Materiales de Construcción

Vol. 70, Issue 339, July–September 2020, e221

ISSN-L: 0465-2746
https://doi.org/10.3989/mc.2020.10619

Behaviour of a high-performance self-compacting concrete

(HPSCC) with ternary mixtures of nano- and
microsilica in the presence of chlorides

E. Reyesa*, J. Massanab, F. Alonsob, N. Leóna, A. Moraguesa

a. Dpto. Ingeniería Civil, Construcción, Escuela Técnica Superior de Ingenieros de Caminos, Canales y Puertos, Universidad
Politécnica de Madrid, (Madrid, Spain)
b. Dpto. de Ingeniería Agroforestal, Escuela Técnica Superior de Ingeniería Agronómica, Alimentaria y Biosistemas. Universidad
Politécnica de Madrid, (Madrid, Spain)
* [email protected]

Received 30 July 2019

Accepted 21 January 2020
Available on line 18 May 2020

ABSTRACT: In this paper, the influence of additions of nanosilica (nSi) and microsilica (mSi) on the behav-
iour of binary and ternary mixtures in chloride environments is studied. The main objective is to obtain high-
performance self-compacting concrete (HPSCC) with a high durability which can meet specific demands in such
aggressive environments. Ten blends were manufactured using Portland cement (CEM I 52.5 R) and additions
of nSi and mSi in binary and ternary mixtures. The results of three tests frequently used to evaluate resistance to
chloride penetration– electrical resistivity, migration and chloride diffusion –were studied and compared. Both
binary and ternary mixtures presented significant improvements in chloride resistance, generally in proportion
to the total content of the addition. In all the ternary mixtures, high resistivity is obtained, which indicates that
such mixtures have a notably low chloride penetrability. Furthermore, these mixtures provided extremely low
chloride diffusion coefficients even at small addition ratios.
KEYWORDS: High-performance concrete; Active addition; Durability; Diffusion; Corrosion.

Citation/Citar como: Reyes, E.; Massana, J.; León, N.; Moragues, A. (2020) Behaviour of a high-performance self-
compacting concrete (HPSCC) with ternary mixtures of nano- and microsilica in the presence of chlorides. Mater.
Construcc. 70 [339], e221 https://doi.org/10.3989/mc.2020.10619

RESUMEN: Comportamiento frente a cloruros de un hormigón autocompactable de alta resistencia con mezclas
ternarias de nano y micro sílice. En este artículo se estudia la influencia en el comportamiento frente a cloruros
de la adición de nanosílice (nSi) y microsílice (mSi) en mezclas binarias y ternarias. El principal objetivo es
obtener un hormigón autocompactante de altas prestaciones con una alta durabilidad frente a estos ambientes
agresivos con cloruros. Se prepararon diez dosificaciones usando cemento Portland (CEM I 52.5 R) y adiciones
de nSi y mSi en mezclas binarias y ternarias. Se estudiaron y analizaron tres ensayos frecuentemente utilizados
para evaluar la resistencia a la penetración de cloruros, resistividad eléctrica, migración y difusión de cloruros.
Tanto las mezclas binarias como ternarias presentaron mejoras significativas en la resistencia a cloruros, en
general proporcional al contenido de adición. En todas las mezclas ternarias se ha obtenido alta resistividad,
indicando una penetrabilidad a cloruros notablemente baja. Estas mezclas proporcionaron un coeficiente de
difusión de cloruros extremadamente bajo, incluso para bajos contenidos de adición.
PALABRAS CLAVE: Hormigón de altas prestaciones; Adición activa; Durabilidad; Difusión; Corrosión.

ORCID ID: E. Reyes (https://orcid.org/0000-0002-1284-7335); J. Massana (https://orcid.org/0000-0003-2960-0491);

F. Alonso (https://orcid.org/0000-0002-5095-385X); N. León (https://orcid.org/0000-0001-6427-798X); A. Moragues
(https://orcid.org/0000-0001-7819-3066).

Copyright: © 2020 CSIC. This is an open-access article distributed under the terms of the Creative Commons
Attribution 4.0 International (CC BY 4.0) License.
2 • E. Reyes et al.

1. INTRODUCTION frequent mechanisms of degradation in the struc-

tural elements of reinforced concrete. Therefore,
Many studies have been carried out on self-­ when the microstructure of cement-based materials
compacting concrete (SCC) (1–11). This type becomes denser and has a refined porous network, it
of concrete is mainly used to aid in proper form- becomes more resistant to aggressive agents in gen-
work placement, particularly for formwork with eral and chlorides in particular.
complex shapes, and in structural elements with Conventionally, pozzolanic additions have been
strongly reinforced sections in which the vibration used to densify, reduce and refine the porous struc-
process is complex and might move reinforcements. ture, due to the combined effect of the pozzolanic
Simultaneously, SCC mitigates some operator reactions and the filling effect associated with the
health risks. The mixture dosage is designed to small size of the material. Among the most interest-
achieve a high degree of flowability that enables ing pozzolanic additions is microsilica (mSi), whose
good consolidation through the action of the con- incorporation in cement-based mixtures has resulted
crete’s own weight without exudation or segregation in sound performance (21). However, the increase
(12, 13). Due to this property, SCC use has increased in structure service life currently demanded by
in construction and building, as well as in the pre- sustainability-oriented technical requirements calls
fabricated concrete industry, where the absence of for cementitious materials with significant improve-
vibration reduces production costs and increases the ments in durability. For this reason, the influence
service life of moulds. of nanosilica (nSi) additions on concrete durability
The advantageous workability and cohesion of has been recently examined, since a greater, faster
SCC are mainly due to a high content in fines, a activity can be expected because of the significant
reduction in the content and size of coarse aggre- increase of specific surface due to the small size of the
gates and the action of superplasticizer additives nanoparticles. The study by Björnstrom et al. (2004)
that provide the necessary flowability for casting (22) showed that colloidal nanosilica has beneficial
work (13). Nepomuceno et al. (14) carried out a accelerating effects for the formation of hydrated
comprehensive study to correlate fresh and hard- products in cement. Some research has concluded
ened properties of different mineral additions in that nanometric silica exhibits notable pozzolanic
binary blends, and they proposed a methodol- activity, increasing the formation of C-S-H gels, due
ogy for designing SCC mixtures. Various methods to the secondary C-S-H generated by the reaction
have been studied for designing a self-compacting between portlandite and pozzolans (23, 24). Rong
mixture, depending on the properties required; a et al. (2015) (25) studied the influence of nanosilica
number of methods were recently examined by Shi, additions on the mechanical performance, hydra-
Wu, Lv and Wu (2015) (15). When an SCC has a tion process and microstructure evolution of ultra-
high compressive strength and high durability, it high-performance cementitious composites. Their
is designed as a high-performance self-compacting results showed that mechanical properties improved
concrete. HPSCC is generally prepared with a high with the increase of the nanosilica content up to 3%,
cement content, from 400 to 600 kg/m3 (13), and a combined with a more homogenous, denser micro-
low water/cement ratio, which can be improved with structure and acceleration of the hydration process,
the use of inactive additions (particularly limestone which was mainly attributed to the pozzolanic and
powder used as a filler (16, 17)) or active additions. filler effects of nanosilica. A review of nanosilica’s
Some case examples of the latter type are fly ash contribution to the mechanical and durability prop-
(18), blast furnace slag (19), microsilica (17, 18) and erties of cement-based materials can be consulted
nanosilica (18, 20). in Balapour et al. (26). These nanoparticles can
Significant strides have been made in the last also facilitate the nucleation of hydrated cement
two decades in improving the mechanical and crystals, generating changes not only in the rate of
resistant performance of concrete, particularly its hydration, but also in the morphology and size of
compressive strength. Therefore, high- and ultra- the hydrated compounds (17). However, the use of
high-strength concretes are relatively easily obtained nSi presents considerable technical difficulties, such
now. However, making the mixture self-compacting as a high water demand, which makes the mixing
and producing high-durability concrete for applica- operation problematic (even more so in the case of
tions that require a service life of over 100 years is SCC). Therefore, identifying improvements is not
considerably more difficult. The main weakness in as simple as it might appear. Numerous published
concrete durability involves the connectivity and size authors have compared the behaviour of nano-
of the porous network, which determines the ingress and microsilica (nSi and mSi) additions. Jalal et al.
of external aggressive substances that can affect (18) studied the influence of these additions on the
the cement matrix and the steel embedded in the mechanical properties, water absorption, electrical
structural concrete. Prominent among these aggres- resistivity and chloride ion penetration of binary
sive agents are chlorides. When chlorides reach the and ternary HPSCC mixtures and found that the
steel, they cause depassivation and corrosion, two refinement of the microstructure obtained using nSi

Materiales de Construcción 70 (339), July–September 2020, e221. ISSN-L: 0465-2746. https://doi.org/10.3989/mc.2020.10619

 Behaviour of a high-performance self-compacting concrete (HPSCC) • 3

and mSi could improve strength and durability prop- using nano- and micrometric additions together
erties, particularly at greater ages. Gesoglu et al. (27) is combined with the enhancement of reactivity
focused on the effect of using nSi and mSi in binary derived from increased specific surface. The main
and ternary blends on the mechanical properties of objective is to obtain HPSCC with notably high
low-binder, ultra-high performance cementitious durability which can respond to specific demands.
composites. The results obtained with mixtures As previously mentioned, further knowledge of
containing nSi and mSi showed better mechanical how a concrete behaves when attacked by chlorides
behaviour than binary mixtures. Similarly, in previ- is one of the fundamental aspects in determining
ous work (28) the authors examined the fresh and concrete durability (1, 2). This work provides the
mechanical behaviour of binary and ternary mix- results of three tests frequently used to assess resis-
tures of HPSCC. Ternary mixtures showed better tance to chloride penetration: electrical resistivity
fresh behaviour than binary mixtures containing the (3–10), chloride migration and chloride diffusion.
same total amount of additions. Furthermore, these The model described in EHE (13) is used to estimate
ternary mixtures present the best mechanical behav- the service life of the structure from the values of
iour. However, there are few published works that the chloride diffusion coefficient. Analysis and com-
evaluate the effects of nSi, mSi and ternary combi- parison of the results obtained from the HPSCC
nations on the durability properties of HPSCC, such mixtures allow the study to find optimised dosages
as freeze-thaw and carbonation-process resistance, for the fabrication of high-sustainability structural
which are very important for concrete structures in elements in chloride-rich environments.
some aggressive environments (25, 29). For example,
freeze-thaw resistance is considered fundamental for 2. EXPERIMENTAL STUDY
both precast and on-site structural elements used in
natural environments with frequent frost action, 2.1. Materials used and mixture proportioning
such as high-mountain environments.
HPSCC seems to be a promising material for In this study, 10 blends were designed (31) for
many applications and structures. However, its per- obtaining HPSCC with CEM I 52.5 R (PC) cement
formance must be studied before it is widely adopted and two mineral admixtures, nSi and mSi, in different
in construction. Also, the behaviour of structural percentages in reference to cement weight. Table 1
elements made with HPSCC has to be more thor- presents the chemical composition and physical
oughly understood, and design provisions in step properties of the cementitious materials used. The
with the latest advances are needed. This paper nSi was dispersed in water, under the trade name
forms part of a larger research effort into the influ- of Levasil® 200/40%, with 40% solids by weight,
ence of binary and ternary mixtures of mSi and nSi a specific surface area of 200 m2g−1 and a particle
in different proportions on the properties of fresh- size of approximately 15 nm (see Table 1). The mSi
and hardened-state HPSCC (28, 29). Designing used was Elkem Microsilica® MS 940 U, a material
self-compacting mixtures with these additions is composed of nonporous amorphous spheres of
a challenge, due to the notable loss of workability SiO2 with submicron size and small agglomer-
they cause. Adequate SCC flowability was achieved ates, with a specific surface area of 15–30 m2g−1
in this study due to chemical additives compatible and a particle size of approximately 0.15 micron
with the mixture in suitable proportions. Although (see Table 1). Even when some of the spheres may be
there are some studies of the behaviour of SCC in found separately, the majority tend to form agglom-
resisting chlorides (30), there is a lack of published erates of primary particles with a size range from
research about the durability of SCC in chloride- 0.1 to 1.0 micron. Additions (nSi and mSi) were not
rich environments, despite the increase of its use. In used as a cement replacement, but as an additional
this paper, the chloride resistance of HPSCC pre- cementitious material.
pared with binary and ternary mixtures using mSi The aggregates and proportioning used in the
and nSi is studied at the reference age of 28 days. mixtures were as follows: 1160 kg/m3 of siliceous
In these mixtures, the effect of greater compactness sand (termed A) measuring less than 4 mm in diam-
due to the increased packing density derived from eter, with a fineness modulus of 3.30; 585 kg/m3

Table 1. Properties of Portland cement (PC) and mineral additions (nSi, mSi).

Loss on Density Specific surface

SiO2 Al2O3 Fe2O3 CaO MgO SO3 K2O Na2O ignition (%) (g·cm−3) area (m2g−1)
PC 19.20 6.07 1.70 63.41 2.56 3.38 0.2 0.33 2.09 3.5 0.42
nSi 99.90 - - - - - - - 0.10 1.29 200
mSi 94 - - - - - - - - 0.7 30

Materiales de Construcción 70 (339), July–September 2020, e221. ISSN-L: 0465-2746. https://doi.org/10.3989/mc.2020.10619

4 • E. Reyes et al.

of rolled gravel (G) measuring 6 mm to 12 mm in and HAC[nmSi]-2.5/5). Table 2 shows the mixture
diameter; and 100 kg/m3 of limestone filler (LF) proportions.
with a granulometry in accordance with UNE The tests performed on the blends to determine
12620:2003+A1:2009 (32) and a maximum diam- self-compacting were as follows: slump-flow diam-
eter of 63 μm. All the SCCs were prepared using a eter [df (mm)], V‑funnel flow time [TV (s)] and L-box
water-to-cementitious material ratio (w/cm) of 0.36 height ratio [Cbl] (12, 13). The values obtained from
and 450 kg/m3 of cement. the tests for all the mixes were within the range that
Additionally, two chemical additives were used: guarantees self-compactability. Once hardened,
SIKA ViscoCrete® 5720, based on polycarboxyl- compressive strength, tensile strength and modu-
ate polymers (solids content 36%, density 1.09 kg/l), lus of elasticity were obtained to characterize the
as superplasticizer (SP) and SIKA Stabilizer® 4R mechanical and resistant properties of all blends.
(solids content 3–3.5%, density 1.03 kg/l) as stabi- Table 3 shows the fresh properties corresponding to
lizer (MV), which controls concrete viscosity with a self-compacting tests and the compressive strength
constant amount in mixtures of 0.15% by weight of at 28 days for all mixtures. Detailed information on
cement (wt.%). the results is available in previously published works
Ten HPSCC dosages were designed. In order to (28, 29).
obtain reference values to compare the results, an
HPSCC without mineral additions was prepared 2.2. Test programme
(CEM I 52.5 R as the sole cementitious mate-
rial), identified as HAC. Three dosages were made Pore-size distribution and total porosity were
with 2.5%, 5% and 7.5% nSi, which were respec- evaluated by Mercury Intrusion Porosimetry (MIP),
tively identified as HAC[nSi]-2.5, HAC[nSi]-5 and according to ASTM D4404-84 (2004), using a
HAC[nSi]-7.5. Three more were made with 2.5%, 5% Micromeritics Autopore IV 9500 at a maximum
and 7.5% mSi, which were identified as HAC[mSi]-2.5, pressure of 33,000 psi, with a range of 5 nm to
HAC[mSi]-5 and HAC[mSi]-7.5. Lastly, the remain- 180 µm.
ing three dosages were made by using admixtures of For electrical resistivity tests, cylindrical moulds
both nSi and mSi (ternary mixtures) in percentages, 200 mm long and 100 mm in diameter were used to
respectively, of 2.5%/2.5%, 5%/2.5% and 2.5%/5% prepare a total of 30 specimens (three specimens per
(identified as HAC[nmSi]-2.5/2.5, HAC[nmSi]-5/2.5 dosage). Electrical resistivity was determined by a

Table 2. HPSCC mixture proportions.

Material HAC HAC HAC HAC HAC HAC HAC HAC

(kg/m3) (%) HAC HAC [nSi]-5 [nSi]-7.5 [mSi]-2.5 [mSi]-5 [mSi]-7.5 [nmSi]-2.5/2.5 [nmSi]-5/2.5 [nmSi]-2.5/5
nSi - 11.25 22.5 33.75 11.25 22.5 11.25
- - -
2.5% 5% 7.5% 2.5% 5% 2.5%
mSi 11.25 22.5 33.75 11.25 11.25 22.5
- - - -
2.5% 5% 7.5% 2.5% 2.5% 5%
Water 162 166 170.1 174.1 166 170.1 174.1 170.1 174.1 174.1
SP (%) 2 3.30 4 6 2.30 2.50 2.70 3.60 4.80 3.90

Table 3. Results of self-compacting and compressive strength tests.

HAC HAC HAC HAC HAC HAC HAC HAC HAC

Self-compacting tests HAC [nSi]-2.5 [nSi]-5 [nSi]-7.5 [mSi]-2.5 [mSi]-5 [mSi]-7.5 [nmSi]-2.5/2.5 [nmSi]-5/2.5 [nmSi]-2.5/5
Slump-flow diameter 650 720 635 565 787 817 795 685 675 752
df (mm)
V-funnel flow time 8 11 13 17 10 7 5 12 12 10
TV (s)
L-box height ratio 0.98 0.96 0.81 0.85 0.95 0.95 1 0.89 0.97 0.97
Cbl
Slump-flow class SF1 SF2 SF1 SF1 SF3 SF3 SF3 SF2 SF2 SF2
Compressive strength 62.46 63.79 65.01 68.86 65.73 68.37 70.35 82.17 71.15 69.25
(28 days)
(MPa)

Materiales de Construcción 70 (339), July–September 2020, e221. ISSN-L: 0465-2746. https://doi.org/10.3989/mc.2020.10619

 Behaviour of a high-performance self-compacting concrete (HPSCC) • 5

non-destructive test carried out under room con- In order to determine the chloride migration coef-
ditions on samples saturated to constant weight. ficient, two slices 100 mm in diameter and 50 mm in
Mortars were tested at 7, 28 and 365 days to analyse length for each mixture were tested according to the
their evolution over time after a 28-day curing pro- NT BUILD 492 (1999) standard (35). These slices
cess at room temperature (20 ± 2°C) and 95% rela- were taken from the central part of the cylindrical
tive humidity. This test was carried out according to specimens (diameter 100 mm, height 200 mm) after
UNE 83988-1 (33) using Giatec RCON equipment. 28 days of curing, prepared according to the stan-
Such equipment measures electrical resistance with dardised procedure and subjected to a difference in
frequencies from 1 to 30 kHz. Electrical resistivity electric potential for a specified time. Each slice was
was then calculated using Equation [1]. Each mix- then cracked axially by the indirect traction method
ture’s resistivity was calculated as the average value with a press. Then a solution of silver nitrate 0.1 M
from three measurements. was applied to the surface of the specimens, and the
depth of chloride penetration was measured. The
ρe = K ⋅ Re [1] non-steady-state migration coefficient (Dnssm) is
obtained by Equation [3]:
where:
ρe: electrical resistivity (Ω m) Dnssm =
0.0239( 273 + T ) L  X d − 0.0238
( 273 + T ) L X  [3]
d

K: cell constant (m) ( U −2 ) t  U −2 


Re: electrical resistance (Ω)

The cell constant K is found by Equation [2]: where:

Dnssm: non-steady-state migration coefficient
S (× 10−12 m2/s)
K= [2] U: absolute value of the voltage applied (V)
L
T: average temperature of the anode solution
between its initial and final value (°C)
where: L: specimen thickness (mm)
K: cell constant (m) Xd: average depth of chloride penetration (mm)
S: surface area of the specimen through which the t: test duration (hours)
electrical charge passes (m2)
L: specimen height (m) The chloride diffusion test was carried out
according to the CEN/TS 12390-11 standard (36).
The test entailed non-destructive testing and is The specimens were obtained from cutting the cylin-
an indirect measure of the connectivity and tortu- drical specimens cured at 28 days and according to
osity parameters of concrete porosity, which have a the standard procedure. All the cuts were made with
significant influence on resistance against chloride the Struers Secotom-10 precision diamond cut-off
penetration. The electrical resistivity value is related machine, with a 0.8 mm-thick disc. Liquid petro-
to chloride penetrability through the determina- leum jelly was used to cool the cutting disc with-
tion of the electrical charge (coulombs) that passes out washing away the chlorides. Layers 2 mm thick
through a specimen in a determined time, as in were obtained. Then, these layers were ground to
ASTM C1202 (34). Table 4 shows the quality clas- be analysed by a chemical procedure to determine
sification according to this value. chloride concentration by using a Metrohm model
With the aim of analysing the evolution of the 916 Ti-Touch titrator. The total chloride concen-
concretes’ resistivity over time, the specimens were tration was measured according to the standard
tested at the ages of 7, 28 and 365 days. test method in UNE-EN 14629:2007 (37) at sev-
eral depths to find the apparent chloride diffusion
coefficient.
Table 4. Quality classification of concrete according to The test results yielded the surface chloride con-
chloride penetrability.
centration on the exposed surface (% mass) (Cs)
Electrical charge that Electrical and the apparent diffusion coefficient (m2/s) (Da),
passes (coulombs) resistivity (Ω cm) Chloride obtained by fitting Equation [4] by means of a
ASTM C1202 UNE 83988-1 penetrability non-linear regression according to the least squares
> 4000 <5 High method.
2000 to 4000 5 to 10 Moderate
1000 to 2000 10 to 20 Low  x 
100 to 1000 20 to 200 Very low ( ) (
C x, t = Cs − Cs − Ci ⋅ erf  )  [4]
 4 ⋅ D ⋅ t 
< 100 > 200 Negligible a

Materiales de Construcción 70 (339), July–September 2020, e221. ISSN-L: 0465-2746. https://doi.org/10.3989/mc.2020.10619

6 • E. Reyes et al.

where: 3. RESULTS AND DISCUSSION

C(x, t): total chloride concentration at depth x and
time t (% mass) 3.1. Pore network
Ci: initial chloride concentration (% mass)
x: depth from the exposed surface (m) Table 5 shows the total porosity and the macro-
t: exposure time (s) and micropore percentages for all mixtures at 7 and
erf: error function defined in Equation [5] 28 days. In this research, macropores are considered
to be pores with a diameter larger than 50 nm in
accordance with IUPAC 1984 (39).
() 2
( )
z
erf z = ⋅ ∫ exp −u 2 du [5] As can be observed, the results show that the use
π 0
of nSi and mSi may have a positive effect on improv-
ing the microstructure and reducing the internal pore
structure at both ages and in all mixtures as com-
Once the apparent diffusion coefficient is pared to the reference mixture (HPSCC). It could
obtained, the chloride penetration coefficient (KCl) be said that nSi causes a reduction in pore diameter,
is calculated following Equation [6] (13): though it is associated with a slight decrease in total
porosity. This effect might be explained by the for-
mation of a larger amount of C-S-H gels and gel
 Cth − Cb  pores due to the high pozzolanic activity of the nSi
K Cl = α ⋅ 12 ⋅ Dt  1 −  [6] addition in regard to its great surface/volume ratio,
 Cs − Cb 
reducing pore size. In the case of mixtures with mSi,
a significant reduction of total porosity is observed
(which might be due to a filling effect of the addi-
where: tion), while pore size remains unchanged. When nSi
a: unit conversion factor equal to 56157 and mSi additions are combined in ternary mixtures,
Dt: apparent diffusion coefficient for time t expressed an overlap of both effects is observed. In this case,
in cm2/s the behaviour might be explained by an enhanced
Cth: critical chloride concentration at which the pas- hydration process (due to the presence of the nSi
sive barrier of steel is broken, in % of cement weight addition) and the final packing efficiency (related to
Cs: chloride concentration on the surface of the con- the continuity of the particle size distribution of the
crete, expressed in % of cement weight components).
Cb: content of chloride from materials (initial chlo-
ride concentration) 3.2. Electrical resistivity
The chloride penetration coefficient enables the As explained above, in order to measure electri-
corrosion initiation time to be predicted for a given cal resistivity, a voltage was applied between two
concrete cover thickness according to Equation electrodes, and the current transmitted through
[7] (13): the water solution within the pores was measured.
This method was used mainly to determine chloride
2 penetrability. Polder in 2001 (40) described meth-
 d 
t= ods to assess concrete resistivity on site for various
 [7]
 K Cl  purposes related to reinforcement corrosion and
protection. Polder discussed practical aspects, pro-
posed guidelines for calibration and interpretation
where: and gave reference values for several cases. Basheer
t: corrosion initiation time in years et al. in 2002 (41) monitored the behaviour of con-
d: concrete cover thickness in mm cretes containing alternative cementitious materials
KCl: apparent chloride penetration factor in m/years (metakaolin, microsilica, pulverised fuel ash and
ground granulated blast furnace slag) in a chloride
The free chloride concentration of the same exposure regime. For this purpose, they measured
concrete specimens was measured experimentally the changes in resistance between pairs of stainless
following the recommendations of RILEM TC 178- steel electrodes embedded in the concrete at differ-
TMC (38). The free chloride concentration at sev- ent depths from the exposed surface. The method
eral depths enabled the effective chloride diffusion provided useful information that could be related to
coefficient to be obtained by following a procedure the chloride diffusion characteristic of the different
similar to that described for the apparent chloride mixtures. Rajabipour et al. (42) proposed a method
diffusion coefficient. The bound concentration was to interpret electrical conductivity measurements in
determined by subtracting the free chloride concen- concrete to assess water penetration. The procedure
tration from the total concentration. enables the position of moisture to be resolved, and

Materiales de Construcción 70 (339), July–September 2020, e221. ISSN-L: 0465-2746. https://doi.org/10.3989/mc.2020.10619

 Behaviour of a high-performance self-compacting concrete (HPSCC) • 7

Table 5. Total porosity and percentage of macro- and micropore values for all mixtures.

Total porosity Macropores Micropores

(%) (Ø > 50 nm) (Ø < 50 nm)
HAC 7 9.24 24.02 75.98
28 9.31 16.83 83.17
HAC[nSi]-2.5 7 9.91 23.49 76.51
28 8.10 13.32 86.68
HAC[nSi]-5 7 9.03 8.60 91.40
28 8.34 13.05 86.95
HAC[nSi]-7.5 7 9.98 8.49 91.51
28 7.43 11.87 88.13
HAC[mSi]-2.5 7 9.00 15.31 84.69
28 7.15 21.63 78.37
HAC[mSi]-5 7 9.09 15.07 84.93
28 6.97 27.03 72.97
HAC[mSi]-7.5 7 8.36 18.88 81.12
28 7.34 16.32 83.68
HAC[nmSi]-2.5/2.5 7 9.18 15.96 84.04
28 7.36 21.40 78.60
HAC[nmSi]-5/2.5 7 9.65 10.87 89.13
28 8.61 14.08 85.92
HAC[nmSi]-2.5/5 7 8.83 9.34 90.66
28 7.13 12.73 87.27

it is significantly superior to previous approaches. resistivity. The use of two sizes of additions may
McPolin et al. in 2005 (43) carried out an investiga- contribute to a better packing of the component
tion to monitor the rate of ingress of chlorides in particles of the material. The difficulty of com-
concrete with additions to mixtures. They obtained paction of mixtures with a high nanosilica content
the chloride profiles and determined the resistivity could explain the results.
of the concrete in order to assess if resistivity could In general, an increase in electrical resistivity was
be used to determine the presence of chlorides. The obtained with time. However, the resistivity values
results of the resistivity profiles corresponded well at 365 days of curing for the HAC[nSi]-7.5 and
with the chloride profiles. More recently, in 2014 HAC[nmSi]-2.5/5 mixtures were lower than those
Andrade et al. (44) developed an alternative model obtained at 28 days. This may be due to a notably
to calculate the service life of reinforced concrete low capillary absorption of the mixtures containing
based on electrical resistivity. a greater total amount of nanosilica addition, which
The average values of electrical resistivity at 7, 28 would result in especially good chloride resistance .
and 365 curing days for mixtures with nSi, mSi and It may also be observed that the addition of
ternary mixtures of nmSi are shown, respectively, in nanosilica produced a significant increase in resis-
Figure 1, Figure 2 and Figure 3. All concretes with tivity at seven days of curing: up to 675.3% for the
additions presented a significantly higher electrical HAC[nSi]-7.5 mixture, in comparison with HAC
resistivity than did the concrete without additions. alone. This result could be of significant impor-
An increase in resistivity was observed when the tance in applications where low chloride penetrabil-
addition percentage was increased, both for binary ity at an early age is required. The percentages of
and ternary mixtures, regardless of the type of addi- increment decreased at later ages of curing, and the
tion. Among the mixtures with a higher addition improvement lessened with a lower content of total
content, the highest resistivity figures were obtained addition. Even so, the smallest increase in resistivity
for the mixtures with 7.5% microsilica, and the low- was 42% for HAC[nSi]-2.5 at 365 days of curing.
est figures were found for the concrete with 7.5% In the case of mixtures with mSi, a significant
nanosilica. However, in the case of the ternary mix- increase in electrical resistivity from 7 to 28 days
tures, the mixture containing the highest amount of of curing was observed. The percentages of these
microsilica was not that which showed the highest increases in electrical resistivity with respect to the

Materiales de Construcción 70 (339), July–September 2020, e221. ISSN-L: 0465-2746. https://doi.org/10.3989/mc.2020.10619

8 • E. Reyes et al.

Figure 1. Average values and error bars of electrical resistivity at 7, 28 and 365 curing days for mixtures with nSi additions.

Figure 2. Average values and error bars of electrical resistivity at 7, 28 and 365 curing days for mixtures with mSi additions.

Materiales de Construcción 70 (339), July–September 2020, e221. ISSN-L: 0465-2746. https://doi.org/10.3989/mc.2020.10619

 Behaviour of a high-performance self-compacting concrete (HPSCC) • 9

Figure 3. Average values and error bars of electrical resistivity at 7, 28 and 365 curing days for mixtures with ternary
mixtures of nmSi.

reference range from 53.01% (HAC[mSi]-2.5) to the amount of mSi). This may explain the resistivity
142.73% (HAC[mSi]-7.5) at seven days of curing results; the best values at both ages came from the
and vary from 193.36% (HAC[mSi]-2.5) to 905% ternary mixtures, possibly because of their having a
(HAC[mSi]-7.5) at 28 days. continuous particle size distribution from the small-
Among the ternary mixtures, HAC[nmSi]-5/2.5 est particle size (addition) to the largest particle size
presented the highest electrical resistivity, with (coarse aggregate) in the dosage of the mixture.
925.16 Ωm after 28 days of curing. These concretes, Table 6 shows the classification of chloride pen-
like the binary mixtures with the addition of mSi, etrability at 7, 28 and 365 days of curing, according
experienced the highest increase in their electrical to the ASTM C1202 standard (34). The reference
resistivity from 7 to 28 days. concrete (HAC) ranges from moderate penetrability
The improvement in resistivity behaviour at 7 and 28 days to significantly low penetrability at
achieved with nSi at seven days may be attributed 365 days of curing. The concretes with the nanosilica
to pore refinement, a reduction in the percentage of addition are classified as having a significantly low
macropores due to the pozzolanic reaction and the penetrability as of seven days of curing. Binary mix-
packing effect observed at early ages. Binary mix- tures with microsilica exhibit low penetrability after
tures with nSi showed a pore size reduction propor- seven days of curing and notably low penetrability
tional to the amount of addition. In binary mixtures after 28 days of curing. Lastly, ternary mixtures are
with mSi, the microstructure produced had a signifi- classified as having an especially low penetrability as
cantly lower total porosity at 28 days due to a slower of seven days of curing.
pozzolanic reaction, reaching resistivity values simi- Given these results, it is noteworthy that con-
lar to binary mixtures with nSi at this age. The aver- cretes with nanosilica are classified in the same
age pore size of binary mixtures with mSi was similar level in both binary and ternary mixtures, providing
to that of the reference concrete. Lastly, the ternary notably low chloride penetrability from early ages.
mixtures developed a microstructure that is a result It is also important to mention that the classifica-
of the combination of both effects, exhibiting a tion places all concretes, including the reference con-
small average pore size (proportional to the amount crete, in the same category at 365 days. Therefore,
of nSi) and a lower total porosity (proportional to if only this category were considered, the use of

Materiales de Construcción 70 (339), July–September 2020, e221. ISSN-L: 0465-2746. https://doi.org/10.3989/mc.2020.10619

10 • E. Reyes et al.

Table 6. Classification of chloride penetrability at 7, 28 and 365 days of curing, according to the ASTM C1202 standard.

HAC HAC HAC

HAC HAC HAC HAC HAC HAC [nmSi]- [nmSi]- [nmSi]-
Mixtures HAC [nSi]-2.5 [nSi]-5 [nSi]-7.5 [mSi]-2.5 [mSi]-5 [mSi]-7.5 2.5/2.5 5/2.5 2.5/5
Resistivity (kΩ cm) 6.62 19.65 24.40 51.31 10.15 10.13 16.06 23.64 29.44 45.59
(7 days)
Chloride penetrability Moderate Very low Very low Very low Low Low Low Very low Very low Very low
(7 days)
Resistivity (kΩ cm) 8.33 24.30 33.00 76.96 24.43 44.03 83.70 57.52 92.52 87.40
(28 days)
Chloride penetrability Moderate Very low Very low Very low Very low Very low Very low Very low Very low Very low
(28 days)
Resistivity (kΩ cm) 21.87 31.05 39.16 68.07 38.02 52.11 98.60 66.66 95.03 72.77
(365 days)
Chloride penetrability Very low Very low Very low Very low Very low Very low Very low Very low Very low Very low
(365 days)

different additions would seem irrelevant after one In Figure 5 a significant decrease can be observed
year. This is because the classification establishes with the increase of the content of addition. The
a rather wide range for the real values of resistiv- decrease was 85.89% for HAC[mSi]-7.5. The coef-
ity in each category and assigns chloride penetra- ficient in this type of concrete presented lower val-
bility levels on the sole basis of whether the value ues than in concretes with nanosilica, except for
lies within this wide range . This should be noted, HAC[mSi]-7.5, which has a slightly higher coeffi-
since concretes that differ by as much as 350% in cient (16.73%) than HAC[nSi]-7.5.
electrical resistivity (as is the case of HAC[mSi]-7.5 Figure 6 shows that ternary specimens had val-
in comparison with the reference concrete) lie within ues of under 1×10−12 m2/s in all cases. In the case
the same category. In addition, this classification of HAC[nmSi]-2.5/2.5, with 5% total addition, the
does not allow the service life of a concrete to be chloride migration coefficient was similar to that
estimated. obtained with 7.5% of either addition separately.
The percentages of decrease of the chloride migra-
3.3. Chloride migration coefficient tion coefficient in these concretes are 84.45% for
HAC[nmSi]-2.5/2.5, 90.42% for HAC[nmSi]-5/2.5
Ion transport in mortar specimens was evaluated and 89.50% for HAC[nmSi]-2.5/5, in comparison
though chloride migration tests according to the with the reference concrete HAC. When comparing
NT BUILD 492 standard (18), which enabled the ternary mixtures with a 7.5% total addition content,
non-steady-state migration coefficient (Dnssm) to a slightly lower chloride migration coefficient was
be obtained. obtained with the higher percentage of nanosilica
Figure 4, Figure 5, and Figure 6 show the aver- addition, HAC[nmSi]-5/2.5.
age values of the chloride migration coefficient for To study how the amount of addition influences
three specimens containing mixtures with nSi and the chloride migration coefficient, three regression
mSi and ternary mixtures of nmSi, respectively, curves have been calculated, one for each type of
compared with HAC, all at 28 curing days. concrete with additions (see Figure 7). Here a good
As can been observed in Figure 4, the chloride fit may be observed by means of an exponential
migration coefficient decreases significantly in com- regression with an R2 of 0.995 for concrete with nSi,
parison with the reference with the increase in the 0.981 for concrete with mSi and 0.976 for concrete
nanosilica content. The values vary from 43.11% with both additions. Binary and ternary mixtures fol-
for HAC[nSi]-2.5 to 87.91% for HAC[nSi]-7.5. The low similar trends, indicating that the effectiveness of
chloride migration coefficient for the reference con- silica against migration depends mainly on its pozzo-
crete is 6.29x10−12 m2/s. Since this lies below 10×10−12 lanic activity. Results are similar for the same amount
m2/s, the concrete is considered notably resistant to of total addition, depending only slightly on particle
chloride penetration. A chloride migration coeffi- size. As the amount of total addition increases, the
cient below this reference value means the estimated chloride migration coefficient tends to present similar
service life of the concrete is high. Therefore, reduc- values. This may be observed in the case of all con-
ing the coefficient to values such as 0.76×10−12 m2/s cretes with the 7.5% addition: their values range from
might represent a significant increase in the useful 0.60·10‑12 m2/s for HAC[nmSi]-5/2.5 to 0.89·10−12 m2/s
life of concrete structures. for HAC[mSi]-7.5 and thus are close to overlapping.

Materiales de Construcción 70 (339), July–September 2020, e221. ISSN-L: 0465-2746. https://doi.org/10.3989/mc.2020.10619

 Behaviour of a high-performance self-compacting concrete (HPSCC) • 11

Figure 4. Average values and error bars of chloride migration coefficient at 28 curing days for mixtures with nSi compared with HAC.

Figure 5. Average values and error bars of chloride migration coefficient at 28 curing days for mixtures with mSi compared with HAC.

Materiales de Construcción 70 (339), July–September 2020, e221. ISSN-L: 0465-2746. https://doi.org/10.3989/mc.2020.10619

12 • E. Reyes et al.

Figure 6. Average values and error bars of chloride migration coefficient at 28 curing days for mixtures with ternary mixtures of
nmSi compared with HAC.

Figure 7. Regression curves of the influence of addition amount on the chloride migration coefficient.

Given the results, the use of addition provides might mean a qualitative advance in the durability
HPSCC with good resistance against chloride of the material, which would result in a significant
migration, as the published literature has shown increase of service life or a notable reduction of
(45–47). However, reducing the coefficient from the coating thicknesses of reinforced structural ele-
43% to 50% by using a 2.5% addition, as shown ments in aggressive environments.
in this work, represents a significant improvement. Other published works have identified relation-
This enhanced behaviour is due to the densification ships between the chloride migration coefficient
of the microstructure of the concrete (28, 29) and and electrical resistivity (44, 48). Figure 8 shows the

Materiales de Construcción 70 (339), July–September 2020, e221. ISSN-L: 0465-2746. https://doi.org/10.3989/mc.2020.10619

 Behaviour of a high-performance self-compacting concrete (HPSCC) • 13

potential correlation between these two parameters HAC. However, concrete with nSi has an amount of
for the concretes after 28 days of curing. As may be free chlorides similar to that of HAC, though it has
observed, the potential correlation presents a good a notably lower amount of total chlorides, which
fit (R2 = 0.9576). means that the concentration of combined chlorides
is lower than in the reference concrete. This figure
3.4. Total, free and binding chloride concentration shows that the addition of mSi leaves fewer free
profiles chlorides at the same depths, giving the concrete an
increase in its chloride binding capacity.
This section shows the results of chloride dif- The penetration profiles of the concretes with a
fusion testing according to the CEN/TS 12390-11 total addition of 5% are shown in Figure 10, con-
standard (36), studying how the total amount of firming the behaviour of the mSi. In this case, the
silica addition influences the penetration profile. amount of chlorides obtained with HAC[mSi]-5
Figures 9a, 9b, 9c, 10a, 10b, 10c, 11a, 11b and 11c and HAC[nmSi]-2.5/2.5 is similar. However, the
show the profiles of total (a), free (b) and combined percentage of free chlorides is lower for the case
(c) chlorides of the concretes with a total addition of binary concrete with mSi. The ternary mixture
of 2.5%, 5% and 7.5%, respectively, in comparison HAC[nmSi]-2.5/2.5 presented the second-lowest
with the reference concrete, HAC. percentage of free chlorides and the lowest content
Figure 9 shows that HAC[nSi]-2.5 and of combined chlorides.
HAC[mSi]-2.5 have a similar total chloride penetra- The case of the concretes with a total addition of
tion depth, significantly lower than the depth for 7.5% is shown in Figure 11. All cases presented sim-
ilar profiles of total, free and combined chlorides
and closely grouped depths of penetration. In this
case, with 7.5% nSi the chloride binding capacity
was close to the capacity with 7.5% mSi. Likewise,
regardless of the combination of additions, the
behaviour of these concretes was closely grouped.
This means that, for this amount of total addition,
the size of the addition does not significantly influ-
ence chloride penetration, which mainly depends on
the activity of the silica.

3.5. Chloride diffusion coefficient

Figure 12, Figure 13 and Figure 14 show the

chloride diffusion coefficient for mixtures with nSi,
mixtures with mSi and ternary mixtures of nmSi,
respectively, compared with HAC.
Figure 12 also shows a significant decrease in the
chloride diffusion coefficient in comparison with
Figure 8. Correlation curves between the chloride migration the coefficient for the reference concrete, in a range
coefficient and electrical resistivity. from 46.78% with HAC[nSi]-2.5 to 83.58% with

Figure 9. Chloride concentration profiles of the concretes with a total addition of 2.5% compared with HAC:
a) total chlorides, b) free chlorides, c) combined chlorides.

Materiales de Construcción 70 (339), July–September 2020, e221. ISSN-L: 0465-2746. https://doi.org/10.3989/mc.2020.10619

14 • E. Reyes et al.

Figure 10. Chloride concentration profiles of the concretes with a total addition of 5% compared with HAC: a) total chlorides,
b) free chlorides, c) combined chlorides.

Figure 11. Chloride concentration profiles of the concretes with a total addition of 7.5% compared with HAC: a) total chlorides,
b) free chlorides, c) combined chlorides.

Figure 12. Chloride diffusion coefficient for mixtures with nSi compared with HAC.

Materiales de Construcción 70 (339), July–September 2020, e221. ISSN-L: 0465-2746. https://doi.org/10.3989/mc.2020.10619

 Behaviour of a high-performance self-compacting concrete (HPSCC) • 15

HAC[nSi]-7.5. The chloride diffusion coefficient when the percentage of total addition is 7.5%. Then
behaviour of these concretes is somewhat different the concretes exhibit similar behaviour with compa-
from their chloride migration coefficient behaviour; rable degrees of total, free and combined chloride
when the amount of addition was increased from concentration, as well as similar chloride diffusion
2.5% to 5%, the decrease in the diffusion coefficient coefficients. This could mean that the behaviour of
was only 8.12%, whereas the effect was much stron- 7.5% additions does not depend on silica size.
ger in the case of the migration coefficient. Figure 15 shows the correlation between the chlo-
Figure 13 shows a similar trend in the case of ride diffusion coefficient and the chloride migration
binary concretes with nanosilica, although they coefficient for the concretes studied.
exhibit a nearly linear decrease of their chloride The figure shows a linear relationship with a
diffusion coefficient with the percentage of addi- fairly good fit, R2 = 0.9571, between the parameters.
tion. In this case, the percentages of decrease were Therefore, given that the information provided by
50.99% for HAC[mSi]-2.5, 75.54% for HAC[mSi]-5 the migration test follows the same trend as that of
and 86.09% for HAC[mSi]-7.5. the chloride diffusion test, such information could
Figure 14 shows that for ternary mixtures the be used in studies of concrete behaviour with respect
chloride diffusion coefficient was reduced by 78 to to chloride penetration. This is important due to the
87% in comparison with the value obtained for HAC. reduction in time, since use of the chloride migra-
The concrete with the lowest diffusion coefficient is tion test provides quantitative results in a maximum
the same that presented the lowest chloride migration of five days, whereas with a natural diffusion test
coefficient, HAC[nmSi]-5/2.5; it presented a value of 90 days of exposure are necessary before proceed-
2.03×10−13 m2/s. The other ternary mixtures exhibited ing to the chloride valuation to find the diffusion
similar values: 3.43×10−13 for HAC[nmSi]-2.5/2.5 and coefficient.
2.74×10−13 m2/s for HAC[nmSi]-2.5/5. Figure 16 shows the potential correlation
The study of total, free and combined chloride between resistivity and the chloride diffusion coef-
penetration presented in the previous section might ficient for the concretes studied. The fit is good
explain why the chloride diffusion coefficient is (R2 = 0.9605).
higher for binary concretes with nSi than for binary These accurate fits might allow relations to be
concretes with mSi when the percentage of addition established for inferring values, although the experi-
is 2.5% or 5%. This might be caused by a greater mental campaign would have to be extended in
chloride binding capacity provided by the addition order to validate the correlations in HPSCC with
of mSi to concrete. However, this trend changes micro- and nanosilica additions.

Figure 13. Chloride diffusion coefficient for mixtures with mSi compared with HAC.

Materiales de Construcción 70 (339), July–September 2020, e221. ISSN-L: 0465-2746. https://doi.org/10.3989/mc.2020.10619

16 • E. Reyes et al.

Figure 14. Chloride diffusion coefficient for mixtures of nmSi compared with HAC.

Figure 15. Correlation line between the chloride Figure 16. Correlation curve between resistivity and
diffusion coefficient and the chloride migration coefficient chloride diffusion coefficient for the mixtures.
for the mixtures.

Structure service life may be estimated using with nSi. As the amount of addition increases,
the model described in EHE 08 (13) and the chlo- the time to chloride penetration to the same depth
ride diffusion coefficient obtained by this study. increases, too. As a reference, a continuous line is
The chloride penetration coefficient KCl is calcu- represented at 20 mm of depth, so that the time
lated by Equation [7]. The calculation assumes the taken to start corrosion with a coating of this
worst-case values from EHE 08 (13): Cth= 0.6% thickness may be established. For HAC the cor-
and Cs = 0.7%. From the KCl calculated, the depth rosion initiation time is slightly greater than 100
of chloride penetration is obtained as a function of years, while when only 2.5% nSi is added the esti-
time and can be represented. mated time is 219 years. This indicates that the
Figure 17 shows the estimated chloride pene- use of nSi additions could reduce coating thick-
tration by natural diffusion as a function of time ness while maintaining the service life established
for the reference concrete and binary concretes by regulations.

Materiales de Construcción 70 (339), July–September 2020, e221. ISSN-L: 0465-2746. https://doi.org/10.3989/mc.2020.10619

 Behaviour of a high-performance self-compacting concrete (HPSCC) • 17

The estimated times of chloride penetration in Lastly, the estimate of chloride penetration for
binary concretes with mSi are shown in Figure 18. ternary concretes is shown in Figure 19. In this
As in the case of nSi, the penetration time decreases type of concrete, HAC[nmSi]-5/2.5 presented the
as the amount of addition in the concrete increases. best chloride resistance, as might be expected from
The corrosion initiation times for a coating of the diffusion coefficient results. The corrosion ini-
20 mm range from 238 years for HAC[mSi]-2.5 to tiation time for a coating of 20 mm is 543 years for
839 years for HAC[mSi]-7.5. HAC[nmSi]-2.5/2.5, 918 years for HAC[nmSi]-5/2.5
and 681 years for HAC[nmSi]-2.5/5.
Table 7 shows the estimated corrosion initiation
time for a structural element coating of 20 mm of each
concrete studied, listed according to the total addition.
Adding 2.5% nano- or microsilica yields similar
estimated corrosion initiation times regardless of
the size of the addition. Both additions double the
corrosion initiation time for HAC, although micro-
silica provides a slightly higher value.
In the case of concretes with a total addition
of 5%, the estimated corrosion initiation time for
binary concrete with mSi is 85% higher than that of
binary concrete with nSi. This might be due to the
difficulty of compaction of mixtures with a high nSi
content. However, the longest corrosion initiation
time is obtained with the ternary mixture, probably
due to a packing effect and the low porosity created
by a wider particle size distribution (29).
HAC[nmSi]-5/2.5 has the longest estimated ser-
vice life of all the mixtures featuring a total addi-
tion of 7.5%. This mixture represents the best
combination of nano- and microsilica addition for
resisting chloride penetration, probably due to the
optimisation of the packing effect, overcoming the
Figure 17. Estimated chloride penetration by natural
diffusion as a function of time for the reference concrete and difficulties in compaction caused by considerable
binary concretes with nSi. percentages of nSi.

Figure 18. Estimated chloride penetration by natural Figure 19. Estimated chloride penetration by natural
diffusion as a function of time for the reference concrete and diffusion as a function of time for the reference concrete and
binary concretes with mSi. ternary concretes with nmSi.

Materiales de Construcción 70 (339), July–September 2020, e221. ISSN-L: 0465-2746. https://doi.org/10.3989/mc.2020.10619

18 • E. Reyes et al.

The experimental results and the estimated ser- the addition of nanosilica. However, when classified
vice life may be interpreted by taking into account by resistivity, all mixtures, both binary and ternary,
that nano- and microsilica additions reduce the exhibit a notably low chloride penetrability.
amount of interconnected pores, since their incor- Similarly, binary mixtures prepared with micro-
poration in cement blends leads to a refinement of silica exhibit lower chloride migration coefficients
the porous network and changes the ionic concen- than binary mixtures with nanosilica. Significantly
tration in the pore solution (49, 50). These addi- lower values are obtained with ternary mixtures,
tions combine pozzolanic action, which increases however, even in the case of the mixture with the
the amount of C-S-H gels, with the filling effect, as smallest amount of each addition.
the small size of the addition particles blocks small The addition of microsilica provides the concrete
pores and prevents their interconnection. with a greater capacity for chloride combination
The results shown in Table 7 indicate that ter- than does the addition of nanosilica for percentages
nary mixtures might be the best for improving the less than or equal to 5%.
durable properties of concrete, since in the worst Binary mixtures prepared with microsilica have
case their estimated service life ranges from over five lower chloride diffusion coefficients than binary
times higher to over eight times higher than that of mixtures with nanosilica at 28 days. However, the
HAC. This could be explained by considering that combination of both additions in ternary mix-
in ternary mixtures pores are filled more efficiently, tures results in significantly low chloride diffusion
causing them to reduce interconnection to a greater coefficients, even at small addition ratios. Further
extent, as the particle size continuity is greater (29). future study including additional parameters, such
Additionally, ternary mixtures show good self- as chloride diffusion coefficients at different ages, is
compacting properties (see Table 3) while demand- necessary in order to examine the influence of nSi
ing a lower SP content than nSi binary mixtures that and mSi on the durability of concrete mixtures, in
contain the same total amount of additions, and particular at early ages, where the microstructure
they do not present the same handling difficulties as improvements obtained are promising.
blends with nS alone. Furthermore, these mixtures The chloride diffusion coefficients and the con-
present the best mechanical behaviour. centrations of total, free and combined chlorides
were similar in each concrete containing a total
4. CONCLUSIONS addition of 7.5%, regardless of the amount of silica
addition.
HPSCC with the addition of nano- and microsil- The addition of nano- and microsilica in binary
ica in binary and ternary mixtures presents signifi- and ternary mixtures leads to notably low chloride
cant improvements in chloride-resistant behaviour. diffusion coefficients, with a significant increase in
This improvement is in general proportional to the the estimated service life according to Spanish regu-
total content of addition. The use of nSi and mSi lations. These results might also allow the thickness
may have a positive effect on improving the micro- of cover to reinforcement to be reduced without
structure and reducing the internal pore structure leading to negative consequences for structure
from early stages. durability.
In binary mixtures the addition of microsilica
presents higher values of electrical resistivity than ACKNOWLEDGEMENTS

Table 7. Estimated corrosion initiation The authors wish to express their gratitude to
time for a 20-mm coating. the Spanish Ministry of Science and Innovation
Corrosion Increase with
(Ministerio de Ciencia e Innovación), project
Amount of initiation respect to RTI2018-100962-B-I00, for financial support.
addition Concrete time (years) HAC (years)
0 HAC 116.72 0 REFERENCES
2.5 HAC[nSi]-2.5 219.31 102.59
1. Siddique, R. (2013) Compressive strength, water absorp-
2.5 HAC[mSi]-2.5 238.13 121.41 tion, sorptivity, abrasion resistance and permeability
5 HAC[nSi]-5 258.79 142.07 of self-compacting concrete containing coal bottom
ash. Construc. Build. Mat. 47, 1444–1450. https://doi.
5 HAC[mSi]-5 477.22 360.49 org/10.1016/j.conbuildmat.2013.06.081.
5 HAC[nmSi]-2.5/2.5 543.01 426.29 2. Łaźniewska-Piekarczyk, B. (2014) The methodology for
assessing the impact of new generation superplasticiz-
7.5 HAC[nSi]-7.5 710.88 594.15 ers on air content in self-compacting concrete. Construc.
7.5 HAC[mSi]-7.5 839.01 722.29 Build. Mat. 53, 488–502. https://doi.org/10.1016/j.
conbuildmat.2013.11.092.
7.5 HAC[nmSi]-5/2.5 918.46 801.74 3. Mohammed, M.K.; Dawson, A.R.; Thom, N.H. (2014)
7.5 HAC[nmSi]-2.5/5 681.36 564.64 Carbonation of filler typed self-compacting concrete and
its impact on the microstructure by utilization of 100% CO2

Materiales de Construcción 70 (339), July–September 2020, e221. ISSN-L: 0465-2746. https://doi.org/10.3989/mc.2020.10619

 Behaviour of a high-performance self-compacting concrete (HPSCC) • 19

accelerating techniques. Construc. Build. Mat. 50, 508–516. 21. Li, Z.; Peng, J.; Ma, B. (1999) Investigation of chloride
https://doi.org/10.1016/j.conbuildmat.2013.09.052. diffusion for high-performance concrete containing fly
4. Druta, C.; Wang, L.; Lane, D.S. (2014) Tensile strength ash, microsilica and chemical admixtures. ACI Mater. J.
and paste-aggregate bonding characteristics of self-consol- 96, 391–396.
idating concrete. Construc. Build. Mat. 55, 89–96. https:// 22. Björnstrom, J.; Martinelli, A.; Matic, A.; Börjesson, L.;
doi.org/10.1016/j.conbuildmat.2014.01.010. Panas, I. (2004) Accelerating effects of colloidal nano-
5. Jalal, M.; Pouladkhan, A.; Harandi, O.F.; Jafari, D. (2015) silica for beneficial calcium-silica-hydrate formation in
Comparative study on effects of Class F fly ash, nano silica cement. Chem. Phys. Lett. 392 [1–3], 242–248. https://doi.
and silica fume on properties of high performance self- org/10.1016/j.cplett.2004.05.071.
compacting concrete. Construc. Build. Mat. 94, 90–104. 23. Jo, B-W.; Kim, C-H.; Tae, G-H.; Park. J-B. (2007)
https://doi.org/10.1016/j.conbuildmat.2015.07.001. Characteristics of cement mortar with nano-SiO2 par-
6. Güneyisi, E.; Gesoglu, M.; Al-Goody, A.; Ipek, S. (2015) ticles. Construc. Build. Mat. 21, 1351‑1355. https://doi.
Fresh and rheological behavior of nano-silica and fly ash org/10.1016/j.conbuildmat.2005.12.020.
blended self-compacting concrete. Construc. Build. Mat. 95, 24. Kawashima, S.; Hou, P.; Corr, D.J.; Shah, S.P. (2013)
29–44. https://doi.org/10.1016/j.conbuildmat.2015.07.142. Modification of cement-based materials with nanopar-
7. Hemalatha, T.; Ram Sundar, K.R.; Ramachandra Murthy, ticles. Cem. Concr. Compos. 36, 8‑15. https://doi.
A.; Iyer, N. R. (2015) Influence of mixing protocol on org/10.1016/j.cemconcomp.2012.06.012.
fresh and hardened properties of self-compacting con- 25. Rong, Z.; Sun, W.; Xiao, H.; Jiang, G. (2015) Effects of
crete. Construc. Build. Mat. 98, 119–127. https://doi. nano-SiO2 particles on the mechanical and microstruc-
org/10.1016/j.conbuildmat.2015.08.072. tural properties of ultra-high performance cementitious
8. Puentes, J.; Barluenga, G.; Palomar, I. (2015) Effect of composites. Cem. Concr. Compos. 56, 25‑31. https://doi.
silica-based nano and micro additions on SCC at early org/10.1016/j.cemconcomp.2014.11.001.
age and on hardened porosity and permeability. Construc. 26. Balapour, M.; Joshaghani, A.; Althoey, F. (2018) Nano-
Build. Mat. 81, 154‑161. https://doi.org/10.1016/j. SiO2 contribution to mechanical, durability, fresh and
conbuildmat.2015.02.053. microstructural characteristics of concrete: A review.
9. Leung, H.Y.; Kim, J.; Nadeem, A.; Jaganathan, J.; Anwar, Construc. Build. Mat. 181, 27–41. https://doi.org/10.1016/j.
M.P. (2016) Sorptivity of self-compacting concrete contain- conbuildmat.2018.05.266.
ing fly ash and silica fume. Construc. Build. Mat. 113, 369– 27. Gesoglu, M.; Güneyisi, E.; Asaad, D.S.; Muhyaddin, G.F.
375. https://doi.org/10.1016/j.conbuildmat.2016.03.071. (2016) Properties of low binder ultra-high performance
10. Tian, J.; Wang, W.; Du, Y. (2016) Damage behaviors of cementitious composites: Comparison of nanosilica and
self-compacting concrete and prediction model under cou- microsilica. Construc. Build. Mat. 102, 706–713. https://
pling effect of salt freeze-thaw and flexural load. Construc. doi.org/10.1016/j.conbuildmat.2015.11.020.
Build. Mat. 119, 241–250. https://doi.org/10.1016/j. 28. Bernal, J.; Reyes, E.; Massana, J.; León, N.; Sánchez, E.
conbuildmat.2016.05.073. (2018) Fresh and mechanical behavior of a self-compact-
11. Asadollahi, S.; Saeedian, A.; Dehestani, M.; Zahedi, F. ing concrete with additions of nano-silica, silica fume and
(2016) Improved compressive fracture models for self-con- ternary mixtures. Construc. Build. Mat. 160, 196–210.
solidating concrete (SCC). Construc. Build. Mat. 123, 473– https://doi.org/10.1016/j.conbuildmat.2017.11.048.
480. https://doi.org/10.1016/j.conbuildmat.2016.07.030. 29. Massana, J.; Reyes, E.; Bernal, J.; León, N.; Sánchez-
12. Specification and Guidelines for Self-Compacting Espinosa, E. (2018) Influence of nano- and micro-silica
Concrete. (2005) European Federation for Specialist additions on the durability of a high-performance self-
Construction Chemicals and Concrete Systems EFNARC. compacting concrete. Construc. Build. Mat. 165, 93–103.
13. Instrucción de Hormigón Estructural (EHE). (2010) https://doi.org/10.1016/j.conbuildmat.2017.12.100.
Ministerio de Fomento, Secretaría General Técnica, Madrid. 30. Ryan, P.C.; O’Connor, A. (2016) Comparing the dura-
14. Nepomuceno, M.C.S.; Oliveira, L.A.; Lopes, S.M.R. bility of self-compacting concretes and conventionally
(2014) Methodology for mix design of self-compacting vibrated concretes in chloride rich environments. Construc.
concrete using different mineral additions in binary blends Build. Mat. 120, 504–513. https://doi.org/10.1016/j.
of powders. Construc. Build. Mat. 26, 317–326. https://doi. conbuildmat.2016.04.089.
org/10.1016/j.conbuildmat.2011.06.027. 31. UNE-EN 197-1:2011 Cement - Part 1: Composition, spec-
15. Shi, C.; Wu, Z.; Lv, K.; Wu, L. (2015) A review on mixture ifications and conformity criteria for common cements.
design methods for self-compacting concrete. Construc. Asociación Española de Normalización y Certificación
Build. Mat. 84, 387–398. https://doi.org/10.1016/j. AENOR. www.aenor.es.
conbuildmat.2015.03.079. 32. UNE 12620:2003+A1:2009 Aggregates for concrete.
16. Ye, G.; Liu, X.; De Schutter, G.; Poppe, A.-M.;, Taerwe, L. Asociación Española de Normalización y Certificación,
(2007) Influence of limestone powder used as filler in SCC AENOR. www.aenor.es.
on hydration and microstructure of cement pastes. Cem. 33. UNE 83988-1:2008 Concrete durability. Test meth-
Concr. Compos. 29, 94–102. https://doi.org/10.1016/j. ods. Determination of the electrical resistivity. Part 1:
cemconcomp.2006.09.003. Direct test (reference method), Asociación Española
17. Barluenga, G.; Palomar, I.; Puentes, J. (2015) Hardened de Normalización y Certificación AENOR. www.
properties and microstructure of SCC with mineral addi- aenor.es.
tions. Construc. Build. Mat. 94, 728–736. https://doi. 34. ASTM C1202-18 (2018) Standard Test Method for
org/10.1016/j.conbuildmat.2015.07.072. Electrical Indication of Concrete’s Ability to Resist
18. Jalal, M.; Pouladkhan, A.; Harandi, O.F.; Jafari, D. (2015) Chloride Ion Penetration. ASTM International.
Comparative study on effects of Class F fly ash, nano silica 35. NT BUILD 492 (1999) Concrete, mortar and cement-
and silica fume on properties of high performance self- based repair materials: Chloride migration coefficient from
compacting concrete. Construc. Build. Mat. 94, 90–104. non-steady-state migration experiments. Nordtest. ISSN
https://doi.org/10.1016/j.conbuildmat.2015.07.001. 0283–7153.
19. Owsiak, Z.; Grzmil, W. (2015) The evaluation of the influ- 36. CEN/TS 12390-11:2010 Testing hardened concrete - Part
ence of mineral additives on the durability of self-com- 11: Determination of the chloride resistance of con-
pacting concretes. KSCE J. Civil Engineer. 19, 1002–1008. crete, unidirectional diffusion. European Committee for
https://doi.org/10.1007/s12205–013-0336-7. Standardization.
20. Ghafari, E.; Costa, H.; Júlio, E.; Portugal, A.; Durães, 37. UNE-EN 14629:2007 Products and systems for the pro-
L. (2014) The effect of nanosilica addition on flowabil- tection and repair of concrete structures - Test methods
ity, strength and transport properties of ultra-high per- - Determination of chloride content in hardened concrete,
formance concrete. Mater. Design. 59, 1‑9. https://doi. Asociación Española de Normalización y Certificación.
org/10.1016/j.matdes.2014.02.051. AENOR. http://www.aenor.es.

Materiales de Construcción 70 (339), July–September 2020, e221. ISSN-L: 0465-2746. https://doi.org/10.3989/mc.2020.10619

20 • E. Reyes et al.

38. RILEM Publications. (2002). Analysis of total chloride 45. Romero Mendoza, H.R. (2011) Deterioro del hormigón
content in concrete. Mater. Struct. 35, 583–585. https://doi. sometido a ensayos acelerados de hielo-deshielo en presen-
org/10.1007/BF02483128. cia de cloruros, PhD dissertation, Escuela Técnica Superior
39. Sing, K.S.W.; Everett, D.H.; Haul, R.A.W.; Moscou, de Ingeniería de Caminos, Canales y Puertos, Universidad
L.; Pierotti, R.A.; Rouquerol, J.; et al. (1985) IUPAC Politécnica de Madrid, Madrid, Spain.
(Recommendations 1984), Reporting physisorption data 46. Jalal, M.; Mansouri, E.; Sharifipour, M.; Pouladkhan,
for gas/solid systems with special reference to the determi- A.R. (2012) Mechanical, rheological, durability and
nation of surface area and porosity. Pure Appl. Chem. 57, microstructural properties of high performance self-com-
603–619. https://doi.org/10.1351/pac198557040603. pacting concrete containing SiO2 micro and nanoparticles.
40. Polder, R.B. (2001) Test methods for on-site measurement Mater. Design. 34, 389–400. https://doi.org/10.1016/j.
of resistivity of concrete. RILEM TC-154 technical recom- matdes.2011.08.037.
mendation. Constr. Build. Mater. 15, 125–131. https://doi. 47. Li, L.G.; Zhu, J.; Huang, Z.H.; Kwan, A.K.H.; Li, L.J.
org/10.1016/S0950-0618(00)00061-1. (2017) Combined effects of micro-silica and nano-silica on
41. Basheer, P.; Gilleece, P;. Long, A.; Mc Carter, W. (2002) durability of mortar. Construc. Build. Mat. 157, 337–347.
Monitoring electrical resistance of concretes containing https://doi.org/10.1016/j.conbuildmat.2017.09.105.
alternative cementitious materials to assess their resistance 48. Zahedi, M.; Ramezanianpour, A.A.; Ramezanianpour,
to chloride penetration. Cem. Concr. Compos. 24, 437–449. A.M. (2015) Evaluation of the mechanical properties and
https://doi.org/10.1016/S0958-9465(01)00075-0. durability of cement mortars containing nanosilica and
42. Rajabipour, F.; Weiss, J.; Shane, J.D.; Mason, T.O.; Shah, rice husk ash under chloride ion penetration. Construc.
S.P. (2005) Procedure to interpret electrical conductivity Build. Mat. 78, 354–361. https://doi.org/10.1016/j.
measurements in cover concrete during rewetting. J. Mater. conbuildmat.2015.01.045.
Civ. Eng. 17, 586–594. https://doi.org/10.1061/(ASCE)​ 49. Alonso-Domínguez, D.; Álvarez-Serrano, I.; Reyes, E.;
0899-1561(2005)17:5(586). Moragues, A. (2017) New mortars fabricated by electro-
43. McPolin, D.; Basheer, P.A.M.; Long, A.E.; Grattan, K.T.V.; static dry deposition of nano and microsilica additions:
Sun, T. (2005) Obtaining progressive chloride profiles in Enhanced properties. Construc. Build. Mat. 135, 186–193.
cementitious materials. Constr. Build. Mater. 19, 666–673. https://doi.org/10.1016/j.conbuildmat.2017.01.011.
https://doi.org/10.1016/j.conbuildmat.2005.02.015. 50. Olsson, N.; Lothenbach, B.; Baroghel-Bouny, V.; Nilsson,
44. Andrade, C.;. D’Andrea, R.; Rebolledo, N. (2014) Chloride L-O. (2018) Unsaturated ion diffusion in cementitious
ion penetration in concrete: the reaction factor in the elec- materials – The effect of slag and silica fume. Construc.
trical resistivity model. Cem. Concr. Compos. 47, 41–46. Build. Mat. 108, 31–37. https://doi.org/10.1016/j.cemconres.​
https://doi.org/10.1016/j.cemconcomp.2013.09.022. 2018.03.007.

Materiales de Construcción 70 (339), July–September 2020, e221. ISSN-L: 0465-2746. https://doi.org/10.3989/mc.2020.10619