Construction and Building Materials 199 (2019) 50–62

Contents lists available at ScienceDirect

Construction and Building Materials

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

Special Issue
Banthia-Basheer Special Issue on Advances in Science and Technology of Concrete
Self-compacting concretes using fly ash and fine limestone powder:
Shrinkage and surface electrical resistivity of equivalent mortars
A. Duran-Herrera a,⇑, J. De-León-Esquivel a, D.P. Bentz b, P. Valdez-Tamez a
a
Universidad Autónoma de Nuevo León, Facultad de Ingeniería Civil, San Nicolás de Los Garza, N.L. 66455, Mexico
b
Engineering Laboratory at the National Institute of Standards and Technology (NIST), Gaithersburg, MD 20899-8615, USA

h i g h l i g h t s

New knowledge about the substitution of OPC by fly ash(FA) and limestone filler(LF).

Appropriate combinations of FA + LF counteracts the setting delays in FA concretes.

Appropriate combinations of FA+LF improve resistivity and volume stability.

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

Article history: Self-compacting concrete (SCC) has become a preferred option for many projects that should satisfy strict
Received 23 December 2016 fresh stage properties that are of major concern in quality assurance. To ensure stable and robust fresh
Received in revised form 10 November 2018 stage properties, typically a significant amount of fine materials is incorporated, but this often increases
Accepted 23 November 2018
shrinkage. For this purpose, fly ash (FA) has been used, but because it can induce delays in times of set-
Available online 11 December 2018
ting, it is not extensively used. Under this scenario, micro limestone powders (L) have been effectively
used to counteract the delays in the times of setting of concrete with high volumes of fly ash.
Keywords:
For a fixed water/powder ratio equivalent to a water-to-cement ratio of 0.40 in a 100% cement mixture,
Fly ash
Micro limestone powder
a total of thirteen mortars were produced to evaluate the synergetic effects of twelve portland cement
Self-compacting concrete substitutions by FA + L on times of setting, compressive strength, shrinkage and electrical resistivity.
Setting times Results indicate appropriate FA + L combinations to counteract delays in times of setting and to signifi-
Shrinkage cantly improve electrical resistivity and volume stability.
Surface electrical resistivity Ó 2018 Published by Elsevier Ltd.

1. Introduction improve the densification of the cementitious matrix, which leads

to a decrease of the concrete’s permeability and consequently to an
Since its first use in the late 1980s, Self-Consolidating Concrete increase of its durability [2].
(SCC) has been used as a replacement of conventional Vibrated In terms of durability, the incorporation of fly ash in concrete
Concrete (VC) because of its numerous advantages, including labor (typically in substitutions of 25–30% by mass of cement), provides
savings, construction cost reductions, and ease of placing in heavily several benefits that could be associated to the following beneficial
reinforced members, among others [1]. One of the most important effects of this SCM-filler in fresh and hardened stages of concrete:
differences between SCC and VC is the incorporation of higher workability improvement, reduction in water demand, less segre-
amounts of fillers in the SCC, as this type of concrete demands a gation, reduced heat of hydration, increased long term strength,
higher powder content for its formulation. significant reduction of permeability and chloride resistance,
These fillers often include supplementary cementitious materi- reduced expansion due to alkali-silica reaction, increased sulfate
als (SCMs) in replacement of portland cement to improve the fresh resistance, among others [3,4].
and hardened stage properties of the SCC, since these powders can Although these benefits, fly ash in concrete are not readily used
by the construction industry, mainly due to a decrease in strength
at early ages. Even tough with the incorporation of high amounts of
⇑ Corresponding author.
fly ash, this problem could be overcome, as long as a significant
E-mail address: [email protected] (A. Duran-Herrera).

https://doi.org/10.1016/j.conbuildmat.2018.11.191
0950-0618/Ó 2018 Published by Elsevier Ltd.
 A. Duran-Herrera et al. / Construction and Building Materials 199 (2019) 50–62 51

reduction of the water/binder ratio, and a superplasticizer to pro- 3. Procedure

vide also an appropriate consistency are incorporated. The use of
3.1. Materials and mixture proportions
a wide range of Portland cement substitutions by fly ash were
widely studied in research projects and field applications in the The materials used in this study are: an ASTM C150 [14]. Type I Portland cement
1990s, in concrete mixtures with key parameters such as the Port- (PC); an ASTM C618 [15] type F fly ash (FA) and a commercial micro limestone pow-
land cement substitution by fly ash per cubic meter, the amount of der (L), the latter with an average particle size of 0.7 mm (according to the manufac-
turer’s technical data sheet). The physical and chemical characteristics of these
fly ash per cubic meter, the water-binder ratio and the 28 day com-
powders are given in Table 1. Also, the corresponding particle size distributions
pressive strength, in a range between 53 and 58%, 125 and 230 kg/ (PSDs) as measured using laser diffraction are shown in Fig. 1. Table 2 presents
m3, 0.32 and 0.45 and 20 to 40 MPa, respectively. The optimum fly the parameters associated with the PSDs of PC, FA and L; each parameter indicates
ash content will vary on a case-by-case basis. Fly ash contents up the particle size of the materials for which 10%, 50% and 90% of the total sample fall
to 50% may be suitable for most elements without any adverse below and are referenced as D10, D50 (also known as the median), and D90, respec-
tively. From these results, it can be observed that the FA particles are 4 times coar-
impact in construction schedule [4]. Fly ash contents of 40–50%
ser than the portland cement particles; while the micro limestone particles are up
are suitable for slabs, that merely require a broom finish, but the to 11 times finer than the portland cement particles.
level of replacement should be reduced (25%) for slabs that A limestone from Nuevo León, Mexico that meets ASTM C33 requirements [16]
requires trowels to avoid unwanted delays in finishing [5]. was used as coarse and fine aggregates. To obtain the target flowability and stabil-
ity, two ASTM C494 [17] admixtures were used: a Type F water reducer and a Type S
The low water-to-cementitious materials ratios commonly
anti-wash out admixture as a viscosity modifier with a density or specific gravity of
used for SCC range from 0.3 to 0.40 [6], and in combination with 1.10 and 1.23, and a solids content of 42.1 and 44.5% respectively.
a higher powder content, often lead to an increase of the total Reference mortar proportions were established from a Self-Consolidating Con-
shrinkage of the concrete [7], which is typically the combination crete (SCC) with a water-to-cement (w/c) ratio of 0.4 by mass and with optimized
of two processes nominally present at different stages of the life proportions to meet the following target properties: slump flow of 56 cm ± 1 cm
according to ASTM C1611 [18], static segregation of 2.6% ± 0.2% using the segrega-
of concrete: autogenous shrinkage and drying shrinkage. The first
tion column according to ASTM C1610 [19], and a flow of 51 cm ± 1 cm for the
one is more significant at early ages, and concerns volumetric J-Ring passing ability test determined according to ASTM C1621 [20]. Reference
changes that involve self-desiccation inducing capillary stresses mortar proportions kept a w/c of 0.4, and were obtained by the Concrete Equivalent
and leading consequently, to cracking. The second one occurs at Mortar (CEM) method [21]. Self-Consolidating Concrete (Ref SCC) and reference
a later age along with the contribution of temperature and con- concrete equivalent mortar (Ref CEM) proportions are given in Table 3, where
aggregates are presented in dry condition, and where water labeled as reaction
crete restraint and is due to moisture loss to the surrounding tap water corresponds to the water dosed for the hydration of the Portland cement,
environment by evaporation [8,9]. Shrinkage can cause cracking according to the fixed target water to-cement ratio (effective water).
that could lead to the failure of the concrete structure, because In addition to Ref CEM made just with cement, twelve (12) mortars, distributed
cracks in concrete will reduce mechanical properties and durabil- in three groups, were produced to evaluate different combinations of PC + FA + L. In
order to eliminate volumetric variations attributed to the significant differences in
ity [10].
the specific gravities of the powders (PC, FA and L), based on the Ref CEM mixture,
Due to the advantages offered by the incorporation of industrial volumes of water and powders were maintained constant in all of the mixtures. The
byproducts into concrete, fly ash (FA) obtained from coal combus- composition of the paste fraction of these mortars varies according to the propor-
tion is one of the most commonly used SCMs for concrete, chiefly tions presented in the experimental matrix in Table 4.
because of the benefits it provides to the rheological, mechanical
and durability performance, and for its contributions to concrete 3.2. Experimental methods

industry sustainability. However, the early-age reactivity of FA is Fresh stage characterization consisted of: flow as a measure of the mortar work-
generally lower than that of portland cement, such that it con- ability with the cone specified in ASTM C 1437 [22]; times of setting according to
tributes to delays in times of setting and to slower strength ASTM C 403 [23]; unit weight and air content obtained through the procedures
development. described in ASTM C 185 [24]. When determining the unit weight and air content,
the mold was filled in one layer and 12 S were applied around the perimeter of the
In high volume fly ash concretes with 40–60% portland cement
mold with a rubber head mallet weighing 0.6 kg ± 0.2 kg. Results of single tests are
replacement, the implementation of judicious combinations of a reported in Table 5; all the flow results meet the targeted consistency of
fine limestone powder with a average particle size of about one 17 cm ± 1 cm. For the air content, the precision and bias section of ASTM C185
micrometer (5–15% of the total cementitious content) in combina- reports a single-operator within laboratory standard deviation of 0.56% for air con-
tion with fly ash, could diminish, delay or lead to similar setting tents in the range of 8–19%. No uncertainty is reported for unit weight
measurements.
times and to a superior potential durability [11–13]. After fresh stage characterization, mortar test specimens were cast and cured
Based on several research or field application reports, this for the determination of the following hardened stage properties:
research work evaluated twelve substitutions of Portland cement
by a binary powder mix of fly ash (FA) and limestone filler (L) for (1) Compressive strength: measured on triplicate specimens at 7 d, 28 d and
56 d using 50 mm standard cubes. Due to the self-consolidating nature of
a fixed total substitutions of 70, 55 and 40% for mixtures of groups
the mortar, all of the specimens were cast with only one layer of mortar
A, B and C respectively, with FA and L substitutions in the range of
15–60% and 5–15% respectively.
Table 1
Chemical composition and specific gravity of PC, FA and L.

2. Research significance Component (%) PC FA L

SiO2 17.66 67.48 0.20
Limestone powder has been used previously in order to coun- Al2O3 4.72 23.94
Fe2O3 2.26 4.63
teract the delays in setting times encountered in most concretes
CaO 63.68 1.76
made with FA; however, only a few of these works have evaluated MgO 1.20 1.16
the effects of the combination of limestone powder and FA on the SO3 5.12 –
shrinkage and durability of concrete. Results of shrinkage, both Na2O 0.66 1.14
autogenous and drying shrinkage, as well as a durability index K2O 0.74 0.99
TiO2 0.24 0.94
(surface electrical resistivity) are presented in this study in order P2O5 0.09 0.03
to demonstrate the beneficial synergistic effects of a FA produced MnO 0.08 –
in Northeast Mexico and a commercial micro limestone powder (SiO2 + Al2O3 + Fe2O3) % 96.05
(L), regarding times of setting, compressive strength, volume sta- CaCO3, % ˃99
Specific gravity, g/cm3 3.10 2.03 2.68
bility and surface electrical resistivity.
52 A. Duran-Herrera et al. / Construction and Building Materials 199 (2019) 50–62

Fig. 1. Particle size distributions of PC, FA and L.

Table 2
calculated and estimated for ages of 28 d and 52 w, respectively, according
D10, D50 and D90 values in mm of PC, FA and L.
to the procedures described in the standard test method (Section 12, note 2
Material D10 (mm) D50 (mm) D90 (mm) and Fig. 1).
(4) Surface electrical resistivity: was obtained with a four-point resistivity
PC 3.0 18.4 46.2 meter operating on the principle of the Wenner probe and with 38 mm
FA 14.0 71.2 242.1 inter probe spacing according to the AASHTO TP 95-11 specification [28].
L 0.6 1.7 4.7 For this test, 3 cylindrical specimens of 10 cm in diameter and 20 cm in
length were made for each mortar mixture and were stored in water satu-
rated with calcium hydroxide. Measurements were taken at ages of (1, 2, 3,
4, 5, 7, 14, 21, 28, 35, 42, 49 and 56) d. The measured electrical resistance
(O) was obtained as the average of 12 measurements (with 4 measurements
without any compaction energy; afterwards, the standard procedure per cylinder). Thereafter, surface electrical resistivity was calculated with
described in ASTM C109 [25] was followed for storage, curing and for the Eq. (1):
determination of compressive strength.
(2) Autogenous strain: measured on triplicate specimens according to the stan-
dard procedure from ASTM C1898 [26]. For each mortar mixture, the first
q¼Rk ð1Þ
reading was taken at the final time of setting, and thereafter up to an age where R is the measured resistance and k is the geometric correction factor used for
of 28 d. Because the standard procedure of this test requires significant care, cylindrical specimens as shown in Eq. (2):
precision and permanent monitoring, intermediate mixtures 1PC-FA-10L,
2PC-FA-10L and 3PC-FA-10L were not considered for this test since their 2pa
k¼ ð2Þ
results should be between those exhibited by the corresponding 5L and 1:10  0:730
d=2
þ ðd=a
7:82
Þ2
15L mixtures.
(3) Drying shrinkage: Measurements were made on triplicate specimens and This formula is only valid when d=a  4 and L=a  5, where d is the diameter of
ultimate drying shrinkage was determined according to the procedure the cylinder (mm), a is the inter probe distance of the Wenner apparatus and L is the
described in ASTM C596 [27]. Average and ultimate drying shrinkage were length of the cylinder (mm) [29].

Table 3
Reference concrete equivalent mortar proportions (Ref CEM) per cubic meter of concrete/mortar, obtained from SF1 Self Consolidating Concrete (Ref SCC). Aggregates in dry
condition.

Materials Ref SCC Ref CEM

Tap water Effective, L/m3 166.2 221.7
Total, L/m3 191.3 255.1
3
Type I Portland cement (kg/m ) 420.0 560.2
HRWA Amount L/m3 2.7 3.6
Dosage (ml/kg) of PC 6.5 6.5
%* 0.71 0.71
VMA Amount L/m3 0.10 0.13
Dosage ml/m3 100.0 133.4
%* 2.9  102 2.9  102
Coarse aggregate kg/m3 696.7 –
Fine aggregate kg/m3 1136.3 1532.2
*
Percentage of the total binder weight.
 A. Duran-Herrera et al. / Construction and Building Materials 199 (2019) 50–62 53

Table 4 result in a reduction of the unit weight by 5.4, 5.8 and 6.9 respec-
Experimental matrix and powder proportions in mortars, % in volume. tively. When the binary cementitious system of PC + FA was fur-
Group Identification PC FA L ther replaced with micro limestone filler (L) in additions of 5, 10
REF REFERENCE 100 0 0 and 15%, the unit weight reduction resulted between 4.1 to 2.7,
4.4 to 2.7 and 4.9 to 1.9 respectively. While the flows for all the
A 1PC-FA 70 30 0
1PC-FA-5L 70 25 5 mortar were similar with results in the range of 16.5–18, a general
1PC-FA-10L 70 20 10 analysis of these results confirm that fly ash (FA) could increase the
1PC-FA-15L 70 15 15 consistency (mainly group C, mixtures 3PC-FA and 3PC-FA-SL) and
B 2PC-FA 55 45 0 suggest that limestone filler (L) slightly increases the viscosity of
2 PC-FA-5L 55 40 5 the paste phase, improving accordingly the stability of the mixture,
2 PC-FA-10L 55 35 10 the benefit is derived in the flow reduction for mixtures PA-FA-ISL
2 PC-FA-15L 55 30 15
(Groups A, A and C).
C 3PC-FA 40 60 0 The entrapped air content of 3% obtained for reference mixture
3PC-FA-5L 40 55 5
3PC-FA-10L 40 50 10
is in agreement with the expected approximate volume suggested
3PC-FA-15L 40 45 15 by ACI 211.1 (as the maximum aggregate size decreases air content
increases). Results for groups A, B and C, suggest that the addition
of FA significantly increases the air content from 3.0 to 9.9, 9.1 and
Table 5 9.1% for substitutions of PC by FA of 30, 45 and 60%.
Measured fresh stage properties of mortars. This adverse effect is attributed to a collateral reaction between
the fly ash and the policarboxilate superplasticizer and/or the
Group Identification Flow (cm) Unit weight (kg/m3) Air content (%)
viscosity-modifying admixture. The substitution of FA by L sup-
REF REFERENCE 17.0 2188 3.0
ports this assertion since as this substitution increases the trapped
A 1PC-FA 16.0 2070 9.9 air content decreases. It is well known in the technical literature
1PC-FA-5L 17.0 2099 8.9
that for each increase of entrapped/entrained by one percent, there
1PC-FA-10L 17.0 2122 8.1
1PC-FA-15L 16.5 2130 8.0 is generally a decrease of about 4–6% in the compressive strength
of concrete; then, a fraction of the significant compressive strength
B 2PC-FA 17.0 2061 9.1
2PC-FA-5L 17.0 2091 8.1 loss observed for groups A, B and C in Fig. 3 could be attributed to
2PC-FA-10L 17.0 2111 7.4 the increase in air content (between 8 and 36%). For future work,
2PC-FA-15L 16.5 2130 6.8 possibly the use of an entrained air reducer will improve compres-
C 3PC-FA 18.0 2036 9.1 sive strength and electrical resistivity.
3PC-FA-5L 18.0 2081 7.3
3PC-FA-10L 17.0 2124 5.7
3PC-FA-15L 17.5 2147 4.9 4.2. Times of setting

FA causes delays in the times of setting as reported previously

4. Results and discussion [30], a characteristic observed in Fig. 2, where the results of the ini-
tial and final times of setting are reported for the thirteen mortar
4.1. Fresh properties mixtures evaluated in this study.
In Fig. 2, it is evident that the incorporation of L reduced the
The measured fresh properties for the thirteen mortar mixtures retardation caused by FA. It can be observed that for a 5% substitu-
are shown in Table 5. For groups A, B and C, the substitution of tion of FA by L, times of setting similar to those of the reference
portland cement (PC) by fly ash (FA) by 30, 45 and 60% by mass, mortar were obtained. Further substitutions of FA by L lead to

Fig. 2. Measured times of setting of mortars.

54 A. Duran-Herrera et al. / Construction and Building Materials 199 (2019) 50–62

Fig. 3. Compressive strength of mortar mixtures.

Table 6 greater reductions in times of setting. The acceleration can be

Measured autogenous strain of mortars. attributed to the fineness of the L, because it has a smaller particle
Group Identification Autogenous strain at Strain Standard size than the portland cement, representing a greater surface area
28 d (mm/m)* reduction deviation available as nucleation centers that promote the formation of port-
%* (mm/m) land cement hydration products [31,32].
REF REFERENCE 378 0 26 Results confirms that the setting times increases as the substi-
A 1PC-FA 230 39 12 tution of PC by FA increases and the results obtained for groups
1PC-FA-5L 146 61 15 A and B clearly illustrates that the replacement of FA by L counter-
1PC-FA-15L 192 49 13 acts this increase.
B 2PC-FA 135 64 38
2PC-FA-5L 82 78 11
2PC-FA-15L 157 58 41 4.3. Compressive strength
C 3PC-FA 60 84 23
3PC-FA-5L 53 86 4 The compressive strength results at ages of 7 d, 28 d, and 56 d
3PC-FA-15L 123 67 17 are given in Fig. 3. Groups A, B and C show results from five binary
* blends (PC + FA), and ternary blends with varying dosages of L that
Reduction of shrinkage regarding the reference mortar, in percentage.

Fig. 4. Autogenous strain of Group A.

 A. Duran-Herrera et al. / Construction and Building Materials 199 (2019) 50–62 55

go to 15% in substitution of the FA volume in the cementitious In general, results of compressive strength indicate that as PC
system. by FA substitutions increases, strength decreases, also that the sub-
Compared to the reference mortar mixture, the substitutions of stitutions of FA by L counteracts this reduction mainly at 28 days
PC by FA evaluated in groups A, B and C (30%, 45%, and 60% FA by and later ages.
volume) produced reductions of the compressive strength of 31%, For each average result, the standard errors calculated as an
61%, and 63% at 7 d, of 30%, 37%, and 54% at 28 d, and of 23%, estimate of the uncertainty with regard to the estimation of the
30%, and 48% at 56 d, respectively. compressive strength were within the following ranges: 1.09–
On the other hand, the effect of L as a filler agent and nucleation 3.98% for the reference mixture, 0.11–4.83% for group A, 0.10–
center, combined with the pozzolanic effect of FA, led to higher 14.74% for group B, and 0.10–11.13% for group C.
compressive strengths at 56 d for mixtures 1PC-FA-5L, 2PC-FA-
15L and 3PC-FA-10L in comparison to their respective reference 4.4. Autogenous strain
(1PC-FA, 2PC-FA and 3PC-FA), specifically leading to increases of
10%, 4%, and 4%, respectively. To obtain reliable results, only the ten mixtures identified in
The standard deviations for the compressive strengths reported Table 6 were evaluated. The total autogenous strains at 28 d are
in Fig. 3 were generally within the range of variations permitted by given in Table 6, the negative sign corresponds to shrinkage. For
ASTM C109 (8.7% for three cubes and 7.6% for two cubes). groups A, B and C, autogenous strains measured up to an age of

Fig. 5. Autogenous strain of Group B.

Fig. 6. Autogenous strain of Group C.

56 A. Duran-Herrera et al. / Construction and Building Materials 199 (2019) 50–62

28 d are also presented in Figs. 4–6. A greater substitution of PC by The results of mixture 3PC-FA-5L confirm that L can also con-
FA or FA + L produces a greater reduction in the measured autoge- tribute to a slight autogenous expansion after the final time of set-
nous strain. Reductions were between 39% and 86% in comparison ting, which can be caused by the fineness of L, as it can absorb and/
with the reference mixture, which can be attributed to the reduc- or adsorb water. However, shrinkage will appear when this water
tion of Portland cement (the most reactive component) in the is subsequently (re)absorbed by the anhydrous and hydrated
cementitious systems of PC + FA or PC + FA + L, which led to a cement phases because of the capillary tensions inside of the paste
reduction in self-desiccation. Substitutions of FA by L also have a pores [37,38]. Other authors attribute this expansion to the forma-
significant influence, with the mixtures with 5% of L having the tion of calcium hydroxide crystals and/or ettringite and carboalu-
higher reductions in autogenous shrinkage for groups A, B and C. minate phases [9,39]. In this regard, it is worth noting that the
The mortar mixtures with FA contents of 45% and 60% by vol- presence of the fine limestone powder will stabilize the ettringite
ume (2PC-FA and 3PC-FA) showed a slight expansion after the formed at early ages [40]. This in turn will maximize the contribu-
final time of setting, which was more noticeable in mixture tion of ettringite (needles) to the early-age autogenous expansion
3PC-FA. Previous research works on the subject have attributed and minimize the contribution to autogenous shrinkage that may
this behavior to the formation of ettringite and/or to the reab- subsequently occur when these needle-like structures dissolve
sorption of the bleed water [33,34]. The reduction of the total (removing their expansive restraint) during their conversion to
shrinkage can be also attributed to the larger particle size of monosulfoaluminate phases.
the FA used in this study, in comparison to the portland cement Fig. 7 presents the ultimate autogenous strain at 28 days; for
[35,36]. group A, the substitutions of PC by FA indicate that as the substitu-
tion increases the reduction increases approximately by 150 (39%),

Fig. 7. Ultimate Autogenous Strain at 28 days.

Fig. 8. Drying shrinkage of group A.

 A. Duran-Herrera et al. / Construction and Building Materials 199 (2019) 50–62 57

and that the substitution of FA by 5 or 15% of L increases the reduc- The calculated standard deviations of the autogenous strain
tion to 240 (63%) and 180 mm (47%) respectively. measurements presented in Figs. 4–6 were within the following
For group B and C, the substitution of PC by FA leads to higher ranges: 3–41 mm/m for the reference mixture, 3–31 mm/m for
reductions of autogenous shrinkage 250 (66%) and 320 mm (86%) group A (Fig. 4), 1–41 mm/m for group B (Fig. 5), and 1–47 mm/m
respectively. For this two groups, the substitution of FA by 5% of for group C (Fig. 6).
L improves (295 mm  78%) and maintain (320 mm  86%) the
reduction respectively. The reductions were less significant when 4.5. Drying shrinkage
FA was substituted by 15% of L, 225 mm (59%) and 265 mm (70%)
for groups B and C respectively. Results of drying shrinkage up to an age of 28 d are presented in
The substitutions of PC by FA in 30, 45 and 60% by mass reduced Figs. 8–10 for Groups A, B and C, respectively. Average results at 28 d
autogenous shrinkage by 150 (40%), 250 (66%) and 320 mm (84%), a and estimations at 52 weeks (52 w) are presented in Fig. 11; a sum-
benefit that suggests that for a substitution between 30 and 60% mary of the results at 28 d and 52 w is presented in Table 7. Results
could lead to an average reduction of 100 mm per each 15% of show that drying shrinkage is reduced by 17%, 32%, and 45% in
substitution. groups A, B, and C, respectively when the substitutions of PC by FA

Fig. 9. Drying shrinkage of Group B.

Fig. 10. Drying shrinkage of Group C.

58 A. Duran-Herrera et al. / Construction and Building Materials 199 (2019) 50–62

or FA + L increases. This reduction can be attributed to the larger par- values in the range of 55–82%, 50–69%, and 32–56% for series A, B,
ticle size of the FA, which is 4 times larger than the portland cement and C respectively. For groups A and B, Fig. 10 indicates that drying
particles. This physical characteristic could lead to average pore shrinkage decreases as L increases and evidence an inverse trend
sizes larger than 50 nm in the mortar mixtures, originating a reduc- for results of group C.
tion in shrinkage, since the pore solution within pores smaller than For each group (A, B or C), results in Fig. 11 illustrate that reduc-
50 nm will induce significant capillary stresses [41,42]. tions in drying shrinkage are approximately the same for mixtures
It is also observed that drying shrinkage slightly increased when PC-FA, PC-FA-5L, PC-FA-10L and PC-FA-15L. Based on this consider-
FA was substituted by L. This increase was more significant for ation, the average drying shrinkage for mixture R and for groups A,
group C, when greater amounts of FA or FA + L were used. This B and C ((PC-FA + PC-FA-5L + PC-FA-10L + PC-FA-15L)/4) were 870,
increase was attributed to the resulting finer pore structure due 720, 600 and 540 mm at 28 days, and 1520, 1230, 1005 and 785 mm
to the filler effect of L. For groups A and B, at 28 d, substitutions at 56 days respectively. This results indicate that as the substitu-
of FA by L did not significantly influence drying shrinkage. tion of PC by FA of FA + L increases by 30, 45 and 60%, drying
Comparing 28 d drying shrinkage for each single mix with the shrinkage decreases by an average of 17, 31 and 38% at 28 days
estimations at 52 w, results indicate that drying shrinkage could and 19, 34 and 48% at 56 days respectively.
increase 74% for the reference and 82%, 77%, and 55% as a result The standard deviations of the 28 d average drying shrinkage
of the substitutions of PC by FA in series A, B, and C, respectively. presented in Figs. 8–10 were within the following ranges: 16–
In comparison with the potential increments estimated between 21 mm/m for the reference mixture, 5–20 mm/m for group A
28 d and 52 w for mixtures R, 1PC-FA, 2PC-FA and 3PC-FA, substi- (Fig. 8), 5–32 mm/m for group B (Fig. 9), and 6–24 mm/m for group
tutions of FA by L lead to lower drying shrinkage estimations, with C (Fig. 10).

Fig. 11. Average and ultimate drying shrinkage calculated and estimated at 28 d and 52 w, respectively.

Table 7
Measured drying shrinkage of mortars.

Group Identification Shrinkage, mm/m (%) Reduction (%)* Standard deviation (mm/m)
28 d1 52 w2 28 d 52 w
REF REFERENCE 871(0.09) 1517(0.15) 0 0 20
A 1PC-FA 724(0.07) 1318(0.13) 17 13 8
1PC-FA-5L 709(0.07) 1283(0.13) 19 15 16
1PC-FA-10L 712(0.07) 1106(0.11) 18 27 20
1PC-FA-15L 719(0.07) 1215(0.12) 17 20 12
B 2PC-FA 591(0.06) 1049(0.11) 32 31 8
2PC-FA-5L 600(0.06) 1011(0.10) 31 33 24
2PC-FA-10L 622(0.06) 932(0.09) 29 39 11
2PC-FA-15L 620(0.06) 955(0.10) 29 37 12
C 3PC-FA 480(0.05) 742(0.07) 45 51 6
3PC-FA-5L 543(0.05) 845(0.09) 38 44 5
3PC-FA-10L 546(0.06) 723(0.07) 37 52 12
3PC-FA-15L 580(0.06) 876(0.09) 33 42 24
*
Reduction of shrinkage regarding the reference mortar, in percentage.
1
Calculated according to ASTM C596-09, Section 12.
2
Estimated according to ASTM C596-09, Note 2 and Fig. 1.
 A. Duran-Herrera et al. / Construction and Building Materials 199 (2019) 50–62 59

4.6. Surface electrical resistivity When portland cement is replaced by FA, the concentration of
the alkali ions generally decreases (Na+ and K+), leading to a lower
The surface electrical resistivity measured at different ages ionic conductivity and therefore to a greater electrical resistivity
(between 1 d and 56 d) is presented in Figs. 12, 13 and 14 for [47].
specimens corresponding to groups A, B and C, respectively. Elec- Moreover, for groups A, B, and C, it can be observed that the
trical resistivity at early ages decreases as PC by FA substitution resistivity decreases when FA is substituted by L. The presence of
increases. At later ages, the electrical resistivity of the FA and FA L increases the concentration of OH ions that will act as an elec-
and L systems increases significantly above that measured for the trolyte in the pore solution of the mortars/concrete and will
reference. This later age behavior is attributed chiefly to the decrease the electrical resistivity because these ions are the more
densification of the cementitious matrix and to the pore refine- conductive ions within the pore solution [48,49].
ment produced by the pozzolanic reaction of FA [43–45]. Surface In Figs. 12–15, results of surface electrical resistivity illustrate
electrical resistivity is an indirect measure of both porosity and dif- that in general, the substitutions of PC by FA of FA + L significantly
fusivity. The electrical current flowing through the hydrated paste increases this durability index. The effect is more significant as FA
is due to an electrolytic process mainly resulting from the flow of or FA + L increases. For the three groups of mixtures (A, B and C),

ions present in the pore solution (Na+, K+, Ca2+, SO2+ 4 , OH ) [46]. the substitutions of PC by FA leaded to the highest increments.

Fig. 12. Surface electrical resistivity vs curing time of Group A.

Fig. 13. Surface electrical resistivity vs curing time of Group B.

60 A. Duran-Herrera et al. / Construction and Building Materials 199 (2019) 50–62

Fig. 14. Surface electrical resistivity vs curing time of Group C.

Fig. 15. Surface Electrical Resistivity at 28 days and 56 days.

For mixtures with FA + L the effect were less significant and tends 5. Conclusions
to decrease as L increases.
Regarding the surface electrical resistivity results obtained for From this study, the following conclusions can be drawn:
group A mixtures, results in Figs. 12–14 indicate that as the substi-
tution of PC by FA or FA + L increases, the age when mixtures 1. The results for electrical resistivity indicate a significant
PC-FA, PC-FA-5L, PC-FA-10L and PC-FA-15L exceeded the surface improvement in the cementitious matrix imperviousness due
electrical resistivity of reference mixture was accelerated by 6, to the replacement of PC by FA. As the substitution of PC by
12, 16 and 14 days respectively for mixtures of groups B and C. FA increases, the electrical resistivity results at later ages
The standard errors calculated as an estimate of the uncertainty demonstrate an increasing benefit.
with regard to the estimation of the surface electrical resistivity 2. The results confirm that the addition of L is an effective option
were within the following ranges: 0.12–1.26% for the reference to counteract the delays in the times of setting that result from
mixture, 0.19–2.45% for group A (Fig. 12), 0.16–3.06% for group B replacing PC by FA and identify the more appropriate combina-
(Fig. 13), and 0.11–2.43% for group C (Fig. 14). tions of FA + L for this purpose. As for the compressive strength,
 A. Duran-Herrera et al. / Construction and Building Materials 199 (2019) 50–62 61

the results show that some of the tested combinations can lead [5] K.H. Obla, R.L. Hills, R.S. Martin, HVFA Concrete – An Industry Perspective
August, ACI Concrete International, 2003, pp. 29–30.
to a better performance at advanced ages, which is attributable
[6] K. Khayat, D. Mitchell, Self-Consolidating Concrete for Precast, Prestressed
to the effects of the micro limestone powder acting as a filler in Concrete Bridge Elements NCHRP REPORT 628, National Cooperative Highway
the cementitious matrix and as a nucleation center for the Research Program, 2009.
hydration products from the surrounding cement particles. [7] F. Aslani, L. Maia, Creep and shrinkage of high-strength self-compacting
concrete: experimental and analytical analysis, Mag. Concr. Res. Instit. Civ.
3. In all cases, the FA substitution by L produced a decrease in the Eng. 65 (17) (2013) 1044–1058.
electrical resistivity; as the substitution levels increased, the [8] R. Loser, A. Leeman, Shrinkage and restrained shrinkage cracking of self-
electrical resistivity further decreased. compacting concrete compared to conventionally vibrated concrete, Mater.
Struct. 42 (1) (2008) 71–82.
4. Substitutions of PC by FA led to significant reductions in autoge- [9] D.P. Bentz, M.A. Peltz, Reducing thermal and autogenous shrinkage
nous shrinkage ranging between 39% and 84% for substitutions contributions to early-age cracking, ACI Mater. J. 105 (4) (2008) 414–420.
between 30% and 60%. In relation to mixtures based on the bin- [10] S. Zhutovsky, K. Kovler, Effect of internal curing on durability-related
properties of high performance concrete, Cem. Concr. Res. 42 (1) (2012)
ary blend (PC + FA), substitutions of FA by L resulted in either 20–26.
slight reductions or increases of autogenous shrinkage, which [11] D.P. Bentz, J. Tanesi, A. Ardani, Ternary blends for controlling cost and carbon
were not greater than 84 mm/m and 63 mm/m, respectively. content, Concr. Int. 35 (8) (2013) 51–59.
[12] L. Gurney, D.P. Bentz, T. Sato, W.J. Weiss, Using limestone to reduce set
These reductions were more evident for PC by FA substitutions retardation in high volume fly ash mixtures: improving constructability for
of 30%; for substitutions of 45% and 60%, substitutions higher sustainability, Transp. Res. Rec. J. Transp. Res. Board No 2290 Concr. Mater.
than 5% of FA by L resulted in increased autogenous shrinkage. (2012) 139–146.
[13] D.P. Bentz, T. de la Sato, I. Varga, W.J. Weiss, Fine limestone additions to
5. Substitution of PC by FA reduced 28 d drying shrinkage between
regulate setting in high volume fly ash mixtures, Cem. Concr. Compos. 34
17% and 45%, the reduction becoming more significant as the (2012) 11–17.
substitution of PC by FA increased. For series A, substitutions [14] ASTM Standard C150/C150M - 12, Standard Specification for Portland Cement
of FA by L produced less significant reductions, 35% on average. vol. 04.01 (2012).
[15] ASTM Standard C618-05, Standard Specification for Coal Fly Ash and Raw or
In series B and C, substitutions of FA by L showed quite similar Calcined Natural Pozzolan for Use in Concrete vol. 04.02 (2005).
trends to those of series A for monitoring times between 3 d and [16] ASTM Standard C33-13, Standard Specification for Concrete Aggregates vol.
15 d. However, at later ages, in both cases, the PC + FA mixtures 04.02 (2013).
[17] ASTM Standard C494-09, Standard Specification for Chemical Admixtures for
as well as the PC + FA + L mixtures showed practically the same Concrete vol. 04.02 (2009).
shrinkages at 28 d. At this age, average reductions for series B [18] ASTM Standard C1611-14, Standard Test Method for Slump Flow of Self-
and C were 18% and 31%, respectively. Regarding projected Consolidating Concrete vol. 04.02 (2014).
[19] ASTM Standard C1610-14, Standard Test Method for Static Segregation on
52 w results, substitutions of PC by FA indicate that ultimate Self-Consolidating Concrete Using Column Technique vol. 04.02 (2014).
drying shrinkage could increase between 55% and 82 % and that [20] ASTM Standard C1621-14, Standard Test Method for Passing Ability of Self-
substitutions of FA by L increase drying shrinkage between 32% Consolidating Concrete by J-Ring vol. 04.02 (2014).
[21] A. Schwartzentruber, C. Catherine, Method of Concrete Equivalent Mortar – a
and 81%. Estimations at 52 w tend to be lower as the substitu- novel tool to help in formulation of concrete with admixtures, Mater. Struct.
tions of PC by FA or PC by FA + L increases. 33 (8) (2000) 475–482.
[22] ASTM Standard C1437-13, Standard Test Method for Flow of Hydraulic
Cement Mortar vol. 04.01 (2013).
Conflict of interest [23] ASTM Standard C403-08, Standard Test Method for Time of Setting of
Concrete Mixtures by Penetration Resistance vol. 04.02 (2008).
[24] ASTM Standard C185-08, Standard Test Method for Air Content of Hydraulic
The authors of the manuscript titled ‘‘Self-Compacting Con- Cement Mortar vol. 04.01 (2008).
cretes Using Fly Ash and Fine Limestone Powder: Shrinkage and [25] ASTM Standard C109-13, Standard Test Method for Compressive Strength of
Surface Electrical Resistivity of Equivalent Mortars” submitted to Hydraulic Cement Mortars (using 2-in or [50-mm] cube specimens) vol. 04.01
(2013).
the Journal of Construction and Building Materials certify that this [26] ASTM Standard C1698-09, Standard Test Method for Autogenous Strain of
manuscript have not being submitted to any other journal, and Cement Paste and Mortar vol. 04.02 (2014).
that they have NO affiliations with or involvement in any organiza- [27] ASTM Standard C596-09, Standard Test Method for Drying Shrinkage of
Mortar Containing Hydraulic Cement vol. 04.01 (2009).
tion or entity with any financial interest, or non-financial interest
[28] AASHTO TP 95-11, Standard Method of Test for Surface Resistivity of
in the subject matter or materials discussed in this manuscript. Concrete’s Ability to Resist Chloride Ion Penetration, American Association of
State Highway and Transportation Officials, Washington, DC.
[29] W. Morris, E. Moreno, A. Sagues, Practical evaluation of resistivity of concrete
Acknowledgements in test cylinders using a Wenner array probe, Cem. Concr. Res. 26 (1996) 1779–
1787.
[30] A. Durán-Herrera, C.A. Juárez, P. Valdez, D.P. Bentz, Evaluation of sustainable
The authors wish to express their gratitude to the Consejo high-volume fly ash concretes, Cem. Concr. Compos. 33 (1) (2011) 39–45.
Nacional de Ciencia y Tecnología (CONACYT) for providing the [31] D.P. Bentz, Modeling the influence of limestone filler on cement hydration
scholarship to Juan De-León for his MSc studies. Also to Dave Carr using CEMHYD3D, Cem. Concr. Compos. 28 (2) (2005) 124–129.
[32] B. Yilmaz, A. Olgun, Studies on cement and mortar containing low-calcium fly
and Francisco Pazos from OMYA and Israel Olivares from Repre-
ash, limestone, and dolomitic limestone, Cem. Concr. Compos. 30 (3) (2008)
sentaciones Técnicas S.A. de C.V., for providing the micro limestone 194–201.
powder used in this study. As well as to Filiberto Marín from [33] Institute Japan Concrete, Technical committee report on autogenous
shrinkage, in: Tazawa (Ed.), Proc. of Int. Workshop, E&FN, London, 1998, pp.
EUCOMEX S.A. de C.V., for providing the high range water reducer
1–63.
and viscosity modifier admixtures, and to Ramón Álvarez from [34] P. Lura, Autogenous Deformation and Internal Curing of Concrete PhD Thesis,
Concretos La Silla, S.A. de C.V., for providing the portland cement. Delft University of Technology, Delft, 2003.
[35] C. Jiang, Y. Yang, Y. Wang, Y. Zhou, C. Ma, Autogenous shrinkage of high
performance concrete containing mineral admixtures under different curing
References temperatures, Constr. Build. Mater. 61 (2014) 260–269.
[36] D.P. Bentz, O.M. Jensen, K.K. Hansen, J.F. Olsen, H. Stang, C.-J. Haecker, Influence
of cement particle size-distribution on early age autogenous strains and
[1] H. Okamura, M. Ouchi, Self-compacting concrete – development, present and
stresses in cement-based materials, J. Am. Ceram. Soc. 84 (1) (2001) 129–135.
future, in: First International RILEM Symposium on Self-Compacting Concrete,
[37] O. Esping, Effect of limestone filler BET (H2O)-area on the fresh and hardened
1999, pp. 3–14.
properties of self-compacting concrete, Cem. Concr. Res. 38 (7) (2008) 938–
[2] M. Usyal, K. Yilmaz, Effect of mineral admixtures on properties of self-
944.
compacting concrete, Cem. Concr. Compos. 33 (7) (2011) 771–776.
[38] B. Craeye, G. De-Schutter, B. Desmet, J. Vantomme, G. Heirman, L. Vandewalle,
[3] V.M. Malhotra, P.K. Mehta, High Performance, High-Volume Fly Ash Concrete,
O. Cizer, S. Aggoun, E.H. Kadri, Effect of mineral filler type on autogenous
second ed., Sustainable Development Inc, Ottawa, Canada, 2005.
shrinkage of self-compacting concrete, Cem. Concr. Res. 40 (6) (2010) 908–
[4] Michael Thomas, Optimizing the Use of Fly Ash in Concrete, Portland Cement
913.
Association, Skokie, Ill, 2007.
62 A. Duran-Herrera et al. / Construction and Building Materials 199 (2019) 50–62

[39] T. Sato, J.J. Beaudoin, The effect of nano-sized CaCO3 addition on the hydration [44] I. De la Varga, R.P. Spragg, C. Di Bella, J. Castro, D.P. Bentz, J. Weiss, Fluid
of cement paste containing high volumes of fly ash, in: Proceeding of the 12th transport in high volume fly ash mixtures with and without internal curing,
International Congress on the Chemistry of Cement, 2007, p. 12. Cem. Concr. Compos. 45 (2014) 102–110.
[40] M. Zajac, A. Rossberg, G. Le Saout, B. Lothenbach, Influence of limestone and [45] A. Raif Boga, I. Bekir Topçu, Influence of fly ash on corrosion resistance and
anhydrite on the hydration of portland cements, Cem. Concr. Compos. 46 chloride ion permeability of concrete, Constr. Build. Mater. 31 (2012) 258–264.
(2014) 99–108. [46] K.O. Ampadu, K. Torii, M. Kawamura, Beneficial effect of fly ash on chloride
[41] W. Wongkeo, P. Thongsanitgarn, A. Chaipanich, Compressive strength and diffusivity of hardened cement paste, Cem. Concr. Res. 29 (4) (1999) 585–590.
drying shrinkage of fly ash-bottom ash-silica fume multi-blended cement [47] C. Shi, Effect of mixing proportions of concrete on its electrical conductivity
mortars, Mater. Des. 36 (2012) 655–662. and the rapid permeability test (ASTM C1202 or ASSHTO T277) results, Cem.
[42] P. Chindaprasirt, S. Homwuttiwong, V. Sirivivatnanon, Influence of fly ash Concr. Res. 34 (3) (2004) 537–545.
fineness on strength, drying shrinkage and sulfate resistance of blended [48] M. Ghrici, S. Kenai, M. Said Mansour, Mechanical properties and durability of
cement mortar, Cem. Concr. Res. 34 (2004) 1087–1092. mortar and concrete containing natural pozzolana and limestone blended
[43] R.B. Polder, W.H.A. Peelen, Characterization of chloride transport and cements, Cem. Concr. Compos. 29 (7) (2007) 542–549.
reinforcement corrosion in concrete under cyclic wetting and drying by [49] C. Andrade, Calculation of chloride diffusion in concrete from ionic migration
electrical resistivity, Cem. Concr. Compos. 24 (5) (2002) 427–435. measurements, Cem. Concr. Res. 23 (5) (1998) 1655–1667.