Influence of the Integration of Geopolymer Wastes on the Characteristics of Binding Matrices Subjected to the Action of Temperature and Acid Environments

Recycling geopolymer waste, by reusing it as a raw material for manufacturing new geopolymer binding matrices, is an interesting asset that can add to the many technical, technological and environmental advantages of this family of materials in the construction field. This can promote them as promising alternatives to traditional materials, such as Portland cements, which are not so environmentally friendly. Recent studies have shown that the partial replacement of reactive aluminosilicates (metakaolin and fly ash) up to a mass rate of 50% by geopolymer waste does not significantly affect the compressive strength of the new product. In line with these findings, this paper investigates the effects of aggressive environments, i.e., high temperatures (up to 1000 °C) and acid attacks (pH = 2), on the characteristics of these new matrices. Different techniques were used to understand these evolutions: mineralogical analysis by X-ray diffraction (XRD), thermogravimetry-differential thermal analysis (TGA-DTA), mechanical characterization and scanning electron Microscopy (SEM) observations. The results are very satisfactory: in the exposure temperature range explored, the new matrices containing geopolymer waste suffered losses in compressive strength similar to those of the matrices without waste (considered as materials reference). On the other hand, the new matrices exhibited good chemical stability in acid media. These results confirm that the reuse of geopolymer waste is a promising recycling solution in the construction sector.


Introduction
In the building and civil engineering sectors, geopolymers are considered a promising alternative to conventional materials based on Portland cements [1,2]. These materials hold the attention of scientists and engineers owing to their numerous technical, technological and environmental advantages. These environmentally friendly materials are characterized by a simple implementation protocol at temperatures below 100 • C (meaning a low carbon footprint), great diversity and availability of raw materials, good properties in the fresh state, excellent mechanical resistance in the hardened state, high fire resistance, good resistance to acids and sulphates [3][4][5][6][7][8] and excellent immobilization capability of toxic and radioactive elements [9][10][11].
These inorganic materials with amorphous behavior under X-ray have a microstructure made up of a three-dimensional stack of aluminate (AlO 4 ) and silicate (SiO 4 ) tetrahedra, The properties and composition of the activating solution also have a significant impact on the thermal performance of geopolymer materials. Thus, the nature of the microstructure of materials is determined by the amount of soluble silica provided by the solution and influences their performance, both before and after exposure to high temperatures [40,[42][43][44]. Lahoti et al. [44] studied the thermal behavior of geopolymers based on F-type fly-ash activated by sodium silicate solutions of different compositions.
The residual compressive strengths of materials exposed to 900 • C show drastic variations. On the one hand, an 80% reduction was associated with the formation of cracks due to the combined effects of the evaporation of moisture water and of the crystallization of nepheline, a particular sodium aluminosilicate. On the other hand, a 150% improvement was attributed to the densification of the matrix by the formation of inter-particle bonds due to viscous sintering.
The effects of the nature and the concentration of the alkali cation on the thermal performance of geopolymers have been reported in several works [31,[45][46][47]. The geopolymers produced by activators based on potassium silicate show better thermal stability than those obtained from sodium silicate [45]. Moreover, an excessive content of alkali cations lowers the recrystallization temperature of certain phases, such as nepheline, which affects the mechanical performance of the materials [1,44,48].
Another important property that characterizes these materials is their good resistance to acidic media. Several works have focused on the study of this phenomenon [11,[49][50][51][52][53][54]. Although the experimental conditions varied from one study to another, in terms of the concentration of the etching solution used, the nature of the acids (HCl, H 2 SO 4 or acetic acid), the results showed that geopolymers are generally chemically stable with relatively low short-term mechanical strength losses.
The main mechanism causing loss of resistance in an acidic environment is identified with the replacement of the alkaline cations (Na + or K + ) in excess in the material by H + protons. Then the de-alumination of the geopolymer structure follows. This leads to an increase in the pH of the etching batch and to the formation of a lacunar siliceous structure that is less mechanically resistant [50,53].
This work is divided into two parts. The first is devoted to the study of the behavior of new binders after exposure to 200, 400, 600 and 800 • C for 2 h. This behavior is evaluated by DTA-TGA analysis. The second part is dedicated to the evaluation of the characteristics of these new binders subjected for 28 days to the chemical action of acid media. After these various treatments, the residual compressive strengths were measured, microstructural analyses by X-ray diffraction and observations by optical microscopy and by scanning electron microscopy were also performed.

Raw Materials and Constituents
The starting materials were: -Aluminosilicate powders: a commercial metakaolin (MK, from Imerys-France) and fly-ash from coal combustion (FA, from the Jarf Lasfar thermal power station in Morocco). -An alkaline activating solution (AAS) with SiO 2 /Na 2 O molar ratio = 1.2 and 63% by weight of water. This solution is prepared from a mixture of sodium hydroxide (98% purity), commercial sodium silicate (SiO 2 /Na 2 O molar ratio of 2) and water. -Hydrochloric acid (37% HCl) was mixed with distilled water to prepare the acid solutions.
The chemical composition of the starting reactive aluminosilicates MK and FA indicate that they consist mainly of silica SiO 2 and alumina Al 2 O 3 , with small amounts of Fe 2 O 3 and CaO in FA ( Table 1). The amount of SO 3 comes from the sulphates present in the material (CaSO 4 , in particular). The mineralogical analysis of MK and FA by DRX (Figure 1) highlights the presence of an intense dome between 15 and 40 • , which indicates the existence of an amorphous aluminosilicate phase, with the presence of quartz and residual  and CaO in FA ( Table 1). The amount of SO3 comes from the sulphates present in the material (CaSO4, in particular). The mineralogical analysis of MK and FA by DRX ( Figure 1) highlights the presence of an intense dome between 15 and 40°, which indicates the existence of an amorphous aluminosilicate phase, with the presence of quartz and residual kaolinite in MK and mullite and quartz in FA. The D90 particle size distribution parameters are 24.64 and 80.2 μm for metakaolin and fly ash, respectively ( Figure 2).     and CaO in FA ( Table 1). The amount of SO3 comes from the sulphates present in the material (CaSO4, in particular). The mineralogical analysis of MK and FA by DRX ( Figure 1) highlights the presence of an intense dome between 15 and 40°, which indicates the existence of an amorphous aluminosilicate phase, with the presence of quartz and residual kaolinite in MK and mullite and quartz in FA. The D90 particle size distribution parameters are 24.64 and 80.2 μm for metakaolin and fly ash, respectively ( Figure 2).

Material Manufacturing
The manufacture of the materials consists of three stages [24]: 1.
The first stage is preparing the reference geopolymer binders. To this end, two types of materials were developed: A fly-ash-based geopolymer, FAref, obtained by mixing fly-ash powder (FA) and alkaline activation solution (AAS) at an L/S ratio of 0.58. This mixture is cured at 80 • C for 20 h.

2.
To simulate geopolymer waste, the manufactured materials (MKref and FAref) are stored for one year in the open air at room temperature, then they are ground to obtain a fine powder D < 80 µm ( Figure 3) referenced MKRG for materials based on metakaolin and FARG for those based on fly-ash. 3.
The third stage consists in preparing the new matrices by substituting the starting aluminosilicates (fly-ash and metakoalin) with the waste (MKRG and FARG) up to a mass rate of 50%. These mixtures are then subjected to the action of the activation solution under the conditions described in stage 1. Three systems were, thus, prepared: FA/FARG, MK/MKRG and MK/FARG. Note that the materials referenced MKref and FAref are fresh matrices with 0% waste. Table 2 summarizes the data for the manufacturing of the materials.

Material Manufacturing
The manufacture of the materials consists of three stages [24]: 1. The first stage is preparing the reference geopolymer binders. To this end, two types of materials were developed: -A metakaolin-based geopolymer, MKref, obtained by mixing metakaolin powder (MK) and alkaline activating solution (AAS) at a Liquid/Solid ratio (L/S) ratio of 0.83. This mixture is hardened at 60 °C for 5 h.  Table 2 summarizes the data for the manufacturing of the materials.  Analyses of the diffractograms (Figure 4) of MKref and FAref and those of the geopolymer waste (FARG and MKRG) do not indicate any change in the mineralogical composition. We note the presence of a large vitreous dome between 15 and 40° that testifies to the presence of a large quantity of amorphous phase that characterizes the geopoly-  Analyses of the diffractograms (Figure 4) of MKref and FAref and those of the geopolymer waste (FARG and MKRG) do not indicate any change in the mineralogical composition. We note the presence of a large vitreous dome between 15 and 40 • that testifies to the presence of a large quantity of amorphous phase that characterizes the geopolymeric phase, as well as that of crystallized phases coming from the starting aluminosilicates and resisted alkaline attack. These are quartz and mullite for fly-ash-based materials (FAref and FARG) and residual quartz and kaolinite for metakaolin-based materials (MKref and MKRG) [22,24,25]. meric phase, as well as that of crystallized phases coming from the starting aluminosilicates and resisted alkaline attack. These are quartz and mullite for fly-ash-based materials (FAref and FARG) and residual quartz and kaolinite for metakaolin-based materials (MKref and MKRG) [22,24,25].

Fire Resistance
The fire resistance of geopolymer matrices was evaluated according to a protocol used in the literature [31,55]. The samples were first stored for 28 days after demolding. A control batch was stored at room temperature in air. A second batch of samples was dried at 105 ± 1 °C for 24 h in an electric muffle oven, and then it was heated up to the treatment temperature (at a rate of 6.67 °C/min) for a dwell of 2 h before naturally cooling down to room temperature inside the oven in order to avoid thermal shock. Four treatment temperatures were selected: 400, 600, 800 and 1000 °C ( Figure 5). These specimens were loaded in compression to determine the compressive strengths.

Fire Resistance
The fire resistance of geopolymer matrices was evaluated according to a protocol used in the literature [31,55]. The samples were first stored for 28 days after demolding. A control batch was stored at room temperature in air. A second batch of samples was dried at 105 ± 1 • C for 24 h in an electric muffle oven, and then it was heated up to the treatment temperature (at a rate of 6.67 • C/min) for a dwell of 2 h before naturally cooling down to room temperature inside the oven in order to avoid thermal shock. Four treatment temperatures were selected: 400, 600, 800 and 1000 • C ( Figure 5). These specimens were loaded in compression to determine the compressive strengths. meric phase, as well as that of crystallized phases coming from the starting aluminosilicates and resisted alkaline attack. These are quartz and mullite for fly-ash-based materials (FAref and FARG) and residual quartz and kaolinite for metakaolin-based materials (MKref and MKRG) [22,24,25].

Fire Resistance
The fire resistance of geopolymer matrices was evaluated according to a protocol used in the literature [31,55]. The samples were first stored for 28 days after demolding. A control batch was stored at room temperature in air. A second batch of samples was dried at 105 ± 1 °C for 24 h in an electric muffle oven, and then it was heated up to the treatment temperature (at a rate of 6.67 °C/min) for a dwell of 2 h before naturally cooling down to room temperature inside the oven in order to avoid thermal shock. Four treatment temperatures were selected: 400, 600, 800 and 1000 °C ( Figure 5). These specimens were loaded in compression to determine the compressive strengths.

Resistane to Acid Attack
The behavior of these matrices in an acid environment was previously investigated [50]. After demolding, the samples were dried for 24 h at room temperature in air and then immersed for 28 days in various solutions maintained at 25 • C in hermetic plastic bags HCl solution (AC) with pH = 2. This solution was not renewed during the treatment period.

-
The basic chemical composition of the powders was determined by X-ray fluorescence by using a Thermo ARL 9800XP spectrometer equipped with an X-ray tube with 3 kW maximum power (30 kV and 80 mA). - The mineralogical composition of the materials was determined using XRD on powder, using a Bruker D8 copper anticathode diffractometer (CuKα 2θ (λKα = 1.5418 Å) operating at 40 mA and 40 kV. The results were collected for 2θ in the range 10 to 70 • with an increment of 0.059 • . The identification of the crystalline phases was made using version 2.1 of the DIFFRAC EVA software. -Differential thermal analysis (DTA) and thermogravimetric analysis (TGA) were performed by using a Shimadzu DTG-60 H instrument operating at a heating rate of The fracture surfaces of the materials before and after the chemical treatment were analyzed using a Tescan VEGA 3 scanning electron microscope and Keyence ZHX-700 type digital optical microscope. - Compressive strength testing was conducted on a Perrier type tensile machine, equipped with a 200 kN capacity loading cell. Four tests were performed for each material formulation. - The pH of the AC and AD acid media was determined using a standard pH-measurement device. Three pH measurements were performed for each material formulation in each medium.

Fire Resistance
DTA-TGA TGA-DTA thermal analyses are suitable techniques for determining the various phenomena that occur in a material subjected to the action of heat. Analyses were performed between room temperature and 1000 • C on samples from the three systems FA/FARG, MK/MKRG and MK/FARG ( Figure 6). DTA/TGA analyses show that FARG and MKRG materials simulating waste exhibit the same thermal behavior as MKref and FAref reference materials in the fresh state with the exception of water loss slightly higher for the fresh materials. Regardless of the nature of the sample, most of the measured mass loss on the TGA curves occurs below 300 • C. This loss is marked on the DTA curves by a strong endothermic effect around 75 • C, which is assigned to the loss of free water [30,41]. In the particular case of the metakaolin-based materials, MK/MKRG and MK/FARG (Figure 6b,c), this effect includes a second endothermic peak with lower intensity around 125 • C, which is attributed to the dehydration of zeolite structures present in the geopolymer gel [56,57].
Beyond 300 • C, two distinct behaviors are observed: the samples of the MK/MKRG system do not undergo any mass loss (Figure 6b), whereas those of the FA/FARG or MK/FARG system (Figure 6a,c) show a second loss, weaker and marked by an endothermic effect centered around 400 • C, which is attributed to the dehydroxylation of portlandite Ca(OH) 2 [41,54]. Finally, we note that the increasing addition of recycled geopolymers, to the detriment of precursor sources of aluminosilicate, increases the overall mass loss. This could be explained by the effect of substituting FA and MK materials with others (FARG and MKRG) with a higher loss on ignition (Table 1). to the detriment of precursor sources of aluminosilicate, increases the overall mass loss. This could be explained by the effect of substituting FA and MK materials with others (FARG and MKRG) with a higher loss on ignition (Table 1).

Mineralogical Analysis
XRD mineralogical analysis has been used to demonstrate phase transformations that occur after exposure to different temperatures. The results recorded before and after heat treatment at 400, 600, 800 and 1000 °C are shown in Figure 7.
It is noted that the diffractograms of the reference matrices FAref and MKref (without waste), as well as those of the samples with 50 wt.% of waste, presented, before heat treatment, a large glass dome in the range 17 < 2θ < 33°, which characterizes the amorphous

Mineralogical Analysis
XRD mineralogical analysis has been used to demonstrate phase transformations that occur after exposure to different temperatures. The results recorded before and after heat treatment at 400, 600, 800 and 1000 • C are shown in Figure 7.
It is noted that the diffractograms of the reference matrices FAref and MKref (without waste), as well as those of the samples with 50 wt.% of waste, presented, before heat treatment, a large glass dome in the range 17 < 2θ < 33 • , which characterizes the amorphous geopolymer phase [20,24]. Analysis also reveals the presence of crystalline phases, such as quartz and mullite in fly-ash-based matrices (Figure 7a,b), and quartz and residual kaolinite in metakaolin-based matrices (Figure 7c-e). No microstructural change was observed after exposure of these materials to 400 • C, which reflects the good thermal stability of the geopolymer gel at this temperature.
icates: unlike metakaolin, fly-ash contains appreciable amounts of the oxides Fe2O3 (5.80%) and K2O + Na2O (2.66%) ( Table 1). These oxides are known for their role as fluxes and mineralizing agents that contribute to the crystallization of nepheline [1,40,48,58]. The substitution of fly-ash by FARG increases the contents of these oxides in the materials of the FA/FARG system, which favors the neo-crystallization phenomenon in these matrices.
At the highest temperatures, all the exposed materials are subjected to important transformations: amplification of nepheline crystallization at the expense of the amorphous geopolymer phase (disappearance of the glassy dome) and gradual dissolution of quartz and mullite. At 1000 °C, the materials appear to be exclusively composed of nepheline. Moreover, the treatment at 600 • C has different consequences depending on the nature of the matrices. We notice the crystallization of a new phase, nepheline NaAlSi 3 O 8 in fly-ash-based materials, the characteristic peaks of which are amplified following the introduction of waste. This phenomenon is not observed in binders based on metakaolin. This is attributed to the difference in the chemical composition of the starting aluminosilicates: unlike metakaolin, fly-ash contains appreciable amounts of the oxides Fe 2 O 3 (5.80%) and K 2 O + Na 2 O (2.66%) ( Table 1). These oxides are known for their role as fluxes and mineralizing agents that contribute to the crystallization of nepheline [1,40,48,58]. The substitution of fly-ash by FARG increases the contents of these oxides in the materials of the FA/FARG system, which favors the neo-crystallization phenomenon in these matrices.
At the highest temperatures, all the exposed materials are subjected to important transformations: amplification of nepheline crystallization at the expense of the amorphous geopolymer phase (disappearance of the glassy dome) and gradual dissolution of quartz and mullite. At 1000 • C, the materials appear to be exclusively composed of nepheline.

Mechanical Behavior
The mechanical behavior of the matrices was studied by analyzing their compressive strengths in order to demonstrate the effect of the addition of recycled geopolymers on the performance of the materials exposed to temperatures of 400, 600, 800 and 1000 • C. The results are shown in Figure 8. To better assess the effect of the two factors, namely the substitution rate and the temperature, the evolution of the normalized residual compressive strength is depicted in Figure 9.

Mechanical Behavior
The mechanical behavior of the matrices was studied by analyzing their compressive strengths in order to demonstrate the effect of the addition of recycled geopolymers on the performance of the materials exposed to temperatures of 400, 600, 800 and 1000 °C. The results are shown in Figure 8. To better assess the effect of the two factors, namely the substitution rate and the temperature, the evolution of the normalized residual compressive strength is depicted in Figure 9. Prior to the heat treatment implementation, the MKref and FAref reference matrices had compressive strengths of 63 and 54 MPa, respectively. The high content of reactive amorphous phases of the starting aluminosilicates (fly-ash and metakaolin) and the adequate activation conditions applied having favored the phenomenon of geopolymerization, a compact microstructure was obtained (as can be observed on the fractographies shown in Figure 10) as well as high compressive strengths [14,20].
The introduction of recycled geopolymer waste up to a substitution rate of 40 wt.% at the expense of reactive aluminosilicates had no noticeable effect on the strength of the Prior to the heat treatment implementation, the MKref and FAref reference matrices had compressive strengths of 63 and 54 MPa, respectively. The high content of reactive amorphous phases of the starting aluminosilicates (fly-ash and metakaolin) and the adequate activation conditions applied having favored the phenomenon of geopolymerization, a compact microstructure was obtained (as can be observed on the fractographies shown in Figure 10) as well as high compressive strengths [14,20].
The introduction of recycled geopolymer waste up to a substitution rate of 40 wt.% at the expense of reactive aluminosilicates had no noticeable effect on the strength of the new matrices. This behavior is due to the establishment of interfacial bonds between the gel of the newly formed geopolymer and the particles of integrated waste.
For a substitution rate of 50 wt.%, a reduction of about 20% in the compressive strength occurs although the observations by optical microscopy and SEM of the fracture surfaces of the specimens show a compact structure similar to that of the reference matrices (Figures 10 and 11). These features are assigned to the reduction in the volume of the neo-formed geopolymer gel, which is the source of the material's cohesion. A phenomenological model was previously proposed to explain this good behavior and the mechanisms governing the good adhesion between the reference matrices and the geopolymer waste aggregates [24]. new matrices. This behavior is due to the establishment of interfacial bonds between the gel of the newly formed geopolymer and the particles of integrated waste. For a substitution rate of 50 wt.%, a reduction of about 20% in the compressive strength occurs although the observations by optical microscopy and SEM of the fracture surfaces of the specimens show a compact structure similar to that of the reference matrices (Figures 10 and 11). These features are assigned to the reduction in the volume of the neo-formed geopolymer gel, which is the source of the material's cohesion. A phenomenological model was previously proposed to explain this good behavior and the mechanisms governing the good adhesion between the reference matrices and the geopolymer waste aggregates [24]. The trend of compressive strength variation as a function of the treatment temperature is in line with the data reported by A. Celik [59] and M. Lahoti [43] for metakaolinbased geopolymer matrices and by X. Jiang [41] for fly-ash-based materials. The increase in the temperature of the heat treatment results in a degradation of the mechanical strength of the matrices. However, there is a weak influence of increasing the exposure temperature for additions less than 40 wt.% (Figures 8 and 9).
Exposure to 400 °C causes a significant decrease in mechanical strength of about 40%, regardless of the system studied. This reduction is mainly due to the increase in porosity caused by the escape of free and zeolitic water revealed by TGA/DTA ( Figure 6) and confirmed by microscopic and naked eye observations that reveal the development of fine microcracks within the matrices (Figures 10 and 11) [7,41,44]. The trend of compressive strength variation as a function of the treatment temperature is in line with the data reported by A. Celik [59] and M. Lahoti [43] for metakaolin-based geopolymer matrices and by X. Jiang [41] for fly-ash-based materials. The increase in the temperature of the heat treatment results in a degradation of the mechanical strength of the matrices. However, there is a weak influence of increasing the exposure temperature for additions less than 40 wt.% (Figures 8 and 9).
Exposure to 400 • C causes a significant decrease in mechanical strength of about 40%, regardless of the system studied. This reduction is mainly due to the increase in porosity caused by the escape of free and zeolitic water revealed by TGA/DTA ( Figure 6) and confirmed by microscopic and naked eye observations that reveal the development of fine microcracks within the matrices (Figures 10 and 11) [7,41,44].
The treatment at 600 • C causes approximately 50% degradation of the mechanical strength of the materials based on metakaolin (MK/MKRG and MK/FARG), which is lower than that (of about 60%) recorded for the fly-ash-based materials (Figures 8 and 9). This difference is attributed to the decomposition of the geopolymer binding phase due to the early crystallization of nepheline that occurs in the fly-ash-based matrices (Figure 7) [7,32]. The treatment at 600 °C causes approximately 50% degradation of the mechanical strength of the materials based on metakaolin (MK/MKRG and MK/FARG), which is lower than that (of about 60%) recorded for the fly-ash-based materials (Figures 8 and 9). This difference is attributed to the decomposition of the geopolymer binding phase due to the early crystallization of nepheline that occurs in the fly-ash-based matrices (Figure 7) [7,32].  Exposure to 800 or 1000 °C further decreases the mechanical strength due to the significant degradation of the microstructure of the various matrices. Indeed, a pronounced development of the porous phase and the appearance of macrocracks (Figures 10 and 11) resulting from the total transformation of the binding geopolymer phase into nepheline, Figure 11. Surfaces of the materials as a function of the recycled geopolymer rate at different temperatures: RT (a), 400 • C (b) and 800 • C (c).
Exposure to 800 or 1000 • C further decreases the mechanical strength due to the significant degradation of the microstructure of the various matrices. Indeed, a pronounced development of the porous phase and the appearance of macrocracks (Figures 10 and 11) resulting from the total transformation of the binding geopolymer phase into nepheline, are noted (Figure 7) [7,43,44]. Figure 12 shows the evolution of the pH of the material immersion baths over four weeks, for the two acidic mediums AC and AD. In the case of AC solutions, the evolution of the pH appears to be independent of the nature of the immersed material: the pH of the solutions increases rapidly and reaches the base range during the first week before stabilizing beyond the third week. It is also noted that the pH values increase when the waste rate in the materials is increased. This development indicates a strong and rapid interaction of the material with the acid bath.

Resistane to Acid Attack
According to J. Kwasny et al. [53], the increase in pH is due to the substitution of the Na + and K + alkaline cations of the geopolymers by the H + protons of the acidic medium. This mechanism is discussed by X.X. Gao et al. [50]. According to these authors, on contact with the acid solution, the materials release alkaline species present in excess after geopolymerization.
These species then form compounds of a basic nature. This is because the excess alkali hydroxide in materials reacts with carbon dioxide present in the air to form carbonate species of the alkaline element, such as NaHCO 3 and Na 2 CO 3 , that dissolve in solution and neutralize the H + proton. The increase in the pH of solutions after immersion of materials containing waste can therefore be explained by the gradual increase in the amount of alkaline species provided by the waste. Exposure to 800 or 1000 °C further decreases the mechanical strength due to the significant degradation of the microstructure of the various matrices. Indeed, a pronounced development of the porous phase and the appearance of macrocracks (Figures 10 and 11) resulting from the total transformation of the binding geopolymer phase into nepheline, are noted (Figure 7) [7,43,44]. Figure 12 shows the evolution of the pH of the material immersion baths over four weeks, for the two acidic mediums AC and AD. In the case of AC solutions, the evolution of the pH appears to be independent of the nature of the immersed material: the pH of the solutions increases rapidly and reaches the base range during the first week before stabilizing beyond the third week. It is also noted that the pH values increase when the waste rate in the materials is increased. This development indicates a strong and rapid interaction of the material with the acid bath.

Resistane to Acid Attack
According to J. Kwasny et al. [53], the increase in pH is due to the substitution of the Na + and K + alkaline cations of the geopolymers by the H + protons of the acidic medium. This mechanism is discussed by X.X. Gao et al. [50]. According to these authors, on contact with the acid solution, the materials release alkaline species present in excess after geopolymerization.
These species then form compounds of a basic nature. This is because the excess alkali hydroxide in materials reacts with carbon dioxide present in the air to form carbonate species of the alkaline element, such as NaHCO3 and Na2CO3, that dissolve in solution and neutralize the H + proton. The increase in the pH of solutions after immersion of materials containing waste can therefore be explained by the gradual increase in the amount of alkaline species provided by the waste.  Analysis of the pH in the AD medium provides further lessons. Indeed, all the baths analyzed from the third week had an acidic pH, which indicates that the materials release less basic species in the solutions. This suggests that the neutralization of basic species by the acidic solution is limited in time [50]. Figure 13 shows the X-ray diffractograms of the geopolymeric materials of the three systems before and after acid attack for four weeks. No change from non-immersed materials (treated at room temperature) was noted. These results are explained because the amorphous phase of the geopolymers and the crystalline phases were not affected and no new phase was detected. They are in good agreement with other works that emphasized the chemical stability of cross-linked structures of geopolymeric binding matrices [50][51][52][53][54] in aggressive acidic environments. They also prove that the substitution of reactive aluminosilicates by recycled geopolymers had no significant effect on the chemical stability of the newly formed matrices. In order to better assess the impact of the acid attack after four weeks immersion, the changes in compressive strength and normalized strength were determined (Figures 14  and 15). In order to better assess the impact of the acid attack after four weeks immersion, the changes in compressive strength and normalized strength were determined (Figures 14 and 15). It is noted that regardless of the system and the substitution rate considered, the immersion of materials in acid solutions led to about a 20% reduction in strength compared to non-immersed materials. These losses in mechanical strength are similar to those noted for the reference materials (without waste) and of the same order as those reported in the literature [11,[50][51][52][53][54]. They are attributed to a de-alumination mechanism of the geopolymer phase, which results in the formation of a lacunar siliceous structure [50,53]. However, this phenomenon appears to be very limited compared to the low mass losses noted after the acid attack (1.4% at most). This observation is confirmed by the optical microscopy analyses of the fracture surfaces (Figure 16), which show that the matrices after immersion are similar to those not attacked (a small increase in the size of the pores). It is noted that regardless of the system and the substitution rate considered, the immersion of materials in acid solutions led to about a 20% reduction in strength compared to non-immersed materials. These losses in mechanical strength are similar to those noted for the reference materials (without waste) and of the same order as those reported in the literature [11,[50][51][52][53][54]. They are attributed to a de-alumination mechanism of the geopolymer phase, which results in the formation of a lacunar siliceous structure [50,53]. However, this phenomenon appears to be very limited compared to the low mass losses noted after the acid attack (1.4% at most). This observation is confirmed by the optical microscopy analyses of the fracture surfaces (Figure 16), which show that the matrices after immersion are similar to those not attacked (a small increase in the size of the pores).

Conclusions
This work is a continuation of the studies we are conducting on the recyclability/recycling of geopolymer materials. The present study was performed to assess the impact

Conclusions
This work is a continuation of the studies we are conducting on the recyclability/recycling of geopolymer materials. The present study was performed to assess the impact

Conclusions
This work is a continuation of the studies we are conducting on the recyclability/recycling of geopolymer materials. The present study was performed to assess the impact of the integration of geopolymer waste up to 50 wt.% on fire resistance, and resistance toward chemical attack by immersion in acidic environments was also assessed. The results indicate that: -Regardless of the substitution rate, the hardened materials showed, at room temperature (RT), good mechanical performance with compressive strengths greater than 45 MPa. -Regardless of the waste substitution rate, the materials showed good fire resistance similar to that of waste-free materials. They maintained compressive strength greater than 20 MPa after treatment at 600 • C. The degradations observed included increased porosity resulting from the release of residual water and the crystallization of nepheline, which depletes the binding phase. - The crystallization of nepheline was earlier (600 • C) in the fly-ash-based materials FA/FARG. This is due to the high content of Fe 2 O 3 , K 2 O and Na 2 O oxides in the starting fly-ash, known for their role as fluxes and mineralizing agents. However, this phenomenon was only observed from 800 • C for the metakaolin-based materials MK/MKRG and MK/FARG. -Regardless of the waste substitution rate, the materials demonstrated very good resistance to acid attack (pH = 2) with resistance losses not exceeding 28%. - The increase in the pH of the attack baths confirms the mechanism of release of sodium ions present in excess in the materials, which enriches the baths in alkaline species by their combination with atmospheric CO 2 . - The materials were chemically stable in the media studied. Indeed, no significant impact on the mineralogical composition nor on the mass losses was recorded. - The increase in the rate of substitution of geopolymer waste had the effect of increasing the concentration of excess sodium ions in the material and, thus, that of the pH.
These results are encouraging and motivating for safeguarding the environment through the reuse of geopolymer wastes as additives in geopolymeric building materials.