Construction and Building Materials 349 (2022) 128759

Contents lists available at ScienceDirect

Construction and Building Materials

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

Rebar corrosion and ASR durability assessment of fly ash concrete mixes
using high contents of fine recycled aggregates
Andrés Barragán-Ramos *, Camilo Ríos-Fresneda , Juan Lizarazo-Marriaga , Nicole Hernández-
Romero
Universidad Nacional de Colombia, Bogotá, Colombia

A R T I C L E I N F O A B S T R A C T

Keywords: This research assessed the potential alkali silica reaction (ASR) of fine recycled aggregates (FRA) derived from
Concrete durability precast concrete slabs using petrographic examination and expansion tests, the effects of incorporating FRA on
Recycled concrete aggregate some chloride related transport properties and the corrosion risk of fly ash blended concretes. Mixes incorpo­
Durability of concrete
rating 20 % fly ash (FA) were produced using water-to-binder ratios of 0.45 and 0.50 and FRA replacements of 0
ASR
Steel rebar corrosion
%, 20 %, 60 % and 100 % to test concrete’s mechanical performance, chloride penetrability and rebar corrosion
risk using electrochemical techniques at different ages. FRA presented an ASR expansion 227 % higher than
natural aggregates. Using high contents (>60 %) of FRA didn’t increase corrosion risk and a comparable risk to
conventional concrete was observed when incorporating FA.

1. Introduction from the parent concrete [6]. OAM consist mainly of fully or partially
hydrated cement particles with high porosity paste adhered to natural
Construction is responsible for causing considerable environmental aggregates, which causes higher water absorption, lower density and
impact due to the extraction of natural resources and the associated lower abrasion resistance than natural aggregates [7]. The presence of
pollution resulting from their fabrication ang processing activities [1]. OAM in RCA also produces higher porosity in concrete’s microstructure
Construction also generates large amounts of waste and debris due to the that reduce their durability performance [8–10].
eventual demolition of buildings that have completed their service life. Corrosion consists of electrochemical reactions at the surface be­
In the USA, 569 million tons of construction and demolition waste tween metal and an electrolyte solution. In concrete, the solid and liquid
(CDW) were generated in 2014, mainly from demolition activities, In products resulting from cement hydration form a thin layer over the
China, 30 % – 40 % of total waste consists of CDW, producing around rebar surface, known as the passive film, which protects the metal from
3000 million tons [2]. In Colombia, CDW correspond to an estimate of further corrosion. Regardless of the formation of this film, external ions
100 thousand tons per day [3]. can be transported through the concrete and in practice, corrosion is
When CDW is separated and recycled concrete aggregates (RCA) are never equals to zero but should be kept at very low levels to maintain the
used, the resulting concrete known as recycled aggregate concrete (RAC) service life of reinforced concrete [11]. The stability of metal could be
can achieve similar properties than those of conventional concrete. described in terms of its electrochemical potential (EP), which is defined
Nevertheless, little acceptance has been observed in the use of fine RCA as the ease of ionizing an atom of the metal. EP is dependent on the pH
(FRA) as the durability-related properties could not guarantee an and factors such as the oxygen and water availability. There are ranges
acceptable service life [4]. It has been reported that the replacement of of electrochemical potential and pH in which the products of corrosion
FRA increases water absorption by immersion and capillary action, and are either dissolved ions such as Fe2+, solid oxides or hydroxides. The
reduces both, carbonation and chloride migration resistances [5]; which Pourbaix diagram for iron, indicates regions of potential and PH in
are directly related to corrosion risk of the steel rebar reinforcement. The which the metals will be immune, will actively corrode or will form
negative effects of incorporating RCA in the durability of concrete could passive films. Steel bars embedded in concrete are considered to be in a
be attributed to the high amounts of old adhered mortar (OAM) derived state of passivity due to the high alkalinity of concrete [11].

Abbreviations: RCA, Recycled concrete aggregates; FRA, Fine recycled aggregates; ASR, Alkali silica reaction; FA, Fly ash; OAM, Old attached mortar.
* Corresponding author.
E-mail address: [email protected] (A. Barragán-Ramos).

https://doi.org/10.1016/j.conbuildmat.2022.128759
Received 11 April 2022; Received in revised form 9 July 2022; Accepted 8 August 2022
Available online 17 August 2022
0950-0618/© 2022 The Authors. Published by Elsevier Ltd. This is an open access article under the CC BY-NC-ND license (http://creativecommons.org/licenses/by-
nc-nd/4.0/).
A. Barragán-Ramos et al. Construction and Building Materials 349 (2022) 128759

Chloride ions can be present in concrete due to the use of chloride- both ASR in aggregates and corrosion in concrete mixes. This paper
contaminated components, the use of de-icing salts or from environ­ contributes to the knowledge on the relationship between aggregate
mental exposure in coastal areas. When the chloride ions in concrete are mineral composition, alkali-silica reactivity, concrete durability per­
present in the pore solution, they are called free chlorides and are formance, steel rebar corrosion and potential mitigation techniques by
responsible for initiating the corrosion process [12]. Studies have shown performing an extensive experimental research program. This paper also
that there is a linear relationship between the chloride penetrability and provides support for further research regarding RAC structural viability
the increasing replacement of RCA [13]. It has been reported that RAC from a durability perspective.
using 100 % FRA showed a migration coefficient increase of 33.8 %
compared to conventional concrete [14]. As the migration coefficient is 2. Experimental program
directly linked to concrete porosity, a reduction of chloride ion pene­
tration resistance is expected when FRA are used [13,15,16]. 2.1. Materials
Alkali-Aggregate reactions are detrimental chemical reactions in
concrete that produce internal expansion and premature loss of 2.1.1. Aggregates
serviceability. In very aggressive environments with high pH (pH > Coarse natural aggregates (CNA) with a maximum size of 12.7 mm
12.5), the silica in the reactive mineral phases within the aggregates and fine natural aggregates (FNA) with a maximum size of 4.76 mm
could react deleteriously with the highly basic alkali hydroxides (K+, (Sieve No.4) were obtained from a local quarry. These natural aggre­
Na+) dissolved within the micropores in the matrix of hardened concrete gates had a fluvial sedimentary origin, regular size, and partially
[17]. Studies have identified that alkali-aggregate reaction can occur in rounded edges. Fine recycled aggregates (FRA) were obtained from a
RCA when the OAM or the original natural aggregates were reactive local concrete manufacturer. FRA were derived from prefabricated
[18] and it increases for higher replacement ratios of RCA in new con­ structural concrete slabs designed to achieve a compressive strength of
crete [19]. The extent of the reaction depends on the supply of reactants, 35 MPa and mechanically crushed to achieve a maximum size of 4.76
if the reactant (unstable siliceous phases) had been depleted in the mm. The physical properties measured on both natural and recycled
previously affected concrete source (i.e., highly ASR-damaged RCA ag­ aggregates such as fineness modulus, specific gravity, absorption (ASTM
gregates), a slower “secondary” expansion is expected in new RAC C128) [30], and materials finer than 75 µm (Sieve No. 200) (ASTM
[19–21]. Due to this effect, researchers have found that RAC with C117) [31] are summarized in Table 1.
slightly damaged or highly damaged RCA may have a lower number of Regarding particle size distribution, natural aggregates showed a size
cracks than conventional concrete [21]. Different behavior has also been distribution satisfying the required gradation limits established in ASTM
observed depending on the size of the reactive recycled particles as the C33 [32]. Due to the homogeneous crushing process, the amount of FRA
reactants are consumed faster using FRA than coarse recycled aggre­ particles retained on sieves>1 mm was inferior to the recommended on
gates, showing an expansion reduction effect over time[20]. A delay in ASTM C33. Particle size distribution for both natural and recycled ag­
expansion was also observed using highly damaged FRA and attributed gregates are shown on Fig. 1.
to the residual cement paste with a lower alkali content obstructing the
diffusion of alkalis from the new cement paste [20]. 2.1.2. Cementitious materials for concrete mixes
It has been established that the use of supplementary cementitious Commercially available blended Portland cement (C) type UG – “Uso
materials (SCM) such as fly ash (FA), silica fume, ground granulated General” (General use cement) according to Colombian regulation NTC
blast furnace slag (GGBFS), or metakaolin, can prevent ASR by reducing 121 [33] was used in the preparation of all concrete mixes. Class F fly
the alkali contribution from Portland cement, inducing pozzolanic re­ ash (FA) meeting the required physical properties defined in ASTM C618
actions that produce calcium silicate hydrate (C–S–H) and lowering [34] was used as a supplementary cementitious material. The physical
the pH of the pore solution [22]. Studies have found that the use of FA is properties measured for the binder (cement and fly ash) are shown in
effective to mitigate ASR on RCA; however, a higher content may be Table 2. Both types of cementitious materials are used extensively by
required compared to conventional concrete due to the additional alkali local concrete and cement manufacturers.
content provided by the OAM [18,22]. The use of fly ash (FA) can also The chemical oxide analysis performed for FA using X-ray Fluores­
improve RAC’s durability to chloride penetration [23,24], as it can cence (XRF) is shown in Table 3.
enhance pozzolanic reactions that result in a denser concrete micro­
structure due to the pozzolanic reaction between FA, Portland cement 2.1.3. Water reducing admixture
hydrating products and OAM [10,15]. The use of FA is also notoriously To avoid possible fresh-state performance loss and achieve the target
favorable to the increase in RAC’s electrical resistivity [25]. Different slump value, a high-range water-reducing admixture super-plasticizer
studies have concluded that the principal negative effect of incorpo­ (SP) made from polycarboxylates was incorporated on all concrete
rating FA to RAC is an increase in carbonation depth, as it causes a mixes. The SP composition does not include chlorides to avoid corrosive
reduction in concrete’s pH due to the pozzolanic reaction between reactions on reinforced concrete and met the requirements of American
portlandite and FA, which reduces the amount of Ca(OH)2, creating standard ASTM C494 [35] for type A water reducing-admixtures and
favorable conditions for carbonation to occur [10,15,25–29]. type F high-range water-reducing admixtures.
While the effects of incorporating FRA in concrete’s durability have
been documented, the relationship between possibly expansive aggre­
gates due to ASR and corrosion in steel reinforcement has not been 2.2. Preparation of concrete mixes
assessed. The objective of this paper was to perform a microscopic
characterization of the FRA and evaluate their potential alkali-silica Mixes were cast using water/binder (W/B) ratios of 0.45 and 0.50,
reactivity. RAC mixes were also produced using different FRA contents
to identify its effects on concrete’s durability, most specifically the Table 1
properties related to chloride penetration resistance. Additionally, Physical properties of the aggregates.
electrochemical measurement techniques were carried out on steel bars Type of Apparent Absorption Fineness Materials finer
embedded into concrete specimens to measure linear polarization Aggregate Density Module than 75 µm
resistance (LPR) which is directly correlated to the corrosion rate of steel (g/cm3) (%) - (%)

reinforcement and it’s an indicative of the passivation of steel embedded CNA 2,32 4,62 – –
in concrete. This research also aimed to determine the effectivity of FNA 2,45 2,09 2,38 5,66
FRA 2,12 9,48 3,25 9,40
incorporating SCM (specifically class-F FA) as a mitigation technique for

2
A. Barragán-Ramos et al. Construction and Building Materials 349 (2022) 128759

mixing water was compensated to maintain the same effective W/B

ratio. SP dosage was defined in the range specified by the manufacturer
(0.15 % to 1.2 % of binder weight) and trial mixes were performed with
different dosage values until the desired slump was obtained. Each
concrete mix was performed following the Two-Stage Mixing Approach
(TSMA). This method was proposed by Tam et al. (2005), and different
studies have concluded that it may improve recycled aggregate concrete
durability and mechanical performance [36–39]. Specimens were
demolded after 24 h and cured in moist conditions until testing.

2.3. Test Procedures

2.3.1. Aggregate petrographic testing

Petrographic analysis was carried out following the American stan­
dard ASTM C295 [40]. Three thin sections were prepared for each
aggregate type. The first one was prepared selecting particles retained
on the #8 sieve, the second section was prepared using particles retained
Fig. 1. Grading curves for the aggregates including ASTM C33 limits.
on the #16 and #30 sieve, and the third section was prepared using
particles retained on the #50 and #100 sieves. The second and third
Table 2 sections were prepared using a different size on each half of the section.
Physical properties of used cement and fly ash. A “Concrete” lithology was defined for the FRA particles which included
Material Fineness Density Water Loss on Ignition
fine aggregates, cementing materials and residual clinker and hydration
Content (1000 ◦ C) products (OAM). The natural aggregates contained in these particles
ASTM ASTM ASTM D7348 ASTM D7348 also included rocks from different lithological origins. The rocks con­
C204 C188 tained in FRA are mostly from sedimentary origin, but igneous and
(cm2/g) (g/cm3) (%) (%)
metamorphic origins were also identified.
Cement 7014,25 2,82 – –
Fly Ash 2,05 0,41 10,9
–
2.3.2. Alkali-silica reaction in mortar bars
Mortar mixes using Type I ASTM C150 Portland reference cement
(OPC), FRA and FNA were formulated to assess the presence of ASR. The
Table 3 procedure and mixing proportions described in ASTM C1260 [41] and
Chemical composition of fly ash. ASTM C1567 [42] were followed to produce 25x25x285 mm mortar
SiO2 Al2O3 Fe2O3 CaO MgO SO3 Na2O bars. Three specimens were tested for each mortar mix incorporating
(%) (%) (%) (%) (%) (%) (%) different replacement ratios of FRA (0 %, 20 %, 60 %, 80 %, 100 %) and
Fly Ash 55,14 25,33 7,76 1,43 0,78 0,43 0,18 FA (0 %, 20 %). The proportions following ASTM C1260 for each mortar
mix are shown on Table 5. The longitudinal expansion of the specimens
cured in an aggressive 1 N NaOH solution at 80 ◦ C was measured during
aiming for an average design compressive strength of 21 MPa and 28 a 28-day period.
MPa as these are the most produced concrete strengths in standard
structures. Fourteen mixes were formulated using FRA replacing 2.3.3. Hardened concrete
different contents (0 %, 20 %, 60 % and 100 % by weight) of NFA. Compressive strength (7, 28 and 90 days) and modulus of elasticity
Additional six recycled concrete mixes were made replacing 20 % of (90 days) were evaluated on cylindrical specimens (100 × 200 mm)
cement weight by fly ash. The mixture proportions and designations of according to ASTM C39 [43] and ASTM C469 [44] respectively. Rapid
the different concrete mixes are summarized in Table 4. An optimal chloride penetration tests (RCPT) (28 and 90 days) were carried out on
volume of 51 % fine and 49 % coarse aggregates that best fits the sug­ cylindrical specimens (100 × 50 mm) according to ASTM C1202 [45].
gested grading was defined for all FRA contents. Specimens were also produced to perform water absorption and percent
The selected slump for all concrete mixtures was 127 mm ± 25.4 voids tests (28 and 90 days) following the procedure of ASTM C642
mm (5 ± 1 in.). The aggregates were used in air-dry state and the [46].

Table 4
Mix proportions of concrete mixes.
Mix Mix W/B* Cement Fly Ash Water Coarse Aggregates Natural Fine Aggregates Recycled Fine Aggregates Super-Plasticizer
3 3 3 3 3 3
kg/m kg/m kg/m kg/m kg/m kg/m % Binder Mass

1 28-CONTROL 0.45 500 0 225.0 667.7 694.9 0.0 0.32

2 28-RA20 0.45 500 0 225.0 658.6 548.4 137.1 0.35
3 28-RA60 0.45 500 0 225.0 639.7 266.3 399.5 0.35
4 28-RA100 0.45 500 0 225.0 619.8 0.0 645.1 0.35
5 28-RA20-FA20 0.45 400 100 225.0 647.0 538.7 134.7 0.42
6 28-RA60-FA20 0.45 400 100 225.0 628.4 261.6 392.4 0.42
7 28-RA100-FA20 0.45 400 100 225.0 608.8 0.0 633.7 0.42
8 21-CONTROL 0.5 450 0 225.0 688.5 716.6 0.0 0.18
9 21-RA20 0.5 450 0 225.0 679.2 565.5 141.4 0.18
10 21-RA60 0.5 450 0 225.0 659.7 274.7 412.0 0.20
11 21-RA100 0.5 450 0 225.0 639.1 0.0 665.2 0.20
12 21-RA20-FA20 0.5 360 90 225.0 668.7 556.8 139.2 0.25
13 21-RA60-FA20 0.5 360 90 225.0 649.6 270.4 405.6 0.25
14 21-RA100-FA20 0.5 360 90 225.0 629.3 0.0 655.0 0.25

3
A. Barragán-Ramos et al. Construction and Building Materials 349 (2022) 128759

Table 5 time and applying Faraday’s law as following:

Mortar mix proportion for ASR mortar-bar method testing. ( )
icorr *ta *MFe μm
Mix Mix Fine Natural Fine Recycled Ordinary Fly CR = *106 (5)
ZFe *F*dFe year
ID Name Aggregates Aggregates Portland Ash
Cement Where ta is the number of seconds in a year (31536000 s/year), MFe is
g g g g
the molar mass of iron (0.055847 kg/mol), ZFe is the number of loss
1 100-NA 990 0 440 0 electrons (2 for iron), F is Faraday’s constant (96485C/mol), and dFe is
2 100-RA 0 990 440 0
the density value for iron (7860 kg/m3). Therefore, the previous equa­
3 20-RA- 792 198 352 88
20-FA tion can be rewritten as:
4 60-RA- 594 594 352 88 ( ) ( )
μm A
20-FA CR = icorr *1161 (6)
5 80-RA- 198 792 352 88 year cm2
20-FA
The main advantage of using LPR is that it is a non-destructive
method and requires only a few minutes to obtain the corrosion rate;
2.3.4. Electrochemical measurement of corrosion however, since the actual corroding area of steel is hard to determine, an
Corrosion consists of an anodic reaction in which a metal is oxidized underestimation of the actual corrosion current density is usual [11]. To
end electrons are released, and a cathodic reaction in which the elec­ perform a better estimation of the corroding area, smooth (non-corru­
trons are consumed or reduced. By equating these two reactions, a gated) steel bars were used. The bars were also painted with epoxy paint
corrosion current ICorr, which is a direct measurement of the rate of to control the exposed area in concrete in which electrode reduction
dissolution or oxidation of the metal, and a corrosion potential or open occurs.
circuit potential, ECorr, can be measured using specialized equipment Linear Polarization resistance (LPR) measurements were performed
[11]. The Linear Polarization Resistance (LPR) is the resistance of a using a potentiostat as described in ASTM G59 [49] on 6 mm steel bars
specimen to oxidation while an external potential is applied: embedded on Ø3-inch cylindrical specimens from the same concrete
mixes submerged in a 3 % NaCl solution as shown in Fig. 2 (a). Initial
ΔE
Rp = (1) LPR was measured for three specimens of each concrete mix after 28
ΔI
days of moist curing conditions in 20 ◦ C tap water. Following the initial
Where ΔE is the change in electrical potential and ΔI is the change in LPR tests, specimens were submerged in a 3 % NaCl solution and tested
electrical current. The corrosion current and the corrosion current after 1, 2 and 3 months of chloride solution immersion. A three-point
density (iCorr) can be calculated using the mixed-theory potential electrode system shown in Fig. 2 (b) was used for LPR measurements.
developed by Stern and Geary, using the following equations: The working electrode was defined as the specimen under testing, in this
B case the embedded steel bars. The counter electrode was a graphite bar
Icorr = (2) that completed the current path. The used reference electrode was a
Rp
saturated calomel electrode (SCE). Experiments were performed at a
B voltage scan rate of 0.36 V/h (0.1 mV/s) from a potential of − 30 mV to
icorr = (3) 30 mV. Since LPR tests measure the total specimen electrical resistance
Rp A
(concrete plus steel bar) and corrosion density calculation requires the
βa βc resistance value of only the embedded steel, Electrochemical Impedance
B= (4) Spectroscopy (EIS) tests were also performed to measure the medium
2.3(βa + βc )
(concrete) electrical resistance.
A is the surface area of the polarized zone, B is Stern-Geary constant
and βa and βc are anodic and cathodic Tafel constants, respectively. The 3. Results and discussion
Stern-Geary constant should be determined empirically, however, for
corrosion of steel rebar in concrete, B is assumed to be equal to 0.026 V 3.1. Aggregate petrographic testing
for active corrosion and 0.052 V for passive corrosion [11]. The elec­
trical resistance measured by the LPR is the sum of the polarization The content of each predominant lithology for the employed aggre­
resistance and the electrolyte resistance RΩ . For environments such as gates are summarized on Table 7. It is possible to identify that the main
concrete, RΩ is significant in magnitude to the total polarization resis­ difference between natural and recycled aggregates is the presence of a
tance and should be considered and subtracted in the measurements “concrete” aggregate type which contains aggregates with old-attached
[47]. According to some researchers, corrosion current densities over 1 mortar (OAM) and cement or hydration products in smaller particle
µA/cm2 are identified as the level of high corrosion risk and corrosion sizes. Sedimentary rocks without OAM such as chert were also identified
current densities below are defined as passive corrosion in the system for the larger FRA sizes. For the aggregates retained on the #100 sieve,
[11]. A corrosion risk classification based on the measured corrosion the “concrete” aggregate type was no longer identified and both FRA and
current density has been formulated [48] as shown on Table 6. FNA had a similar composition with a high presence of feldspar and
The corrosion current density can be related to the corrosion rate plagioclase. A high content of straight and high strained quartz was
(CR) in µm/year by determining the mass loss during the polarization observed for both types of aggregates due to the predominant sedi­
mentary origin of the aggregates.
Table 6 The mineral origin of the FNA was more heterogeneous, having a
Corrosion risk according to corrosion current density. Retrieved significant content of metamorphic and igneous aggregates. Since the
from López, Pérez et al. (2006) [48]. natural aggregates had a fluvial origin, particles are usually transported
Current Density –icorr Risk of Corrosion many kilometers before being deposited, which results in a wide variety
of minerals with diverse origins. The larger sizes of FRA also showed
2
(µA/cm ) combinations of sedimentary, igneous, and metamorphic aggregates
<0.1 Low bonded to hydration products from the parent concrete as shown in
0.1–0.5 Moderate Fig. 3. Both types of aggregates had low porosity and non-homogeneous
0.5–1 High
shapes and minor levels of carbonation were observed on FRA. For FRA
>1 Very High

4
A. Barragán-Ramos et al. Construction and Building Materials 349 (2022) 128759

(a) (b)
Fig. 2. (a) Cylindrical specimens with embedded steel bar. (b) Three-point electrode system used for PR tests.

FNA). FRA showed approximately 10 % higher content of potentially

Table 7
reactive minerals. In average, a potentially reactive component was
Predominant lithologies for each aggregate size.
identified for 45.7 % of all FRA particles and 36.8 % of all FNA particles.
Sieve Fine Recycled Fine Natural Coarse Natural Based only on the mineral composition of the aggregates (excluding
Size Aggregates Aggregates Aggregates
the possible reactive effects of OAM), for this research program the
Lithologies % Lithologies % Lithologies % concrete mixes made with both types of aggregates could be similarly
#8 Concrete 64.9 Chert 20.7 Quartz 66.7 reactive depending on the cementing material used. Apart from the
(2.38 Sandstone hydration products on the “concrete” lithology from FRA, no significant
mm) Chert 20.5 Diorite 16.7 Silicate 12.0 difference was observed on the mineral composition of the aggregates. It
Mudstone
is important to notice that the petrographic analysis was used to identify
Silicate 6.6 Quartz 8.7 Chert 10.7
Siltstone Sandstone the lithology, origin and reactive minerals of the natural aggregates and
- - Silicate 9.3 - - the original aggregate portion of FRA (excluding OAM). For FRA, it was
Siltstone possible to assess the existence of hydration products from a “concrete”
#16 Concrete 61.2 Chert 17.3 lithology but due to the chemical composition of this cementitious
(1.19 Chert 11.2 Quartz 11.3
mm) Sandstone
matrix, the potentially reactive alkaline ions cannot be identified using
Silicate 8.6 Plagioclase 11.3 petrographic analysis and further assessment using ASTM C1260 [41]
Siltstone and ASTM C1567 [42] was formulated.
- - Quartz Shale 9.3
#30 Concrete 67.1 Straight 22.0
(595 Quartz 3.2. Alkali-silica reaction of mortar bars
µm) Chert 10.7 High strained 13.3
Quartz
According to ASTM standards, mortar bars cast using potentially
Quartz 5.3 Plagioclase 16.0
#50 Concrete 43.3 Straight 19.3 expansive aggregates will have an expansion>0.10 % after 16 days of a
(297 Quartz hot and alkaline curing. The obtained bar expansion for each mix is
µm) Straight 10.7 High strained 16.0 plotted against the specimen age in Fig. 4. Results showed that the used
Quartz Quartz
FRA are potentially expansive, having an expansion of approximately
Feldspar 12.7 Feldspar 15.3
#100 Straight 48.7 Straight 16.0
0.25 % at 16 days. The used NFA showed an expansion of approximately
(149 Quartz Quartz 0.11 % at 16 days, staying slightly above the defined limit. These results
µm) Plagioclase 17.3 Plagioclase 20.7 are consistent with other researchers who have reported that alkali-
Feldspar 10.7 Feldspar 15.3 aggregate reaction can occur when the OAM or the original natural
Hornblende 14.7
aggregates were reactive [18] and increases for higher replacement ra­
tios of FRA [19].
particles, no signs of previous ASR related expansion or cracking was Petrographic analysis revealed a very similar mineral composition
identified. between FNA and FRA, this would indicate that the high alkalinity
Regarding alkali-silica reaction (ASR), multiple potentially alkali- content of the OAM is responsible for the observed expansion. Previous
silica reactive lithologies such as chert, quartzites and phyllites were studies have also reported that FRA derived from previously ASR-
identified. The main identified reactive components of these lithologies damaged concrete showed lower expansion than FRA derived from
are summarized for each particle size in Table 8. For both types of fine non-damaged parent concrete as the alkali supply is higher [19–21].
aggregates, the main reactive mineral in the particles was high strained This is consistent with the obtained results as the petrographic exami­
quartz observed in metamorphic aggregates (29.6 % in FRA and 21 % in nation revealed no previous cracking or expansion in the OAM or ag­
FNA) followed by microcrystalline quartz which was present in the gregates related to ASR, suggesting that the OAM of the employed FRA
sedimentary rocks like chert and mudstone (16.1 % in FRA and 15.5 % in supplied a high number of free alkalis that had not been consumed
previously. It has also been reported that the reactants are consumed

5
A. Barragán-Ramos et al. Construction and Building Materials 349 (2022) 128759

Fig. 3. PPL - XPL Microphotography of RCA with Quartz Sandstone (Qa), cement with sedimentary and metamorphic aggregates (S + M), Cement with Sedimentary
and Igneous aggregates (S + I). and Cement with Sedimentary aggregates (S).

Table 8
Potentially reactive components identified in FNA and FRA.
Reactive components content (%) Fine Recycled Aggregates Fine Natural Aggregates
Sieve Size #8 #16 #30 #50 #100 Avg #8 #16 #30 #50 #100 Avg.

Mycrocristalline Quartz 27.3 21.3 16.0 13.3 2.7 16.1 32.7 24.7 8.7 8.0 3.4 15.5
Highly Strained Quartz 48.7 37.3 38.7 16.0 7.4 29.6 26.0 20.7 24.0 18.0 16.1 21.0
Total 76.0 58.7 54.7 29.3 10.1 45.7 58.7 45.4 32.7 26.0 19.5 36.5

Fig. 4. Expansion of mortar bars for different mortar mixes.

faster using FRA, showing an expansion reduction effect over time [20]; effective to detect potential ASR reactivity since the alkali content of
however, this reduction in the expansion rate was not observed and the OAM cannot be identified only by petrographic/microscopic
mortar mix containing 100 % FRA showed the same expansion rate even examination.
after 28 days of curing in a very aggressive environment. This may be Results from the mortar bars including fly ash, revealed an innocuous
attributed to the initial alkali content of the aggregates and the achieved expansive behavior even when employing 80 % replacement of FRA.
particle suspension of fine aggregates inside the cement matrix, which This would indicate that FA is a very effective pozzolan for reducing
may allow a higher reactant interchange. Results indicate that the ASR expansive negative reactions when using FRA. This has been observed
observed on mixes with FRA does not depend only on the mineral by other researchers who claimed that fly ash reduces the alkalinity of
composition of the aggregates, but it is also highly affected by the alkali recycled aggregates and neutralizes possible chemical reactions that
supply of the OAM. Mortar bar expansion methods may also be more result in bar expansion by reducing the pH of the fluids saturating the

6
A. Barragán-Ramos et al. Construction and Building Materials 349 (2022) 128759

mortar bars [18]. Results indicate that the use of high contents of respectively than each control mix and 6 % and 9 % higher respectively
reactive FRA may also be possible when employing supplementary than mixes 28-RA100 and 21-RA100. These results are not consistent
cementitious materials such as FA. The effectivity of FA may depend on with the reported by other researchers, who have stated that fly ash
the initial alkali content of the OAM or aggregates and the employed reduced WA and porosity due to the filling of pores, pozzolanic reaction,
content of FA. and its positive effect in density [24,28]; however, it is important to
notice that the employed FA presented a high loss on ignition value
(10.9 %) which indicates the presence of residual unburned coal and
3.3. Percent voids and water absorption by immersion organic materials.

Results showed an overall reduction in porosity and water absorp­

tion by immersion (WA) in concrete mixes with longer curing duration 3.4. Concrete compressive strength
and an increase in concrete porosity and WA for higher contents of FRA
as shown in Fig. 5 and Fig. 6. Concrete compressive strength (CCS) was negatively affected by the
At 90 days, concrete mixes 21-RA100 and 28-RA100 showed a inclusion of FRA as shown in Fig. 7. At 28 days, 0.45 W/B mixes with 20
porosity increase of 12 % and 25 % respectively. For both groups, %, 60 % and 100 % FRA had a CCS reduction of 11 %, 15 % and 16 %
replacing 20 % of FRA increased porosity by less than 2 % at 90 days. when compared to the control mix while at 90 days these reduction
Mixes using 20 %, 60 % and 100 % FRA replacement showed a WA values were 4 %, 8 % and 13 % respectively. This is consistent with the
increase of 6 %, 18 % and 34 % respectively for 0.45 W/B mixes and 4 %, results reported by multiple researchers who also have described a loss
17 % and 24 % respectively for 0.50 W/B mixes. These results are of CCS when including only FRA in concrete [54,55]. The reduction of
consistent with different researchers who reported an increase in CCS when incorporating FRA has been widely documented in technical
porosity and WA by immersion for higher contents of FRA [14,50] which literature and is mainly attributed to the increase in concrete porosity
is often attributed to the lower particle density of FRA. In this research, due to OAM and a weaker aggregate adherence. As the microstructure of
FRA particles had a higher water absorption due to the “concrete” li­ recycled concrete has three interfacial transition zones (ITZ): (1) ITZ
thology identified in the petrographic examination. The presence of between natural aggregate and old cement matrix (Old ITZ). (2) ITZ
high-absorption residual OAM bonded to the aggregate particles, may between natural aggregate and new cement matrix (New ITZ) and (3)
lead to a higher water requirement in the mixture, hence increasing the ITZ between the new and the old cement matrices (New ITZ) [53,56], a
resulting concrete’s porosity [14,28,51]. Another possible cause for the weaker aggregate-matrix interface bond strength is expected from the
high absorption levels is the presence of micro-cracks in the concrete interaction of these ITZ than conventional concrete. Petrographic ex­
structure, which tend to create a path for water flow [52,53]. As FRA amination also showed a significant content of the “concrete” lithology
was identified as potentially reactive, these cracks may have been the in FRA and particles composed entirely of residual mortar without any
result of possible expansion related to ASR, particularly in mixes without aggregate. As the FRA replacement ratio increases, these particles are
FA. expected to have a much lower adherence and internal strength than
WA increased 14.38 % for mix 28-RA100 and 13.53 % for 21-RA100. natural aggregates.
These results are similar to the obtained by another researcher where the CCS for 0.50 W/B mixes was not reduced significantly and mix 21-
measured WA for concrete specimens using 100 % FRA incorporation RA20 showed 10 % higher CCS at 28 days and 21 % higher CCS at 90
and a W/B ratio of 0.48 was 16.5 % [14]. At 90 days water absorption days. A possible increase of strength by incorporating low values of FRA
was higher for 0.45 W/B mixes. This may be the result of the hydration with higher W/B ratios has also been described in other works, and
of the old-attached mortar in FRA at early ages due to free water content attributed to an improved microstructure of the Interfacial Transition
in 0.50 W/B mixes. Fly ash incorporation increased the porosity of Zone with increased bond strength between the new cement paste and
concrete for all tested replacement ratios of FRA. This increase is more particles [57–59]. Old mortar particles could hydrate and contribute to
evident for concrete mixes with 0.50 W/B ratio since at 90 days, the the binding strength of the microstructure for higher W/B ratios as has
porosity value of mix 21-RA100-FA20 was 24 % higher than the control been described by other researchers [58].
mix and 12 % higher than mix 21-RA100. Fly ash also increased WA by FA had a negative effect in CCS for all concrete mixes, specially at
immersion for both groups of RAC mixes. The WA values for mixes 28- younger specimen ages. At 7 days mixes 28-RA100-FA20 and 21-RA100-
RA100-FA20 and 21-RA100-FA20 were 40 % and 35 % higher FA20 showed a reduction of 44 % and 27 % respectively when compared

Fig. 5. Voids percentage (Porosity) results for incrementing values of FRA.

7
A. Barragán-Ramos et al. Construction and Building Materials 349 (2022) 128759

Fig. 6. Water absorption by immersion results for incrementing values of FRA.

Fig. 7. Compressive strength (F’c) for increasing replacement ratios of fine RCA at different testing ages.

(b)
(a)
Fig. 8. Strength development curves for increasing ratios of fine RCA of (a) 0.45 W/B mixes. (b) 0.50 W/B mixes.

8
A. Barragán-Ramos et al. Construction and Building Materials 349 (2022) 128759

to control mixes. At 90 days, the negative effect of including FA was reduced as the replacement ratio of FRA increased. Based on these re­
reduced. FA could reduce the water content in early hydration and cause sults, for low replacement ratios up to 20 %, the modulus of elasticity
a delayed pozzolanic reaction that occurs between the FA and Ca(OH)2 could be calculated with the same coefficient of estimation k than nat­
in RCA, which absorbs CaO and produces CSH, contributing to CCS ural aggregate concrete without expecting a significant overestimation
development [28,60]. Additionally, the high content of unburned of this parameter. For medium or high replacement ratios, a more
organic materials measured on the FA may also contribute to a loss in conservative factor could be formulated but this would result in larger
mechanical performance. The CCS development curve for mixes without structural elements in concrete buildings. Nevertheless, since only one
FA is shown in Fig. 8. At 28 days, only mix 28-CONTROL had achieved type of recycled aggregate was tested, it was not possible to establish
the CCS design value in the 0.45 W/B group, however all mixes with FRA how the calculated coefficient is related to the aggregate type or the
surpassed the CCS design value at 90 days. At 28 days, all 0.50 WB mixes source concrete.
without FA had achieved the design F’c of 21 MPa. This may indicate
that FRA could be used in low replacement ratios (up to 20 %) without
significantly affecting CCS. 3.6. Rapid chloride migration test (RCPT)

Results from RCPT tests at 28 and 90 days and the classification

3.5. Modulus of elasticity limits defined in ASTM C1202 are shown in Fig. 11. Compared to the
control mixes, at 90 days the incorporation of 20 %, 60 % and 100 %
Modulus of elasticity (Ec) results at 90 days are shown in Fig. 9. For FRA increased the charge passed by 10 %, 15 % and 57 % for 0.45 W/B
both mix groups, the incorporation of FRA up to 60 % did not have a mixes respectively and 5 %, 10 % and 12 % for 0.50 W/B respectively.
significantly negative effect and even increased Ec for the 0.45 W/B These results are consistent with other research in which the increase of
group. When compared to 28-CONTROL, Ec for mixes 28-RA20 and 28- FRA resulted in higher charge passed which is associated with a higher
RA60 increased by 8 % and 2 % respectively and reduced 11 % for 28- chloride penetration risk [62–64]. As FRA is more porous than natural
RA100. On the other hand, compared to 21-CONTROL, 0.50 W/B mixes aggregates, a reduction of chloride ion penetration resistance has been
presented reductions of 2 %, 6 % and 17 % when incorporating 20 %, 60 observed by different studies and may be attributed to the increase in
% and 100 % FRA, respectively. A decreasing relation between the porosity and capillarity in the microstructure of concrete [13]. The
content of FRA and the Ec of concrete samples has been observed by porosity and water absorption increased for mixes with higher FRA
other researchers [55,61] and attributed to the fact that the inclusion of content, this may induce the presence of micro-cracks and inter­
FRA reduces the stiffness of both the aggregates and the mortar. Since connecting capillary pores that could increase chloride permeability. It
concrete’s modulus of elasticity is dependent on the modulus of elas­ is important to note that at 90 days, a “Moderate” classification was
ticity of the aggregate, a reduction may also be expected due to the FRA obtained for most mixes with values between 2000 and 4000 Coulombs.
microstructure. Additionally, the presence of the “concrete” lithology The negative effect was more notorious for 0.45 W/B mixes probably
and OAM particles without aggregate may result in higher strain levels due to the possible enhanced hydration process of OAM in the 0.50 W/B
in the concrete microstructure as the particle adherence and bond mixes which was noted in physical testing.
strength in the new ITZs is lower than conventional concrete. As con­ The use of FA reduced the chloride penetration risk, effectively being
crete mixes with a lower modulus of elasticity are more susceptible to a mitigator for the detrimental effects of incorporating FRA. Mix 28-
cracking, the observed lower modulus of elasticity may also have an RA100-FA20 had a reduction of 55 % and mix 21-RA100-FA20 pre­
impact in the durability of RCA structures. sented a reduction of 22 % when compared to their respective corre­
The factor for modulus of elasticity estimation, k, for the relation spondent mixes without FA and achieved similar values than mixes 28-
( ′ )0.5
Ec = k F c was also obtained for each concrete mix. This factor is CONTROL and 21-CONTROL. Multiple researchers had already noted
widely used in structural design practice. The obtained results are shown the positive effect of incorporating FA in concrete’s conductivity and, in
on Fig. 10. From these results it is possible to identify that k is similar to consequence, the reduction of chloride ion migration [28,64]. This has
the suggested value for sedimentary aggregates in Colombia’s building been attributed to the positive pozzolanic reaction between FA and the
code (3.600), however, when compared to the value suggested by the Ca(OH)2 in FRA and the reduction in electrical conductivity of hardened
American Concrete Institute for normal weight concrete (4.700), all of concrete. Since higher porosity and water absorption was obtained for
the mixes from both mix groups showed a lower k value. Factor k was mixes including FA, the positive effect of FA in reducing electrical

Fig. 9. Modulus of Elasticity (Ec) for increasing replacement ratios of fine RCA at 90 days.

9
A. Barragán-Ramos et al. Construction and Building Materials 349 (2022) 128759

Fig. 10. Factor for Ec estimation “k” for increasing replacement ratios of FRA at 90 days.

Fig. 11. RCPT results and ASTM C1202 limits for increasing contents of FRA.

Fig. 12. Corrosion current density of steel bars embedded in concrete specimens with increasing contents of FRA.

10
A. Barragán-Ramos et al. Construction and Building Materials 349 (2022) 128759

conductance of concrete would be more dependent on the pozzolanic negative effect should be observed in this testing period scale unless
reactions of specimens with FA rather than on its filler properties alone. corrosive reactions are triggered.
Results also indicate that for all concrete mixes, the corrosion rate of
steel bars showed values between 0.7 µm/y and 4.4 µm/y with an
3.7. Potentiodynamic polarization resistance
average value of 2.3 µm/y except for one atypically high measurement
at 120 days for mix 28-CONTROL. From these results it could be noted
The measured corrosion current density and the corrosion risk clas­
that the corrosion rate of steel bars was not modified by using FRA for
sification limits are shown graphically in Fig. 12. For most concrete
the measured time, and additionally, since no clear negative tendency
mixes, a “moderate” corrosion risk was observed at all testing ages. After
could be observed and all mixes had a “moderate” risk classification, it
immersion in the 3 % NaCl solution, a clear negative effect of incorpo­
could not be concluded that incorporating FRA would accelerate
rating incrementing levels of FRA could not be determined. For the 0.45
corrosion in the embedded steel bars when comparing to conventional
W/B ratio, at 28 days, steel embedded in RAC mixes showed higher
concrete, even for corrosive environments with high chloride contents.
corrosion current density, but as the age increased, the difference with
Measurements were also made to estimate the corrosion electric
the control mix was less notorious. For the 0.50 W/B ratio, steel
potential (ECORR), also called the open circuit potential, which corre­
embedded in mix 21-CONTROL presented the lower corrosion current
sponds to the potential measured when no (net) current is flowing in the
density at 120 days, however a detrimental corrosion effect couldn’t be
absence of electrical connections as shown in Fig. 14. This potential is
determined from the incorporation of FRA. No reference could be found
measured against the reference calomel electrode, and the results are
in previous research works incorporating FRA; however, it has been
negative since a relative excess of electrons is present in the embedded
observed that reducing the W/B ratio will reduce corrosion current
steel bar due to the oxidation reactions in the interface. Higher electrical
density as higher resistivity in concrete mixes will increase polarization
potential values are an indicator of higher rates of corrosion due to an
resistance.
increase in electron exchange.
The effects of using FA varied depending on the age of the specimen.
Similar values of corrosion potential can be observed for steel bars
While at 28 days steel bars embedded in all mixes from the 0.45 W/B
embedded in 0.45 and 0.50 W/B ratios, obtaining absolute results be­
group with FA presented a reduction in corrosion current, this behavior
tween 0.2 V and 0.425 V. No clear relationship could be established
was less evident for older specimens submerged in the NaCl solution.
between the replacement ratio of FRA and the corrosion potential of
Different studies of corrosion in steel bars embedded in FA concrete have
steel bars as the corrosion electrical potential did not increase with
noted that this mineral addition could reduce corrosion rate as the
specimen age after 3 months of curing in an aggressive environment.
modification of pH inside concrete provides passive protection to the
Since corrosion is a slow process, longer measurements should be made
steel bar and reduces chloride penetration [16,23], however, in this
after years of exposure to very aggressive environments. However, as no
research program, these positive effects were different depending on the
significant differences were noticed on this test or the other chloride-
W/B ratio, the age of the specimen and the FRA replacement ratio.
penetration related tests that would reject the use of FRA in concrete
The corrosion rate of the embedded steel bars for different concrete
from corrosion related problems, the assessed aggregate cannot be
specimen ages was also calculated from the current density. Results are
stablished as a corrosive material that enhances this reaction in rein­
shown in Fig. 13. The corrosion rate was calculated in micrometers per
forcement steel bars. Furthermore, as observations from the petro­
year (µm/y) and can be interpreted as the steel bar radius that would be
graphic analysis and RAS assessments indicated that the FRA are
affected each year by corrosion. Results show that corrosion rate of steel
possibly expansive due to their high alkalinity, it is expected that the
bars is not a directly time-dependent property. This is expected as it
inclusion of FRA increases the pH of concrete due to the high alkali
corresponds to a steel property rather than a concrete property and no

(a) (b)
Fig. 13. Corrosion rate results for steel bars embedded in concrete specimens for (a). 0.45 W/B mixes and (b). 0.50 W/B mixes.

11
A. Barragán-Ramos et al. Construction and Building Materials 349 (2022) 128759

(a) (b)
Fig. 14. Corrosion Potential measurements at different specimen ages for (a) 0.45 W/B mixes. (b). 0.50 W/B mixes.

supply provided by OAM. Assuming pH values for the concrete mixes with an average lifespan of 100 years should not have steel corrosion
around 12, results from the corrosion potential measurements would levels that would represent a risk in the structural integrity when
indicate that the steel bars are in a state of passivity according to the adequate cover is provided.
Pourbaix diagram for iron as shown in Fig. 15.
These values were obtained for 6 mm diameter steel bars imbedded 4. Conclusions
in 76 mm diameter concrete cylinders, representing an approximate
concrete cover of 35 mm which is a lower value than the specified for This research work aimed to assess the effects of incorporating in­
aggressive environments in most building codes. Since the measure­ cremental levels of fine recycled concrete aggregates (FRA) in the
ments at 120 days were made after 3 months of immersing the speci­ durability properties of concrete with different water/binder ratio (W/
mens in an aggressive 3 % NaCl solution, RAC used in regular buildings B) related to steel rebar corrosion and the possible enhancing effects of

Fig. 15. Schematic Pourbaix Diagram for Iron including the measured potential region for the steel rebar.

12
A. Barragán-Ramos et al. Construction and Building Materials 349 (2022) 128759

using fly ash (FA). The main conclusions obtained from the mechanical the work reported in this paper.
and durability assessments are the following:
Data availability
• The assessed FRA presented higher absorption and higher voids in
aggregates than natural aggregates. This aggregate loss in perfor­ Data will be made available on request.
mance is the responsible for the observed reduction in concrete’s
fresh state and hardened state properties. References
• The employed FRA is potentially expansive from Alkali-Silica re­
actions (ASR), petrographic analysis revealed no significant differ­ [1] B. Estanqueiro, J. Dinis Silvestre, J. de Brito, M. Duarte Pinheiro, Environmental
life cycle assessment of coarse natural and recycled aggregates for concrete, Eur. J.
ence in the mineral part of both natural and recycled aggregates. This
Environ. Civ. Eng. 22 (2018) 429–449, https://doi.org/10.1080/
would indicate that the old-attached mortar (OAM) is responsible for 19648189.2016.1197161.
enhancing ASR reactions due to its high alkalinity, especially when it [2] J. Xiao, Recycled Aggregate Concrete Structures (2018), https://doi.org/10.1007/
is derived from parent concrete without previous ASR damage. The 978-3-662-53987-3_16.
[3] S. Suárez-Silgado, J.D.A. Molina, L. Mahecha, L. Calderón, Diagnóstico y
use of FA was an effective mitigator against this detrimental reaction propuestas para la gestión de los residuos de construcción y demolición en la
in all scenarios. The structural use of FRA for structural concrete ciudad de Ibagué (Colombia), Gestión y Ambient. 21 (2018) 9–21. 10.15446/ga.
should require Alkali-aggregate testing similar and mitigation v21n1.69637.
[4] J. De Brito, F. Agrela, R.V. Silva, Construction and Demolition Waste, Elsevier Ltd
assessment using supplementary cementitious materials. The use of a (2019), https://doi.org/10.1016/B978-0-08-102480-5.00001-4.
reactive testing method such as the mortar-bar method would be [5] F. Cartuxo, J. De Brito, L. Evangelista, J.R. Jiménez, E.F. Ledesma, Increased
preferable since the alkali content of OAM cannot be assessed using durability of concrete made with fine recycled concrete aggregates using
superplasticizers, Materials (Basel). 9 (2016), https://doi.org/10.3390/
petrographic/microscopic methods. ma9020098.
• Both mechanical and durability testing revealed a better relative [6] B.J. Zhan, D.X. Xuan, W. Zeng, C.S. Poon, Carbonation treatment of recycled
performance of incorporating FRA for the 0.50 W/B ratio. At early concrete aggregate: Effect on transport properties and steel corrosion of recycled
aggregate concrete, Cem. Concr. Compos. 104 (2019), 103360, https://doi.org/
ages, the higher free water content balanced the absorption rate of
10.1016/j.cemconcomp.2019.103360.
FRA and probably contributed to the hydration of old-attached [7] K. Mcneil, T.H. Kang, Recycled Concrete Aggregates : A Review 7 (2013) 61–69,
mortar and reduction of the effective W/B ratio in the mix, result­ https://doi.org/10.1007/s40069-013-0032-5.
[8] C. Shi, Y. Li, J. Zhang, W. Li, L. Chong, Z. Xie, Performance enhancement of
ing in a lesser difference between compressive strength, modulus of
recycled concrete aggregate - A review, J. Clean. Prod. 112 (2016) 466–472,
elasticity, porosity and water absorption of mixes with 60 % and 100 https://doi.org/10.1016/j.jclepro.2015.08.057.
% FRA incorporation than the 0.45 W/B group. The inclusion of 20 % [9] L. Evangelista, J. De Brito, Concrete with fine recycled aggregates: A review, Eur. J.
FRA had a positive effect in most tested physical and mechanical Environ. Civ. Eng. 18 (2014) 129–172, https://doi.org/10.1080/
19648189.2013.851038.
properties. [10] K.P. Verian, W. Ashraf, Y. Cao, Properties of recycled concrete aggregate and their
• Testing revealed that FRA can be incorporated in concrete for low or influence in new concrete production, Resour. Conserv. Recycl. 133 (2018) 30–49,
medium replacement ratios without significantly reducing the me­ https://doi.org/10.1016/j.resconrec.2018.02.005.
[11] A. Poursaee, Corrosion of Steel in Concrete Structures, 1st ed., Elsevier, 2016.
chanical properties, and the desired compressive strength can be 10.1016/B978-1-78242-381-2.00002-X.
achieved at long-term with an adequate mix design. The incorpora­ [12] B. Martín-Pérez, H. Zibara, R.D. Hooton, M.D.A. Thomas, Study of the effect of
tion of FA had a negative effect at early ages, however, at 90 days the chloride binding on service life predictions, Cem. Concr. Res. 30 (2000)
1215–1223, https://doi.org/10.1016/S0008-8846(00)00339-2.
difference was lower, and some mixes achieved the design strength. [13] M. Bravo, J. De Brito, J. Pontes, L. Evangelista, J. De Brito, J. Pontes,
• Chloride penetration tests revealed a detrimental effect of incorpo­ L. Evangelista, Durability performance of concrete with recycled aggregates from
rating FRA in concrete, presenting higher concrete conductivity. construction and demolition waste plants, Constr. Build. Mater. 77 (2015)
357–369, https://doi.org/10.1016/j.conbuildmat.2014.12.103.
However, the use of 20 % FA proved to be a highly effective mitigator
[14] L. Evangelista, J. de Brito, Durability performance of concrete made with fine
in reducing chloride permeability, obtaining at 90 days a “moderate” recycled concrete aggregates, Cem. Concr. Compos. 32 (2010) 9–14, https://doi.
risk classification according to the ASTM C1202 for mixes made with org/10.1016/j.cemconcomp.2009.09.005.
[15] S.C. Kou, Influence of Fly Ash as Cement Replacement on the Properties of
even 100 % FRA.
Recycled Aggregate Concrete, ASCE Libr. (2007).
• Electrochemical corrosion measurements revealed that the use of [16] S.-C. Kou, C.-S. Poon, Long-term mechanical and durability properties of recycled
FRA is not a corrosion trigger by itself. As the OAM in recycled aggregate concrete prepared with the incorporation of fly ash, Cem. Concr.
particles has a high alkalinity, steel rebar in concrete remained in a Compos. 37 (2013) 12–19, https://doi.org/10.1016/j.cemconcomp.2012.12.011.
[17] B. Fournier, M.-A. Bérubé, Alkali-aggregate reaction in concrete: a review of basic
state of passivation. At early and later ages, all concrete mixes with concepts and engineering implications, Can. J. Civ. Eng. 27 (2011) 167–191,
both natural and recycled aggregates presented a “moderate” risk of https://doi.org/10.1139/cjce-27-2-167.
corrosion even when exposed to a corrosive environment with high [18] M.B. Santos, J. De Brito, A.S. Silva, A review on alkali-silica reaction evolution in
recycled aggregate concrete, Materials (Basel). 13 (2020), https://doi.org/
chloride presence. Corrosion testing also revealed that concrete 10.3390/ma13112625.
made with FRA may be used for a service lifespan of>100 years [19] Y. Zhu, A. Zahedi, L.F.M. Sanchez, B. Fournier, S. Beauchemin, Overall assessment
without risk of corrosion for low exposure categories to chlorides. of alkali-silica reaction affected recycled concrete aggregate mixtures derived from
construction and demolition waste, Cem. Concr. Res. 142 (2021), 106350, https://
doi.org/10.1016/j.cemconres.2020.106350.
CRediT authorship contribution statement [20] C. Trottier, R. Ziapour, A. Zahedi, L. Sanchez, F. Locati, Microscopic
characterization of alkali-silica reaction (ASR) affected recycled concrete mixtures
induced by reactive coarse and fine aggregates, Cem. Concr. Res. 144 (2021),
Andrés Barragán-Ramos: Conceptualization, Investigation, Meth­
106426, https://doi.org/10.1016/j.cemconres.2021.106426.
odology, Visualization, Writing – original draft. Camilo Ríos-Fresneda: [21] C. Trottier, A. Zahedi, R. Ziapour, L. Sanchez, F. Locati, Microscopic assessment of
Conceptualization, Methodology, Supervision, Visualization, Writing – recycled concrete aggregate (RCA) mixtures affected by alkali-silica reaction
(ASR), Constr. Build. Mater. 269 (2021), 121250, https://doi.org/10.1016/j.
review & editing. Juan Lizarazo-Marriaga: Conceptualization, Meth­
conbuildmat.2020.121250.
odology, Visualization, Writing – review & editing. Nicole Hernández- [22] M.P. Adams, J.H. Ideker, Using Supplementary Cementitious Materials to Mitigate
Romero: Resources, Investigation, Visualization, Writing – review & Alkali-Silica Reaction in Concrete with Recycled-Concrete Aggregate, J. Mater. Civ.
editing. Eng. 32 (2020) 04020209, https://doi.org/10.1061/(asce)mt.1943-
5533.0003277.
[23] K. Kim, M. Shin, S. Cha, Combined effects of recycled aggregate and fly ash towards
Declaration of Competing Interest concrete sustainability, Constr. Build. Mater. 48 (2013) 499–507, https://doi.org/
10.1016/j.conbuildmat.2013.07.014.
[24] R. Kurda, J. de Brito, J.D. Silvestre, Water absorption and electrical resistivity of
The authors declare that they have no known competing financial concrete with recycled concrete aggregates and fly ash, Cem. Concr. Compos. 95
interests or personal relationships that could have appeared to influence (2019) 169–182, https://doi.org/10.1016/j.cemconcomp.2018.10.004.

13
A. Barragán-Ramos et al. Construction and Building Materials 349 (2022) 128759

[25] R. Kurda, J. De Brito, J.D. Silvestre, Carbonation of concrete made with high [48] R. López, C. José, T. Pérez, Q. Andrés, A.T. Acosta, M. Martínez, M. Wilfredo,
amount of fly ash and recycled concrete aggregates for utilization of CO2, J. CO2 M. Molina, L. Eduardo, A. Aguilar, E.Z. Cíntora, J. Genescá, L. Benjamín, V. Salas,
Util. 29 (2019) 12–19, https://doi.org/10.1016/j.jcou.2018.11.004. Durabilidad De La Infraestructura De Concreto Reforzado Expuesta a Diferentes
[26] M. Limbachiya, M.S. Meddah, Y. Ouchagour, Use of recycled concrete aggregate in Ambientes Urbanos De México 292 (2006) 149.
fly-ash concrete, Constr. Build. Mater. 27 (2012) 439–449, https://doi.org/ [49] ASTM International, ASTM G59-97(2020), Standard Test Method for Conducting
10.1016/j.conbuildmat.2011.07.023. Potentiodynamic Polarization Resistance Measurements, (2020).
[27] K.P. Verian, Using recycled concrete as coarse aggregate in pavement concrete, [50] C. Alexandridou, G.N. Angelopoulos, F.A. Coutelieris, Mehcanical and durability
(2012) 192. performance of concrete produced with recycled aggregates from Greek
[28] R. Kurda, J. de Brito, J.D. Silvestre, Combined influence of recycled concrete construction and demolition waste plants, (2017).
aggregates and high contents of fly ash on concrete properties, Constr. Build. [51] D. Pedro, J. De Brito, L. Evangelista, Influence of the use of recycled concrete
Mater. 157 (2017) 554–572, https://doi.org/10.1016/j.conbuildmat.2017.09.128. aggregates from different sources on structural concrete, Constr. Build. Mater. 71
[29] J. Sim, C. Park, Compressive strength and resistance to chloride ion penetration (2014) 141–151, https://doi.org/10.1016/j.conbuildmat.2014.08.030.
and carbonation of recycled aggregate concrete with varying amount of fly ash and [52] H. Guo, C. Shi, X. Guan, J. Zhu, Y. Ding, T.C. Ling, H. Zhang, Y. Wang, Durability of
fine recycled aggregate, Waste Manag. 31 (2011) 2352–2360, https://doi.org/ recycled aggregate concrete – A review, Cem. Concr. Compos. 89 (2018) 251–259,
10.1016/j.wasman.2011.06.014. https://doi.org/10.1016/j.cemconcomp.2018.03.008.
[30] ASTM International, ASTM C128-15, Standard Test Method for Relative Density [53] R. Wang, N. Yu, Y. Li, Methods for improving the microstructure of recycled
(Specific Gravity) and Absorption of Fine Aggregate, (2015). concrete aggregate : A review, Constr. Build. Mater. 242 (2020), 118164, https://
[31] ASTM International, ASTM C117-17, Standard Test Method for Materials Finer doi.org/10.1016/j.conbuildmat.2020.118164.
than 75-μm (No. 200) Sieve in Mineral Aggregates by Washing, (2017). [54] C.J. Zega, Á.A. Di Maio, Use of recycled fine aggregate in concretes with durable
[32] Standard Specification for Concrete Aggregates (2018). requirements, Waste Manag. 31 (2011) 2336–2340, https://doi.org/10.1016/j.
[33] ICONTEC - Instituto Colombiano de Normas Técnicas, NTC 121 - Especificación de wasman.2011.06.011.
desempeño para cemento hidráulico, (2021). [55] D. Pedro, J. de Brito, L. Evangelista, Structural concrete with simultaneous
[34] ASTM International, ASTM C618-19, Standard Specification for Coal Fly Ash and incorporation of fine and coarse recycled concrete aggregates: Mechanical,
Raw or Calcined Natural Pozzolan for Use in Concrete, (2019). durability and long-term properties, Constr. Build. Mater. 154 (2017) 294–309,
[35] Concrete (2019). https://doi.org/10.1016/j.conbuildmat.2017.07.215.
[36] V.W.Y. Tam, X.F. Gao, C.M. Tam, Microstructural analysis of recycled aggregate [56] J. Xiao, W. Li, Y. Fan, X. Huang, An overview of study on recycled aggregate
concrete produced from two-stage mixing approach, Cem. Concr. Res. 35 (2005) concrete in China (1996–2011), Constr. Build. Mater. 31 (2012) 364–383, https://
1195–1203, https://doi.org/10.1016/j.cemconres.2004.10.025. doi.org/10.1016/j.conbuildmat.2011.12.074.
[37] V.W.Y. Tam, C.M. Tam, Assessment of durability of recycled aggregate concrete [57] S. Ghorbani, S. Sharifi, S. Ghorbani, V.W. Tam, J. de Brito, R. Kurda, Effect of
produced by two-stage mixing approach, J. Mater. Sci. 42 (2007) 3592–3602, crushed concrete waste’s maximum size as partial replacement of natural coarse
https://doi.org/10.1007/s10853-006-0379-y. aggregate on the mechanical and durability properties of concrete, Resour.
[38] V.W.Y. Tam, C.M. Tam, Diversifying two-stage mixing approach (TSMA) for Conserv. Recycl. 149 (2019) 664–673, https://doi.org/10.1016/j.
recycled aggregate concrete: TSMAs and TSMAsc, Constr. Build. Mater. 22 (2008) resconrec.2019.06.030.
2068–2077, https://doi.org/10.1016/j.conbuildmat.2007.07.024. [58] L. Evangelista, J. de Brito, Mechanical behaviour of concrete made with fine
[39] A. Abd Elhakam, A.E. Mohamed, E. Awad, Influence of self-healing, mixing method recycled concrete aggregates, Cem. Concr. Compos. 29 (2007) 397–401, https://
and adding silica fume on mechanical properties of recycled aggregates concrete, doi.org/10.1016/j.cemconcomp.2006.12.004.
Constr. Build. Mater. 35 (2012) 421–427, https://doi.org/10.1016/j. [59] S. Manzi, C. Mazzotti, M.C. Bignozzi, Short and long-term behavior of structural
conbuildmat.2012.04.013. concrete with recycled concrete aggregate, Cem. Concr. Compos. 37 (2013)
[40] ASTM International, ASTM C295 / C295M-19, Standard Guide for Petrographic 312–318, https://doi.org/10.1016/j.cemconcomp.2013.01.003.
Examination of Aggregates for Concrete, (2019). [60] A. Poursaee, Corrosion of steel bars in saturated Ca(OH)2 and concrete pore
[41] ASTM International, ASTM C1260–14 Standard Test Method for Potential Alkali solution, Chall. J. Concr. Res. Lett. 1 (2010). http://www.challengejournal.
Reactivity of Aggregates (Mortar-Bar Method), Annu. B. ASTM Stand. 04 (02) com/index.php/cjcrl/article/view/105.
(2014) 1–5, https://doi.org/10.1520/C1260-14.2. [61] D. Pedro, J. de Brito, L. Evangelista, Evaluation of high-performance concrete with
[42] ASTM International, ASTM C1567–21, Standard Test Method for Determining the recycled aggregates: Use of densified silica fume as cement replacement, Constr.
Potential Alkali-Silica Reactivity of Combinations of Cementitious Materials and Build. Mater. 147 (2017) 803–814, https://doi.org/10.1016/j.
Aggregate (Accelerated Mortar-Bar Method), Annu. B. ASTM Stand. 04 (02) (2021) conbuildmat.2017.05.007.
774–778, https://doi.org/10.1520/C1567-13.2. [62] Y. Kim, A. Hanif, S.M.S. Kazmi, M.J. Munir, C. Park, Properties enhancement of
[43] ASTM International, ASTM C39 / C39M-21, Standard Test Method for Compressive recycled aggregate concrete through pretreatment of coarse aggregates -
Strength of Cylindrical Concrete Specimens, (2021). Comparative assesssment of assorted techniques, (2018).
[44] ASTM International, ASTM C469 / C469M-14e1, Standard Test Method for Static [63] G.H.N. Suarez, Propiedades mecánicas y de durabilidad del concreto elaborado con
Modulus of Elasticity and Poisson’s Ratio of Concrete in Compression, (2014). agregados finos reciclados sometidos a carbonatación acelerada, ECI. 4 (2017)
[45] ASTM International, ASTM C1202-19, Standard Test Method for Electrical 9–15.
Indication of Concrete’s Ability to Resist Chloride Ion Penetration, (2019). [64] R.V. Silva, J. De Brito, R. Neves, R. Dhir, Prediction of chloride ion penetration of
[46] ASTM International, ASTM C642–13, Absorption, and Voids in Hardened Concrete, recycled aggregate concrete, Mater. Res. 18 (2015) 427–440, https://doi.org/
Standard Test Method for Density, 2013. 10.1590/1516-1439.000214.
[47] D.A. Jones, Principles and Prevention of Corrosion, Prentice Hall, 1996 https://
books.google.com.co/books?id=i8pRAAAAMAAJ.

14