A Sustainable Strategy for Co-Melting of Electroplating Sludge and Coal Gasification Slag: Metals Recovery and Vitrified Product Valorization

Abstract

The high concentrations of heavy metals in electroplating sludge (ES) result in its dual properties as both hazardous waste and a potential secondary resource. Effective strategies are urgently needed for the simultaneous detoxification and utilization of ES. In this study, a sustainable strategy for co-melting of ES and coal gasification slag (CGS) was proposed. By optimizing the mass ratio of ES to CGS (m(ES)/m(CGS) = 1) and adding 7.5 wt% B₂O₃, a low-temperature vitrification system was established at 1250 °C, enabling the recovery of 97.31 ± 0.61% Cu, 99.17 ± 0.43% Ni, and 81.84 ± 0.33% Fe in the alloy phase within 90 min. Melt structure analysis indicated that CaO and [BO₃] promoted the depolymerization of the silicate network, facilitating the amorphous phase transition and enhancing fluidity. Meanwhile, residual carbon from CGS functioned as a reductant, reducing metal minerals in the mixture to form alloys that were simultaneously separated during co-melting. Compared with the raw sample, heavy metals in the vitrified product were effectively immobilized, exhibiting low risk of heavy metal leaching. Furthermore, high-value-added glass-ceramic materials were successfully prepared from the vitrified product. Therefore, the proposed strategy could serve as a sustainable solution for the treatment of ES and CGS.

1. Introduction

Electroplating serves as a critical supporting technology for a wide range of modern industries [1]. However, its extensive application inevitably leads to the generation of substantial electroplating sludge (ES) as a byproduct of wastewater treatment [2]. Due to its complex composition, high heavy metal content, and toxicity, ES is classified as hazardous waste worldwide [3]. The global production of ES exceeds 20 million tons annually, posing significant environmental risks [4]. Moreover, the grades of typical metals in ES are higher than those of primary metal ores, making it a promising secondary resource with considerable utilization potential [5,6]. Thus, effective strategies for resource recovery and treatment of ES are urgently needed to support the sustainable development of the electroplating field.

Current treatment methods for ES include solidification/stabilization (S/S) and resource utilization [7]. S/S is recognized as an effective method for immobilizing heavy metals, thereby reducing the risks of leaching and migration [8,9]. However, it results in the loss of metal resources and increased waste volume [10]. Resource utilization methods mainly include hydrometallurgical leaching, bioleaching, and material reuse. However, secondary waste generation, long processing times, and incomplete metal recovery limit their practical applicability [11,12]. Compared with the aforementioned methods, melting shows greater potential for the comprehensive utilization of ES. It could convert waste into stable and homogeneous glass, significantly reducing waste volume while immobilizing heavy metals [13]. Moreover, controlled transformation of the metal phase enables selective recovery of target metals [14,15], and the vitrified product could be further utilized [16].

The appropriate chemical composition plays a crucial role in melting. Previous studies showed that regulating composition within the CaO-Al₂O₃-SiO₂ ternary system could significantly reduce the melting temperature and promote vitrification of hazardous wastes [17,18]. Thermal treatment could also restructure heavy-metal-bearing compounds [19,20], and metal oxides tend to convert into chemically stable spinel phases during melting [21,22,23]. However, vitrified products might still pose a risk of heavy metal release. Metals could selectively transform into discrete metallic phases that are readily separable [24,25]. With the synergistic action of multiple chemical species in the melt, the complex silicate network could be depolymerized, reducing melt viscosity and promoting the separation of easily separable metallic phases from the melt [26,27]. Thus, achieving simultaneous metal recovery during melting requires effective regulation of the chemical composition in the melt.

Notably, the high calcium content in ES often leads to an imbalance in the CaO-SiO₂-Al₂O₃ ternary system, resulting in elevated melting temperatures [28]. Coal gasification slag (CGS), a byproduct of the coal chemical industry, is rich in SiO₂ and Al₂O₃ [29]. It is reported that more than 70 million tons of CGS are produced every year in China. However, the majority of CGS is simply dumped or arbitrarily landfilled, leading to significant waste of land resources, the release of dangerous elements, and numerous pollution problems [30]. Co-melting of ES and CGS could adjust the compositional ratio of the CaO–SiO₂–Al₂O₃ system, thereby potentially lowering the system’s melting point. In addition, CGS contains a certain amount of residual carbon, which has the potential to act as a reductant [31]. Therefore, due to the complementary characteristics of ES and CGS, vitrification and metal separation may be achieved simultaneously.

In this study, a novel strategy was proposed for simultaneously recovering metals and valorizing the vitrified product through the co-melting of ES and CGS. The effects of diverse factors on the melt phase transition and metal recovery were analyzed. Various analytical techniques, including X-ray photoelectron spectroscopy (XPS), Fourier-transform infrared spectroscopy (FTIR), scanning electron microscopy (SEM), and thermodynamic calculations, were employed to clarify the mechanisms of the complex silicate network transformation and synchronous metals recovery. Furthermore, the feasibility of using the harmless vitrified product for the preparation of glass-ceramics was explored.

2. Materials and Methods

2.1. Materials

The ES used in this study was obtained from an electroplating enterprise in Guangdong Province, China.

Table 1. Chemical composition of ES and CGS (wt%).

	Composition
	Ca	Si	Al	Cu	Ni	Fe	Cr	Zn	C	S	O	Others
ES	21.39	0.97	1.05	4.70	7.43	4.08	0.51	0.63	4.05	3.80	41.51	9.88
CGS	13.24	19.88	9.13	0.08	0.13	8.76	0.28	0.05	5.59	0.18	39.61	3.07

showed that the ES contained 4.70 wt% Cu, 7.43 wt% Ni, 4.08 wt% Fe, 21.39 wt% Ca, and 3.08 wt% S. Due to the low crystallinity of the metallic phase, only diffraction peaks of CaSO₄ and CaCO₃ were detected in Figure S1a. The CGS used in this study was obtained from a coal gasification enterprise in Ningxia Province, China. According to the results presented in Table 1 and Figure S1b, the CGS contained 8.76 wt% Fe, 13.24 wt% Ca, 19.88 wt% Si, 9.13 wt% Al, and 5.59 wt% C, with its phase composition comprising amorphous slag and residual carbon. Before the experiment, the samples were dried at 105 °C for 24 h and then ground to particles finer than 200 mesh.

Chemical reagents including boric anhydride (B₂O₃, AR), acetic acid (HAc, AR), hydrofluoric acid (HF, AR), hydroxylamine hydrochloride (NH₂OH·HCl, AR), and ammonium acetate (NH₄Ac, AR) were purchased from Aladdin Co., Ltd. (Shanghai, China). Sulfuric acid (H₂SO₄, AR), nitric acid (HNO₃, AR), perchloric acid (HClO₄, AR), sodium hydroxide (NaOH, AR), and hydrogen peroxide (H₂O₂, AR) were purchased from Guangzhou Chemical Reagent Co., Ltd. (Guangdong, China).

2.2. Co-Melting Experiment

Co-melting experiments were conducted in a laboratory-scale tube furnace (GSL-1700X-S60, Hefei Kejing Material Technology Co., Ltd., Hefei, China). The pretreated ES and CGS were mixed with B₂O₃ in a certain proportion and put into corundum crucibles. The crucibles were placed into a tube furnace, and nitrogen was introduced at a flow rate of 100 mL/min. Subsequently, the samples were heated to a specified temperature at a heating rate of 5 °C/min and maintained for the corresponding time. Finally, the samples were taken out after cooling naturally to room temperature in the furnace. All co-melting experiments were performed in triplicate under each set of conditions. The schematic diagram of the experimental setup for the co-melting treatment of ES and CGS is shown in Figure 1.

The alloys in the melting product were separated from the slag by crushing. The slag was crushed, ground, and stored in a dry environment. The residual heavy metal content was determined through digestion of slag samples, and the digestion method was shown in Text S1. The Cu, Ni, and Fe recovery efficiencies were calculated using Equation (1).

$$\mathsf{\eta }=\left({\displaystyle \frac{{\mathrm{m}}_0\times {\mathrm{c}}_0-{\mathrm{m}}_1\times {\mathrm{c}}_1}{{\mathrm{m}}_0\times {\mathrm{c}}_0}}\right)\times 100\%$$

(1)

where η is the recovery efficiency of metals (Cu, Ni, and Fe), %; m₀ is the weight of raw sample, g; c₀ is the element weight fraction in the raw sample, mg/kg; m₁ is the weight of the slag after melting, g; c₁ is the element weight fraction in the slag after melting, mg/kg.

2.3. Analysis and Characterization

The chemical compositions of the raw materials and slag were detected by X-ray fluorescence spectrometry (XRF, ARL Advant’X, Thermo Fisher Scientific, Waltham, MA, USA). The mass fractions of carbon and sulfur in the ES and CGS were determined using a carbon-sulfur analyzer (HIR944, Huake Instrument, Shanghai, China). The heavy metal content in the digested solutions of the samples was determined by inductively coupled plasma optical emission spectrometry (ICP-OES, Avio 560 Max, PerkinElmer, Waltham, MA, USA). The crystalline phases of the samples were characterized by X-ray diffraction (XRD, D8 Advance, Bruker, Karlsruhe, ND, USA) using Cu Kα radiation at 40 kV and 30 mA, and the percentage of amorphous phase of the slag samples was calculated following the Chinese standard GB/T 18046-2017 [32]. The chemical states of elements in the samples were analyzed by X-ray photoelectron spectroscopy (XPS, K-Alpha+, Thermo Fisher Scientific, Waltham, MA, USA), with spectral calibration referenced to the C 1s peak at 284.8 eV. Spectral fitting was carried out using the CasaXPS 2.3.26 software. Fourier-transform infrared spectra were measured by FT-IR spectrometer (FT-IR, Nicolet iS20, Thermo Fisher Scientific, Waltham, MA, USA) in the wavenumber range of 400 to 4000 cm⁻¹. The elemental distributions of the alloy and vitrified product were analyzed by field emission scanning electron microscopy (FE-SEM, Ultra 55, Zeiss, Oberkochen, Germany) combined with energy dispersive spectroscopy (EDS, TM3030, Hitachi, Tokyo, Japan). The crystallization characteristics of the vitrified product were analyzed using thermogravimetry differential scanning calorimetry (TG-DSC, STA 449 F3, Netzsch, Selb, Germany). The Vickers hardness was determined using a Vickers indentation (Vickers hardness tester, MH-6, Zwick/Roell, Ulm, Germany) with a load of 1.96 N and a dwell time of 10 s. FactSage 7.1 (Thermfact/CRCT, Montreal, Canada; and GTT-Technologies, Aachen, Germany) was employed for thermodynamic calculations and phase diagram analysis.

2.4. Vitrified Product Heavy Metals Analysis

2.4.1. Leaching Toxicity Test

The leaching toxicity test on the vitrified product was conducted according to the Chinese standard HJ/T 299-2007 [33]. Concentrated sulfuric acid and concentrated nitric acid were mixed at a mass ratio of 2:1 and diluted to pH of 3.20 ± 0.05 to prepare the leaching solution. The ground vitrified product was added to the leaching solution at a liquid-to-solid ratio of 10:1 mL/g. Subsequently, the sample was placed on a shaker at a speed of 30 rpm for 18 h. Finally, the leachate was collected, centrifuged, and filtered, and its heavy metal content was determined.

2.4.2. Environmental Risk Assessment

In this work, the modified BCR sequential extraction procedure [34] was employed to analyze heavy metals in vitrified product (Text S2). The risk assessment code (RAC) and the potential ecological risk index (RI) were utilized to evaluate the individual ecological risks of heavy metals in the vitrified product and the overall potential risk of heavy metals, respectively [35]. The RAC values of heavy metals in the vitrified product were calculated using Equation (2).

$$\mathrm{RAC}={\displaystyle \frac{\mathrm{F}1}{{\mathrm{C}}_{\mathrm{Total}}}}\times 100\%$$

(2)

where C_Total is the total content of the heavy metals in the vitrified product.

RI indicates the comprehensive potential ecological risk of heavy metals in the vitrified product. It was calculated using Equation (3).

$$\mathrm{RI}=\sum _{\mathrm{i}=1}^{\mathrm{n}}{\displaystyle \frac{{\mathrm{T}}_{\mathrm{i}}\times {\mathrm{A}}_{\mathrm{i}}}{{\mathrm{D}}_{\mathrm{i}}}}$$

(3)

where A_i is the proportion of active fractions (F1 + F2 + F3) of the heavy metals in the vitrified product, D_i is the proportion of residual fraction (F4) of the heavy metals in the vitrified product, and T_i is the toxicity response factors for the corresponding heavy metals, with assigned values for Cu, Ni, Cr, and Zn being 5, 5, 2, and 1, respectively [36]. The risk categories of RAC and RI are presented in Table S1.

3. Results and Discussion

3.1. Phase Transformation and Metal Recovery

3.1.1. Effect of CGS to ES Mass Ratio

The effects of different m(CGS)/m(ES) ratios on metal recovery efficiency and slag phase transformation were investigated at 1400 °C. Figure 2a showed the phase composition of slags produced from different mixtures. When the m(CGS)/m(ES) ratio was 0.5, multiple calcium-containing crystalline phases including Ca₂Fe₂O₅ (PDF#38-0408), CaSO₃ (PDF#36-0530), CaS (PDF#08-0464), and CaFeO₂ (PDF#21-0917) were observed in the slag. This was attributed to the high Ca content in the mixture that induced the formation of calcium-containing phases with high melting points. As the ratio increased to 0.75, 1, or 1.25, the slag was transformed into an amorphous glassy state. However, further increases in the CGS proportion led to the reappearance of various crystalline phases, including SiO₂ (PDF#16-0380) and CaAl₂Si₂O₈ (PDF#05-0528), as well as several metal sulfides and oxides. The optimized mass ratio of CGS to ES enhanced the melt fluidity and effectively promoted the aggregation and sedimentation of the metallic alloys. As shown in Figure 2b, when the m(CGS)/m(ES) ratio was 0.75,1 or 1.25, the metal recovery efficiency remained at a high level. Among these ratios, when the m(CGS)/m(ES) ratio was 1, the recovery efficiencies of Cu, Ni, and Fe reached maximum values of 96.33 ± 0.22%, 99.56 ± 0.13%, and 79.93 ± 1.02%, respectively. Therefore, the optimal mass ratio of CGS to ES was determined to be 1.

3.1.2. Effect of Smelting Temperature

The effects of melting temperature on the formation of vitrified product and metal recovery were investigated at a m(CGS)/m(ES) ratio of 1 and a holding time of 90 min. As shown in Figure 2c, the slag formed at 1250 °C exhibited a high degree of crystallinity, with complex aluminosilicates and various heavy metal phases observed. As the treatment temperature increased, the intensity of the diffraction peaks corresponding to crystalline phases gradually diminished and completely disappeared at 1350 °C, indicating a transition to an amorphous structure. However, at 1450 °C, a distinct crystallization peak of CaS (PDF#08-0404) reappeared in the XRD pattern. This phenomenon may be attributed to the enhanced mobility and reduced size of polymeric silicate structures at elevated temperatures, which facilitated the crystallization of CaS [37]. When the smelting temperature was below 1350 °C, the metal recovery efficiency increased significantly but slowed down at temperatures above 1350 °C (Figure 2d). Specifically, as the temperature rose from 1250 °C to 1350 °C, the recovery efficiencies of Cu, Ni, and Fe increased from 79.21 ± 2.17%, 89.01 ± 0.87%, and 58.01 ± 3.48% to 96.70 ± 0.47%, 98.98 ± 0.44%, and 77.02 ± 1.28%. The recovery efficiencies of Cu, Ni, and Fe at 1400 °C were 96.33 ± 0.22%, 99.59 ± 0.03%, and 79.96 ± 0.97%, respectively. The increased smelting temperature promoted the depolymerization of silicate networks in the melt, thereby reducing the melt viscosity (Figure S2) and contributing to improved recovery of metal alloys.

3.1.3. Effect of B₂O₃ Addition

Low-temperature melting is an effective strategy for reducing the energy consumption of the melting process. Accordingly, a small quantity of B₂O₃ was added to the co-melting system as fluxing agent. The incorporation of B₂O₃ into the CaO-SiO₂-Al₂O₃ system could facilitate the transformation of high-melting-point phases into low-melting-point phases, reduce the flow temperature and viscosity, and promote the formation of a glassy phase [38]. Based on the characteristics of the typical smelting process for copper-nickel metallurgy and the viscosity characteristics of the co-melting system (Figure S3), 1250 °C was selected as the target temperature for melting treatment. The phase transition of slag samples with different B₂O₃ additions is shown in Figure 2e. The slag exhibited a high degree of crystallization with no B₂O₃ addition. When the B₂O₃ content reached 2.5 wt%, an amorphous phase began to form, and at contents exceeding 5 wt%, the slag was completely transformed into an amorphous glassy state. The enhanced melt fluidity simultaneously promoted the recovery of metals (Figure 2f). As the B₂O₃ addition increased from 0 wt% to 7.5 wt%, the recovery rates of Cu, Ni, and Fe increased significantly from 79.21 ± 2.17%, 89.01 ± 0.87%, and 58.01 ± 3.48% to 97.31 ± 0.62%, 99.17 ± 0.43%, and 81.84 ± 0.34%, respectively. However, when the B₂O₃ content was further increased to 10 wt%, the recovery efficiencies of Cu, Ni, and Fe slightly decreased to 94.97 ± 0.36%, 97.72 ± 0.72%, and 78.80 ± 0.51%, respectively. The decrease in the recovery of Cu, Ni, and Fe with increasing B₂O₃ addition could be attributed to the increased chemical dissolution of these metals into the slag [39].

3.1.4. Effect of Holding Time

When the mass ratio of m(CGS)/m(ES) was 1, the B₂O₃ content was 7.5 wt%, and the melting temperature was 1250 °C, the slag samples subjected to melting treatment for 30–150 min were completely transformed into a glassy state (Figure 2g). As shown in Figure 2h, the recovery efficiencies of Cu, Ni, and Fe initially increased with time and then stabilized. After 30 min of melting, the recovery efficiencies of Cu, Ni, and Fe were 86.47 ± 0.44%, 85.74 ± 0.83%, and 69.20 ± 1.10%, respectively. When the holding time was extended to 90 min, the recovery efficiencies increased to 97.31 ± 0.62%, 99.17 ± 0.43%, and 81.84 ± 0.34%, respectively, and then remained stable. Extending the holding time favored the forward progression of metal reduction and sulfidation reactions. Additionally, it enhanced atomic diffusion and facilitated the formation of a stable alloy structure through the continuous migration, aggregation, and growth of alloy particles, ultimately leading to their separation from the slag.

In summary, when ES was mixed with CGS at a mass ratio of 1:1 with 7.5 wt% B₂O₃ during co-melting at 1250 °C for 90 min, the metal recovery efficiencies of Cu, Ni, and Fe were as high as 97.31 ± 0.62%, 99.17 ± 0.43%, and 81.84 ± 0.34%, respectively. This demonstrated the high feasibility for metals recovery through the co-melting of ES and CGS. However, it is noteworthy that although 5 wt% B₂O₃ addition resulted in complete vitrification and high metal recovery, the dosage of 7.5 wt% was selected as optimal for maximizing metal recovery (Cu, Ni, Fe) and minimizing their residual content in the slag.

3.2. Synergistic Mechanism of Co-Melting

3.2.1. Activation of Potential Components in Silicate Network for Favorable Melt

The non-metallic phase formed during the co-melting of ES and CGS was predominantly composed of CaO, SiO₂, and Al₂O₃ (Table S1). According to the CaO-SiO₂-Al₂O₃ ternary phase diagram generated by thermodynamic calculations (Figure 3a), the distinct melting behaviors of ES and CGS could be explained by the potential components. Specifically, the theoretical melting point of ES was close to that of CaO, whereas the melting point of CGS exceeded 1450 °C due to its high SiO₂ and Al₂O₃ contents. When ES and CGS were blended at designed ratios (CGS/ES = 0.5–2), the chemical compositions of the mixtures fell within the low-temperature eutectic zone. It indicated that the combination of CaO-rich ES and SiO₂-Al₂O₃-rich CGS effectively reduced the melting temperature, enhancing the feasibility of vitrification.

Potential components activated within the silicate network enhanced melt fluidity. In the CaO-SiO₂-Al₂O₃ ternary system, SiO₂ primarily acted as a network former, CaO functioned as a network modifier, and Al₂O₃ exhibited dual roles, serving both as a network former and a modifier [40]. Specifically, [AlO₄] tetrahedra contributed to network formation, while [AlO₆] octahedra acted as network modifiers. To clarify the specific function, XPS analyses were conducted (Figure 3b–f). As shown in Figure 3b, the O 1s binding energy decreased from 531.8 eV when m(CGS)/m(ES) was 2 to 531.1 eV when the ratio was 1, and subsequently increased to 531.3 eV when the ratio was 0.5. Figure 3c showed the XPS fitting spectra of O 1s. Two peaks were divided in O 1s high-resolution spectra of different samples; the peak centered on 532.00 eV was assigned to the bridging oxygen (BO), while the peak at 531.00 eV represented non-bridging oxygen (NBO) [41]. When m(CGS)/m(ES) was 2, the NBO fraction was 39.02%, reaching a maximum of 62.42% when the ratio was 1 (Figure 3d). The increase in NBO was attributed to the release of free oxygen ions (O²⁻) from CaO, which promoted the conversion of BO to NBO, thereby disrupting the aluminosilicate network [42]. Figure 3e showed the XPS spectra of Al 2p. The peaks in 73.4–74.5 eV were assigned to [AlO₄] units with network-forming functions, while those in 74.1–75.1 eV corresponded to [AlO₆] units with network-modifying functions [43]. As shown in Figure 3f, the proportion of [AlO₄] increased from 48.72% to 58.98% as the mass ratio of CGS to ES decreased from 1 to 0.5. The [AlO₄] tetrahedral units were structurally similar to [SiO₄] units and could integrate to form complex Al-O-Si anionic groups, thereby increasing the complexity of the silicate network in the melt [44]. Moreover, the amorphous phase content in the slag followed the same trend as NBO content, reaching a maximum when m(CGS)/m(ES) was 1 (Figure S4). The synergistic activation of CaO and Al₂O₃ in the silicate network promoted the conversion of BO to NBO, leading to the formation of the favorable melt.

3.2.2. Positive Effects of B₂O₃ on Vitrification and Silicate Network Depolymerization

Based on the optimized CaO-SiO₂-Al₂O₃ system, the effects of B₂O₃ on the complex silicate network and melting properties were studied. The FTIR spectra of the melting products with different B₂O₃ contents produced at 1250 °C were shown in Figure 4a. The band at 600–800 cm⁻¹ was attributed to the bending vibration of [AlO₄] or B-O-B, while those at 800–1200 cm⁻¹ were assigned to the stretching vibrations of [SiO₄]; additionally, the bands at 1210 cm⁻¹ and 1350 cm⁻¹ corresponded to the stretching vibration of [BO₄] and the asymmetric stretching vibration of [BO₃], respectively [45,46]. The FTIR spectrum of the product with 0% B₂O₃ exhibited the distinct bending vibration band of [AlO₄] and the stretching vibration band of [SiO₄], indicating that the slag primarily consisted of a Si-O polymer network. In addition, the presence of crystalline phases in the product caused partial ordering, which resulted in irregular absorption peaks. Notably, after the addition of B₂O₃, the FTIR spectrum became smoother, and the vibration bands associated with the B-O bond appeared in the range of 1200–1600 cm⁻¹. With the increase of B₂O₃ content, the vibrations corresponding to the [BO₃] units gradually intensified, while those associated with the tetrahedral [BO₄] units remained weak. This suggested that [BO₃] was the main structural unit of boron in the melt. The unique planar triangular structure of [BO₃] disrupted the continuous three-dimensional network when it was incorporated into the Si-O network, thereby promoting the depolymerization of the complex silicate network [47].

Furthermore, the FTIR bands in the range of 800–1200 cm⁻¹ were analyzed by Gaussian function to obtain the content of Qⁿ species, thereby characterizing changes in the aggregation degree of the slag network. A smaller value of n referred a lower degree of polymerization in the network structure [48]. Among them, the characteristic frequencies of Q⁰ (NBO/Si = 4), Q¹ (NBO/Si = 3), Q² (NBO/Si = 2), Q³ (NBO/Si = 1), and Q⁴ (NBO/Si = 0) species are in the range of 840–890 cm⁻¹, 900–950 cm⁻¹, 960–1030 cm⁻¹, 1050–1100 cm⁻¹, and 1190 cm⁻¹, respectively [49]. The relative percentages of Q⁰, Q¹, Q², Q³, and Q⁴ are shown in Figure 4b–g. The percentage of low-polymerization Si-O tetrahedral structures (Q⁰ and Q¹) in the slag increased with higher B₂O₃ content. Specifically, without B₂O₃, the combined percentage of Q⁰ and Q¹ was only 27.83%, whereas it significantly increased to 51.36% when the B₂O₃ content was 10 wt%. Correspondingly, the Q⁴ unit was present only in the slag without B₂O₃, and the percentage of Q³ units decreased with increasing B₂O₃ content. Without B₂O₃, the percentage of Q³ was as high as 43.27%, but it decreased to 24.30% when the B₂O₃ content was 10 wt%. In addition, the viscosity of the melt with 10 wt% B₂O₃ decreased from 20.77 Pa·s to 3.79 Pa·s compared with the slag without B₂O₃ at 1250 °C (Figure 4h). The depolymerization of the melt reduced the degree of polymerization and the thermodynamic stability of the system, thereby promoting the vitrification of the product, which was consistent with the enhanced characteristic peaks of the amorphous phase observed in the XRD pattern (Figure 4i).

3.3. Mechanism of Metals Recovery as Alloy During Co-Melting

Gibbs free energy calculations were conducted to evaluate the thermodynamically feasible reactions during the co-melting process. The CGS contained amorphous carbon (Figure 5a), whereas ES analysis confirmed the presence of CaSO₄ (Figure S1a). Accordingly, three categories of thermodynamically favorable reactions were identified: (i) reduction of Cu, Ni, and Fe oxides, (ii) reduction of sulfur species, and (iii) sulfidation of transition metals. As shown in Figure 5b, the reduction reactions of CuO, NiO, and FeO were all thermodynamically feasible at 1250 °C, while the reductions of CuO and NiO occurred more readily than that of FeO, consistent with their lower Gibbs free energy values under identical conditions. In addition, CaSO₄ underwent carbothermal reduction to CaS at relatively lower temperatures (Figure 5c), suggesting that CaS formation was the predominant reduction pathway in the co-melting system. As shown in Figure 5d, CaS further reacted with Cu, Ni, and Fe oxides, while Cu₂O and NiO were also spontaneously sulfidized by FeS. This preferential sulfidation could be attributed to the more negative Gibbs free energies of formation of Cu₂S and NiS compared with FeS, in agreement with the relative electronegativity differences among the cations [50]. Alloy composition and chemical state analyses revealed the coexistence of metallic Cu, Ni, and Fe together with their corresponding sulfides, in agreement with the thermodynamic predictions (Figure S5). Moreover, SEM analysis confirmed that metallic copper, nickel, iron, and matte phases were stably distributed within the alloy (Figure S6). The observed enrichment of Cu, Ni, and Fe in the alloy phase was consistent with the Hume-Rothery rules [51], which predict the favored formation of Cu-Ni-Fe solid solutions with high solubility. The co-melting of CGS and ES thus thermodynamically promotes the synchronous migration of metals into stable alloy.

Due to differences in chemical bonding and surface tension, the metal and slag phases were immiscible [52]. During the melting process, alloy particles settled and aggregated under the influence of gravity and melt fluidity. As shown in Figure 5e, a well-defined alloy phase and a dense glass phase formed in the melt system, showing a clear separation interface between the alloy and glass phase. The distributions of Cu and Ni were highly consistent with that of the alloy phase. Fe was mostly concentrated in the alloy phase, with a small amount present in the glass phase composed of Ca, Si, and Al. Thus, the co-melting of ES and CGS simultaneously achieved vitrification and the recovery of metals as alloy.

In summary, adjusting the content of active species promoted the depolymerization of the complex silicate network in the co-melting system of ES and CGS. The favorable melt exhibited excellent fluidity, which facilitated the separation of the alloy phases. The comprehensive mechanism is shown in Figure 6.

3.4. Harmless Vitrified Product Valorization

The co-melting process facilitated the transfer of metals from ES and CGS into the alloy phase, leaving only minor residues in the vitrified product. The stabilization of these residual metals was further assessed using the modified BCR sequential extraction procedure. As shown in Figure 7, the labile (F1 + F2) fractions of Cu, Ni, and Zn in the raw sample were 87.6%, 50.7%, and 82.6%, respectively, but were markedly reduced after vitrification. Correspondingly, the stable (F3 + F4) fractions of Cu, Ni, Fe, Cr, and Zn increased to 98.0%, 90.8%, 94.7%, 93.7%, and 99.3%, respectively, indicating that nearly all heavy metals were immobilized within the glassy matrix. In addition, both the RAC and RI indicated low risk levels (

Table 2. RAC and RI of the raw sample and vitrified product.

Index	Sample	Heavy Metal
		Cu	Ni	Fe	Cr	Zn
RAC	Raw sample (%)	67.84	35.75	0.68	10.86	63.98
	Vitrified product (%)	ND	7.52	2.87	5.46	ND
RI	Raw sample	168.184
	Vitrified product	19.10

), while leaching toxicity tests confirmed that heavy metals in the vitrified product were below detection limits (Table S4).

The preparation of glass-ceramics is an effective method for utilizing harmless vitrified products by converting them into high-value products [53]. Figure 8a showed the DSC analysis of the vitrified product. The endothermic peak at 734 °C corresponded to the glass transition temperature (T_g), while the exothermic peak at 850 °C corresponded to the crystallization temperature (T_c). The XRD patterns of the vitrified product and the resulting glass-ceramics were shown in Figure 8b. After heat treatment at 750 °C for 2 h, crystalline phases began to precipitate. When the temperature was increased to 850 °C, both the number and intensity of crystalline phase diffraction peaks increased significantly. The crystallinity of the sample treated at 850 °C for 2 h, calculated using Jade 6.0, reached 69.92%, with the main crystalline phases identified as SiO₂, CaAl₂Si₂O₈, CaSiO₃, and Ca₂SiO₄. Meanwhile, the microstructure of the crystallized product is clearly visible in the SEM image (Figure 8c). The performance of the glass-ceramic product was also evaluated. Its Vickers hardness, water absorption rate, acid corrosion resistance, and alkali corrosion resistance were 6.91 GPa, 0.04%, 99.61%, and 99.62%, respectively, demonstrating excellent overall performance. In summary, the glass-ceramic was successfully fabricated via one-step heat treatment, providing a sustainable pathway for the valorization of the harmless vitrified product.

3.5. Implications

Current co-melting studies primarily focus on immobilizing heavy metals within a glass matrix or producing spinel phases with enhanced stability, while the recovery of metals is generally limited [54,55,56]. In contrast, the co-treatment strategy developed in this study takes full advantage of the inherent CaO-SiO₂-Al₂O₃ complementarity of ES and CGS as well as the residual carbon in CGS, enabling efficient utilization of ES and CGS. The produced alloy ingot contained 29.73 wt% Cu, 37.88 wt% Ni, 22.63 wt% Fe, 0.52 wt% Cr, and 0.16 wt% Zn, along with sulfur and other minor elements. Meanwhile, the glass-ceramic prepared by one-step process exhibited a Vickers hardness of 6.91 GPa, along with excellent acid and alkali resistance, achieving 99.61% and 99.62%, respectively.

The technical synergy of stabilizing waste, recovering metals, and valorizing product through the co-melting of ES and CGS, as demonstrated in this study, lays a solid foundation for a “treat-waste-with-waste” circular economy. In addition, the vitrification temperature was reduced to 1250 °C, which is 150–250 °C lower than typical melting temperatures (1400–1500 °C) reported for ES vitrification [10]. The corresponding economic and environmental performance, including profitability and carbon emissions, are summarized in

Table 3. Assessment of profit and carbon emissions.

Category	Detail	Amount (kg)	Cost (Dollars/USD)	Profit (USD)	Carbon Emission Factor (kg CO2/Ton)	Carbon Emission (kg CO2)
Resource	ES	500	200	100	0	0
	CGS	500	10	5	0	0
	B2O3	75	−300	−22.5	0	0
	Electricity	400 kW h	−0.1 USD/kW h	−40	0.58 kg CO2/kW h	232
Labor cost	Laborer	2	−5 USD	−10	2 kg CO2/human	4
Product	Glass ceramic	710	50	35.5	−7.76	−5.51
	Alloy	113	5000	565	−1990	−224.87
			Profit	633	Carbon emission	5.62

. These evaluations were conducted based on standard price benchmarks in China and the China Products Carbon Footprint Factors Database, respectively [57,58].

4. Conclusions

In this work, a sustainable strategy for the co-melting of ES and CGS was proposed. By fully utilizing the compositional characteristics of these two waste materials, efficient metal recovery and valorization of the vitrified product were achieved. The conclusions were summarized as follows:

(1)

When ES and CGS were co-melted at a mass ratio of 1:1 with 7.5 wt% B₂O₃ at 1250 °C for 90 min, vitrification was achieved, and 97.3% Cu, 99.2% Ni, and 81.8% Fe were simultaneously recovered.

(2)

The synergistic mechanism of CaO, Al₂O₃, and B₂O₃ in the silicate network was investigated through characterization. Active species CaO, [AlO₆], and [BO₃] promoted the depolymerization of the complex Si-O network structure, thereby facilitating crystal phase melting and enhancing melt fluidity.

(3)

The co-melting of ES and CGS simultaneously reduced Cu, Ni, and Fe oxides to metals and sulfurized them into matte phases, both of which were subsequently incorporated into the alloy phase as a solid solution. Moreover, the favorable melting environment promoted efficient metal recovery.

(4)

Residual heavy metals such as Cu, Ni, Cr, and Zn in the vitrified product were effectively immobilized, and a high-performance glass-ceramic was prepared by heat-treating the vitrified product at 850 °C for 120 min.