SINTEF REPORT

TITLE

SINTEF Building and Infrastructure

COIN Concrete Innovation Centre
Address:
Location:
Telephone:
Fax:

NO-7465 Trondheim
NORWAY
Richard Birkelands vei 3
+47 73 59 52 24
+47 73 59 71 36

Enterprise No.: NO 948 007 029 MVA

COIN P1
SP 1.2 F

Advanced cementing materials

Controlling hydration development

Accelerating admixtures for concrete

State of the art
AUTHOR(S)

Roar Myrdal
CLIENT(S)

Aker Kvrner Engineering and Technology, Borregaard LignoTech,

maxitGroup, Norcem A.S, Norwegian Public Roads Administration,
Rescon Mapei AS, Spenncon AS, Unicon AS, Veidekke ASA and
The Research Council of Norway
REPORT NO.

CLASSIFICATION

CLIENTS REF.

SBF BK A07025 Unrestricted

Terje F. Rnning

CLASS. THIS PAGE

ISBN

PROJECT NO.

Unrestricted

978-82-536-0989-8

3D006020

ELECTRONIC FILE CODE

FILE CODE

NO. OF PAGES/APPENDICES

35

PROJECT MANAGER (NAME, SIGN.)

CHECKED BY (NAME, SIGN.)

Harald Justnes

Harald Justnes

DATE

APPROVED BY (NAME, POSITION, SIGN.)

2007-12-05

Tor Arne Hammer, Centre Manager

ABSTRACT

Accelerating admixtures affect the rates of reactions between cement and water to give an overall
increase in the hydration rate. Thus, the use of accelerators in concrete provides a shortening of setting
time and/or an increase in early strength development. This report gives a historical and technical
overview of liquid accelerators used for concrete, their mode of actions and the characteristics of todays
commercial accelerators. Accelerators for both normal and sprayed concrete are covered. Antifreeze
admixtures for winter concreting are also discussed.
Calcium chloride, the beste accelerator ever, is no longer an option due to severe corrosion hazzards in
steel reinforced concrete. Inorganic salts of nitrate (for setting) and thiocyanate (for hardening) are
common alternatives, although they are less efficient than calcium chloride. Many commercial
admixtures today are made of mixtures of nitrate and thiocyanate. Some also contain alkanolamines and
carboxylic acids or their salts. Combinations of chemicals may give interesting synergistic effects.
Alkali-free shotcrete accelerators are based on aluminum sulphate as main ingredient, often combined
with diethanolamine and organic acids (oxalic, formic) and/or inorganic acids (phosphorous,
hydrofluoric).
The report gives recommendations for further research and development of accelerators for concrete.
KEYWORDS
GROUP 1
GROUP 2
SELECTED BY AUTHOR

ENGLISH

Materials technology
Concrete
Accelerator
Cement
Hydration

NORWEGIAN

Materialteknologi
Betong
Akselerator
Sement
Hydratisering

Foreword
COIN - Concrete Innovation Centre - is one of presently 14 Centres for Research based
Innovation (CRI), which is an initiative by the Research Council of Norway. The main objective
for the CRIs is to enhance the capability of the business sector to innovate by focusing on longterm research based on forging close alliances between research-intensive enterprises and
prominent research groups.
The vision of COIN is creation of more attractive concrete buildings and constructions.
Attractiveness implies aesthetics, functionality, sustainability, energy efficiency, indoor climate,
industrialized construction, improved work environment, and cost efficiency during the whole
service life. The primary goal is to fulfill this vision by bringing the development a major leap
forward by more fundamental understanding of the mechanisms in order to develop advanced
materials, efficient construction techniques and new design concepts combined with more
environmentally friendly material production.
The corporate partners are leading multinational companies in the cement and building industry
and the aim of COIN is to increase their value creation and strengthen their research activities in
Norway. Our over-all ambition is to establish COIN as the display window for concrete innovation
in Europe.
About 25 researchers from SINTEF (host), the Norwegian University of Science and Technology NTNU (research partner) and industry partners, 15 - 20 PhD-students, 5 - 10 MSc-students every
year and a number of international guest researchers, work on presently 5 projects:
1
2
3
4
5

Advanced cementing materials and admixtures

Improved construction techniques
Innovative construction concepts
Operational service life design
Energy efficiency and comfort of concrete structures

COIN has presently a budget of NOK 200 mill over 8 years (from 2007), and is financed by the
Research Council of Norway (approx. 40 %), industrial partners (approx 45 %) and by SINTEF
Building and Infrastructure and NTNU (in all approx 15 %). The present industrial partners are:
Aker Kvrner Engineering and Technology, Borregaard LignoTech, maxitGroup, Norcem A.S,
Norwegian Public Roads Administration, Rescon Mapei AS, Spenncon AS, Unicon AS and
Veidekke ASA.
For more information, see www.sintef.no/coin

3
n
d
a
n
d
a
c
c
e
l
e
r
a
t
o
r
s
i
n
g
e
n
e
r
a
l
4
2.1
Cla
ssif
icat
ion
of
acc
eler
ator
s
4
2.2
Be
nefi
ts
pro
vid
ed
by
acc
eler

4
C
a
l
c
i
u
m
c
h
l
o
r
i
d
e
1
0
5
C
a
l
c
i
u
m
n
i
t
r
a
t
e
a
n
d
o
t
h
e
r
n
i
t
r
a
t
e
/
n

1
4
8
C
a
r
b
o
x
y
l
i
c
a
c
i
d
s
a
n
d
t
h
e
i
r
s
a
l
t
s
1
5
9
A
l
k
a
n
o
l
a
m
i
n
e
s
1
7
1
0

1
2
C
o
m
m
e
r
c
i
a
l
a
c
c
e
l
e
r
a
t
o
r
s
f
o
r
n
o
r
m
a
l
c
o
n
c
r
e
t
e
2
5
1
3
C
o
m
m
e
r
c
i

u
m
a
l
u
m
i
n
a
t
e
c
e
m
e
n
t
s
2
8
1
5
R
e
c
o
m
m
e
n
d
a
t
i
o
n
s
f
o
r
f
u
t
u
r
e
R
&
D
2
8

0
1
7
R
e
f
e
r
e
n
c
e
s
3
1

Introduction

The rates of chemical reactions between clinker materials in cements and water, often referred to
as cement hydration reactions, may be altered by adding small amounts of chemical substances to
the cement- water mix. Substances affecting these rates to give an overall increase in the hydration
rate, i.e. an accelerating effect, are termed accelerating admixtures or simply accelerators. Hence,
an accelerator is added to concrete for the purpose of shortening setting time and/or increasing
early strength development. In the first case the main action of the accelerator occurs in the plastic
state of the cement paste, while in the latter case the accelerator acts primarily in the hardened
state. Some accelerators affect either setting or hardening, while several accelerate both setting
and hardening.
The aim of this report is to provide an overview of chemical admixtures reported to accelerate
setting and/or hardening of Ordinary Portland Cement (OPC) and OPC based concrete. Both
accelerators for normal concrete and nozzle added accelerators for sprayed concrete (shotcrete)
will be discussed. First, an outline of accelerators in general is given including an overview of
accelerator consumption in Norway over the last years, followed by a description of specific
accelerators and their mode of action during cement hydration. Also commercial accelerators are
treated.
As almost 100 % of concrete accelerators on the market are in liquid form (a few are sold as
readily soluble powders), and the fact that most operators at concrete mixing plants prefer to
handle accelerators in liquid form, this report does not treat powder accelerators insoluble in
water.
The effect of accelerating admixtures on high alumina cements or CAC (Calcium Aluminate
Cement) is covered only briefly.

2
2.1

Background and accelerators in general

Classification of accelerators

Accelerators may be classified in numerous ways. For instance, the European norm EN 9342:2001 [1] makes a distinction between setting and hardening properties when defining
accelerators for concrete:
Set accelerating admixture
Admixture which decreases the time to commencement of transition of the mix from the plastic to
the rigid state.
Hardening accelerating admixture
Admixture which increases the rate of development of early strength in the concrete, with or
without affecting the setting time.
Setting is normally determined by measuring the time until a certain mechanical resistance is
obtained when a penetration needle is forced into the cement paste (e.g. the Vicat apparatus). A
distinction is even made between initial and final setting time. The level of hardening is normally
determined by measuring the compressive strength of hardened cement paste, mortar and concrete
prisms.

Setting and hardening may also be determined indirectly by measuring the heat evolution or
temperature increase generated from the chemical reactions between cement and water. The
hydration of cement, being an exothermic reaction, produces heat and if the hydration is
accelerated, heat is either produced earlier (setting) and/or at a faster rate (hardening). This is
illustrated in Figure 2.1.
It should, however, be realised that a lot of accelerators affect both setting and hardening,
sometimes depending on the dosage of the accelerator.

Q
or
T

Time, t
Figure 2.1 The effects of setting and hardening accelerators upon the rate of heat
o
evolution Q (W/kg) or temperature T ( C) during hydration of cement.
Reference: Cement paste without accelerator.
Setting: Cement paste with setting accelerator.
Hardening: Cement paste with hardening accelerator.
The setting accelerator produces heat earlier than the reference, but the slopes of the
curves are parallel (equal dQ/dt).
The hardening accelerator starts the production of heat at the same time as the
reference, but the slope is steeper (higher dQ/dt).

From a more practical point of view, accelerators are sometimes classified simply according to the
type of concrete in which they are used:
Accelerator for normal concrete
Accelerator added during mixing of the concrete at the concrete plant
Shotcrete accelerator for sprayed concrete
Accelerator added in the nozzle of the equipment during spraying
Another way of making distinctions between different accelerators is to divide them into groups

according to their physical form and chemical nature, for instance:

Liquid form:
Water soluble inorganic and organic salts and compounds
Slurry form:
Insoluble inorganic salts and compounds dispersed in water
Powder form:
Insoluble inorganic salts and compounds
A detailed overview based on the chemical nature of liquid accelerators is presented in Chapter 3.

2.2

Benefits provided by accelerators

The purposes of using accelerators and the advantages resulting from the use of accelerators are
many. The benefits of a reduced setting time may include [2]:
1 Earlier finishing of surfaces
2 Reduction of hydraulic pressure on forms
1 More effective plugging of leaks against hydraulic pressure
The benefits of an increase in the early strength may include [2]:
2 Earlier removal of forms
3 Reduction of the required period of curing and protection
4 Earlier placement in service of a structure or a repair
5 Partial or complete compensation for the effects of low temperatures on strength
development

2.3

Consumption of accelerators in Norway

Accelerator consumption normal concrete

Even though the benefits of using accelerators are obvious, helping the contractor to meet
specified construction schedules, the total volume of accelerators used for normal concrete is
surprisingly low. According to the European Federation of Concrete Admixture Associations
accelerators for normal concrete make up about 4 % of all admixtures sold in Europe [3]. The
same picture is seen in Norway. In 2006 accelerators (shotcrete accelerators excluded) stood for
only 3.4 % of all admixtures for normal concrete sold in Norway [ 4]. Figure 2.2 shows the trend
in consumption over the last 15 years. There was an increase in the use of accelerators since early
1990s. This is to some extent caused by the increase in cement consumption in the same period
(see Figure 2.3). However, it is shown in Figure 2.4 that despite the increase in cement
consumption, the average consumption of accelerators for normal concrete has increased to about
0.5 kg/ton cement or more the last years compared to less than 0.1 kg/ton cement in the early
1990s. The very high application of accelerators in 2002 (see Figure 2.4) must be linked to
specific building projects that particular year.

Tons

Figure 2.2

7
um
pti
on
of
acc
ele
rat
ors
for
nor
ma
l
co
ncr
ete
in
the
per
iod
T
19
ot
91
al

N
20
o06
r(D
w
ata
efro
gi
m
aN
nCC
cA
o[4]
n).
s
1
5

21
0

1
7

1000000

199

1200
1000
per ton of cement (g/ton)

Average amounts of accelerators

1400

800
600
400
200
0
1991

1994

1997

2000

2003

2006

Figure 2.4 Ratio between total application of accelerators for normal concrete and the total use
of Portland cements in Norway in the period 19912006 (Data from NCCA [4]).
Accelerator consumption sprayed concrete
Due to the large volumes of shotcrete production and the high dosage of shotcrete accelerators
(typically more than 5 % by cement weight), shotcrete accelerators is by far the most used class of
accelerator in terms of volume. For instance, the volume of shotcrete accelerators sold in Norway
in 2006 was approximately 10 times that of accelerators used for normal concrete in the same
period (see Figures 2.2 and 2.5). Unlike the situation for normal concrete, there is no trend of
increase in shotcrete accelerator consumption per ton cement used during the 1990s. On the other
hand, a very clear increase in accelerator consumption relative to cement consumption has been
observed since 2003 (see Figure 2.6). It seems to be a consensus among Norwegian shotcrete
operators today, although not documented, that higher dosages of shotcrete accelerators are
needed to compensate for slower strength development in the concrete nowadays compared to the
late 1990s or early 2000s. The reason for this trend has not yet been clarified, but may be partly
caused by the introduction of alkali-free shotcrete accelerators the very same period (see Chapter
13) and/or the increasing use of blended cements.
7000
6000

Tons

5000
4000
3000
2000
1000
1991

1994

1997

2000

2003

2006

Figure 2.5 Total Norwegian application of accelerators for sprayed concrete (shotcrete) in
the period 19912006 (Data from NCCA [4]).

3
,

Average amounts of shotcrete accelerators

9
,

1,5
1991

n
Figure 2.6
3

a
t
e

38 P

M e
r
S c
h
A l
1 H o
r
4y a
dr
o t
xi e
d
e9 S
i
5
2 C l
hl i
or c
id a
e, t
br e
6o
m10
idAl
e u
a m
7n i
d n
fl a
u t
or e

10

Both type of anion and type of accompanying alkali or alkali earth cation play a role. A screening
of different combinations of some of the anions and cations within this group, based on high and
equivalent molar concentrations, has given the following accelerating effects on tri-calcium
silicate (C3S) hydration in decreasing orders [6]:

2+

2+

2+

Anions:

Br Cl > SCN > I > NO3 > ClO4

Cations:

Ca > Sr > Ba > Li > K > Na Cs > Rb

This ranging points out calcium chloride to be an effective accelerator. Indeed, it has been widely
used for many decades. However, due to its tendency to promote corrosion of steel, it is no longer
recommended for use in steel reinforced concrete. Today, most commercial accelerators for

normal concrete (shotcrete excluded) contain sodium and calcium salts of nitrate (NO3 ), nitrite

(NO2 ) and/or thiocyanate (SCN ) (see Chapter 12).

Silicates and aluminates listed above show quick setting or flash setting properties and are used a
lot in shotcreting. To some extent, potassium carbonate has also been used for this purpose [5].
Potassium carbonate has also been used in binary and ternary antifreeze admixtures [7].
Hydroxides are seldom used as such, but may be part of aluminate solutions to increase pH. High

pH (high concentration of OH ) reduces the risk of precipitation of insoluble aluminium

hydroxide according to the equilibrium

Al(OH)4 = Al(OH)3 + OH

(1)

Soluble organic salts and compounds

Soluble organic salts and compounds can be divided into two main groups:
1 Carboxylic and hydroxycarboxylic acids and their salts
2 Alkanolamines
Among the first group, calcium formate is a well known accelerator, while triethanolamine (TEA)
is the most common among the alkanolamines. Alkanolamines are typically used as ingredients in
accelerator blends, and rarely, if ever, as a sole ingredient. Probably, the most common use of TEA
is to compensate for the retarding effect of some plasticizers.

Calcium chloride

Since use of chloride containing admixtures is no longer an option for steel reinforced concrete,
only a short description of calcium chloride will be given. Calcium chloride (CaCl 2) is
characterised by:

11

1 CaCl2 accelerates both setting and hardening of OPC [2].

2 CaCl2 is the oldest chemical admixture for OPC based concrete; reported as early as 1886
[8].
3 CaCl2 is the cheapest and the most effective accelerator to date [9].
4 Although not fully understood , the mode of action of CaCl2 in concrete and its effect on
the hydration kinetics are well documented [2, 8].
5 The effect of CaCl2 on other cement paste properties is known; for instance
freeze/thaw resistance, sulphate resistance and porosity [2, 10].
6 CaCl2 accelerates both the reaction between tricalcium aluminate (C 3A) and
gypsum (CaSO4 2H2O) and the hydration of tricalcium silicate (C3S) [11].
CaCl2 does not accelerate the hydration of high-alumina cements [2, 12], but has been reported to
accelerate the pozzolanic reaction between Ca(OH)2 and rice husk ash [13]. It is also reported that
CaCl2 activates the hydration of slag [14]. However, a comparison between CaCl2 and Na2SO4
(sodium sulphate) as activators for the pozzolanic reaction of fly ash, showed that Na 2SO4
increased both early and later strengths, while CaCl2 mainly increased later strengths [15].
Despite the research on CaCl2/cement paste interactions for several decades, it has been stated that
the mechanisms responsible for the accelerating effect of CaCl 2 on the hydration of C 3S still are
poorly understood [16]. The new chloride-free accelerators, replacing CaCl 2, are even less
understood, and also less effective. The general situation today can be summarised like this [17]:
Chloride-based accelerators
Severe corrosion hazards
Low cost/performance ratio

Chloride-free accelerators
No (or low) corrosion hazards
High cost/performance ratio

The development of chloride -free accelerators with low cost/performance ratios has been a
challenge for manufacturers of admixtures for many years, and still is.

Calcium nitrate and other nitrate/nitrite salts

Nitrates

According to the ranging of accelerating anions described above, nitrate (NO3 ) is less effective
than chloride (see Chapter 3). The isothermal conduction calorimeter curves shown in Figure 5.1,
comparing the effects of different calcium salts on the rate of hydration of synthetic C 3S, clearly
show that calcium nitrate (Ca(NO3)2) is less effective than CaCl2 on both setting and hardening
[8].
Ramachandran [18] found that Ca(NO3)2 acted as an accelerator of setting in cement paste at low
concentrations, but showed no effect on early strength development. According to El-Didamony et
al [19] Ca(NO3)2 is more effective than CaCl2 as accelerator for the hydration of belite (-C2S).
This has been confirmed by Justnes [20, 21] who has carried out a series of research studies on
+
technical grade calcium nitrate (including a small amount of ammonium, NH 4 ) as concrete
admixture since early 1990s. His findings regarding mode of action in OPC paste are as follows:

12

1 Set acceleration is achieved by low dosages.

2 Must be combined with other components to function as a hardening accelerator.
3 No correlation between set accelerating efficiency and content of C3A in the cement.
4 Initial setting time decreases with increasing content of belite in the cement.
5 Initial setting time decreases with decreasing alkali content in the cement.
6 The mechanism for accelerating setting is two-fold:
(1) An increased calcium concentration leads to a faster super-saturation of the water with
respect to calcium hydroxide (Ca(OH)2).
(2) A decreased sulphate concentration leads to slower/less formation of ettringite which
will shorten the onset of C3A hydration.

Figure 5.1 Isothermal conduction calorimeter curves for C3S with various calcium salts. Dosage is
2 % anhydrous salt by weight of C3 S. The numbers indicate the slopes of the
temperature-time curves from the end of the dormant period to the peak temperature
relative to that of untreated C3S, which assigned a value of 1.0 [8].
Rixom [22] investigated the relationship between temperature and setting and found that CaCl 2
o
can compensate for approximately 15 C drop in temperature, while Ca(NO3)2 can compensate for
o
a 10 C drop.
Calcium nitrate at relatively high dosages has proven to be an effective inhibitor against chloride
induced corrosion of steel reinforcement in concrete [20]. Accordingly, calcium nitrate is a
multifunctional admixture for concrete.
Today, most suppliers of concrete admixtures offer calcium nitrate based accelerators (see Chapter
12).
The behaviour of alkali nitrates in concrete has also been investigated, but not as thoroughly as

Ca(NO3)2. It seems that sodium nitrate (NaNO3) has hardly any accelerating effect on setting

13

behavior [20, 23], but is used to some extent in winter concreting admixtures, so-called antifreeze
admixtures (see Chapters 10 and 12).
Nitrites

Nitrite (NO2 ) differs chemically from nitrate (NO3 ) only in the oxidation state of nitrogen:
Nitrite (ox. state +5) versus nitrate (ox. state +7).
Calcium nitrite (Ca(NO2)2) has probably been the most popular non-chloride setting accelerator in
the USA since late 1960s [8]. Many commercial accelerators today, and particularly antifreeze
admixtures (see Chapters 10 and 12), contain calcium and/or sodium salts of nitrite. Even strength
development effects produced in concrete are reported to be almost comparable to those obtained
with CaCl2 [9]. Despite this widespread use, reports describing the mode of action and
mechanisms involved in cement paste setting are lacking.
Ca(NO2)2 has also proved to be an effective anodic corrosion inhibitor for metals embedded in
concrete [24]. This is probably the main research area today regarding use of Ca(NO 2)2 in
concrete.
Nitrites suffer from one big commercial disadvantage compared to nitrates. Even though calcium
nitrite seems to be the best non-chloride admixture to date [9], nitrites are toxic and
environmentally unfriendly. This means that nitrites, at least for future applications, will be
discouraged due to health and environmental requirements and regulations.

Thiocyanates

Unlike nitrates and nitrites, alkali and calcium salts of thiocyanic acid, HSCN, i.e. thiocyanates

(SCN ), hardly affect setting of OPC, only hardening [25]. Justnes [20] found that a smaller
amount of sodium thiocyanate (NaSCN) together with technical grade calcium nitrate can
counteract the retarding effect of lignosulphonate based plasticizers and give 1 day strength values
in the same range of what was found for samples accelerated by CaCl2.
NaSCN are probably the most recent non-chloride hardening accelerating admixture introduced on
the concrete market (since early 1980s) [8]. Although quite effective, NaSCN is expensive
compared to technical grade calcium nitrate and CaCl2 [26]. Nevertheless, roughly 10,000 tons of
NaSCN are used wordwide every year to accelerate hardening of concrete [26]. Unfortunately,
NaSCN shows some hazardous properties and has to be labelled Xn Harmful [26]. Another
drawback for alkali-thiocyanate based admixtures is the high amount of alkalis introduced into the
concrete and the corresponding potential for alkali-aggregate-reactions (AAR) [7].
Surprisingly little has been published regarding mode of action in cement paste and reaction
mechanisms involved. Most of the information is available only in the patent literature, where it is
seen that thiocyanate salts are often combined with other accelerating admixtures [27]. Some
examples of commercial accelerators containing thiocyanate salts are shown in Chapter 12.
Wise et al [25], probably the first who published research on alkali-free thiocyanates, studied the
effect of different thiocyanates on the hydration of OPC at low temperatures and found:

14

1 Calcium thiocyanate (Ca(SCN)2) is the most effective thiocyanate for promoting strength
development. (In accordance with the ranging of ionic combinations in Chapter 3)
o

2 The addition of 3 % Ca(SCN)2 (by cement weight) to a paste hydrated at 5 C resulted

in 28-days compressive strength that was 74 % of the strength of the paste cured at room
temperature.
3 The least effective salt on compressive strength development was potassium thiocyanate
(KSCN). The paste with KSCN, although producing the highest heat development after
3 days at room temperature, had strength values as low as 2/3 of the reference value.
4 Ammonium thiocyanate (NH4SCN) releases offensive ammonia vapour and therefore is
not a viable chemical for practical applications.
A possible solution to this last point could be to try out amines with less offensive vapours than
ammonium; for instance, alkanolamines like di- or triethanolamines which are also accelerators
for cement hydration (see Chapter 9).
Unfortunately, Wise et al [25] used very high dosages in their study between 1.5 and 3 % by
weight of cement. This is probably far too high for many practical applications (high cost and high
amount of alkalis if alkali thiocyanates are used). Furthermore, practical experiences recommend a
maximum dosage below 1 % by cement weight. One supplier of a NaSCN based accelerator states
that a dosage higher than 0.8 % of a 40 % solution (i.e. 0.32 % active material) does not give
further increase in acceleration, but can in certain cases have a retarding effect [28].
Abdelrazig et al [29] estimated the degree of hydration of OPC pastes by differential scanning
calorimetry and found that addition of NaSCN produced more calcium hydroxide than an equal
molar amount of calcium chloride between 6 and 18 hours after mixing.
Since NaSCN is regarded a corrosive chemical [26], studies have been carried out in order to find
out to what extent addition of this compound causes corrosion of steel reinforcement in concrete.
Nmai and Corbo [30] found that NaSCN at maximum dosage levels recommended (0.75 to 1.0 %)
either reduced or did not affect the corrosion rate of steel in concrete.

Alkali carbonates

Alkali carbonates are easily soluble powders, and have been used to some extent in antifreeze
admixtures for concrete (see Chapter 10), in shotcrete accelerators (see Chapter 11) and in
mortars. Potassium carbonate (K2CO3) is probably the most common. These carbonates are
characterised by their quick-setting properties at moderate to high dosages, probably by promoting
the flash-setting of tricalcium aluminate (C3A) [5]. One study, comparing sodium, potassium and
lithium carbonates in OPCs, showed [31]:
1 At low dosages sodium and potassium carbonates retarded the setting times.
2 At higher dosages (> 0.1 %) sodium and potassium accelerated the setting.
3 Lithium carbonate acted as an accelerator at all concentrations studied.
The quick-setting properties of alkali carbonates can be eliminated by adding a retarder. In this
way it is possible to formulate accelerators to be used in normal concrete [7]. However,
controlling setting time with carbonate based accelerators is a difficult task.

15

Schwarz [32] found that high dosages of K 2CO3 in combination with high dosages of potassium
citrate are highly efficient accelerators for aluminate and ferrite hydration in OPCs. It is assumed
that citrate promotes the dissolution of the ferrite phases by acting via surface complexation and
ligand-promoted dissolution [32]. Citric acid and its salts are normally used as retarders for cement
hydration, but may act as accelerators a high dosages [33].

Carboxylic acids and their salts

Accelerating properties are found among several types of carboxylic acids and their salts:
1 monocarboxylic
2 dicarboxylic
3 hydroxycarboxylic
Popovics [2] claims that water soluble organic materials belonging to the carboxylic acid group
are some of the most effective chloride-free accelerators for cement hydration. Without being
specific regarding type of acid, he states that one acid with a short hydrocarbon chain is effective
due to its ability to promote the dissolution of calcium ions from the cement compounds [2].
Monocarboxylic (formates and acetates)
Accelerating properties have been reported for the lower monocarboxylic acids (C1 to C4) and
their salts (formic, acetic, propionic and butyric), while higher carboxylic acid homologes (>C4)
are retarders [5]. It seems that those belonging to C1 are the most efficient ones, and that the
calcium salt of formic acid, calcium formate (Ca(HCOO) 2), are the most common. Dodson [8]
began research on Ca(HCOO)2 in early 1960s and found that although Ca(HCOO) 2 was not as
effective as CaCl2, it does produce satisfactory increases in early strength as well as that of later
ages. According to Dodson [8] Ca(HCOO)2 also accelerates setting of OPC. Heikal [ 34] found
that the initial and final setting times of both OPC and pozzolanic cement pastes are shortened
with the addition of Ca(HCOO)2.
Hewlett [35] has compared the calcium salts of chloride, nitrite and formate, and related their
accelerating efficiencies to the mobility of the anions in order to penetrate into the silicate
particles. Consequently, their size is of major importance. Taken this consideration into account,
Ca(HCOO)2 is less suitable with a relatively large anion size (0.45 nm compared to 0.34 nm for
nitrite and 0.27 nm for chloride) [35]. Heikal [34] claims that the accelerating effect of
Ca(HCOO)2 on calcium silicate hydrate formation is mainly attributed to the fact that the diffusion

2+

rate of HCOO ions is much higher than for the Ca ions, and that the HCOO ions can penetrate
into the hydrated layers covering C3S and -C2S grains. As a consequence, the precipitation of
Ca(OH)2 is accelerated as well as the decomposition of calcium silicates [34].
Justnes [20] found that Ca(HCOO)2 and calcium acetate (Ca(CH3COO)2) was less effective
accelerators than Ca(NO3)2 at equimolar concentrations of calcium ions, and suggests the reason
2+
to be possible complex formation of Ca with the accompanying organic anion, leading to a
2+
lowering of the Ca concentration in the paste fluid.
Ramachandran [36] reports that Ca(HCOO)2 accelerates the hydration of C3S, and one explanation

is that HCOO ions interfere with the protective layer normally formed on the surface of the
cement particles. Furthermore, the effect of Ca(HCOO)2 as an accelerator is influenced by

16

the C3A/SO3 ratio in the cement. According to Gebler [37] cements that are undersulfated
(C3A/SO3>4.0) provide the best potential for Ca(HCOO)2 to accelerate the early age strength of
concrete.
Rixom [9] claims that Ca(HCOO)2 accelerates the setting and hydration of all types of cement, but
at dosages beyond 2 % (by cement weight) especially with respect to the first 24 hours the
effect is not significant.
The biggest problem in using Ca(HCOO)2 as an accelerating admixture for concrete is its low
solubility in water (approx. 16g/100g H2O at room temperature). Therefore, it is not commercially
available in liquid form, and must be added to concrete in solid powdered form [9]. This can create
dispensing problems. Sodium formate (NaHCOO), which has also accelerating properties, is
approximately 3 times more soluble in water, but the potential problem of the added alkali at
relatively high dosages discourages its use [8].
Dicarboxylic (oxalic acid)
Dicarboxylic acids are acids in which there are two carboxyl groups. The simplest of these acids,
ethanedioic acid (IUPAC name) or oxalic acid (trivial name), consists simply of two carboxyl
groups (COOH) linked together. Although chemically related, the properties of oxalic acid are
very different than for ethanoic acid (acetic acid):
COOH

CH3

COOH

COOH

Oxalic acid (ethanedioic)

Acetic acid (ethanoic)

pKa1=1.27

pKa=4.76

From the pKa values it is seen that oxalic acid is a much stronger acid than acetic acid. The
inductive effect of one carboxyl group in oxalic acid is expected to enhance the acidity of the
other [38].
Oxalic acid is reported to decrease the setting time of OPC and increase early compressive
strength [36]. Taylor [39] found that calcium oxalate is precipitated when oxalic acid is added to
2+
C3S, and the concentration of Ca ions close to the surface of the C3S is thereby kept low until all
the oxalic acid has been used up.
Ramachandran [18] compared hydrated cement with and without oxalic acid using the DTA
technique and found the existence of calcium oxalate in the cement containing oxalic acid, and
suggested that much of the strength development may be due to the formation of calcium oxalate
during hydration.
In a fundamental study on the kinetics of C3S hydration, Odler [40] found that when adding oxalic
acid to the C3S/water system the initial hydration progresses with a fast rate as long as the
liberated calcium hydroxide is bound to oxalic acid and the liquid phase stays undersaturated with
respect to Ca(OH)2.
Singh et al [41] tested several dosages of oxalic acid in OPC and found that this acid acted as an
accelerator at dosages from 0.5 % to 4 % (by cement weight), but acted as an retarder if present
below 0.05 %. The accelerating effect is probably caused by the ability of carboxylate to form

17

ionic bonds with free lime and calcites of cement and form calcium oxalate [41]. When present in
small amounts in cement it seems that the retarding effect of oxalic acid is similar to that found for
acetic acid at low dosages [41].
Hydroxycarboxylic (glycolic and lactic acids)
Although most hydroxycarboxylic acids and their salts retard the setting of OPC [35], there are
examples of hydroxycarboxylic acids with accelerating properties in cement. This information is
typically disclosed in the patent literature [42, 43]. s early as 1964 Zach et al [42] found that the
two first members of the series of -hydroxycarboxylic acids (with a hydroxyl group adjacent to
the carboxyl group see Table 8.1 below) caused a significant increase in the late (28 days)
compressive strength of concrete when admixed at low dosages, typically from 0.01 % to 0.1 %
by cement weight. Neither the mechanism of action in cement paste, nor the effect on early
strength was discussed by Zach et al [42].
Table 8.1
Characteristics of two -carboxylic acids [44] reported to accelerate cement hydration
Trivial name

Other name

IUPAC name

Chemical formula

pKa

Glycolic acid

Hydroxyacetic acid

2-hydroxyathanoic acid

CH2(OH)COOH

3.83

Lactic acid

Milk acid

2-hydroxypropanoic acid

CH3CH(OH)COOH

3.85

Schutz [43] found that high dosages of sodium glycolate and sodium lactate (sodium salts of
glycolic and lactic acid respectively) shortened setting time and speeded early strength
development. Mechanisms causing the accelerating ability of -hydroxycarboxylic acids are
probably unknown, but may be linked to the ability of these acids to complex metals. For instance,
glycolic acid is used as a component of household cleaners due to its dissolution capabilities for
hard-water scale (calcium and magnesium) [45].

Alkanolamines

Alkanolamines are amines containing either one (primary amine NRH 2), two (secondary amine
NR2H) or three (tertiary amine NR3) hydroxy alkyl groups (R) . Showing accelerating effect on
OPC hydration, these organic chemicals are sometimes used as ingredients in admixture blends.
They are all weak bases.
Triethanolamine (TEA)
TEA, N(CH2CH2OH)3 , is both a tertiary amine and a tri-alcohol, and is probably the most
common alkanolamine for concrete purposes. TEA was introduced as early as 1934, when it was
interground as an addition with OPC in combination with calcium lignosulphonate [8]. This was
done to increase the early compressive strength of the concrete made from the cement. Since 1958
TEA has become a common ingredient of some water reducing admixtures to off-set possible set
retardation caused by the water reducer [8]. One of the most characteristic features of TEA is its
ability to either accelerate or retard cement hydration depending on dosage. TEA tends to

18

accelerate C3A hydration kinetics, but act as a retarder on the hydration of C3S [31, 35]. Rixom
[46] has summarised some of the features of TEA in OPC:
1 In the presence of TEA the reaction between C3A and gypsum is accelerated.
2 The subsequent conversion of the ettringite to monosulphate by reaction with C 3A is also
accelerated by TEA.
3 The formation of the hexagonal aluminate hydrate and conversion to the cubic form
is accelerated by TEA.
4 There may be a chelating effect whereby TEA reacts with the ferrite phase of OPC.
5 There is some evidence for the formation of a surface complex between C 2S and C3S
initial hydrates and TEA.
Hewlett [35] screened potential accelerators and found flash-set of OPC admixed with 0.5 mole
and 1.0 mole TEA per kg cement (i.e. approximately 0.75 % and 1.5 % by weight of cement
respectively). At lower dosages the effect of TEA on C 3A and C3S can be summarised like this
[35]:
1 At low dosages (0.1 0.5 % by weight of cement) TEA accelerates the hydration
of C3A in OPC.
2 At dosages higher than 0.5 % TEA causes retardation of C3S hydration.
Dodson [8] claimed that the most popular addition rate of TEA is in the range of 0.010 % to
0.025 % on the weight of cement. Figure 9.1 shows the effect of dosage.

Figure 9.1 The effect of dosage of TEA on setting and hardening of OPC concretes [8]. The
curves show the resistance to penetration as a function of time for four different

dosages of TEA and a reference without TEA.

19

Setting time (minutes)

The data in Figure 9.1 indicate that at a dosage of 0.02 %, TEA acts as an accelerator, at 0.25 % as
a retarder, at 0.5 % as a severe retarder and at 1.0 % as a flash-set accelerator. Similar behaviour
was found by Ramachandran [31] when measuring the initial and final setting characteristics of
OPC mortar treated with 0 0.5 % TEA (see Figure 9.2).
1
0
0
0
0

Final
1
0
0
0

100

10

1
0,01

Initial

0,1

Dosage (% by weight of cement)

wfound
noaccelerating effects). However, his
h significant effectresults indicate that DEA, at proper
Figure 9.2 Effect of dosage of TEA oni on the initialdosages, could be an interesting
initial and final setting of OPC mortar t setting times atpotential accelerator, and even better
e low dosages, butthan TEA. In that respect it is
(Data from [31]).
significant
interesting to note that several alkaliP retardation
atfree shotcrete accelerators contain
Although it has been shown that TEAo higher dosagesDEA (see Chapters 11 and 13). Another
r for all threerelated amine, methyldiethanolamine
decreases the strength of OPC, it actst types. Besides,
differently with blast furnace cements.l both MEA and(MDEA), N(CH3)(CH2CH2OH)2, is
used in commercial accelerators for
It has been found that TEA (in a mortar a TEA lowered thenormal concrete (see Chapter 12).
containing 30 % OPC and 70 % slag) n compressive
accelerated the hydration kinetics byd strength after 3
whileTriisopropanolamine (TIPA)
increasing the rate of reaction of days,
c cement
pastes
gypsum with slag in the presence of
e containing DEAIn 1991 TIPA was suggested as a
lime [31].
mwere close to thatstrength enhancer for OPC [27]. Later,
e of the controlin 1997, Ichikawa [48] studied the
Heren [47] was the first to publish datan sample
[47].effect of 0.01 % TIPA (by weight of
(in 1995) on the effect oft Unfortunately, cement) in cements with different
monoethanolamine
(MEA), Heren did notcompositions of clinker minerals, and
NH2(CH2CH2OH), and diethanolaminea publish resultshe found that the strength enhancement
of cement mortar by TIPA depends on
(DEA), NH(CH2CH2OH)2, on cementn on lower dosages
the quantity of ferrite phase (C4AF).
hydration. He compared MEA, DEAd (potential
forNo effect was obtained on the cement
and TEA at dosages 0.1 to 1.0 % in dosages
prepared

20

from the clinker with only 1 % C4AF. Sandberg [49] carried out similar tests on 10 different OPCs
all containing normal distributions of clinker minerals and found approximately 10 %
increase in 28 days strength for all samples containing 0.02 % TIPA (by weight of cement).
Aggoun [50] tested the effect of the admixture combinations TIPA / TEA / calcium nitrate on
setting time and early strength in OPC, and found that the combination amine / calcium nitrate
produced significant and promising results with respect to both setting and hardening acceleration.
Moreover, regardless of the cement type used, Aggoun [50] found that the performance of TIPA as
a hardening accelerator is greater than that of TEA.
TIPA has also been evaluated as a possible grinding aid component for cement production (a
common role of TEA today) [51].

10 Antifreeze admixtures
Winter concreting practice has remained almost unchanged since the 1930s [52]. The concrete
ingredients must be heated, the substrate must be thawed, and the concrete itself has to be
thermally protected until it has cured. An alternative approach to cold weather concreting is to use
antifreeze admixtures. This is not yet fully recognized by standard practice, except in Russia
where the use of antifreeze admixtures (often containing chloride salts) has been acceptable
practice for nearly five decades [53, 54]. In the last two decades antifreeze admixtures (without
chlorides) have been tested in North America, Scandinavia, Finland, China and Japan [7, 17, 54].
Antifreeze admixtures are sometimes considered a special class belonging to the general category
of accelerating admixtures. One of their mode of actions is to affect the physical condition of the
mixing water, i.e. being capable of depressing the freezing point of water in the fresh concrete, but
the main action is to accelerate the hydration of cement like conventional accelerators [55]. The
principal objective of using such admixtures is twofold [8]:
(1) To give the OPC in the concrete an opportunity to hydrate to a point that if the remaining water
in the pores freezes, it has sufficient volume of empty pores into which it can expand.
(2) To give the concrete sufficient strength to resist damage by the expansion of the freezing
water.
It should be noticed that, in order to obtain a significant lowering of the freezing point, one has to
dissolve a considerable quantity of solute in the solvent. Dodson [8] calculated that an addition of
2 % anhydrous CaCl2 (by the weight of cement) to a typical concrete mix design would lower the
freezing point of the remaining water assuming that 30 % of the originally added water was
o
consumed during the early hydration by less than 1 C. Consequently, larger dosages must be
used compared to conventional admixtures in order to obtain significant freezing point depression.
For example, 8 % NaNO2 (by weight of cement) has been used to keep water in the liquid state in
o
concrete at a temperature of 15 C [7].
Korhonen [56] screened a series of inorganic and organic chemicals in OPC mortars in the
o
o
temperature range 0 C to 20 C, and found the following relation between the freezing point (FP)

of the water and the molality (M) of the chemical solved in the water:

21

FP( C) = 1.76M 1.28

(10.1)

This relation is independent of type of chemical used, provided the chemical remains soluble at
low temperatures. (Molality is defined as number of moles of particles (ions, molecules) dissolved
into 1 kg of water). Equation 10.1 is a good first estimate of expected freezing point whenever the
chemical composition of the admixture is known.
Generally, antifreeze admixtures utilize both freezing point depressants and accelerators [8, 54,
55, 56, 57]:
Freezing point depressant (may act as weak accelerator or even retarder)
1 Sodium nitrite, NaNO2
2 Sodium chloride, NaCl
3 Weak electrolytes
4 Non-electrolytic organic compounds, such as high molecular weight alcohols,
and urea (carbamide, (NH2)2CO)
Accelerator (acts also as freezing point depressant)
1 Binary and ternary admixture systems that contain potash, K2CO3
2 Additives based on calcium chloride (CaCl 2) combined with sodium chloride (NaCl),
sodium nitrite (NaNO2), calcium nitrate (Ca(NO3)2) and urea (carbamide, (NH2)2CO)
Generally, for concrete cured at low temperatures higher strengths are obtained by binary
admixtures at lower dosages than when the individual components are used singly [7]. Good
results have been obtained with the combination of K 2 CO3 and a retarder [7]. Korhonen [57]
evaluated a number of potential chemical compounds and their ability to promote concrete
o
strength at 5 C, and concluded that the best antifreeze admixture formulation was 3 parts of
NaNO3 and 1 part of Na2SO4, and that the recommended dosages varied from 6 to 8 % by weight
of cement. However, a major disadvantage of this admixture is that it adds a lot of alkalis to the
concrete mix.

Korhonen [57] also reported results from an evaluation of the commercial accelerator PolarSet ,
intended for concreting under very cold conditions. This admixture is a blend of calcium bromide,
calcium nitrate, calcium nitrite, diethylene glycol and methyl-diethanolamine (see Table 12.1).

The technical data sheet of PolarSet states [58]:

PolarSet is specially formulated to reduce concrete setting times and increase early strengths for
concrete in very cold conditions, and may be used to reduce the time that concrete must be
protected against freezing in ambient temperatures as low as 7C (20F).

The results of Korhonen [57] showed that PolarSet did perform well at temperatures above 0 C,
but its strength performance at sub-zero temperatures did not meet the preset acceptance criteria
o
(even at 8 % dosage): An admixture that would promote concrete strength at 5 C to meet or
o
exceed the strength of admixture-free concrete at +5 C.

22

11 Shotcrete accelerators
In modern shotcreting a liquid accelerator is added as a separate component at the nozzle during
spraying, providing rapid stiffening of the concrete, final setting within few minutes, and high
early strength [59, 60, 61]. Accelerator dosages are higher than those for normal concrete,
typically in the range of 6 to 8 % admixture by weight of cement, or even higher. The active dry
matter contents of commercial liquid shotcrete accelerators are typically in the range of 40 to 50
1%
(see Table 13.1). Shotcrete accelerators can be divided into two main groups:
1 Alkaline accelerators (conventional high pH)
2 Alkali-free accelerators (modern low pH )
A conventional shotcrete accelerator is normally a one-component material, i.e. an alkaline
compound dissolved in water as the sole active ingredient, while modern alkali-free accelerators
are blends of several ingredients dissolved in water.
Conventional alkaline shotcrete accelerators (alkali salts dissolved in water) [60]
1 Sodium silicate (water glass, nNa2OSiO2, n3.3)
2 Alkali aluminates (NaAl(OH)4, KAl(OH)4)
3 Alkali carbonates and hydroxides (Na2CO3, K2CO3 and NaOH, KOH)
Alkali carbonates and hydroxides mainly accelerate C 3S hydration, while sodium aluminates react
with gypsum in the cement, thus preventing ettringite formation around the cement grains,
allowing almost instantaneous reaction of the C3A [60]. Water soluble sodium silicate (water
glass) provides rapid setting by reaction with calcium in the cement to form insoluble calcium
silicates [60, 61].
Among the conventional accelerators, only water glass is used in todays Norwegian shotcreting.
In modern shotcrete technology, particularly in Northern parts of Europe, alkali aluminates,
carbonates and hydroxides are considered problematic due to:
1 Highly caustic materials that require caution in handling
2 Addition of a lot of alkalis and the risk of promotion of alkali-aggregate reaction
3 Decrease in final strength
Water glass, the classical shotcrete accelerator used for decades and still having a significant
market share, is characterized by [60]:
1 Compatible with all types of cement
2 Less caustic than aluminates (pH lower than that of fresh concrete)
3 Low early strengths
4 Decrease in the E-modulus with time
5 Reduced final strength at high dosages
6 Decrease in the water proofing characteristics due to the extraction of lime when the
concrete is subjected to a continuous exposure to moisture
7 High viscosity at low temperature which requires heating of the material during

shotcrete operations

23

For these reasons the use of water glass as shotcrete accelerator has been restricted in Germany
and Austria [60].
Alkali-free shotcrete accelerators (weakly acidic solutions of aluminum salts) [61, 62]
Alkali-free accelerators were originally developed to solve some of the problems caused by the
alkaline accelerators. However, it seems that the large market share of alkali-free accelerators in
Scandinavia today is mainly caused by the capability of these accelerators to give very high early
strengths, rather than their lack of alkali problems.
The chemical compositions of alkali-free accelerators are closely guarded trade secrets and
therefore little published information is available. Normally, the main ingredient is aluminum
sulphate [61, 62]. Admixtures containing large amounts of aluminum salts and complexes are
3+
weakly acidic (pH=2-3) due to the hydrolysis caused by aluminum ions, Al , given by the
equilibrium [61]:
3+

Al

+ H2O = Al(OH)

2+

+H

(11.1)

Aluminum sulphate, Al2(SO3)3 nH2 O (n15-18), is a low cost bulk chemical with high
solubility. At high dosages aluminum sulphate causes flash setting of OPC [35]. Brge [62]
investigated the mechanism of action of aluminum sulphate in cement and found that aluminum
sulphate reacts with lime and produces ettringite and aluminum hydroxide according to:
Al2(SO3)3 + 6 Ca(OH)2 + 32 H2O = 3CaO Al2O3 3CaSO4 32H2O + Al(OH)3

(11.2)

Ettringite is responsible for the rapid setting [62, 63]. As a by-product of the reaction, Al(OH) 3 is
formed which is also known to be an accelerator [62].
Maltese [64, 65] investigated to what extent type of calcium sulphate setting regulator in OPC
(gypsum, -hemihydrate, -hemihydrate or anhydrite) affected the setting behavior of shotcrete
accelerated by an inorganic acid based alkali-free accelerator containing aluminum sulphate. He
found that the lower the instantaneous dissolution rate of the calcium sulphate, the more efficient
the accelerator (see Table 11.1).

Table 11.1
Alkali-free accelerator and type of setting regulator in OPC (data from [64])

-hemihydrate

Dissolution rate of
Final setting time of OPC paste
setting regulator
Without accelerator With accelerator
(Ca 2+
ions, g/l min)
92
5 h 30 min
8 h 30 min

-hemihydrate
Gypsum
Anhydrite

60
48
7

Type of setting
regulator in OPC

5h
5 h 30 min
5 h 30 min

6 min 30 s
3 min 5 s
1 min 18 s

24

Ferric sulphate, TEA and sodium fluoride are also reported as ingredients in shotcrete accelerators
[35]. According to patent literature alkali-free shotcrete accelerators may contain ingredients like:
1 Amine and glycolic acid [66]
2 Hydrofluoric acid, diethanolamine, methyldiethanolamine and ethylenediamine
tetraacetic acid (complex binder) [67]
3 Magnesium sulphate, phosphorous acid, oxalic acid, maleic anhydride and diethanolamine
[68]
4 Citric and malic acids and their alkali metal salts [69]. (Malic acid should not be confused
with maleic or malonic acid, both of which are different types of dicarboxylic acids.)
5 Al(OH)xRy, where x+y=3, R=organic acid (formic, acetic, 4-toluenesulphonic, lactic
or oxalic) or nitric acid [70]
Some information on ingredients in commercial shotcrete accelerators is disclosed in their Safety
Materials Data Sheets provided by the producers (see Chapter 13).
In order to gain very high early strengths in shotcrete Myrdal [71] tried to combine a nozzle added
alkali-free accelerator with a dormant accelerator that was added in the concrete during mixing.
The idea was that the dormant should not react with the cement until activated or ignited by
the alkali-free accelerator. A doubling of early compressive strengths was obtained using a
proprietary dormant blend of water soluble carbonate, alkanolamine, hydroxycarboxylate and
phosphonate [71].
It is generally believed that the high content of aluminum sulphate in alkali-free accelerators may
promote an internal form of sulphate attack in the concrete, but so far no conclusive results have
been found concerning the durability of field shotcrete [72, 73].

25

12 Commercial accelerators for normal concrete

Suppliers of accelerators seldom disclose the ingredients of their accelerators, neither in their open
technical documents nor in their responses to direct inquiries. Occasionally, the Technical Data
Sheet (TDS) of the product gives some information on the main chemical ingredient.
Due to stricter safety regulations over the last years more information is often found in the
Material Safety Data Sheet (MSDS) of the product. One should notice that information given in a
MSDS can vary from country to country for the very same product. This might be a result of
specific national regulations or national safety practices. For instance, more detailed information
on chemical ingredients can sometimes be found in Swedish MSDSs compared to MSDSs from
continental Europe for the same chemical admixture. This unfortunate practice should be stopped
by proper authorities so that users of admixtures can receive identical information regardless of
national belonging.
Table 12.1 shows information on ingredients in a selection of well known commercial accelerators
in Europe and North America. As the information is based on MSDSs only (as of November
2007), one should bear in mind that additional ingredients may be present in the products, e.g.
ingredients not disclosed either due to lack of demand in regulations for that particular chemical,
or other reasons. Obviously, an accelerator of which the MSDS states content of ingredients in the
range of 10 % must contain additional ingredients.
The information given in Table 12.1 shows that:
1 Most commercial accelerators for normal concrete consist of calcium nitrate and/or sodium
thiocyanate, either as sole component or in combination
2 Nitrite is not a common ingredient, and used mainly in North America
3 It is common to add small amounts of amine (typically alkanolamine)
4 The only carboxylic acid found was lactic acid
5 No information on calcium formate in MSDSs for liquid accelerators has been found
(Calcium formate is sold in powder form on the American market [74])

13 Commercial shotcrete accelerators

As for accelerators for normal concrete, information on ingredients of shotcrete accelerators can
be found in MSDSs. Table 13.1 on next page gives an overview of well known commercial alkalifree shotcrete accelerators on the market in Europe and North America. Some characteristics can
be found:
1 All contain aluminum salts
2 Almost all (probably all) contain aluminum sulphate
3 Many contain diethanolamine
4 Some contain fluorides
5 Few contain organic acids (oxalic, formic) and/or inorganic acid (phosphorus)

26

Catexol 10

Table 12.1
A selection
ofCatexol 20
commercial
liquid nonchloride
*)
accelerators
for
M concrete
S
D
S
=
M
at
er
ial
Sa
fe
ty
D
at
a
S
he
et
**)

N
o
=
N
o
in
fo
r
m
ati
on
gi
ve
n
in
th
e
M
S
D
S

27

Table 13.1
A selection of commercial liquid alkali-free accelerators for shotcrete
*)

Name of product

Information on ingredients given in MSDS

Type of ingredient
% by weight Manufacturer
Aluminum sulphate

30 45

Diethanolamine

1 4

Aluminum sulphates/fluorides

15 30

Diethanolamine

1 5

Oxalic acid di-hydrate

16

Meyco SA 164 [87]

Meyco SA 170 [88]

Aluminum sulphates/fluorides
Aluminum sulphate, hexadecahydrate
Aluminum sulphate

20 40
25 50
30 50

Meyco SA 172 [89]

Formic acid

0.5 2.5

Meyco SA 170
Aluminum sulphate

30 50
20 30

Maleic anhydride

1 4.5

Oxalic acid di-hydrate

1 4.5

Phosphorous acid
Aluminum sulphate

1 4.5
20 30

Diethanolamine

1 4.5

Phosphorous acid

1 5

Aluminum sulphate
Aluminum salts
Aluminum salts

No**)
No
No

Aluminum sulphate

35 50

Formic acid

2 10

2,2-iminodiethanol (diethanolamine)

1 3

Glycerin (propane-1,2,3-triol)
Aluminum sulphate

15
20 25

Aluminum fluoride

5 9.99

Aluminum sulphate

25 34.99

Manganese sulphate

0.5 0.99

Aluminum sulphate

12.5 19.99

Inorganic fluoride

2.5 2.99

Manganese sulphate

1 2.49

Aluminum sulphate (anhydrous)

> 60.0

Diethanolamine

3.0 7.0

Meyco SA 160 [85]

Meyco SA 163 [86]

Meyco SA 175 [90]

Meyco SA 180 [91]

SigunitL 50 AF [92]
SigunitL 53 AF S [92]
SigunitL 5450 AF (N) [93]

SiguniteL 72 AF VP [92]

Mapequick AF 2000 [94]

Mapequick AF D 02 [94]

Mapequick AF 400 [94]

Sure-Shot AF [95]

*)

MSDS = Material Safety Data Sheet

BASF

Sika

Mapei

Euclid Chemical

**)

No = No information given in the MSDS

28

14 Accelerators for calcium aluminate cements

Many inorganic salts affect the setting of calcium aluminate cements (CACs), and are effective in
concentrations of 0.1 to 0.5 % by weight of cement [96, 97]. The effect on setting times differ
greatly both in direction and in relative magnitudes from those observed with Portland cements.
The setting times of CAC caused by various admixed ions are reported to increase in the
following order [96]:
+

2+

2+

2+

Cations:

Li Na < control (without accel.) < K Ca

Anions:

OH control (without accel.) < Cl < NO3 < Br < CH3COO (acetate)

< Mg

< Sr

< NH4

This means that LiOH is an effective accelerator, while Ca(NO 3) 2 tend to retard setting of CAC.
Generally, lithium salts like Li2CO3, LiOH, LiCl and LiNO3 are most efficient and widely used
accelerators for CAC [96, 98, 99].
The accelerating effect of lithium salts has been explained by the precipitation of lithium
aluminate hydrate which acts as a heterogeneous nucleation substrate [96].
Alkali-earth metal salts of chloride, nitrate and formate may act as weak accelerators or retarders
depending on concentration and temperature [96, 100].
It seems that the chemical mechanisms of action of these admixtures are not well understood.

15 Recommendations for future R&D

Future R&D should aim at developing more efficient non-chloride accelerators for concrete, and it
is considered important to focus the efforts on:
1 Chemicals that speed up early hardening of OPC blended with pozzolanic materials
2 Technical grade bulk chemicals
3 Safety (chemicals friendly to humans and natural environment)
4 Low alkali content
5 Synergistic effects of chemical blends
Acceleration of early hardening
There is a well pronounced problem that todays non-chloride hardening accelerators do not
perform well enough, especially in slow hardening blended cements. There is a lack of
accelerators that efficiently speed up the strength development during the first 24 hours. It seems
that research on setting accelerators is not that important. Todays most common setting
accelerator based on technical grade Ca(NO3)2 fulfils the requirements in the market.
Technical grade bulk chemicals
From a commercial point of view it is important that the main chemical in a formulated admixture
is easily available at acceptable low cost. The use of more expensive and specific chemicals, as
minor parts of admixture formulations, is possible as long as the total material cost of the chemical
formulation is kept low. Total material cost higher than roughly EUR/kg seems unrealistic.

29

Safety
The development of future admixtures must take both human and environmental safety into
consideration. This means that caustic, as well as highly acidic, admixtures are out, e.g. shotcrete
accelerators based on alkali aluminates and acidic alkali-free accelerators with very low pH should
be avoided. For normal concrete, both nitrite and thiocyanate based accelerators on the market
today are (or should be) labelled Xn, Harmful . The worst labelling of a potential admixture for
the future would probably be Xi, Irritant. Other labels describing risk to humans and nature should
be avoided:
Xi

Xn

Irritant

Harmful

Toxic

Corrosive

Dangerous for environment

Acceptable Unacceptable

Low alkali content

The term alkali-free is already well established for shotcrete accelerators. It would be extremely
difficult to enter the European market today with a new non-aluminate, non-silicate or noncarbonate shotcrete accelerator containing alkali metals above 1 % Na 2O-equivalents. Attempts at
developing alkali-free hardening accelerators for normal concrete should also be aimed at,
especially if the accelerator is to be used at high dosages, as for antifreeze admixtures.
Synergistic effects
A lot of potential chemicals have already been tested as accelerators for concrete. Future R&D
efforts should focus on, in a systematic way, synergistic effects occurring when several chemicals
are mixed. For instance, a synergistic effect is reported with a mixture of TEA and calcium
formate [27].
Potential accelerating chemicals and their combinations
Although a harmful material labelled Xn, thiocyanate should not be abandoned completely.
Thiocyanate is a well established accelerator in Europe, and should be investigated further, but as
minor component in formulations together with nitrate, alkanolamines and/or hydroxycarboxylic
acids and their salts. Alkali metal thiocyanate, like todays sodium thiocyanate, should be avoided.
Instead, soluble alkali-free thiocyanates can be made by neutralizing thiocyanic acid with a nonalkaline base, for instance Ca(OH)2 or alkanolamine:
NH(CH2CH2OH)2
Diethanolamine

HSCN

[NH2(CH2CH2OH)2] SCN

(15.1)

Thiocyanic acid

In this way an alkali-free accelerator can be formulated using ingredients that separately show
accelerating effects as well. Possible synergistic effects of such combinations have not been
published earlier.
Similar approach can be utilized for developing soluble non-alkali salts of carboxylic acids, e.g.
formic and oxalic acids. The effect of non-calcium formate on the hydration of OPC with different
C3A/SO3 ratios should be compared to the C 3A/SO3 sensitivity reported for calcium formate (see

Chapter 8).

30

A parallel R&D activity should be a screening of alkanolamines. From literature it seems that
DEA and TIPA are the most promising, while other amines can be found in MSDSs of commercial
accelerators (see Table 12.1).
A systematic investigation of the effects of carboxylic acids (and their salts) on OPC hydration is
highly appreciated, especially the properties of -hydroxycarboxylic acids and their salts.
Further development of antifreeze admixtures should focus on combinations of chemicals already
tested and reported. An ideal antifreeze admixture would consist of a temperature depressant
(major part) and a non-chloride accelerator (minor part). It is important to find temperature
depressants with low alkali metal content. Technical grade calcium nitrate is a promising
temperature depressant. At high dosages, however, short setting times may require the addition of
a retarder in the admixture. The hardening accelerator part can be found among the normal nonchloride accelerators. However, an accelerator to be combined with a temperature depressant
2+
based on a calcium salt cannot form sparingly soluble salts with Ca . This solubility requirement
excludes many carboxylic acids, e.g. oxalic acid and to a certain degree formic acid.
If better alkali-free shotcrete accelerators are to be developed, one should investigate the effects of
phosphorous and carboxylic acids in commercial shotcrete accelerators (see Table 13.1). As for
accelerators for normal concrete synergistic effects may occur and should be investigated.
Due to a possible risk of internal sulphate attack caused by the latge amounts of aluminum
sulphate in most alkali-free accelerators for shotcrete, an ideal shotcrete accelerator would be both
alkali-free and low-sulphate. Possible alternatives to aluminum sulphate could be aluminum salts
of inorganic acids (e.g. nitric and phosphorous) and carboxylic acids (e.g. formic).
A major challenge linked to the development of a low-sulphate aluminum salt based accelerator
for shotcrete would be to dissolve technical grade aluminum hydroxides in acids by cost-effective
means.
The effect of water soluble silicon complexes on hydration of OPC is another intriguing
possibility that might be considered.
As part of the research and documentation of any new accelerator, possible leaching of the
chemical compound(s) from hardened concrete should be investigated.

16 Conclusions
The best all-round accelerator to date has been calcium chloride. Due to restrictive regulations
prohibiting the use of chlorides in steel reinforced concrete, there has been a continuous search for
an alternative, one without its limitations.
In Europe calcium nitrate has replaced calcium chloride as a setting accelerator, while sodium
thiocyanate has replaced calcium chloride as hardening accelerator. In North America calcium
chloride is partly replaced by calcium and sodium nitrite and calcium formate (in powder form).
Alkali shotcrete accelerators have partly replaced the conventional alkali silicates and aluminates,
particularly in Northern part of Europe.

31

The mechanisms of action of the new non-chloride accelerators are not fully understood. The
synergistic effect of combinations of chemicals on hydration rates is a relatively new research
field, and few results have been published in open reports. There is a trend that different
carboxylic acids and their salts are used in commercial accelerators. Near future research will
probably focus on synergistic effects of chemicals already tested as sole admixtures.

17 References
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16

17
18
19

EN 934-2:2001, Admixtures for concrete, mortar and grout - Part 2: Concrete admixtures
Definitions, requirements, conformity, marking and labelling, Brussels, Belgium, 2001.
nd
Popovics S, Concrete Materials Properties, Specifications and Testing, 2 edition,
Chapter 6, William Andrew Publishing, Noyes, 1992, pp 236-246.
European Federation of Concrete Admixture Associations, EFCA Environmental
Declaration, EFCA doc.300 ETG, January 2006, http://www.efca.info/publications.html
Statistical data provided by the Norwegian Committee for Concrete Admixtures (NCCA),
2007.
The American Concrete Institute, Committee Report ACI 212.3R-04, Chemical Admixtures
for Concrete, 2004.
Taylor H F W, Cement Chemistry, Chapter 11, Academic Press Limited, London, 1990, pp
358-359.
Rixom R and Mailvaganam N, Chemical Admixtures for Concrete, Chapter 6, E&FN Spon,
London, 1999, pp 208-211.
Dodson V, Concrete Admixtures, Chapter 4, Van Nostrand Reinhold, New York, 1990, pp
73-102.
See ref. 7, Chapter 7, pp 312-313.
See ref. 7, Chapter 5, pp 182-187.
Tenoutasse N, The Hydration Mechanism of C3A and C3S in the Presence of Calcium
th
Chloride and Calcium Sulphate, Proceedings 5 International Symposium on the Chemistry
of Cement, Cement Association of Japan, Tokyo, Vol 2, 1969, pp 372-378.
Nilforoushan M R and Sharp J H, The effect of additions of alkaline-earth metal
chlorides on the setting behaviour of a refractory calcium aluminate cement, Cement and
Concrete Research, 25, 1995, pp 1523-1534.
Singh N B et al, Effect of lignosulfonate, calcium chloride and their mixture on the
hydration of RHA-blended Portland cement, Cement and Concrete Research, 32, 2002, pp
387-392.
Collepardi M et al, The Influence of Calcium Chloride on the Properties of Cement Pastes,
Il Cemento, 2, 1973, pp 83-92.
Shi C and Day R L, Pozzolanic reaction in the presence of chemical activators. Part 1.
Reaction kinetics, Cement and Concrete Research, 30, 2000, pp 51-58.
Juenger M C G et al, A soft X-ray microscope investigation into the effects of calcium
chloride on tricalcium silicate hydration, Cement and Concrete Research, 35, 2005, pp 1925.
th
Collepardi M, The World of Chemical Admixtures in Concrete, Proceedings 18 Conference on
Our World in Concrete & Structures, Singapore, 25-27 August 1993, pp 63-71.
Ramachandran V S, Elucidation of the role of chemical admixtures in hydrating cements by
DTA technique, Thermochimica Acta, 4, 1972, pp 343-366.
El-Didamony et al, Hydration characteristics of -C2S in the presence of some accelerators,
Cement and Concrete Research, 26, 1996, pp 1179-1187.

32

20
21

22
23
24
25
26

27
28
29
30
31
32

33
34
35
36
37
38
39
40
41
42

Justnes H, Kalcijev nitrat kot mnogo-funkcianalni dodatek betonu (Calcium Nitrate as

th
Multifunctional Concrete Admixture in Slovenian), 14 Slovenian colloquium on Concrete,
th
Ljubljana, Slovenia, 29 May, 2007, pp. 21-28.
Justnes H and Nygaard E C, Setting Accelerator Calcium Nitrate Fundamentals,
rd
Performance and Applications, Proceedings 3 CANMET/ACI International
Conference Advances in Concrete Technology, Auckland, New Zealand, ACI
International SP-171 (Edited by V M Malhotra), 1997, 325-338.
See ref. 7, Chapter 5, pp 178-181.
th
Hewlett P C (ed), Leas Chemistry of Cement and Concrete, Chapter 15, 4 Edition, Arnold,
London, 1998, pp 877-883.
Ramachandran V S, Concrete Admixtures Handbook. Properties, Science and Technology,
nd
Chapter 3, 2 Edition, Noyes Publications, USA, 1995, pp 860-862.
Wise T et al, The effect of thiocyanate on the hydration of Portland cement at low
temperatures, Thermochimica Acta, 264, 1995, pp 157-171.
Schfer S, Sodium Thiocyanate. The alternative to non chloride accelerator, Supplementary
Papers Eight CANMET/ACI International Conference on Superplasticizers and Other
Chemical Admixtures in Concrete, Sorrento, Italy (Compiled by M Venturino), Oct 29
Nov 1, 2006, pp 381-387.
See ref. 24, Chapter 17, pp 1045-1049.
http://www.mapei.it/Referenze/Multimedia/Antifreeze%20AK_eng.pdf
Abdelrazig B E I et al, Estimation of the degree of hydration in modified ordinary Portland
cement pastes by differential scanning calorimetry, Thermochimica Acta, 145, 1989, pp
203-217.
Nmai C K and Corbo J M, Sodium Thiocyanate and the Corrosion Potential of Steel in
Concrete and Mortar, ACI Course Manual SCM-23 (90) How to Effectively Use the
Newest Admixtures, ACI, Michigan, USA, 1990, pp 67-75.
See ref 24, Chapter 5, pp 185-285.
Schwarz W, Modified microstructure via chemical activation of ferrite and aluminate
phases in Portland cement-based formulations, The Modelling of Microstructure and Its
Potential for Studying Transport Properties and Durability (Edited by H Jennings et al),
Kluwer Academic Publishers, The Netherlands, 1996, pp 373-398.
Pacanovsky J T et al, Hydration control of cementitious systems, United States Patent No
5,634,972, June 3, 1997.
Heikal M, Effect of calcium formate as an accelerator on the physiochemical and
mechanical properties of pozzolanic cement pastes, Cement and Concrete Research, 34,
2004, pp 1051-1056.
Hewlett P C, Leas Chemistry of Cement and Concrete, Fourth Edition, Arnold, London,
1998, 837-901.
See ref 24, Chapter 3, pp 96-102.
Gebler S, Evaluation of Calcium Formate and Sodium Formate as Accelerating admixtures
for Portland Cement Concrete, ACI Journal, Proceeding V, Vol 80, No 5, 1983, pp 439-444.
Roberts J D and Caserio M C, Basic Principles of Organic Chemistry, Chapter 18, W A
Benjamin Inc, USA, 1977, pp 791-849.
See ref 6, Chapter 11, pp 350-351.
Odler I and Drr, Early hydration of tricalcium silicate. II. The induction period, Cement
and Concrete Research, 9, 1979, pp 277-284.
Singh N K et al, Effects of hydroxyethyl cellulose and oxalic acid on the properties of
cement, Cement and Concrete Research, 33, 2003, pp 1319-1329.
Zach S et al (du Pont), Monohydroxy Organic Acid Compounds for Improving the
Properties of Cement, Concrete and Mortars, United States Patent No 3,144,347, August
11, 1964.

33

43
44
45
46
47
48

49
50
51
52
53

54
55
56
57
58
59
60
61
62

Schutz R J (Sika Chemical Corporation), Admixtures and Method for Accelerating the
Setting of Portland Cement Compositons, United States Patent No 4,264,367, April 28,
1981.
Wikipedia, http://en.wikipedia.org/wiki/Glycolic_acid and
http://en.wikipedia.org/wiki/Lactic_acid
DuPont, Glycolic Acid. Properties, Uses, Storage and Handling, DuPont Speciality
Chemicals, USA, 1998.
See ref. 7, Chapter 5, pp 166-177.
Heren Z and lmez H, The Influence of Ethanolamines on the Hydration and Mechanical
Strength Properties of Portland Cement, Cement and Concrete Research, 26, 1996, pp 701705.
Ichikawa M et al, Effect of Triisopropanolamine on Hydration and Strength Development of
th
Cements with Different Character, Proceedings 10 International Congress on the
Chemistry of Cement, Gothenburg (Edited by H Justnes), Sweden, June 2-6, 1997, paper no.
3iii005, 8 pp.
Sandberg P J and Doncaster F, On the Mechanism of Strength Enhancement of Cement
Paste and Mortar with Triisopropanolamine, Cement and Concrete Research, 34, 2004, pp
973-976.
Aggoun S et al, Effect of some admixtures on the setting time and strength evolution of
cement pastes at early ages, Construction and Building Materials, 22, 2008, pp 106-110.
Jeknavorian A A et al, Determination of Grinding Aids in Portland Cement by Pyrolysis Gas
Chromatography-Mass Spectrometry, Cement and Concrete Research, 28, 1998, pp 13351345.
Korhonen C J, Off-the-Shelf Antifreeze Admixtures, Technical Report ERDC/CRREL TR-0207, US Army Corps of Engineers, Engineer Research and development Center, National
Technical Information Service, USA, April 2002, 38 pp.
Farrington S A and Christensen B J, The Use of Chemical Admixtures to Facilitate
Placement of Concrete at Freezing Temperatures, Proceedings from Seventh
CANMET/ACI International Conference on Superplasticizers and Other Chemical
Admixtures in Concrete (Edited by V M Malhotra), ACI International SP-217-5, USA,
2003, pp 71-85.
Korhonen C J, Antifreeze Admixtures for Cold Regions Concreting. A Literature Review,
Special Report 90-32, US Army Corps of Engineers, Engineer Research and development
Center, National Technical Information Service, USA, September 1990, 19 pp.
See ref 24, Chapter 11, pp 740-799.
Korhonen C J, Expedient Low-Temperature Concrete Admixtures for the Army, Special
Report 99-17, US Army Corps of Engineers, Engineer Research and development Center,
National Technical Information Service, USA, November 1999, 20 pp.
Korhonen C J et al, Antifreeze Admixtures for Concrete, Special Report 97-26, US Army
Corps of Engineers, Engineer Research and development Center, National Technical
Information Service, USA, October 1997, 22 pp.
http://www.na.graceconstruction.com/product.cfm?mode=c&id=1&did=1
See ref. 24, Chapter 15, pp 1009-1017.
See ref. 7, Chapter 6, pp 252-265.
Myrdal R, Modern Chemical Admixtures for Shotcrete, Proceedings from Third International
Symposium on Sprayed Concrete (Edited by N Barton et al), Gol, Norway, 26-29
September, Norwegian Concrete Association, 1999, pp 373-382.
Brge T A, Mode of action of alkali-free sprayed shotcrete accelerators, Shotcrete:
Engineering Developments, Proc. International Conference on Engineering Developments
in Shotcrete (Edited by E S Bernard), Hobart, Tasmania, April 2-4, 2001, Balkema
Publishers, The Netherlands, 2001, pp 79-85.

34

63
64
65
66
67
68
69
70
71

72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89

Paglia C et al, The influence of alkali-free and alkaline shotcrete accelerators within cement
systems. I. Characterization of the setting behavior, Cement and Concrete Research, 31,
2001, pp 913-918.
Maltese C et al, Effects of setting regulators on the efficiency of an inorganic acid based
alkali-free accelerator reacting with a Portland cement, Cement and Concrete Research, 37,
2007, pp 528-536.
Maltese C et al, A case history: Effect of moisture on the setting behaviour of a Portland
cement reacting with an alkali-free accelerator, Cement and Concrete Research, 37, 2007,
pp 856-865.
Angelskaar T (MBT Holding AG), Cement Accelerator and Method, International PCT
Patent No WO 03/045872 A1, June 5, 2003.
Sommer M, (Sika AG), Alkalifreier Abbinde- und Erhrtungsbeschleuniger, European
Patent No EP 1 167 317 A1, January 2, 2000.
Angelskaar T (Construction Research & Technology GmbH), Admixture, International PCT
Patent No WO 2004/106258 A2, December 9, 2004.
Farrington S A et al (Sandoz-Patent-GmbH), Rasche abbindende Zementzubereitungen,
German Patent DE 195 34 929 A1, March 21, 1996.
Valenti S (Sandoz Ltd), Aluminium Accelerator for Cement, UK Patent No GB 2 307 475 A,
November 19, 1996.
Myrdal R et al, A New Chemical Approach to Obtain High Early Strength in Shotcrete,
Shotcrete: Engineering Developments, Proc. International Conference on Engineering
Developments in Shotcrete (Edited by E S Bernard), Hobart, Tasmania, April 2-4, 2001,
Balkema Publishers, The Netherlands, 2001, pp 193-196.
Paglia C et al, An evaluation of the sulfate resistance of cementitious material accelerated
with alkali-free and alkaline admixtures. Laboratory vs. field, Cement and Concrete
Research, 32, 2002, pp 665-671.
Paglia C et al, The influence of alkali-free and alkaline shotcrete accelerators within cement
systems. Influence of the temperature on the sulphate attack mechanisms and damage,
Cement and Concrete Research, 33, 2003, pp 387-395.
http://www.fritzpak.com/products/product_nca.html
http://www.basf-admixtures.com/NR/rdonlyres/6376A357-F14D-4516-9F69C951510AAF28/0/Pozzolith_NC534_MSDS606.pdf
http://www.basf-admixtures.com/NR/rdonlyres/5EDF7876-A3C0-47FD-8E9E2B0253D1D26A/0/Pozzutec_20_Plus_MSDS606.pdf
http://www.sika.se/upload/S%E4kerhetsdatablad/SikaRapid-1%20VIB.PDF
http://sika.no/sika_rapid_2_(1).asp
http://sika.no/sika_rapid_3_(1).asp
http://hms-chess.resconmapei.no/UK/index.htm
http://www.euclidchemical.com/fileshare/ProductFiles/msds/721_99_U.pdf
http://www.euclidchemical.com/fileshare/ProductFiles/msds/019N_99_U.pdf
http://www.aximconcrete.com/default.aspx?pageid=86&ProductId=27&GroupId=45
http://www.aximconcrete.com/default.aspx?pageid=86&ProductId=28&GroupId=45
http://www.ugc.basf.com/NR/rdonlyres/C078F785-AF93-4412-BDDD92BD39891463/91547/MeycoSA160_20070222.PDF
http://www.ugc.basf.com/NR/rdonlyres/AE617443-4CF0-46C4-9A539E4E4D205246/91550/MeycoSA163_20070222.PDF
http://www.ugc.basf.com/NR/rdonlyres/595BC15F-B346-4DD0-A20797F4DC611168/91551/MeycoSA164_20070222.PDF
http://www.basf-cc.no/NR/rdonlyres/3498E380-3A73-4817-BB8B38E4D1479B4A/79702/MeycoSA172.pdf
http://www.ugc.basf.com/NR/rdonlyres/E0E59041-1C53-4BE1-B4C1D82F596A112C/91553/MeycoSA172_20070222.PDF

35

90

http://www.ugc.basf.com/NR/rdonlyres/FC42580A-003A-4525-B36772598988C506/91555/MeycoSA175_20070222.PDF
91 http://www.ugc.basf.com/NR/rdonlyres/DBCADFFA-685D-44FD-9D478C16BC21150F/91557/MeycoSA180_20070222.PDF
92 http://www.sika.ch/stm/stm-produkte-4/con-produkte-tunnel-spritzbeton
93 http://www.sika.no/upload/No/Construction%20%28Bygg%20og%20Betong%29/HMS_dat
ablad/HMS_Sigunit-L5450_AF_N.pdf
94 http://www.mapei.it/
95 http://www.euclidchemical.com/fileshare/ProductFiles/msds/019S_99_C.pdf
96 See ref. 35, Chapter 13, pp 747-749.
97 See ref. 6, Chapter 10, pp 330-331.
98 Rodger S A and Double D D, The chemistry of hydration of high alumina cement in the
presence of accelerating and retarding admixtures, Cement and Concrete Research, 14,
1984, pp 73-82.
99 Scian A N et al, Mechanochemical activation of high alumina cements Hydration
behaviour. I, Cement and Concrete Research, 21, 1991, pp 51-60.
100 Nilforoushan M R and Sharp J H, The effect of additions of alkaline-earth metal
chlorides on the setting behaviour of a refractory calcium aluminate cement, Cement and
Concrete Research, 25, 1995, pp 1523-1534.