

Replacement, Solid Volumea, percent