Construction and Building Materials 263 (2020) 120671

Contents lists available at ScienceDirect

Construction and Building Materials

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

Effect of fine recycled concrete aggregate on the mechanical behavior of

self-compacting concrete
Víctor Revilla-Cuesta a,⇑, Vanesa Ortega-López a, Marta Skaf b, Juan Manuel Manso a
a
Department of Civil Engineering, University of Burgos, Spain
b
Department of Construction, University of Burgos, Spain

h i g h l i g h t s g r a p h i c a l a b s t r a c t

SCC flowability was maintained by

adding fine RCA and adjusting
effective w/c ratio.

Up to 50% fine RCA, SCC presented
optimal mechanical properties for
structural use.

Strength development was slower the
higher the fine RCA content.

The addition of fine RCA increased the
fracture strain of the SCC.

Adjustment to the theoretical
formulas improved by increasing the
RCA content.

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

Article history: The high flowability of Self-Compacting Concrete (SCC) is achieved by adding large amounts of fine aggre-
Received 11 March 2020 gate. Therefore, the addition of fine Recycled Concrete Aggregate (RCA) in this type of concrete can very
Received in revised form 26 June 2020 noticeably change its behavior. SCCs with different percentages of fine RCA (0%, 25%, 50%, 75%, and 100%)
Accepted 18 August 2020
and 100% coarse RCA were manufactured in this study, to evaluate their performance, and to analyze the
Available online 6 September 2020
effect of fine RCA in an SCC when a high amount of coarse RCA is also added. Both the fresh properties
(flowability, density, and air content) and their mechanical behavior (strengths, non-destructive tests,
Keywords:
stress–strain curves, and Poisson coefficient) at different curing ages were studied. These mechanical
Self-compacting concrete
Recycled concrete aggregate
properties were compared with the values calculated using the formulas from two of the most common
Flowability structural design standards. High values of strength and modulus of elasticity were obtained up to a fine
Mechanical performance at different curing RCA content of 50%. Additionally, any increase in fine RCA increased flowability and elastic and plastic
ages deformability of the SCC. The theoretical values overestimated the experimental ones by around 25%.
Stress–strain curves From the mechanical point of view, SCC with up to 50% fine RCA could be used for structural applications,
Design values although service requirements regarding deformability recommend that its content should be limited to
25%.
Ó 2020 Elsevier Ltd. All rights reserved.

⇑ Corresponding author at: Department of Civil Engineering, University of Burgos,

1. Introduction
Escuela Politécnica Superior, Calle Villadiego s/n, 09001 Burgos, Spain.
E-mail addresses: vrevilla@ubu.es (V. Revilla-Cuesta), vortega@ubu.es Construction sector activities are generally characterized as
(V. Ortega-López), mskaf@ubu.es (M. Skaf), jmmanso@ubu.es (J.M. Manso). large-scale projects that consume vast amounts of natural

https://doi.org/10.1016/j.conbuildmat.2020.120671
0950-0618/Ó 2020 Elsevier Ltd. All rights reserved.
2 V. Revilla-Cuesta et al. / Construction and Building Materials 263 (2020) 120671

resources, something that the production figures of the sector lar- these procedures has been demonstrated in vibrated concrete with
gely reflect. On average, within the European Union, around 2700 high contents of coarse RCA [28,29], even at early ages [30].
Mt of Natural Aggregate (NA) are placed on the market each year, Despite the low coarse aggregate content of SCC, non-destructive
of which 120 million were consumed in Spain. Moreover, 16 Mt of tests are also valid for this type of concrete when NA is used [31].
bituminous concrete, and 22 Mm3 of hydraulic concrete were pro- The aim of this study is to evaluate the effect of adding different
duced, in 2018, in this country [1]. High levels of resource exploita- amounts of fine RCA in an SCC with a constant amount of coarse
tion have continued over time, leading in some areas to significant RCA. SCC requires a high proportion of fine aggregate to reach
scarcities [2]. flowability, so it is more sensitive to the effects of fine RCA. In addi-
Construction and Demolition Waste (CDW) represents 34.7% of tion, the interaction between coarse and fine RCA can influence its
all building waste generated in Europe, over 800 Mt/year, where performance. The coarse RCA content was defined in a preliminary
the countries with the highest production levels are France and analysis, involving tests on the compressive strengths and the elas-
Germany (227 and 207 Mt/year respectively) [3]. In Spain, with tic moduli of mixes with 0%, 50% and 100% of coarse RCA. Based on
the upturn of the construction sector, CDW production exceeded this analysis, the amount of 100% coarse RCA was chosen for the
20 Mt in 2017 [3]. Demolition processes generated around 0.9 t optimum sustainability of the product. Subsequently, five different
of CDW per m2 of demolished housing, reaching 1.2 t/m2 in indus- SCCs with 100% coarse RCA and fine RCA percentages of 0%, 25%,
trial buildings, 40% of which was deposited in illegal landfill sites 50%, 75% and 100% were manufactured and analyzed in this study.
[4]. CDWs usually present a mixture of concrete with ceramic resi- Firstly, the evaluation of the fresh concrete behavior of these
dues and glass and/or gypsum, which worsen their behavior as a SSCs verified their compliance with all the recommendations of
raw material for other uses [5]. Recycled Concrete Aggregate the European Federation of National Associations Representing
(RCA), a particular type of CDW, is usually produced from the producers and applicators of specialist building products for Con-
crushing of selected concrete elements, such as precast concrete crete (EFNARC) [32]. Subsequently, the temporal evolution of the
rejects, and it demonstrates better mechanical behavior, because mechanical properties was evaluated. Additionally, the theoretical
it contains fewer contaminants and its properties are less variable values of these mechanical properties, estimated from their com-
[6]. pressive strengths, were calculated and compared with the exper-
The two above-mentioned problems -scarcity of natural imental values, in order to validate the calculation methods. The
resources and abundance of waste- could be mitigated through a validity of RCA as a material for manufacturing SCC for structural
simple strategy: using waste as a raw material in different con- usage was therefore demonstrated, in accordance with its dosage
struction sector applications, in substitution of NA, among which (in particular, the percentage of fine RCA), which must be selected
the manufacture of hydraulic and bituminous concretes is notable to meet the main purpose of the concrete design (self-
[7–13]. This strategy would also help to reduce the ecological compactability and/or strength).
impact of quarrying as well as the carbon footprint of the construc-
tion sector [5,14–16]. Among the various potential usages of RCA,
2. Materials
this study is focused on its validation in the manufacture of Self-
Compacting Concrete (SCC).
The characteristics of the materials used in this study are
The total aggregate volume of concrete is 65–70%, making RCA
described in this section. An analysis of the physical properties
an ideal material for applying the above strategy. In fact, the liter-
and the chemical composition of the RCA is also included, to eval-
ature contains several studies of conventional concretes made with
uate its suitability for concrete manufacture.
proper dosages of coarse RCA that have demonstrated a good
behavior [17,18]. The addition of fine RCA has a more harmful
effect, causing a very noticeable decrease in the mechanical prop- 2.1. Cement, water, filler and admixtures
erties of the concretes in which it is incorporated [19]. Although
there are some studies on high-performance concretes, such as A conventional Portland cement CEM I 52.5 R was used, with a
SCC [20], the available bibliography is scarce, fundamentally in density of 3.12 Mg/m3 and a clinker content of around 98%. Mix
relation to fine RCA, a remarkable aspect, considering the impor- water that contained no chemical compounds with potentially
tance of the fine aggregate fractions within this type of concrete adverse effects on concrete behavior was taken from the water
[21]. supply of the city of Burgos, Spain.
The aspects commented on above are due to the particular char- The finest fraction of the granulometry (<0.063 mm) was pro-
acteristics of RCA. The coarse fraction of RCA, which is larger than vided by the addition of limestone filler with a CaCO3 content
4 mm, is mainly characterized by three interrelated aspects. Firstly, higher than 98%. According to the manufacturer’s specifications,
the mortar adhered to the surface of the RCA means that, in com- this filler had a density of 2.77 Mg/m3 and a water absorption rate
parison with the NA, its density and hardness are lower [22,23]. of 0.54% over 24 h (0.37% in 10 min).
Secondly, RCA has a contact surface between the aggregate and Two additives, fundamental components for providing the SCC
the mortar, known as the Interfacial Transition Zone (ITZ) [24], with its characteristic self-compactability, were used, which had
which is weaker and less dense than the ITZ of NA [22,25]. Finally, demonstrated good behavior in a similar study [33]. The first
the high water absorption levels of RCA can be due to the unhy- one, a plasticizer, enhanced concrete flowability. The second, a vis-
drated cement in the attached mortar [23,25]. The behavior of cosity regulator, maintained SCC flowability over longer periods,
the fine RCA, less than 4 mm in size, is strongly conditioned by improving the results of the in-fresh state tests. In all dosages, total
the mortar particles, as well as some pollutants (clay, gypsum additions of admixture amounted to 2.2% by weight of cement.
and mica), mainly introduced during the crushing process, which
cause, among other aspects, higher levels of water absorption than 2.2. Aggregates (RCA, siliceous gravel and siliceous sand)
in the coarse fraction [26].
The addition of RCA affects not only both the fresh and the hard- The RCA was produced from concrete components with a char-
ened behavior of the concrete, but also the estimation of compres- acteristic strength of 45 MPa, manufactured with siliceous aggre-
sive strength through non-destructive testing, such as the hammer gate, which had subsequently been rejected by the prefabrication
rebound index and Ultrasonic Pulse Velocity (UPV), mainly due to industry and crushed. Supplied in sizes between 0 and 31.5 mm
the presence of attached mortar [27]. However, the validity of from a local CDW management company, the RCA was sieved
 V. Revilla-Cuesta et al. / Construction and Building Materials 263 (2020) 120671 3

and separated into fine RCA (0/4 mm) and coarse RCA (4/12.5 mm) adjusted to the Fuller curve, using the granulometric modules of
fractions, both of which were used in this study. the aggregates. Rather than the addition of admixtures to the
In Table 1, the main properties of the RCA are compared with mix design, the water content in the mixtures was carefully
the recommended limit values from the Spanish concrete standard increased to achieve a proper self-compactability. That decision
EHE-08 [34]. The comparison shows that the RCA fulfilled most of was taken to avoid admixture segregation that occurred when its
the requirements, except for density and water absorption, which proportion in the mix exceeded 2.2% of cement mass.
explains the limitation that is specified in the standard of a maxi-
mum content of 20% total coarse RCA. The chemical composition of 3.2.1. Preliminary analysis
fine RCA obtained by X-ray fluorescence (XRF) spectrometry and The optimum amount of coarse RCA to be added to the mixtures
the X-ray diffraction (XRD) pattern are shown in Table 2 and was decided after a preliminary analysis. Three mixes with 0%, 50%
Fig. 1, respectively. The predominance of silicon oxide and calcium and 100% coarse RCA (labelled C0, C50 and C100 respectively) were
carbonate can be observed in its composition. The hydrated used to mold 10  20-cm cylindrical samples (2 samples in each
cementitious components (CSH) are not visible in the XRD test test). Their compositions are shown in Table 3. The compressive
results, although the content in both aluminum and iron oxides strength and the modulus of elasticity of the samples were tested
denotes the presence of old mortar in the fine RCA. at 7 and 28 days. The results were evaluated with a one-way
Siliceous gravel and sand were used to complete the coarse ANOVA (5% significance level), which showed no significant differ-
(preliminary analysis) and the fine aggregate fraction of each con- ences between the compressive strength and the modulus of elas-
crete mix. Their properties are also summarized in Table 1. ticity of the mixes with 50% and 100% of coarse RCA (Table 4). So,
The granulometric curves of all aggregates are shown in Fig. 2. It an amount of 100% of coarse RCA was used in the analysis of the
can be seen that the fine RCA shows higher percentages of fine par- fine RCA mix performance, in order to maximize the sustainability
ticles than the siliceous sand. of the SCC designed.

3.2.2. Mixes with fine RCA

3. Methodology
Having defined the reference dosage (100% coarse RCA and 0%
fine RCA), the siliceous sand 0/4 mm was progressively replaced
In this section, the design of the experiment is explained. Firstly,
with fine RCA by volume, in percentages of 25%, 50%, 75%, and
a preliminary analysis is shown to define the coarse RCA content in
finally 100%. As indicated above, the water content was adjusted
the mixes. Secondly, the dosage of the SCCs with different percent-
to each mix, thereby increasing the water-to-cement (w/c) ratio
ages of fine RCA are defined. Thirdly, the set of tests performed on
and the effective w/c ratio, calculated from the water absorption
the mixes is discussed. Finally, the type of analysis of the experi-
rate over 10 min. The values for these ratios and the mix design
mental data is likewise explained.
can be seen in Table 5. The mixes were labelled M0, M25, M50,
M75, and M100 depending on the percentage of fine RCA. In this
3.1. Experiment design table, the dosage of the mixtures is shown in two different ways.
Firstly, in a comparative way, by weight, with respect to the ref-
Firstly, a preliminary analysis served to define the amount of erence mix M0, so that any changes to the fine RCA and the
coarse RCA to use in the mixes. Based on this analysis, it was water content may be easily observed. Secondly, the dosage of
decided to use 100% coarse RCA (see Section 3.2). Later, the fine the mixtures adjusted to 1 m3 is shown, so that they may be
RCA contents (25%, 50%, 75% and 100%), the set of tests to carry easily reproduced. Fig. 3 shows the overall particle size (cor-
out and the number of samples in each test (application of the rected by volume) of mixes M0, M50, and M100, and their Fuller
power method, with a significance level of 5% and a power of curve.
20%) were defined by analyzing other similar research studies
[36,37]. In addition, testing ages were established according to 3.3. Experimental procedure
the standards [34,38,39], in order to evaluate the effect of fine
RCA over time. Therefore, age and fine RCA content were the two A three-stage mixing process was performed for proper hydra-
factors evaluated in this study. tion of all the components, maximization of flowability, and to pre-
vent the absorption of admixture into the aggregates [40]. First,
3.2. Mix design coarse RCA and fine fractions (siliceous sand 0/ 4 mm and/or fine
RCA) were loaded in the concrete mixer bowl with half of the
The dosage of all mixes was defined by the specifications of mix-water and mixed for 30 s. Then, the cement, filler and the rest
EHE-08 [34], based on EC2 [39]. The overall particle size was of the water were added, followed by another 30 s of mixing.

Table 1
Aggregates’ physical properties.

Test Regulations Siliceous Siliceous Coarse Fine EHE-08 limit [34]

[35] gravel sand RCA RCA
Saturated-Surface-Dry (SSD) density (Mg/m3) EN-1097-6 2.62 2.58 2.42 2.37  2.6 kg/dm3
Water absorption 24 h (%) EN-1097-6 0.84 0.25 6.25 7.36 Coarse RCA content < 20%: <7%
Coarse RCA content > 20%: Combination
of RCA and NA coarse fractions: <5%
Water absorption 10 min (%) – 0.66 0.18 5.28 6.03 If RCA content is lower than 20%: <5.5%
Fines content (%) EN-933-1 0.11 1.82 0.17 4.83 <1.50% (coarse fraction)
Equivalent sand (%) EN-933-8 – – – 83 >75
Bulk density (Mg/m3) EN-1097-3 – – 1.26 1.23 –
Los Angeles coefficient (%) (size 10/14 mm) EN-1097-2 – – 35 – <40
Sand friability test (%) (size 0/4 mm) UNE-146404 – – – 16
4 V. Revilla-Cuesta et al. / Construction and Building Materials 263 (2020) 120671

Table 2
Chemical composition (% in weight) of fine RCA through XRF.

SiO2 CaO Al2O3 Fe2O3 SO3 MgO K2O TiO2 P2O5 Others (CO2 . . .)
50.80 20.00 3.74 1.12 1.00 0.63 0.60 0.15 0.06 21.9

Fig. 1. XRD of fine RCA.

Fig. 2. Granulometry of aggregates used.

Table 3 Having completed the mixing process, in-fresh state tests, fol-
Mix design of the mixes of the preliminary analysis (kg). lowing the EFNARC recommendations, were performed [32] and
Mix C0 C50 C100 the specimens were molded to carry out the hardened state tests.
Cement 300 300 300
The tests, the standards, the age of the test, and the type and the
Filler 180 180 180 number of test samples at each dosage are summarized in Table 6.
Water 140 155 165 The samples were held in a wet chamber at a humidity of 95 ± 5%
Siliceous gravel 4/12.5 mm 570 285 0 and at temperature of 20 ± 2 °C until the time of the test. In addi-
Coarse RCA 0/12.5 mm 0 265 525
tion, images from a Scanning Electron Microscope (SEM) were used
Siliceous sand 0/4 mm 1150 1150 1150
Admixture 1 2.20 2.20 2.20 to evaluate the quality of the ITZ in one of the worst performing
Admixture 2 4.40 4.40 4.40 specimens of mix M75.
The mechanical properties measured by these tests were evalu-
ated both descriptively, which allows visual detection of differ-
ences in the behavior of the mixes, and by statistical analysis
Finally, the additive dissolved in 0.2 L of water was added and through one-way ANOVA at a significance level of 5%. This statisti-
mixed for the last 30 s. The aggregates were used under laboratory cal analysis was to study the effect of each factor (age and fine RCA
conditions, at an average temperature and humidity of 20 °C and content) for each mechanical property under study and to establish
45%, respectively, thus adjusting the experimental procedure to homogeneous groups, i.e., factor values for which a certain
the most economically advantageous methodology [41]. mechanical property is significantly equal.
 V. Revilla-Cuesta et al. / Construction and Building Materials 263 (2020) 120671 5

Table 4
Hardened properties (average values) and one-way ANOVA of the preliminary analysis.

Mixes One-way ANOVA. Factor analyzed: coarse

RCA content
Test Age (days) C0 C50 C100 P-value Homogeneous groups
Compressive strength (MPa) 7 67.6 57.8 55.7 0.0002 C50 and C100
28 69.8 61.5 59.3 0.0082
Modulus of elasticity (GPa) 7 39.7 35.9 34.8 0.0001
28 43.4 40.0 39.1 0.0002

Table 5
Mix design.

Comparative dosage with mixture M0 (kg) Dosage (kg/m3)

Mix M0 M25 M50 M75 M100 M0 M25 M50 M75 M100
Cement 300 300 300 300 300 300 295 288 282 276
Filler 180 180 180 180 180 180 176 173 169 166
Water 165 187 205 230 250 165 183 198 215 232
Coarse RCA 0/12.5 mm 525 525 525 525 525 525 518 508 497 487
Fine RCA 0/4 mm 0 265 530 795 1060 0 260 510 748 977
Siliceous sand 0/4 mm 1150 865 575 290 0 1150 850 555 272 0
Admixture 1 2.20 2.20 2.20 2.20 2.20 2.20 2.18 2.14 2.10 2.05
Admixture 2 4.40 4.40 4.40 4.40 4.40 4.40 4.32 4.24 4.15 4.06
Approximate volume (l) 1000 1018 1038 1060 1083 1000 1000 1000 1000 1000
Approximate weight (kg) 2327 2307 2282 2262 2237 2327 2289 2238 2189 2144
w/c 0.561 0.624 0.688 0.763 0.839 0.561 0.624 0.688 0.763 0.839
Effective w/c 0.459 0.470 0.482 0.506 0.530 0.459 0.470 0.482 0.506 0.530

Fig. 3. Mix design granulometry.

3.4. Theoretical calculations Eq. (2), and the medium flexural strength (fct,m,fl), as a function of
the beam height (h) in mm by applying Eq. (3). These last two
The experimental values of each mechanical property were strengths were compared with the values of splitting tensile
compared with the theoretical values obtained from the formulas strength and flexural strength, experimentally obtained at 28 days.
of the Spanish structural concrete standard EHE-08 [34], an adap-
tation of the Eurocode 2 regulation [39], based on the recommen- f c;k ¼ f c;m  8 ð1Þ
dations of the International Federation for Structural Concrete 8 9
(CEB-FIP) [42]. <f ¼ 0:30  f c;k if f c;k  50MPa =
2=3
ct;m
According to EHE-08 [34], experimental compressive strengths ð2Þ
:f ¼ 0:58  f c;k if f c;k > 50MPa ;
1=2
at 28-days can be considered a medium 28-day compressive ct;m

strength (fc,m), with which the characteristic compressive strength   

(fc,k) can be estimated, according to Eq. (1). Having done so, the h
f ct;m;fl ¼ max 1:6   f ct;m ; f ct;m ð3Þ
medium tensile strength (fct,m) can be determined, according to 1; 000
6 V. Revilla-Cuesta et al. / Construction and Building Materials 263 (2020) 120671

Table 6 ticity) measured in this study. These formulas can be used as a

Tests, regulations, and samples used for each mix. basis for more general models taken from the literature, for accu-
Test Regulations Test Number and type of rate estimation of the properties of SCC made with RCA.
[35] age sample tested per On the other hand, the value of the dynamic modulus of elastic-
(days) mixture ity (Ed) at different ages of curing can be calculated from the test
Slump flow EN 12350-8 Sample of fresh concrete results with Eq. (9) [33], using the density of the concrete (q),
V-funnel EN 12350-9 the UPV (Vl), and the Poisson coefficient (m). All the above-
2-bar L-box EN 12350-10
Sieve segregation EN 12350-11
mentioned properties were precisely established with the tests
Fresh density EN 12350-6 performed.
ð1 þ mÞ  ð1  2  mÞ
Air content EN 12350-7
Compressive strength EN 12390-3 1, 7, 3 cylindrical samples Ed ¼ q  V 2l  ð9Þ
28, 90 10  20 cm at each age 1m
Hammer rebound index EN 12504-2 2 cylindrical samples
10  20 cm at each age
Ultrasonic Pulse EN 12504-4 2 prismatic samples 4. Results and discussion
Velocity (UPV) 10  10  40 cm at
each age The experimental results of the in-fresh and the hardened
Hardened density EN 12390-7 28 2 cubic samples
10  10  10 cm
states, as well as the theoretical estimation of the mechanical prop-
Splitting tensile EN 12390-6 7, 28, 2 cylindrical samples erties from the concrete design standards are presented in this
strength 90 15  30 cm at each age section.
Flexural strength EN 12390-5 2 prismatic samples
10  10  40 cm at
4.1. Fresh properties
each age
Static modulus of EN 12390-13 3 cylindrical samples
elasticity and 10  20 cm at each age It is clear from the bibliography that the addition of RCA clearly
Poisson coefficient reduces the flowability of the SCC, due to its higher water absorp-
Stress–strain curves – 90 2 cylindrical samples tion, resulting in a lower effective w/c ratio for any given w/c ratio
10  20 cm
without RCA [20,33,43]. In addition, the crushing process of the
Abrasion resistance EN 1340 2 cubic samples
EN 14157 10  10  10 cm RCA produces irregular shapes, which means that the aggregate
particles will not flow easily in the paste [44,45]. Although the
effective w/c ratio decreases with the coarse RCA, its irregular
EHE-08 [34] specifies the formula for estimating the static con- shaped particles have the most notable effects [33,46]. The pre-
crete elasticity modulus at any age, Eq. (4). The expression contains dominant effect of the fine RCA results from its high level of water
the following factors: curing time in days (t); the 28-day secant absorption that decreases the effective w/c ratio [43]. Regarding
modulus of elasticity (Ec,m) calculated with Eq. (5); the medium the maximum diameter obtained in the slump-flow test, the
compressive concrete strength at t days (fc,m (t)), obtained with decrease in initial flowability is as much as 36% with 100% coarse
Eq. (6); and, the secant modulus of elasticity at t days (Ec,m (t)). RCA and up to 19% with 100% fine RCA [43–45].
The units of strength and the modulus of elasticity in all the The adjustment of the water absorbed by the RCA solves the
expressions can only be introduced in Megapascals (MPa). water absorption problem, preventing the effective w/c ratio from
decreasing [19,37]. Another possibility is to increase the propor-
!0:3 tion of the finest aggregate fraction (<0.125 mm), thereby increas-
f c;m ðtÞ
Ec;m ðt Þ ¼  Ec;m ð4Þ ing the volume of paste. Balanced proportions of fine and coarse
f c;m aggregates are therefore fundamental for good performance of
SCC [21].
qffiffiffiffiffiffiffiffi In the concretes under study, the water content was adjusted by
Ec;m ¼ 8; 500  3
f c;m ð5Þ increasing the percentage of fine RCA aggregate, thereby increasing
the effective w/c ratio. In addition, that increase of fine RCA aggre-
( "
 1=2 #) gate also led to an increase in the fine fraction (<0.125 mm) within
28
f c;m ðt Þ ¼ exp 0:2  1   f c;m ð6Þ the overall granulometry of the mixtures. Both aspects compen-
t sated the negative effects of RCA and even improved the slump
flow and the passing ability of the SCC. Some of the in-fresh state
The structural concrete design guidelines of the American Con-
tests are shown in Fig. 4 and their results, presented in Table 7,
crete Institute (ACI) [38] contain formulas, Eqs. (7) and (8), for cal-
are consistent with the above-mentioned aspects.
culating the tensile strength of concrete and its modulus of
As reported in similar studies [43], the lower density of the RCA,
elasticity from its compressive strength, respectively. The variables
compared to the NA, and the increase in water content decreased
of the equation refer to compressive strength, (fc), tensile strength,
the fresh density of the SCC [46,47]. The air content, very similar
(ft), and the modulus of elasticity, (E). These expressions are easier
in all mixtures, mainly depended on the admixtures and the chem-
to apply than those indicated above, and their validity is also ana-
ical reactions they produced, with no clear influence of fine RCA.
lyzed in this study.
Based on the results of the in-fresh state tests and the EFNARC
0:5 recommendations [32], the developed SCCs presented:
f t ¼ 0:56  f c ð7Þ

qffiffiffiffi
A slump-flow class SF2 (maximum diameter between 660 and

E ¼ 4; 700  f c ð8Þ 760 mm).

A viscosity class, according to the slump-flow test, VS2 (time to
Simple regression equations are provided, to complement this reach a diameter of 500 mm greater than 2 s) and, according to
analysis, which were developed for the mechanical properties the V-funnel test, VF2 (time to empty the V-funnel greater than
(splitting tensile strength, flexural strength, and modulus of elas- 8 s), except for the M0 mixture which was class VF1.
 V. Revilla-Cuesta et al. / Construction and Building Materials 263 (2020) 120671 7

Fig. 4. In-fresh state tests for M50 mix. (a) slump-flow; (b) V-funnel; (c) sieve segregation.

Table 7
In-fresh state test results.

Test/Mix M0 M25 M50 M75 M100

Viscosity t500 slump flow test (s) 3.40 3.60 4.00 4.20 4.80
Slump flow (mm) 680 695 730 740 755
Viscosity V-funnel test (s) 6.40 8.40 10.20 12.60 15.20
Passing ability L-box test H2/H1 0.86 0.88 0.92 0.93 0.94
Sieve segregation (%) 1.70 1.52 1.67 1.53 1.36
Air content (%) 4.80 3.75 4.35 4.00 4.15
Fresh density (Mg/m3) 2.40 2.31 2.25 2.16 2.12

A passing-ability class PA1, according to the 2-bar L-box test 4.2.1. Density
(blocking ratio greater than 0.80). The lower density of RCA compared to NA reduced the hardened

A segregation-resistance class SR2 (sieve segregation less than density of the SCC, as the RCA content increased (Table 8), in line
15%). with the existing bibliography [33,37,48]. The decrease of the hard-
ened density compared to the fresh density showed no clear
The maximum diameter in the slump-flow test increased by trends, although it was higher in the mixtures with a higher NA
11% (from 680 mm for M0 to 752 mm for M100), following the content (mixture M0 showed a decrease of 4.2%, while this
addition of 100% fine RCA, which in turn increased the effective decrease for mixture M50 was only 0.4%), which may be due to
w/c ratio by 15% (from 0.46 to 0.53). In this case, the passing ability the lower water evaporation, because of the higher water absorp-
was also increased by 9% (from 0.86 to 0.94). The fines content of tion of fine RCA [33].
the mix M25 influenced the flowability of the SCC. A fine RCA con-
tent of 25% and an effective w/c ratio of 0.47, slightly higher than
0.46 (corresponding to mix M0), led to a higher slump flow and
passing ability. However, it was not possible to compensate all 4.2.2. Compressive strength, UPV, and hammer rebound index
the negative effects of RCA by increasing the proportion of water The addition of coarse RCA is known to reduce the compressive
and fines, as viscosity was negatively affected, resulting in higher strength of concrete [49], although reductions of just 5 MPa for
slump-flow viscosity and emptying times (V-funnel test). This per- 100% coarse RCA content were obtained in some studies [45,50].
formance can be explained by two aspects: the irregular shapes of The reduction was higher when fine RCA was added [36], and in
the coarser fine RCA particles that move more slowly within the some studies the decrease caused by 100% fine RCA in SCC reached
cement paste than the rounded siliceous sand particles; and the 26% compared to the reference concrete (0% coarse and fine RCA)
higher proportion of cement paste, resulting from the increased [20,51]. Nevertheless, if the effective w/c ratio is reduced, the loss
volumes of fines, following the addition of RCA, leaving the SCC of strength can be compensated, especially when only coarse RCA
with a more viscous consistency. is used [33,46]. In the present study, as expected, the compressive
Resistance to segregation was not negatively affected by the strengths of the concretes under study decreased as their fine RCA
addition of fine RCA and was even improved by the increase of content increased and with it, the w/c ratio (Table 5), as shown in
the fine RCA content. In fact, the sieve segregation of the mixture Fig. 6. At 28 days, the decrease in strength, with regard to the ref-
M100 (1.36%) was 25% lower than that of the mixture M0 erence mixture M0 (60 MPa), was 7% for mixture M25 (56 MPa),
(1.70%). Even though fine RCA had a higher content of particles 27% for mixture M50 (44 MPa), and 47% and 51% for mixtures
smaller than NA (0.125 mm), the use of fine RCA produced higher M75 and M100, respectively (strengths of 31 and 30 MPa). Regard-
long-term water absorption levels that reduced segregation. ing the overall trend, mixture M25 showed a much lower than
expected strength decrease, possibly due to its very similar effec-
tive w/c ratio to mix M0.
The compressive strength results showed less dispersion with
4.2. Hardened state behavior higher additions of fine RCA content, although that dispersion
increased at advanced ages. The strengths of the mixtures with
The analysis of the mechanical properties aims to evaluate the high RCA contents were more closely limited at early ages.
effect of fine RCA on the development of strength at different cur- The compressive strength developed at 1 and 7 days, with
ing ages. Some of these tests, such as for splitting tensile strength, regard to the strength at 28 days, were in percentage terms 78%
flexural strength, and modulus of elasticity, are shown in Fig. 5. and 94% for the reference mix, M0; 60% and 90% for M50; and
8 V. Revilla-Cuesta et al. / Construction and Building Materials 263 (2020) 120671

Fig. 5. Mechanical tests: splitting tensile strength (left); flexural strength (middle); modulus of elasticity (right).

Table 8
Hardened density of the mixes.

Test/Mix M0 M25 M50 M75 M100

Hardened density (Mg/m3) 2.30 2.29 2.24 2.13 2.10
Decrease with respect to its fresh density (%) 4.2 0.9 0.4 1.4 1.0
Decrease with respect to the M0 density (%) 0 0.4 2.6 7.4 8.7

Fig. 6. Compressive strength of the SSC mixtures.

47% and 86% for M100. Strength development, in percentage terms, (Fig. 10) and the flexural strength (Fig. 11), a widely reported
was slower as the fine RCA content increased. behavior in the literature [33,37,51]. Nevertheless, the harmful
The UPV and the hammer rebound tests, two conventional effect of fine RCA can be compensated, and if the dosage is cor-
methods for estimating the compressive strength of non-recycled rectly modified, it is possible to obtain a splitting tensile strength
concrete, were measured at all ages. Pearson’s symmetrical corre- equal to that of the reference concrete [27]: reducing the water
lation matrix between these three properties for all mixtures and content, increasing the cement content, or using alternative bin-
ages simultaneously is shown in Fig. 7. A clear linear relationship ders such as fly ash or silica fume [53,54].
between these variables can be observed, as in other similar stud- In this study, the decrease was mainly observed for the flexural
ies [28,52], although an individual analysis carried out in each mix strength, with different behavior in two groups (fine RCA up to 50%
showed that the exponential model had the best overall fit. Table 9 and above 50%). Compared to M0, the decrease of the flexural
shows the least square exponential adjustment of compressive strength at 28 days of the mixture M100 was 30% (3.4 versus
strength (variable ‘‘y”) as a function of the UPV (variable ‘‘x”) and 4.9 MPa) and was only 22% for the splitting tensile strength (2.5
the hammer rebound index (variable ‘‘z”). The adjustment had a versus 3.2 MPa). However, there were some exceptions, such as
coefficient R2 higher than 0.90 in 80% of the cases. Mixture M100 the higher flexural strength of mix M100 compared to M75 at
had the coefficients R2 closest to 1.00. The validity of these indirect 90 days, due to the low adhesion between the cement paste and
measures was not influenced by the addition of RCA. Figs. 8 and 9 the coarse aggregate, in the mixes with high fine RCA content. It
show this adjustment for each mixture and it may be noted that was observed that high additions of fine RCA also favored a more
the increase in RCA content increased the curvature of the models. uniform strength over time, and a less predictable behavior.
Some studies have shown that the theoretical tensile strength
4.2.3. Splitting tensile strength and flexural strength value, calculated according to the EC2 [39], was lower than the
The results, obtained at 7, 28, and 90 days, showed that the experimental value in SCC made with coarse RCA, whilst the exper-
addition of fine RCA decreased both the splitting tensile strength imental value was overestimated in SCCs with high fine RCA con-
 V. Revilla-Cuesta et al. / Construction and Building Materials 263 (2020) 120671 9

Fig. 7. Pearson’s symmetrical correlation matrix for compressive strength, hammer rebound index and UPV.

Table 9
Exponential adjustment between UPV and hammer rebound index and compressive strength.

Test/Mix M0 M25 M50 M75 M100

UPV (km/s) 1 day 3.94 3.76 3.73 3.27 3.19
7 days 4.28 4.00 4.03 3.63 3.68
28 days 4.32 4.20 4.09 3.80 3.71
90 days 4.34 4.30 4.09 3.79 3.71
Exponential adjustment: y = 3.666 exp(0.643x) y = 1.854 exp(0.811x) y = 0.155 exp(1.376x) y = 0.312 exp(1.232x) y = 0.130 exp(1.461x)
compressive strength-UPV
Coefficient R2 UPV 0.96 0.87 0.99 0.95 0.98
Hammer rebound index 1 day 46 40 34 27 24
7 days 56 52 42 33 33
28 days 58 54 42 39 37
90 days 58 56 45 39 40
Exponential adjustment: y = 17.443 exp(0.021z) y = 11.669 exp(0.029z) y = 5.236 exp(0.048z) y = 4.804 exp(0.050z) y = 3.871 exp(0.054z)
compressive strength- hammer
rebound index
Coefficient R2 hammer rebound 0.98 0.99 0.92 0.85 0.98
index

Fig. 8. Adjustment of the compressive strength as a function of the UPV.

tent [37]. The theoretical tensile strength values as per EHE-08 mated with Eq. (7) from the ACI [38], it can be observed that the
[34], shown in Fig. 12 and calculated with Eq. (2), were higher than theoretically calculated values also overestimated the results. The
the experimentally obtained results for all the mixtures in this SCCs with the highest fine RCA content were those that showed
study, although the adjustment was improved by increasing the the least difference between both values: in M0, the theoretical
fine RCA content (the experimental tensile strength of M100 was value at 28 days (4.3 MPa) was 34% higher than the experimental
higher than the theoretical value). In relation to the values esti- value (3.2 MPa), while this overestimation in the mixture M100
10 V. Revilla-Cuesta et al. / Construction and Building Materials 263 (2020) 120671

Fig. 9. Adjustment of the compressive strength as a function of the hammer rebound index.

Fig. 10. Splitting tensile strength.

was only 20% (3.0 versus 2.5 MPa). As the increase in compressive STS ¼ 8:46  0:79  CS þ 0:04  CS2 ð10Þ
strength after 7 days was higher than the increase in splitting ten-
sile strength, the adjustment was better for advanced ages. Simi-  2
19:26
larly, the design values of flexural strength (Fig. 13) also FS ¼ 2:55  ð11Þ
CS
overestimated the experimental values, once again with a better
fit as fine RCA was added. From all the above, it can be deduced
that both the splitting tensile strength and the flexural strength 4.2.4. Static modulus of elasticity and Poisson coefficient
were, in general, lower than expected, especially in the case of con- The presence of adhered mortar and the ITZs, which were weak
cretes with few recycled fines. The use of the equations from the and not very dense, meant that the SCC with RCA was, in general,
standards applied to the test mixtures would imply strength over- more deformable than the SCC with NA [25,33], observations sup-
estimations of around 25%, especially with fine RCA contents up to ported by the results from this study that are shown in Fig. 14. A
50%. higher fine RCA content led to a lower modulus of elasticity at all
The standard formulas [34,38] relate both tensile strength and ages: at 28 days, M100 presented a modulus of elasticity of 18.8
flexural strength to compressive strength by means of a square GPa, 54% lower than the mix M0 (40.6 GPa). The addition of
or cubic root. Nevertheless, the best-fit model (coefficient R2 of 100% fine RCA to an SCC with 100% coarse RCA, as performed in this
0.86) of the splitting tensile strength (STS, in MPa) as a function study, led to a greater decrease in the modulus of elasticity than
of compressive strength (CS, in MPa), shown in Eq. (10), established when this amount of fine RCA is added to an SCC with no coarse
a polynomial relationship between both variables. In contrast, the RCA, which is around 24% [37,51].
flexural strength (FS, in MPa), Eq. (11), with a coefficient R2 of 0.88, The trends for the modulus of elasticity over the curing, once
has an inverse relationship with the compressive strength. again showed a different behavior between the mixtures with
 V. Revilla-Cuesta et al. / Construction and Building Materials 263 (2020) 120671 11

Fig. 11. Flexural strength.

Fig. 12. Relationship between the experimental tensile strength and the estimated value as per EHE-08 (left) and ACI (right).

low/medium percentages and those with high percentages of fine

RCA: the mixtures with low percentages of fine RCA (M0 and
M25) showed a very marked increase in the modulus of elasticity
up to 28 days, while the increase in concretes with higher contents
(M75 and M100) was practically uniform over time.
The adjustment of the theoretical values, obtained with Eq. (4),
to the experimental values was quite poor (Fig. 15). In mixtures
with low percentages of fine RCA (M0 and M25), the design values
were lower than the experimental values (at 28 days, the design
value of mixture M25 was 32.4 GPa, 15% lower than the experi-
mental value, 38.2 GPa). The theoretical expressions overestimated
the modulus of elasticity in the mixtures with high contents of fine
RCA (M75 and M100), by approximately 40% at all ages of curing.
Mixture M50 was in an intermediate situation: its theoretical value
at 7 days (28.2 GPa) was higher than the experimental value (29.9
GPa), with the opposite situation at 90 days (theoretical value of
30.7 GPa and experimental value of 31.7 GPa). The same situation
was observed for the ACI [38] formula, once again showing that
mixture M50 had the best fit. These results are in line with the con-
Fig. 13. Relationship between the experimental flexural strength and the estimated
clusions of other studies, which also showed that the theoretical
value as per EHE-08. expressions underestimated the modulus of elasticity in SCCs with
12 V. Revilla-Cuesta et al. / Construction and Building Materials 263 (2020) 120671

Fig. 14. Static modulus of elasticity.

Fig. 15. Relationship between compressive strength and modulus of elasticity according to EHE-08 (left) and ACI (right).

low percentages of RCA and overestimated it in mixtures with high (0.2) was higher than the experimental values, except for concretes
percentages of RCA [37,55]. with less than 50% fine RCA at 7 days, that had slightly higher
Eq. (12) shows that there is an exponential relationship experimental values.
between the modulus of elasticity (ME, in GPa) and the logarithm
of compressive strength (CS, in MPa), similar to that shown by 4.2.5. Stress–strain curves
EHE-08 [34] in Eq. (6). Its coefficient R2 is 0.96. The stress–strain curves that are used to evaluate the plastic
behavior of concrete usually show an elastic behavior at low load
ME ¼ expð0:62 þ 1:05  lnðCSÞÞ ð12Þ
levels and reduced fracture strain values (theoretical value of
Increasing percentages of fine RCA increased both the deforma- 3500 mm/m) [34,38,39]. Higher fines contents will usually result
bility of the concrete in the load direction and the volume variation in higher strain values after the ultimate strength value of the con-
under load, as shown by the lower Poisson coefficient (Fig. 16) of crete is reached [59]. When this aspect was addressed in recycled
the concrete with additions of fine RCA: after 28 days, the differ- concrete, it was concluded that coarse RCA reduced fracture strain
ence of the Poisson coefficient between mixtures M0 (0.20) and [60]. No reference to this behavior was found for SCCs made with
M100 (0.16) was 19%. This trend was similar to the one obtained coarse and fine RCA, such as those reported in this article. Cylindri-
in vibrated concretes with only coarse RCA [56,57], although a con- cal specimens of 10 and 20 cm in diameter and height, respec-
stant w/c ratio appeared to prevent the Poisson coefficient from tively, were subjected to compressive strength tests until failure
decreasing [58]. The temporal evolution of this coefficient was sim- at 90 days, yielding the curves shown in Fig. 17. The stress and
ilar in all the mixtures: a marked decrease in the first days, less strain values were recorded at a frequency of 15 Hz.
noticeable as time goes by, although M100 showed a much more The stress–strain curves reflected the decrease, both in com-
remarkable decrease after 28 days than the rest of the mixtures. pressive strength (ultimate strength) and in the modulus of elastic-
The conventional value of the Poisson coefficient for concrete ity, as the fine RCA content increased, as previously discussed
 V. Revilla-Cuesta et al. / Construction and Building Materials 263 (2020) 120671 13

Fig. 16. Poisson coefficient.

Fig. 17. Stress–strain curves of the mixtures.

(Section 4.2.2). Other relevant aspects of the structural design with (3300 mm/m) were the closest to the theoretical value. Thus,
this type of concrete are: the mixtures with higher fine RCA contents presented higher
safety design coefficients, as they exceeded the theoretical

The proportional limit (the point where linear elastic strain values calculated with the formulas from the standards. This
ends) was produced for a deformation of 720 mm/m for M0 behavior was contrary to the one observed when only coarse
and 580 mm/m for M100, increasing the plastic deformation RCA was added [60].
with the fine RCA content.
The strain values corresponding to the ultimate strength

An increase in the fine RCA content led to a higher fracture increased with higher contents of fine RCA in the mixtures
strain: mixtures M0 and M25 showed strain values that were (2650 mm/m for mixture M0 and 3250 mm/m for mixture
18.9% lower than mixtures M75 and M100 (3000 mm/m ver- M100). The ratios between the fracture strain and peak strain
sus 3700 mm/m). The strain values of mixture M50 for mixtures M0 and M100 were 1.16 and 1.13, respectively,
14 V. Revilla-Cuesta et al. / Construction and Building Materials 263 (2020) 120671

and the remaining strain after breakage increased in SCCs with

low fine RCA contents. The reference specimen, mixture M0,
showed the highest safety factor after failure.

4.2.6. Dynamic modulus of elasticity through the UPV

The dynamic modulus of elasticity decreased with the addition
of fine RCA. According to other studies, in mixtures with 50% coarse
RCA, the decrease was approximately 4%, reaching 9% for 100%
replacements [33,61]. No other study has been found in which this
property is evaluated in SCCs with fine RCA. The results of this
study (Fig. 18) showed the same overall trend as the static modu-
lus of elasticity (Fig. 14), both in terms of fine RCA content and its Fig. 19. Abrasion test. Left: footprint on the M50 mix. Right: sample set up in the
temporal evolution. The results for all properties based on the cal- testing machine.
culation of this modulus of elasticity, according to Eq. (9), were
consistent with each other.
Two groups of concretes may be distinguished: high and low/ there were two different groups of concretes, according to the
medium fine RCA content. Mixtures with fine RCA content of less results of this test: high and low/medium fine RCA content.
than 50% had a lower dynamic modulus of elasticity through the
UPV than the static modulus of elasticity (e.g. the values of the sta- 4.2.8. Statistical analysis of mechanical properties
tic and dynamic modulus of elasticity of M25 at 28 days were 38.2 The descriptive analysis was completed with a statistical analy-
and 36.5 GPa respectively). Meanwhile mixtures M75 and M100 sis based on one-way ANOVA (Table 10). It showed that both age
showed higher stiffness under dynamic loading: the dynamic mod- and fine RCA content significantly influenced the behavior of the
ulus of elasticity through the UPV of mixture M100, at 28 days, was mixes in all tests (p-value lower than the significance level consid-
44.1% higher than the static modulus of elasticity (27.1 GPa versus ered, 0.05). The homogeneous groups revealed that, in general,
20.1 GPa). mixes M0 and M25, as well as M75 and M100, have no significant
difference in their mechanical properties regardless the age. In
contrast, the mechanical properties of mixes with low fine RCA
4.2.7. Abrasion resistance contents (M0 and M25) hardly showed significant variations
Resistance to abrasion was evaluated by the footprint caused by between 28 and 90 days, while the mechanical properties of mixes
an abrasive disc on the surface of the material on cubic samples of M75 and M100, at 7, 28 and 90 days, can be considered statistically
10  10  10 cm at 90 days (Fig. 19), which measures the hardness equal in several tests. Mix M50 showed an intermediate behavior.
of the surface cement paste [62]. According to other studies, the
addition of fine RCA increased the size of the footprint [63]. Never- 4.2.9. SEM analysis
theless, the use of RCA from a concrete with a strength of over As commented in the introduction, the presence of adhered
70 MPa can improve the surface resistance of the concrete [64]. mortar in the RCA leads to the formation of an ITZ that is less dense
In this research, the increase in fine RCA generated a cement and of poorer quality than the ITZ between the NA and the cemen-
paste with a lower surface hardness, which increased the footprint titious matrix [22,25], as may be observed in the image of Fig. 21,
size, as shown in Fig. 20. These results were in line with the results obtained from a low-strength M75 test specimen. In this image,
of the 90-day rebound index: a higher rebound index was pro- two particles of siliceous aggregate of the RCA can be seen in both
duced in those mixtures with a smaller footprint width (mix the lower right and left-hand-side corners; the old cementitious
M75 had a lower rebound index, 39, and a footprint size of matrix (mortar adhered to the NA) of a darker color in the central
21 mm that was 27.3% larger than the M0 footprint). Once again, part, corroborated by the microanalysis spectrum; and, the new

Fig. 18. Dynamic modulus of elasticity through the UPV.

 V. Revilla-Cuesta et al. / Construction and Building Materials 263 (2020) 120671 15

Fig. 20. Resistance to abrasion test results. The hammer rebound index and the increase in footprint size, in relation to mixture M0, are shown between brackets.

Table 10
One-way ANOVA of mechanical properties.

Test Factor Condition P-value Homogeneous groups

Compressive strength Fine RCA percentage Age of 1 day 0.0001 M75 and M100
Age of 7 days 0.0001 M0 and M25; M75 and M100
Age of 28 days 0.0002 M0 and M25; M75 and M100
Age of 90 days 0.0001 M0 and M25; M75 and M100
Age 0% fine RCA 0.0007 7, 28 and 90 days
25% fine RCA 0.0003 7, 28 and 90 days
50% fine RCA 0.0001 7, 28 and 90 days
75% fine RCA 0.0002 7, 28 and 90 days
100% fine RCA 0.0001 28 and 90 days
Splitting tensile strength Fine RCA percentage Age of 7 days 0.0017 M0 and M25; M50, M75 and M100
Age of 28 days 0.0233 M75 and M100
Age of 90 days 0.0282 M75 and M100
Age 0% fine RCA 0.0370 None
25% fine RCA 0.0230 None
50% fine RCA 0.0239 28 and 90 days
75% fine RCA 0.0351 7 and 28 days
100% fine RCA 0.0349 28 and 90 days
Flexural strength Fine RCA percentage Age of 7 days 0.0173 M25 and M50; M75 and M100
Age of 28 days 0.038 M0, M25 and M50; M75 and M100
Age of 90 days 0.017 M0 and M25
Age 0% fine RCA 0.0001 28 and 90 days
25% fine RCA 0.0004 28 and 90 days
50% fine RCA 0.0001 28 and 90 days
75% fine RCA 0.0003 7, 28 days and 90 days
100% fine RCA 0.0015 7 and 28 days
Static modulus of elasticity Fine RCA percentage Age of 7 days 0.0002 M75 and M100
Age of 28 days 0.0001 M0 y M25; M75 y M100
Age of 90 days 0.0002 M0 y M25; M75 y M100
Age 0% fine RCA 0.0260 28 and 90 days
25% fine RCA 0.0173 28 and 90 days
50% fine RCA 0.0104 28 and 90 days
75% fine RCA 0.0256 7, 28 and 90 days
100% fine RCA 0.0394 7, 28 and 90 days
Poisson coefficient Fine RCA percentage Age of 7 days 0.0083 M25 and M50
Age of 28 days 0.0158 M0 and M25
Age of 90 days 0.0118 M0 and M25
Age 0% fine RCA 0.0285 None
25% fine RCA 0.0327 None
50% fine RCA 0.0211 None
75% fine RCA 0.0441 28 and 90 days
100% fine RCA 0.0303 None
Dynamic modulus of elasticity through the UPV Fine RCA percentage Age of 7 days 0.0023 M25 and M50; M75 and M100
Age of 28 days 0.0016 M75 and M100
Age of 90 days 0.0010 M0 y M25; M75 y M100
Age 0% fine RCA 0.0344 28 and 90 days
25% fine RCA 0.0465 None
50% fine RCA 0.0351 28 and 90 days
75% fine RCA 0.0222 28 and 90 days
100% fine RCA 0.0455 7, 28 and 90 days
Abrasion resistance Fine RCA percentage Age of 90 days 0.0043 M0, M25 and M50; M75 and M100
16 V. Revilla-Cuesta et al. / Construction and Building Materials 263 (2020) 120671

Fig. 21. SEM analysis of mix SA75.

cementitious matrix of a lighter color on the right-hand side and
In general, recycled SSCs with high fine RCA contents showed
towards the upper right-hand-side corner. It is clear that the ITZ more uniform mechanical properties over time than those with
has failed due to its poor quality; the new cement paste shows less fine RCA. According to one-way ANOVA, mixes with 0% and
low adhesion to the RCA, both with the mortar phase and the silic- 25% of fine RCA showed the same strength and stiffness values
eous aggregate phase. This poor behavior of the ITZ, when adding at 28 and 90 days, meanwhile mixes with 75% and 100% of fine
large quantities of fine RCA, explained the variable behavior of RCA had the same values at 7, 28, and 90 days.
the mixtures with high contents of fine RCA, especially in relation
The adjustment of the experimental values to the theoretical
to the splitting tensile strength and the flexural strength. design values provided by the regulations was different for each
property and each recycled SCC. The theoretical values of tensile
5. Conclusions strength were higher than the experimental values, both from
the European standard [34,39,42] and the ACI regulations
In this study, the physical and mechanical behavior, at different [38], a fit that was improved by increasing the fine RCA content.
curing ages, of a Self-Compacting Concrete (SCC) manufactured The theoretical modulus of elasticity, calculated with both the
with coarse and fine Recycled Concrete Aggregate (RCA) has been European and the USA standard, underestimated the experi-
studied. A preliminary statistical analysis showed an acceptable mental value in mixtures with up to 25% fine RCA, while the
mechanical behavior of the concretes manufactured with 100% of modulus of elasticity was overestimated by percentages greater
coarse RCA. Therefore, SCCs with 100% RCA in the coarse fraction than 75%.
and different percentages (0%, 25%, 50%, 75% and 100%) of fine
Indirect measurements, such as the Ultrasonic Pulse Velocity
RCA in substitution of Natural Aggregate (NA), 0/4 mm, were per- (UPV) and the hammer rebound index yielded accurate com-
formed. The conclusions relating to the effect of the incorporation pressive strength estimates for the concretes under analysis,
of fine RCA on these recycled SSCs are set out below: with the most linear adjustment for the mixtures with low con-
tents of fine RCA. The best statistical adjustment was obtained

Increasing the mix water according to RCA water absorption for the mixture with 100% fine and coarse RCA.
was insufficient to maintain the flowability of the SCC manufac- In view of the above, SCC of optimum flowability and with ade-
tured with NA 0/4 mm when fine RCA was added. Instead, the quate mechanical behavior can be produced using high RCA con-
effective water/cement ratio had to be increased. The increase tents in such a way that the SCC is valid for use in structural
in the fines content of the RCA, in comparison with NA, was components. In the present study, the combination of 100% coarse
favorable in the slump-flow test and the passing ability. RCA and a 50% fine RCA was the limit value, after which a notable

All developed recycled SCCs had fresh property values within worsening of the mechanical properties ensued, suggesting that
the limits of the EFNARC recommendations [32]. SCC were the use of a fine RCA content higher than 50% in structural con-
obtained in slump-flow class SF2, viscosity class VS2 and VF2, cretes would not be advisable in structural applications from a
passage ability class PA1, and resistance to segregation class mechanical point of view. Nevertheless, when service require-
SR2. ments are considered, the higher deformability of the SCC due to

The mechanical properties of the SCC worsened as the percent- higher fine RCA contents, shown in the stress–strain curves, is a
age of fine RCA increased at all ages of curing. On the one hand, good reason for limiting the content of fine RCA to 25%.
fine RCA contents of 0% and 25% allowed to obtain compressive
strengths above 50 MPa (high strength concrete). On the other
hand, the SCC with a high content of fine RCA (75% and 100%) CRediT authorship contribution statement
showed a worsening of their mechanical properties with com-
pressive strengths below 35 MPa. The one-way ANOVA showed Víctor Revilla-Cuesta: Conceptualization, Data curation, Formal
that there was no significant difference between the mechanical analysis, Investigation, Methodology, Software, Writing - original
properties for mixes with fine RCA percentages of 0% and 25%, draft. Vanesa Ortega-López: Conceptualization, Methodology, Pro-
and for 75% and 100%. ject administration, Funding acquisition, Resources, Supervision,

The stress–strain curves showed that the mixtures with 75 and Writing - review & editing. Marta Skaf: Data curation, Formal anal-
100% fine RCA were much more deformable, with fracture ysis, Investigation, Software, Project administration, Writing -
strains of 3.7‰, 18.9% higher than the fracture strain of the mix- review & editing. Juan Manuel Manso: Funding acquisition,
ture with 0% fine RCA, which had a value of 3‰. Resources, Supervision, Validation, Writing - review & editing.
 V. Revilla-Cuesta et al. / Construction and Building Materials 263 (2020) 120671 17

Declaration of Competing Interest [23] V.W.Y. Tam, K. Wang, C.M. Tam, Assessing relationships among properties of
demolished concrete, recycled aggregate and recycled aggregate concrete
using regression analysis, J. Hazard. Mater. 152 (2008) 703–714, https://doi.
The authors declare that they have no known competing finan- org/10.1016/j.jhazmat.2007.07.061.
cial interests or personal relationships that could have appeared [24] J.T. San-José, J.M. Manso, Fiber-reinforced polymer bars embedded in a resin
concrete: study of both materials and their bond behavior, Polym. Compos. 27
to influence the work reported in this paper.
(2006) 315–322, https://doi.org/10.1002/pc.20188.
[25] K.P. Verian, W. Ashraf, Y. Cao, Properties of recycled concrete aggregate and
Acknowledgements their influence in new concrete production, Resour. Conserv. Recycl. 133
(2018) 30–49, https://doi.org/10.1016/j.resconrec.2018.02.005.
[26] A. Yacoub, A. Djerbi, T. Fen-Chong, Water absorption in recycled sand: new
The authors wish to express their gratitude to: the Spanish Min- experimental methods to estimate the water saturation degree and kinetic
istry MCI, AEI, EU and ERDF [grant number FPU17/03374]; the filling during mortar mixing, Constr. Build. Mater. 158 (2018) 464–471,
https://doi.org/10.1016/j.conbuildmat.2017.10.023.
Junta de Castilla y León (Regional Government) and ERDF [grant [27] V. Revilla-Cuesta, M. Skaf, F. Faleschini, J.M. Manso, V. Ortega-López, Self-
numbers UIC-231, BU119P17]; Youth Employment Initiative (JCyL) compacting concrete manufactured with recycled concrete aggregate: an
and ESF [grant number UBU05B_1274]; and finally, the University overview, J Clean Prod 262 (2020), https://doi.org/10.1016/j.
jclepro.2020.121362 121362.
of Burgos [grant numbers SUCONS, Y135.GI].
[28] M. Kazemi, R. Madandoust, J. de Brito, Compressive strength assessment of
recycled aggregate concrete using Schmidt rebound hammer and core testing,
References Constr. Build. Mater. 224 (2019) 630–638, https://doi.org/10.1016/
j.conbuildmat.2019.07.110.
[29] B.B. Mukharjee, S.V. Barai, Influence of Nano-Silica on the properties of
[1] ANEFA, ANEFHOP, ASEFMA, Informes de situación económica sectorial.
recycled aggregate concrete, Constr. Build. Mater. 55 (2014) 29–37, https://
Sectorial economic progress reports, (2018).
doi.org/10.1016/j.conbuildmat.2014.01.003.
[2] M. Sandanayake, G. Zhang, S. Setunge, Estimation of environmental emissions
[30] R. Latif Al-Mufti, A.N. Fried, The early age non-destructive testing of concrete
and impacts of building construction – a decision making tool for contractors,
made with recycled concrete aggregate, Constr. Build. Mater. 37 (2012) 379–
J. Build. Eng. 21 (2019) 173–185, https://doi.org/10.1016/j.jobe.2018.10.023.
386, https://doi.org/10.1016/j.conbuildmat.2012.07.058.
[3] Eurostat, Waste statistics, (2017).
[31] M.C.S. Nepomuceno, L.F.A. Bernardo, Evaluation of self-compacting concrete
[4] J.I. Tertre, Construction and Demolition Waste (CDW), Conference Paper,
strength with non-destructive tests for concrete structures. Appl. Sci. 9 (2019)
Congreso Nacional del Medio Ambiente, CONAMA2016 (2016).
5109, doi: 10.3390/app9235109.
[5] M. Behera, S.K. Bhattacharyya, A.K. Minocha, R. Deoliya, S. Maiti, Recycled
[32] EFNARC, Specification Guidelines for Self-compacting Concrete, European
aggregate from C&D waste & its use in concrete - a breakthrough towards
Federation of National Associations Representing producers and applicators of
sustainability in construction sector: a review, Constr. Build. Mater. 68 (2014)
specialist building products for Concrete, (2002).
501–516, https://doi.org/10.1016/j.conbuildmat.2014.07.003.
[33] F. Fiol, C. Thomas, C. Muñoz, V. Ortega-López, J.M. Manso, The influence of
[6] C. Shi, Y. Li, J. Zhang, W. Li, L. Chong, Z. Xie, Performance enhancement of
recycled aggregates from precast elements on the mechanical properties of
recycled concrete aggregate - a review, J. Clean. Prod. 112 (2016) 466–472,
structural self-compacting concrete, Constr. Build. Mater. 182 (2018) 309–323,
https://doi.org/10.1016/j.jclepro.2015.08.057.
https://doi.org/10.1016/j.conbuildmat.2018.06.132.
[7] A. Gonzalez-Corominas, M. Etxeberria, Effects of using recycled concrete
[34] EHE-08, Instrucción de Hormigón Estructural. Structural Concrete Regulations,
aggregates on the shrinkage of high performance concrete, Constr. Build.
Ministerio de Fomento, Gobierno de España (2010).
Mater. 115 (2016) 32–41, https://doi.org/10.1016/j.conbuildmat.2016.04.031.
[35] EN-Euronorm, Rue de stassart, 36. Belgium-1050 Brussels, European
[8] V. Ortega-López, J.A. Fuente-Alonso, A. Santamaría, J.T. San-José, Á. Aragón,
Committee for Standardization.
Durability studies on fiber-reinforced EAF slag concrete for pavements, Constr.
[36] S.C. Kou, C.S. Poon, Properties of self-compacting concrete prepared with
Build. Mater. 163 (2018) 471–481, https://doi.org/10.1016/
coarse and fine recycled concrete aggregates, Cem. Concr. Compos. 31 (2009)
j.conbuildmat.2017.12.121.
622–627, https://doi.org/10.1016/j.cemconcomp.2009.06.005.
[9] M. Skaf, E. Pasquini, V. Revilla-Cuesta, V. Ortega-López, Performance and
[37] S.A. Santos, P.R. da Silva, J. de Brito, Mechanical performance evaluation of self-
durability of porous asphalt mixtures manufactured exclusively with electric
compacting concrete with fine and coarse recycled aggregates from the
steel slags, Materials. 12 (2019) 3306, https://doi.org/10.3390/ma12203306.
precast industry, Materials 10 (2017) 904, https://doi.org/
[10] A.S. Brand, J.R. Roesler, Steel furnace slag aggregate expansion and hardened
10.3390/ma10080904.
concrete properties, Cem. Concr. Compos. 60 (2015) 1–9, https://doi.org/
[38] ACI-318-19, Building Code Requeriments for Structural Concrete, (2019).
10.1016/j.cemconcomp.2015.04.006.
[39] EC-2, Eurocode 2: Design of concrete structures. Part 1-1: General rules and
[11] C. Pellegrino, F. Faleschini, Experimental behavior of reinforced concrete
rules for buildings, CEN (European Committee for Standardization) (2010).
beams with electric arc furnace slag as recycled aggregate, ACI Mater. J. 110
[40] E. Güneyisi, M. Gesoǧlu, Z. Algin, H. Yazici, Effect of surface treatment methods
(2013) 197–205.
on the properties of self-compacting concrete with recycled aggregates,
[12] H. Qasrawi, The use of steel slag aggregate to enhance the mechanical
Constr. Build. Mater. 64 (2014) 172–183, https://doi.org/10.1016/
properties of recycled aggregate concrete and retain the environment, Constr.
j.conbuildmat.2014.04.090.
Build. Mater. 54 (2014) 298–304, https://doi.org/10.1016/
[41] I. González-Taboada, B. González-Fonteboa, J. Eiras-López, G. Rojo-López, Tools
j.conbuildmat.2013.12.063.
for the study of self-compacting recycled concrete fresh behaviour:
[13] A.R. Khan, S. Fareed, M.S. Khan, Use of recycled concrete aggregates in
workability and rheology, J. Clean. Prod. 156 (2017) 1–18, https://doi.org/
structural concrete, Sustain. Constr. Mater. Technol. 2 (2019).
10.1016/j.jclepro.2017.04.045.
[14] B. Fronek, P. Bosela, N. Delatte, Steel slag aggregate used in portland cement
[42] CEB-FIP, Model Code 2010 (Volumes 1 and 2), (2012).
concrete, Transp. Res. Rec. (2012) 37–42, https://doi.org/10.3141/2267-06.
[43] D. Carro-López, B. González-Fonteboa, J. De Brito, F. Martínez-Abella, I.
[15] A. Santamaria, F. Faleschini, G. Giacomello, K. Brunelli, J.T. San José, C.
González-Taboada, P. Silva, Study of the rheology of self-compacting
Pellegrino, M. Pasetto, Dimensional stability of electric arc furnace slag in civil
concrete with fine recycled concrete aggregates, Constr. Build. Mater. 96
engineering applications, J. Clean. Prod. 205 (2018) 599–609, doi: 10.1016/j.
(2015) 491–501, https://doi.org/10.1016/j.conbuildmat.2015.08.091.
jclepro.2018.09.122.
[44] O. Kebaïli, M. Mouret, N. Arabia, F. Cassagnabere, Adverse effect of the mass
[16] I.Z. Yildirim, M. Prezzi, Chemical, mineralogical, and morphological properties
substitution of natural aggregates by air-dried recycled concrete aggregates on
of steel slag, Adv. Civ. Eng. 463638 (2011), https://doi.org/10.1155/2011/
the self-compacting ability of concrete: evidence and analysis through an
463638.
example, J. Clean. Prod. 87 (2015) 752–761, https://doi.org/10.1016/j.
[17] V.W.Y. Tam, M. Soomro, A.C.J. Evangelista, A review of recycled aggregate in
jclepro.2014.10.077.
concrete applications (2000–2017), Constr. Build. Mater. 172 (2018) 272–292,
[45] Z.J. Grdic, G.A. Toplicic-Curcic, I.M. Despotovic, N.S. Ristic, Properties of self-
https://doi.org/10.1016/j.conbuildmat.2018.03.240.
compacting concrete prepared with coarse recycled concrete aggregate,
[18] C. Thomas, J. Setién, J.A. Polanco, A.I. Cimentada, C. Medina, Influence of curing
Constr. Build. Mater. 24 (2010) 1129–1133, https://doi.org/10.1016/
conditions on recycled aggregate concrete, Constr. Build. Mater. 172 (2018)
j.conbuildmat.2009.12.029.
618–625, https://doi.org/10.1016/j.conbuildmat.2018.04.009.
[46] S. Manzi, C. Mazzotti, M.C. Bignozzi, Self-compacting concrete with
[19] L. Evangelista, J. De Brito, Concrete with fine recycled aggregates: A review,
recycled concrete aggregate: study of the long-term properties, Constr.
Eur. J. Environ. Civ. Eng. 18 (2014) 129–172, https://doi.org/10.1080/
Build. Mater. 157 (2017) 582–590, https://doi.org/10.1016/
19648189.2013.851038.
j.conbuildmat.2017.09.129.
[20] S. Santos, P.R. da Silva, J. de Brito, Self-compacting concrete with recycled
[47] T. Uygunoʇlu, I.B. Topçu, A.G. Çelik, Use of waste marble and recycled
aggregates – a literature review, J. Build. Eng. 22 (2019) 349–371, https://doi.
aggregates in self-compacting concrete for environmental sustainability, J.
org/10.1016/j.jobe.2019.01.001.
Clean. Prod. 84 (2014) 691–700, https://doi.org/10.1016/j.
[21] A.M. Matos, L. Maia, S. Nunes, P. Milheiro-Oliveira, Design of self-compacting
jclepro.2014.06.019.
high-performance concrete: study of mortar phase, Constr. Build. Mater. 167
[48] L.A. Pereira-De-Oliveira, M.C.S. Nepomuceno, J.P. Castro-Gomes, M.F.C. Vila,
(2018) 617–630, https://doi.org/10.1016/j.conbuildmat.2018.02.053.
Permeability properties of self-Compacting concrete with coarse recycled
[22] S. Nagataki, A. Gokce, T. Saeki, M. Hisada, Assessment of recycling process
aggregates, Constr. Build. Mater. 51 (2014) 113–120, https://doi.org/10.1016/
induced damage sensitivity of recycled concrete aggregates, Cem. Concr. Res.
j.conbuildmat.2013.10.061.
34 (2004) 965–971, https://doi.org/10.1016/j.cemconres.2003.11.008.
18 V. Revilla-Cuesta et al. / Construction and Building Materials 263 (2020) 120671

[49] R.V. Silva, J. De Brito, R.K. Dhir, The influence of the use of recycled aggregates concrete, Constr. Build. Mater. 226 (2019) 345–359, https://doi.org/10.1016/
on the compressive strength of concrete: a review, Eur. J. Environ. Civ. Eng. 19 j.conbuildmat.2019.07.311.
(2015) 825–849, https://doi.org/10.1080/19648189.2014.974831. [57] B. Wu, H. Jin, Compressive fatigue behavior of compound concrete containing
[50] P. Revathi, R.S. Selvi, S.S. Velin, Investigations on fresh and hardened properties demolished concrete lumps, Constr. Build. Mater. 210 (2019) 140–156, https://
of recycled aggregate self compacting concrete, J. Inst. Eng. Ser. A 94 (2013) doi.org/10.1016/j.conbuildmat.2019.03.188.
179–185, https://doi.org/10.1007/s40030-014-0051-5. [58] C. Zhou, Z. Chen, Mechanical properties of recycled concrete made with
[51] M. Gesoglu, E. Güneyisi, H.Ö. Öz, I. Taha, M.T. Yasemin, Failure characteristics different types of coarse aggregate, Constr. Build. Mater. 134 (2017) 497–506,
of self-compacting concretes made with recycled aggregates, Constr. Build. https://doi.org/10.1016/j.conbuildmat.2016.12.163.
Mater. 98 (2015) 334–344, https://doi.org/10.1016/ [59] F. Aslani, S. Nejadi, Mechanical characteristics of self-compacting concrete
j.conbuildmat.2015.08.036. with and without fibres, Mag. Concr. Res. 65 (2013) 608–622, https://doi.org/
[52] D.B. Lad, S.A. Faroz, S. Ghosh, Fusion of Rebound Number and Ultrasonic Pulse 10.1680/macr.12.00153.
Velocity Data for Evaluating the Concrete Strength Using Bayesian Updating, [60] Y. Chen, Z. Chen, J. Xu, E.M. Lui, B. Wu, Performance evaluation of recycled
in: P.V. Varde, R.V. Prakash, G. Vinod (Eds.), International Conference on aggregate concrete under multiaxial compression, Constr. Build. Mater. 229
Reliability, Safety and Hazard, ICRESH 2019, Springer, 2020, pp. 437–448. (2019), https://doi.org/10.1016/j.conbuildmat.2019.116935 116935.
[53] B.M. Vinay Kumar, H. Ananthan, K.V.A. Balaji, Experimental studies on [61] L.A. Pereira-de Oliveira, M. Nepomuceno, M. Rangel, An eco-friendly self-
utilization of coarse and finer fractions of recycled concrete aggregates in compacting concrete with recycled coarse aggregates, Inf. Constr. 65 (2013)
self compacting concrete mixes, J. Build. Eng. 9 (2017) 100–108, https://doi. 31–41, https://doi.org/10.3989/ic.11.138.
org/10.1016/j.jobe.2016.11.013. [62] M.K. Ismail, A.A.A. Hassan, Abrasion and impact resistance of concrete before
[54] F. Aslani, G. Ma, D.L. Yim Wan, G. Muselin, Development of high-performance and after exposure to freezing and thawing cycles, Constr. Build. Mater. 215
self-compacting concrete using waste recycled concrete aggregates and rubber (2019) 849–861, https://doi.org/10.1016/j.conbuildmat.2019.04.206.
granules, J. Clean. Prod. 182 (2018) 553–566, https://doi.org/10.1016/j. [63] D. Pedro, J. de Brito, L. Evangelista, Structural concrete with simultaneous
jclepro.2018.02.074. incorporation of fine and coarse recycled concrete aggregates: mechanical,
[55] J.A. Ortiz, A. de la Fuente, F. Mena Sebastia, I. Segura, A. Aguado, Steel-fibre- durability and long-term properties, Constr. Build. Mater. 154 (2017) 294–309,
reinforced self-compacting concrete with 100% recycled mixed aggregates https://doi.org/10.1016/j.conbuildmat.2017.07.215.
suitable for structural applications, Constr. Build. Mater. 156 (2017) 230–241, [64] G. Duarte, M. Bravo, J. de Brito, J. Nobre, Mechanical performance of shotcrete
https://doi.org/10.1016/j.conbuildmat.2017.08.188. produced with recycled coarse aggregates from concrete, Constr. Build. Mater.
[56] J. Xie, W. Chen, J. Wang, C. Fang, B. Zhang, F. Liu, Coupling effects of recycled 210 (2019) 696–708, https://doi.org/10.1016/j.conbuildmat.2019.03.156.
aggregate and GGBS/metakaolin on physicochemical properties of geopolymer