Cement and Concrete Research 32 (2002) 1605 – 1611

Influence of cement and superplasticizers type and dosage

on the fluidity of cement mortars—Part I
S. Chandra*, J. Björnström
Chalmers University of Technology, Civil Engineering Department, Applied Concrete Chemistry, 412 96 Göteborg, Sweden

Received 10 January 2002; accepted 29 April 2002

Abstract

Concrete quality is controlled by the flow behavior of cement paste, which is related to the dispersion of cement particles.
Superplasticizers (SPs) provide the possibility of a better dispersion of cement particles, thereby producing paste of higher fluidity. With the
development of high strength, high performance concrete, SPs are becoming indispensable. SPs are adsorbed on the cement particles. This
adsorption is uneven and depends upon the clinker composition of cement and the type of SP used. This work is focused on the study of the
influence of lignosulfonic acid (LS)- and melamine sulfonic acid (SMF)-based SPs on the fluidity of mortars made with ordinary Portland
(OPC), low alkali (LAC) and white cement (WC) at different water to cement ratio. It is shown that LS are more effective than SMF in
providing better fluidity. Further WC has given the highest fluidity among the cements used. It is attributed to the lower C3A + C4AF and
alkali content, and higher SO3 content.
D 2002 Elsevier Science Ltd. All rights reserved.

Keywords: Adsorption; Clinker mineral; Workability; Cement paste; High-range water reducers

1. Introduction availability of data, especially the molecular weight of SPs.

Cements, on the other hand, vary in composition and
The properties of concrete are governed by its flow fineness, which influence upon the hydration process and
behavior, which is controlled by the dispersion of cement thereby on the flow behavior of cement paste. In this work,
particles. It is widely known that better fluidity is achieved influence of different dosage of lignosulfonate and sulfo-
by the addition of superplasticizers (SPs). The role of SPs in nated melamine formaldehyde SPs, on the fluidity of
concrete has therefore become increasingly important. For mortars made with ordinary Portland cement (OPC), low
instance, the contribution of SP in the development of high alkali cement (LAC) and white cement (WC) has been
strength concrete is remarkably larger than that of cement. studied. Possible mechanism of interaction is discussed.
The SPs are adsorbed on the cement particles, which
deflocculate and separate, releasing trapped water from
cement flocks. A variety of SPs have been developed and 2. Materials and methods
are available in the market. These belong to different basic
groups, such as lignosulfonic acid (LS), melamine form- 2.1. Cements
aldehyde sulfonic acid (SMF), naphthalene formaldehyde
sulfonic acid (SNF) and polycarboxylic acid (CE). Apart Three types of cements were used. OPC and LAC were
from the SPs of different basic groups, there can also be supplied by Cementa, Sweden. WC was supplied by Aal-
differences in SPs from the same group depending upon borg Portland Cement, Denmark. The properties are shown
their synthesis, which influences upon the molecular weight in Table 1.
and chemical configuration. The problem, which is encoun-
tered in the interpretation of results, very often is the non- 2.2. SPs

* Corresponding author. Tel.: +46-3-1772-2301; fax: +46-3-1772-2853. LS, Cementa P40, was supplied by Cementa.
E-mail address: [email protected] (S. Chandra). SMF, Cementa Flyt 92M, was supplied by Cementa.

0008-8846/02/$ – see front matter D 2002 Elsevier Science Ltd. All rights reserved.
PII: S 0 0 0 8 - 8 8 4 6 ( 0 2 ) 0 0 8 3 9 - 6
1606 S. Chandra, J. Björnström / Cement and Concrete Research 32 (2002) 1605–1611

Table 1 Hobart mixer. Fluidity was measured on the cement mortars

Composition and properties of the cements used: OPC, LAC, WC
made with different types of cements and SPs at different
OPC LAC WC water to cement ratios. The dosage of SP mentioned in this
C2S, % 13 22.8 25 study was calculated to the weight of cement. Flow was
C3S, % 64 55.7 63 measured at 20
C by pull out spread of the mortar from a
C3A, % 5 2.1 4
cone of top diameter 70 mm, bottom diameter 100 mm and
C4AF, % 10 13.1 1
K2O + Na2O, % 1.6 0.5 0.3 height 60 mm. The spread was the average of two perpen-
SO3, % 3.2 2.4 3–5 dicularly crossing diameters. From the spread ( F), relative
Specific surface area, m2/kg 337 304 410 flow area (Ã) was calculated by Eq. (1) [1].
Density, kg/m3 3200 3210 3150
~
* Data presented were supplied by the manufacturers. A ¼ ðF=r0 Þ2  1 ð1Þ

Standard grade sands 1, 2 and 3 was used. The mortars where r0 = 50 mm, bottom cone radius.
were made with 1:3 cement to sand ratio, where equal part Flow area and fresh density of mortars with OPC, LAC
of each sand grade was mixed. Mortars were mixed in a and WC with different dosage of LS (P40) and SMF
(Cementa Flyt 92M) were measured at different water to
cement ratios. Air content was calculated. Flow area and

Fig. 2. Flow area (a), fresh density (b) and air content (c) with respect
Fig. 1. Relation between flow area (a), fresh density (b) and air content (c) to water to cement ratio at different dosage of lignosulfonate SP, P40.
for OPC, LAC and WC at different water to cement ratio. For OPC.
 S. Chandra, J. Björnström / Cement and Concrete Research 32 (2002) 1605–1611 1607

ratio. Similar trends were observed with the fresh density up

to 0.45 water to cement ratio. At 0.50 water to cement ratio,
the fresh density of WC was slightly higher, while the air
content was slightly lower than for the other two cements.
This was due to better compaction of the mortar because of
enhanced fluidity.
Fig. 2 shows the results of mortars made with OPC and
0.3, 0.5 and 0.8 wt.% of lignosulfonate (P40) SP. It is seen
from Fig. 2a –c that the addition of P40 has increased the
flow. However, an increase in the dosage from 0.3% to 0.5%
did not result in a significant increase on the flow, marginal
changes are, however, noted. At 0.8% P40, the flow was
also similar to 0.3% and 0.5%, but has significantly
increased at 0.6 water to cement ratio. Substantial differ-
ences on fresh density were observed up to 0.45 water to
cement ratio, exceeding, which no appreciable difference in
the fresh density could be noticed. At 0.35 water to cement
ratio, the density of mortars with 0.3% and 0.5% P40 are

Fig. 3. Flow area (a), fresh density (b) and air content (c) with respect
to water to cement ratio at different dosage of lignosulfonate SP, P40.
For LAC.

fresh density data presented are based on triplicate measure-

ments. The average standard deviation in determining flow
area is 1.12 and for the fresh density determinations 5.88.
The results are shown in Figs. 1– 7.

3. Results and discussions

The results of the tests performed with three cements

without SP are shown in Fig. 1a –c. It is seen here that there
is no significant difference on the fluidity of mortars up to
0.45 water to cement ratio. At 0.50 water to cement ratio,
the fluidity of WC increased compared to the other cements,
whereas the cement OPC and LAC have shown the same Fig. 4. Flow area (a), fresh density (b) and air content (c) with respect to
fluidity. High fluidity was marked at 0.60 water to cement water to cement ratio at different dosage of lignosulfonate SP, P40. For WC.
1608 S. Chandra, J. Björnström / Cement and Concrete Research 32 (2002) 1605–1611

some differences when the dosage increased to 0.8% com-

pared to 0.3% and 0.5%. It is attributed to the increase in
cohesiveness of mortar with increased dose of P40.
Fig. 4 shows the influence of P40 on WC. It is seen from
Fig. 4a that there is no significant difference in the fluidity
up to 0.40 water to cement ratio for different doses of P40.
At 0.45 water to cement ratio, there is a substantial
difference in fluidity between 0.3% and 0.5%. Whereas
between 0.5% and 0.8%, there is practically no difference.
There was no substantial difference in the densities and air
content for the three dosages of P40 at all the water to
cement ratios (Fig. 4b,c). It is, however, observed that, at
0.35 water to cement ratio, the density was low and the air
content was high. It is because of the inadequate fluidity and
thereby bad compaction of the mass in the density measure-
ments. At 0.40 water to cement ratio, the density increased
for the three dosages. With higher water to cement ratios,

Fig. 5. Flow area (a), fresh density (b) and air content (c) with respect to
water to cement ratio at different dosage of SMF SP, Flyt 92M. For OPC.

similar and lower than that obtained with 0.8% P40. This is
in agreement with the air content, which is higher for 0.3%
and 0.5% and lower for 0.8% P40. Increase in the density at
0.45 water to cement ratio follows with lower air content. At
0.35 water to cement ratio, the lower density and conse-
quently higher air content is due to inadequate packing of
the mortar in the pot during weighing for density deter-
mination. With increase in the water to cement ratio to 0.50
wettability increases, which provides better compaction,
thereby density increases and air content decreases com-
pared to with 0.35 water to cement ratio. With further
increase in the water to cement ratio, flowability increased
but there is not so much decrease in the density as the mass
was very cohesive and there was no bleeding. Because of
this the air content is low and stable.
Fig. 3 shows the influence of P40 on LAC. It is seen from
the figure that the flow of mortars with 0.3%, 0.5% and 0.8% Fig. 6. Flow area (a), fresh density (b) and air content (c) with respect to
P40 were similar. The density and air content have, however, water to cement ratio at different dosage of SMF SP, Flyt 92M. For LAC.
 S. Chandra, J. Björnström / Cement and Concrete Research 32 (2002) 1605–1611 1609

cement ratio and then decreased at 0.60 water to cement

ratio. At 0.8% SMF, there is a slight increase in the density
at 0.4 and 0.45 water to cement ratio and then a decrease.
This decrease in density is due to the increase in the fluidity
(Fig. 6b). The air content for 0.3% and 0.5% SMF was the
same at all the water to cement ratios. It, however, decreased
with 0.8% water to cement ratio up to 0.50, beyond which it
became constant (Fig. 6c).
Fig. 7 shows the influence of SMF on WC. It is seen here
that there is no appreciable difference on fluidity, density and
air content with increasing dosage. The fluidity increased
with increase in water to cement ratio (Fig. 7a). But there was
practically no difference in the density (Fig. 7b). The air
content decreased with the increase in the fluidity and
exceeding 0.55 water to cement ratio, there was practically
no air in the mixture (Fig. 7c). It shows that the increase in
the dose of SMF does not increase the fluidity, 0.3% can be
translated as the optimum dose. WC cement has shown the
highest fluidity compared to the other cements.

3.1. Compatibility of cements with LS and SMF

The fluidity of a cement paste is related to the hydration

of the cement, which, in its turn, is connected with the
cement composition and fineness. Cement particles contain
several mineral phases of different reactivity, as well as a
variety of chemical and structural defects. Their initial
hydration will likely generate a surface with important
variation in the surface charge density, both in size and
magnitude. These localized surface charges promote floc-
culation of hydrating cement particles, but they can be
effectively neutralized and separated by the anionic charge
of the SP molecules.
The hydration of interstitial phases is affected by the
concentration of Ca2 + , OH  and SO42  ions in the
mixing water. The concentration of those ions depends upon
Fig. 7. Flow area (a), fresh density (b) and air content (c) with respect to the amounts of alkali sulfate, gypsum and free lime in the
water to cement ratio at different dosage of SMF SP, Flyt 92M. For WC.
cement just after mixing with water, after that, it depends
upon the hydration reaction of C3S, alite. The hydration of
the density decreased, as the fluidity increased. The air interstitial phases is affected in particular by the lime –
content decreased simultaneously. It was noted that the saturation ratio. Since small crystals of ettringite, which
cohesiveness of the mixture increased with the increase in are produced in high concentration conditions, cover the
the P40 dosages. unreacted interstitial phase, the hydration reaction rate slows
Identical tests were performed with SMF (Flyt 92M) and down. On the contrary, in low ion concentration conditions,
three cements OPC, LAC and WC. Fig. 5 shows the large amounts of ettringite are produced in the shape of large
influence of SMF on OPC. It is seen here that, with OPC, needles. In this case, the hydration of the interstitial phase
there is no difference in the flow, density and air content continues to produce large amounts of ettringite, which
with increase in the dosages of SP (Fig. 5a –c). Increase in causes the stiffness and pseudo-setting [2,3].
the water to cement ratio increased the fluidity equally. A few points relevant in the interaction of SP with
Fig. 6 shows the influence of SMF on LAC. It is seen cements, which are sometimes overlooked, are worth men-
from Fig. 6a that, with LAC, the fluidity increased with the tioning [4].
increase in the water to cement ratio. It was higher for 0.8% First, the relative sizes of the particles in a cementitious
compared to 0.3% and 0.5%. At 0.55 and 0.60 water to system of SP molecules differ typically by two or three
cement ratio, the fluidity increased more for 0.5% and 0.8% orders of magnitude; the average diameter of cement par-
compared to 0.3% (Fig. 6a). The density was lower for ticles are typically of 10 mm, whereas the size of the SP
0.3%. For 0.5%, it was almost constant up to 0.55 water to molecules is of the order of a few nanometers.
1610 S. Chandra, J. Björnström / Cement and Concrete Research 32 (2002) 1605–1611

Second, since the initial surface hydration of cement content of the WC. This agrees with the results reported by
particles is extremely rapid (t1/2 < 1 min), the action of Hanna et al. [7]. However, a comparison with the fineness
SP molecules will mainly occur on the surface of the shows a contradiction. The fineness of WC is the highest,
hydrated particles. yet it has shown the highest fluidity. This may be due to the
Third, as complex chemical entities, SP molecules can fact that C3A + C4AF and the alkali are very low and the
themselves participate in chemical processes. sulfate content is high. Because of the competitive adsorp-
tion between SP and SO3, most of the SP remained in the
3.2. Depletion effect solution, poisoning the hydration process so much that
the fineness became insignificant. It has been reported that
As the hydration reaction proceeds, the amount of free the extent of adsorption is influenced more by the sulfate
water decreases, and so does the distance between the content than by the fineness. In high sulfate content
hydration surfaces of the neighboring cement (hydrate) cement, adsorption of SP is less than in low sulfate content
particles. As the interparticle volume becomes smaller, the cement [8]. Similar results have been reported by Simard
concentration of SP molecules becomes higher, the concen- et al. [9].
tration of SP molecules confined in this volume may create
a substantial osmotic pressure effect. The latter would either 3.3. Influence of alkali content
tend to expel the SP molecules from the confined interpar-
ticle volume or create a water flow to dilute the polymer In the absence of SP, cements containing high levels of
molecules in that region. The first effect would lead to alkali (e.g., Na2SO4 or K2SO4) will usually exhibit poorer
induce a particle – particle attraction, while the latter would rheological behaviors than cements having low alkali
induce additional particle –particle repulsion. contents, other conditions being the same. Also, the water
The variation in lime saturation ratio with different water reduction with admixtures will be more readily achieved
to cement ratio varies with the cement type used. The with LAC [7,10]. Several effects may be promoted by the
variation of the lime saturation ratio when using a lignin alkalis, namely: flocculation of cement (or other) fine
sulfonic acid based admixture is smaller than the variation particles induced by the electrolytes, formation of new
observed when using no admixture or SMF. The reason is hydrates containing alkali ions (e.g., syngenite) or
that lignin sulfonic acid based admixture binds up the Ca2 + , increase in the reactivity of mineral phases (particularly
and the concentration of Ca2 + in the pore solution is C3A). WC has shown the highest fluidity followed by
therefore lowered. It is seen from the experiments that at LAC and OPC. One of the reasons for this is due to its
up to 0.55 water to cement ratio there was not significant low alkali content.
increase in the fluidity, but exceeding this, the fluidity In the presence of SP, it was found that the addition of
increased substantially. It seems that, up to the water to alkali sulfates (Na2SO4) can lead to improvements in the
cement ratio of 0.55, the Ca2 + were blocked by the LS, rheological properties of cement paste [11,12]. While the
indicating the formation of an interlayer complex [5]. With results shown may not be generalized, they are consistent
an increase in the water to cement ratio, more alite hydrates with the concepts of SO42  /SP competition. The presence
and thereby more Ca2 + ions are produced. Lime saturation of SO42  ions leads to a decreased absorption of the SP,
in the pore solution increases, poisoning the hydration leaving more of the latter available on the solution phase
process. Subsequently the fluidity increases. for paste fluidification; the fluidity of the paste increases,
The admixture is not evenly adsorbed on cement par- accordingly, with the amount of Na2SO4 added. In another
ticles. It adsorbs more readily on C3A, C4AF than on C3S study Andersson et al. [12] further reported that the
and C2S [6]. Comparisons of SMF and LS have shown that adsorption of SP was reverted by addition of potassium
SMF is even more unevenly adsorbed on the clinker hydroxide (KOH).
minerals than LS is.
The more even the adsorption of SMF, the higher the 3.4. Influence of water to cement ratio
fluidity. The admixture is so unevenly adsorbed on cement
containing much C3A + C4AF that the amount of admix- The water to cement ratio controls the concentration of
ture adsorbed on alite and belite is relatively decreased, ions in the pore solution. At low water to cement ratio,
thereby lowering the fluidity of the paste. The more even the surface of interstitial phases especially of C3A and
the adsorption of admixture on cement minerals, the C4AF is adsorbing the SP; thus, very little SP is in the
higher the fluidity of the paste will be. Thus, the amount pore solution. But with an increase in the water to cement
of the admixture adsorbed on the cement sometimes ratio, more alite hydrates and thereby more Ca2 + ions
fluctuates greatly, as it depends on the mineral composi- are produced. Lime saturation in the pore solution in-
tion of the clinker. creases, poisoning the hydration process. Subsequently,
In this work, the highest fluidity was observed with WC the fluidity increases.
both for LS and SMF, followed by the cements LAC and It has been observed in the experiments that, up to a
OPC. The reason is the low C3A + C4AF and low alkali water to cement ratio of 0.45, there was no significant
 S. Chandra, J. Björnström / Cement and Concrete Research 32 (2002) 1605–1611 1611

increase in the fluidity, but when it exceeded 0.45, there was Acknowledgments
substantial increase in the fluidity.
This work has received financial support from the
3.5. Fresh density and air content Knowledge Foundation (KK-stiftelsen, Stockholm), which
is gratefully acknowledged.
Fresh density is a measure of compactness. Generally, it
is low at 0.35 water to cement ratio, while the air content is
high. It is due to the bad compaction because of inadequate References
fluidity. At 0.40– 0.45 water to cement ratio, the fluidity
increases, thereby the density increases due to better com- [1] H. Okamura, K. Ozawa, S. Matsuo, K. Shimokawa, Evaluation of
superplasticizers for self compacting concrete with mortar test, JCA
paction. But, at higher water to cement ratios, the density
Proc. Cem. Concr. 48 (1994) 374–379.
decreases gradually. The air content also decreases gradually [2] H. Uchikawa, S. Uchida, K. Ogawa, S. Hanehara, Influence of
with the increase in fluidity. The decrease in density is due CaSO42H2O, CaSO41/2H2O, and CaSO4 on the initial hydration
to decrease in the solid content in the mass as the water to of clinker with different burning temperature, Cem. Concr. Res. 14
cement ratio is increased, a dilution effect. Sometimes, the (1984) 645 – 656.
density is high observed at high water to cement ratios, it is [3] H. Uchikawa, Hydration of cement and structure formation and pro-
perty of cement paste in the presence of organic admixture, Confer-
due to mortar segregation. ence in Tribute to Micheline Moranville Regourd at Beton Canada
Conference, Sherbrooke, Canada, 1994, pp. 63–117.
[4] C. Jolicouer, M.A. Simard, Admixture – cement interactions: phenom-
4. Conclusions enology and physico-chemical concepts, Cem. Concr. Compos. 20
(2/3) (1998) 87 – 102.
[5] V.S. Ramachandran, F.R. Feldman, Effect of calcium lignosulfonate
The addition of a LS-based SP resulted in higher fluidity on tricalcium aluminate and its hydration products, Mater. Struct. 5
of the mortar compared to when a SMF-based SP was used. (1972) 67 – 76.
This is because the variation of lime saturation rate in the [6] H. Uchikawa, S. Hanehara, T. Shirasaka, D. Sawaki, Effect of admix-
case of LS is smaller than that in the case of SMF. Further ture on hydration of cement, absorptive behavior of admixture, and
fluidity and setting of fresh cement paste, Cem. Concr. Res. 22 (1992)
SMF is much more unevenly adsorbed than LS on the
1115 – 1129.
clinker minerals of cement. Higher fluidity was observed [7] E. Hanna, K. Luke, D. Perraton, P.C. Aitcin, Rheological behavior of
with WC for both LS and SMF than in the case of LAC and Portland cement in the presence of a superplasticizer, in: V.M. Malho-
OPC cements. This is attributed to the lower C3A + C4AF tra (Ed.), Proc. 3rd Int. Conf., Superplasticizer and other Chemical
and alkali content and higher sulfate content in WC com- Admixtures in Concrete, ACI SP vol. 119, 1989, pp. 171 – 188.
pared to the LAC and OPC. WC has a particle size [8] T. Nawa, H. Eguchi, Effect of cement characteristics on the fluidity of
cement paste containing chemical admixture, 9th Int. Cong. Chem.
distribution with a large amount of fine material. This Cem. IV, New Delhi, India, (1999) 597–603.
should have given a low fluidity in the mortar but the [9] M.A. Simard, P.C. Nkinamubanzi, C. Jolicouer, D. Perraton, P.C.
observations showed the opposite. The reason may be that Aitcin, Calorimetry, rheology and compressive strength of superplas-
the C3A + C4AF and alkali content was so low and sulfate ticized cement paste, Cem. Concr. Res. 23 (1993) 939 – 950.
content so high that much of the SP remained in the [10] V. Dodson, Concrete admixtures, Van Nostrand Reinhold, New
York, 1990.
solution. When it adsorbed on alite and belite, it poisoned [11] T. Nawa, H. Eguchi, Y. Fukaya, Effects of alkali sulfates on the
the hydration process so much that the fineness did not play rheological behavior of cement pastes containing superplasticizers,
a significant role. superplasticizers and other chemical admixtures in concrete, in:
Many factors are influencing the fluidity and hydration V.M. Malhotra (Ed.), Proc. 4th CANMET/ACI Int Conf., ACI SP
process of the cement paste some of these factors may vol. 119, 1989, pp. 405 – 424.
[12] P.J. Andersson, A. Kumar, D.M. Roy, D. Wolfe-Confer, The effects of
also have synergistic effects. This makes a theory based calcium sulphate concentration on the adsorption of a superplasticizer
on one parameter irrelevant, and it makes it difficult to on a cement; methods, Zeta potential and adsorption studies, Cem.
point out one parameter, which is responsible to produce a Concr. Res. 16 (1986) 255 – 259.
particular property.