Rheological Properties of

Microfine Cement Grouts

UIf HSkansson, Lars Hiissler and Hbkan Stille

A b s t r a c t ~ T h i s paper describes a measuring technique, based on R~sumd--Ce rapport d~crit une technique de mesure, fond~ sur la
rotational viscometry, that can be used to evaluate the flow properties viscomdtrie rotationnelle, qui peut $tre utilisde pour dvaluer les
of grouts. The technique has been used on cement grouts, assuming propridt~s d'dcoulement des enduits de jointoiement. La technique
the Bingham model. The paper discusses field equipment that can est utilisde sur des enduits en ciment, en admettant le module de
be used easily to estimate the flow properties on site. The results Bingham. Le rapport traite de l'dquipement que l'on peut facilement
shown indicate the effect on the flow properties by different admixtures utiliser pour est imer les p ropridt~s d "dcoulement sur site. Les rdsultats
and additives. It is concluded that an increase in specific surface obtenue montrent l'effet de diffdrents m~langes et additifs sur les
and addition of bentonite will increase both the yield stress and the propridtds d'~coulement. Ils rdvglent aprgs analyse q'une
plastic viscosity. Plasticizing admixtures reduce the yield stress augmentation en surface spdcifique et addition de bentonite accrottrait
and the plastic viscosity. By combining bentonite and plasticizers la contrainte de d~bit ainsi que la viscositd plastique. Introduire des
in the grout, improved flow properties (i.e., low yield stress, low mati~res plastiques darts les rn~langes r~duit la contrainte de d~bit
plastic viscosity) can be achieved. et la viscositd plastique. En combinant bentonite et plastificateurs
darts les enduits de jointoisment, les propriL, tds d~coulement (e.g.,
faible contrainte de tidbit, faible viscositd plastique) peuvent ~tre
amdlior~es.

Introduction Very littleresearch work isreported using grouts based on these cements
R heology is the science of defor- on the rheology of cement grouts;how- because the specificsurface will influ-
mation and flow of matter. In ever, m u c h have been done during the ence the theology.
order to simulate and predict a last20 years concerning cement pastes This paper focuses only on the rheo-
flow situation, it is important to have (forconcrete)and oilwell cement slur- logy ofmicrofine cement-based grouts.
knowledge concerning the theological ries (Banfill 1990). There are m a n y
behaviour, i.e., the constitutive rela- similarities in these fields, although Rheology of Cement Grouts
tionship, of the fluids. In most cases, research in the latter mainly focuses
the behaviour can be expressed by vari- on higher cement concentrations. The rheology of suspensions is very
ous theological models (Bird et al. 1960). W h e n cement grouts (suspensions) complex because of the interplay of
Different types of viscometers (Van are used to permeate into joints or various physical and chemical pro-
Wazer et al. 1963) are used in order to pores, it is not only the rheelogical cesses arising from both the solid and
find the theological model (and inher- propertiesthat are important and that the fluid phases (Nguyen 1985). Inter-
ent parameters) to which the measured influence the success of the grouting particle forces between the solids re-
data best can be fitted. operation, but also: sult in a yield stress (shear strength)
If a theological model can be found that must be exceeded in order to ini-
* The stabilitywith regard to sedi- tiate flow. Below the yield stress, the
(in the range of practical interest), it mentation (bleeding);and
will simplify the calculations signifi- suspension behaves as a (weak) solid.
* The size of the cement particles. Thus, the yield stress can be regarded
cantly, e.g., the volume flow rate as a
function of applied pressure. How- A cement grout is "stable~ if the as the material property that repre-
ever, in all calculations the geometry of sedimentation due to gravity is zero or sents the tranRition between solid-like
the problem must be known, and this kept at a rnlnlmuIn. A commonly used and fluid-like behaviour. Cement
olden is very difficult to achieve when criterion is that the volume of clear grouts are not only suspensions; they
dealing with a geological material. In water on top of a 1000-ml graduated are also chemically reactive during the
the case of both rock and soil grouting, cylinder, divided by the original grout hydration process, in a way that deft-
many simplifications must be made in volume, must be less than 5% after two nitely will influence their behaviour
order to define the geometry of the hours (Deere et al. 1985). It is gener- and introduce changes in their theol-
conduits (Hiissler 1991). ally considered that a stable grout is ogy over time.
more favourable than an unstable grout There are many different theologi-
in practical grouting. cal models that can be used to charac-
Micro-cement is becoming increas- terize suspensions (Nguyen 1985). In
ingly popular in replacing chemical this work, however, we have concen-
Present address: Ulf HEkansson, Royal trated on keeping the model as simple
Institute of Technology, Dept. of Soil and grouts, because the particles are much
RockMechRnics,100 44 Stockholm,Sweden; smaller and therefore can penetrate as possible. Because the geometry of
Lars HRssler, Svenska Golder Associates, into narrower voids. The fineness of the media that is to be grouted is so
Inc., Uppsala, Sweden; and H~Lkan Stille, the particles is often described by their difficultto define, a complex rheologi-
Royal Institute of Technology, Stockholm, specific surface (i.e., surface area by cal model is not considered appropri-
Sweden. weight). Care must be taken when ate at our present stage of knowledge.

Tunnelling and Ur~dergroum]Space Technology, VoL 7, No. 4, pp. 453-458, 1992. 0886-7798/92 $5.00 + .00
Printed in Great Britain. Pergamon Press Ltd 453
 The simplest rheelogical model that
includes a yield stress is the Bingham yield stress [Pa]
model. Although simple, it has been
reported by many authors that this : ~: : : [
.... : ..... ~; ....... : ...... : ...... !
model is appropriate for some cement Pl
grouts and slurries (Banfill 1981, . . . . . . . . . . ~. . . . . . . : ...... "....... i
---~ . . . . . . ix . . . . . . ~ . . . . . . ~. . . . . .
Littlejohn 1982, Lombardi 1985, Deere . . . . . . . . . . ;.." .\ . . . . . . . . . . . ; ...... ii
et al. 1985, I-I~ssler et al. 1987, and ...! ...... ! .... ~..? ...... ! ...... P2 I~
HSLkansson et al. 1991). 20 M
2r

The Bingham model consists of two ............... -x: ...... "....... i 4 1'
•-~ ...... ~ ....... ~ ..... ~ ...... i L
parameters, a (constant) plastic vis-
cosity component and a yield stress "'" ....... " ....... ".... ~'&" .... i" ....... : ....... " ....... : ....... " ........

component. The Bingham model is I0

expressed as
~=~o+P~Y (I) i i i i. . . ,. . . . ~: . .i. . . =. . .:. . .~. . . i. . .: . . . . .:.~. :. . . . . . "2.:-,, _ xa " ,I. L. .i.i. i. i. i: ~ii .i . . . . . . :.........
where ~ is the shear stress, ~ the
plasticviscosity,~othe yieldstress,and i ' i i ' 1
y the shear rate. 10 20 30 40
The plastic viscosity and the yield
stress will influence the volume flow Penetration - L [m]
~ate because they both affect the veloc-
ity profile, as shown in Figure 1 (pipe Figure 2. M a x i m u m penetration as a function o f the yield stress. Bold line: r =
flow). 0.1 mm. Dashed line: r = 0.5 mm.
In the zone where the shear stress is
lower than the yield stress, a "plug"
will form. When the pressure gradient • Specific surface (i.e., grind Measuring Technique--
along the pipe decreases, the plug will fineness); Laboratory Equipment
grow until it reaches the wall and the • C e m e n t type (i.e., m i n e r a l The rheological properties are evalu-
flow stops. Thus, the yield stress has composition); ated by using a Brookfield Rheoset,
the most important effect of limiting • Cement hydration (i.e., time rotational viscometer (HAkansson et
the distance that the grout will pen- dependency); al. 1991). The viscometer is of the
etrate (Lombardi 1985). • Mixing t i m e a n d m i x i n g concentric-cylinder type, whereby an
The ru~ximlun distance a grout, with intensity; and inner cylinder (bob) rotates while the
a yield stress, will penetrate in a pipe • Temperature. outer cup is stationary (see Fig. 3). The
can be expressed by viscometer measures the torque needed
The water/cement ratio and the spe-
cific surface significantlyinfluencethe to rotate the inner cylinder at a certain
L _APr (2)
rheologicalproperties. Theyield stress rotational speed. The main advantage
2~o of this device is that the same fluid can
will increase with increasing solids
where AP (P1- P2) isthe pressure differ- concentration (i.e., decreasing w/c) and be measured continuously for extended
ence between the ends of the pipe and increasing specific surface (vom Berg periods of time. This means that not
r is the pipe radius. Eq. 2 is also valid 1979). Although this paper mainly only can the properties be evaluated,
for flow in a rectangular channel, by focuses on how to change the proper- but also their variation with time.
substituting r for the channel width b ties with additives and admixtures, it When measuring a fluid that has a
(Hiissler 1991). Moreover, it is valid also concerns the effect of varying the yield stress, three distinct conditions
for radial flow between two parallel specific surface. Can Occur:
plates, by the same substitution as The mixing procedure must be con- % ~ Xo (3)
above (Lombardi et al. 1985). In Fig- sistent and it is important that the % < %~ % (4)
ure 2, the relationship is illustrated mixing equipment be appropriate (i.e.,
with a pressure difference of 1 MPa %<% (5)
high-speed mixers). Inadequate mix-
and a radius of 0.1 rnm (bold line) and ing will result in nonreproducible prop- where Zbis the shear stress at the bob
0.5 mr, (dashed line), respectively. erties. Banfill (1981) showed that a and z¢ is the shear stress at the cup
The most important factors influ- m i x i n g t i m e of about five minUteS is wall. First, if~ bis lower than the yield
encing the theology of a cement grout sufficient to obtain constant properties stress (Eq. 3), there is no flow at all.
are (vom Berg 1979): for beth the yield stress and the plastic Second, only in the part of the cup
• Water/cement ratio (w/c); viscosity. where the yield stress is exceeded (R <
Ro) does flow occur, while the rest of the
fluid is stationary (Eq. 4). Finally,
when ~¢is higher than the yield stress
(Eq. 5), there is flow throughout the
cup. The situation with only partial
flow (i.e., with a stationary zone) in the
r ~ (r) v (r)
cup is illustrated in Figure 3.

l Z
~o

- ~o
When there is flow throughout the
cup, the properties can be evaluated by
using

Vb-~c =2~o I n Rc + 2~tB~o (6)

Rb

Figure 1. B i n g h a m flow in a pipe (v = velocity, r ~-p l u g radius.

454 TUNNELLINGANDUNDERGROUNDSPACETECHNOLOGY Volume 7, Number 4, 1992

 the grout is left to rest for awhile, the
properties will be different from those
achieved when the grout is continously
~o is e x c e e d e d agitated. Asm, mlng that, under a nor-
real grouting operation, the grout is
continuously in motion, it is the ~dis-
turbed" (or dynamic) properties that
r o t a t i n g bob, radius = R b ..!.:.!:!:i:i:!:!:i:!:i:!:. are to be evaluated.
When using a rotational viscom-
eter, thixotropy can be seen by a differ-
~o is not e x c e e d e •.•••••••!i!ii•••••!!•••••••••iii••i•i•i?i•i{ii!i•iii•i!i!!iiiiiiiii!!iiii•i!{i!•ience
•i (hysteresis) in the curve obtained
(stationary zone) by increasing the angular velocity
(RPM), compared with the curve show-
ing decreasing velocity. In order to
Figure 3. Top view showing the stationary zone (Ro < R < R e) in the cup (R = R~). evaluate the "disturbed" properties,
only these achieved through decreas-
ing angular velocity are considered
herein. Moreover, a stirring device is
which is known as the Reiner-Riwlin higher (up curve) or the opposite (down connected to the measuring cup, which
equation. ~b is the angular velocity of curve). A sequence takes about five can be used to agitate the sample after
the bob. 1~ and R are the radius of the minutes; the properties are evaluated each sequence.
bob and cup, respectively. The shear after each.
stress at the bob, needed for Eq. 6 to Three different curve fitting meth-
apply, can be calculated by using ods are used in the evaluation: two for Measuring Technique---Field
the curvilinear behaviour according to Equipment
R Eq. 8, and one for the linear behaviour The yield stress of a grout can be
v~,mln = ~o - -
¢

(7) according to Eq. 6. Results from each determl, ed by using the fact that there
< method are compared with the mea-
sured data, and the yield stress and the
is no flow when the mA~elmLlmshear
stress is below the yield stress. A pipe,
When the shear stress at the bob is plastic viscosity are stored (after each with a known radius and a rough inner
less than ~ . a stationary zone ap- sequence) for the method that com- wall, is immersed in the fluid to a
• S a m

pears m the cup and the following equa- plies best. certain level (hl). By observing the
tion applies: height that the fluid reaches inside the
Problems Encountered pipe (h 2) (as shown in Fig. 5), the yield
R stress can be estimated. When the
Slip at the bob-fluid interface is fre- propagation inside the pipe stops, the
- % = 2~ ° / n _~o + 2~B ! (8) quently observed when measuring sus-
% "plug" reaches the wall, where the shear
pensions while using a bob with a stress is equalto the yield stress. Math-
It is important to note that R ° is a smooth surface. I f slip occurs, the
e m a t i c a l l y , t h e m e t h o d c a n be
function of~. An increasing shear stress evaluated properties will not be repre- described by performing a vertical force
will increase the effective gap in the cup sentative for the fluid. To reduce the balance, which leads to:
(Ro- Rb). This change of geometry leads effect of slippage, a thin layer of fine
to a curvilinear relationship between ~b sand (silt) is (in this study) attached to
the surface of the bob. •0=YR( ~---~ (9)
and ~b, as can be seen in Figure 4. Eq.
7 is used as a criterion to distingni~h Thixotropic behaviour is commonly 2 ~hl-Ah!
between when Eq. 6; or Eq. 8 can be encountered when using a low water/ where All is (hi-h2) , y is the specific
applied, as shown in Figure 4. cement ratio or when bentonite is weight of the grout, and R is the pipe
The whole measurement is divided added. Thixotropy means that the rheo- radius. It can be noted that if the grout
into a number of sequences whereby logical properties are dependent not reaches all the way up inside the pipe
the a n g u l a r velocity of the bob is only on the shear rate, but also on the (Ah = 0), there is no yield stress. Also,
changed, either from a low RPM to a duration of shear and shear history. If of course, the shorter the penetration

":< • •

<:H• :
~b :.~• +
" ...... " "H" • : • •

:,'.j•:::'!$ < •" • / • :.S•: k " :• :: I

H H
H /

i:< ,: ••: i•
I
I
hl ::: ::
~b, m i n !

H I
., ,.,..,,, ..,.,,., , :. • :.. I
H..,.....,...,..:.....,...,, .,

1:o
• •• •: 7L• •• . ...........
-(

g/b
Figure 4. A Bingham fluid will yield a linear and non- Figure 5. The yield stress can be estimated by observing
linear curve when using a concentric-cylinder viscometer. how high the fluid reaches inside the pipe.

Volume 7, Number 4, 1992 TUN~FCLn~OANDUNDERGROUNDSPACE TECHNOLOGY455

 ~
distance, the higher the yield stress.
The graphical representation of Eq. 9, .."/g

with a pipe radius of 3 ram, is shown in : : . /

Figure 6. The density of the fluid, ...... . ." /

which must be known, can be found by ~ 20 " - p = 1600 kg/m 3 Ii. . . . . . . . . . . . . . . . . . . . . :"
using a simple field instrument called
a "mud balance" (Deere 1982). o= 14oo kg/m 3 i / ! ,"
The method can be used in combina- ~ 15 --p = 1200 kg/m 3 i '"" ~ ' . : " ~ : : ' ~ " . . . .
tion with a Marsh cone, using the proce- L . . . . i. . . . .-" ",: :. . . . . . .
dure proposed by Lombardi with the 4_) •
:
.
• ...
•
.j
f •

cone and plate (Lombardi 1985). When 10 ............... .............. ..:.": ; . . . . . . . . . . . ................
a Marsh cone is used, beth the plastic ,---t

viscosity and the yield stress will influ- '~>t 5 • ..-"7 ~ • ;

ence the thne it takes for a certain
volume to flow out. In order to separate
the two properties, one of them must be
known prior to performing the Marsh
cone test. The yield stress can be found 5 10 15
with the pipe (as described above); a Ah [cm]
nomogramis thenused (Lombardi 1985)
to find the plastic viscosity when the Figure 6. Yield stress as a function o f z3h for different densitites. Pipe radius R
Marsh flow time is measured. = 3 m m and H I = 32 cm.

Materials and Procedure

Results and Discussion creased activity in the hydration pro-
Two Portland micro-cements and cess. Also, the particles will be nega-
two slag micro-cements were used in Fineness of Cement (Specific Surface)
tively charged by adsorption, which
this study. The physical properties of The fineness ofthe cement will influ- means that they will repel each other
these cements are shown in Table 1. ence the properties by increasing both and not come close enough to form
The plasticizers that were used are: the yield stress and plastic viscosity, agglomerates.
1. Sulphonated naphthalene form- with an increase in specific surface. Plasticizers will affect the properties
aldehyde resin (SNF), (liquid with solid Figure 7 shows the properties evalu- over only a Hm~ted period oftime. It is
content 35.0%), HP, Rescon. ated from cements having three differ- believed that the metal ions that are
2. Sulphonated melamine formal- ent specific surface values, with w/c = 1. produced during the hydration process
dehyde resin (SMF) (liquid with solid are responsible for this phenomenon.
c o n t e n t 39.2%), P e r a m i n 0149, Effect of Bentonite Addition ofplasticizers reduces the
Perstorp. yield stress and the plastic viscosity.
Bentonite is often used to reduce
Ordinary tapwater at a tempera- sedimentation of the grout. Adding Figure 9 shows the result from two of
ture of 20°C was used throughout the bentonite will increase the yield stress the cements, with and without plasti-
study. The grout was mixed with a and the plastic viscosity. The bento- cizer and with w/c = 1.
high-speed industrial mixer for five nite will also introduce considerable The amount of plasticizer (SNF) is
minutes before the measurements thixotropy, depending on the amount expressed by the liquid weight of the
started. After each measuring sequence added. Figure 8 shows the effect of plasticizer divided by cement weight.
(increasing or decreasing RPM), the bentonite added by water weight, with
grout was agitated for 10-15 seconds. w/c = 1. Effect of Combining Bentonite and
The bob and the cup radius were 5.2 Plasticizer
mm and 47 ram; respectively. As noted Effect of Plasticizers By combining bentonite and a plasti-
above, only the decreasing RPM curves cizer, improved flow properties can be
are shown in the diagrams. All of the The main effect of plasticizers is to
disperse the cement, i.e., particles that achieved. Figure 10 shows the results
grouts used were "stable" according to with Cement #1 and SMF-plastizicer.
the criteria mentioned. The bentonite have joined together to form agglomer-
ates are separated. A radical increase As shown, both the yield stress and the
was mixed with the water for one plastic viscosity initially can be main-
minute before the cement was added. in the specific surface available for
water will take place, leading to in- tained at a very low value (almost simi-
lar to that of chemical grouts).

Table 1. Physical properties of cements used in the study.

Specific Surface Max grainsize

Name Type (Sq. m/kg) [,am] Producer

Degerhamn Po~land, SR* 600 64 Cementa, Sweden

Mikrodur Po~land 1150 16 Dyckerhoff, Germany

Alofix Slag 800 15 Onoda, Japan

Cement #1 Slag 1270 6 Origny, France

*SR = Sulphur-resistant

456 ~ - ~ E L L ~ a ANDUNDERGROUNDSPACETECHNOLOGY Volume 7, Number 4, 1992

 16 0,5
Mikrodur Mikrodur
14
...... Alofix 0,4 ...... Alofix
12
e Cementa Cementa
m. 1o m
m 0,3
8 >.,
j
0,2
u , ~ - "J#%%"
@
0,I
2
0 • i i • i • | • ! • ! i • 0,0 ! i i | i i | •

0 30 60 90 120 150 180 210 240 0 30 60 90 120 150 180 210 240

time [mln] time [mln]

Figure 7. The effect of specific surface at w / c = 1.

16 ! 0,5
I
Mikrodur, 1% B Mikrodur, 1% B
14
...... Alofix, 2% B 0,4 ...... Alofix, 2% B I
I;' i
Cementa, 3% B /
10
m
~ 0,3
8
o,2
w
o
o

~" 2

, , • , • : • i , i • , • i 0,0 • i , l , i • ! , ! , , • l •

0 30 60 90 120 150 180 210 240 0 30 60 90 120 150 180 210 240

tim [rain] time [mln)

Figure 8. The effect of bentonite (B) by water weight at w /c = 1.

16 0.5
Mikrodur Mikrodur
14 =.==== Mikrodur 1% P
0.4 "=='== Mikrodur 1% P
12 Alofix
i ...... Alofix 2.5% P - - Alofix
0,3 ...... Alofi× 25 ~ P

•~ 0,2
¢% ~ • #,, v %
•~ 0,1

ILC..."
a ~ m. .m. .-. . . . . . . "
0 o,o ~'. " : ' ; ' ? - . . . . . . . . . .
0 3 '0 6'0 s ' 0 " I ~0 1 5' 0 " 1 8' 0 2 1' 0 240 0 30 60 90 120 150 180 210 240
t£-~ [mln] time [mln]

Figure 9. The effect of SNF-plasticizer (P).

16 0,30
14 ' 2% B, 4% P 2% B, 4% P
2% B, 3.5% p 0,25 -" ..... 2% B, 3.5% P
12
2% B, 3% P 2% B, 3% P
,~ 0,20
-- 10
i
~ 8 0,15
.,.i !i
> 6
u o.1o
IQ

"~ 0,05
{ 2
0 , • , • , • | • , • i • i • , • i , i , 0,00 i ~ , i , |

0 30 6 0 90 1 2 0 1 5 0 1 8 0 2 1 0 2 4 0 2 7 0 3 0 0 0 30 60 90 120150180210240270300
time [m.£n] time [re.in]

Figure 10. The effect of plasticizer (P) and bentonite (B) combined (cement #1).

V o l u m e 7, N u m b e r 4, 1992 ~ G AND UNDERGROUNDSPACE TECHNOLOGY457

 2O References
18 B~n~n~ P. F.G. 1983. A viscometric study
m 16 of cement pastes containing super-
plasticizers with a note on experimental
14 techniques. Magazine of Concrete
~ 12 Research 33(114), 37-47.
I BAnfill p. F. G., ed. 1990. RheologyofFresh
10 Cementand Concrete. London:E.& F. N.
8 Sport.
Bird, R. B.; Stewart, W. E.; and Lightfoot,E.
> 6 N. 1960. Transpert Phenomena. New
4 York: Wiley and Sons.
@
Deere, D. U. mad Lombardi, G. 1985. Grout
slurries--4hick or thin? Issues in dam
0 ~, | , I , I I I • I • I • I . I • I . grouting, 156-164. New York: ASCE.
0 2 4 6 8 10 12 14 16 18 20 H~kansson, U.; ttiisaler, L.; and Stille, H.
yield value- 1991. A Technique for Measuring the
Rheological Properties of Injected Grout.
viscometer [Pa] BeFo Report 241:1/91. (In Swedish).
Stockholm: Swedish Rock Engineering
Figure 11. Yield stress from viscometer a n d pipe. Also shown is Research Foundation (BeFo).
the "perfect fit" line. I-I~aler L., Stille, H., end H~kansson, U.
1987. Simulation of grouting in jointed
rock. In Prec. ofthe 6th Int. Congress on
Comparison of Laboratory and Field 4. Plasticizing admi~rtures reduce Rock Mechanics,Montreal, Canada, Vol.
the yield stress and plastic viscosity, 2, 943-946. Rotterdam: A. A. Balkema.
Equipment Hiissler L. 1991. Grouting of rock--
but the time during which they affect
Some preliminary results from a simulation and classification. Ph.D.
the properties is limited. thesis, Department of Soft end Rock
comparison between yield stress evalu-
5. By comblnlngbentonite and plas- Mechanics, Royal Institute of Tech-
ated with the viscometer and pipe are
ticizers in a cement grout, the flow prop- nology, Stockholm, Sweden.
shown in Figure 11.
erties can be improved significantly. Littlejolm G.S. 1982. Design of cement
based grouts. Grouting in geoteehnlcal
Conclusions The results of the study are pre- engineering. In Prec. of an ASCE Conf.
sented in this relatively inconclusive in New Orleans. New York: ASCE.
A rotational viscometer can be used
form in order to show, mo~n)y in a Lombardi G. 1985. The role of cohesion in
to evaluate the rheological properties
qualitatively m~nner, how different cement grouting of rock. 15th Congres
of grouts, and to observe the effect of des Grandes Barrage, Lausanne.
manipulations can change the flow
different additives and admixtures. Nguyen, Q. D. 1983. Rheology of concen-
properties. Cement grout rheology is
The presented field equipment, to- trated bauxite residue suspensions.
extremely complex, and much more
gether with a Marshcone, can be used Ph.D. thesis, Monash University,
work is needed to achieve a better un-
to estimate the properties on-site. Australia.
derstanding of this field. [] Van Waser, J. It.; Lyons, J. W.; Kim, tCY.;
The main findings in this study are:
and ColweU,R.E. 1963. "Viscosityand
1. There is an increase in both yield Flow Measurement: a Laboratory
stress and plastic viscosity with in- Acknowledgment Handbook of Rheology." Monsanto
creasing specific surface. The majority of this work has been Chemical Company. St. Louis, Missouri:
2. Depending on the amount added, financed by the Swedish Rock Engi- Interscience Publishers.
benttenite will increase the yield stress neering Research Foundation (BeFo). Veto Berg, W. 1979. Influence of specific
and the plastic viscosity. surface and concentration of solids upon
Contributions have also been made by the flow behaviour of cement pastes.
3. The addition of benttenite also cooperation with the contracting indus-
introduces more thixotropy than exists Magazine of Concrete Research 31(109),
try through Stabilater AB, Sweden. 211-216.
without the addition of bentonite.

458 TUNN~.T.mQANDUsu~.Rm~Otn~VSPACET~CHNOLOGY Volume 7, Number 4, 1992