Journal of Building Engineering 78 (2023) 107550

Contents lists available at ScienceDirect

Journal of Building Engineering

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

Effects of incorporating silica glass powder as a supplementary

cementitious material on selected properties of portland
cement mortar
Zine el-Abidine Kameche a, *, Mohammed Djelil a, Benamar Dahmani b
a
Smart Structures Laboratory (SSL), Department of Civil Engineering and Publics Works, Faculty of Sciences and Technology, University of Ain-
Témouchent, P.B. 284, 46000, Algeria
b
Spectrochemistry and Structural Pharmacology Laboratory (SSPL), Department of Chemistry, Faculty of Sciences, University of Tlemcen, P.B. 119,
13000, Algeria

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

Keywords: The use of waste glass powder as a partial replacement for cement is an emerging practice for
Cement mortar sustainable cement-based construction. Several studies have demonstrated that finely ground-
SCM powdered glass could be an interesting supplementary cementitious material (SCM). However,
Waste glass the performance of such material is influenced by the chemical composition of glass and its
Silica-glass powder pozzolanic reactivity. In this context, the goal of this study is to investigate the use of silica-glass
Sorptivity
powder (SGP), which is characterized by a high content of silica (>96%), as a partial replacement
Chloride ions diffusion
of Portland cement in mortar. In this research work, mortars containing 7.5%, 10, 12.5, 15 and
20% SGP in replacement of an amount of cement, with three different water-to-binder ratios of
0.55, 0.60 and 0.65 are studied. Selected properties have been investigated at both fresh and
hardened stages, which include workability, air content, flexural and compressive strengths
estimated by destructive and non-destructive methods, water absorption by capillarity, perme­
ability to chlorides and chloride ions diffusion. The experimental results confirm that incorpo­
rating SGP improves the mechanical strength of the mortar, notably in the long-term, and can
considerably enhance its durability properties, especially by a decrease in sorptivity of the mortar
and by an increase of its resistance to chloride ions diffusion. Correlations between selected
physical and mechanical properties and certain durability indices are also established and dis­
cussed in this study.

CRedit autorship contribution statement

Z.E.A. Kameche, M. Djelil and B. Dahmani, contributed to the design, data collection and analysis, manuscript drafting and final
approval of the manuscript.

1. Introduction
Cement-based materials are the most consumed building material. This is mainly due to the availability of their constituents and
their good engineering properties in both fresh and hardened states. Portland cement is a core element that is used as a primary binder

* Corresponding author.
E-mail address: [email protected] (Z.e.-A. Kameche).

https://doi.org/10.1016/j.jobe.2023.107550
Received 13 May 2023; Received in revised form 31 July 2023; Accepted 12 August 2023
Available online 19 August 2023
2352-7102/© 2023 Elsevier Ltd. All rights reserved.
Z.e.-A. Kameche et al. Journal of Building Engineering 78 (2023) 107550

in the production of conventional cementitious material. Producing Portland cement is energy-intensive as it consumes 110–220 kWh
of electricity for one tonne of cement [1]. In addition, it plays an essential role in the emission of carbon dioxide as it releases 0.64
tonne of CO2 per one tonne of cement [2]. In 2019 alone, the cement industry emitted about 1.5 billion tonnes of CO2 into the at­
mosphere for 4 billion tonnes of produced worldwide Portland cement [3,4]. It is worth mentioning that the production of Portland
cement involves a huge amount of natural raw materials (e.g., coal, limestone and clay) which is extremely harmful to the environment
[5]. Minimizing the quantity of Portland cement required for the manufacture of cementitious materials (mortars or concretes) is a
promising option for sustainable cement-based construction because it offers many environmental benefits, such as saving raw ma­
terials and reducing both the carbon dioxide emissions and energy demand in cement industry [4].
There is an increased focus in recent years on alternative binder cementitious materials that are derived from waste materials. This
process provides an additional environmental benefit that decreases the demand for valuable landfill areas, thereby ensuring envi­
ronmental sustainability [6–8]. The waste recycled glass material is one of the most promising waste materials that can be used as an
alternative for traditionally used supplementary cementitious materials, like fly ash, silica fume, slag …, etc.) [9–11]. Recycled waste
glass fulfils the basic requirements for pozzolanic materials, especially the presence of a higher amount of silica [9]. Khan et al. [5]
indicated that recycled glasses can be reused in mortar and concrete in – mainly – three different ways: (i) raw materials for cement
production, (ii) partial or full replacement of aggregate, and (iii) partial replacement of cement. The use of waste glass as a raw
material for cement production is permitted, but only for a very low percentage of glass. Limited reuse of waste glass (less than 5%) as
raw material is essential to avoid the contamination of clinker by the sodium oxide (Na2O) present in the glass but renderssuch a
process less attractive [12,13]. Moreover, the glass cannot be used as a partial or full replacement of aggregate without considering
related expansion and cracking caused by alkali-silica reaction (ASR). The expansion due to ASR is also observed even if the fine glass
aggregate is used as part of the full replacement of natural fine aggregate in cement-based mortar and concrete [14]. Several studies
have demonstrated that there is no expansion occurred in mortar specimens when the particle size of glass powder is smaller than 75
μm [15–17]. Therefore, the use of glass powder as cement replacement and/or as cement addition is considered as the most effective
way for sustainable reuse of waste glass in cement-based construction materials [6,11]. Fine glass powder has also been used as a
partial substitution of cement and silica fume in ultra-high-performance concrete (see for instance Refs. [18–20]) and as a precursor in
alkali-activated materials (see for instance Ref. [21]).

1.1. Background
The performance of Portland cement-based materials made with waste glass powder is particularly influenced by the glass particle
size, the colour of the glass, and its chemical composition [22,23]. Glasses with different colours have different chemical compositions
[22]. So, the effect of the colour glass can be related to the chemical composition of colouring elements that may affect the pozzolanic
reaction of glasses [6]. Several studies have assessed the effect of glass powders on the plastic properties of concrete [24–30] or mortar
[31–36]. Although there is no established consensus on their effects on the workability of concrete or mortar, it is well established that
particle size and morphology of the ground glass can have an important effect [14]. The general trend observed was that the higher the
particle finenesses, the more positive effect of the glass powder on the workability [37]. Moreover, several research works have studied
the effect of grounded waste glass on air content in fresh cementitious materials. Some studies report that the air content is not affected
by the incorporation of glass into concrete [25,36], while others conclude that air content is increased [26]. Afshinnia and Rangaraju
[26] have indicated that this trend can be attributed to the effects of elongated glass particles that entrapped air which result in higher
air content. In terms of hardened properties, the results from the literature regarding the effect of replacing cement with waste glass
powder on the compressive strength of mortar and concrete are inconclusive [11]. Du and Tan [38] have found that the incorporation
of 15% of soda-lime glass as a cement replacement in concrete mixtures resulted in an increase in compressive strength of 32%, 25%
and 24% compared respectively with the control mixes at 7, 28 and 90 days. Gignozzi et al. [23] compared the compressive strength of
mortar containing 25% of GP (in replacement of cement) originally made with soda-lime glass, glass coming from fluorescent lamps,
and funnel glass from cathode ray tube and crystal items. They found that all mortar samples showedlower compressive strengths
compared to the control mixes, except for mortar made with funnel glass which achieved a compressive strength higher than the
control mix after 90 days of curing. While, another study concluded that the use of 20–30% of glass powder, prepared from scrap glass
collected from shops of glass, as fine aggregate or binder in concrete will not cause any detrimental effect on the compressive strength
[39]. Almeshal et al. [40] have studied the mechanical performance of eco-friendly cement-based glass powder mortar exposed to an
aggressive environment (i.e. NH4NO3 solution). They found that the compressive strength of modified mortars containing 10% of glass
powder was improved. While modified mortars containing 20% of glass powder as cement replacement showed similar performance in
terms of residual strength compared to control mixes. On the other hand, the available research works related to the influence of glass
powder on the flexural strength of cementitious materials are relatively limited [22]. Despite this, some studies revealed that flexural
strength shows a similar tendency to compressive strength [41–43]. In terms of durability properties, the use of waste glass powder as
cement replacement may affectthe water capillary absorption of cementitious materials, which is an important indicator of durability,
especially when the material is in aggressive environments. It was found that the pozzolanic reactivity of glass powder induces
decreasing water absorption which may lead to improved durability characteristics of concrete in the long term [8,25,29,41,44,45].
Carsana et al. [46] have compared the performance of ground waste glass in terms of durability tests with other supplementary
cementitious materials, such as natural pozzolana, fly ash, and silica fume. They found that ground glass improved the resistance to
chloride penetration, and the resistance to sulphate attack of mortars more than natural pozzolana and similarly to fly ash. The mortar
containing glass powder showed inferior performance only when compared to ordinary Portland cement and silica fume. Tayeh et al.
[47] found that the incorporation of 10% of glass powder as cement replacement enhanced the sulphate resistance compared to control
mix. In addition, Wang et al. [48] indicated that the incorporation of finely crushed waste liquid crystal glass particles in concrete

2
Z.e.-A. Kameche et al. Journal of Building Engineering 78 (2023) 107550

mixtures resulted in a decrease in chloride ions penetration. The general trend observed was found to be proportional to the amount of
glass content, the higher the dosage of glass powder, the lower the chloride ions penetration. The same tendency was observed by Chen
et al. [49] where waste E-glass particles are used in cementitious mixtures. In terms of frost resistance, chloride ions penetrability and
chloride ions diffusion, the investigation conducted by Lee et al. [50] concluded that the durability properties of concrete would
appreciably impoved when glass powder and glass sludge are used in concrete. In contrast, De Castro and De Brito [51] have observed a
slight increase in chloride ions penetration when replacing fine aggregate in concrete by crushed waste glass.

1.2. Research significance

Although it has already been proved that Portland cement-based material made with glass powder (GP) is a promising building
material, additional studies are still required to investigate the appropriateness of each type of glass utilisation as an SCM. As stated
previously, this is of particular interest as the glass’s chemical composition plays a fundamental role in the development of pozzolanic
reactions [23]. Furthermore, from the literature review, it is noted that the random use of glass particles in cement-based materials
does not necessarily give a satisfactory or optimised performance [6]. This may also explain the fact that, to date, there is no consensus
in the literature on the influence of recycled waste glass on the hardened and durability properties of cement-based materials [14,22].
Recognising the fact that the majority of the literature covered soda-lime-silica glass or do not explicitly mention the chemical
composition of the waste glass tested [11,52], there is still a need to improve the understanding of the effects of different types of glass
that have different chemical compositions on the performance of modified cement-based materials.

1.3. Objectives
The objective of this study is to evaluate the properties of mortars containing silica-rich glass powder (SGP) as a partial replacement
of Portland cement. Selected properties are investigated at both fresh and hardened stages. This type of glass is widely used in (i)
various thermally stressed light bulbs, (ii) tubing for the production of semiconductors and gas lasers, (iii) optical instruments and
apparatus, and (iv) for high-temperature industrial components such as glass used in laboratory glassware (e.g., crucibles, beakers,
evaporators, …, etc.) [53]. The silica glass is characterized by a high content of silica (in this study SiO2 > 96%) which makes it highly
resistant to thermal shock (very rapid heating and cooling without cracking). To the best of the authors’ knowledge, no such study has
been conducted on this type of glass, i.e. silica-rich glass.

2. Methodology
Several studies have demonstrated that the use of finely grounded glass as a cement replacement material is the most effective way
for the sustainable reuse of waste glass in cement-based materials [6,11]. Thus, the recycled silica-richglass will be used as a partial
replacement of Portland cement in different ordinary mortars tested in this study. The collected waste glass was crushed and then
finely ground in a ball mill to ensure that the glass powder is homogenous. Previous studies indicated that glass powder with a particle
size of 0–75 μm can be recognised as a reactive pozzolan and thus can be used as an SCM in cement-based materials [10,54]. Hence, the
maximum size of the recycled glass powder adopted in this study was 75 μm. Furthermore, it is well established that the major
challenges in using recycled materials in cement-based materials are related to the amount of impurities, amount of cement to be
replaced, uncontrollable water content for desired workability and appropriate activation of inert pozzolanic materials [7,8,55].
Therefore, prismatic and cylindrical samples were made with different percentages of cement replacement (i.e., 0, 7.5, 10, 12.5, 15 and
20%, by weight of cement) and considering different water-to-binder ratios (i.e., w/b = 0.55, 0.6 and 0.65), making a total of 18
mortar mixtures.
Moreover, and as stated previsiouly in section 1.1, the performance of Portland cement-based materials made with waste glass
powder is particularly influenced by the glass particle size and its chemical composition. Thus, a series of laboratory tests were
conducted to assess the properties of mortars containing various amounts of SGP. In this study, the plastic properties of mortar
evaluated include the workability (flow test) and air content. The hardened properties include (i) mechanical properties (the flexural
and compressive strengths), (ii) ultrasonic properties (UPV) and (iii) durability properties (the water absorption, permeability to
chloride and chloride ions diffusion). Pozzolanic strength activity index (SAI) was also calculated in this study for all the modified
mortars at different ages. Furthermore, a comparison between experimentally measured compressive strength and predicted
compressive strength using the modulus of elasticity calculated from the UPV values is discussed. Indeed, three correlations are
assessed and discussed for all mortars with different w/b ratios. The first correlation is between sorptivity coefficients and compressive
strengths. The second correlation is between the sorptivity coefficient and that one of chloride ions diffusion, while the third one is
between the air content (air trapped) in fresh mortar and the coefficient of chloride ions diffusion. The optimum content of silica
ground glass powder used as a Portland cement replacement will also be identified. In this study, ASR is not considered, since several
previous research works have demonstrated that once the glass powder is sufficiently fine and the glass particles are with a diameter of
less than 75 μm, no deleterious alkali-silica reaction would occur [15–17].

3. Materials
3.1. Cementitious materials
In this study, CEM I 42.5 N Portland cement meeting the requirement of EN 196-6 was used. The Blaine-specific surface area and the
density of the cement were found to be equal to 3320 cm2/g and 3.11 g/cm3, respectively. The fine glass powder has a density equal to
2.97 g/cm3 and a Blaine-specific surface equal to 4280 cm2/g. The particle size distribution of materials was determined using a laser

3
Z.e.-A. Kameche et al. Journal of Building Engineering 78 (2023) 107550

particle analyser. The particle size distribution was carried out for the recycled waste GP, the cement used and, in order to do a
comparison, for a silica fume manufactured by Sika Company. The particle size distribution results, illustrated in Fig. 1, indicate that
the maximum diameter of SGP particles is equal to 75 μm. It is also observed that 50% of GP particles have diameters ≤21.92 μm when
the mean diameter of silica fume (D50%) is around 0.35 μm only. As seen in Fig. 1, although the maximum diameter of the cement
particles is equal to 125 μm, the particle size distribution of SGP is considered closer to that of the cement, compared to that of the silica
fume which is with particles that have a dimension of 10 μm as maximum diameter. It is clearly shown in Fig. 1 that more than 80% of
silica fume particles are with nanometric level (<1 μm), while only 4.5% of GP particles, tested in this work, are with nanometric
dimensions.
Chemical compositions of Portland cement and recycled silica glass powder were studied using X-ray fluorescence spectrography
(XRF). The results, expressed as percentage by mass of oxides, are given in Table 1. As seen in Table 1, the recycled glass powder
obtained is mainly composed of silica (>96%). This recycled glass with a high SiO2 content is therefore comparable to silica fume
which presents a percentage of SiO2 > 90%. So, the studied glass can be classified as silica glass (SG) content. Moreover, the equivalent
reactive component (SiO2 + Al2O3 + Fe2O3 = 97.48% (see Table 1)) is much higher than 70% considered by the standard as the
minimum limit for a Pozzolanic activity of material [56]. It is important to point out that there is no specific standard that covers glass
powder material. Thus, in general, glass powder properties are compared to requirements for fly ash (e.g., EN 450-1), silica fume (e.g.,
EN 13263-1), and slag (e.g., EN 15617-1) [31]. Moreover, Na2 Oeq content is far less the limit of 5% imposed in the standard of fly ash.
This optional requirement for the alkali content is of particular importance due to potential alkali silica reaction (ASR) related
concerns.
For a more complete characterisation of the silica glass powder used in this study, and in particular, to determine its amorphous
phase, X-ray diffraction (XRD) and Fourier Transform Infrared Spectroscopy (FT-IR) tests were carried out. X-ray powder diffraction
analysis was made using a Bruker D8 Advance diffractometer with a fine long focus CuKα1 anode tube (WL = 1.54060 Å) operated at
40 kV and 40 mA over therange 5–70◦ (2θ) (2 Theta/Theta coupled), step size of 0.02◦ and scan rate of 0.5 s/step. X-ray diffraction
pattern for the recycled waste glass powder is shown in Fig. 2. It is clear that the glass powder studied herein contains large amounts of
SiO2 and it is amorphous since no clear crystalline peaks could be found (only with a broad peak at 22◦ (2θ)) which represents
amorphous silica. This confirmed that when glass is finely ground, it may be used as an SCM’s since it is essentially made by amorphous
silica [57].
From the literature, amorphous forms of silica, such as silica fume have also been investigated as SCM’s. In order to study the effect
of the silica particle size on the formation of C–S–H, Riccardo et al. [58] have used two different silica sources: silica fume and
nano-silica with particle size range 5–20 nm. XRD analysis, performed on these amorphous forms of silica, indicate X-ray diffraction
patterns similar to that obtained for the silica glass powder tested in this study, where a peak is formed between 10◦ and 30◦ (2θ) and
with a maximum in intensity corresponding to 22◦ (2θ).
Fourier Transform Infrared Spectroscopy (FT-IR) absorbance patterns of the silica glass powder was made using an Agilent Res­
olution Pro. The FT-IR spectrum of SiO2 is given in Fig. 3 which consists of bands at 1380, 1029, 916 and 798 cm− 1. The peak at 1029
cm− 1 of glass powder is the absorption peak of Si–O–Si groups and Si–O vibration modes of isolated Si–OH groups respectively. Similar
bands were observed by Cheng et al. [57]. The peak at 791 cm− 1 may be assigned to the O–Si–O vibration mode of SiO2. The IR
spectrum of SiO2, thus demonstrates the presence of all the vibrational bands of O–Si–O and Si–O–Si typical of amorphous SiO2. The
band observed at 1380 cm− 1 may be attributed to the presence of CO2− 3 .

3.2. Aggregates material

Two types of non-reactive aggregates are selected: one fine sand (0–1 mm) which is siliceous in nature and the other which is coarse
sand (0–4 mm) and limestone in nature. This combined sand has a maximum size of 4 mm, density of 2620 kg/m3, fineness modulus of
2.58, water absorption of 0.81% and sand equivalent equal to 80%.

Fig. 1. Laser granulometry curves of silica fume, silica glass powder and cement.

4
Z.e.-A. Kameche et al. Journal of Building Engineering 78 (2023) 107550

Table 1
Chemical compositions of Portland cement and glass powder, obtained from the XRF test.

Oxide (%) SiO2 MgO SO3 K2O Na2O Al2O3 Fe2O3 CaO LOI

CEM I 42.5 N 19.28 2.03 2.47 0.53 0.22 3.91 2.62 64.9 7.35
Glass powder 96.91 0.07 0.05 0.11 1.08 0.50 0.07 1.17 0.21

Fig. 2. X-ray diffraction spectrum of silica glass powder.

Fig. 3. FT-IR absorbance patterns of the silica glass powder.

3.3. Mixtures proportioning

In this paper, all mortars were prepared according to the European standard EN 196-1 [59] which describes the procedure of
mixing and casting used for the preparation of mortars. So, mortars were cast in prismatic moulds ((4 x 4 x 16) cm3) and the test
specimens were demolded after 24 h and were kept in the curing room (in tap water at 20 ± 2 ◦ C) up to the testing day. Table 2
summarises the relative proportions of the constituents used in the mortar mixtures wherein glass powder was used as a substituent for
an amount of cement in mortar mixtures prepared with non-reactive aggregate. The experimental investigations were carried out by
considering three series of mortars. The mortar series refers to the water-to-binder ratio (w/b) used in mortar mixtures. The series M1,
series M2 and series M3 were produced using a water-to-binder ratio of 0.55, 0.60 and 0.65%, respectively. Moreover, five silica-glass
powder (SGP) dosages (7.5%, 10, 12.5, 15 and 20% (by weight of cement)) are considered for each mortar series as a partial
replacement of Portland cement. Mortars without any SGP were prepared as reference mixtures and named M1-CM, M2-CM and
M3-CM for mixtures prepared with a w/b ratio of 0.55, 0.60, and 0.65, respectively. Whereas, mortars with SGP were named
MX-SGPY. X was used to identify the w/b ratio, while Y was used to indicate the percentage of the glass powder used as SCM (e.g.
M1-SGP7 identifies mortar specimen with a w/b ratio of 0.55 and containing 92.5% of CEM I + 7.5% of SGP as a binder).

4. Test procedures
4.1. Workability
In this study, the flow test was used to measure the workability of mortar according to NF P18-452 [60] and NF P15-437 [61]. In

5
Z.e.-A. Kameche et al. Journal of Building Engineering 78 (2023) 107550

Table 2
Mixtures proportions of all mortar series mixtures.

N◦ Series Mixture ID Water to binder ratio binder Aggregates Water

(w/b) (g)
Portland Cement Silica-Glass Siliceous sand Limestone sand
(g) powder (g) (g) (g)
(40%) (60%)

1 M1 CM (100% C) 0.55 450 0 540 810 247,5

2 SGP7 (92.5% C+7.5% 0.55 416.25 33.75 540 810 247,5
SGP)
3 SGP10 (90% C+10% 0.55 405 45 540 810 247,5
SGP)
4 SGP12 (87.5% 0.55 393.75 56.25 540 810 247,5
C+12.5% SGP)
5 SGP15 (85% C+15% 0.55 382.5 67.5 540 810 247,5
SGP)
6 SGP20 (80% C+20% 0.55 360 90 540 810 247,5
SGP)
7 M2 CM (100% C) 0.60 450 0 540 810 270
8 SGP7 (92.5% C+7.5% 0.60 416.25 33.75 540 810 270
SGP)
9 SGP10 (90% C+10% 0.60 405 45 540 810 270
SGP)
10 SGP12 (87.5% 0.60 393.75 56.25 540 810 270
C+12.5% SGP)
11 SGP15 (85% C+15% 0.60 382.5 67.5 540 810 270
SGP)
12 SGP20 (80% C+20% 0.60 360 90 540 810 270
SGP)
13 M3 CM (100% C) 0.65 450 0 540 810 292,5
14 SGP7 (92.5% C+7.5% 0.65 416.25 33.75 540 810 292,5
SGP)
15 SGP10 (90% C+10% 0.65 405 45 540 810 292,5
SGP)
16 SGP12 (87.5% 0.65 393.75 56.25 540 810 292,5
C+12.5% SGP)
17 SGP15 (85% C+15% 0.65 382.5 67.5 540 810 292,5
SGP)
18 SGP20 (80% C+20% 0.65 360 90 540 810 292,5
SGP)

this test method, the fluidity of a mortar is appreciated according to the time required for the mortar to reach a uniform distribution in
the receivers, which describes how freshly mixed mortar can be easily placed, consolidated, and finished. This method consists in
pouring the fresh mortar in a prismatic receiver that is divided into two unequal volumes by a removable partition. Once the separating
partition is removed vibrator starts automatically and the test consists in measuring the time necessary for a mortar to flow and reach a
mark engraved on the inner face of the mold under vibration. The effect of the glass powder dosage on the workability of mortar was
assessed through the comparison of the workability of mortar mixtures containing SGP with the workability of control mortars
(without any SGP).

Fig. 4. Device for measuring ultrasonic pulse velocity (UPV) in mortar specimens.

6
Z.e.-A. Kameche et al. Journal of Building Engineering 78 (2023) 107550

4.2. Air content

In this paper, air content test was used to evaluate the effect of SGP dosage on the amount of air introduced into the tested mortars
during the manufacturing process. Air content tests were conducted on fresh mortar mixturesand measured using the aerometer ac­
cording to NF EN 413-2 [62]. This parameter is of particular importance as it indicates the presence of air in the microstructure of
mortar in the hardened state. The air in the mortar microstructure providesspace for the expansion of water to occur without stress.
This is the reason why higher air content is related to a higher resistance to cracking caused by freeze-thaw damage [63], as the
freeze-thaw damage caused by the volume increase of water after freezing can be prevented [14].

4.3. Ultrasonic pulse velocity (UPV) tests and dynamic modulus of elasticity
UPV tests of mortar specimens were conducted using a portable ultrasonic non-destructive digital indicating tester (PUNDIT)
shown in Fig. 4. In this study, the direct transmission technique is chosen as it is considered as the most sensitive transmission
technique among other techniques (e.g. semi-direct or indirect transmission mode). According to EN 12504-4 [64], its 54 kHz
transducers were placed on each opposing end-surfaces of the specimen tested (surfaces of (4 x 4) cm2), and the propagation time of
ultrasound pulses through the length of the (4 x 4 x 16) cm3 mortar specimen was measured with an accuracy of 0.1 μm (see Fig. 4). To
maximise the accuracy of the transit time measurement, the end-surfaces of prismatic mortar specimens were made smooth and flat,
which would provide a good coupling between the transducer face and the specimen surface. Moreover, a thin film of Vaseline was
applied to the surface of both the transmitter and receiver transducers to ensure full contact between the transducers and the
end-surfaces. The ultrasonic pulse velocity (UPV in m/s) was computed by dividing the distance between the transducers (i.e., length of
the specimen in m) by the transit time (s). The length of the mortar specimen was measured with an accuracy of 0.1 mm.
The dynamic modulus of elasticity (Edyn ) was evaluated theoretically from the UPV values using the following equation (Eq. (1)):

ρ(1 + μ).(1 − 2μ) 2

Edyn = V (1)
(1 − μ)

where ρ is the density (kg/m3), μ is the dynamic Poisson’s ratio and herein taken as 0.2, and V is the pulse velocity (m/s) of mortar.

4.4. Mechanical properties

Mechanical resistance (compressive strength and flexural strength) is important data for the design of cementitious material. The
compressive strength results are also used to qualify and quantify the effectiveness of mineral additions replaced by cement in a
cementitious system [65–68].
After UPV tests, the flexural strength (FS) and the uniaxial compressive strength (UCS) were performed on the same specimens
using a computer-controlled mechanical hydraulic press. First, the determination of the flexural strength was carried out by using a
three-point bending test. The distance between the points of support was 100 mm, and the total length of the specimen tested was 160
mm. The compressive strength of the mortar was determined on the specimen halves obtained from bending tests. The obtained test
samples of mortar were subjected to axial compression through steel plates of a square shape and dimension of (4 x 4) cm2. Note that
the load was applied at a uniform rate of the order of 15 N/s, according to EN 1015-11 [69]. All the experiments were carried out in
triplicate and the mean values of UPV, FS and UCS were presented in the results.

4.5. Pozzolanic activity

According to ASTM C311 standard [70], the pozzolanic activity of fine SGP mortars with different w/b ratios is evaluated with the
Strength Activity Index (SAI). This latter is the ratio between the strength of SGP mortar and the strength of corresponding reference
mortar (CM mortar with the same w/b ratio and at the equal age). The Strength Activity Index (SAI in percentage) was calculated
following the ASTM C311 [70] for different SGP mortars tested in this study, using the following equation (Eq. (2)):
SAI = (A / B) • 100 (2)

where Ais the average compressive strength of mortar specimens containing SGP (in MPa), at different curing ages (7, 28, 60 and 90
days), and B is the average compressive strength of control mortar specimens (in MPa) at the equal age to that considered in A.

4.6. Water capillary absorption

Capillary absorption is the tendency of mortar to take up water in the absence of external hydraulic pressure. This is believed to
occur almost entirely in the capillary pores of the material. Liquid water is pulled into these pores by surface tension. Absorption is the
main mechanism by which dry or partially dry mortar takes up water. The smaller the pores, the greater the suction. In this study,
sorptivity was carried out on prismatic test specimens (4 x 4 x 16) cm3 at the age of 90 days, for each mortar tested, according to the
procedure recommended by AFPC-AFREM [71].
The test - itself - consists in placing the specimens, previously dried in an oven (at T = 80 ± 5 ◦ C), face downwards in a shallow
water bath. The water level was adjusted so that the face of the (4 x 4) cm2 was submerged to a depth of approximately 5 mm. During
the test, water was drawn into the core of the mortar specimen by capillarity forces, and weighing of these specimens was carried out at
5 min, 10, 15, 30, 60 and 120 min from the start of the test.
The absorption of water into mortar under capillary force is generally dependent on the square root of time and can be assessed by
using the following expression (Eq. (3)):

7
Z.e.-A. Kameche et al. Journal of Building Engineering 78 (2023) 107550

Sw
Ca = √̅ (3)
t

where Ca is the sorptivity coefficient and Sw is the depth of water penetration (in mm).
The water penetration is measured by the equation given below (Eq. (4)):
M − M0
Sw = (4)
A

where M is the mass of the specimen at a given time (g), M0 is the initial mass of the specimen (g) and A is the section of the specimen
(cm2).

4.7. Permeability to chlorides

When chloride ions penetrated the internal concrete, it causes the corrosion of the steel reinforcement and therefore reduces the
service life of the structure. For this reason, the chloride ions permeability test is widely used to predict the quality of concrete structure
durability life. In this study, the chloride ion permeability test was performed according to the American standard power method
(ASTM C1202 [72]), which is based on the method of the diffusion cell. The test consists of sealing, after water saturation under
vacuum condition, a disk of mortar between the two compartments of the migration cell, and putting it for 6 h under 60 V direct
current to speed up both ends of mortar-disk solution ion transportation, then the power of the whole process was recorded. The
chloride permeability resistance for the reference mortars as well as for the modified mortars, i.e., M1-SGP, M2-SGP and M3-SGP, was
studied after 90 days of wet curing.

4.8. Chlorides ion diffusion

The apparent diffusion coefficient of chloride ionscan be measured directly through a migration test under saturated conditions.
This test was performed according to the AFPC-AFREM (1997) procedure [73]. The mortar discs (110 mm in diameter and 5 cm in
thick extracted from cylinders of 110 × 220 mm in size) were placed between two compartments of the migration cell (placed in the
middle of the sink). To provide a one-way transfer of chlorides (Cl− ) passing through the discs during the diffusivity measurements, the
lateral surface of each disc was coated with two layers of impermeable epoxy resin [74,75]. So, one end of the mortar disk was
immersed in a solution (in the first compartment) of 0.1 mol NaOH and 0.5 mol NaCl, and the other end of the disk in 0.1 mol NaOH (in
the terminal compartment). To accelerate the penetration of chlorides, the mortar disk was put under a potential difference of 30 V
(direct current).
During the test, regular 10 ml samples were used to determine the concentration of chlorides in the downstream solution using the
potentiometric titration technique. The chloride ions diffusion of the reference mortars as well as of the modified mortars, i.e., series
M1, series M2 and series M3, was studied after 90 days of wet curing. The diffusion of chloride ions in mortar is considered stationary
when the speed of movement of ions becomes constant. The passage time of the chloride ions is measured after 180 days when the
steadystate is reached. The effective diffusion coefficient De may be estimated using the following equation (Eq. (5)):
( )
R.T.L
De = •J (5)
z.F.E.C0

where R is the gas constant (8,3144 J/mol− 1. K− 1), T is the thermodynamic temperature (293.15 K), L is the specimen thickness (0.05
1
± 0.001) m, z is the valence of the chloride ion (equal to 1), F is the Faraday constant (96,480 J V− mol− 1), E is the measured electric
field of 30 V, C0 is the concentration in the upstream compartment (0.513 mol/L) and J is the chloride ion flux (mol/m2. s) which may

Fig. 5. Flow test results of the different studied mortars with different dosages of SGP and for different w/b ratios.

8
Z.e.-A. Kameche et al. Journal of Building Engineering 78 (2023) 107550

be calculated using Eq. (6):

(Δc ).V
J= (6)
S.t

where Δc is the concentration difference in the downstream compartment (mol/L) for a time t (s), V is the volume of the downstream
compartment (1.82E-3 m3), and S is the specimen section (78.5E-4 m2).

5. Results and discussion

5.1. Effect of incorporating the SGP on the plastic properties of mortars
The results of the workability of all tested mixtures (control mortars and modified mortars prepared with partial replacement of
cement by silica-glass powder) are presented in Fig. 5. From this figure, it is observed that the time required for the mortar to reach
uniform distribution in the receivers decrease as the dosage of SGP increases. For example, as is shown in Fig. 5, in the case of SGP-
modified mortars with a w/b ratio equal to 0.55, the times decreased by − 6%, − 10, − 20, − 28 and − 44% compared to that of CM
mortar, respectively for SGP substitution level of 7.5%, 10, 12.5, 15 and 20%. The incorporation of the GP tends to increase the
workability of mortar. Patel et al. [76] have indicated that, in the fresh state, the glass powder behaves as a filler and the workability of
mortar is mainly influenced by the particle size and morphology of the glass powder which are different from that of cement. More
precisely, as particle size decreases, the workability of mortar increases. This can be considered as a plausible reason for the observed
contradictory results regarding the effects that glass powder has on workability [31–35]. From the obtained results, it can be concluded
that particle size and Blaine-specific surface area of the glass powder, considered herein, are enough to show a positive effect of waste
glass finely grounded on the workability of the mortars.
Moreover, it is worth mentioning that the considered w/b ratios (i.e., w/b of 0.55, 0.60, and 0.65) showed a similar tendency in
terms of increasing in workability of mortars with the dosage rate of cement replacement. However, the maximum increase in
workability was observed at 20% replacement of cement with SGP, in the case of mortar with a high water-to-binder ratio (i.e., w/b =
0.65). So, more workability improvement was observed in M3-SGP modified mortars as compared to control mortars, M1-SGP and M2-
SGP modified mortars.
On the other hand, Fig. 6 shows the evolution of air entrapped in the fresh mortar mixtures prepared with different SGP dosages. It
is noted that each result presented in Fig. 6 corresponds to the arithmetic mean value of three measurements obtained by three
entrapped air tests performed for each mortar mixture. Overall, the results show that the entrapped air in the mortar at the fresh state
decreases when the w/b ratio increases. Moreover, the air content values are within the limits of the relevant standards (<8%), and the
use of finely powdered glass, with a maximum particle size of 75 μm, results in a lower air content when compared to that obtained in
the case of the reference mortars. It is shown from Fig. 6 that the greater the amount of glass powder in the mortar, the more the
quantity of entrained air introduced into the material decreases. For instance, mixtures containing 20% of SGP as cement replacement,
showed about 38%, 66% and 54% lower air content compared to that of the control mortars, respectively for mixtures with w/b ratios
of 0.55, 0.60 and 0.65 (i.e., M1GP20, M2GP20 and M3GP20). This indicates that SGP-modified mortars may have lower open porosity
(i.e., porosity accessible to water). This means that SGP mortars may be more susceptible to freeze-thaw damage, as a lower content of
entrained air in the material leads to a lower resistance to cracking caused by damage due to the freeze-thaw phenomenon. The
observed trend may be attributed to the fact that the fine particles of the glass powder fill the voids spaces available between the
particles of the mortar thus reducing the air trapped in these voids. This finding agrees with the earlier field investigation by Du and
Tan [77], Parghi and Alam [41] and Jubeh et al. [8] which indicated that the use of fine particles of the glass powder led to better
backing and less retention of air voids.

Fig. 6. Air content in different fresh mortars tested with different dosages of SGP and for different w/b ratios.

9
Z.e.-A. Kameche et al. Journal of Building Engineering 78 (2023) 107550

5.2. Effect of incorporating the SGP on the mechanical properties of mortars

Fig. 7 shows the responses of mechanical tests conducted in this study. Flexural and compressive strengths were determined at the
ages of 7, 28, 60 and 90 days, using (4 x 4 x 16) cm3 prismatic specimens. For each mortar mixture, the given results of the mechanical
strengthare the arithmetic mean values of three measurements obtained from three tests performed on three samples.
Fig. 7 shows that the flexural strength of all tested mortars (CM and SGP modified mortars) increases with the age of the tested
specimens. The same trend is observed in Fig. 7 corresponding to the evolution of compressive strength of mortars (with different w/b

Fig. 7. Flexural and compressive strengths of different series of mortars after different ages of curing: (a) Series M1, (b) Series M2 and (c) Series M3.

10
Z.e.-A. Kameche et al. Journal of Building Engineering 78 (2023) 107550

ratios) according to their ages (at 7, 28, 60 and 90 days). The other remark that is retrieved from the graph is that the evolution that is
related to the age of the material has the same tendency for all mortars tested with different w/b ratios (i.e., w/b = 0.55, 0.6, and 0.65)
and different SGP dosages (i.e., 0, 7.5, 10, 12.5, 15 and 20%).
Moreover, at 7 days of curing (at earlier age), flexural and compressive strengths of all SGP-modified mortars are lower than that of
the CM mortars (without SGP). For instance, modified mortars with a w/b ratio equal to 0.60 and with 15% and 20% of SGP (i.e., M2-
SGP15 and M2-SGP20 mortars) have flexural strengths lower than that of M2-CM by 23% and 26%, respectively. The compressive
strengths are less sensitive to the incorporation of SGP showing a reduction of 5% and 8%, as compared to M2-CM, respectively. This
may be due to the - aforementioned - filler effect of glass powder which takes place at the earlier age of curing. Because of the
dissolution of alkalis from the glass particle, Dyer and Dhir [78] confirmed that the replacement of cement by glass powder leads to an
acceleration in cement hydration. However, the insufficient amount of released alkalis does not make it possible to compensate the
incomplete hydration of cement and the early age strength reduction. Similar results were found by Du and Tan [79] and Schwarz et al.
[80].
At 28 days of hardening, flexural and compressive strengths followed the same trend as a function of SGP dosages, particularly in
the case of mortars prepared with a w/b ratio of 0.55. Whereas, the amount of water in the case of mortars with w/b ratios of 0.60 and
0.65 appears to enhance the pozzolanic reaction, thus leading to a development of the microstructure of the mortar, and consequently,
to higher mechanical strength. Carsana et al. [46] found that the increase in fineness of GP improved significantly the strength of
mortar and confirmed that GP showed the best pozzolanic reactivity in comparison to that of mortars with natural pozzolana and fly
ash. Therefore, when comparing the compressive strengths of the mix at later stages of hardening, the SGP-modified mortars exhibit
compressive strengths greater than or equal to that of the control mix (depending on the SGP-dosage used in mortar). Similar results
have been found in the literature for cement-based materials containing silica fume. The latter, considered as a silica-rich SCM and
which allowed to an improvement in mechanical resistance, a decrease in permeability and thermal conductivity [81–83]. Further­
more, comparing this effect of SGP on the hydration of mortar at early age, with the effect of silica fume, Wei et al. [84] have found,
using Thermogravimetric Analysis (TGA), that silica fume accelerates the C2S, C3A and the C4AF-early hydration and stabilizes
ettringite, resulting in reduced cement and concrete damage caused by delayed ettringite formation.
After the 28th day of curing, the evolutions of the flexural and compressive strengths, for all the mortars containing silica-glass
powder, are greater compared to that of the control mortars (see Fig. 7). Incorporating glass powder as a replacement of cement
leads to a dense microstructure of mortar, and has increased the compressive strength. This conclusion is more noticeable when the w/
b ratio is greater. This can be explained by the better hydration of binder (cement + SGP) when the amount of water is greater. This
trend may be related to the higher water demand of silica glass powder to be activated as pozzolanic material [55]. Moreover, the use
of fine sand (0–1 mm) in the composition of the mortars, which is with a high water absorption coefficient compared to that of crushed
sand (0–4 mm), binder hydration may be incomplete in the case of low w/b ratio (when the quantity of mixing water is insufficient).
Consequently, the physical and mechanical properties of mortar may be influenced. Fine sand with high water absorption requires a
larger quantity of mixing water to prepare mortars. This can be seen in the case of mortar with w/b equal to 0.55, where a quantity of
binder can always remain anhydrous which leads to a delay in the evolution of the mechanical strength at the age of 60 days or more,
compared to mortars prepared with a higher water-to-binder ratio in which the hydration of the binder is considered almost complete
(e.g., the case of mortars with w/b of 0.65). This finding is consistent with previous studies on cement-based materials containing silica
fume [8,85].
It is observed in Fig. 7 that the compressive strength of all the mortar specimens containing SGP increases significantly, at the age of
90 days, when compared to that of CM mortars. The same trend can also be observed in the case of flexural strength measured at 90
days. For the mortar specimens containing SGP with w/b of 0.55, the increases in compressive strength between 28 and 90 days are
equal to 18.1%, 22.9, 22.4, 19.5 and 14.3%, respectively for M1-SGP7, M1-SGP10, M1-SGP12, M1-SGP15 and M1-SGP20 mortars,
while this increase in compressive strength is equal only to 7.5% in the case of control mortar (M1-CM). Similarly, in the case of
mortars with a w/b ratio of 0.65, the increases in compressive strength between 28 and 90 days, are equal to 10.7%, 14, 14.6, 17 and

Fig. 8. The evolution of compressive strength versus the water-to-binder ratio.

11
Z.e.-A. Kameche et al. Journal of Building Engineering 78 (2023) 107550

16.7% for M3-SGP7, M3-SGP10, M3-SGP12, M3-SGP15 and M3-SGP20, respectively. Concerning the control mortar (M3-CM), the
results demonstrate growth in compressive strength, between the 28th and 90th day of curing, with only a percentage of 6.7% (see
Fig. 7).
On the other hand, Fig. 8 presents the evolution of compressive strength versus the water-to-binder ratio for mortars of 90-day age.
The results of regression analysis show that the curves of experimental data can perfectly be approximated using a linear function,
where the regression coefficient R2 is often between values of 0.97 and 0.99 and that for different dosages of SGP. So, a straight-line fit
can represent the relationship between the w/b ratio and the compressive strength of SGP-modified mortars. Fig. 8 has clearly shown
that the compressive strength is more influenced by the water-to-binder ratio than the SGP content. Therefore, it can also be concluded
that compressive strength has a strong correlation with the water-to-binder ratio of mortar. In addition, as seen in Fig. 7, the

Fig. 9. Compressive strength activity index of different SGP-modified mortars.

12
Z.e.-A. Kameche et al. Journal of Building Engineering 78 (2023) 107550

compressive strengths of all mortars have the same tendency as the flexural strengths up to 90 days, where the optimum development
in mechanical strength is observed in the case of SGP12 compared to CM mortars.
The findings demonstrated that the silica-glass powder, tested in this study, improves the mechanical strength of the mortar in the
long term due to its greater pozzolanic reactivity. Previous research papers confirmed that the pozzolanic activity becomes more
important if the fineness of GP is greater [27,28,54,86]. Thus, the enhancement, at a later age, of the flexural and compressive
strengths of mortars with the addition of silica-glass powder can be attributed to the improvements in the cementitious paste which is
due to the pozzolanic reactions of SGP along with calcium hydroxide (CH) leading to improved C–S–H in the mortar. A similar
conclusion was reported in the research work of Parghi and Alam [41].
In addition, it is demonstrated thatwhatever the w/b ratio considered herein, mortars prepared with an SGP dosage equal to 12.5%,
always represent the optimum waste glass powder content for achieving the best flexural and compressive strengths. Therefore, an
overdose of silica-glass powder (higher than 12.5%) leads to a decrease in the mechanical strength, because the chemical bonds of the
C–S–H gel can be minimized with a lower amount of cement (replaced by SGP) in the mortar, and therefore, a lower mechanical
resistance is recorded in the short and long term. Du and Tan [38] have indicated that, at a higher GP replacement level, no further
pozzolanic reaction would occur and glass powder would become inactivated inert filler leading to a compressive strength similar to

Fig. 10. UPV for different mortar series at (a) 7, (b) 28 and (c) 90 days of curing.

13
Z.e.-A. Kameche et al. Journal of Building Engineering 78 (2023) 107550

the control material. Moreover, with higher silica-rich glass powder substitution, the Ca/Si ratio in the formed
calcium-silicate-hydrates (C–S–H) was lower and relatively more alkalis were bonded in the C–S–H phase [84]. As reported by Du and
Tan [84], with high cement substitution by GP, when the porosity of the material increases because of the higher effective w/b ratio,
the pH value in the interstitial solution would decrease. However, after a longer curing age, it remains unclear as to the porosity and
performance cement-based materials with high volume GP [44].

5.3. Pozzolanic activity of SGP-modified mortar

The Strength Activity Index (SAI) calculated at different curing ages (at 7, 28, 60 and 90 days) for all the series of SGP-mortar
specimens (M1, M2 and M3), with different water-to-binder ratios, are presented in Fig. 9. The results show that, except for the
M2-SGP12 and M3-SGP12 mortars, all the SGP-mortar specimens at 7 and 28 days-age are with a pozzolanic Strength Activity Index
less than those of the control mortars. At 7 days of curing, all SGP-mortar specimens show the lowest strengths and SAI than those of
the control mortar. In this case, SAI values are around 84–99%. It is important to note that these values are higher than the minimum
value indicated for pozzolanic constituents in ASTM C618 [56] which is 75%. It can be concluded from this SAI results that the SGP has
not a pozzolanic reactivity at early ages (7–28 days) and only a 12,5% SGP substitution made it possible to reach a strength value
comparable to 100% Portland cement at 28-day age when w/b ratio equal to 0.6 and 0.65. As seen in Fig. 8, values of SAI become
higher when SGP-mortar specimens are at 60 and 90 days of curing. So, the SGP shows considerable improvement in the strengths of
cement mortar, and therefore, it can be concluded that the pozzolanic reactivity of SGP can be improved in mortar after 28 days of
curing and become more noticeable with the age of specimen mortar. The series-M3 specimens (with w/b = 0.65) show the highest
pozzolanic SAI values than those of the M1 and M2 specimens series, especially when SGP substitutions are between 10% and 15% at
60 and 90 days of curing ages. Pozzolanic SAI values in these cases of series-M3 specimens are around 103–109%.
Parghi and Alam [41] investigated the Strength Activity Index of different mortars with different compositions of SCM’s and they
found that pozzolanic SAI values for recycled GP specimens (less than 75 μm and with 76% GP particles passing through 45 μm sieve)
were around 107–140% at 7, 28, 56 and 90 days. With a replacement up to 30% fine GP (less than 38 μm) in concrete at 90 days. Shao
et al. [41] investigated the mechanical properties of concrete and values of pozzolanic SAI achieved were greater than 90%. Shi et al.
[41] achieved SAI of about 74% at 28 days by using 25% of coarse GP (glass particles diameter ≤100 μm). Chen et al. [49] reported
that less or no pozzolanic reaction can be shown when the GP particles are greater than 75 μm, and a higher pozzolanic reaction is
found when the particle size is between 36 μm and 75 μm. The present study with 12.5% SGP replacement shows pozzolanic SAI values
around 108–110% at 90 days for all mortars with different w/b ratios.

5.4. Effect of incorporating the SGP on the ultrasonic properties of mortar

In this part of the study, the effect of silica-glass powder content on the ultrasonic properties of SGP-modified mortars was assessed
using a non-destructive method, namely the ultrasonic pulse velocity (UPV) technique. Considering different w/b ratios, UPV tests
were carried out on mortar at 7, 28, and 90 days of curing. Three measurements were taken for each test specimen and the mean values
are presented in Fig. 10. A total of about 162 average results of UPV were considered herein ranging from 3.3 to 3.85 km/s.
Fig. 10 shows that the UPV of mortars tested increases with the decrease in the w/b ratio. This can be explained by the fact that
mortar made with a lower w/b ratio showed relatively dense microstructure with fewer and smaller pores. Likewise, a higher w/b ratio
certainly will lead to higher porosity, which can slow down the speed of ultrasonic wave propagation. The results showed also that the
increase of the ultrasonic pulse velocity had almost the same tendency for the three series of mortars (i.e., M1, M2 and M3) (see
Fig. 10). The UPV of all mortar mixtures increased with the increasing amount of SGP. A similar tendency was also observed by
Ref. [87], for waste liquid crystal glass powder.
Moreover, the results showed that the velocity of ultrasonic waves through the material increases accordingly with the curing age
(see Fig. 10). In addition, the UPV increased faster in the case of SGP-modified mortars compared to that of the control mortars. From
Fig. 10, it can be seen that the increasing rate of pulse velocity at 7 days is higher than that recorded after 28 days of curing time. The

Fig. 11. The evolution of UPV versus the water-to-binder ratio.

14
Z.e.-A. Kameche et al. Journal of Building Engineering 78 (2023) 107550

increasing rate of pulse velocity is slower between the ages of 28 and 90 days. This result is consistent with the evolution of the
compressive strength of the mortar as a function of its age. As it is well-known that the mechanical resistance is influenced by the
hardening time of the cementitious material, the velocity of the ultrasonic pulses through this material increases with curing time [88,
89]. Indeed, since the velocity of ultrasonic waves through solid matter is higher than that through voids in the material, when.
The hydration of the paste is greater, the pore volume of cementitious material is lower and therefore the pulse velocity is higher
[89].
On the other hand, Fig. 11 shows the correlation between ultrasonic pulse velocity readings and water-to-binder ratios for the mixes
CM, SGP7, SGP12 and SGP20. It is observed that, whatever greater dispersion between the curves plotted, there is a perfect fit for
different SGP-modified mortars with R2 values often greater than 0.96. Although there is a scatter between the trend lines plotted for a
correlation between the ultrasonic wave velocity and the w/b ratio, a linear equation can be used to assess the UPV for each SGP
dosage as a function of the w/b ratio. It can also be concluded from the obtained curves (Fig. 11) that the use of a linear equation for all
the dosages of SGP to assess the UPV from some mortar criteria as the w/b ratio is not valid. Moreover, for a given value of the w/b
ratio, the obtained results show that UPV increased with increasing the SGP content, which is not the case for the compressive strength
evolution versus the w/b ratio as discussed above. This can be attributed to the fact that the compressive strength is not influenced only
by the w/b ratio, but also by other criteria such as the pozzolanic reactivity behaviour over time. The UPV through the material seems
to be directly influenced by the w/b ratio.
In addition, using the dynamic modulus of elasticity Edyn calculated from the obtained UPV values, the compressive strengths of
mortars at the 90th day of age are theoretically estimated and compared to the experimental uniaxial compressive strength. As can be
seen in Fig. 12 and Table 3, the trends between the compressive strength (CS) calculated from the UPV results and those measured
experimentally using a hydraulic press were found to be linear when considering each dosage level of SGP separately. Moreover, the
silica-glass powder content influence strongly the correlation between CS calculated and CS experimentally measured.
Fig. 12 shows that all the results obtained are below the diagonal line which divides the figure into two parts (which separates the
measured compressive strengths (CS) and those calculated from the readings of the ultrasonic pulse velocity). This indicates that the
calculated CS are always higher than those measured experimentally. It is also shown in Fig. 12 that the slopes of the straight lines
corresponding to the SGP15 and SGP20 mortars are lower in comparison to those of the CM, SGP7, SGP10 and SGP12 mortars. Because
the calculated CS results increase continuously with the silica-glass powder content, unlike those measured experimentally in the case
of SGP15 and SGP20 mortars which decrease when the SGP dosage increases. The calculated compressive strengths based on the
ultrasonic method are significantly influenced by the porosity of the material which decreases with the increase in SGP dosage. This
trend is different when observing the measured compressive strengths which can be influenced by other physical and/or chemical
characteristics of the material such as weaker bonds between the C–S–H gel, due to a replacement of high quantities of cement by SGP.
Last, but not least, it is to note that although the variation between the different silica-glass powder dosages considered herein was
relatively low, it was difficult to propose a single trend line between the calculated compressive strengths and those measured from the
ultrasound results for mortars modified with glass powder.

5.5. The effect of incorporating the SGP on the durability properties of mortars
5.5.1. The water absorption by capillarity
Fig. 13 shows the evolution of the mass of water absorbed by the different mortars tested as a function of the square root of time. It
is important to note that the measurements of the water absorbed were recorded by weighing carried out at 5, 10, 15, 30, 60 and 120
min after the start of the test. It is important to note that the values illustrated in Fig. 13 are arithmetic means of three values, obtained
by measurements carried out on three specimens of (4 x 4 x 16) cm3 of each mortar tested at the age of 90 days.
As seen in Fig. 13, the amount of water absorbed by the mortar specimens increases with time. It is also observed that the absorption
of control mortar is greater than that of the modified mortars which contain a percentage of the silica-glass powder. For instance, the
masse of water absorbed by mortars with 0.65 w/b ratio reaches, after 120 min, the values of 5.2 g, 4.64, 4.37, 4.19, 4 and 3.86 g of

Fig. 12. Comparison of measured CS and predicted CS for each dosage level of SGP.

15
Z.e.-A. Kameche et al. Journal of Building Engineering 78 (2023) 107550

Table 3
Empirical equations between calculated CS and measured CS.

Mixture ID Correlation coefficient (R2) Regression equation (y = ax − b)

CM 0.97 CS(measu.) = 1.34 × CS(cal.)‒ 16.3

SGP7 0.99 CS(measu.) = 1.98 × CS(cal.)‒ 48.5
SGP10 0.99 CS(measu.) = 1.48 × CS(cal.)‒ 45.4
SGP12 0.96 CS(measu.) = 1.47 × CS(cal.)‒ 27
SGP15 0.99 CS(measu.) = 0.67 × CS(cal.)‒ 8.1
SGP20 0.99 CS(measu.) = 1.18 × CS(cal.)‒ 22

Fig. 13. The masse of absorbed water by hardened mortars with different dosages of SGP: (a) Series M1, (b) Series M2 and (c) Series M3.

water for M3-CM, M3-SGP7, M3-SGP10, M3-SGP12, M3-SGP15 and M3-SGP20, respectively. The results also highlighted that,
whatever the water-to-binder ratio of the mortar, the higher the dosage of SGP, the lower the absorption of the mortar mixture.
Therefore, the mass of absorbed water decreases for the mortars containing SGP, compared to the control mortars. This trend is more
noticeable when the mortar has a high w/b ratio (see Fig. 13). The reduction in water absorption (influenced by the percentage of SGP)
can be explained by the great fineness of the glass powder compared to cement. Moreover, the pozzolanic reaction of silica-glass
powder may produce new additional reaction products, such as calcium silicate hydrate (C–S–H) or calcium aluminates hydrate (C-
A-H) [90]. In the mortar at the age of 90 days, the pozzolanic reaction products lead certainly to a densification of the cement matrix.
This will reduce the capillary porosity as the compactness of the mortar mixture increase. This result is in agreement with the study by

Fig. 14. Capillary Sorptivity of different mortar series for different dosages of SGP.

16
Z.e.-A. Kameche et al. Journal of Building Engineering 78 (2023) 107550

Tumidajskiet al. [91]. From a literature review, and with the aim of comparison with other works carried out on the influence of silica
fume on the different properties of mortar, Oltulu and Sahin [49] studied the pore-structure development of cement mortars containing
silica fume and other nano-powders. The addition of nano-silica resulted in an increased amount of C–S–H and other hydrated
products. However, it has been shown that an increased content of silica fume in Portland cement produced in C–S–H with a lower C/S
ratio [49].
The sorptivity coefficient characterizes the absorption capacity of the tested mortars over time. When the porous network is thin,
the absorption kinetics and the absorption coefficient are certainly lower [91].
The sorptivity coefficient results presented in Fig. 14 are obtained by applying Eq. (3) provided insection 4.6. The obtained
sorptivity coefficients indicate that the control mortars (M1-CM, M2-CM and M3-CM) represent the highest sorptivity coefficients
about 0.219, 0.274 and 0.347 mm/min0.5, respectively. The results indicate that quantities of water absorbed by M3-CM are greater
than those absorbed by M1-CM and M2-CM. This can be attributed to the fact that a mortar with a high water-to-binder ratio has a large
open porosity (porosity accessible to water), which leads certainly to a higher sorptivity coefficient.
Furthermore, it is observed from Fig. 14 that the incorporation of the SGP slows down the absorption kinetics. More precisely,
increasing the amount of glass powder in the mortar led to a decrease in thesorptivity coefficients indicating that the residual glass
powder could fill the micro-cracks and pores in the mortar leading to the decrease of the voids being otherwise occupied with water.
For instance, in the case of M1, the decreases in sorptivity coefficients of SGP mortars compared to M1-CM are equal to 6%, 10, 12.3, 21
and 22.8% for M1-SGP7, M1-SGP10, M1-SGP12, M1-SGP15 and M1-SGP20, respectively. Similarly, the decrease in sorptivity of M3
modified mortars compared to M3-CM are about 10.4%, 11.8, 17.8, 20.2 and 25.6% for M3-SGP7, M3-SGP10, M3-SGP12, M3-SGP15
and M3-SGP20, respectively. This trend is mainly due to the finer particles and high specific surface area of silica-glass powder used
herein. The finer the silica-glass powder particles, the lower capillary absorption as they physically fill capillary pores [28,31]. A
similar conclusion has been reported by Parghi and Alam [41] and Du and Tan [44]. In addition, a reduced water-to-binder ratio which
further contributes to the microstructure densification can be considered as a factor contributing to the reduction of the sorptivity of
mortars that have been made with silica-glass powder (see Fig. 14).
Furthermore, the correlation between sorptivity coefficients and the compressive strength values is proposed in this study for each
SGP dosage. The correlation is established by considering experimental results obtained from different w/b ratios. It is seen from
Fig. 15 that by increasing the sorptivity coefficient of the mortar, its compressive strength decreases. In addition, the evolution of the
compressive strength as a function of the coefficient of capillary absorption may be represented through a linear regression, where a
straight line fits perfectly with the experimental results (with a coefficient of regression R2 often greater than 0.97). For instance,
Fig. 15 shows the linear equation of the sorptivity coefficient as a function of compressive strength in the case of SGP12 mortar which
represents the optimum value of 12,5% to substitute cement by SGP. It is important to point out that specific equations for each dosage
level of SGP lead to higher R2. However, if the fitting is performed by considering the experimental results of all mortar mixtures, the
coefficient of correlation would increase considerably, which means that the use of a general equation for the prediction of the
sorptivity coefficient from the compressive strength is not valid.

5.5.2. The permeability to chloride

To assess the beneficial effects of hydration of SGP in relation to durability of mortar, resistance to chloride penetration were
considered in this study. Thus, to perform a quantitative assessment of concrete and mortar permeability, the chloride ion permeability
test is commonly used. According to the American standard power method (ASTM C1202), the principle of this experimental test is to
determine the total charges passing through the sample of cementitious material subjected to a difference in voltage of 60 V after 6 h. In
this study, the chloride ion tests were carried out after 90 days of wet curing. The chloride ion conduction results, represented by the
total electric charge Q (in coulombs) crossing the mortar sample, are shown in Fig. 16. So, the total charges measured on each mortar
studied are correlated with its chloride ion permeability. The increase in the total electrical charges indicates that the mortar is more
permeable to chloride.
From Fig. 16, a quite perceptible difference can be observed in terms of the chloride ions permeabilities of the control mortars (CM)

Fig. 15. The correlation between sorptivity coefficients and compressive strengths.

17
Z.e.-A. Kameche et al. Journal of Building Engineering 78 (2023) 107550

Fig. 16. Permeability to chloride expressed in terms of conduction of chloride ions of all series of mortar mix at 90 days of curing. (Chloride ion permeability ranges
are separated by dotted lines indicating high, moderate, low and very low permeability).

prepared with different w/b ratios. The permeability to chloride of the reference mortars tested (M1-CM, M2-CM and M3-CM) follows
the classic trend well-known in the literature: the permeability of the material increases with the increase in the w/b ratio. A higher w/
b ratio leads to an improvement in the volume and size of the pores. However, the mortars containing the same rate of silica-glass
powder have chloride permeabilities almost similar whatever the w/b ratio, indicating a reduction in the interconnectivity of the
pores due to the pozzolanic reaction of silica-glass powder.
At the age of 90 days, the chloride ions permeability of the mortars incorporating SGP is considerably reduced compared to that of
the control mortars. A proportional reduction in the total electric charge Q with the increase in the dosage of SGP is observed (see
Fig. 16). So, according to ASTM C1202 [72], results indicate that the chloride ion permeability is high for the mortars without SGP
because of the high w/b ratios used in this study (between 0.55 and 0.65). Whereas the chloride ion permeability of mortars containing
SGP are ranging from the class of moderate to low permeability depending on the percentage of SGP replacing an amount of cement.
This result is in full accordance with several research works carried out on concretes, in which the authors observed low conduction of
chloride ions in concretes based on finely ground glass waste, in particular for high substitution rates (greater than 20%) [92,93].
In addition, obtained results show that the decrease in the chloride ions conduction in the presence of SGP is significantly greater, as
long as the dosage of SGP is increased. The lowest permeability values are observed in the case of mortars containing 20% of the glass
powder, particularly when the w/b ratio is lower. It is also observed from Fig. 16 that mortars incorporating 20% of SGP exhibit
permeability of 4–5 times lower than those of control mortar depending on their w/b ratios. This decrease in permeability in the
presence of SGP can be related to pore refinement, pore blocking and filling effects of the glass pozzolanic reaction product [5,25,94].
The densification of the cementitious matrix can lead to a reduction in the penetration of chloride ions. Furthermore, because of its
high fineness, SGP develops a filler effect as well as a pozzolanic effect, which promotes refining mortar pores. This could explain the
observed low permeability when the mortar contains a quantity of glass powder. The durability of the mortar based on SGP against
aggressive agents is, therefore, improved because of its low permeability to chlorides which increases its resistance to the penetration
of external agents. The same trend was observed for cement-based material made with waste liquid crystal glass [48], E-glass particles
[49] and glass powder and glass sludge wastes [50].

5.5.3. The chloride ions diffusion

The passage of the chlorides was carried out after 90 days of curing. The coefficients of chloride ions diffusion (De) for all the

Fig. 17. Coefficient of chloride ions diffusion of all mortar series tested after 90 days of curing.

18
Z.e.-A. Kameche et al. Journal of Building Engineering 78 (2023) 107550

mortars studied are presented in Fig. 17. The diffusion coefficients of the control mortars were found to be equal to 5.2E-11, 7.9E-11
and 1.5E-10 m2/s, respectively for M1-CM, M2-CM and M3-CM. The passage time of the chloride ions is greater for the mortars
containing silica-glass powder. In the case of mortars containing 20% of the SGP in replacement of cement, the passage time lasted
more than 180 days. This duration was considered sufficient for the steady state of chloride diffusion to be reached. The long chloride
passage time indicates good resistance to chloride penetration through mortars based on glass powder.
Moreover, the obtained results show that, whatever the water-to-binder ratios considered herein, all the control mortars tested
have higher diffusion coefficients compared to that of mortars containing silica-glass powder, which indicates the great capacity of
glass powder to reduce Cl− diffusion through the mortar. According to Fig. 17, the lowest diffusion coefficients are recorded in the case
of mortars incorporating 20% glass powder. This decrease in the Cl− diffusion coefficient is mainly due to the pozzolanic activity of the
glass powder and its ability to fix chlorides. So, the fixation of chlorides on the walls of hydrates of cement and glass powder (C–S–H)
by adsorption, which have a high specific surface, contributes to the reduction of the diffusion coefficient of chloride ions.
In addition, because of the pozzolanic reaction, mortars based on glass powder have a higher resistance against the penetration of
chlorides due to the development of a denser microstructure [92]. A high rate of portlandite (Ca(OH)2) in the hydrated matrix gen­
erates a large volume of continuous pores. Therefore, the consumption of lime during the hydration of the silica glass powder
(pozzolanic reactions) leads to a relative decrease in the volume of the interconnected capillary pores making the porous structure of
the material denser, more tortuous and discontinuous. Matos and Sousa-Coutinho [31] investigated the durability properties of mortar
with 10 and 20% cement replacement by GP and found that chloride ions diffusion decreased with GP inclusion.
Furthermore, Fig. 17 shows that the increase in the Cl− diffusion coefficient is also observed with the increase in the w/b ratio in all
mortar mixtures including the control mortars and the mortars incorporating SGP. A similar trend was observed by Leng et al. [95] in
the case of concretes incorporating fly ash and slag. In the case of SGP20 mortar, the increase in the chloride diffusion coefficient is
very low showing that the influence of the w/b ratio on the coefficient of Cl− diffusion is not significant.
By comparing the chloride ions conduction results (the total charge Q) with those of the diffusion (Cl− diffusion coefficient), it is
remarkable that the two tests show relatively the same improvement of the permeability to chloride ions of the mortars based on the
SGP.
In this section of the study, two correlations are assessed and discussed. The first correlation is between the coefficients of sorptivity
(Ca) and chloride ions diffusion (De) (see Fig. 18), while the second is between the air content in the mortar at the fresh state and the
coefficient of Cl− diffusion (De) (see Fig. 19). Recall here that the measurements of water capillary absorption and the diffusion of
chloride ions were carried out on test specimens of mortar after 90 days of curing.
In Fig. 18, the correlation between the coefficients of sorptivity (Ca) and chloride ions diffusion (De) is assessed by considering the
experimental data of all the control mortars and those of the modified mortars with 7, 12,5 and 20% of SGP with different w/b ratios.
The specific equations for each dosage level of SGP show that the experimental data of the sorptivity coefficient can correctly be
approximated by a logarithmic function (Ln) using the coefficient of Cl− diffusion.
The regression coefficients (R2) for all tested mortars are higher or equal to 0.98. It is also observed from Fig. 18 that, the more the
SGP dosage is greater, the more the correlation is higher. Furthermore, the water-to-binder ratio influences strongly the tendency of
the curves obtained by regression analysis. When the water-to-binder ratio is low, the microstructure in the transition zones (near the
aggregate surface) is improved. The size, quality and preferential orientation of C–S–H crystals are considerably improved because of
the pozzolanic reactions of the SGP. All these phenomena result in stronger and more uniform interfacial transition zones with less
potential for micro cracks leading to the grain refinement of hydrated binder paste in the transition zone.
The water absorption by capillarity has a strong correlation with the coefficient of chloride ions diffusion through the hardened
mortar (X in m2/s). Mathematical analysis of data shows good relation of sorptivity by logarithmic function versus the coefficient of
Cl− diffusion as shown below:

Ca(CM) = 0.12 Ln(De) + 3.07 (R2 = 0.98),

Ca(SGP7) = 0.106 Ln(De) + 2.77 (R2 = 0.98),

Ca(SGP12) = 0.047 Ln(De) + 1.4 (R2 = 0.98),

Ca(SGP20) = 0.074 Ln(De) + 2.1 (R2 = 0.99).

From the obtained logarithmic functions (Ln), given for different dosage level of SGP, between the sorptivity coefficient (Ca) and
that one of the Cl− diffusion (De), it can be concluded that, despite obtaining a regression coefficient (R2), which is often higher than
0.97 whatever the GSP dosage, it is difficult to give only one logarithmic equation for all the modified mortars tested to approximate
the sorptivity coefficient using the coefficient of chloride ions diffusion.
Fig. 19 shows the correlation between the air content and the coefficient of chloride ions diffusion. The regression analysis is
assessed by considering the experimental data of all the control mortars and those of the modified mortars with 7, 12,5 and 20% of SGP
with different w/b ratios. It was observed that the relationship between the percentage of air entrapped in fresh mortar has a strong
correlation with the coefficient of Cl− diffusion through the hardened mortar. This correlation can be described mathematically by a
powerlaw in function of the coefficient of Cl− diffusion (De) as shown below:
0.35
Air content(CM) = 1E-03 (De)− (R2 = 0.99)
0.45
Air content(SGP7) = 9E-05 (De)− (R2 = 0.98)

19
Z.e.-A. Kameche et al. Journal of Building Engineering 78 (2023) 107550

Fig. 18. Relationship between the coefficients of sorptivity and Cl− diffusion for mortars tested at 90 days of curing age.

Fig. 19. Relationship between the air content and the coefficient of Cl− diffusion for mortars tested at 90 days of curing age.

0.36
Air content(SGP12) = 4E-04 (De)− (R2 = 0.99)
0.58
Air content(SGP20) = 1E-06 (De)− (R2 = 0.99)

In Fig. 19, only the power equation for the mortar with 12.5% of silica-glass powder which is considered the optimum SGP dosage,
allowing to obtain the best mechanical strength of the mortar tested. But it is clear, from all the obtained equations in this correlation,
that it is difficult, if not impossible, to give only one power law to predict the Air content in the fresh mortar from the coefficient of Cl−
diffusion for all the different modified-mortars with different dosage of SGP.

6. Conclusion
This paper studied the potential use of silica-glass powder (SGP) as a cement replacement for sustainable cement-based con­
struction. Based on observed results, the following conclusions can be drawn.

• Modified mortars with SGP exhibit better flow behaviour which confirms that finely grounded recycled GP behaves as a filler in the
fresh state.
• The use of finely powdered silica-glass (≤75 μm) can lead to lower air content. The general trend observed was that, the larger the
amount of SGP in the mortar mixture, the lower quantity of air entrapped in the fresh mortar. The entrapped air in the fresh mortar
decreases when the w/b ratio increases.
• Flexural and compressive strengths of all SGP mortars are lower than that of control mortars at the earlier age. This trend is reversed
at later ages due to the pozzolanic effect of SGP finely ground. The incorporation of SGP as cement replacement led to an
improvement in flexural and compressive strengths of mortars at later ages, except for the case of mortars made with 20% of SGP.
The strength performance of SGP was shown to be slightly better for mortars with a lower w/b ratio. The optimum SGP content for
achieving the best mechanical strength was found to be 12.5%.

20
Z.e.-A. Kameche et al. Journal of Building Engineering 78 (2023) 107550

• The UPV through the SGP-modified mortars, at different ages, are higher than those of CM mortars. UPV increased with the
increasing amount of SGP. This trend is significantly noticeable in the case of mortars with a lower w/b ratio.
• The compressive strengths calculated from the UPV results are always higher than those measured experimentally, particularly in
the case of mortar mixtures with higher substitution levels (i.e. SGP15 and SGP20).
• The incorporation of SGP in mortar can prevent water adsorption. It is concluded that, whatever the w/b ratio, the higher the
dosage of SGP, the lower the absorption of the mortar. This trend is particularly noticeable for higher w/b ratio.
• A comparative evaluation of the sorptivity coefficients indicate that incorporating SGP slows down the absorption kinetics which
may be attributed to the fact that fine SGP can physically fill capillary pores, which implies that the mortar becomes denser and less
permeable.
• Correlation analysis between the sorptivity coefficient (Ca) and the compressive strength ofmortars indicates that the relationship
between them can be approximated by a linear fit with a high correlation.
• The partial substitution of cement by SGP considerably decreases the chloride ion permeability. Mortars incorporating 20% of SGP
showed 4 to 5 times lower permeability compared to CM mortars. Moreover, the mortars containing the same substitution level of
SGP have almost similar chloride permeability ranges, regardless of the w/b ratio.
• In terms of the chloride ions diffusion, the general trend observed was that the larger amount of SGP, the lower the Cl− diffusion
coefficient. Moreover, it was found that the Cl− diffusion increased with increasing of the w/b ratio, except in the case of M1-
SGP12, M1-SGP15, and M1-SGP20 where the influence of w/b is not significant.
• Considering a specific equation for each SGP dosage level, the coefficient of chloride ions diffusion of mortars had a strong cor­
relation with the air entrapped in the fresh mortar and the sorptivity coefficient. Correlation analyses have demonstrated that the
coefficient of chloride ions diffusion of mortars can be approximated by a power law function using the percentage of air entrapped
in the fresh mortar and by a logarithmic function using the sorptivity coefficient.

7. Recommendations for future research

This study intended to evaluate the effects of incorporating silica-rich glass powder as an SCM on selected properties of Portland
cement mortar, mainly, workability, air content, mechanical properties, ultrasonic properties, sorptivity and resistance to Cl− pene­
tration. The performance of mortars made with SGP seems to be promising. However, further investigation based on X-ray diffraction
(XRD), thermal analysis (TGA/DTA), scanning electron microscopy (SEM), calorimetry and mercury intrusion porosimetry (MIP) is
still needed to better understand the effects of this silica-rich glass on the microstructure evolution, mineralogy, pozolanic activity in
different hardened SGP-modified mortars (with different porosities and different pH of pore solution) and their changes with curing
age. It is also very important to test mortars with a high SGP contents (up to 50%).
Furthermore, studying the effects of this type of glass powder on the behaviour of SGP-modified mortars in terms of durability
parameters is vital for the effective use of silica glass in different applications. On the other hand, the use of glass powder as sup­
plementary cementitious material can be considered as environmental-friendly practice, as it reduces the amount of cement in
cementitious materials. However, milling and grinding waste glass to obtain finely ground-powdered glass – essential to increase the
surface and reactivity of the glass particle – is an energy-intensive process. More research is required to assess the effectiveness of such
procedures for sustainable construction practices.

Declaration of competing interest

The authors declare that they have no known competing financial interests or personal relationships that could have appeared to
influence the work reported in this paper.

Data availability

Data will be made available on request.

References
[1] P.A. Alsop, Cement Plant Operations Handbook: for Dry Process Plants, Tradeship Publications Ltd, 2007.
[2] Cement Industry Federation, Australia’s Cement Industry. Available from: <http://www.cement.org.au/AustraliasCementIndustry.aspxy>.” .
[3] D.N. Huntzinger, T.D. Eatmon, A life-cycle assessment of Portland cement manufacturing: comparing the traditional process with alternative technologies,
J. Clean. Prod. 17 (7) (2009) 668–675.
[4] P. Shoaei, F. Ameri, H. Reza Musaeei, T. Ghasemi, C.B. Cheah, Glass powder as a partial precursor in Portland cement and alkali-activated slag mortar: a
comprehensive comparative study, Construct. Build. Mater. 251 (2020), 118991.
[5] M.N.N. Khan, A.K. Saha, P.K. Sarker, Reuse of waste glass as a supplementary binder and aggregate for sustainable cement-based construction materials: a
review, J. Build. Eng. 28 (October 2019) (2020), 101052.
[6] S. Chandra Paul, B. Šavija, A.J. Babafemi, A comprehensive review on mechanical and durability properties of cement-based materials containing waste recycled
glass, J. Clean. Prod. 198 (2018) 891–906.
[7] B.A. Tayeh, Effects of marble, timber, and glass powder as partial replacements for cement, J. Civ. Eng. Constr. 7 (2) (2018) 63.
[8] A.I. Jubeh, D.M. Al Saffar, B.A. Tayeh, Effect of recycled glass powder on properties of cementitious materials contains styrene butadiene rubber, Arabian J.
Geosci. 12 (2) (2019).
[9] L.M. Federico, S.E. Chidiac, Waste glass as a supplementary cementitious material in concrete - critical review of treatment methods, Cem. Concr. Compos. 31
(8) (2009) 606–610.
[10] L.A. Pereira-De-Oliveira, J.P. Castro-Gomes, P.M.S. Santos, The potential pozzolanic activity of glass and red-clay ceramic waste as cement mortars components,
Construct. Build. Mater. 31 (2012) 197–203.

21
Z.e.-A. Kameche et al. Journal of Building Engineering 78 (2023) 107550

[11] A. Mohajerani, J. Vajna, T.H.H. Cheung, H. Kurmus, A. Arulrajah, S. Horpibulsuk, Practical recycling applications of crushed waste glass in construction
materials: a review, Construct. Build. Mater. 156 (2017) 443–467.
[12] G. Chen, et al., Glass recycling in cement production—an innovative approach, Waste Manag. 22 (7) (2002) 747–753.
[13] Z. Xie, Y. Xi, Use of recycled glass as a raw material in the manufacture of Portland cement, Mater. Struct. 35 (8) (2002) 510–515.
[14] E.T. Bueno, J.M. Paris, K.A. Clavier, C. Spreadbury, C.C. Ferraro, T.G. Townsend, A review of ground waste glass as a supplementary cementitious material: a
focus on alkali-silica reaction, J. Clean. Prod. 257 (2020), 120180.
[15] G. Lee, T.-C. Ling, Y.-L. Wong, C.-S. Poon, Effects of crushed glass cullet sizes, casting methods and pozzolanic materials on ASR of concrete blocks, Construct.
Build. Mater. 25 (5) (2011) 2611–2618.
[16] H. Du, K.H. Tan, Use of waste glass as sand in mortar: Part II–Alkali–silica reaction and mitigation methods, Cem. Concr. Compos. 35 (1) (2013) 118–126.
[17] H. Du, K.H. Tan, Effect of particle size on alkali–silica reaction in recycled glass mortars, Construct. Build. Mater. 66 (2014) 275–285.
[18] N.A. Soliman, A. Tagnit-Hamou, Partial substitution of silica fume with fine glass powder in UHPC: filling the micro gap, Construct. Build. Mater. 139 (2017)
374–383.
[19] N.A. Soliman, A. Tagnit-Hamou, Development of ultra-high-performance concrete using glass powder – towards ecofriendly concrete, Construct. Build. Mater.
125 (2016) 600–612.
[20] A.M. Tahwia, A. Essam, B.A. Tayeh, M.A. Elrahman, Enhancing sustainability of ultra-high performance concrete utilizing high-volume waste glass powder,
Case Stud. Constr. Mater. 17 (October) (2022), 01648.
[21] A. Adesina, et al., Fresh and mechanical properties overview of alkali-activated materials made with glass powder as precursor, Clean. Mater. 3 (November
2021) 2022.
[22] Y. Jiang, T.C. Ling, K.H. Mo, C. Shi, A critical review of waste glass powder – multiple roles of utilization in cement-based materials and construction products,
J. Environ. Manag. 242 (November 2018) (2019) 440–449.
[23] M.C. Bignozzi, A. Saccani, L. Barbieri, I. Lancellotti, Glass waste as supplementary cementing materials: the effects of glass chemical composition, Cem. Concr.
Compos. 55 (2015) 45–52.
[24] A.A. Aliabdo, A.E.M. Abd Elmoaty, A.Y. Aboshama, Utilization of waste glass powder in the production of cement and concrete, Construct. Build. Mater. 124
(2016) 866–877.
[25] R.U.D. Nassar, P. Soroushian, Strength and durability of recycled aggregate concrete containing milled glass as partial replacement for cement, Construct. Build.
Mater. 29 (2012) 368–377.
[26] K. Afshinnia, P.R. Rangaraju, Impact of combined use of ground glass powder and crushed glass aggregate on selected properties of Portland cement concrete,
Construct. Build. Mater. 117 (2016) 263–272.
[27] A. Shayan, A. Xu, Performance of glass powder as a pozzolanic material in concrete: a field trial on concrete slabs, Cement Concr. Res. 36 (3) (2006) 457–468.
[28] B. Taha, G. Nounu, Properties of concrete contains mixed colour waste recycled glass as sand and cement replacement, Construct. Build. Mater. 22 (5) (2008)
713–720.
[29] N. Schwarz, H. Cam, N. Neithalath, Influence of a fine glass powder on the durability characteristics of concrete and its comparison to fly ash, Cem. Concr.
Compos. 30 (6) (2008) 486–496.
[30] I.M. Metwally, Investigations on the performance of concrete made with blended finely milled waste glass, Adv. Struct. Eng. 10 (1) (2007) 47–53.
[31] A.M. Matos, J. Sousa-Coutinho, Durability of mortar using waste glass powder as cement replacement, Construct. Build. Mater. 36 (2012) 205–215.
[32] H. Rashidian-Dezfouli, K. Afshinnia, P.R. Rangaraju, Efficiency of Ground Glass Fiber as a cementitious material, in mitigation of alkali-silica reaction of glass
aggregates in mortars and concrete, J. Build. Eng. 15 (July 2017) (2018) 171–180.
[33] K. Afshinnia, P.R. Rangaraju, Influence of fineness of ground recycled glass on mitigation of alkali-silica reaction in mortars, Construct. Build. Mater. 81 (2015)
257–267.
[34] K. Afshinnia, P.R. Rangaraju, Efficiency of ternary blends containing fine glass powder in mitigating alkali-silica reaction, Construct. Build. Mater. 100 (2015)
234–245.
[35] C. Shi, Y. Wu, C. Riefler, H. Wang, Characteristics and pozzolanic reactivity of glass powders, Cement Concr. Res. 35 (5) (2005) 987–993.
[36] G.M.S. Islam, M.H. Rahman, N. Kazi, Waste glass powder as partial replacement of cement for sustainable concrete practice, Int. J. Sustain. Built Environ. 6 (1)
(2017) 37–44.
[37] J. xin Lu, Z. hua Duan, C.S. Poon, Fresh properties of cement pastes or mortars incorporating waste glass powder and cullet, Construct. Build. Mater. 131 (2017)
793–799.
[38] H. Du, K.H. Tan, Waste glass powder as cement replacement in concrete, J. Adv. Concr. Technol. 12 (11) (2014) 468–477.
[39] A. Ghayoor Khan, B. Khan, E. Abdul Ghayoor khan, B. Khan, Effect of partial replacement of cement by mixture of glass powder and silica fume upon concrete
strength, Int. J. Eng. Work. 4 (7) (2017) 124–135.
[40] I. Almeshal, M.M. Al-Tayeb, S.M.A. Qaidi, B.H. Abu Bakar, B.A. Tayeh, Mechanical properties of eco-friendly cements-based glass powder in aggressive medium,
Mater. Today Proc. 58 (2022) 1582–1587.
[41] A. Parghi, M. Shahria Alam, Physical and mechanical properties of cementitious composites containing recycled glass powder (RGP) and styrene butadiene
rubber (SBR), Construct. Build. Mater. 104 (2016) 34–43.
[42] M. Kamali, A. Ghahremaninezhad, Effect of glass powders on the mechanical and durability properties of cementitious materials, Construct. Build. Mater. 98
(2015) 407–416.
[43] B. Taha, G. Nounu, Utilizing waste recycled glass as sand/cement replacement in concrete, J. Mater. Civ. Eng. 21 (12) (2009) 709–721.
[44] H. Du, K.H. Tan, Properties of high volume glass powder concrete, Cem. Concr. Compos. 75 (2017) 22–29.
[45] K.I.M. Ibrahim, Recycled waste glass powder as a partial replacement of cement in concrete containing silica fume and fly ash, Case Stud. Constr. Mater. 15
(June) (2021), e00630.
[46] M. Carsana, M. Frassoni, L. Bertolini, Comparison of ground waste glass with other supplementary cementitious materials, Cem. Concr. Compos. 45 (2014)
39–45.
[47] B.A. Tayeh, D.M. Al Saffar, A.S. Aadi, I. Almeshal, Sulphate resistance of cement mortar contains glass powder, J. King Saud Univ. - Eng. Sci. 32 (8) (2020)
495–500.
[48] H.Y. Wang, H.H. Zeng, J.Y. Wu, A study on the macro and micro properties of concrete with LCD glass, Construct. Build. Mater. 50 (2014) 664–670.
[49] C.H. Chen, R. Huang, J.K. Wu, C.C. Yang, Waste E-glass particles used in cementitious mixtures, Cement Concr. Res. 36 (3) (2006) 449–456.
[50] H. Lee, A. Hanif, M. Usman, J. Sim, H. Oh, Performance evaluation of concrete incorporating glass powder and glass sludge wastes as supplementary cementing
material, J. Clean. Prod. 170 (2018) 683–693.
[51] S. De Castro, J. De Brito, Evaluation of the durability of concrete made with crushed glass aggregates, J. Clean. Prod. 41 (2013) 7–14.
[52] S.M.A. Qaidi, Y.Z. Dinkha, M.M. Arbili, J.H. Haido, B.A. Tayeh, Properties of sustainable green concrete containing waste glass as eco-friendly aggregate: a
review, AIP Conf. Proc. 2660 (November) (2022).
[53] J. Brinker, Silica Glass and its Application, vol. 11, 1991.
[54] R. Idir, M. Cyr, A. Tagnit-Hamou, Pozzolanic properties of fine and coarse color-mixed glass cullet, Cem. Concr. Compos. 33 (1) (2011) 19–29.
[55] D.M. Al Saffar, T.A. Tawfik, B.A. Tayeh, Stability of glassy concrete under elevated temperatures, Eur. J. Environ. Civ. Eng. 0 (0) (2020) 1–12.
[56] ASTM C618-05, Standard Specification for Coal Fly Ash and Raw or Calcined Natural Pozzolan for Use in Concrete, 2005.
[57] Z.H. Cheng, A. Yasukawa, K. Kandori, T. Ishikawa, FTIR study on incorporation of CO2 into calcium hydroxyapatite, J. Chem. Soc., Faraday Trans. 94 (10)
(1998) 1501–1505.
[58] R. Maddalena, C. Hall, A. Hamilton, Effect of silica particle size on the formation of calcium silicate hydrate [C-S-H] using thermal analysis, Thermochim. Acta
672 (September 2018) (2019) 142–149.
[59] European Committee for Standardization, EN 196-1 : Methods of Testing Cement–Part 1: Determination of Mechanical Strengths, 2006.

22
Z.e.-A. Kameche et al. Journal of Building Engineering 78 (2023) 107550

[60] AFNOR, NF P18-452 : Concretes - Measuring the Flow Time of Concretes and Mortars Using a Workabilitymeter, 2006.
[61] AFNOR, NF P15-437 : Hydraulic Binders. Testing Technics. Characterization of Cements by Fluidity Measurement under Mortar Vibration, 1987.
[62] AFNOR, NF EN 413-2 : Masonry Cement - Part 2 : Test Methods, 2017.
[63] S. Chatterji, Freezing of air-entrained cement-based materials and specific actions of air-entraining agents, Cem. Concr. Compos. 25 (7) (2003) 759–765.
[64] AFNOR, NF EN 12504-4 : Testing Concrete in Structures - Part 4 : Determination of Ultrasonic Pulse Velocity, 2021.
[65] R.F. Feldman, H. Cheng-Yi, Properties of portland cement-silica fume pastes I. Porosity and surface properties, Cement Concr. Res. 15 (5) (1985) 765–774.
[66] S.G. Xiaofeng Cong David Darwin, Steven L. McCabe, Role of silica fume in compressive strength of cement paste, mortar, and concrete, ACI Mater. J. 89 (4)
(1992).
[67] P.J.P. Gleize, A. Müller, H.R. Roman, Microstructural investigation of a silica fume–cement–lime mortar, Cem. Concr. Compos. 25 (2) (2003) 171–175.
[68] R.M. de Gutiérrez, L.N. Díaz, S. Delvasto, Effect of pozzolans on the performance of fiber-reinforced mortars, Cem. Concr. Compos. 27 (5) (2005) 593–598.
[69] AFNOR, NF EN 1015-11 : Methods of Test for Mortar for Masonry—Part 11: Determination of Flexural and Compressive Strength of Hardened Mortar, 2019.
[70] ASTM C311, Standard Test Methods for Sampling and Testing Fly Ash or Natural Pozzolans for Use in Portland-Cement Concrete, 2011.
[71] Recommended Methods for Measuring Quantities Associated with Durability, Report of the AFPC-AFREM Technical Days ‘Concrete Durability’, 1998, 11-12
December 1997,” Toulouse, France.
[72] ASTM C1202, Standard Test Method Electrical Indication of Concrete’s Ability to Resist Chloride Ion Penetration, 2019, pp. 1–8.
[73] AFPC-AFREM, Recommended methods for the measurement of magnitudes associated with sustainability, in: Proceedings of the AFPC-AFREM Technical Days,
1997. Concrete Durability.
[74] F. G. B. Touil A. Bezzar, A. Khelidj, and S. Bonnet, “Effect of temperature on chloride diffusion in saturated concrete,” ACI Mater. J., vol. 114, no. 5.
[75] A. Khelidj, S. Bonnet, F. Rosquoet, F. Schoefs, Chloride Spread on a Harbor Structure. Report No. 2-4 Concrete in a Marine Environment MEDACHS -Interreg IIIB
Atlantic Space – Project, 2005, 197–9.
[76] D. Patel, R.P. Tiwari, R. Shrivastava, R.K. Yadav, Effective utilization of waste glass powder as the substitution of cement in making paste and mortar, Construct.
Build. Mater. 199 (2019) 406–415.
[77] H. Du, K.H. Tan, Concrete with recycled glass as fine aggregates, ACI Mater. J. 111 (1) (2014) 47–57.
[78] T.D. Dyer, R.K. Dhir, Chemical reactions of glass cullet used as cement component, J. Mater. Civ. Eng. 13 (6) (2001) 412–417.
[79] H. Du, K.H. Tan, Waste glass powder as cement replacement in concrete, J. Adv. Concr. Technol. 12 (11) (2014) 468–477.
[80] N. Schwarz, M. DuBois, N. Neithalath, Electrical conductivity based characterization of plain and coarse glass powder modified cement pastes, Cem. Concr.
Compos. 29 (9) (2007) 656–666.
[81] R. Yu, P. Spiesz, H.J.H. Brouwers, Effect of nano-silica on the hydration and microstructure development of Ultra-High Performance Concrete (UHPC) with a low
binder amount, Construct. Build. Mater. 65 (2014) 140–150.
[82] H.-W. Song, S.-W. Pack, S.-H. Nam, J.-C. Jang, V. Saraswathy, Estimation of the permeability of silica fume cement concrete, Construct. Build. Mater. 24 (3)
(2010) 315–321.
[83] Y. Xu, D.D.L. Chung, Cement of high specific heat and high thermal conductivity, obtained by using silane and silica fume as admixtures, Cement Concr. Res. 30
(7) (2000) 1175–1178.
[84] Y. Wei, W. Yao, X. Xing, M. Wu, Quantitative evaluation of hydrated cement modified by silica fume using QXRD, 27Al MAS NMR, TG–DSC and selective
dissolution techniques, Construct. Build. Mater. 36 (2012) 925–932.
[85] J.Y. Choi, M.K. Joo, B.C. Lho, Effects of silica fume content and polymer-binder ratio on properties of ultrarapid-hardening polymer-modified mortars, Int. J.
Concr. Struct. Mater. 10 (2) (2016) 249–256.
[86] R. Idir, M. Cyr, A. Tagnit-Hamou, Use of fine glass as ASR inhibitor in glass aggregate mortars, Construct. Build. Mater. 24 (7) (2010) 1309–1312.
[87] T.T. Chen, W.C. Wang, H.Y. Wang, Mechanical properties and ultrasonic velocity of lightweight aggregate concrete containing mineral powder materials,
Construct. Build. Mater. 258 (2020), 119550.
[88] K.K. Phoon, T.H. Wee, C.S. Loi, Development of statistical quality assurance criterion for concrete using ultrasonic pulse velocity method, Materials J. 96 (5)
(1999) 568–574.
[89] A.A. Ikpong, The relationship between the strength and non-destructive parameters of rice husk ash concrete, Cement Concr. Res. 23 (2) (1993) 387–398.
[90] A. Nadeem, S.A. Memon, T.Y. Lo, Mechanical performance, durability, qualitative and quantitative analysis of microstructure of fly ash and Metakaolin mortar
at elevated temperatures, Construct. Build. Mater. 38 (2013) 338–347.
[91] P.J. Tumidajski, Effect of slag, silica fume, and finishing on the sorptivities of field concrete, Can. J. Civ. Eng. 33 (8) (Aug. 2006) 1022–1026.
[92] T.H. Wee, A.K. Suryavanshi, S.S. Tin, Evaluation of rapid chloride permeability test (RCPT) results for concrete containing mineral admixtures, Materials J. 97
(2) (2000) 221–232.
[93] T. Iyoda, Y. Dan, Y. Sagawa, H. Hamada, The effect of curing period on the durability of concrete using blast-furnace slag blended cement, ACF Int. Conf. 3
(2008).
[94] M. Mejdi, W. Wilson, M. Saillio, T. Chaussadent, L. Divet, A. Tagnit-Hamou, Investigating the pozzolanic reaction of post-consumption glass powder and the role
of portlandite in the formation of sodium-rich C-S-H, Cement Concr. Res. 123 (July) (2019), 105790.
[95] F. Leng, N. Feng, X. Lu, An experimental study on the properties of resistance to diffusion of chloride ions of fly ash and blast furnace slag concrete, Cement
Concr. Res. 30 (6) (2000) 989–992.

23