Construction and Building Materials 23 (2009) 2487–2491

Contents lists available at ScienceDirect

Construction and Building Materials

journal homepage: www.elsevier.com/locate/conbuildmat

Effect of nano-silica on rheology and fresh properties of cement pastes and mortars
Luciano Senff a,*, João A. Labrincha b, Victor M. Ferreira c, Dachamir Hotza a,d, Wellington L. Repette e
a
Materials Science and Engineering Graduate Program (PGMAT), Federal University of Santa Catarina (UFSC), 88040-900 Florianópolis, SC, Brazil
b
Department of Ceramics and Glass Engineering (CICECO), University of Aveiro (UA), 3810-193 Aveiro, Portugal
c
Department of Civil Engineering (CICECO), University of Aveiro (UA), 3810-193 Aveiro, Portugal
d
Department of Chemical Engineering (EQA), Federal University of Santa Catarina (UFSC), 88040-900 Florianópolis, SC, Brazil
e
Department of Civil Engineering (ECV), Federal University of Santa Catarina (UFSC), 88040-900 Florianópolis, SC, Brazil

a r t i c l e i n f o a b s t r a c t

Article history: Amorphous nano-silica (nS) particles (0–2.5 wt%) by cement were incorporated in cement pastes and
Received 7 August 2008 mortars, and their effect on the fresh state behaviour was analysed. Rheological tests showed that after
Received in revised form 14 December 2008 75 min from the mixing start, the mortar having 2.5 wt% nS shows insufficient flowability to allow its
Accepted 8 February 2009
continuous monitoring in a Viskomat PC viscometer. The influence of nS content was better observed
Available online 17 March 2009
on yield stress when compared with plastic viscosity values (the first increased about 66.5% while the
latter just increased 3.6%). With nS addition, spread, setting time and the moment to reach the maximum
Keywords:
temperature decreased 33%, 60% and 51.3%, respectively, when compared with samples without nS. X-ray
Nano-silica
Portland-cement pastes
diffraction showed presence of calcium hydroxide after 9 h in the sample with 2.5 wt% nS. The air content
Mortars increased 79% and apparent density decreased 2.4% when nS was added.
Fresh properties Ó 2009 Elsevier Ltd. All rights reserved.
Rheology

1. Introduction
In recent years, the use of nano-particles has received particular
attention in many fields of applications to fabricate materials with
new functionalities. When ultra-fine particles are incorporated
into Portland-cement paste, mortar or concrete, materials with dif-
ferent characteristics from conventional materials were obtained
[1–3]. The performance of these cementitious based materials is
strongly dependent on nano-sized solid particles, such as particles
of calcium–silicate–hydrates (C–S–H), or nano-sized porosity at
the interfacial transition zone between cement and aggregate par-
ticles. Typical properties affected by nano-sized particles or voids
are strength, durability, shrinkage and steel-bond [4]. Nano-parti-
cles of SiO2 (nS) can fill the spaces between particles of gel of C–
S–H, acting as a nano-filler. Furthermore, by the pozzolanic reac-
tion with calcium hydroxide, the amount of C–S–H increases,
resulting a higher densification of the matrix, which improves
the strength and durability of the material. Previous research
[2,5–10] indicates that the inclusion of nano-particles modifies
fresh and hardened state properties, even when compared with
conventional mineral additions. Colloidal particles of amorphous
silica appear to considerably impact the process of C3S hydration
[5]. Nano-silica decreased the setting time of mortar when com-
* Corresponding author.
E-mail address: Isenff@gmail.com (L. Senff).
pared with silica fume (SF) [6] and reduced bleeding water and
segregation, while improving the cohesiveness of the mixtures in
the fresh state [7]. When combined with ultra-fine fly ash it as-
sures better performance than that achieved by the use of silica
fume alone [2]. Besides, the compressive strength of mortar or con-
crete with silica fume is improved when compared with formula-
tions without addition [8–10].
Some authors defend that the appropriate percentage of nS has
to be small (1–5 wt%) due to agglomeration caused by difficulties
to disperse the particles during mix [6,8,10,11], while others indi-
cate that the improvement of the properties can also be achieved
with higher dosages, of about 10 wt%, if proper adjustments are
made to the formulation in order to avoid excessive self-desicca-
tion and micro-cracking that could hinder the strength [12].
When fine particles are added to cement, materials have a
strong tendency to form settlements or agglomerates when in con-
tact with water. These phenomena affect the rheological behaviour
of the paste [13] and the ultimate hardened properties [6]. Thus,
there is a need to increase the repulsive forces between adjacent
colloidal particles, by adding proper chemical admixtures or by
adding extra water to disperse the solid particles in aqueous
solution.
This paper reports the effects of nano-sized amorphous silica on
rheological properties, setting time, air content and apparent den-
sity of cement-based mortars. The heat of hydration and phase’s
determination (XRD) of cement pastes were also estimated.

0950-0618/$ - see front matter Ó 2009 Elsevier Ltd. All rights reserved.
doi:10.1016/j.conbuildmat.2009.02.005
2488 L. Senff et al. / Construction and Building Materials 23 (2009) 2487–2491

2. Experimental procedure Table 2

Chemical composition of nS.
2.1. Materials
Constituents Content
The Portland cement used was CEM I – 52.5R, as classified by the EN 197-1 stan- SiO2 (wt%) 99.4
dard [14]. Its chemical composition is shown in Table 1. The superplasticizer (SP) Na2O (wt%) 0.45
admixture was a polycarboxylic acid based (Glenium 51) with density between Al2O3 (wt%) 0.075
1.067 and 1.107 g/cm3 and solids content between 28.5 and 31.5 wt%. The nS slurry Sulphate (wt%) <0.1
(LevasilÒ 300/30%) contains 30 wt% solids and 1.21 g/cm3 density. Nano-SiO2 parti- Fe (ppm) 25
cles have an average size of 9 nm and specific surface area of 300 m2/g. Its chemical Ca (ppm) 10
composition is shown in Table 2. A commercial sand, composed by four size frac- Zn, Pb and Cu (ppm) <0.1
tions (1.2, 0.6, 0.3 and 0.15 mm), each one corresponding to 25 wt%, was used as
aggregate in the mortar.

2.2. Paste and mortar mixing Table 3

Mortar formulations.
Fresh mortars were prepared with binder/aggregate weight ratio (B/A) of 1:2
nS (wt%) W/B Mixture components
and water/binder (W/B) ratio of 0.35. The cement paste was prepared with the same
W/B ratio. Mortars and paste were produced with 0%, 1.0%, 1.5%, 2.0% and 2.5% nS in Water (ml) Cement (g) Sand (g) nS (g) SP (g)
weight, replacing cement. The amount of SP was 2.0 wt% of the binder (cement plus
0 0.35 87.5 250.00 500 0 5.0
nS). These compositions are shown in Table 3. The preparation of mortars involved:
1.0 0.35 87.5 247.50 500 2.50 5.0
(a) weighing of the components, (b) mixing the solid components inside a plastic
1.5 0.35 87.5 246.25 500 3.75 5.0
bag for 1 min, (c) adding nS and superplasticizer into water, (d) pouring the solid
2.0 0.35 87.5 245.00 500 5.00 5.0
components into water and (e) mechanical mixing for 3.5 min at 60 rpm. The
2.5 0.35 87.5 243.75 500 6.25 5.0
amount of water incorporated with nS was taken into account by reducing the
nominal content of water added to the preparation of the mortar. The paste formu-
lation are shown in Table 4.
Table 4
2.3. Testing procedures Paste formulations.

The flow table test is widely used to evaluate the workability of cement pastes, nS (wt%) W/B Mixture components
since it is easy to operate, and allows some parameters such as viscosity to be in- Water (ml) Cement (g) nS (g) SP (g)
ferred. However, the use of a rheometer permits a more accurate determination
and discrimination of the rheological parameters for distinct operational situations. 0 0.35 87.5 250 0 5
The rheological behaviour of the mixtures was represented by the Bingham equa- 2.5 0.35 87.5 243.75 6.25 5
tion (s = so + lpc) where s (Pa) is the shear stress, s0 (Pa) is the yield stress, lp
(Pa s) is the plastic viscosity and c (s 1) is the shear rate. Bingham model can be rep-
resented as a torque (T) as a function of rotation speed (N). In this case, the Bingham
model is expressed by (T = g + hN), where g and h are directly proportional to the
yield stress and plastic viscosity, respectively. 180
The rheological behaviour of fresh mortars was evaluated in a Viskomat PC vis-
cometer. The total testing time ranged from 75 to 150 min, depending on the nS up curve
content. The rheometer maximum speed used was 100 rpm and at every 15 min
150
the speed was brought to zero, kept as that for 30 s, and after the speed was increas-
ing during the 30 s until reach to 100 rpm. In all measures g and h were obtained 120
from Bingham’s model and using the down curve, since it shows a more regular
T (N mm)

behaviour, as shown in Fig. 1. The upper curve more irregular behaviour is affected
by the preceding resting period where some restructuring exists. In this work the 90
rheological behaviour was measured in mortars.
Flow table tests EN 1015-3 [15], air content EN 1015-7 [16] and apparent den- down curve
sity were carried out immediately after mixing, in a situation that corresponds to
60
the initial testing time of the rheological measurements. The apparent density
was measured into the recipient with known volume 245 ml. The setting time (ini-
tial and final) was determined by Vicat’s needle, according to EN 196-3 [17]. The
30 T = g + hN
measurement of the heat of hydration of cement paste was determined through a
quasi-adiabatic calorimeter (Langavant). The XRD (Rigaku Geigerflex, Cuka) was 0
determined at the moment where the maximum temperature is reached, and after 0 20 40 60 80 100
1 and 3 days curing. The scanning speed was 0.02°/s. N (rpm)

3. Results and discussion Fig. 1. Bingham’s model applied on rheogram of mortar.

The influence of the nS content on the rheological behaviour of

mortars can be seen in Fig. 2. There is an evident increase of torque increase as it will be shown below. By reducing mineral addition
values at early testing times when nS is incorporated into the mix- (nS) and keeping the equal SP content in mixture, more SP is avail-
ture. The mortar with nS showed the higher torque along all the able to help the dispersion of cement particles [18]. Although
testing period (Fig. 2a) due to the plastic viscosity and yield stress increasing slightly, the value of the plastic viscosity (h) of nS-con-
taining mortars is not considerably affected by the content of nS.
Table 1 Its variation along the testing time is also minor (Fig. 2b). However,
Chemical composition of Portland cement CEM I – 52.5R. differences to the mortar without nS are clear. Yield stress in-
Constituents Content (wt%) creases by increasing the nS content in the mixture (Fig. 2c). After
SiO2 20.9
75 min testing, the mortar with 2.5% nS shows very high yield
Al2O3 4.60 stress values, over the maximum torque applied by the rheometer,
Fe2O3 3.15 so the test could not proceed further with time. By sequentially
CaO 62.0 increasing the nS content (0, 1.0, 1.5, 2.0 and 2.5 wt%) the progres-
MgO 2.00
sion of yield stress after 75 min testing is 126.4%, 4.3%, 14.7%,
SO3 3.60
+38.1% and +66.5%, respectively. This percentage is calculated tak-
 L. Senff et al. / Construction and Building Materials 23 (2009) 2487–2491 2489

a 300
0% a 280 1.6

1% 260 1.4
250 1.5%
2%
240 1.2
2.5%

h (N mm min)
220

Spread (mm)
Torque (N mm)

200 1.0
200
0.8
150 180
0.6
160
spread 0
100 140 0.4
spread 15
h
120 0.2
50
100 0.0
0.0 1.0 1.5 2.0 2.5
0
0 15 30 45 60 75 90 105 120 135 150
nS content (wt%)

Time (min)
b 280 60
260
b 1.6
0% 240
50
1 .4 1.0%
220 40

Spread (mm)
1 .2 1.5%

g (N mm)
h (N mm min)

2.0% 200
1.0 2.5% 30
180
0.8
160 20
0.6
140 spread 0
0 .4 spread 15 10
120 g
0 .2 100 0
0.0 1.0 1.5 2.0 2.5
0.0
0 15 30 45 60 75 90 105 120 135 150 nS content (wt%)
Time (min) Fig. 3. Influence of nS content on spread (after 0 and 15 strokes) and rheological
(h, g) parameters estimated after mixing.
c 180
0%
160 550
1.0%
140 1.5%
500
120 2.0%
g (N mm)

2.5% 450
100
Setting time (min)

80 400
60
350
40
20 300

0 250
0 15 30 45 60 75 90 105 120 135 150
initial (min)
Time (min) 200
final (min)

Fig. 2. Influence of nS content on the time evolution of rheological behaviour of the 150
mortar: (a) torque; (b) h and (c) g. 0.0 1.0 1.5 2.0 2.5
nS content (wt%)
ing as reference the maximum test time for the 2.5 wt% nS-con- Fig. 4. Variation of setting time (initial and final) on the mortar with the nS content.
taining sample.
When nS (wt%) is incorporated into the mortar in the fresh state 55
it has a direct influence on the water amount required in the mix-
ture. This behaviour confirms the fact that additions of high surface 50 nS = 2.5%
area mineral particles to cement mixtures cause the need for high-
45
Temperature (∫C)

er amounts of water or chemical admixtures in order to keep the nS = 0%

workability of the mixture [2,3]. If the water content is kept con-
40
stant, as in the actual conditions, an increase of nS content will pro-
mote the packing of particles, decreasing the volume between 35
them and decreasing the free water. Therefore, there is a higher
internal friction between solid particles, which contributes to the 30
increase of torque. Hence, values of yield stress are seriously af-
25
fected and nS content can be considered as a limiting factor that
shortens the open testing time. Setting time is also anticipated 20
and the dormant period was decreased, as it will be shown below. 0 5 10 15 20 25 30 35 40 45 50 55 60 65 70
The addition of nS to the mixture reduces the spread on flow Age (h)
table (Fig. 3), due to the increase of cohesion in the mortar. Mortars
with nS = 2.5 wt% shows a reduction of 32.8% in the spread Fig. 5. Influence of nS (wt%) content on temperature of cement paste hydration.
2490 L. Senff et al. / Construction and Building Materials 23 (2009) 2487–2491

a oΔ
o Ca3SiO5
Δ Ca2SiO4 o 9h
o Δ
▲ Ca(OH)2 o
Δ
Etringite o
Δ o
o o o o o
o o o
Δ o CEM I 52.5
o oΔ o Δ
o o
Δ
o o o o o o
Δ o 9h
o
o oΔ
Δ
o Δ o
▲
Δ ▲ o
Δ ▲ o o o ▲ o o o
o ▲ 1d

oΔ
o
▲
o oΔ
Δ o ▲
▲ o o Δ ▲o
Δ o o ▲ o o o 3d

4 6 8 10 12 14 16 18 20 22 24 26 28 30 32 34 36 38 40 42 44 46 48 50 52 54 56 58 60 62 64 66 68 70
2θ

oΔ
b o Ca3SiO5
Δ Ca2SiO4 o 9h
o Δ
▲ Ca(OH)2 o
Δ
Etringite o
Δ o
o o o o o
o o
Δ o CEM I 52.5
o oΔ o Δ
o o
Δ
▲ Δ o o o o o o o
9h
o
oΔ
o oΔ
▲ Δ o
▲ o
▲ o Δ ▲
Δ o o o ▲
o o o 1d

oΔ
oo
o ΔΔ
▲ o
Δ ▲
▲
o o ▲o
Δ o ▲ o ▲
o 3d

4 6 8 10 12 14 16 18 20 22 24 26 28 30 32 34 36 38 40 42 44 46 48 50 52 54 56 58 60 62 64 66 68 70
2θ

Fig. 6. XRD of cement paste (a) without nS and (b) with nS.

diameter in comparison to the mortar without nS. By increasing
the nS content from 1 to 2.5 wt%, spread on flow table reduces
by 19.6%, while plastic viscosity only increases 3.6%. The progres-
sion of yield stress values is even larger (157%). Therefore, the
variation of spread is closely associated with yield stress when
nS is added into the mortar.
The addition of nS reduces the setting time (Fig. 4) and the
beginning of the acceleration period of cement hydration (Fig. 5).
The results showed that it is possible to establish a relationship be-
tween initial setting time and the time for the beginning of the
acceleration period during cement hydration. With 2.5 wt% nS, both
the initial setting time and the time to reach the maximum temper-
ature upon hydration decrease, respectively, 60% and 51% when
compared with samples without nS. This corresponds to an oppos-
ing delaying action of the superplasticizer, due to the adsorption of
superplasticizer on the surface of cement particles [19]. According
to Björnström et al. [5] it has not yet been established whether
the faster hydration of cement in the presence of nS is due to its
 L. Senff et al. / Construction and Building Materials 23 (2009) 2487–2491
6 2.30
5 the mixture, the setting time and the moment to reach the maxi-
Air content (%)
4 tively. The XRD showed the presence of CH, already after 9 h, in
3 2.25
2 and increases the air content in these mortars.
air content
1 Apparent density
0 2.20
0.0 1.0 1.5 2.0
nS content (wt%)
Fig. 7. Influence of nS content on air content and apparent density on the mortar.
chemical reactivity upon dissolution (pozzolanic activity) or to a
high surface activity. The acceleration of cement hydration and con-
sequently reduction of setting time and dormant period was also
detected indirectly by an increase of torque values upon rheological
tests. In addition, the presence of calcium hydroxide (CH) in the
samples with nS addition is possible to be observed with 9 h, using
the XRD shown on Fig. 6. According to Taylor [20], the rapid growth
of crystalline CH and fall of calcium ions in solution occur at the end
of induction period and suggest that the precipitation of CH is re-
lated to the start of the acceleratory stage.
The increasing nS addition also decreases the fresh apparent
density state of mortar (Fig. 7). This reduction can be attributed
to the combined effect of air-entraining and the replacement of
denser cement particles (3.1 g/cm3) by lighter nS particles
(2.21 g/cm3). With 2.5 wt% nS, there is an increase on the air-en-
trained content of 79%, when compared with mortars without nS
(Fig. 7). Therefore, the effect of entrained air must be considered
to the reduction of fresh apparent density [3]. Nevertheless, the
lower density values and higher air contents were not enough to
counteract the faster increase of torque and yield stress when nS
is added (see Fig. 2).
4. Conclusions
Nano-SiO2 modified the characteristics of fresh mortars. In for-
mulations having fixed values of W/B and SP amount, the presence
of nS decreases the amount of lubricating water available in the
mixture. The yield stress increases considerably when nS is incor-
porated in the paste, being the most affected rheological parame-
ter. On the other hand, changes on plastic viscosity are less
intense. The addition of nS also reduces the spread diameter on
the flow table of these mortars, due to the gain in cohesiveness
of the paste. The relationship between spread and yield stress val-
ues better describes the effects of nS addition. For example, 2.5 wt%
of nS causes a spread diameter reduction of 19.6%, while the yield
stress progresses 157%. At the same time, the plastic viscosity in-
creases only 3.6%.
2491
By adding nS, the beginning of setting is anticipated and the
2.29 dormant period is reduced. When 2.5 wt% of nS was added into
Apparent density (g/cm 3 )
2.28
mum hydration temperature decreased 60% and 51.3%, respec-
2.27
2.26
samples with nS addition. Therefore, the nS addition contributes
to an increased production of CH at early age compared with sam-
2.24 ples without nS. The nS addition decreases the apparent density
2.23
2.22
2.21 Acknowledgements
2.5 The authors acknowledge the support of CAPES, Brazil. The
authors also thank Weber-Portugal, BASF, H.C. Starck Empowering
High Tech Materials and Cimpor, for providing raw materials.
References
[1] Li H, Gang H, Jie X, Yuan J, Ou J. Microstructure of cement mortar with nano-
particles. Compos Part B: Eng 2004;35(2):185–9.
[2] Older I. Lea’s chemistry of cement and concrete. 4th ed. London: Arnold; 1998.
[3] Neville AM. Properties of concrete. 4th ed. England: ELBS with Addison Wesley
Longman; 1996.
[4] Collepardi S, Borsoi A, Ogoumah Olagot JJ, Troli R, Collepardi M, Cursio AQ.
Influence of nano-sized mineral additions on performance of SCC. In:
Proceedings of the 6th international congress, global construction, ultimate
concrete opportunities, Dundee, UK; 5–7 July 2005.
[5] Björnström J, Martinelli A, Matic A, Börjesson L, Panas I. Accelerating effects of
colloidal nano-silica for beneficial calcium–silicate–hydrate formation in
cement. Chem Phys Lett 2004;392(1–3):242–8.
[6] Qing Y, Zenan Z, Deyu K, Rongshen C. Influence of nano-SiO2 addition on
properties of hardened cement paste as compared with silica fume. Constr
Build Mater 2007;21(3):539–45.
[7] Collepardi M, Ogoumah Olagot JJ, Skarp U, Troli R. Influence of amorphous
colloidal silica on the properties of self-compacting concretes. In: Proceedings
of the international conference in concrete constructions – innovations and
developments in concrete materials and constructions, Dundee, Scotland, UK;
9–11 September 2002. p. 473–83.
[8] Li H, Zhang M, Ou J. Abrasion resistance of concrete containing nano-particles
for pavement. Wear 2006;260(11–12):1262–6.
[9] Jo BW, Kim CH, Tae G, Park JB. Characteristics of cement mortar with nano-SiO2
particles. Cement Concr Compos 2007;21(6):1351–5.
[10] Li G. Properties of high-volume fly ash concrete incorporating nano-SiO2.
Cement Concr Res 2004;34(6):1043–9.
[11] Shih Jeng-Ywan, Chang Ta-Peng, Hsiao Tien-Chin. Effect of nanosilica on
characterization of Portland cement composite. Mater Sci Eng 2006;424(1–
2):266–74.
[12] Jo BW, Kim CH, Tae G, Park JB. Characteristics of cement mortar with nano-
SiO2 particles. Constr Build Mater 2007;21(6):1351–5.
[13] Park CK, Nohb MH, Park TH. Rheological properties of cementitious materials
containing mineral admixtures. Cement Concr Res 2005;35(5):842–9.
[14] EN 197-1:2004. Cement: composition, specifications and conformity criteria
for common cements.
[15] EN 1015-3:2007. Methods of test for mortar for masonry. Part 3:
determination of consistency of fresh mortar (by flow table).
[16] EN 1015-7: 2007. Methods of test for mortar masonry – part 7: determination
of air content of fresh mortar.
[17] EN 196-3:2005. Methods of testing cement: determination of setting time and
soundness.
[18] Ramachandran VS, Beaudoin JJ. Handbook of analytical techniques in concrete
science and technology, techniques, and applications. New York: Noyes/
William Andrew; 2001.
[19] Ramachandram VS. Concrete admixtures handbooks, properties, science and
technology. New Jersey: Noyes; 1984.
[20] Taylor HFW. Cement chemistry. Academic Press; 1990.