Characterization and Property Evaluation of Glasses Made from Mine Tailings, Glass Waste, and Fluxes

Abstract

The present investigation introduces a novel approach, using As-Zn-Fe mining tailings (MT) and recycled bottle glass (cullet) to enable the manufacturing of a new glass for ornamental articles, with characteristics similar to those of soda–lime–silicate glass (SLS), and at the same time, immobilizing potentially toxic elements (PTEs) from mining tailings, which cause environmental pollution with severe risks to human health. The glass used was obtained from transparent glass bottles collected from urban waste, which were later washed to remove impurities and then crushed until they reached No. 70 mesh (212 μm) level; in the case of mining tailings, the sample used comes from the ore benefit process, with 96.8% of particles below the No. 50 mesh level (300 μm). Six mixtures were made by varying the composition of the mining tailings and glass, K₂CO₃ and H₃BO₃ as fluxes were also used in constant proportion. The mixtures were melted at 1370 °C, and later, the glass samples were cast on a steel plate at room temperature. The characteristics of the glasses were studied using thermal analysis (TA), X-ray diffraction (XRD), Fourier transform infrared (FTIR), and scanning electron microscopy (SEM). Likewise, their chemical resistance in acid and basic media and density were evaluated. The results unequivocally demonstrate the feasibility of manufacturing glasses with a light green color, the increase in the content of mining tailings increased the apparent T_g from 625 to 831 °C. Glasses with 17 and 21.3% MT presented lower density values due to a better-polymerized glass structure, attributed to the increase in SiO₂ and Al₂O₃ and the decrease in alkaline oxides, which allowed for the retention of PTEs in their structure.

1. Introduction

Mexico, a country rich in mineral resources, is a significant global player, ranking among the top ten producers of 16 different metals and the first silver producer worldwide [1]. However, the environmental impact of mining activity is a pressing concern. The large amount of waste generated by mining operations, such as tailings and slags, is a significant source of environmental pollution. The deposition of mine tailings, in particular, leads to the release of potentially toxic elements (PTEs) such as Pb, As, Zn, Cu, Cr, Ni, Co, Fe, and Cd, causing high ecological pollution [2,3,4]. This urgent environmental issue necessitates innovative solutions, such as the one presented in this research.

The historical deposition of large quantities of mining waste containing iron (Fe), arsenic (As), and lead (Pb) in semi-arid soils in Durango has led to a severe problem. The Comarca Lagunera, an economically important region between the states of Coahuila and Durango in northern Mexico, is now one of the areas with the highest accumulation of arsenic in the world. With its severe human health implications, this alarming situation underscores the urgent need to prioritize research efforts and find sustainable solutions [5,6].

On a more positive note, there has been a significant shift in the approach towards environmental protection. This shift has increased attention to the treatment, recycling, and safe use of mine tailings [7]. As a result, various innovative applications for mine tailings have been explored, such as their use in unpaved road bases, environmentally friendly bricks, partial replacement of cement, paste backfill, and ceramic articles [8,9,10,11,12]. These developments offer potential solutions to the environmental challenges posed by mining and inspire hope for a more sustainable future.

In this sense, vitrification is a valuable method for treating hazardous mining tailings because it provides a stable way to solidify industrial waste [2,9,13,14,15,16]. During vitrification, waste is transformed into glassy material to retain its toxic components, captured in the glassy matrix [9].

According to Terry Lay et al. [14], the development of formulations for vitrifying waste represents a challenge, as aspects such as (a) acceptability of the waste, (b) melt processability, (c) durability of the vitreous product, and (d) economics must be considered.

The main challenge in converting As-Zn-Fe mining tailings from Durango State into glassy materials is ensuring that the resulting product has good chemical stability thus immobilizing the PTEs, and can be processed at reasonably low temperatures (1350–1450 °C). Soda–lime–silicate glass (SLS) is a stable glass type commonly used in various applications. It can be recycled as waste glass to lower glass melting temperature, reducing energy consumption by up to 25% [17]. According to Baccarin and Bragança [18], the advantage of using waste glass resides in its chemical compatibility with the raw materials used in the ceramics industry, which has been an incentive to partially or fully replace the flux with this alternative source of alkalis. SLS has a strong glass-forming ability and can be used to vitrify tailings. A second challenging aim of this study was to recycle clear SLS bottles.

Likewise, the reason for choosing H₃BO₃ as flux results from the need to reduce the melting point of silica (usually 1730 °C) and as a component to minimize the leaching of PTEs. B₂O₃ is one of the best glass formers and has wide flexibility to accommodate various modifier oxides and for obtaining acceptable quality glasses for potential applications, such as borosilicate glasses for the immobilization of radioactive waste, shields for use in industrial and medical radiation facilities and laboratory glass [14,19,20]. On the other hand, the use of K₂O from K₂CO₃ could ensure better chemical resistance behavior and would give a larger moldability range, due to the greater ionic volume of potassium [21].

Therefore, this research explores the feasibility of recycling As-Zn-Fe mining tailings and clear SLS to produce new light green-colored glass with good chemical resistance that allows PTE retention using the vitrification process. These results could provide a beneficial solution to the health and environment of Durango society.

2. Materials and Methods

2.1. Materials

Clear bottles of SLS urban waste glass were collected and washed with hot water. Then, the glass was crushed using a ceramic mill with porcelain balls as a grinding media and sieved to a particle size of 212 μm mesh.

Fifty samples of approximately 2 kg each of mining tailings from the silver ore concentration process were gathered from various locations in the tailings dam, mixed to ensure consistency, and then 5 kg as the representative sample was selected. The mine is located northeast of the state of Durango; however, due to company policies, its name cannot be revealed.

2.2. Glass Synthesis

Due to the high content of silica in raw materials (values of main oxides are presented in

Table 1. Chemical compositions of raw materials: mining tailings (MT) and reference glass (GR).

Raw Materials	Oxide Content (Weight %)
	SiO2	Al2O3	CaO	MgO	Na2O	K2O	Fe2O3	TiO2	MnO	P2O5	ZnO
MT	90.13	4.06	0.075	0.07	0.176	2.23	1.64	0.08	0.012	0.052	0.021
GR	71.65	1.21	11.99	0.94	12.99	0.94	0.23	-	-	-	-

), typical fluxes such as alkaline earth oxides and B₂O₃ were found to be necessary to lower the melting temperature of the silicate glasses [14]. For that reason, reagents K₂CO₃ and H₃BO₃ (Fermont brand, purity > 99%) were used in the bath compositions to produce K₂O and B₂O₃, respectively. The conventional melting–casting technique was employed to synthesize the glasses under study. The proportion of mining tailings was increased in the compositions while the waste glass decreased. The fluxes were consistently used in the same proportion for each composition, as shown in

Table 2. Investigated composition to produce glasses.

Sample	Composition (Weight %)
	Mining Tailings	Glass Waste	K2O	B2O3
MTG1	4.2	81.0	7.0	7.8
MTG2	8.5	76.7	7.0	7.8
MTG3	12.8	72.4	7.0	7.8
MTG4	17.0	68.2	7.0	7.8
MTG5	21.3	63.9	7.0	7.8
MTG6	25.6	59.7	7.0	7.8
GR	0.0	100.0	0.0	0.0

. The SLS glass was used as the reference glass (GR) for comparison. Therefore, the GR glass served as the same starting raw material.

The raw materials and fluxes were mixed and melted in a Pt crucible using an electrical furnace (Lindberg/Blue BF51433). First, the samples were heated up to 400 °C to decompose boric acid and prevent losses due to volatilization [22]. They were maintained at this temperature for 1 h, then heated to 1350 °C and held at that temperature for an additional hour. Finally, the glasses were cast onto a steel plate at room temperature.

2.3. Analytical Methods

Before use, the particle size range of the raw materials (MT and SLS glass) was determined using sieve analysis, and the results are shown in Figure 1 as can be seen, at least 90% of the particles have a size less than 180 μm for both materials.

Raw materials and samples underwent a comprehensive characterization using X-ray diffraction (XRD). Powder samples were placed in the sample holder of a Philips X’Pert-MPD, with Cu-Kα radiation of 1.5405 Å, from 10 to 90, 2θ degrees and scanning rate 2° per min. Results are shown in Figure 2. Likewise, MT were also analyzed for potentially toxic elements using an Agilent 4100 MP-AES Spectrometer, the results are presented in

Table 3. Concentration of potentially toxic elements in mining tailings.

Element	Concentration (mg/L)
Iron	1631.00
Arsenic	4.14
Zinc	0.29
Lead	0.24
Cadmium	0.19
Manganese	0.15

.

On the other hand, to understand the structural and microstructural characteristics, the glasses by infrared spectroscopy (FTIR), X-ray diffraction (XRD), and scanning electron microscopy (SEM) were analyzed. These analyses provided significant insights into the composition and properties of the glasses. For FTIR analysis, a Perkin Elmer Frontier equipped with an ATR (attenuated total reflection) device covering the range from 2800 to 400 wavenumber (cm⁻¹) was used to analyze a powdered glass sample (grounded in an agate mortar). For XRD analyses, glass powder sieved at a particle size less than 212 μm was placed in the sample holder of a Bruker, D8 Advance equipment, with copper anode and CuK_α radiation of 1.5405 Å, from 5 to 90, 2θ degrees and scanning rate 2° per min. SEM examined glass microstructure (Jeol microscope JSM-6610 equipped with an EDS detector, JEOL Ltd., Tokyo, Japan). The sample, a piece of each glass, was mounted into quick setting resin, then sandpapered of different grains (80, 120, 240, 320, 500, 800, 1200, and 2400), subsequently were diamond-paste polished, and lastly, were Au-coated to make them conductive. Likewise, to know the thermal characteristics of each glass, differential thermal analysis (DTA, Netzsch STA 449 F1 Jupiter) was carried out from room temperature up to 1000 °C using an N₂ atmosphere and a heating rate of 10 °C/min. Inflection point method was used for determination of glass transition temperature (T_g), and the crystallization temperature (T_c) was taken as the higher point of each peak.

Furthermore, the chemical resistance of each glass and reference glass (GR) was evaluated in acid and basic media, in strict accordance with Russian standard GOST 10134-82 [23], using HCl 1N and NaOH 1N solutions, respectively. First, a known quantity of grounded glasses was put into PE bottles with a corresponding solution in a 1:10 (powder: solution) ratio. The PE containers were introduced into a stove at 96 °C for three hours. Finally, the solution was filtered, and the retained powders were dried and weighed. Weight losses (%) were estimated with the Equation (1). Additionally, As, Pb, and Zn in the solution were analyzed by ICP-OES spectrophotometer.

$$\% losses={\displaystyle \frac{initial weight-final weight}{initial weight}}\times 100$$

(1)

Density was measured using a 25 mL calibrated glass pycnometer, 5 g of powdered glass and toluene as the immersion liquid. It was calculated using Equation (2) [24].

$$\delta ={\displaystyle \frac{(m_1-m)(d_1-d_a)}{(m_3+m_1-m-m_2)}}+d_a$$

(2)

where δ = density; d_a = air density; d₁ = immersion liquid density (toluene = 0.8623 g/cm³ at 25), m = pycnometer mass; m₁ = pycnometer mass plus glass mass; m₂ = pycnometer plus glass plus liquid mass; and m₃ = pycnometer plus liquid mass. The chemical resistance and density tests, crucial for understanding the durability and physical properties of the glasses, were carried out in triplicate to ensure the reliability of the results. The values were averaged to provide a comprehensive overview. The produced glass properties were then compared with SLS waste, providing a solid basis for the conclusions drawn from this research.

3. Results and Discussion

3.1. Glass Characterization

From the content of the raw materials, the theoretical composition of each glass produced can be calculated, which is presented in

Table 4. Calculated glass compositions from content of raw materials.

Sample	Composition (Weight %)
	SiO2	Al2O3	CaO	MgO	Na2O	K2O	B2O3	Fe2O3
MTG1	61.82	1.15	9.72	0.76	10.53	7.91	7.80	0.26
MTG2	62.17	1.27	9.20	0.73	9.98	7.86	7.80	0.32
MTG3	63.41	1.40	8.69	0.69	9.43	7.97	7.80	0.38
MTG4	64.19	1.52	8.19	0.65	8.89	8.02	7.80	0.44
MTG5	64.98	1.64	7.68	0.62	8.34	8.08	7.80	0.50
MTG6	65.85	1.76	7.18	0.58	7.80	8.13	7.80	0.56

. Compositions MTG1–MTG5, which were melted at 1350 °C, formed glass that could be cast onto a steel plate. However, the composition MTG6, due to its high viscosity, was vitrified and attached to the crucible, making it impossible to overturn. This was due to the increase in silica and alumina as well as the decrease in flux oxides. The synthesized glasses presented a slight green color, turning to a yellow-green color with an increase in tailing quantity in the composition. The literature describes that clear SLS glasses have a low iron content (less than 0.5 weight % [25]); at low concentrations of Fe₂O₃ (less quantity of tailings), there are Fe²⁺ ions, since it is known that Fe²⁺ absorbs the red light region resulting in the glass appearing blue-green [26,27]. In contrast, at higher concentrations of Fe₂O₃, Fe³⁺ ions could be present in glasses, causing that coloration to turn yellow, according to what has been established in many pieces of research [27,28,29]. In addition, Figure 3 presents the XRD patterns of synthesized glass samples. It can be seen that a broad diffuse scattering at different angles and an absence of crystallization peaks indicate the complete transformation of crystalline mining tailings (which contain SiO₂ as the main crystalline phase) into an amorphous structure.

The FTIR analysis provides crucial insights into the structural arrangement of the glass network, making it a significant part of our research. The FTIR transmittance spectra of both glasses and GR are presented in Figure 4. Five main bands are observed at 430, 695, 790, 1010, and 1425 cm⁻¹, revealing a lack of a sharp form that indicates a general disorder in the silicate network [30]. The band at 430 cm⁻¹ corresponds to the bending vibrations of Si-O-Si and Si-O-Al [7,9]. At 695 cm⁻¹, a band appears, which can be attributed to B-O-B bond-bending vibrations from the pentaborate group or the bending vibrations of BO₃ triangles. Similarly, the band at 790 cm⁻¹ is attributed to the tetrahedral structure’s symmetric stretching and bending vibrations, including X-O-X bonds (where X = Si, Fe, Al, B) [7,9,31,32]. The broad and intense band at 1010 cm⁻¹ is assigned to the asymmetric stretching vibrations of Si-O-Si and the symmetric stretching vibration of the O-B-O bonds for tetrahedral units [7,9,32]. The band at 1425 cm⁻¹ is due to the B-O bond stretching vibrations and the B-O bridging between the (B₃O₆)³⁻ boroxol rings to the modes of the boron–oxygen triangular units [BO₃], which are the chains or rings of metaborate, pyroborate and orthoborate groups [33,34]; this band only appears in synthesized glasses, which have in their composition B₂O₃ as flux, and is not present in the reference glass. It is observed that the bands at 430, 790, and 1010 cm⁻¹ become more intense with the increase in tailings in the glass. This increase in intensity is a significant finding and is explained by the increasing amounts of silicate and aluminate groups, which, according to Almasri et al.’s explanations [32], leads to more intense bands. Similarly, due to the increasing aluminate groups and considering that Al³⁺ acts as a network former, non-bridge oxygens decrease, as proposed by Fernandes et al. [35], leading to a better-polymerized glass structure.

SEM analyses confirmed the presence of a single glassy phase in each synthesized glass. Figure 5 displays the MTG1 and MTG5 glass samples; the top image of the micrograph is a photograph of the glass after pouring. Based on the average results from EDS on at least five points of each sample (because boron content could not be quantified), the semi-quantitative analyses revealed an increase of Si and Al from MTG1 to MTG5. This observation aligns with the FTIR results, which showed more intense bands due to increased silicate and aluminate groups. Additionally, Figure 6 presents an elemental mapping of the MTG3 glass, a significant finding indicating a uniform distribution of the main elements in glasses. This suggests that the different groups of B₂O₃ (chains and rings detected by FTIR) could be incorporated into the silicate phase, mainly comprised of tetrahedral units of SiO₂, forming a single glassy phase. These results provide clear evidence of the uniform distribution of elements in the glass, a key characteristic of a well-polymerized glass structure.

The synthesized glasses’ thermal behavior was compared to GR, and the DTA results are shown in Figure 7. The glass transition (T_g) ranged from 625 to 831 °C. It was expected that the T_g of all the glasses would have a similar behavior to the MTG1 glass (T_g = 625 °C), with a slight increase due to the increase in silica and alumina in the composition; however, for the MTG2–MTG5 glasses, an apparent Tg is shown. With an increase in mining tailings, T_g also increased, exceeding the T_g of the GR (T_g = 808 °C). Glasses MTG1, MTG2, and MTG3 exhibited exothermic peaks at 820, 890, and 958 °C, respectively, which can be considered as the crystallization temperatures (T_c). On the other hand, compositions MTG4 and MTG5 showed a behavior similar to SLS glass (GR), where an apparent Tg was observed but not a Tc peak. According to Terry Lay et al. [14], glass stability is often characterized by the difference between the beginning of the glass transition region and the appearance of the first crystallization peak. Also, the absence of exothermic peaks can indicate a lack of crystallization. Hence, compositions MTG4 and MTG5 (with 17 and 21.3% of mining tailings, respectively), with the highest Al₂O₃ content, might be considered as stable as GR glass, instilling confidence in the research findings.

3.2. Properties Evaluation

The chemical resistance and the density of the synthesized glasses were measured and compared against the GR glass, and the results are presented in

Table 5. Chemical resistance and density results of studied glasses.

Sample	Basic Medium		Acid Medium		Density
	Average (% Losses)	Standard Deviation (σ)	Average (% Losses)	Standard Deviation (σ)	Average (g/cm3)	Standard Deviation (σ)	Calculated from Table 4
MTG1	−9.07	3.21	1.5	2.55	2.35	0.021	2.56
MTG2	−6.11	7.14	0.86	0.59	2.27	0.003	2.55
MTG3	−8.27	8.11	0.65	0.99	2.21	0.027	2.54
MTG4	−0.62	3.16	0.23	0.52	2.18	0.086	2.52
MTG5	−0.79	3.97	0.18	0.15	2.12	0.122	2.51
GR	−3.41	3.96	0.18	0.1	2.11	0.075	2.50

; according to the standard, samples with values higher than 5% should not be used in humid atmospheres; therefore, compositions MTG4 and MTG5 presented better behavior. In this sense, in the basic medium, the samples presented negative values, which means that the samples gained weight during the test. The values varied from −0.62 to −9.07%. The MTG4 (−0.62%) and MTG5 (−0.79%) samples presented the best behavior, even better than that of the GR (−3.41%). Considering that the test was carried out with an aqueous solution of NaOH, the weight gain could be due to the incorporation of free water molecules into the leached layers of the glass surface, according to work established by Paul A. [36]; also, several researchers have explained the mechanism as the dissolution of the SLS glass network due to a hydrolysis reaction, as is shown in Equation (3) [13,37,38]:

Si-O-Si- + OH- ↔ -Si-OH + -Si-O-

(3)

These results indicate that the synthesized glasses, particularly MTG4 and MTG5, could be suitable for applications requiring high chemical resistance, such as in the pharmaceutical or chemical industries.

The glass’s resistance varied from 0.16 to 1.15% weight loss in an acidic environment. It was observed that as the mining tailings content increased, there was better chemical resistance. Glasses MTG4 and MTG5 showed the best results, with tailings content of 17% and 21.3%, respectively, similar to the reference glass. The reaction mechanism is the same as proposed by several researchers, who established that acid attack involves an ionic diffusion process due to the proton domain. The attack is related to the exchange of alkaline/alkali-earth ions (such as Na⁺/K⁺ and Ca²⁺/Mg²⁺ in the glass) with H⁺ and H₃O⁺ ions from the acid and water, as shown in Equations (4) and (5) [13,37,38].

Si-O-M⁺ + H⁺ ↔ -Si-OH + M⁺

(4)

Si-O-M⁺ + H₃O⁺ ↔ -Si-OH + H₂O + M⁺

(5)

However, these chemical resistance results should be taken with caution, since the dispersion of the reported values is high (high standard deviation).

On the other hand, it is well known that the introduction of alkaline fluxes, such as K₂O into the batch, brings a detriment of glass properties, including chemical resistance, as the introduction of B₂O₃ and Al₂O₃ (from flux and tailings, respectively) can increase corrosion resistance [21]; therefore, the mixtures used in the present investigation had good chemical resistance, despite the high percentage of alkaline flux in each composition. Likewise, it can be seen from these results that the chemical attack in the basic medium was more aggressive than acid medium for all glasses. The above-mentioned agrees with the literature data, which establishes that oxide glasses generally have good resistance to attack by acid solutions and low resistance to alkaline attack [36,39,40].

After the chemical attack of the MTG4 and MTG5, the filtered solution was analyzed, and the As, Pb, and Zn content was presented in

Table 6. Concentration of As, Pb and Zn in the lixiviated solution after chemical attack of MTG4 and MTG5 glasses.

Analyte	Concentration (mg/L)
	MTG4 (HCl)	MTG5 (HCl)	MTG5 (NaOH)
As	0.66	0.51	3.81
Pb	0.16	0.31	Not detected
Zn	0.19	0.18	1.03

. The As values are better than those reported by Liu et al. of 0.88 mg/L, after soaking glasses in water at 35 °C for 28 days [13].

Finally, the experimental density values (Table 5) ranged from 2.35 to 2.12 g/cm³, while calculated densities according to Fluegel [41] ranged from 2.56 to 2.50. Although the values clearly differ, as observed, there is a decrease in density with the increase in mine tailings, which can be attributed to a greater degree of structural compaction of the glassy matrix, due mainly to the increase in silica content and decrease in alkaline oxides, see Table 3. MTG4 and MTG5 presented similar values to SLS reference glass (GR). Chemical resistance and density results confirm that a well-polymerized glass structure was obtained.

4. Conclusions

In this study, the feasibility of obtaining new light green glasses from mining tailings (incorporating up to 25.6 weight %) and waste glass was demonstrated. The results indicated that glasses MTG4 and MTG5 had excellent thermal stability with a T_g of 818 and 831 °C, respectively. Also, the glasses boosted good chemical behavior due to the better-polymerized glass structure, attributed to the increase in SiO₂ and Al₂O₃ and the decrease in alkaline oxides. It could be proved that glasses retain potentially toxic elements (As, Pb, and Zn) in their structures. Therefore, tailing vitrification is a viable process for reducing both environmental pollution and waste volume in the northern region of Mexico, minimizing the health risks of the population. Finally, we can establish that for each ton of glass produced, up to 213 kg of mining tailings and 639 kg of cullet could be recovered, although, for industrial considerations, dilatometric and viscosity analyses are required. Future work will be focused on the potential of producing glass-ceramics using a parent heat treatment and their characterization toward applying them as glazes on ceramic bodies.