Construction and Building Materials 197 (2019) 241–250

Contents lists available at ScienceDirect

Construction and Building Materials

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

Effect of carbonated aggregates on the mechanical properties and

thermal conductivity of eco-concrete
Imen Rahmouni a,b,⇑, Geoffrey Promis b, Abdelhamid R’mili a, Hassen Beji b, Oualid Limam a
a
Laboratory of Civil Engineering (LGC-ENIT), University of Tunis El-Manar, ENIT-BP 37, 1002 Tunis le Belvédère, Tunisia
b
Laboratory of Innovative Technology (LTI), GC Department, University of Picardie Jules Verne, Avenue des Facultés-Le Bailly, 80025 Amiens Cedex, France

h i g h l i g h t s

New ecological concretes made with carbonated aggregates.

Measurement of mechanical and thermal properties of eco-concretes studied.

Prediction and validation of thermal conductivity with HAC model.

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

Article history: This paper presents a study on the effect of carbonated-aggregates on mechanical properties and thermal
Received 25 June 2018 conductivity of eco-concretes. These carbon negative eco-aggregates are produced by accelerated carbon-
Received in revised form 13 November 2018 ation where industrial by-products or incinerators’ waste are combined with carbon dioxide (CO2). These
Accepted 23 November 2018
manufactured aggregates have particular characteristics giving them a certain interest in the field of civil
Available online 28 November 2018
engineering such as in replacement of natural aggregates in concrete. Morphological, physical, mechan-
ical and thermal characterizations were performed on eco-aggregates in order to use them in the produc-
Keywords:
tion of concrete. Then, concretes based on these carbonated-aggregates were formulated. In the
Incinerators’ waste
Carbon negative aggregates
experimental program, different mixtures were prepared by partially replacing natural aggregates by car-
Accelerated carbonation technology bonated ones resulting in concretes with unit weights varying between 1750 and 2290 kg/m3. For all pro-
Formulation duced concrete mixes, the water to cement (w/c) ratio was kept constant. Compressive strength, modulus
Mechanical properties of elasticity, thermal conductivity, water absorption and sorptivity were measured. Results have been
Thermal conductivity compared to control concrete mixes produced using ordinary aggregates and expanded clay. The results
show that the tested eco-concretes have sufficient mechanical strength and interesting thermal conduc-
tivity. Therefore, they could be used for construction projects.
Ó 2018 Elsevier Ltd. All rights reserved.

1. Introduction approaching closed-loop works: extracted product => used product

=> recycled product => reused product.
Huge quantities of aggregates are consumed annually given At the same time, the rapid growth of the urban fabric has led to
their frequent usage in several applications. Therefore, it is essen- an increase in the quantities of industrials by-products and waste
tial to ensure the supply of such important construction material in [13]. These materials represent an economic and ecological burden
a sustainable way for future generations [1,2]. since they need more landfill spaces and cause environmental pol-
Recycled aggregates seem to be a good alternative for natural lution [14]. If it is not possible to eliminate waste or industrial by-
aggregates, which explains the worldwide interest in researching products, it is necessary to find the best conditions for salvaging
their properties for possible usage as civil engineering materials these materials [15]. Therefore, scientists and engineers should
[3–12]. In fact, recycled aggregates offer the advantage of perceive waste as a potential source to exploit.
Moreover, industrial and incinerators waste contains heavy
metals and carbon dioxide (CO2) that are harmful to health and
environment. CO2, being a greenhouse gas accumulates in the
⇑ Corresponding author at: Laboratory of Innovative Technology (LTI), GC
atmosphere and contributes to global warming. Waste recycling
Department, University of Picardie Jules Verne, Avenue des Facultés-Le Bailly,
80025 Amiens Cedex, France.
is effective for reducing greenhouse gases and other pollutants.
E-mail address: imenrahmouniimen@hotmail.fr (I. Rahmouni). When recycled products are used in the production of civil engi-

https://doi.org/10.1016/j.conbuildmat.2018.11.210
0950-0618/Ó 2018 Elsevier Ltd. All rights reserved.
242 I. Rahmouni et al. / Construction and Building Materials 197 (2019) 241–250

neering materials, natural deposits are preserved for longer on the surface of the grains and progressively fill the capillary voids. Carbonation
has many advantages such as agglomeration of particles for the formation of solid
periods.
aggregates (hardening of the matrix makes it possible to obtain resistant finished
In order to respond simultaneously to the vital need for con- materials) and physical (pollutants) and chemical (formation of insoluble metal car-
struction and the universal necessity to conserve resources and bonates and decrease of PH) stabilization. For all these reasons, the recovery of
protect the environment against pollution and the release of CO2, waste from heat treatments (rich in CaO) is particularly suitable for the carbonation
it is essential to reuse waste and industrial by-products, which process. Based on this observation, the University of Greenwich has developed the
ACT [18,19] that allows thermal residues of minerals to react with carbon dioxide.
when combined with CO2, can be transformed into aggregates.
The precipitation of carbonates (limestone) generates a permanent fixation of CO2
These artificial aggregates could be used in many construction and the formation of inert aggregates. This process aims to carbonate incinerator
applications such as the substitution of natural aggregates in the waste in an atmosphere enriched with carbon dioxide to generate aggregates. After
production of concrete if they do not alter its properties. This dual being transported and stored at the production centre, the waste is pre-treated by
carbonation and then mixed with cement, sand and water to form a binder paste.
benefit of reducing carbon dioxide emissions and recycling waste
The materials are then carbonated in a rotating reactor thus forming negative car-
is an interesting alternative. Thus, the University of Greenwich bon lightweight aggregates. The carbon dioxide is transformed into carbonic acid,
and Carbon8systems developed a new technology, known as the which will then be able to initiate reactions with calcium hydroxide to form cal-
accelerated carbonation technology (ACT). ACT process has been cium carbonates [20–22].
shown to be effective in making incineration residues inert. The
process is currently directed to category 1 landfills (hazardous 2.2. Characterization of carbonated aggregates
waste). In addition to limiting the quantities of buried materials,
ACT would make it possible to recycle carbon dioxide present in Three different waste sources have been selected from southeast England and
northwest France. Aggregates, labelled as C8 AGG (Fig. 1-a), were produced from
the incineration fumes because of the carbonation reaction that residues of air-pollution control residues (APCr) [18] with cement used as binder.
consumes a lot of CO2. Thus, this new type of aggregates can par- The second waste used to produce aggregates, labelled as PA (Fig. 1-b), is paper
tially or fully replace natural aggregates, which preserves natural ash [19]. These aggregates were produced without the addition of a binder but with
deposits. These eco-aggregates have bulk densities less than addition of quarry fines, rich in calcium oxide CaO [20,21]. The last tested aggre-
gates, labelled as BAk (Fig. 1-c) were manufactured using waste from a French bio-
2000 kg/m3, which classify them as lightweight aggregates.
mass centre [23].
In order to promote these eco-aggregates within the building In addition to the three types of eco-aggregates presented above, expanded clay
and public works sector, it is necessary to characterize their prop- (labelled as EC) and typical natural aggregates (labelled as O) were tested in this
erties as well as those of the products they make part of their com- study and served as control materials.
position (concretes based on carbonated aggregates). For this Physical tests such as gradation analysis, specific gravities (SG), water absorp-
tion and porosity were performed on all tested aggregates. In addition, mechanical
purpose, morphological, physical, mechanical and thermal charac- characteristics (crushing resistance and LA abrasion) and thermal conductivities
teristics of these materials were studied in this research. The paper were also determined.
starts by presenting the industrial process for manufacturing eco- Gradation testing was performed using the dry sieving method according to EN
aggregates. Then, characteristics of all tested aggregates, sample 933-1 standard [24]. The obtained gradation curves for all tested aggregates as well
as that of natural sand (used for the concrete production in this study) are shown in
preparation and concrete formulation are described. Afterwards,
Fig. 2. The particle size analysis shows that the dimensional distribution of aggre-
the mechanical properties and thermal conductivity results for gates produced by the accelerated carbonation process is similar to that of EC
all prepared eco-concretes are introduced. In order to situate the and O control aggregates. The uniformly and curvature coefficients (Cu and Cc,
characterization results for these new sustainable materials with respectively) show that all tested aggregates are uniformly graded (Cu < 4 and
respect to those of traditional building materials a comparative 1 < Cc < 3). The gradation analysis for all tested eco-aggregates confirms their suit-
ability for usage as coarse aggregates in producing concrete.
study is proposed. This comparison was performed in order to
Results of the physical and mechanical tests are summarized in Table 1. The
study the effects of carbonated aggregates on mechanical and ther- specific gravities were determined according to EN 1097-6 standard [25,26]. Car-
mal properties of concretes. This would allow the development of a bonated aggregates have low bulk specific gravity as compared to ordinary aggre-
concrete formulation method based on lightweight aggregates’ gates, which classify them as lightweight aggregates. Their porosity is higher
than that of ordinary aggregates. Carbonated aggregates present moderate abrasion
characteristics.
resistance (high LA coefficients). Crushing resistance was determined according to
standard NF 13055-1 [27]. The results show that the crushing resistance of the dif-
2. Materials and method ferent tested eco-aggregates is smaller than that of ordinary aggregates.
High porosity of carbonated aggregates presents a problem during their mixing
2.1. Aggregate production to produce concrete. Therefore, an important correction for the needed mixing water
should be taken into account when formulating concrete with these eco-aggregates.
Aggregates are produced by accelerated carbonation process, patented by Car- For this reason, measurement of water absorption capacity for the different aggre-
bon8 System industry, an entity of the University of Greenwich [16,17]. Accelerated gates was done on three 400-g-samples immersed in water. The water absorption
carbonation is a rapid and accelerated version of natural carbonation, which coefficient is determined by continuous weight measurements for a period of 7 days.
exposes materials to high concentrations of carbon dioxide. Many wastes are reac- As shown by Fig. 3, an immediate absorption of 22% was measured for PA aggregates.
tive to carbon dioxide especially industrial waste resulting from heat treatment. After 7 days, the EC aggregates reached about 20% water absorption. Concerning C8
Carbonation is a very slow reaction of carbon dioxide in humid environment that aggregates and BAk, an atypical behaviour was noticed, with fast water absorption of
produces carbonic acid as shown by Eq. (1). the order of 5% then a loss of mass related to the disintegration of these aggregates in
water. Compared to natural aggregates, the carbonated aggregates have higher val-
CO2 + H2 O ! H2 CO3 ð1Þ ues of water absorption coefficient, which is due to their internal porosity.
Subsequently, carbonic acid reacts with lime to form limestone as shown by Eq. Fig. 4 shows the variation of water absorption index, defined as iw ð%Þ ¼ WWA24
At

(2). [28], with respect to time. The figure shows that water absorption after 24 h of
immersion is not sufficient for determining the water absorption coefficient of
CaO + H2 CO3 ! CaCO3 + H2 O ð2Þ aggregates. In fact, some aggregates continued to absorb water for a period of three
days before stabilisation.
In case of carbonation of cement matrix, calcium silicates C3S and C2S will inter-
Based on experimental results, a linear relationship, with a coefficient of deter-
act with carbon dioxide and water because of their hydraulic nature. The reaction
mination R2 = 0.96, was found between the 24 h water absorption WA24 and the
produces calcium carbonates CaCO3, which crystallizes into calcite stable poly-
density of aggregates (see Fig. 5). The relationship can be expressed by Eq. (4):
morph and hydrated calcium silicates CSH (non-crystalline compound) as shown
by Eq. (3).

d kg=m3 ¼ 2600  51  W A24 ð%Þ ð4Þ

Carbonated aggregates have higher porosity than natural aggregates. As the

ð3Þ
porosity increases, the water absorption coefficient at 24 h increases. Regardless
of the aggregates type, based on the experimental results, a linear relationship
The entanglement of the CxSHy gel brings the solidity of the aggregates and par- between aggregates’ porosity and water absorption was found as expressed by
ticipates in making them inert, which ensures their chemical stability. CSH develop Eq. (5) (see Fig. 6):
 I. Rahmouni et al. / Construction and Building Materials 197 (2019) 241–250 243

Fig. 1. Studied eco-aggregates (a) C8 AGG, (b) PA and (c) BAk.

Fig. 2. Gradation curves for all tested aggregates.

Fig. 3. Water absorption of all tested aggregates.

Table 1
Aggregates characteristics.

Aggregate’s characteristics C8 AGG PA BAk EC O

Bulk SG qv (g/cm3) 1.1 0.9 1.1 0.7 1.4
Oven dry SG qrd (g/cm3) 1.8 1.4 1.9 1.2 2.5
Saturated surface (SSD) SG 2.1 1.8 2.2 1.4 2.5
qssd (g/cm3)
Apparent SG qa (g/cm3) 2.6 2.4 2.8 1.5 2.6
Porosity (%) 30.2 62.6 49.9 53.8 3.7
Crushing resistance (MPa) 9.7 6.5 8.8 8.9 21.3
Los Angeles coefficient (%) 52.5 51 53 30 21

nð%Þ ¼ 46:7 þ 1:06  W A24 ð%Þ with R2 ¼ 0:86 ð5Þ

Thermal conductivity of the studied aggregates was performed using the

guarded hot plate in steady state method in accordance with standard [29]. With
this method, aggregates are placed in a 25  25  6 cm3 expanded polystyrene
mould and are then covered with a transparent plastic sheet. This experimental
setup results in measuring the conductivity of bulk aggregates kexp. The real ther-
mal conductivity of aggregates kagg was calculated in two steps using the arithmetic
average of Hashin-Shtrikman bounds kU and kL of a two-phase material [23,30]
according to Eqs. (6) and (7). Fig. 4. Variation of water absorption index as a function of time.

kU þ kL
keff ¼ ð6Þ
2
ity and the calculated real thermal conductivity. Good thermal performance was
For k U  kL, the Hashin-Shtrikman lower kL and upper kU bounds for thermal found for paper ash aggregates PA with a thermal conductivity of 0.15 W/mK, which
conductivity of a two phase composite can be written as follows: is similar to that of expanded clay. C8 and BAk aggregates have a higher thermal con-
ductivity of 0.2 W/mK, but still lower than that of ordinary aggregates with a ther-
e 1e mal conductivity of 0.24 W/mK. Results of aggregates’ thermal conductivity kagg
kU ¼ ks þ and kL ¼ ka þ ð7Þ
1
ka ks
þ 1
3ks
e 1
þ 3ke a
ks ka show that carbonated aggregates have lower thermal conductivity than ordinary
aggregates but higher than EC. Thermal conductivity of the solid phase of aggregates
where ks, keff and ka are respectively the thermal conductivity of aggregate of solid shows that C8, BAk and PA aggregates and ordinary aggregates have similar conduc-
phase, bulk aggregate and air. e is the volume fraction of air. The thermal conductiv- tivity values. These carbonated aggregates produced by recycling waste containing
ity of the aggregates was calculated using the volume fraction of inter and intra- metals and minerals that have high thermal conductivities. Eco-aggregates have
particular air. Table 2 shows the measured results for the bulk aggregate conductiv- low thermal conductivities due to their porosity.
244 I. Rahmouni et al. / Construction and Building Materials 197 (2019) 241–250

a constant w/c ratio of 0.5. In total, Seventeen mixtures were formulated in this
study using different proportions of lightweight eco-aggregates. In fact, ordinary
aggregates were replaced by 25, 50, 75 and 100% of eco-aggregates.
Table 3 summarizes the formulations of the various eco-concretes. All produced
eco-concretes in this study have a dry density less than 2000 kg/m3, which classify
them as lightweight concretes according to RILEM’s classification [35]. The density
was determined on samples at 28 days by dividing their dry masses with their vol-
ume. The consistence of the different mixes was measured using the slump test. The
slump values obtained for the different mixes classify them as class S2 according to
European standard NF EN 206-1 standard [36].
For all studied mixes, several specimens were prepared for different tests:

Twelve cylinders 11 cm in diameter and 22 cm in length for standard compres-

sive strength tests.

Twelve 7  7  28 cm3 prismatic beams for bending tensile test.

Six 4  4  16 cm3 prismatic beams for water absorption by immersion and by
capillarity and sorptivity tests.

One 25  25  6 cm3 prismatic beam for thermal tests.

One 10  10  10 cm3 cube for Moisture Buffer Value tests.

It is to note that all tests were carried out on air-dry specimens at the ages of 7,
Fig. 5. Relationship between density and water absorption. 28, 90 and 365 days.

3. Results and discussion

3.1. Unit weight

As shown by Fig. 7, the dry density of the concrete mix

decreases as the percentage of eco-aggregates incorporated in
the mix increases. The unit weights were calculated in fresh and
hardened states. The dry unit weight is presented here for illustra-
tion of the effect of carbonated aggregates on density. When unit
weight air-dry conditions (during mechanical tests) are compared,
it’s obvious that mixture containing only ordinary aggregates has a
2300 kg/m3 dry unit weight upper than dry density of concretes
made by carbonated aggregates. There is a good correlation of
the measurements.
Due to the higher porosity and water absorption of the aggre-
gates, the aggregate-paste interface transition zone is improved
(the width of the area is extremely low). That can be observed on
the images obtained by scanning electron microscopy (SEM)
Fig. 6. Relationship between water absorption and porosity. (Fig. 8). We can also see that cement paste penetrate into open sur-
face pores of the lightweight aggregates providing a good paste-
aggregate bonding. This is due to a porous shell of the aggregate
Table 2 and kinetics of water absorption of the aggregates.
Thermal conductivity results for all studied aggregates. Based on experimental results, there is linear relationship
Aggregates C8 BAK PA EC O between the density of prepared concretes and their porosity
e inter-particular air 0.4 0.44 0.39 0.42 0.46 (Fig. 9) that can be expressed by Eq. (8):
e intra-particular air 0.3 0.31 0.41 0.18 0.03
kexp (W/mK) 0.2 0.2 0.15 0.15 0.24 qb ðg=cm3 Þ ¼ 0:06ð40  nð%ÞÞ ð8Þ
kagg (W/mK) 0.54 0.62 0.53 0.39 0.81
ks (W/mK) 0.9 0.81 0.85 0.5 0.95
3.2. Compressive strength

2.3. Formulation of concretes Compressive strength tests were performed according to NF EN

12390 standard [37] at the ages of 7, 28, 90 and 365 days using
The cement used in this study is a CEMII/B 32.5 R CE NF multipurpose cement as three replicate specimens per mix. As shown by Fig. 10, the 28-
classified by standard EN 197-1 [31]. Rolled siliceous local sand with 0/5 mm gran-
ular class, 2.7 specific gravity and 2.23 finesses modulus was used for all prepared
day compressive strength of concrete decreases as the percentage
mixes. The same mixing sequence was used for all mixtures. Before mixing the var- of incorporated eco-aggregates in the mix increases [38,39]. For
ious constituents for 15 min, aggregates are pre-wetted using the amount of water instance, the control mix made with 100% ordinary aggregates
necessary for their absorption capacity. This amount is equivalent to the water achieved a compressive strength of 33.2 MPa after 28 days, while
absorbed after 30 min of immersion. The aggregates achieve therefore a saturated
the mix with 25% PA aggregates achieved only 21.8 MPa at the
surface-dry (SSD) condition. Then, natural sand and cement are added. Finally,
the mixing water is slowly added and mixed for 4 min. The prepared testing spec- same age, representing a decrease of about 35%. It appears that
imens are demolded after 24 h and stored in a chamber with relative humidity the compressive strength of concretes is proportional to the aggre-
ranging from 60 to 80% and a constant temperature of 23 °C. Conventional concrete gates crushing resistance [40] as well as the bulk density of aggre-
formulation methods are applicable in the case of lightweight concretes [32]; with gates. The aggregate’s resistance depends on the producing process
special attention given to the water absorption of lightweight aggregates [33]. In
this study, Dreux-Gorisse’s method [34] was selected for economic reasons as it
and shell structure of aggregates [41]. A decreasing linear relation-
results in using a smaller quantity of cement and a larger quantity of aggregates, ship, with coefficient of determination exceeding 0.8 was found
which translates in better thermal properties for produced lightweight concretes. between the concrete 28-day compressive strength and eco-
All the mixes were formulated using a constant cement dosage of 350 kg/m3 and aggregates content.
 I. Rahmouni et al. / Construction and Building Materials 197 (2019) 241–250 245

Table 3
Mix proportions for all studied concretes.

Concrete’s Cement Eco-aggregate Ordinary Sand Water Effective W/C Slump (mm)
composition (kg/m3) aggregates
% C8 100 350 1007 – 639 175 0.5 80
75 350 755 252 639 175 0.5 70
50 350 503 503 639 175 0.5 70
25 350 252 755 639 175 0.5 80
% PA 100 350 869 – 658 175 0.5 70
75 350 652 217 658 175 0.5 80
50 350 435 435 658 175 0.5 70
25 350 217 652 658 175 0.5 80
% BAK 100 350 1069 – 639 175 0.5 80
75 350 817 267 649 175 0.5 70
50 350 534 534 639 175 0.5 70
25 350 267 817 639 175 0.5 80
% EC 100 350 475 – 551 175 0.5 90
75 350 356 119 532 175 0.5 90
50 350 238 238 532 175 0.5 80
25 350 119 356 570 175 0.5 80
%O 100 350 – 1103 755 175 0.5 70

 0:3
f
Ec ¼ 22 cm ð9Þ
10
Where Ec is the modulus of elasticity in GPa and fcm is the compres-
sive strength in MPa. Fig. 14 shows that the EC2 equation did not
predict well the measured values. A nonlinear regression of the
measured values using the same form as the EC2 expression
resulted in Equation (10) with an R2 value of 0.57.
 0:6684
f cm
Ec ¼ 17:54  ð10Þ
10
The American Concrete Institute ACI [44] predicts the modulus
of elasticity of concrete through the use of Eq. (11):
qffiffiffiffi
Ec ¼ wc 1:5 0:043 f c for 2:5  f c  40 MPa and 1440
kg
 wc  2500 ð11Þ
m3
Fig. 7. Effect of eco-aggregates content on dry density of concrete.
Where wc is the unit weight of the concrete (kg/m3) and fc and Ec
Table 4 presents the results of the compressive strength for all are the compressive strength and the modulus of elasticity (MPa),
testing ages. As expected, for all tested mixes, as the age increases respectively.
so does the compressive strength [3,42]. The strength development ACI prediction equation was found to underestimate the modu-
is slow after the seven first days. lus of elasticity of concrete incorporating eco-aggregates.
The flexural strength was determined by 3-points flexural test The carbonated aggregates are porous and have lower crushing
on 7  7  28 cm3 specimens at the age of 7, 28, 90 and 365 days. resistance so lower modulus of elasticity. Concrete’s modulus of
Fig. 11 presents the 28-days-flexural strength of the studied con- elasticity is proportional to the aggregate’s modulus of elasticity.
cretes and shows good values for eco-concretes compared to ordi- In micro structural point of view, bond cracks penetrate into aggre-
nary one. gates [39]. In addition, cement paste penetrate into open pores of
As shown by Fig. 12, a brittle type of fracture is obtained for the the lightweight aggregates providing a good paste-aggregate bond-
tested specimens either in compression or in flexural tension. The ing (Fig. 8).
aggregate’s shell remains intact and integral with the matrix while
the interior of the aggregates is detached from the matrix. 3.4. Thermal conductivity

3.3. Modulus of elasticity The guarded hot plate method in steady state [29] was used to
measure the thermal conductivity of all studied concretes [23]. The
As presented by Fig. 13, when the percentage of eco-aggregates tested specimens are 25  25  6 cm3 dried at 70 °C until constant
increases, the modulus of elasticity of concrete decreases. For weight. Thermal conductivities are obtained for four temperatures
example, the modulus of elasticity of the control mix was 10°, 20°, 30° and 40 °C. Fig. 15 shows the measured thermal con-
37.4 GPa, which is a typical value for conventional concrete, while ductivity of all studied concretes as a function of temperature.
that of the mix prepared using 100% PA aggregates was 23.6 – a We can see that temperature has a negligible effect on thermal
reduction of about 36%. conductivity. Concrete incorporating C8 aggregates has a thermal
To predict the modulus of elasticity of concrete from its com- conductivity of 0.7 W/mK at 10 °C, which is almost similar to that
pressive strength, Eurocode EC2 proposes the expression shown of EC concrete. Ordinary concrete has a thermal conductivity of
by Eq. (9) [43]. 1.23 W/mK. Thermal conductivities of eco-concretes as a function
246 I. Rahmouni et al. / Construction and Building Materials 197 (2019) 241–250

Fig. 8. SEM-images for concretes (a) EC, (b) PA and (c) BAk.

Fig. 9. Density of prepared concretes as a function of their porosity.

Fig. 10. Effect of eco-aggregates content on concrete 28-day compressive strength.

of eco-aggregates percentage are presented in Fig. 16. Replacing by Arnoud et al. [46] for heterogeneous materials of hemp in bulk,
ordinary aggregates with carbonated ones reduces the thermal hemp concrete and hemp wools. The heterogeneous material is
conductivity due to the porous structure of carbonated aggregates assimilated to an equivalent homogeneous material composed by
[45]. For example, the thermal conductivity of the mixture contain- an assembly of spherical inclusions. This material could be consti-
ing 75% C8 AGG is 45% lower than that of the control mix, and 21% tuted by two or three inclusions. Concretes based on carbonated
smaller than that of the mix incorporating 50% C8 AGG. aggregates are assimilated to an assembly of 3 inclusions. The
To estimate the thermal conductivity of the eco-concretes, the model is then a tri-composite model with an internal sphere of
auto-coherent homogenization model was used. This model esti- air (ka, ra) surrounded by a spherical inclusion of carbonated aggre-
mates thermal conductivity of heterogeneous materials knowing gates (kpG, rpG) and moreover surrounded by an inclusion of
the conductivity of each component and its volume fraction. This cement matrix (ksl, rsl) (Fig. 17). The equivalent thermal conductiv-
model was developed for mechanical characterization and used ity of the concrete is given by Eq. (12).
 I. Rahmouni et al. / Construction and Building Materials 197 (2019) 241–250 247

Table 4
Compressive strength results for all testing ages.

Concrete 7 days Compressive 28 days Compressive 90 days Compressive 365 days Compressive
strength (MPa) strength (MPa) strength (MPa) strength (MPa)
% C8 100 7.7 17.8 19.9 23.8
75 8.4 20.1 21.7 25.4
50 10.9 20.8 22 26.1
25 19.1 23.6 23.4 26.2
% PA 100 15.2 16.6 17.1 19.8
75 15.3 16.8 17.2 18.0
50 15.4 16.8 17.9 20.3
25 18.5 21.8 24.8 25.3
% BAK 100 10.3 14 17.6 20.6
75 11.9 15.6 18.2 22.6
50 12.3 16 19.6 21
25 12.7 16.3 19.8 21.1
% EC 100 13.2 14.2 16.1 19.6
75 14.3 15.2 16.2 20.3
50 14.4 16.1 16.8 18.8
25 17.2 18.9 20.8 21.4
%O 100 26.9 33.2 38.5 42.2

Fig. 11. Effect of eco-aggregates content on concrete 28-day flexural strength. Fig. 13. Effect of type and quantity of eco-aggregates on the modulus of elasticity of
concretes at 28 days.

Fig. 12. Surface failures of the eco-concretes PA and C8 under flexural test.

keq h
¼1þ   ð12Þ
ksl ka
3þ kpG 1 d
1h
3
þ    
2kpG

ka ka
3 ksl 1 d kpG 1 ksl þ1

where h and d are the volume fractions of carbonated aggregates

and cement respectively. Fig. 14. Modulus of elasticity as function of the compressive strength.
248 I. Rahmouni et al. / Construction and Building Materials 197 (2019) 241–250

Fig. 18. Predicted and measured values of thermal conductivity for all studied eco-
Fig. 15. Thermal conductivity of all studied concretes as a function of temperature.
concrete mixes.

Fig. 19. Effect of aggregates on water absorption of concretes.

Fig. 16. Effect of eco-aggregates percentage on concretes’ thermal conductivity.

Fig. 17. Modelling by auto-coherent homogenization of three-phases. Fig. 20. Effect of eco-aggregates on sorptivity coefficient of all studied concretes.
 I. Rahmouni et al. / Construction and Building Materials 197 (2019) 241–250 249

The experimental results are plotted as a function of the pre- Acknowledgements

dicted values and are presented in Fig. 18. The model under pre-
dicts the thermal conductivities with standard error of 0.3%. The authors acknowledge the University of Greenwich and Car-
bon8Systems for their partnership and for supplying eco-
aggregates for this study.
3.5. Water absorption and sorptivity

Water absorption of all studied concretes was obtained by References

immersion in water until reaching a constant weight. The speci-
mens were oven dried prior to testing. The water absorption was [1] «Indicateurs de développement durable en Champagne-Ardenne: La
production des granulats,» Insee dossier n°34, p. 21/22, Octobre 2011.
found to linearly increase with an increase in eco-aggregates con- [2] Commisariat général au dévéloppement durable-service de l’observation et
tent (Fig. 19). By using carbonated aggregates, which are more por- des statistiques, Impact de l’épuisement des ressources naturelles sur les
ous than ordinary ones, water absorption is important. agrégats économiques, Etudes & documents, 2011, N°56.
[3] J.A. Bogas, J. de Brito, J.M. Figueiredo, Mechanical characterization of concrete
Fig. 20 shows the sorptivity coefficient of tall studied concretes produced with recycled lightweight expanded clay aggregate concrete, J.
as a function of the percentage of eco-aggregates. The capillary Clean. Prod. (2014) 1–9.
absorption coefficients Ac correspond to the coefficients governing [4] N.K. Bairagi, K. Ravande, V.K. Pareek, Behaviour of concrete with different
proportions of natural and recycled aggregates, Resour. Conserv. Recycl. 9
the lines of evolution of the absorption coefficient as a function of (1993) 109–126.
the square root of time. [5] L. Evangelista, J. de Brito, Mechanical behaviour of concrete made with fine
The capillarity sorption increases with an increase in the per- recycled concrete aggregates, Cem. Concr. Compos. 29 (2007) 397–401.
[6] C.S. Poon, S.C. Kou, L. Lam, Use of recycled aggregates in moulded concrete
centage of eco-aggregates in the concrete mix. The porous aggre-
bricks and blocks, Construct. Build. Mater. 16 (2002) 281–289.
gates clearly affect the sorptivity of concretes. [7] A. Rao, K.N. Jha, S. Misra, Use of aggregates from recycled construction and
demolition waste in concrete, Resour., Conserv. Recycl. 50 (2007) 71–81.
[8] O. Kayali, Fly ash lightweight aggregates in high performance concrete,
Construct. Build. Mater. 22 (2008) 2393–2399.
4. Conclusions [9] V. Radonjanin, M. Malešev, S. Marinković, A.E. Saed Al Malty, Green recycled
aggregate concrete, Construct. Build.Mater. 47 (2013) 1503–1511.
[10] A.D. Tran Le, C. Maalouf, T.H. Mai, E. Wurtz, F. Collet, Transient hygrothermal
In this study, new types of eco-concrete were produced. Car-
behaviour of a hemp concrete building envelope, Energy Build. 42 (2010)
bonated aggregates show thermal conductivities similar to that 1797–1806.
of expanded clay. Concretes produced by these aggregates show [11] M. Rahim, O. Douzane, A.D. Tran Le, G. Promis, B. Laidoudi, A. Crigny, et al.,
good thermal conductivities, low densities and sufficient mechan- Characterization of flax lime and hemp lime concretes: hygric properties and
moisture buffer capacity, Energy Build. 88 (2015) 91–99.
ical resistance for certain civil engineering applications. [12] M. Rahim, O. Douzane, A.D. Tran Le, G. Promisand, T. Langlet, Characterization
In fact, this work focused on the effect of carbonated aggregates and comparison of hygric properties of rape straw concrete and hemp
on the mechanical properties and thermal conductivities of con- concrete, Construct. Build. Mater. 102 (2016) 679–687.
[13] B. González-Corrochano, J. Alonso-Azcárate, M. Rodas, Characterization of
cretes made with different percentages of these aggregates. Results lightweight aggregates manufactured from washing aggregate sludge and fly
indicate that compressive strength and modulus of elasticity of ash, Resour. Conservat. Recycl. 53 (2009) 571–581.
eco-concretes are reduced due to the low crushing resistance of [14] M. Behera, S. Bhattacharyya, A. Minocha, R. Deoliya, S. Maiti, Recycled
aggregate from C&D waste & its use in concrete – a breakthrough towards
eco-aggregates as compared to ordinary ones. The compressive sustainability in construction sector: a review, Construct. Build. Mater. (2014).
strength is reduced by as much as 50% for concretes with 100% [15] H. Motz, J. Geiseler, Products of steel slags an opportunity to save natural
eco-aggregates. At the same water to cement ratio, concretes made resources, Waste Manage. 21 (2001) 285–293.
[16] C.D. Hills, P.J. Carey, Production of Secondary Aggregates, United States Patent
with 100% eco-aggregates are less resistant than ordinary concrete.
Application Publication, 2009.
The best performance in compressive strength and modulus of [17] P.J. Gunning, C.D. Hills, P.J. Carey, Production of lightweight aggregate from
elasticity was achieved up to 25% of carbonated aggregates. As industrial waste and carbon dioxide, Waste Manage. 29 (2009) 2722–2728.
[18] P.J. Gunning, C.D. Hills, P.J. Carey, Accelerated carbonation treatment of
the substitution percentage increases, the modulus of elasticity
industrial wastes, Waste Manage. 30 (2010) 1081–1090.
decreases due to the high porosity of the aggregates. [19] A. Antemir, C. Hills, P. Carey, P. Gunning, Secondary aggregate from waste
Both water absorption and sorptivity increases as the amount of treated with carbon dioxide, ICE Proc., Construct. Mater. 164 (2011) 231–239.
carbonated aggregates increase. The carbonated concretes are [20] P. Gunning, C. Hills, A. Antemir, P. Carey, Novel approaches to the valorization
of ashes using aggregation by carbonation, in: 2nd International. Slag
more porous than ordinary concrete with porosity up to 12% for Valorisation Symposium, Leuven, Belgium, 2011, pp. 103–116.
PAC as compared to 3% for ordinary concrete. It is also shown that [21] M.F. Bertos, S.J.R. Simons, C.D. Hills, P.J. Carey, A review of accelerated
there is a reduction in thermal conductivities up to 50% by replac- carbonation technology in the treatment of cement-based materials and
sequestration of CO2, J. Hazard. Mater. 112 (2004) 193–205.
ing natural aggregates by carbonated aggregates. The results indi- [22] M.F. Bertos, X. Li, S.J.R. Simons, C.D. Hills, P.J. Carey, Investigation of accelerated
cate a strong relationship between voids present in eco-aggregates carbonation for the stabilisation of MSW incinerator ashes and the
and thermal conductivity. sequestration of CO2, Green Chem. 8 (2004) 428–436.
[23] A. Bourdot, G. Promis, A.D. Tran Le, O. Douzane, A. Benazzouk, F. Rosquoët, T.
Thermal conductivities were estimated using the auto-coherent Langlet, Hygrothermal properties of blocks based on eco-aggregates:
mixture theory model. The model was found to under predict ther- experimental and numerical study, Construct. Build. Mater. 279 (2013) 289–
mal conductivity with standard error of 0.3%. 1125.
[24] AFNOR NF EN 933-1, Tests for geometrical properties of aggregates - Part 1:
The results of thermal conductivities and mechanical properties
determination of particle size distribution - Sieving method, 2012.
of carbonated concretes, encourage their usage in the construction [25] AFNOR NF EN 1097-3, Tests for mechanical and physical properties of
field. These eco-concretes could be used for the production of aggregates - Part 3: determination of loose bulk density and voids, 2004.
[26] AFNOR NF EN 1097-6, Tests for Mechanical and Physical Properties of
masonry blocks.
Aggregates - Part 6: Determination of Particle Density and Water
Absorption, 2014.
[27] AFNOR NF 13055-1, Lightweight aggregates - Part 1: lightweight aggregates
for concrete, mortar, and grout, 2002.
Conflict of interest
[28] Safiullah Omary, Elhem Ghorbel, George Wardeh, Relationships between
recycled concrete aggregates characteristics and recycled aggregates concretes
We certify that we have no affiliation with or involvement in properties, Construct. Build. Mater. 108 (2016) 163–174.
any organization or entity with any financial interest or non- [29] AFNOR NF EN 12664, Thermal Performance of Building Materials and Products
Determination of Thermal Resistance by Means of Guarded Hot Plate and Heat
financial interest in the subject matter discussed in this Flow Meter Methods—Dry and Moist Products of Medium and Low Thermal
manuscript. Resistance, 2001.
250 I. Rahmouni et al. / Construction and Building Materials 197 (2019) 241–250

[30] A. Benazzouk, O. Douzane, K. Mezreb, B. Laidoudi, M. Quéneudec, Thermal [40] H. Oktay, R. Yumrutas, A. Akpolat, Mechanical and thermophysical properties
conductivity of cement composites containing rubber waste particles: of lightweight aggregate concretes, Construct. Build. Mater. 96 (2015) 217–
experimental study and modelling, Construct. Build. Mater. 22 (2008) 573– 255.
579. [41] T.Y. Lo, W.C. Tang, H.Z. Cui, The effects of aggregate properties on lightweight
[31] EN 197-1, Cement-Part1: composition, specifications and conformity criteria concrete, Build. Environ. 42 (2007) 3025–3029.
for common cements, 2001. [42] K. GaneshBabu, D. SaradhiBabu, Behaviour of lightweight expanded
[32] M. Arnould, M. Virlogeux, Aggregates and Lightweight Concretes, Presse de polystyrene concrete containing silica fume, Cem. Concr. Res. 33 (2003)
l’ENPC, 1986. 755–762.
[33] T. Holm, J. Ries, High-performance Structural Lightweight Concrete, American [43] Eurocode 2: Design of concrete structures part 1-1: General rules and rules for
Concrete Institute, 2004, p. 202. buildings, EN 1992-1-1, Nederlands Normalisatie institute, 2007.
[34] J. Baron, J.P. Olivier, La Durabilité des bétons, Presses de l’École Nationale des [44] ACI-318-08, Building Code Requirements for Structural Concrete, American
Ponts et Chaussées-Paris, 1992, p. 453. Concrete Institute, Farmington Hills MI, 2008.
[35] RILEM, Functional classification of lightweight concrete, Mater. Struct. 5 (27) [45] L.H. Nguyen, A.L. Beaucour, S. Ortola, A. Noumowé, Influence of the volume
(1972) 173–175. fraction and the nature of fine lightweight aggregates on the thermal and
[36] EN 206-1, Concrete-Part1: specification, performance, production and mechanical properties of structural concrete, Construct. Build. Mater. 51
conformity, 2000. (2014) 121–132.
[37] AFNOR NF EN 12390, Testing Hardened Concrete, 2012. [46] L. Arnaud, H. Monnet, C. Cordier, F. Sallet, Modélisation par homogénéisation
[38] Safiullah Omary, Elhem Ghorbel, George Wardeh, Minh Duc Nguyen, Mix auto cohérente de la conductivité thermique du béton et laines de chanvre, in:
design and recycled aggregates effects on the concrete’s properties, Int. J. Civ. A. Lallemand, J.F. Leone (Eds.), Proceedings of the Congrès français de
Eng. (2017). thermique, 15–17 Mai 2000, Elsevier, 2000, pp. 543–548.
[39] M.H. Zhang, O.E. Gjorv, Mechanical properties of high strength lightweight
concrete, ACI Mater. J. 88 (3) (1991) 240–247.