Construction and Building Materials 158 (2018) 464–471

Contents lists available at ScienceDirect

Construction and Building Materials

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

Water absorption in recycled sand: New experimental methods to

estimate the water saturation degree and kinetic filling during mortar
mixing
Aiman Yacoub, Assia Djerbi ⇑, Teddy Fen-Chong
Université Paris-Est, MAST, FM2D, IFSTTAR, 14-20 Boulevard Newton, Cité Descartes, Champs sur Marne, F-77447 Marne la Vallée Cedex 2, France

h i g h l i g h t s

Behavior of recycled sand in mortar mixing.

A new vacuum-based method combined with an evaporation method to measure the WAC.

A new experimental approach to follow the absorption kinetics of the RS.

An adequately pre-saturating the RS can improve the mortar consistency.

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

Article history: Use of recycled concrete aggregates (RCA) as construction and building materials has not yet spread out.
Received 25 January 2017 One of the reasons lies in porosity: a given RCA is more porous than its natural aggregate because of the
Received in revised form 26 June 2017 presence of the attached old cement paste. This higher porosity causes inadequate mixing due to increase
Accepted 4 October 2017
in both RCA water absorption and water exchange between RCA and new cement paste. It is thus neces-
sary to quantify the weight fraction of in-pore water, which is usually done through the so-called water
absorption coefficient (WAC), and to decrease this potential water exchange by pre-saturating the RCA.
Keywords:
This paper aims to investigate mortar mixes made of RCA that is 1–4 mm in diameter called recycled sand
Recycled sand
Water absorption coefficient
(RS). It puts forward a new vacuum-based method combined with an evaporation method to measure the
Water saturation degree WAC, which permits saturating the RS fully and confirming that the standard method (NF EN 1097-6) lar-
Absorption kinetics gely underestimates the WAC (as already pointed out by other works). The evaporation method is also
Slump of mortar used to measure the water content of the pre-saturated RS, in order to evaluate the proportion of the
pores that remain accessible to water even after pre-saturation, which was not done in other works.
This paper then puts forward a new experimental approach to follow the absorption kinetics of the RS
by immersing them into a limestone filler, which allows better reproduction of mixing mortar since
aggregates are added into cement paste and not into water. Last, it is shown that adequately pre-
saturating the RS can improve the mortar consistency.
Ó 2017 Elsevier Ltd. All rights reserved.

1. Introduction
The sustainability of primary resources is threatened by the
construction industry because the amount of construction and
demolition waste (CDW) produced and dumped into landfills is
increasing each year [1]. This situation has forced the concrete
industry to generate effective solutions such as implementing
CDW as recycled concrete aggregates (RCA) to produce new types
of concrete. Furthermore, a recent life cycle assessment study
⇑ Corresponding author.
E-mail address: assia.tegguer@ifsttar.fr (A. Djerbi).
proved that the concrete made of RCA presents the best environ-
mental behavior [2]. CDW is currently used in the road and side-
walks construction and maintenance. However, it needs further
investigations before being used completely as RCA in new con-
crete for building construction field because of the poor porosity
properties of the RCA.
RCA consists of 2 phases: natural aggregates and hardened
cement paste (old binder). The latter is known to mainly cause
RCA to be more porous than natural aggregates [3]. This high
porosity generates an important water absorption coefficient
(WAC) [4,5]. The WAC of the RCA increases with the increase of
the amount of adherent cement paste [6]. Many studies [4–9] were

https://doi.org/10.1016/j.conbuildmat.2017.10.023
0950-0618/Ó 2017 Elsevier Ltd. All rights reserved.
 A. Yacoub et al. / Construction and Building Materials 158 (2018) 464–471
conducted in order to determine the WAC of the RCA and they con-
cluded that the WAC is underestimated by the standard method NF
EN-1097-6. A higher value of WAC leads to a higher water
exchange between RCA and the new cement paste. Indeed, this
water exchange raises the problem of using a convenient W/C ratio
to fabricate new RCA-based concrete with appropriate fresh and
hardened properties.
With regard to the fresh properties, it was found in [10] that the
incorporation of crushed clay bricks as a replacement for the natu-
ral aggregates reduces the W/C ratio and affects mortar flow. Sim-
ilar result was found in [11]: the more the incorporation of brick
waste, the more mixing water needed to obtain the adequate con-
sistency of mortar. It was also found in [12] that a higher WAC of
the RCA causes a decrease in both the yield stress and the plastic
viscosity of the RCA-based mortar. For these reasons, it was
advised in [13,14] to pre-soake the RCA or use the RCA in a dry
state and then add the water needed for the absorption in the
mix. Using the RCA in the dry state increases the initial slump
[5,15–17]. This can be related to the high initial quantity of water
used in the mix to compensate the high WAC. It was found in [18]
that the pre-saturation of the RCA by soaking for 3 and 5 min
enhanced the initial slump value of concrete and that the slump
value increased with the soaking time interval. In [19], it was
shown that the high WAC of fine recycled aggregates reduces the
W/C ratio and reduces the mortar consistency. So it was proposed
to add the recycled fine aggregates in the mixer with a determined
quantity of water and to mix them for 5 min. After leaving the con-
crete mix for 10 min at rest, other components are added. This
technique was found to improve the consistency of the concrete.
A different working procedure was followed in [20] by pre-
saturating recycled sands in a plastic bottle: it was found that
the slump flow decreases with the increase of the pre-saturation
time and pre-saturation water quantity. All this controversy
implies the need to develop a better understanding of the water
absorption kinetics of the RCA as well as the influence of the satu-
ration state of the RCA in order to allow a better use of RCA in the
construction and building field.
We here are aiming to contribute to RCA-based production of
mortar and concrete. To do so, fine RCA is eliminated and only
1–4 mm fraction, called recycled sand (RS) is used. The reason is
that smaller particles contain greater amount of attached mortar
which alters the properties of the new concrete [6], such that many
studies did not recommend the use of fine RCA in producing new
concrete [21,22]. The main contributions of this paper are as
follows:

Determine a more accurate water absorption coefficient that is
denoted here by Reference WAC (RWAC) by using a new
method that combines a vacuum-based method and an evapo-
ration method.

Study the effect of the pre-saturation process on the empty
pores of the RS by determining the water content after pre-
saturation based on the previous evaporation method.
Table 1
The chemical composition of the cement.
SiO2 Al2O3 Fe2O3 CaO MgO SO3
20.2% 5.1% 3% 64% 1% 3.1%
Table 2
The composition of the limestone filler used.
CaCO3 Cl SO2
4 S
98.80% 0.001% 0.001% 0.005%
465

Investigate the absorption kinetics of the RS in the mortar mix
using limestone filler as model material.

Study the influence of the saturation state of RS on the consis-
tency of mortar.
2. Materials
CEM I 52.5 N CE CP2 NF is the type of cement used in this study;
its chemical composition is given in Table 1. The limestone filler
used in the absorption kinetics test is provided by Betocarb, its
chemical composition is shown in Table 2. The recycled sand used
in this study was provided by the French national project Recybe-
ton as a fraction of 0–4 mm. The sand received (natural and recy-
cled) is sieved between 1 and 4 mm before any use. Fig. 1 shows
that the particle size distributions of both sands are similar. Know-
ing that RS has a heterogeneous structure and its properties
depend on the source of supply, it is important to determine its
characteristics prior to any use. Physical properties of the RS and
the NS such as compactness, WAC and density are determined
using the standard method NF EN 1097-6 [23]. The results are
shown in Table 3. The RS has a higher WAC than the NS, the density
and compactness of the RS are lower than the NS. This is expected
since the RS is more porous than the NS. A total volume substitu-
tion of the NS by RS is proposed in this study.
3. Experimental program
3.1. Saturated surface dry (SSD) state measurement by evaporation
The SSD state is defined in [24] as ‘the condition in which the
permeable pores of the aggregate’s particles are filled with water.
This condition is achieved by submerging in water for the pre-
scribed period of time, but without free water on the surface of
the particles’. This quantity of absorbed water by the specimen is
referred to as the water content of the specimen which is
expressed as a percentage weight (mass of the water divided by
the oven dry mass). The SSD state takes into account the absorbed
water only whereas the wet state of RS consists of a film of water
on the surface in addition to the absorbed water, as depicted in
Fig. 2.
The most commonly used methods to determine the SSD state
consist in toweling and drying the particles with a hot air flow.
However these methods are considered subjective since the oper-
ator alone decides when the surface of the material is dry [25]. In
addition, the toweling method is not recommended [26,27]
because of a loss of material remaining attached to the towel.
To determine the quantity of water absorbed in RS, we here
measure the weight variation of a wet sample under drying at con-
stant temperature and relative humidity conditions. According to
the theory of porous media [28,29], the free water and the
absorbed water do not evaporate at the same rate. Evaporation
includes three main phases beginning with a transition phase fol-
lowed by a constant phase (evaporation of the free water) and then
K2O Na2O S Cl CO2 Free CaO Ignition loss
0.75% 0.21% 0% 0.03% 0.4% 1.6% 0.7%
Organic materials Methylene blue test Alkaline equivalents SiO2
0.01% 0.3 g/kg 0.005% 0.1%
466 A. Yacoub et al. / Construction and Building Materials 158 (2018) 464–471
Fig. 1. The particle size distribution of the recycled and natural sand.
Table 3
Physical properties of the recycled and natural sands measured following standard
method NF EN 1097-6.
WAC (%) Density (kg/m3) Compactness
Recycled Sand 6.86 2440 0.56
Natural Sand 2.3 2530 0.63
a decreasing phase (evaporation of the absorbed water). The tran-
sition point between the constant phase and the decreasing phase
is the SSD state of the sample, which can be easily determined by
drawing the derivative of the water content with respect to time,
as shown in Fig. 3. This method was originally proposed in [28]
to determine the water content of fine RCA in the SSD state and
in [29] to measure the water absorption in mineral admixtures of
concrete.
3.2. Determining the reference WAC and the density of the RS
In [4] a new approach called ‘hydrostatic weighing’ was devel-
oped to determine the WAC of RCA. This method is based on mea-
suring the variation of mass of the RCA immersed continuously in
water during a specific period of time. Results showed that dura-
tion more than 24 h is needed to saturate the RCA. A new method,
called real-time assessment of water absorption, was developed in
[30]. In this new method, the weight of a filled pycnometer is mea-
sured over time and water is added once the level of water drops.
This procedure is repeated until no further change in weight hap-
pens. Results also showed that the 24 h duration rule is not appli-
cable in the case of RCA. Another approach, based on a relationship
between the cement paste content in the RCA and its WAC, was
conducted in [31] to determine the WAC of the RCA: it was found
that the WAC calculated is higher than the one determined by the
standard method NF EN 1097-6.
Fig. 2. The difference between the wet state and the SSD state [4].
Fig. 3. Drying curves of a porous material with respect to time: (left: evolution of
the loss of water content, right: evaporation rate) [28].
Based on these approaches, we here propose a new method in
order to determine the RWAC and the density qa of the RS. It is a
combination of the standard methods NF EN 1097-6 [23] for
weight measurement, NF P18-459 [32] for vacuum-based satura-
tion, and the evaporation method to determine the SSD state of
the specimen [29]. The process is illustrated in Fig. 4. The specimen
is placed in an airtight container and vacuumed for 2 h, then water
is added in the container and the RS is saturated under vacuum for
24 h. The tested specimen is immersed in a pycnometer full of
water, and then the pycnometer is closed and weighed (M2). The
pycnometer is emptied, the specimen undergoes a drying process
by evaporation to determine the SSD state (M1) and the pycnome-
ter is refilled with water and closed in the same way as before and
weighed (M3). Afterwards, the specimen is dried in the oven at 80
°C until constant mass (M4). Both RWAC and density are then cal-
culated using the same equations as in the standard method NF EN
1097-6 [23]:

Density: qa = qw  M4 ðM
M4
2 M 3 Þ
; qw = density of water

Reference water absorption coefficient: RWAC = 100  ðM1MM
4
4Þ
.
3.3. Measurement of water absorption in RS after pre-saturation and
its effect on slump
The study of water pre-saturation of RS and its effect on slump
is sketched in Fig. 5 and detailed in Subsections 3.3.1–3.3.3. At first,
the RS specimen is pre-saturated with a specific initial water quan-
tity for 24 h, and the quantity of water absorbed by the RS speci-
men is determined using the evaporation method. Then, mortar
made of pre-saturated RS is fabricated and the slump of the mortar
is measured. In order to interpret the obtained results further,
absorption kinetics of the RS in limestone filler paste are per-
formed by use of a hydrostatic weighing method modified from [4].
3.3.1. Complete experimental process from pre-saturation to water
content measurement
Different approaches were used to find the best pre-saturation
method of the RCA in order to optimize the quantity of water
 A. Yacoub et al. / Construction and Building Materials 158 (2018) 464–471 467

Fig. 4. A new method to determine the RWAC and the density of the RS.

Fig. 5. Study of water absorption in RS by pre-saturation and of mortars consistency.

absorbed: pre-saturation of the RCA in a closed plastic bottle with a
specific imposed quantity of water [20]; use of pre-wetted RCA
[12]; change of the soaking period of the RCA in water [18]. How-
ever these works did not evaluate the quantity of water truly
absorbed in the RCA after the pre-saturation, which is of great
importance because of the subsequent hydration of the RCA-
based mortar mix.
We here intend to do this evaluation after two methods of pre-
saturation. The first one is commonly used: it consists in pre-
saturating a specimen of sand for 24 h by total immersion under
atmospheric pressure. The second one [20] recourses to plastic bot-
tle as depicted in Fig. 6. A specimen of sand is placed in a plastic
bottle and then a quantity of water is added. The plastic bottle is
then closed for 24 h to avoid evaporation of the water. The plastic
bottle is agitated from time to time to ensure an equal distribution
of the water into the sand. In addition to that, the shaking process
helps water to drive out the trapped air and replace it in the pores
of the aggregates. A drying process is then applied to determine the
quantity of water absorbed by the RS using a digital balance placed
in a room at constant temperature and constant relative humidity
(the measurement of the weight change is done for 48 h with a
measure every 2 min). Finally, the specimen is dried in the oven
at 80 °C until a difference in weight less than 0.05%. The results
represent the average of two samples tested. Using the obtained
drying curves, we determine the quantity of water truly absorbed
by the sample, as recalled in Section 3.1.
In the following, the sand specimens at the dry state and SSD
state are referred to by ‘D’ and ‘S’ respectively, and the sand spec-
imens pre-saturated in the plastic bottle are referred to by the ini-
tial weight fraction of water (mass of water/mass of dry sand)
added at the onset of the pre-saturation (see Table 4).
3.3.2. Water absorption kinetics of RS in cement paste model
In [4] the ‘hydrostatic weighting’ method was built up to deter-
mine the WAC of coarse recycled aggregates (and not RS). This
method allows determining the absorption kinetics of the tested
sample by measuring the mass variation of coarse recycled aggre-
gates in water. However, during the fabrication of mortar and
according to the standard NF EN 196-1 [33], the aggregates are
added into a cement paste and not into water. Thus, the absorption
kinetics of the aggregates in a cement paste should be determined
under this configuration. But the absorption kinetics test cannot be
done in a cement paste due to the hydration reaction between the
cement paste and water that competes with the water absorption
in the aggregates. In order to eliminate the hydration reaction
during the evaluation of the absorption kinetics, a limestone filler
paste here replaces a cement paste. There is no hydration reaction
between a limestone filler and water such that the limestone filler
paste represents a model material of a cement paste in the
dormant period. In order to ensure the feasibility of this test, the
density, the viscosity and the stability with time of both pastes
are determined. The viscosity and the density of the limestone

Fig. 6. Pre-saturation method proposed by [20] combined with the drying process in order to determine the quantity of truly absorbed water after pre-saturation.
468 A. Yacoub et al. / Construction and Building Materials 158 (2018) 464–471

Table 4
Compositions of mortars.

Cement (g) Recycled Sand (g) Natural Sand (g) W (g) Water for absorption (g)
M-RS-D 450 1302 – 225 89.32
M-RS-S 450 1302a – 225 –
M-NS-D 450 – 1350 225 31.05
M-NS-S 450 – 1350a 225 –
6.86% 450 1302 – 225 69.14
9.00% 450 1302 – 225 45.77
12.0% 450 1302 – 225 44.14
20.0% 450 1302 – 225 2.610
25.0% 450 1302 – 225 0.000
a
Pre-saturated 24 h before mixing.

filler and cement pastes are found to be close: 13.6 Pa.s and 14 Pa.s
respectively; 1612.45 kg/m3 and 1710 kg/m3 respectively. No
bleeding was noticed in the first 15 min for both pastes. This exper-
iment is thus appropriate to evaluate the real behavior of the sand
in a mortar mix.
Fig. 7 shows how the hydrostatic weighting method is adapted
to determine the absorption kinetics of the aggregates of fraction
1–4 mm (RS) in limestone filler. Use is made of a basket with open-
ing of 0.5 mm. The pre-saturated RS is placed in the bucket
attached to a digital weighing machine with an accuracy of 1 mg.
The bucket is then immersed into a bigger bucket containing lime-
stone filler paste with W/L = 0.5. The variation of mass is then mea-
sured during 15 min.
of water estimated for absorption takes into account the results
obtained in Section 4.2 (Table 6).
Slump test is performed in order to determine the fresh behav-
ior of the mortars. The slump of the mortar was measured using
the mini cone named ‘Mortiers de Béton Equivalent’ [34] (‘Equiva-
lent Mortars to Concrete’). The slump of mortars is measured as a
function of time; a test is done every 15 min for duration of 90 min.
After each test, the specimen is put again in the mixer and remixed
before the next test.
4. Results and discussion
4.1. WAC, RWAC and density

As described in Section 3.2, the new approach aims to deter-

3.3.3. Mixing procedure and fresh behavior of mortar
The mixing procedure is done according to the standard NF EN
196-1 [33]. Fig. 8 shows the detailed mixing procedure. For all
mixes, the W/C ratio used is 0.5 and the S/C ratio is 3. The mass
of water involved in the W/C ratio is the mass of water that causes
hydration of the new cement grains and the mass of water that
causes the workability of mortar; it is called effective water (W).
Table 4 shows the compositions of mortars fabricated with nat-
ural and recycled sands. The amount of water added to the mix-
tures is fixed by taking into consideration the saturation state of
the sand. In the case of mortars containing dry sand (RS-D and
NS-D), the amount of water added in the mix (E) is equal to the
amount of effective water (W) in addition to the amount of water
required for absorption (Eabs): E = Eabs + W. When using the total
immersion method (RS-S and NS-S), the amount of the water
added in the mix is the amount of W minus the excess water in
the pre-saturated sand (Eexcess): E = W–Eexcess. When using the
pre-saturation method as explained in Section 3.3.1, the quantity
mine the RWAC and density of the RS. The results in Table 5 are
the average from the results of three tested specimens.
The RWAC determined by our vacuum-based method is approx-
imately 42.8% higher than the WAC determined by the standard
method (NF EN 1097-6). The density of the RS determined by the
new method is higher than the one determined by the standard
method. These results prove that some pores in the RS need vac-
uum pressure to be accessible and that the RS is not fully saturated
by total immersion, which is consistent with [25] that found that
the 24 h soaking at atmospheric pressure is not enough to saturate
all pores. The use of vacuum soaking is better to determine the
content of absorbed water.
The total water porosity of the RS is calculated according to the
standard method NF-P18-459 [32] and found to be 20.85%. The
volume fraction of water in the pores of the RS water-filled under
vacuum VVwater
pores
, called the saturation degree, is found to be 99.70%.
This indicates that the RS can be considered as fully saturated
using our vacuum method.

Fig. 7. Hydrostatic weighing device [4] adapted for limestone filler instead of water.
 A. Yacoub et al. / Construction and Building Materials 158 (2018) 464–471
Fig. 8. The mixing procedure of the mortar.
Table 5
Physical properties of RS.
Method Absorbed water content
(%)
NF EN 1097-6 WAC = 6.86
Our method RWAC = 9.80
Standard deviation of our method (%) 0.26
4.2. Effect of the pre-saturation on the empty pores
The quantity of water absorbed by the RS is determined by the
drying process described in Section 3.3.1. Table 6 presents the
average from the results of three tested specimens.
Table 6 shows that the quantity of water absorbed in RS placed
in the plastic bottle, increases with the quantity of water initially
added at the beginning of pre-saturation. By comparison with
Table 5, the water content of the specimen pre-saturated by total
immersion for 24 h (RS-S) is almost similar to the one determined
by the standard method NF EN 1097-6, while for a 25% of imposed
water, the water content is higher than the one determined by the
standard method (7.17% > 6.80%). The increase in the capacity of
absorbing water can be due to the shaking process during the
pre-saturating time. The shaking process drove out the trapped
air and replaced it by water in the pores of the RS contrary to
the total immersion method. In addition, this result shows that
the pre-saturation method by plastic bottle can decrease the waste
water. The experiments were stopped at 25% of imposed quantity
of water because we noticed excess water in the plastic bottle for a
higher imposed quantity of water. This excess water will affect the
W/C ratio during the fabrication of mortars.
In Section 4.1, the vacuum-based method was found to com-
pletely fill the pores of the RS. The degree of saturation can then
be estimated for each pre-saturated specimen. Results are shown
in Table 6. The proportion of empty pores decreases with the
increase of the pre-saturation water. The proportion of empty
pores for a pre-saturation water of 25% is lower than the one
pre-saturated by total immersion (RS-S): 13.25% < 17.69%. The
quantity of water needed to pre-saturate the RS can be optimized
by the plastic bottle process.
4.3. Absorption kinetics in limestone filler paste
The absorption kinetics of the sand, at dry and SSD states, is pre-
sented as the increase in mass weight fraction in the sand with
Table 6
Water content and saturation degree of different pre-saturated RS (RS-S refers to the standard immersion method while the others are designated by their initial weight fraction
of water added in the plastic bottle; the proportion of empty pores is equal to 1  Vwater/Vpores.
Specimen Pre-saturation process
RS-S Total immersion
6.86% Plastic bottle
9.00% Plastic bottle
12.0% Plastic bottle
20.0% Plastic bottle
25.0% Plastic bottle
469
respect to time in Fig. 9. The results are the average of three spec-
imens tested. A significant difference in the absorption kinetics
Density between RS and NS is noticed. The absorption kinetics of the satu-
(g/cm3) rated RS is located between the dry NS and saturated NS because
2.44 the WAC of RS is higher than the WAC of NS. Additional tests have
2.66 been done in order to determine the absorption kinetics of the pre-
0.01 saturated RS (Fig. 10). Results show that the absorption kinetics of
the RS decreases with the increase of the pre-saturation water.
None of the specimens has absorption kinetics similar to the
absorption kinetics of the saturated NS.
4.4. Slump of mortar
Figs. 11 and 12 show that the water saturation state and the
type of sand affect the slump of mortars. Results presented in
Fig. 11 show that the slumps of mortars made of NS are higher than
those of mortars made of RS in the SSD state. Same results were
found by studies done on concrete [35] and were related to the
roughness and the irregular shape of the RS.
We notice that the slumps of mortars made of dry sand are
higher than those made of saturated sand. In the case of saturated
sand, the pores of the sand are filled with water before mixing
whereas, in the case of dry sand, the pores of the sand are empty
and the amount of water for absorption is added at the beginning
of mixing in addition to the effective water (W) (see Section 3.3.3).
Thus, this additional water initially increases the W/C ratio and
thus the amount of water available for fluidizing the mixture,
which enhances the workability of mortar before the water for
absorption gets absorbed into the sand. When using the aggregates
in the dry state, the slump of M-NS-D is lower than M-RS-D, the
amount of water added into the mixer being higher in the case
of RS. These results were also found in [16,36] for the slump of con-
cretes and mortars and in [37] when adding 90% compensation
index (the amount of additional water added to the mix expressed
as weight fraction of the WAC) into the concrete mix. It was found
in [38] that using dry RCA gives a strong decrease in the slump
because they did not take into account the quantity of water
needed for the absorption of the RCA into the mix.
Fig. 12 shows the time evolution of the slump of mortars made
of RS pre-saturated in the plastic bottle with different pre-
saturation water quantities. According to the obtained results,
the pre-saturation process enhanced the slump of mortar made
of RS; same results were found on concrete of different pre-
saturation processes in [18]. Slump of mortars varies in the same
Water content (wt%) Vwater/Vpores (%) Empty pores (%)
6.80 82.31 17.69
1.55 18.61 81.39
3.34 33.35 66.65
3.47 38.47 61.53
6.66 73.75 26.25
7.17 86.75 13.25
470 A. Yacoub et al. / Construction and Building Materials 158 (2018) 464–471

Fig. 9. Absorption kinetics of recycled and natural sands in the limestone filler
paste. Fig. 12. Slump of mortars made of pre-saturated RS with plastic bottle.

Table 7
Linear fit of the slump vs. time of RS-based mortars: a is the slope; b is where the line
crosses the y-axis; R2 is the correlation factor.

Specimen a b R2
6.86% 0.3429 2.3429 0.9057
9.00% 0.3179 7.0000 0.9142
12.00% 0.3000 7.7429 0.9932
20.00% 0.2607 8.4143 0.9739
25.00% 0.2464 8.7000 0.9614
RS-S 0.3000 1.8000 0.9375
RS-D 0.8357 6.7571 0.9813

Fig. 10. Absorption kinetics of recycled sand in the limestone filler paste for
different pre-saturation processes.
and because of the high initial slump value. In Figs. 11 and 12,
no important slump loss is observed during the first 5 min. This
result suggests that the absorption by RS in cement paste is much
slower than the absorption by RS in water that is usually adopted
in the standard procedures. Figs. 9 and 10 shows that the absorp-
tion kinetics of the RS in limestone filler paste keeps increasing
after 5 minutes contrary to what was found in [8] where the RCA
absorbed 80% of its water capacity during the first 5 min. This
could be due to the fact that some surface porosity of the RS is
filled with small limestone filler particles causing a reduction in
the absorption capacity and kinetics of the RS.
As shown in Table 7, as the water content of the pre-saturation
of the RS increases the slump loss of mortar decreases. The more
the RS absorbs water during the pre-saturation process, the less
it affects the W/C ratio desired for mixing so that the workability
of the mortar is maintained (Fig. 12). For the 20–25% specimens,
the mortar workability is enhanced when compared to the RS-S
specimen.

5. Conclusion

The following conclusions can be drawn from the test results

and discussions in this paper:
Fig. 11. Slump of mortars made of RS and NS.
way as the percentage of empty pores remaining accessible to
water in the RS. Mortars made with RS with less percentage of
empty pores tend to keep their workability during time.
Table 7 presents the coefficients of the linear fit of slump vs.
time of RS-based mortars, the slope representing the slump loss.
The highest slump loss is observed in the case of dry sand, which
is expected because of the absorption of free water in the mix

The reference water absorption coefficient (RWAC) is deter-
mined by a new method that combines vacuum-based pre-
saturation and evaporation method. The standard method EN
1097-6 underestimates the WAC of the RS.

The RWAC is used to determine the total porosity of the RS in
order to determine the degree of saturation of pre-saturated RS.

The weight faction of in-pore water after pre-saturation is sig-
nificantly lower than the initial weight fraction of water added
at the pre-saturation of RS.
 A. Yacoub et al. / Construction and Building Materials 158 (2018) 464–471

The pre-saturation of the RS using a plastic bottle with 25% of
water leads to a higher water content of the RS than the total
immersion method. The pre-saturation of the RS by the plastic
bottle reduces water wastage and may be applicable in building
site, at least more easily than the vacuum-based method.

The absorption kinetics test in the limestone filler paste allows
following the behavior of the RS in a mortar mix. The proportion
of pores remaining empty after pre-saturation, the absorption
kinetics test of the sand and the slump of mortars evolve in a
logical way. The pre-saturation of the RS decreases its absorp-
tion kinetics in the limestone filler paste, which enhances the
slump of mortars.

The slump values of mortars made of RS depends on the pre-
saturation; in case of pre-saturated RS with the plastic bottle
method, the slump value of mortar increases and the slump loss
decreases as the quantity of pre-saturation water increases.
Results showed that the RS cannot be fully saturated by total
immersion or pre-saturation. It needs vacuum-based method that
seems to be inappropriate to construction sites. The standard
method NF EN 196-1 maybe has to be adapted for the fabrication
of mortar made of RS or new mixing approach should be proposed.
Hardened behavior of mortars made of different pre-saturated RCA
should be investigated also in order to choose the best pre-
saturation method that combines good fresh and hardened
properties.
References
[1] M.V.A. Florea, Z. Ning, H.J.H. Brouwers, Activation of liberated concrete fines
and their application in mortars, Constr. Build. Mater. 50 (2014) 1–12, https://
doi.org/10.1016/j.conbuildmat.2013.09.012.
[2] N. Serres, S. Braymand, F. Feugeas, Environmental evaluation of concrete made
from recycled concrete aggregate implementing life cycle assessment, J. Build.
Eng. 5 (2016) 24–33, https://doi.org/10.1016/j.jobe.2015.11.004.
[3] C. Poon, Z. Shui, L. Lam, Effect of microstructure of ITZ on compressive strength
of concrete prepared with recycled aggregates, Constr. Build. Mater. 18 (2004)
461–468, https://doi.org/10.1016/j.conbuildmat.2004.03.005.
[4] A. Djerbi Tegguer, Determining the water absorption of recycled aggregates
utilizing hydrostatic weighing approach, Constr. Build. Mater. 27 (2012) 112–
116, https://doi.org/10.1016/j.conbuildmat.2011.08.018.
[5] Zhao Zengfeng, Valorisation des sables de béton recyclé pour la fabrication de
mortiers, Thesis, école des Mines, 2014.
[6] M.S. de Juan, P.A. Gutiérrez, Study on the influence of attached mortar content
on the properties of recycled concrete aggregate, Constr. Build. Mater. 23
(2009) 872–877, https://doi.org/10.1016/j.conbuildmat.2008.04.012.
[7] V.W.Y. Tam, C.M. Tam, Crushed aggregate production from centralized
combined and individual waste sources in Hong Kong, Constr. Build. Mater.
21 (2007) 879–886, https://doi.org/10.1016/j.conbuildmat.2005.12.016.
[8] J.P.B. Vieira, J.R. Correia, J. de Brito, Post-fire residual mechanical properties of
concrete made with recycled concrete coarse aggregates, Cem. Concr. Res. 41
(2011) 533–541, https://doi.org/10.1016/j.cemconres.2011.02.002.
[9] J. Rueda, E. Dapena, P. Alaejos, S. Menéndez de Llano, An accelerated test to
assess the quality of recycled concrete sands based on their absorption
capacity, Constr. Build. Mater. 78 (2015) 464–469, https://doi.org/10.1016/
j.conbuildmat.2014.12.039.
[10] F. Bektas, K. Wang, H. Ceylan, Effects of crushed clay brick aggregate on mortar
durability, Constr. Build. Mater. 23 (2009) 1909–1914, https://doi.org/
10.1016/j.conbuildmat.2008.09.006.
[11] J. Silva, J. de Brito, R. Veiga, Recycled red-clay ceramic construction and
demolition waste for mortars production, J. Mater. Civil Eng. 22 (2010) 236–
244, https://doi.org/10.1061/(ASCE)0899-1561(2010) 22:3(236).
[12] V. Corinaldesi, G. Moriconi, Behaviour of cementitious mortars containing
different kinds of recycled aggregate, Constr. Build. Mater. 23 (2009) 289–294,
https://doi.org/10.1016/j.conbuildmat.2007.12.006.
[13] A.R. Khaloo, Properties of concrete using crushed clinker brick as coarse
aggregate, ACI Mater. J. 91 (1994), https://doi.org/10.14359/4058.
[14] M. Mansur, T.H. Wee, S.C. Lee, Crushed bricks as coarse aggregate for concrete,
ACI Mater. J. 96 (1999), https://doi.org/10.14359/649.
471
[15] H. Mefteh, O. Kebaïli, H. Oucief, L. Berredjem, N. Arabi, Influence of moisture
conditioning of recycled aggregates on the properties of fresh and hardened
concrete, J. Clean. Prod. 54 (2013) 282–288, https://doi.org/10.1016/j.
jclepro.2013.05.009.
[16] Z. Zhao, S. Remond, D. Damidot, W. Xu, Influence of fine recycled concrete
aggregates on the properties of mortars, Constr. Build. Mater. 81 (2015) 179–
186, https://doi.org/10.1016/j.conbuildmat.2015.02.037.
[17] C.S. Poon, S.C. Kou, L. Lam, Influence of recycled aggregate on slump and
bleeding of fresh concrete, Mater. Struct. 40 (2007) 981–988, https://doi.org/
10.1617/s11527-006-9192-y.
[18] J. García-González, D. Rodríguez-Robles, A. Juan-Valdés, J. Morán-del Pozo, M.
Guerra-Romero, Pre-saturation technique of the recycled aggregates: solution
to the water absorption drawback in the recycled concrete manufacture,
Materials 7 (2014) 6224–6236, https://doi.org/10.3390/ma7096224.
[19] G.M. Cuenca-Moyano, M. Martín-Morales, I. Valverde-Palacios, I. Valverde-
Espinosa, M. Zamorano, Influence of pre-soaked recycled fine aggregate on the
properties of masonry mortar, Constr. Build. Mater. 70 (2014) 71–79, https://
doi.org/10.1016/j.conbuildmat.2014.07.098.
[20] T. Le, G. Le Saout, E. Garcia-Diaz, S. Remond, Influence of the moisture
condition of recycled concrete aggregates on the interfacial transition zone
(ITZ) and the compressive strength of mortar, in: Paris, 2015.
[21] L. Ferreira, J. de Brito, M. Barra, Influence of the pre-saturation of recycled
coarse concrete aggregates on concrete properties, Mag. Concr. Res. 63 (2011)
617–627, https://doi.org/10.1680/macr.2011.63.8.617.
[22] International Conference on the Use of Recycled Materials in Buildings and
Structures, E. Vázquez, International Union of Laboratories and Experts in
Construction Materials, Systems and Structures, eds., International RILEM
Conference on the Use of Recycled Materials in Buildings and Structures:
Barcelona, Spain, pp. 8–11, November 2004, RILEM Publications, Bagneux,
2004.
[23] NF EN 1097-6, Tests for mechanical and physical properties of aggregates –
part 6: determination of particle density and water absorption, 2014.
[24] ASTM C 127, Standard Test Method for Relative Density (Specific Gravity) and
Absorption of Coarse Aggregate, 2015.
[25] M. Quattrone, B. Cazacliu, S.C. Angulo, E. Hamard, A. Cothenet, Measuring the
water absorption of recycled aggregates, what is the best practice for concrete
production?, Constr Build. Mater. 123 (2016) 690–703, https://doi.org/
10.1016/j.conbuildmat.2016.07.019.
[26] A.S. Mujumdar (Ed.), Handbook of industrial drying, 3rd ed., CRC/Taylor &
Francis, Boca Raton, FL, 2007.
[27] N.H. Ceaglske, O.A. Hougen, Drying granular solids, Ind. Eng. Chem. 29 (1937)
805–813, https://doi.org/10.1021/ie50331a017.
[28] P. Gentilini, O. Yazoghli-Marzouk, V. Delmotte, Y. Descantes, Determination of
the water content of fillerised fine aggregates in the saturated surface dry
state, Constr. Build. Mater. 98 (2015) 662–670, https://doi.org/10.1016/
j.conbuildmat.2015.08.131.
[29] J.-M. Mechling, A. Lecomte, K. Merriaux, Mesure de l’absorption d’eau des
additions minérales des bétons par évaporométrie: measurement of the
absorption of water of the mineral admixtures in concrete by evaporometry,
Mater. Struct. 36 (2003) 32–39, https://doi.org/10.1007/BF02481568.
[30] V.W.Y. Tam, X.F. Gao, C.M. Tam, C.H. Chan, New approach in measuring water
absorption of recycled aggregates, Constr. Build. Mater. 22 (2008) 364–369,
https://doi.org/10.1016/j.conbuildmat.2006.08.009.
[31] T. Le, S. Rémond, G. Le Saout, E. Garcia-Diaz, Fresh behavior of mortar based on
recycled sand – influence of moisture condition, Constr. Build. Mater. 106
(2016) 35–42, https://doi.org/10.1016/j.conbuildmat.2015.12.071.
[32] NF P18-459, Concrete – testing hardened concrete – testing porosity and
density, 2010.
[33] EN 196-1, Methods of Testing Cement – Part 1: Determination of Strength,
European Committee for Standardization, 2006.
[34] A. Schwartzentruber, C. Catherine, La méthode du mortier de béton équivalent
(MBE) – un nouvel outil d’aide à la formulation des bétons adjuvantés, Mater.
Struct. 33 (2000) 475–482.
[35] D. Matias, J. de Brito, A. Rosa, D. Pedro, Mechanical properties of concrete
produced with recycled coarse aggregates – influence of the use of
superplasticizers, Constr. Build. Mater. 44 (2013) 101–109, https://doi.org/
10.1016/j.conbuildmat.2013.03.011.
[36] C.S. Poon, Z.H. Shui, L. Lam, H. Fok, S.C. Kou, Influence of moisture states of
natural and recycled aggregates on the slump and compressive strength of
concrete, Cem. Concr. Res. 34 (2004) 31–36, https://doi.org/10.1016/S0008-
8846(03)00186-8.
[37] M.B. Leite, J.G.L. Figueire do Filho, P.R.L. Lima, Workability study of concretes
made with recycled mortar aggregate, Mater. Struct. 46 (2013) 1765–1778,
https://doi.org/10.1617/s11527-012-0010-4.
[38] A. Ait Mohamed Amer, K. Ezziane, A. Bougara, M. Adjoudj, Rheological and
mechanical behavior of concrete made with pre-saturated and dried recycled
concrete aggregates, Constr. Build. Mater. 123 (2016) 300–308, https://doi.org/
10.1016/j.conbuildmat.2016.06.107.