Cement & Concrete Composites 45 (2014) 6981

Contents lists available at ScienceDirect

Cement & Concrete Composites

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

SCC modication by use of amorphous nano-silica

G. Quercia a,b,, P. Spiesz b, G. Hsken c, H.J.H. Brouwers b
a

Materials innovation institute (M2i), Delft, The Netherlands

Eindhoven University of Technology, Eindhoven, The Netherlands
c
BAM Federal Institute for Materials Research and Testing, Berlin, Germany
b

a r t i c l e

i n f o

Article history:
Received 29 June 2012
Received in revised form 22 July 2013
Accepted 3 September 2013
Available online 8 September 2013
Keywords:
Nano-silica
Self compacting concrete
Durability
Chloride
Freezethaw

a b s t r a c t
In this study two different types of nano-silica (nS) were applied in self-compacting concrete (SCC), both
having similar particle size distributions (PSD), but produced through two different processes: fumed
powder silica and precipitated silica in colloidal suspension. The inuence of nano-silica on SCC was
investigated with respect to the properties of concrete in fresh (workability) and hardened state
(mechanical properties and durability). Additionally, the densication of the microstructure of the hardened concrete was veried by SEM and EDS analyses. The obtained results demonstrate that nano-silica
efciently used in SCC can improve its mechanical properties and durability. Considering the reactivity of
the two applied nano-silicas, the colloidal type showed a higher reactivity at early age, which inuenced
the nal SCC properties.
2013 Elsevier Ltd. All rights reserved.

1. Introduction
Nowadays, the micro-level does not provide enough insights
into building materials. Therefore, all around the world, increasing
amounts of funding are being directed to research projects dealing
with material properties on the nano-level, which is claimed to
have a tremendous potential for the future [1]. The fundamental
processes that govern the properties of concrete are affected by
the performance of the material on nano-scale. The main hydration
product of cement-based materials, the CSH gel, is a nano-structured material [14]. The mechanical properties and the durability
of concrete mainly depend on the renement of the microstructure
of the hardened cement paste and the improvement of the pasteaggregate interface zone (ITZ) [5].
One of the most referred to and used cementitious nano-materials is amorphous silica with a particle size in the nano-range,
even though its application and effects on concrete have not been
fully understood yet. It has been reported that the nano-silica
addition enhances the compressive strength and reduces the
overall permeability of hardened concrete due to its pozzolanic
properties, which result in ner hydrated phases (CSH gel) and
densied microstructure (nano-ller and anti-Ca(OH)2-leaching
effect) [620]. These effects may enhance the durability of concrete
elements and structures.

Corresponding author at: Materials innovation institute (M2i), Delft, The

Netherlands.
E-mail address: [email protected] (G. Quercia).
0958-9465/$ - see front matter 2013 Elsevier Ltd. All rights reserved.
http://dx.doi.org/10.1016/j.cemconcomp.2013.09.001

There are different commercial types of nano-silica additives

available on the market, which are produced in different ways such
as precipitation, pyrolysis, solgel and others [19]. The main
characteristics of nano-silica, such as the particle size distribution,
specic density, specic surface area, pore structure and reactivity
(surface silanol groups) depend on the production method [620].
Despite the presence of several studies describing the main properties and characteristics of concrete containing nano-silica particles,
most of them focus on the application of nano-silica as an antibleeding [2028] and compressive strength enhancing additive
[12,13,2731]. Furthermore, the durability and sustainability of
concrete is becoming of vital importance for the construction
industry. In this context, SCC is a type of concrete that has
generated tremendous interest since its initial development in
Japan by Okamura [21].
SCC was developed to obtain durable concrete structures due to
an increased homogeneity. For this purpose, a concrete with low
yield stress, moderate viscosity and high resistance to segregation,
which also can be cast on-site or into precast molds without
compaction, was developed. The special ow behavior of SCC is
obtained by the use of superplasticizers, high amounts of ne
particles, and, in some cases, viscosity modifying agents (added
to reduce segregation and bleeding). Due to the presence of high
amounts of nes, the pore structure of SCC differs from the pore
structure of traditional concrete. According to Audenaert et al.
[22], the application of SCC might be somewhat risky due to the
lack of knowledge concerning the actual durability of this material.
So far, only a few reports on the effects of nano-silica on the
durability of SCC are available [27,29,31]. In addition, the

70

G. Quercia et al. / Cement & Concrete Composites 45 (2014) 6981

difference in the reactivity of nano-silica due to its production

route has not been reported yet.
In this study, two different types of nano-silica were applied in
SCC, both having similar particle size distributions (PSD), but produced through two different processes: fumed powder silica and
precipitated silica in colloidal suspension. The inuence of the
nano-silica on SCC was investigated with respect to the properties
of the concrete in fresh (workability) and hardened state (mechanical properties and durability indicators). Additionally, the densication of the microstructure of the hardened concrete was
analyzed by SEM and EDS techniques.

2. Materials and methods

2.1. Materials and SCC mix design
The Portland cement used was CEM I 42.5N (ENCI, The Netherlands), as classied by [32]. This cement consists of at least 95%
Portland cement clinker; the initial setting time is 60 min, a water
demand of 38.9% by weight, and a compressive strength of
21 3 N/mm2 after 2 days and 51 4 N/mm2 after 28 days, respectively, are obtained [32]. The coarse aggregates used were composed of broken granite in the fractions 28 mm and 816 mm.
Two different sands were used: dredged river sand (04 mm)
and microsand (01 mm). The microsand (Granite Import B.V.,
the Netherlands) is mainly composed of natural sandstone waste
that is generated during the crushing process of coarser fractions.
A ground limestone powder was applied as ller. Two different
commercial nano-silica additives were selected to produce two different SCC batches: one colloidal nano-silica suspension and one
fumed powder nano-silica. Both nano-silica additives have similar
PSD and specic surface area measured by BET (BrunauerEmmet
Teller method [33]), following the standard DIN-ISO 9277-2005
[34]. Furthermore, a superplasticizer (SP) based on polycarboxylate
ethers was added to adjust the workability of the mix. A summary
of the general characteristics of all materials used in this study is
presented in Table 1 and their PSDs are depicted in Fig. 1.
For the composition of SCC mixes, the mix design concept
described in [35] was used. This design concept makes use of an
optimization algorithm described in [36] to compute the mix proportions of all solid ingredients of the concrete mix, following the
theory of continuously graded granular mixtures. In the optimization process, a distribution modulus (q) of 0.25 was used together
with the following constrains: cement content of 340 kg/m3, w/c
ratio of 0.45 and air content in the fresh mix of 1% by volume.
The cement content and the w/c ratio were selected based on
NEN-EN 206-1 (2008) [37] for the exposure class XS3 (aggressive
exposure to chlorides originating from seawater). In addition, a
ow class of the fresh concrete F7 (630800 mm) was selected as
a target, taking into consideration the Dutch recommendation

Fig. 1. PSD of the used materials, target function based on [36] and composition of
reference mix (dashed line) for Dmin = 60 nm.

BRL 1801 [38] for SCC. An example of the target curve and the composed grading curve of the SCC mix is shown in Fig. 1. In Table 2 the
nal mix proportioning and characteristics of the reference mix
without nano-silica and two mixes with nano-silica addition are
presented.
2.2. Test methods
2.2.1. Fresh concrete properties
Using the provided materials, three SCC mixes (2 batches of
65 dm3 for each mix) were prepared according to the mix proportions listed in Table 2. These SCCs were mixed for 5 min in total in a
compulsory mixer, and subsequently tested for their fresh properties, according to the procedure recommended by the EFNARC
committee [39]. The following fresh-concrete properties of the
SCC were determined: V-funnel time, slump ow, density, packing
density and air content.
2.2.2. Hardened concrete: mechanical properties
To test the properties of the hardened concrete, thirty-eight
cubes (150 mm side length) were cast for each mix, cured sealed
during the rst day, stripped from the mold after 24 h and cured
subsequently in water until their test age was reached, according
to the prescribed storing conditions given by BS-EN 12390-2
[40]. The compressive strength was determined after 1, 3, 7, 28
and 91 days on three cubes for each mix. The compressive strength
test was performed according to BS-EN 12390-3 [41]. In addition,
the splitting tensile strength was determined after 28 days on
three cubes, following the procedure given in BS-EN 12390-6 [42].

Table 1
Properties of the used materials.

a
b

Materials

Specic density (g/cm3)

BET (m2/g)

pH

Solid content (% w/w)

Loss on ignition (LOI)

Computed SSAb (m2/m3)

CEM I 42.5N
Colloidal nano-silica
Powder nano-silica
Limestone powder
Microsand (sandstone)
Sand 04
Granite 28
Granite 816
Superplasticizer

3.14
1.40
2.15
2.71
2.64
2.64
2.65
2.65
1.10

1
50
56

9.5
5.0a

50

35

2.8

0.5

1,699,093
46,110,081
48,175,461
1,234,362
193,514
14,251
1740
515

7.0

4% w/w in water.
SSA is the specic surface area assuming spherical shape for nano-silica and the angularity factor of every other used granular material, following [35].

G. Quercia et al. / Cement & Concrete Composites 45 (2014) 6981

Table 2
Composition and characteristics of the three designed SCC mixes.
Materials

Reference
(kg/m3)

Colloidal nS
(kg/m3)

Powder nS
(kg/m3)

CEM I 42.5N
Nano-silica
Limestone powder
Microsand (sandstone)
Sand 04
Granite 28
Granite 816
Water
SP

340.0
0.0
179.4
125.0
624.3
733.8
274.7
153.0
3.4

340.0
12.8
151.8
141.3
617.9
735.6
274.2
153.0
6.5

340.1
12.8
151.9
141.4
618.0
735.7
274.3
153.0
6.5

Air (%V)-estimated
Density (g/cm3)
w/c
w/p
Powder content (dm3/m3)
Composed surface (m2/m3)
SP content (g/m2)
SP content (% bwoc)

1.0
2.427
0.45
0.267
194.2
277,972
0.0122
1.0

1.0
2.427
0.45
0.270
192.7
547,905
0.0119
1.9

1.0
2.430
0.45
0.270
192.6
554,428
0.0117
1.9

bwoc: based on the weight of cement.

2.2.3. Hardened concrete: durability tests

2.2.3.1. Permeable (water accessible) porosity. The permeable porosity affects the transport properties of uids in concrete and therefore also its durability. It is related to many deterioration processes
driven by the transport properties of concrete [43]. In this context,
three cores (diameter of 100 mm, height of 150 mm) were drilled
with a diamond coring bit from three cubes for each mix. Afterwards, using a diamond saw at medium speed lubricated with
water, six concrete discs (height of approximately 15 mm) for each
SCC mix were cut from the inner layers of the drilled cores. No polishing of the surfaces was performed. In total 18 discs were used to
determine the permeable porosity, following the procedure described in the ASTM 1202 [44]. The vacuum-saturation technique
was applied to ll the accessible pores with water, as this technique is concluded to be the most efcient by Sauddin and Heran
[43].
2.2.3.2. Penetration of water under pressure. The depth of penetration of water under pressure was tested according to BS-EN
12390-8 [45] at the age of 28 days. The samples (three cubes for
the reference and two for each mix with nano-silica) were exposed
to water under pressure (5 bar) for 72 h and subsequently split in
order to determine the maximum depth of the water penetration.
2.2.3.3. Pore size distribution and porosity of the paste. The pore size
distribution was measured using the mercury intrusion porosimetry (MIP) technique (Autopore IV, Micromeretics). The maximum
applied pressure of mercury was 228 MPa, the mercury contact angle was 130 and the equilibrium time was 20 s. The pore size
range of 0.0063900 lm was investigated. Pieces of hardened
mortar (approximately 2 grams in total) were carefully selected
to avoid the presence of coarse sand and gravel. The mortars, extracted from cured SCC samples at 28 days, were rst dried in an
oven (80 C) until a constant mass was reached, and then tested
by MIP technique.
2.2.3.4. Rapid chloride migration test (RCM) and conductivity test.
Two specimens for the RCM test were retrieved from each extracted core, as described in Section 2.2.3.1, giving in total six test
specimens (discs, diameter of 100 mm and height of 50 mm) for
each mix. Three of these specimens were tested at the age of
28 days and the remaining three at 91 days. One day prior to the
RCM test, the specimens were pre-conditioned (vacuum-saturation
with limewater). The RCM test was performed according to NT

71

Build 492 [46], using the test set-up described in [47]. The duration
of the RCM test for all samples was 24 h. After the test, the penetration depth of chlorides was measured on split samples by applying a colourimetric indicator for chlorides (0.1 M AgNO3 solution)
and the values of the chloride migration coefcients (DRCM) were
calculated according to [46].
Before the RCM test, the electrical resistance was measured on
the same saturated samples by using the so-called two electrodes
method [48]. For this, an AC test signal (f = 1 kHz) was applied between two stainless-steel electrodes and the resistance of the concrete sample placed between the electrodes was registered. Finally,
the conductivity of the samples was calculated taking into account
their thicknesses and transversal areas.
2.2.3.5. Chloride diffusion test. As the addition of nano-silica
changes the ionic strength, the pH and the conductivity of the pore
solution [49], the results obtained using the RCM test may be inuenced, as the procedure of this test is based only on experience
with OPC systems [50]. At the same time, the natural chloride diffusion test is only affected by the pore structure (permeability and
tortuosity of the pores), chloride binding and chloride concentration gradient. Thus, the chloride diffusion test may be more reliable
for SCC with nano-silica addition than the RCM test.
For each prepared mix, three specimens (cylinders, diameter of
100 mm and height of 50 mm) were extracted from different cubes
by drilling and then cutting. The diffusion test began 28 days after
casting the concrete, following the procedure described in [51].
Prior to the test, all external faces of the specimens were coated
with an epoxy resin except for one at surface, left uncovered to
allow the chlorides to penetrate the samples just from that side.
Then, the specimens were immersed in a sodium chloride solution
(concentration of 165 g/dm3) for 63 days at room temperature in a
sealed and de-aired container with the uncoated surface on top.
After the exposure period, one specimen from each test series
was split in order to measure the penetration depth of chlorides
(using 0.1 M AgNO3 solution as a colourimetric chloride indicator).
The remaining samples were dry-ground in layers for determining
chloride concentration proles. The grinding was performed on an
area of 73 mm in diameter using the Prole Grinder 1100 (Germann Instruments). The obtained powder was collected for determining chloride proles, following the procedure described in [52].
An automatic potentiometric titration unit was used for the Cl
concentration measurements, applying a 0.01 M AgNO3 solution
as titrant. The obtained chloride concentration values were t to
the solution of Ficks 2nd law, in order to estimate the apparent
chloride diffusion coefcient (Dapp) and the surface chloride
concentration.
2.2.3.6. Freezethaw resistance (surface scaling test). As a further
durability assessment, the freezethaw test was performed on
SCC samples, even though the air content in the fresh mix was less
than the recommended value of 4% by volume [37]. The freeze
thaw resistance, expressed by the surface scaling factor (Sn), was
determined following NEN-EN 12390-9 [53]. Nevertheless, the test
samples differed from the specications in the standard. For practical reasons, cylinders were used instead of slabs. The 150 mm
cubes were cured in water after demolding until the age of 14 days,
when the cores (100 mm in diameter) were extracted and sliced
(two cylinders of 50 mm in height were obtained from each core).
Afterwards, the obtained cylinders were cured under water and
then sealed after 25 days with tight rubber sleeves. The sealed
samples were placed in polyurethane insulations of 10 mm thickness and surface-saturated with demineralized water for 3 days.
Due to a limited volume of the used climate chamber, only three
specimens were tested for each mix, resulting in a total exposed
surface area of 0.024 m2 (the area recommended in [53] is

72

G. Quercia et al. / Cement & Concrete Composites 45 (2014) 6981

0.08 m2). After the saturation, the demineralized water was replaced by a 3 mm layer of 3% by weight NaCl solution, poured on
the top surfaces of the samples and then the freezethaw cycles
were started. The applied temperature prole was following the
recommendations given in NEN-EN 12390-9 [53]. The level of solution on the surface of concrete was adjusted regularly. In total, 56
freezethaw cycles were performed, during which the surface scaling was measured after 7, 14, 28, 42 and 56 cycles.

as a typical characteristic of SCC with increased amount of limestone powder [5658].

As also can be seen in Table 3, the mixes with nano-silica have
higher air content compared to the reference mix, which is caused
by the higher viscosity of the paste due to the application of particles with high specic surface area (SSA). The assumed air entrainment was also conrmed by comparing the difference between the
designed and the measured concrete density (see Tables 2 and 3).

2.2.4. Microstructural characterization and analysis

The microstructural morphology of the prepared concrete was
analyzed using a high resolution scanning electron microscope
(FEI Quanta 600 FEG-SEM) with a Schottky eld emitter gun (at
voltage of 10 keV and 0.6 mbar of low-vacuum pressure). Furthermore, a general chemical analysis was performed using EDAX energy dispersive spectroscopy (EDS) detector. Several fracture
surfaces were investigated in natural conditions (without any sample preparation). The fracture surfaces of the SCC samples were
analyzed in this study to avoid the effect of sample preparation
(cutting, grinding, polishing, etc.), which may alter the microstructure of concrete. In the analyses of the fracture surfaces it is possible to characterize the interfacial transition zone (ITZ) and the
formation of ettringite.

3.2. Hardened concrete: mechanical properties

3. Results and discussion

3.1. Fresh concrete properties
The fresh concrete properties of the tested mixes are presented
in Table 3. All three mixes fulll the requirements for the ow class
F7 (630800 mm of spread diameter), specied in BRL 1801 [38].
Only the mix with colloidal nano-silica resulted in a spread ow
close to the lower limit of this target range. Another interesting
nding is that the SP content, based on the composed surface area,
is nearly constant for all mixes and amounts to an average value of
0.0119 g/m2. This shows that in the present case the required
amount of SP can be calculated if the composed surface area of
all solid ingredients is known. Nevertheless, the SP dosage depends
mainly on the mineralogy and the surface charge of the ne particles [54,55]. Another implication of this result is that the SP
requirement is rather related to changes in the specic surface area
of the mix than to the concentration (by mass) of the nano-silica
addition, as it was previously reported by Sobolev et al. [17]. These
researchers concluded that 0.21% of additional SP is needed for
each 1% of nano-silica added to a standard concrete composition.
Considering the V-funnel time, only the two mixes with nanosilica fulll the requirements of the viscosity class VF2 (funnel time
925 s) as established by the European guidelines on SCC [39].
Although the reference mix does not fulll this viscosity class, it
is in the range for a high powder content SCC mix, as proposed
by Hunger [35]. In addition, no segregation or blocking was observed for any mixes. A long V-funnel time was already reported

3.2.1. Compressive strength

The 1-day compressive strength of the reference mix was
higher than the strength of mixes with nano-silica (Fig. 2a). At this
age, the mix with powder-type nano-silica showed the lowest
compressive strength. This behavior can be related to the difference in the reactivity of both nano-silica used. The colloidal
nano-silica is synthesized at a low temperature and has a higher
concentration of silanol groups on its surface (around 5.3 SiOH/
nm2), which increases its reactivity [59,60]. On the contrary, the
pyrogenic type of nano-silica has a concentration of silanol groups
on its surface in the range of 1.22.5 SiOH/nm2 [59]. It was demonstrated in [9] that the reactivity of nano-silica is controlled by its
specic surface area and the amount of Q2 3 (silanol) SiOH
groups, which cover the nano-particle. These groups constitute
condensation sites for monomeric silica units released from the
clinker phases and, consequently, the setting of the cement paste
is accelerated. Some researchers reported a strong and instantaneous interaction between nano-silica (independently from its
state: colloidal, powder or in slurry) and some chemical species
dissolved in the liquid phase of fresh cement pastes, leading to

Table 3
Properties of the prepared SCC mixes in fresh state.

a
b

Value

Reference

Colloidal nS

Powder nS

Slump ow (mm)b
V-funnel time (s)
Fresh density (g/cm3)
Air content (%V)a
Packing density (%)a

690720
35.0
2.399
1.15
83.55

664701
20.5
2.384
1.79
82.91

685720
24.5
2.392
1.58
83.12

Calculated value.
Maximum and minimum ow diameter measured.

Fig. 2. Mechanical properties of the tested SCC mixes, (a) compressive strength
development and (b) 28 and 91 days splitting tensile strength.

G. Quercia et al. / Cement & Concrete Composites 45 (2014) 6981

the formation of a destabilizing gel with high water retention

capacity (in form of bound water) [6163]. Even in a well-dispersed colloidal dispersion, the nano-particles still exist as clusters
(agglomerates) when incorporated into the highly alkaline environment such as the cement pastes [64]. By using colloidal silica,
it is assumed that the mono-dispersed nano-silica can act as ller
and seed to form CSH gel much more effectively than the
agglomerates of silica particles generated from pyrogenic silica
powders. Nevertheless, Kong et al. [64] revealed that the colloidal
silica reacts to form a gel or coagulates immediately when the cement is mixed in water containing sol due to the rapid increase of
ionic strength in the paste, the rising pH and the adsorption of Ca2+
ions on the silica particles. As a result, no matter which source of
nano-silica is used, it is the behavior of the nal agglomerates,
rather than that of the individual nano-particles, which controls
the owability, the lling ability, the pozzolanic and acceleration
effects on the cement hydration and the improvement of the nal
concrete microstructure.
During the development of strength over time, the mix with
colloidal nano-silica showed the highest increase in the compressive strength from 3 until 91 days. An interesting observation is
that at 7 days, the mix with powder nano-silica shows similar compressive strength as the reference mix, as the result of a pozzolanic
reaction with Ca(OH)2 generated during the OPC hydration. At
91 days, the powder nano-silica developed a higher compressive
strength than the reference SCC. In the case of the SCC with colloidal nano-silica, the 91 days compressive strength was almost the
same as at 28 days. This fact is mainly caused by the higher reactivity of that silica, so in turn it was consumed earlier. It can be deduced from various investigations [23,26,29,65] that concretes
made with nano-silica have a ner pore structure, with a lower
proportion of fairly coarse capillary pores. For the same composition, these concretes are, therefore, generally denser than pure
Portland cement concretes. One reason for this is a lower content
of coarse Ca(OH)2 crystals in the matrix and the aggregate surface
zones of these concretes. A shift of the pore structure towards ner
pores as a result of the pozzolanic reaction, is also observed when
nano-silica is used in concrete [6,13,2326,65]. The reaction of the
coarse Ca(OH)2 crystals to form microcrystalline CSH phases is
recognized to be the cause of this densication of the microstructure. In addition, several researchers reported that concretes with
very ne pozzolans, such as nano-silica, have signicantly higher
impermeability than conventional concretes. This is attributed in
particular to a considerable improvement of the aggregates interfacial transition zone (ITZ) [6,65,66]. In normal concretes there is
an increased concentration of well crystallized Ca(OH)2 and
ettringite needles aligned uniformly towards the aggregate surface
in the ITZ, which increases the porosity when compared to the surrounding matrix. Pozzolans that are substantially ner than cement improve the packing of particles at the aggregate surface
from the outset, which also favors the pozzolanic reaction in this
zone, i.e. the formation of denser and high stiffness CSH phases
[6,13,67]. The changes in the microstructure in the presence of
both types of nano-silica are responsible for the slightly increased
compressive strength (about 10% higher) of SCC. A further explanation and support is presented in Section 3.4.
An important factor to take into account is that the 28-day compressive strength results are statistically close between all the
mixes. The average compressive strength values for all SCCs lay
within the error band of the measurements. The error (standard
deviation) in the compressive strength results is higher for the
SCC with colloidal nano-silica additions (9.4 N/mm2), including
the 91-days strength standard deviation (6.5 N/mm2). The
standard deviations for the reference and powder SCC amount to
1.6 and 4.5 N/mm2, respectively. The standard deviations for
the nano-silica SCC are higher than the recommended range

73

(3.54.8 N/mm2) for high strength concrete according to [68]. This

could be caused by the increased reactivity and the higher viscosity
of the paste, due to the fact that colloidal silica leads to an inhomogeneous dispersion of the mix components. From the industrial
point of view it would be a factor to consider in the application
of colloidal nano-silica in the future. Higher standard deviation of
the 28-day compressive strength should produce SCC mix with
poor standard of control classication in real conditions [68],
based on the fact that higher standard deviations are expected in
the eld than in the laboratory conditions.

3.2.2. Tensile splitting strength

The determined average 28 days tensile splitting strength of the
reference mix was 4.51 N/mm2. The equivalent values for the
mixes with colloidal and powder nano-silica are higher and
amount to 4.92 N/mm2 and 5.48 N/mm2, respectively (Fig. 2b). A
possible explanation for the higher tensile splitting strength of
the samples containing nano-silica is that the bond between the
hardened paste and the aggregates is stronger. The nano-silica improves the quality of the interfacial zone (ITZ) due to precipitation
of smaller and stronger (higher stiffness) CSH gel and an accelerated rate of hydration of the paste, as it is also reported by several
researchers [620,6567]. Similar trends in the tensile splitting
strength at 91 days were observed. Further data are presented in
Section 3.4. The distinctions on the tensile splitting strength
produced by the colloidal nano-silica and the powder nano-silica
originate from their different reactivities. As the powder nano-silica is less reactive at early age, a better wet packing or lling of the
space between the aggregates is obtained due to a lower viscosity
of the paste. Then, with the age and the progress of the pozzolanic
reaction, the ITZ is densied, improving the bonding between the
matrix and the aggregates. Nevertheless, similarly to the compressive strength results, the average tensile splinting strength values
of the SCCs with both types of nano-silica are within the error band
of the measurements. It is only evident at 91 days that the compressive and tensile splitting strength of the SCC are improved by
1318% due to the use of both types of nano-silica. In the case of
the standard deviation values obtained for all the SCC in the tensile
splitting strength, all of them are lower than the maximum recommended value of 0.40 N/mm2 to be classied as a good standard
control according to [68].

3.3. Hardened concrete: durability

3.3.1. Permeable porosity
The results of the measurements of the permeable porosity of
the SCC mixes are presented in Table 4. These results are surprisingly showing that the SCC reference mix has slightly lower total
porosity (12.1%) compared to the mixes containing the two types
of nano-silica (12.5%). The porosity and the tortuosity of the pores
in the hardened cement paste are normally reduced when pozzolanic materials are added, and this inuences many properties
such as the compressive and splitting tensile strength [69]. Nevertheless, Yogendran and Langan [70] stated that for the addition
of micro-silica the total pore volume is not necessarily changed,
but larger pores appear to be subdivided into smaller pores.
Apparently, the same behavior was found for the SCC in the present case. The air content of the fresh mix inuenced, probably,
the nal porosity. Some researchers [43] state that the vacuumsaturation technique is also able to take into account the volume
of the entrapped air voids. This means that the porosity values
shown Table 4 consider also the differences in the air content
of the mixes (the lower the air content the lower the permeable
porosity).

74

G. Quercia et al. / Cement & Concrete Composites 45 (2014) 6981

Table 4
28-day permeable porosities and penetration depths of water under pressure of the
three designed SCC mixes.

Properties

Reference

Colloidal nS

Powder nS

Permeable porosity (% vol.)

Penetration depth of WUPa
(mm)

12.07 0.17
26 7

12.45 0.26
32

12.48 0.23
32

Table 5
Properties of the hardened matrix (mortar) of SCC mixes obtained from MIP
measurements.

WUO: water under pressure (5 bar).

3.3.2. Penetration of water under pressure

The test results of the penetration of water under pressure are
shown in Table 4. According to [29], all tested samples are in the
low permeability range (penetration depth of less than 30 mm).
Additionally, the two SCC with both types of nano-silica presented
penetration depth lower than 5 mm. This implies that the addition
of 3.8% nano-silica results in concrete which is highly resistant to
the penetration of water under pressure of 0.5 MPa (5 bar). Similar
permeability improvements were reported by Raiess-Ghasemi
et al. [29] for conventional concrete with micro and nano-silica
addition. The results also suggest that, despite the similar permeable porosity of all SCC mixes, the nano-silica samples have very
low effective water permeability (less interconnected pores and/
or ner pore structure). A similar phenomenon was already reported by Yogendran and Langan [70] for HPC with micro-silica
addition, where the total porosity was not affected, but the permeability decreased by one order of magnitude.
3.3.3. Pore size distribution and porosity of the concrete matrix
To support the ndings presented in Sections 3.3.1 and 3.3.2 the
hardened mortar (matrix) of the concrete samples was analyzed

Fig. 3. Mercury intrusion porosimetry results of hardened paste extracted from the
tested SCC mixes, (a) cumulative intrusion/extrusion vs. pore size curves, (b) log
differential intrusion vs. pore size curves. Black arrows in the gures indicate
changes or displacement of the curves due to the addition of nano-silica.

Properties

Reference

Colloidal
nS

Powder
nS

Median pore diameter (volume) (nm)

Median pore diameter (area) (nm)
Average pore diameter (4 V/A) (nm)
Bulk density (at 0.1 MPa) (g/ml)
Apparent (skeletal) densitya (g/ml)
Porosity (%)

34.8
20.3
27.8
2.216
2.415
8.79

30.9
17.2
24.3
2.222
2.438
9.31

33.8
17.5
25.9
2.239
2.445
8.99

Determined excluding the Hg accessible pores.

using mercury intrusion porosimetry (MIP). The obtained results

are shown in Fig. 3 and Table 5. In Fig. 3a and b it is possible to notice that the addition of 3.8% of nano-silica slightly increases the
volume of pores lower than 20 nm (indicated by black arrows in
Fig. 3a and b). Similar trends were obtained for the overall parameters that were extracted from the mercury intrusion test. These
parameters are shown in Table 5. It is clearly shown in this table
that the addition of nano-silica decreased the median pore diameter (by volume and area) and the average pore diameter, which is
reduced from 27.8 nm (for the reference) to 24.3 and 25.9 nm for
the colloidal nano-silica and powder nano-silica, respectively.
Apparently, a reduction of 4 nm in the average pore diameter
was enough to signicantly reduce the permeability of the SCC
containing nano-silica. According to Mindess et al. [71], the pores
in cement paste are divided into large capillary pores (10
0.05 lm), medium capillary pores (5010 nm) and gel pores
(<10 nm). This means that the modication of the SCC pore structure due to nano-silica addition is reected by the changes in the
medium capillary pores and the gel pores (the minimum pore size
detected with MIP was 8 nm). Permeability and penetration of
harmful substances into concrete are affected mainly by the large
and medium capillary pores [68]. Apparently, nano-silica made
the pore structure of paste more homogeneous by increasing medium capillary porosities. In similar way, the gel pores volume was
increased (see Fig. 4). The increased gel porosity can be caused by
an increased amount of CSH gel in the paste (acceleration effect
of the nano-silica [9]). The increased gel porosity could result from
a higher hydration degree and from the properties of the gel produced by the pozzolanic reaction of the nano-silica particles. Pozzolanic CSH gels, normally, have a lower Ca/Si ratio (10.8)
and different gel porosities [72]. Similar changes in the pore structure of blended cement with slag and nano-silica addition were
reported in [66].
Another interesting fact, also presented in Table 5, is that the
apparent (skeletal) density of the hardened mortar was higher
for the SCC with nano-silica (both types). This demonstrates that
the addition of nano-silica promotes the densication of the cement matrix. Despite this densication, the apparent porosity of
paste increases with the addition of nano-silica (see Table 5), similarly to the results obtained for the permeable porosity of SCC.
This means that other factors are improving the resistance to water
penetration of the SCC with nano-silica. These factors can be e.g.
changes of the shape of pores (cylindrical vs. bottle-neck type),
changes of the tortuosity or pores interconnectivity and an
improvement of the ITZ of cement paste/aggregates. During the
intrusion cycles, the complete pore space is lled with mercury.
When the pressure is released, the mercury is sucked out of the
pore space except for the ink-bottle and dead-end pores. Isolated
(non-accessible) pores do not contribute to the transport properties of uids and cannot be detected by MIP. Nevertheless, taking
into account the differences between the intrusion and the extrusion cycles of mercury (Fig. 3a) and the denition of effective

G. Quercia et al. / Cement & Concrete Composites 45 (2014) 6981

Fig. 4. 28 and 91 days test results: (a) conductivity and, (b) chloride migration
coefcient (DRCM).

porosity and ink-bottle porosity dened in [73], it is possible to

observe that the addition of nano-silica increased the amount (volume) of ink-bottle pores. The increment is around 7% (+0.0028 ml/
g) for both types of nano-silicas. The transport of species through
the ink-bottle pores is more difcult due to their constrictivity.
These differences can explain the changes observed in the pores
structure of the cement pastes and can partially explain the results
of water pressure penetration test. Similar results were reported by
Shih et al. [74] who found a small difference in the porosity (2.84%
in the volume of pores of cement pastes with nano-silica compared
to pure cement pastes), but they demonstrated that nano-silica
addition of 0.8% (based on the weight of cement) caused that the
microstructure of hardened cement becomes denser in the regions
where the pore size was larger than 10 nm, but becomes looser at
regions where the pore size is smaller than 10 nm. Even though,
the performed MIP analysis gives some evidence of the changes
in the concrete pore structure, the results are close to the repeatability band of the test. For that reason, complementary pore size
distribution analyses and other characterization techniques
enabling to detect changes in the pore structure are recommended
in further research.
3.3.4. Conductivity
Fig. 4a shows the average values of the conductivity, measured
on cylindrical SCC samples that were extracted from the cast cubes.
It is clearly shown that the conductivity of the SCC with nano-silica
addition is reduced by more than 50% compared to the SCC reference mix. Meanwhile, the SCC with the colloidal nano-silica presented a slightly lower conductivity than samples with the
powder nano-silica. This behavior is an indication of the ability
of the water-saturated pore structure of the concrete to transport
electrical charge. Different authors [7577] established that the
conductivity is directly related to the porosity, the pore structure

75

(tortuosity, connectivity and conductivity) and to the pH of the

pore solution (the pH in the presence of amorphous silica is decreased, as was established in [47]). In general, more pore water
as well as more and larger pores with higher degree of connectivity
and lower tortuosity results in higher electrical conductivity values
[48]. The lower conductivity values shown by the SCC with nanosilica are the result of the pore structure renement (less
connected pores) due to the progressive pozzolanic reaction and
higher hydration degree (see the microstructural analysis section).
The small difference between the two types of silica is mainly
related to the higher reactivity of the colloidal nano-silica, which
promotes a more compact and ner microstructure (higher
stiffness CSH gel) than the powder nano-silica. Nevertheless,
the values are in line with the results of the penetration of water
under pressure and the determined compressive strength. Conductivity or its inverse value (resistivity) can also be related to the
compressive strength as it was demonstrated by Andrade et al.
[76]. These authors concluded that the electrical resistivity is a
suitable test for providing the estimated compressive strength
with a high level of reliability. The resistivity is related with the
compressive strength because it reects the ability of the porous
medium to transfer the electrical charge thought the aqueous
phase. This property promotes the satisfactory relation between
the resistivity, porosity and connectivity of pores in saturated
concrete. Due to the renement of the pore structure (progressive
hydration during time), the resistivity increases over time. It was
demonstrated [76] that this evolution is very similar to the
development of the compressive strength in time, and a linear
relationship between the compressive strength and the resistivity
was proposed. Nevertheless, the improvements in the compressive
strength of the SCCs were much lower than the improvement
obtained in the conductivity test. This is probably due to the fact
that the total pore volume of the SCC is not reduced by the pozzolanic reaction promoted by the nano-silica additions, but the pore
structure becomes more discrete. In addition, the electrical conductivity of concrete depends on the pore solution chemistry and
chemical binding of various ions by the hydration products [78],
whereas these phenomena do not affect the strength. The use of
nano-silica dilutes the pore solution and increases the binding of
different ions such Na+ and K+ by the reaction products. Consequently, the pH of the pore solution is reduced (OH concentration
drop). Because the important role of hydroxyl ions in the conductivity of concrete, the reduction of the pH of the pore solution of
concrete due to the use of supplementary cementitious materials
such as nano-silica can increase the electrical resistance of concrete [78].
3.3.5. Rapid chloride migration test (RCM)
The Rapid chloride migration test (performed according to NT
Build 492 [46]) is a commonly used accelerated technique for
determining the chloride transport rate in concrete. The output
of the test the so-called chloride migration coefcient
DRCM can be employed in service-life design models for concrete
elements and structures exposed to chlorides. In this context,
Fig. 4b presents the average values of the calculated chloride
migration coefcients (DRCM) of each SCC mix studied. Like the conductivity test results, the migration coefcients are much lower for
the mixes containing nano-silica. Additionally, the SCC mix with
colloidal nano-silica shows again the best performance. The explanation of this behavior is the same as previously discussed for the
conductivity test results. A ner porosity, greater tortuosity and
more precipitated CSH gel reduce the ingress speed of chlorides
into the concrete. It can also be noticed that the 91 days DRCM values are lower compared to the 28 days DRCM. This can be explained
by the progressing hydration of cement and a related densication
of the microstructure of concrete.

76

G. Quercia et al. / Cement & Concrete Composites 45 (2014) 6981

Table 6
Maximum values of 28-days DRCM coefcients for 100 years of service-life design of concrete, taken from [72].
Minimum concrete cover
depth (mm)

Maximum value of DRCM (10

Steel

Pre-stressed steel

CEM I

35
40
45
50
55
60

45
50
55
60
65
70

12

m2/s)
CEM I + CEM III 2550% GGBS

CEM III 5080% GGBS

CEM II B/V + CEM I 2030% y

ash

XD1, XD2, XD3, XS1

XS2, XS3

XD1, XD2, XD3, XS1

XS2, XS3

XD1, XD2, XD3, XS1

XS2, XS3

XD1, XD2, XD3, XS1

XS2, XS3

3.0
5.5
8.5
12
17
22

1.5
2.0
3.5
5.0
7.0
9.0

2.0
4.0
6.0
9.0
12
16

1.0
1.5
2.5
3.5
5.0
6.5

2.0
4.0
6.0
8.5
12
15

1.0
1.5
2.5
3.6
5.0
6.5

6.5
12
18
26
36
47

5.5
10
15
22
30
39

The 28-day DRCM can be employed in service-life design models

[79,80] for concrete elements and structures exposed to chlorides.
When comparing all the obtained 28-day DRCM to the values suggested in the CUR Durability Guideline [80] for 100 years of service
life (see Table 6), the SCC reference mix is out of the range of the
aimed exposure class XS3. In contrast, both SCC mixes with
nano-silica addition satisfy the exposure class XS3 when a concrete
cover depth of 50 mm is provided. This demonstrates how a relatively small addition of the nano-silica (3.8% based on the weight
of cement) can effectively increase the resistance of concrete to
the intrusion of chlorides, and opens, therefore, a possibility for
reducing the thickness of the concrete cover, based on the values
suggested in [80], which are shown in Table 6. On the other hand,
comparing the obtained DRCM values to similar SCC mixes published in literature, the values obtained for the reference mix are
in line with SCC mixes having high amount of limestone powder
(between 8 and 12  10 12 m2/s at 28 days [22]). The values
obtained for the SCC mixes with nano-silica are comparable to
reported values of SCC composed of slag cement or y ash with
similar w/b ratio and cement content (between 4 and
5  10 12 m2/s [22]).
3.3.6. Chloride diffusion test
In Fig. 5a the obtained apparent chloride diffusion coefcients
(Dapp) of the three SCC mixes are shown. A trend similar to the
DRCM was obtained for the diffusion test. The largest Dapp was computed for the reference SCC, 9.61  10 12 m2/s. The Dapp of
4.45  10 12 m2/s and 3.55  10 12 m2/s were obtained for powder
nano-silica and colloidal nano-silica, respectively. Even though the
chemical balance of the pore solution is changed due to the
addition of nano-silica and the dominating chloride transport
mechanisms in both methods are different (migration due to the
electrical eld and diffusion due to the concentration gradient),
the obtained Dapp and DRCM have comparable values for each tested
SCC.
3.3.7. Freezethaw resistance
The results of the freezethaw surface scaling of the three
selected SCC mixes are shown in Fig. 5b. The failure of the SCC
reference mix, taking into account the maximum scaling criteria
of 1.5 kg/m2 at 28 cycles as recommended in [81], occurred about
the 11th cycle. On the other hand, the SCC mixes with nano-silica
addition resulted in a surface scaling factor lower than the recommended value for non-air entrained concrete after 28 cycles. These
SCC mixes failed the 1.5 kg/m2 criteria after 48 cycles. The freeze
thaw resistance depends on the compressive strength, porosity, air
void content and other parameters such as the air-voids distribution and pore size [68]. A better resistance to the freezethaw
induced damage of the SCC with nano-silica addition can be attributed to different factors, such as the formation of highly stiff CSH

Fig. 5. (a) Chloride diffusion proles of the tested SCC mixes, (b) cumulative scaling
factor (Sn) after 56 freezethaw cycles (3% NaCl solution).

gel (with E modules of 26 GPa reported by [67]) and the rened

pore structure (higher tortuosity and constrictivity of the pores),
which results in a limited intrusion of water and improved
resistance to the changes of temperature near the surface of the
concrete. Despite the better freezethaw resistance of the SCC with
nano-silica compared to the reference mix, its scaling values are
larger than the recommended value of 0.5 kg/m2 after 56 cycles,
suggested by Stark and Wicht [82] for a concrete classied as having good resistance against freezethaw exposure. Nevertheless,
with an air entrainment admixture that guarantees a minimum
air content of 4%, the freezethaw resistance of SCC with nano-silica should result in a mix having a high resistance to freezethaw.
3.4. Microstructural analysis
The objective of the microstructural analysis is to support the
ndings shown in the previous sections. It is important to notice,

G. Quercia et al. / Cement & Concrete Composites 45 (2014) 6981

that the performed microstructural analyses were qualitative, with

the objective to give additional information that can explain the results obtained in the other executed tests. In this context, Fig. 6
shows some selected morphological characteristics of the microstructure of the SCC reference mix. The SCC reference mix has an
apparent dense structure and a relatively good ITZ (Fig. 6a). A good
ITZ is dened in the present investigation as the zone (1020 lm
long) in the vicinity of the sand and gravel aggregates, with a
relatively lower amount of gel and air pores. The microstructure
presented heterogeneous morphologies, with high amounts of
small pores and big sized CSH gel like structures. Additionally,
acicular (needle-shaped) structures (Fig. 6b) were identied and
possibly formed of ettringite or other AFt phases that are rich in
CO3 2 (veried by EDS). The formation of the needle-shaped AFt
phases in cement pastes with a high calcium carbonate concentration was reported in [83]. Another expected phase that was not
possible to identify is monocarboaluminate. The inuence of the
presence of limestone on the hydration of Portland cement was
investigated by [84]. Blending of Portland cement with limestone
was found to inuence the hydrate assemblage of the hydrated
cement. Thermodynamic calculations as well as experimental
observations indicated that in the presence of limestone, monocarbonate instead of monosulfate was stable. Thermodynamic modelling shows that the stabilisation of monocarbonate indirectly also
stabilises ettringite. This leads to a corresponding decrease of the
total volume of the hydrated phase and an increase of porosity.
Similarly, Bonavetti et al. [85] reported that in limestone-blended
cement (20% limestone based on the weight of cement), calcium
monocarboaluminate can be easily detected at 3 days and its
amount increases after 28 days of hydration. The excess of carbonate ions in cement paste causes the transformation of monosulfoaluminate to monocarboaluminate. This promotes the reconversion
of monosulfoaluminate to ettringite in limestone blended cements
due to the sulfates liberated during the carbonate substitution
reaction. The conversion of ettringite to monosulfoaluminate is
deferred by the presence of limestone ller. Probably, similar
hydrates phase assemblies take place in the reference SCC, which
has a limestone concentration of 179.4 kg/m3 (34% based on the
weight of cement). Nevertheless, without complementary characterization techniques such as X-ray diffraction (XRD) and differential thermogravimetric analysis (DTG) it is not possible to conrm
the formation of manocarbonates phases in the reference SCC. In
addition, in the reference SCC, well crystallized hexagonal portlandite (Ca(OH)2) plates were formed in the cement matrix and in the
largest air-voids (Fig. 6c) due to the absence of any pozzolanic
additives. Normally, the presence of signicant amounts of
Ca(OH)2 results in a lower chloride intrusion resistance and lower
compressive strength [13]. These ndings are in line with the
mechanical and durability test results discussed previously.
As it can be observed in Fig. 7a, the SCC with colloidal nano-silica shows a more homogeneous microstructure compared to the
reference mix. A homogenous microstructure is dened by a more
regular morphology and similar size of the hydrates. This microstructure is characterized by apparent compact and small-sized
CSH gel and the absence of abundant gel porosity. As a consequence, a relatively denser ITZ was also conrmed by SEM
(Fig. 7b). It is important to notice, that the needle-type structures
or well grown Ca(OH)2 crystals were not found in the microstructural analysis. The absence of needle-type hydrates in the cement
matrix is unexpected. In general, as was explained for the reference
SCC, ettringite is one of the hydrated phases formed in cement system with high amount of limestone additions. In the case of the
SCC with colloidal nano-silica the concentration of limestone was
slightly decreased (from 179.4 to 151.8 kg/m3), but this decrease
should not inuence the hydrates formation. Similarly, in system
with amorphous nano-silica it was predicted [72] that one of the

77

stable phases formed is ettringite. Ettringite in long-term is only

unstable when the pH of the pore solution falls below 10 due to
further replacement of cement by silica. The dropping of the pore
solution pH to values lower than 10 is predicted for amorphous silica additions higher than 35% by mass [72]. In the present research
only 3.8% (based on the weight of cement) of nano-silica is added.
One possibility that can explain the absence of ettringite is that the
addition of nano-silica decreased the size of ettringite crystals,
making it much more difcult to be identied by the used magnication of the ESEM. Also, smaller ettringite crystals are more
susceptible to be discomposed by drying and vacuum conditions
as was reported by [86]. As was mentioned before, well grown
Ca(OH)2 crystals were not found in the microstructural analysis.

Fig. 6. Microstructure low vacuum (0.6 mbar) FEGSEM photomicrographs of the

SCC reference mix, (a) cement matrix and aggregate ITZ, (b) Ettringite and AFt
needles and (c) precipitated hexagonal Portlandite plate crystals.

78

G. Quercia et al. / Cement & Concrete Composites 45 (2014) 6981

Changes in the amount and size of portlandite crystals are expected in the presence of amorphous silica as it was established
by different authors [620]. The absence of well crystallized
portlandite and the lower volumes of gel pores conrm that the
addition of the nano-silica causes a renement of the microstructure and probably induces the precipitation of small-sized CSH
gel that normally has a high stiffness and lower Ca/Si ratio [67].
The improvement of the microstructure is also reected by the
mechanical properties (compressive and splitting tensile strength)
due to the fact that the pozzolanic gel structure has better mechanical properties than the CSH gel precipitated in OPC concrete.
These differences in the local CSH gel properties were reported
by Mondal et al. [67] using nanoindentation techniques. The study

complemented with 29Si MAS-NMR spectra of the cement pastes

with nano-silica proved that its addition increases the average
chain length of CSH gel. Therefore, it is expected that nano-silica
increases either the amount of high-stiffness CSH gel (E modulus
of 26 GPa) or the strength of the ITZ. Nanoindentation studies
proved that the elastic moduli of the two types of CSH phases
(low and high stiffness type) did not change with the addition of
nano-silica [67]. However, they stated that nano-silica increased
the volumetric amount of high-stiffness CSH gel. For example,
they reported that in samples with 6% by mass of nano-silica, the
volume fraction of high-stiffness CSH was as high as 38% by
vol. of the total gel. In contrast, reference samples (without nanoparticles) showed only 19% by vol. of high stiffness CSH gel.

Fig. 7. Microstructure low vacuum (0.6 mbar) FEGSEM photomicrographs of the

SCC colloidal nano-silica mix, (a) cement matrix, (b) aggregate dense ITZ and (c)
agglomerates in a pore.

Fig. 8. Microstructure low vacuum (0.6 mbar) FEGSEM photomicrographs of the

SCC powder nano-silica mix, (a) cement matrix, (b) aggregates ITZ and (c) AFt
needles precipitates.

G. Quercia et al. / Cement & Concrete Composites 45 (2014) 6981

Thus, the increased volume of the high-stiffness CSH gel might

explain the positive impact of nano-silica on the durability of
SCC in the present study.
In the case of SCC with powder nano-silica, the obtained results
of the microstructural analysis were similar to the colloidal nanosilica. Its microstructure is also found to be rened (as it is shown
in the selected image in Fig. 8a), but not as much as with the colloidal nano-silica. Apparently, due the fact that this nano-silica was
produced at high temperature (more compact, lower concentration
of silanol groups) its pozzolanic reactivity is lower than the colloidal one. Nevertheless, a relatively homogeneous matrix was observed with more small pores distributed in it (Fig. 8b). Even
though the microstructure was rened due to the addition of the
powder silica, it was still possible to observe some remnant needle-shaped AFt phases in the matrix (Fig. 8c). As it was explained
for the reference and SCC with colloidal nano-silica, the presence
of ettringite is expected in blended cements with amorphous silica
additions. In the case of the powder nano-silica, the size of the
ettringite needle appears to be smaller than the needle-type hydrates identied by the reference SCC. This can conrm that the
presence of amorphous nano-silica and probably also the higher
amount of SP in the SCC, could modify the morphology of the
ettringite-type hydrates. Taylor [86] reported that the amount of
available space, pH and sulfate ion concentration control the
ettringite morphology.
In summary, the resistance to the intrusion of chlorides and
water under pressure in the SCC containing nano-silica was improved due to the densication of the microstructure and the
high specic surface area of the produced gel [49]. The high reactivity of the colloidal nano-silica was also conrmed in Fig. 8c,
where small CSH gel precipitates were observed around the
limestone powder and agglomerates of nano-particles. The differences between the SCC with both types of nano-silica observed in
the microstructural analysis can explain the results that were obtained for these two SCCs, where the SCC containing colloidal
nano-silica presented better results in most of the performed
tests. The SCC with colloidal nano-silica appears to have a denser
matrix and ITZ than the SCC with powder nano-silica, leading to a
rened microstructure. The renement of the microstructure was
conrmed by the increased volumes of ink-bottle pores in the
SCC. In addition, this renement causes better mechanical and
durability properties of the tested SCC. Nevertheless, further research is needed to relate the results obtained in this study with
the progressing hydration of the SCC paste in the presence of
nano-silica particles. In addition, further research is also needed
to determine which type of nano-silica is better for a real scale
production of SCC.
4. Conclusions
Based on the mechanical and durability properties obtained by
the tests on a reference SCC and SCC mixes containing two types of
nano-silica, the following conclusions can be drawn:
 The results of the fresh state behavior of SCC demonstrated that
concrete with an addition of 3.8% nano-silica (based on the
weight of cement) shows similar ow properties and viscosity
as the reference mix without nano-silica. Slightly increased air
content in the mixes with nano-silica was measured due to a
higher viscosity of the paste.
 Under the laboratory conditions, the compressive and tensile
splitting strength of the reference SCC was improved by the
addition of both types of nano-silica. The highest compressive
strength was found for the colloidal nano-silica, while the
highest splitting tensile strength was found for the powder type
nano-silica. Small differences in the reactivity at early age

79

(1-day) were conrmed by the compressive strength measurements for both types of nano-silica studied.
The water permeable porosity of the three SCC mixes was found
to be similar. On the other hand, when nano-silica was added,
the concrete becomes almost impermeable to the penetration
of water under pressure. This can be explained by a renement
of the microstructure of the paste due to higher amount of pores
with ink-bottle shape, as conrmed by MIP measurements.
All durability indicators of the SCC studied (conductivity, chloride migration and diffusion coefcients, and freezethaw resistance) were signicantly improved with the addition of 3.8% of
both types of the nano-silica. Moreover, the SCC with colloidal
nano-silica showed slightly better properties than the SCC with
powder nano-silica.
The microstructural analysis of the hardened SCC reveals that
the addition of nano-silica resulted in a homogeneous microstructure, characterized by compact and small-sized CSH
gel. As a consequence, a denser ITZ was produced. The addition
of nano-silica caused a renement of the microstructure (less
interconnected and ner pore structure) and induced the precipitation of small-sized CSH gel, probably having a higher
stiffness and lower Ca/Si ratio.
The improvement of the microstructure resulted in higher
mechanical properties (compressive and splitting tensile
strength) due to the fact that the pozzolanic gel structure shows
better mechanical properties than the CSH gel precipitated in
OPC concrete. In addition, permeability of concrete was signicantly reduced (good resistance against the penetration of chlorides and water under pressure) because of the microstructural
densication and increased tortuosity of the pore system.
The high reactivity and faster pozzolanic behavior of the colloidal nano-silica particles at early age produced a more rened
microstructure than obtained for the SCC with powder nano-silica. Nevertheless, the higher reactivity and agglomerated state
appears to be the reason for the higher standard deviation
obtained in the compressive strength results.

Acknowledgments
This research was carried out under the Project Number
M81.1.09338 in the framework of the Research Program of the
Materials innovation institute (www.m2i.nl) and The European
Communitys Seventh Framework Program, ProMine: Nano-particle products from new mineral resources in Europe, FP7-NMP2008-LARGE-2 under Grant Agreement 228559. The authors also
wish to express their gratitude to the following sponsors of the
Building Materials research group at TU Eindhoven: Rijkswaterstaat Grote Projecten en Onderhoud, Graniet-Import Benelux, Kijlstra Betonmortel, Struyk Verwo, Attero, ENCI, Provincie Overijssel,
Rijkswaterstaat Zee en Delta District Noord, A&G Maasvlakte,
BTE, Alvon Bouwsystemen, V.d. Bosch Beton, Selor, Twee R Recycling, GMB, Schenk Concrete Consultancy, Geochem Research, Icopal, BN International, APP All Remove, Consensor, Eltomation,
Knauf Gips, Hess ACC Systems, Kronos and Joma international
(chronological order of joining).
References
[1] Scrivener KL, Kirkpatrick RJ. Innovation in use and research on cementitious
material. Cem Concr Res 2008;38:12836.
[2] Sanchez F, Sobolev K. Nanotechnology in concrete a review. Constr Build
Mater 2010;24:206071.
[3] Constantinides G, Ulm F. The nanogranular nature of CSH. J Mech Phys Solids
2007;55:6490.
[4] Richarson IG. The nature of the hydration products in hardened cement pastes.
Cem Concr Compos 2000;22:97113.

80

G. Quercia et al. / Cement & Concrete Composites 45 (2014) 6981

[5] Nili M, Ehsani A, Shabani K. Inuence of nano-SiO2 and micro-silica on

concrete performance. In: Proceedings second international conference on
sustainable construction materials and technologies, June 2830, Universita
Politecnica delle Marche, Ancona, Italy; 2010.
[6] Qing Y, Zenan Z, Deyu K, Rongshen Ch. Inuence of nano-SiO2 addition on
properties of hardened cement paste as compared with silica fume. Constr
Build Mater 2007;21:53945.
[7] Senff L, Labrincha JA, Ferreira VM, Hotza D, Repette WL. Effect of nano-silica on
rheology and fresh properties of cement pastes and mortars. Constr Build
Mater 2009;23:248791.
[8] Lin KL, Chang WC, Lin DF, Luo HL, Tsai MC. Effects of nano-SiO2 and different
ash particle sizes on sludge ashcement mortar. J Environ Manage
2008;88:70814.
[9] Bjornstrom J, Martinelli A, Matic A, Borjesson L, Panas I. Accelerating effects of
colloidal nano-silica for benecial calciumsilicatehydrate formation in
cement. Chem Phys Lett 2004;392:2428.
[10] Senff L, Hotza D, Repette WL, Ferreira VM, Labrincha JA. Mortars with nanoSiO2 and micro-SiO2 investigated by experimental design. Constr Build Mater
2010;24(8):14327.
[11] Ji T. Preliminary study on the water permeability and microstructure of
concrete incorporating nano-SiO2. Cem Concr Res 2005;35:19437.
[12] Li G. Properties of high-volume y ash concrete incorporating nano-SiO2. Cem
Concr Res 2004;34:10439.
[13] Gaitero JJ, Campillo I, Guerrero A. Reduction of the calcium leaching rate of
cement paste by addition of silica nanoparticles. Cem Concr Res
2008;38:11128.
[14] Green BH. Development of a high-density cementitious rock-matching grout
using nano-particles. In: Proceedings of ACI session on nanotechnology of
concrete: recent developments and future perspectives, November 7, Denver,
USA; 2006. p. 11930.
[15] Sobolev K, Ferrara M. How nanotechnology can change the concrete world
Part 2. Am Ceram Bull 2005;84(11):1620.
[16] Sobolev K, Ferrara M. How nanotechnology can change the concrete world
Part 1. Am Ceram Bull 2005;84(10):147.
[17] Sobolev K, Flores I, Hermosillo R. Nanomaterials and nanotechnology for highperformance cement composites. In: Proceedings of ACI session on
nanotechnology of concrete: recent developments and future perspectives,
November 7, Denver, USA; 2006. p. 91118.
[18] Belkowitz JS, Armentrout D. An investigation of nano-silica in the cement
hydration process. In: Proceeding 2010 concrete sustainability conference,
national ready mixed concrete association, USA; 2010. p. 115.
[19] Quercia G, Brouwers HJH. Application of nano-silica (nS) in concrete mixtures.
In: Gregor Fisher, Mette Geiker, Ole Hededal, Lisbeth Ottosen, Henrik Stang,
editors. 8th b International Ph.D. symposium in civil engineering, June 2023,
Lyngby, Denmark; 2010. p. 4316.
[20] Rols S, Ambroise J, Pra J. Effects of different viscosity agents on the properties
of self-leveling concrete. Cem Concr Res 1999;29:2616.
[21] Okamura H, Ozawa K. Mix-design for self-compacting concrete. Concr Libr, Jpn
Soc Civ Eng JSCE 1995;25:10720.
[22] Audenaert K, Boel V, De Schutter G. Chloride migration in self compacting
concrete. In: Proceeding fth international conference on concrete under
severe conditions: environment and loading (CONSEC07), June 46, Tours,
France; 2007. p. 191298.
[23] Byung WJ, Chang HK, Jae HL. Investigations on the development of powder
concrete with nano-SiO2 particles. KSCE J Civ Eng 2007;11(1):3742.
[24] Collepardi M, Ogoumah JJ, Skarp U, Troli R. Inuence of amorphous colloidal
silica on the properties of self-compacting concretes. In: Proceedings of the
international conference challenges in concrete construction innovations
and developments in concrete materials and construction, 911 September,
Dundee, Scotland, UK; 2002. p. 47383.
[25] Sari M, Prat E, Labastire JF. High strength self-compacting concrete original
solutions associating organic and inorganic admixtures. Cem Concr Res
1999;29:8138.
[26] Sadrmomtazi A, Barzegar A. Assessment of the effect of nano-SiO2 on physical
and mechanical properties of self-compacting concrete containing rice husk
ash. In: Proceedings second international conference on sustainable
construction materials and technologies, June 2830, Universita Politecnica
delle Marche, Ancona, Italy; 2010. p. 19.
[27] Maghsoudi AA, Arabpour-Dahooei F. Effect of nanoscale materials in engineering
properties of performance self compacting concrete. In: Proceeding of the 7th
international congress on civil engineering, Iran; 2007. p. 111.
[28] Khanzadi M, Tadayon M, Sepehri H, Sepehri M. Inuence of nano-silica
particles on mechanical properties and permeability of concrete. In:
Proceedings second international conference on sustainable construction
materials and technologies, June 2830, Universita Ploitecnica delle Marche,
Ancona, Italy; 2010. p. 17.
[29] Raiess-Ghasemi AM, Parhizkar T, Ramezanianpour AA. Inuence of colloidal
nano-SiO2 addition as silica fume replacement material in properties of
concrete. In: Proceedings second international conference on sustainable
construction materials and technologies, June 2830, Universita Ploitecnica
delle Marche, Ancona, Italy; 2010. p. 18.
[30] Baomin W, Lijiu W, Lai FC. Freezing resistance of HPC with nano-SiO2. J Wuhan
Univ Technol, Mater Sci ed 2008;23(1):858.
[31] Wei X, Zhang P. Sensitivity analysis for durability of high performance
concrete containing nanoparticles based on grey relational grade. Mod Appl
Sci 2011;5(4):6873.

[32] ENCI BV. Betonpocket 2010. The Netherlands: Heidelberg Cement GroupsHertogenbosch; 2009. p. 1288 [in Dutch].
[33] Brunauer S, Emmet PH, Teller E. Adsorption of gases in multimolecular layers. J
Am Chem Soc 1938;62:30919.
[34] DIN ISO 9277. Determination of the specic surface area of solids by gas
adsorption using the BET method. German Institute of Normalization-DIN;
2005. p. 119.
[35] Hunger M. An integral design concept for ecological self-compacting
concrete. Ph.D. thesis. Eindhoven University of Technology, The Netherlands;
2010.
[36] Hsken G, Brouwers HJH. A new mix design concept for earth-moist concrete:
a theoretical and experimental study. Cem Concr Res 2008;38:124659.
[37] Dutch Normalization-Institute. Nederlandse invulling van NEN-EN 2061:
Beton Deel 1: specicatie, eigenschappen, vervaardiging en conformiteit,
Nederlands Normalisatie Instituut, Delft, The Netherlands; 2008. p. 172.
[38] BMC Certicatie BRL 1801: Nationale Beoordelingsrichtlijn Betonmortel. BMC
Certicatie, Gouda, The Netherlands; 2006 [in Dutch].
[39] Specication and guidelines for Self Compacting Concrete-SCC. In: Report,
European Federation of Producers and Contractors of Specialist Products for
Structures EFNARC, Surrey, UK; 2005.
[40] BS-EN 12390-2. Testing hardened concrete Part 2: making and curing
specimens for strength tests. British Standards Institution-BSI and CEN
European Committee for Standardization; 2000. p. 110.
[41] BS-EN 12390-3. Testing hardened concrete compressive strength of test
specimens. British Standards Institution-BSI and CEN European Committee for
Standardization; 2009. p. 122.
[42] BS-EN 12390-6. Testing hardened concrete tensile splitting strength of test
specimens. British Standards Institution-BSI and CEN European Committee for
Standardization; 2000. p. 114.
[43] Sauddin Md, Hearn N. Comparison of ASTM saturation techniques for
measuring the permeable porosity of concrete. Cem Concr Res
2005;35:100813.
[44] ASTM C1202. Standard test method for electrical indication of concretes
ability to resist chloride ion penetration. In: Annual Book of ASTM Standards,
vol. 04.02. Philadelphia: American Society for Testing and Materials; 2005. p.
16.
[45] BS-EN 12390-8. Testing hardened concrete depth of penetration of water
under pressure. British Standards Institution-BSI and CEN European
Committee for Standardization; 2009. p. 110.
[46] Nordtest method NT Build 492. Concrete, mortar and cement-based repair
materials: chloride migration coefcient from non-steady-state migration
experiments. Finland 1999:18.
[47] Spiesz P, Ballari MM, Brouwers HJH. RCM: a new model accounting for the
non-linear chloride binding isotherm and the non-equilibrium conditions
between the free- and bound-chloride concentrations. Constr Build Mater
2012;27:293304.
[48] Polder RB. Test methods for onsite measurement of resistivity of concrete a
RILEM
TC-154
technical
recommendation.
Constr
Build
Mater
2001;15:1251301.
[49] Bentz DP, Jensen OM, Coats AM, Glasser FP. Inuence of silica fume on
diffusivity in cement-based materials I. Experimental and computer modeling
studies on cement pastes. Cem Concr Res 2000;30:95362.
[50] Tang L.Chloride transport in concrete measurement and prediction. PhD
thesis. Chalmers University of Technology, Gothenburg, Sweden; 1996.
[51] Nordtest method NT Build 443. Concrete, hardened: accelerated chloride
penetration. Finland 1995:15.
[52] Yuan Q. Fundamental studies on test methods for transport of chloride ions in
cementitious materials. PhD thesis. Universiteit Gent, Belgium; 2009. p. 1
340.
[53] NEN-EN 12390-9. Testing hardened concrete freezethaw resistance
scaling. CEN European Committee for Standardization and Dutch
Normalization-Institute, Delft, The Netherlands; 2006. p. 129 [in English].
[54] Plank J, Schroe Ch, Gruber M, Lesti M, Sieber R. Effectiveness of
polycarboxylate superplasticizer in ultra-high strength concrete: the
importance of PCE compatibility with silica fumel. J Adv Concr Technol
2009;7(1):125.
[55] Spiratos N, Page M, Mailvaganam NP, Malhotra VM, Jolicoeur C.
Superplasticizer for concrete. Fundamentals, technology, and practice.
Otawa, Canada: Supplementary Cementing Materials for Sustainable
Development Inc.; 2003. p. 1321.
[56] Kordts S, Breit W. Combined test method for assessing the workability of SCCow cone. In: Concrete technology reports; 20042006. p. 715.
[57] Felekoglu B. Utilization of high volumes of limestone quarry wastes in
concrete industry (self-compacting concrete case). Resour, Conserv Recycl
2007;51:77091.
[58] Piechowka-Mielnik M, Giergiczny Z. Properties of Portland-composite cement
with limestone. In: Proceeding of the XIII international conference on cement
chemistry, July 48, Madrid; 2011. p. 17.
[59] ECETOC JACC REPORT No. 51. Synthetic amorphous silica (CAS No. 763186-9).
European Centre for Ecotoxicology and Toxicology of Chemicals, Brussels,
September; 2006. p. 1231.
[60] Iler RK. The colloid chemistry of silica and silicate. New York: Edit. Ithaca;
1955. p. 1250.
[61] Madani H, Bagheri A, Parhizkar T. The pozzolanic reactivity of monodispersed
nanosilica hydrosols and their inuence on the hydration characteristics of
Portland cement. Cem Concr Res 2012;42:156370.

G. Quercia et al. / Cement & Concrete Composites 45 (2014) 6981

[62] Kong D, Su Y, Du X, Yang Y, Wei S, Shah SP. Inuence of nano-silica
agglomeration on fresh properties of cement pastes. Constr Build Mater
2013;37:55762.
[63] Berra M, Carassiti F, Mangialardi T, Paolini AE, Sebastiani M. Effecs of
nanosilica addtion on workability and compressive strength of Portland
cement pastes. Constr Build Mater 2012;35:66675.
[64] Kong D, Du X, Wei S, Zhang H, Yang Y, Shah SP. Inuence of nano-silica
agglomeration on microstructure and properties of the hardened cementbased materials. Constr Build Mater 2012;37:70715.
[65] Zhang M, Ilsam J, Peethamparan S. Use of nano-silica to increase early strength
and reduce setting time of concretes with high volumes of slag. Cem Concr
Compos 2012;34:65062.
[66] Jalal M, Mansouri E, Sharipour M, Pouladkhan AR. Mechanical, rheological,
durability and microstructural properties of high performance selfcompacting concrete containing SiO2 micro and nanoparticles. Mater Des
2012;34:389400.
[67] Mondal P, Shah SP, Marks LD, Gaitero JJ. Comparative study of the effects of
microsilica and nanosilica in concrete. Transportation Research Record:
Journal of the Transportation Research Board, No. 2141. Washington,
DC: Transportation Research Board of the National Academies; 2010. p. 69.
[68] Neville AM. Properties of concrete. 4th ed. Harlow, UK: Prentice Hall/Pearson;
2002. p. 53776.
[69] Garboczi EJ. Permeability, diffusivity, and microstructural parameters: a
critical review. Cem Concr Res 1990;20:591601.
[70] Yogendran V, Langan BW. Utilization of silica fume in high strength concrete.
In:
Proceedings
of
utilization
of
high
strength
concrete,
Stavanger. Norway: Tapir Publisher Trondheim; 1987.
[71] Mindess S, Young JF, Darwin D. Concrete. 2nd ed. Upper Saddle River, New
Jersey: Prentice Hall; 2003.
[72] Lotenbach B, Scrivener K, Hooton RD. Supplementary cementitious materials.
Cem Concr Res 2001;41:124456.
[73] Ye G. Experimental study and numerical simulation of the development of the
microstructure and permeability of cementitious materials. PhD thesis. Delft
University of Technology, Delft; December 18 2003. p. 1189.
[74] Shih JY, Chang TP, Hsiao TCh. Effect of nano-silica on characterization of
Portland cement composite. Mater Sci Eng A 2006;424:26674.

81

[75] Desmet D, Hernandez J, Willain L, Vantomme J. Porosity determination of selfcompacting concrete using forced saturation. In: Proceeding of the XIII
international conference on cement chemistry, July 48, Madrid; 2011. p. 17.
[76] Andrade C, DAndrea R, Lopez JC, Cienfuegos-Jovellano A, Alvarez JM, Millan
JM. Use of electrical resistivity as complementary tool for controlling the
concrete production. In: Proceeding of the XIII international conference on
cement chemistry, July 48, Madrid; 2011. p. 17.
[77] Andrade C, Rio O, Castillo A, Castellote M, DAndrea R. A NDT performance
method based on electrical resistivity for the specication of concrete
durability. In: ECCOMAS thematic conference on computational methods in
tunnelling, August 2729, Vienna, Austria; 2007. p. 19.
[78] Bagheri AR, Zanganeh H, Moalemi MM. Mechanical and durability properties
of ternary concretes containing silica fume and low reactivity blast furnace
slag. Cem Concr Compos 2012;34:66370.
[79] DuraCrete. Probabilistic performance based durability design of concrete
structures. In: DuraCrete nal technical report. Document BE951347/R17;
2000.
[80] CUR Durability Guideline. Duurzaamheid van constructief beton met
betrekking tot chloride-genitieerde wapeningscorrosie. CUR Bouw en Infra.
Gouda, The Netherlands; 2009. p. 165 [in Dutch].
[81] Romero HL, Casati MJ, Galvez JC, Molero M, Segura I. Study of the damage
evolution of concrete under freezethaw cycles using traditional and nontraditional techniques. In: Proceeding of the XIII international conference on
cement chemistry, July 48, Madrid; 2011. p. 17.
[82] Stark J, Wicht B. Dauerhaftigkeit von Beton: Der Baustoff als Werkstoff,
Birkhuser, Basel, Switzerland; 2001 [in German].
[83] Nocun-Wczelik W, Loj G. Effect of nely dispersed limestone additives of
different origin on cement hydration kinetics and cement hardening. In:
Proceeding of the XIII international conference on cement chemistry, July 48,
Madrid; 2011. p. 17.
[84] Lothenbach B, Le Saout G, Gallucci E, Scrivener K. Inuence of limestone on the
hydration of Portland cements. Cem Concr Res 2008;38:84860.
[85] Bonavetti VL, Rahhal VF, Irassar EF. Studies on the carboaluminate formation
in limestone ller-blended cements. Cem Concr Res 2001;31:8539.
[86] Taylor HFW. Cement chemistry. London: Academic Press; 1990.