Incineration Process

One tonne of MSWs and 500 kg of organic MSWs were incinerated in a rotary kiln incinerator (KALFRISA MODEL ROT-1500); incineration temperatures ranged from 600 to 1270ºC in the first combustion chamber. After the incineration process, 10% of the MSWIBA1 and 11% of the MSWIBA2 were produced. The fraction (0–5 mm) of the two MSWIBAs was pretreated after sampling by using magnetic separation to remove metals, then ground in a ball mill to obtain powders with a particle size of less than 80 µm. Ordinary Portland Cement (OPC) class CEMI 42.5 N was obtained from Chlef Cement Plants (ECDE). The physical properties (specific gravity and specific surface area (SSA)) of the materials are presented in Table 2. Standard sand was used to prepare the mortars.

Fused Bead Preparation, Chemical Composition on Oxides

The solution to be analyzed was prepared by the fusion of 0.5 g of BA powder mixed with 1.5 g of sodium carbonate (Na₂CO₃) in a muffle furnace at 1000ºC for 1 h. After the fusion was complete, the fused sample was dissolved in hydrochloric acid (HCl), and the cement standard method testing (1997) was followed in the analysis of the chemical oxides in the solution [26]. The chemical composition of cement Portland was analyzed using XRF analysis.

Loss On Ignition

The loss on ignition (LOI) was measured by igniting the fines at 950±25ºC in a muffle furnace according to EN 196-2 [27].

Mineralization

Heavy metal concentrations were determined after microwave acid digestion. 0.25 g of each BA was placed in Teflon tubes, and 2 mL of hydrogen peroxide (H₂O₂), 6 mL of nitric acid (HNO₃), 3 mL of hydrochloric acid (HCl), and 2 mL of fluoric acid (HF) were added to the Teflon tubes. The mixtures were digested at 180ºC for one hour in the microwave oven (Multiwave GO, Anton Paar); after the mixtures had cooled completely, the excess HF was neutralized using 6 mL boric acid (H₃BO₃) at 160ºC for 30 min. The final digested mixtures were diluted with ultrapure water to 50 mL and then filtered through a membrane filter (0.45 µm).

Inductively Coupled Plasma Mass Spectroscopy

The solubility of heavy metals in BAs and leachates was determined by inductively coupled plasma mass spectroscopy (ICP-MS, Agilent Technologie, 7700). The leaching test (24 h shaking, L/S10, 250 rpm) was carried out according to EN 12457-4 [28].

Ion Chromatography

Ion chromatography (IC, Metrohm) detected chloride and sulfate in leachates. Eluent (1 mM NaHCO₃+3.2 mM Na₂CO₃) was used.

X-ray Diffraction

The mineral phases of the BAs were identified by X-ray diffraction (XRD, Rigaku Miniflex 600) with Cu Kα radiation (λ =1.54059 Å), a detection angle between 5 and 90º (2-theta), and a step size of 0.01. The phases were identified using X’Pert HighScore Plus.

Fourier Transform Infrared Spectroscopy

Fourier transform infrared spectroscopy (FTIR) analysis of compressed KBr-BA pellets was carried out using a Thermo-Scientific Nicolet 8700 spectrometer (USA). The FTIR spectra were recorded in the wave number range 4000-400 cm^-1 with a resolution of 4 cm^-1.

Physical Properties of Cement Pastes

The standard consistency was determined for all cement pastes. The water required for normal consistency was determined by paste penetration tests with a variation in the quantity of water for each test [29], using a manual Vicat apparatus. A good test was achieved after obtaining a distance between the plunger and the base plate containing the paste d = 6±2 mm.

Initial setting time was measured using standard consistency pastes and an automatic Vicat apparatus (Toni Technik, ToniSET) according to EN 196-3. Soundness was determined using the Le Chatelier apparatus following EN 196-3.

Mechanical Properties of Mortar Samples

The mortars were prepared according to EN 196–1 [30]. The samples contained 450 g of cement with (0 wt%, 5 wt%, 10 wt%, and 20 wt%) MSWIBAs, 225 g of water, and 1350 g of sand. These ingredients were mixed in a standard laboratory mixer, and the mixtures obtained were cast into 4×4×16 cm molds, then sealed with polyethylene sheets for 24 h. After demolding, the mortars were stored in a curing room at a temperature of 20ºC and a relative humidity of 96%. The compressive and flexural strengths of the mortars were tested after 2, 7, 28, and 90 days of curing. The results obtained were expressed as the averages of three samples.

Fig. 1 presents the outline of the experimental protocol followed in this study.

Chemical Composition of both MSWIBA, OPC

The chemical composition of the investigated powders (MSWIBA1, MSWIBA2, and OPC) is shown in Table 3. The chemical composition of MSWIBAs is influenced by the MSW composition input in the incinerator, the process parameters (temperature, air, turbulence, residence time), and the type of the incinerator [31, 32]. A comparison of the two bottom ashes shows that MSWIBA1 has a higher SiO₂ content (37.86%) than MSWIBA2 (15.80%) and a lower CaO content (25.23%) than MSWIBA2 (39.06%). The lower SiO₂ content in MSWIBA2 is due to glass sorting before the incineration process, while the higher CaO content in MSWIBA2 is attributed to the composition of the raw MSW, which is rich in organic MSW (putrescible, paper, cardboard). The Al₂O₃, Fe₂O₃, and MgO contents are relatively lower in both MSWIBAs compared to their SiO₂ and CaO contents; a similar composition is reported in the literature [33, 34]. Loss on ignition (LOI) for both MSWIBAs is very high, indicating the presence of large amounts of carbonaceous material in MSWIBA1 and MSWIBA2 [35]. OPC contains the same oxides as the two MSWIBAs, but there is a difference in their contents: it has higher CaO (62.95%) and lower SiO₂ (22.16%).

Heavy Metals and Leachability

Heavy metal concentrations in MSWIBA1 and MSWIBA2 are presented in Table 4. The elements Cr, Ni, Cu, Zn, Cd, Sb, Ba, and Pb are the main heavy metals which are remarkable with higher concentrations in MSWIBA1 than in MSWIBA2. The concentrations of Cr, Ni, Cu, Zn, Cd, Sb, Ba, Pb in MSWIBA2 are 64%, 62%, 15%, 42%, 95%, 51%, 13% and 49% respectively, lower than those in MSWIBA1. Dezhi et al. (2018) found that the bottom ashes contain high concentrations of Pb, Cu, Zn, Cr, and Mn and lower concentrations of Cd [36]. The lower heavy metal concentrations in MSWIBA2 are linked to the composition of the raw MSW, which is free of any metal sources thanks to the selective sorting carried out before the incineration process. These findings align with the results of Dezhi Shi et al. (2018), which demonstrated that removing hazardous waste, including mercury thermometers, fluorescent lamps, and batteries before incineration reduces heavy metal contents in bottom ashes [36].

Fig. 1.

Diagram of experimental protocol.

Fig. 2.

XRD analysis of bottom ashes a) MSWIBA1, b) MSWIBA2.

The contents of heavy metals from leachates of MSWIBA1 and MSWIBA2 are summarized in Table 5. The concentrations of heavy metals in leachates from MSWIBA1 and MSWIBA2 are approximately similar. The leachate from MSWIBA2 contains high concentrations of Ba, Mo, and Cr, while the leachate from MSWIBA1 contains a high concentration of Sb. It can be remarked that the leachate from MSWIBA2 has a higher content of contaminants compared to the leachate from MSWIBA1, which might be due to the high surface area of MSWIBA2 (6636.64 cm²/g). These results are comparable to those of Caprai et al. (2019), who detected excessive leaching of Sb, Cr, Cu, and Mo heavy metals [37].

The concentration of sulfate ion (SO₃^2-) in the leachate from MSWIBA1 is higher than that in the leachate from MSWIBA2. In contrast, the concentration of chloride ion (Cl⁻) in the leachate from MSWIBA2 is higher than that in the leachate from MSWIBA1. The leaching results for the two MSWIBAs indicate that the concentrations of heavy metals and sulfates are below the hazardous limit [28]. According to European landfill limits, chloride in the leachate is in the hazardous waste range, which accords with results reported in the literature. Tang et al. (2015) found that the leaching of chloride and sulfate in the investigated bottom ashes exceeds the relevant limit values [38].

XRD Analysis

For XRD analysis, Fig. 2 shows that the mineralogical composition of the two MSWIBAs differs slightly. The predominant crystalline phases in MSWIBA1 (Fig. 1a)) are quartz (PDF. No 96-500-0036) and clinoenstatite (PDF. No 96-900-2716) with strong peaks; the gypsum (PDF. No 96-901-3170), calcite (PDF.No 96-101-0929), halite (PDF. No 96-900-3309), magnetite (PDF. No 96-900-2324), tricalcium aluminate (PDF. No 96-901-4360), and brownmillerite (PDF. No 96-901-4662) present with weak peaks. The MSWIBA2 (Fig. 1b)) contains alite (PDF. No 96-154-0706) with a high peak, calcite (PDF.No 96-101-0929), and halite (PDF. No 96-900-3309) with many minor peaks, it contains also tricalcium aluminate (PDF.No 96-901-4360), coesite (PDF. No 96-900-7171), and portlandite (PDF. No 96-900-0113). Some of the phases detected in both MSWIBAs are reported in previous studies [39, 40]. The presence of alite (cement component) in MSWIBA2 is due to the high incineration temperature. Based on the chemical analysis, both MSWIBAs contain Si, Ca, Al, Fe, and Mg; the results obtained by XRD analysis confirm this composition, and these elements are distributed in detected phases [41].

Fig. 3.

FTIR analysis of bottom ashes: a) MSWIBA1, b) MSWIBA2.

FTIR Analysis

The FTIR spectra of MSWIBA1 and MSWIBA2 are shown in Fig. 3. It can be seen that both spectra are approximately similar. The large bands around 3435 and 3428 cm⁻¹ are linked to vibrations of O-H stretching of bonded and non-bonded hydroxyl groups and water, indicating some humidity in both MSWIBAs [42]. Carbonate (C-O) vibrations are identified by small peaks centered at 2520, 1797, and 714 cm⁻¹, as well as strong peaks at 1430 and 875 cm⁻¹ [43]. Absorbances at 1045, 510, and 459 cm⁻¹ indicate the presence of Si-O-Si (quartz) [44, 45]. The low peaks located at 2925 cm⁻¹ are related to C-H stretching (as) of the aliphatic methylene group [42], and the peaks observed at 2360 cm⁻¹ are due to CO₂ molecules [46]. These results agree with the chemical and XRD analyses, which confirm that the main components of both bottom ashes are Ca, represented by calcium carbonate [47], and Si, represented by quartz.

Flexural Strengths

Fig. 4 shows the impact of MSWIBA1 and MSWIBA2 on the flexural strengths of mortars at 2, 7, 28, and 90 days of curing. The flexural strengths of the cements containing 5 wt% and 10 wt% levels of MSWIBA1 and MSWIBA2 show a small increase at 2 days and a slight decrease at 7 days compared to the strength values of the control cement, which are 3.46 MPa at 2 days and 6.90 MPa at 7 days. The flexural strengths of cement containing 20 wt% of MSWIBA1 are 3.46 and 5.16 MPa at 2 and 7 days, respectively; the flexural strengths of cement with 20 wt% of MSWIBA2 are 3.41 and 4.77 MPa at 2 and 7 days, respectively. Flexural strength readings at 7 days of curing are lower than that of the control mortar; this is because the cement has been diluted by the addition of 20 wt% of bottom ashes.

After 28 and 90 days of curing, the flexural strengths of the reference cement are 7.68 and 7.79 MPa, respectively. Flexural strength decreases gradually with increasing contents of both bottom ashes in the cement. After 28 days, replacing 5 wt%, 10 wt%, and 20 wt% of the cement with MSWIBA1 gives flexural strengths of 7.02, 6.97, and 6.24 MPa, respectively, while at 90 days, the flexural strengths are 7.21, 6.86, and 5.98 MPa, respectively. On the other hand, the flexural strengths of cements with 5 wt%, 10 wt%, and 20 wt% of MSWIBA2 are 7.30, 6.73, and 6.14 MPa, respectively, at 28 days; furthermore, the flexural strengths are 7.55, 6.79, and 5.60 MPa, respectively, at 90 days.

Compressive Strengths

The effect of MSWIBA1 and MSWIBA2 on the compressive strengths of mortars at 2, 7, 28, and 90 days of curing is depicted in Fig. 5. The compressive strengths of the reference cement at 2 and 7 days are 15.78 and 41.98 MPa, respectively. It can be noticed that the compressive strengths of cements with 5 wt% and 10 wt% of MSWIBA1 after 2 days are 19.68 and 20.86 MPa, respectively, while the compressive strengths after 7 days of these mixtures are 42.94 and 45.12 MPa, respectively. The same applies to the cements prepared with 5 wt% and 10 wt% of MSWIBA2; after 2 days, the compressive strengths are 21.30 and 21.83 MPa, respectively, while the compressive strengths after 7 days are 44.93 and 43.93 MPa, respectively. Compared with the compressive strength results at 2 and 7 days for the control cement, there is an increase in the compressive strengths of cements with 5 wt% and 10 wt% of MSWIBA2. This growth in strength at early ages indicates that the addition of MSWIBA2 accelerates the hydration process during the early period. The hydration occurs through the reaction of different compounds in the cement and BAs (C₃A, C₃S, CaO) with water, forming hydrates like ettringite, CH, and C-S-H. As seen in the previous results for the setting time, the heavy metals delay the hydration process of cement with MSWIBA1. Thus, the compressive strengths at 2 and 7 days for this blended cement increased compared to those of the control cement. This increase is attributed to the nucleation of C-S-H when all trace metals precipitate with calcium hydroxide [50].

Fig. 4.

Effect of bottom ashes on flexural strengths of hardened blended cement: a) MSWIBA1, b) MSWIBA2.

Fig. 5.

Effect of bottom ashes on compressive strengths of hardened blended cement: a) MSWIBA1, b) MSWIBA2.

The compressive strengths of cements with 20 wt% of MSWIBA1 and 20 wt% of MSWIBA2 after 2 days are 16 and 14.29 MPa, respectively, while the compressive strengths after 7 days of these mixtures are 32.59 and 29.11 MPa, respectively. There is a remarkable decrease in compressive strengths compared to those of cements with 5 wt% and 10 wt% of both types of bottom ashes. This reduction is related to the decrease in the quantity of alite (C₃S) when replacing cement with 20 wt% of both MSWIBAs [37]. It is observed that the compressive strengths of cement with 20 wt% MSWIBA2 are slightly higher than those of cement with 20 wt% MSWIBA1. This is explained by the lower contents of heavy metals in MSWIBA2 and its higher fineness, allowing MSWIBA2 to react more easily with cement.

At delay ages, the compressive strengths of cements with 5 wt% and 10 wt% of MSWIBA1 are 57.31 MPa and 57.18 MPa at 28 days, and 66.41 MPa and 63.67 MPa at 90 days, respectively. For cements prepared with 5 wt% and 10 wt% of MSWIBA2, the compressive strengths are 58.90 MPa and 57.80 MPa at 28 days, and 65.86 MPa and 61.12 MPa at 90 days, respectively. By comparing these results with those of the control mortar (60.94 MPa at 28 days and 71.16 MPa at 90 days), it can be seen that there is a slight decrease in compressive strengths because the cement was replaced with small proportions of BAs (5 wt% and 10 wt%). Loginova et al. (2021) found in their study that replacing cement 42.5N with 5 wt% of BA slightly decreased the compressive strengths by 7% and 2% at 28 and 91 days, respectively, compared with the control mortar, which agrees with our results [21]. Loginova et al. (2023) also found a slight decrease in the compressive strength of cement 52.5R blended with 10 wt% of BA after 2 days, with a remarkable decrease in strength after 28 days [17]. For cement preparations containing 20 wt% of MSWIBA1 and 20 wt% of MSWIBA2, the compressive strengths are 42.39 MPa and 36.44 MPa, respectively, after 28 days of curing, while after 90 days, the compressive strengths are 47.28 MPa and 40.91 MPa, respectively. The reduction in compressive strengths is related to the decrease in the content of belite (C₂S), a phase responsible for strength at delay ages, when cement is substituted with a high level of BAs (20 wt%). Cheng et al. (2022) demonstrated that the reduction in compressive strength increases with higher BA content in cement after 28 days of curing [52]. It can also be observed that the compressive strengths of cement with 20 wt% of MSWIBA1 are higher than those of cement with 20 wt% of MSWIBA2; the main explanation for this observation is the lower pozzolanic reactivity of MSWIBA2 at delay ages. The MSWIBA1 contains a high amount of SiO₂ (37.86 %.), so it reacted with CH from the hydration of active compounds of cement and formed additional C-S-H. The results of the compressive strengths of cement types containing high amounts of MSWIBA1 and MSWIBA2 (20 wt%) at 28 days are similar to those obtained in previous studies [17-19, 52]. The results suggest the utilization of bottom ashes as SCM up to 10 wt% replacement because the compressive strength of CEMI 42.5N with 5 wt% and 10 wt% is in accordance with the standards (>40 MPa at 28 days).