Effect of Gold Tailing Addition on the Mechanical Properties and Microstructure of Foam Ceramics

Abstract

Tailings, with their low utilisation rate and value, pose an urgent problem that needs to be solved. This study focused on foam ceramics and glass granules, which were prepared using gold tailings as the main raw materials, and the prepared microcrystalline foam ceramics were investigated. The optimum sample (S6) of foam ceramics consists of 77 wt% of gold tailings, 6 wt% of talc, 9 wt% of bentonite, 2 wt% of calcite, 0.9 wt% of foaming agent and 0.6% of water reducer. The optimum foaming temperature is 1190 °C, and the holding time is 60 min. The compressive strength and bulk density of the samples are 7.4 MPa and 396.4 kg/m³; nucleation temperature, crystallisation temperature, and crystallisation activation energy of glass particles are 700 °C ± 50 °C, 1100 °C ± 50 °C, and 1572 kJ/mol, respectively. Microanalysis results showed that the crystal phases of foam ceramics were mainly quartz, calcium silicate, and aluminium silicate, and crystallinity increases with increasing gold tailing content. Wollastonite appeared in the crystalline phase of microcrystalline foam ceramics. When the glass granule addition reached 30% (2–3 samples), the compressive strength was increased by 19% (8.81 MPa), and the pore size distribution was more uniform.

1. Introduction

In 2020, the total discharge of tailings in China reached 1.295 billion tonnes. The majority of this amount comes from iron tailings, followed by copper and gold tailings, accounting for about 41.66%, 25.84%, and 14.51% of the discharge quantity, respectively [1], and the total amount of the comprehensive utilisation of gold tailings exceeded 56.16 million tonnes, and the comprehensive utilisation rate was about 26%. Based on their chemical characteristics, gold tailings of similar composition are expected to replace the silicon–aluminium oxide that is used as one of the raw materials for foam ceramics.

Foam ceramics [2,3,4] are a kind of building fireproof and thermal insulation materials, consisting mainly of silicon–aluminium oxide, inorganic materials, and a foaming agent. In the past 20 years, people have been exploring the feasibility of using industrial wastes as raw materials for the production of glass and glass ceramics [5]. Solid wastes [6,7,8,9], such as fly ash, bottom ash, and industrial metallurgical slag, have been used to produce casting or sintered glass ceramics. Chen [10] developed a lightweight thermal insulation wall material using iron tailings, bentonite, and rice husk with uniform and moderate pores and compressive strength of up to 7.6 MPa. Yang [11] used 40–55% of iron tailings and 45–60% of waste stone as the main raw materials and controlled the foaming agent dosage and fineness. Zhou [12] used gold tailings as the main raw material and added coal gangue, lightly burned magnesia, and alumina as auxiliary raw materials to prepare closed-pore foam ceramics. Liu [13] used gold tailings with quartz sand and limestone as the main raw materials and controlled the crystallisation amount by controlling the maximum sintering temperature and crystallisation holding time. Cao [14] used gold tailings as the main raw material to prepare glass ceramics with wollastonite as the main crystal phase, and the properties of the products were better when the nucleation and crystallisation temperatures were 892 °C and 976 °C, respectively. Based on powder sintering with SiC as foaming agents, Wu [15] produced glass ceramic foams that exhibit satisfactory compressive strength of about 1.5 MPa using coal pond ash. Fernandes [16] made an attempt to prepare high-compressive strength foams from mixtures of sheet glass cullet in combination with fly ash. The properties of foamed ceramics with different bulk densities are different. Taking the foamed ceramics with fire resistance reaching A1 level and bulk density of 400 kg/m³ as an example, it has the characteristics of high porosity, low thermal conductivity (≤0.12 w/m·k), and high compressive strength (≥4.00 MPa). It can be made into ceiling and floor and other building panels. Microcrystalline foamed ceramic is a potential functional building insulation material with light volume and high strength. By mixing a certain amount of glass particles on the basis of the foaming ceramic process and relying on the glass material to control the combination of ceramic crystallisation and foaming, ceramic materials with better performance can be prepared.

In this paper, foam ceramics, glass particles, and microcrystalline foam ceramics were prepared with gold tailings as the main raw materials, and the effects of gold tailing content on the mechanical properties, such as bulk density and compressive strength, were studied. The mechanism of foaming and crystallisation, mineral phase, and microstructure were studied by X-ray diffraction (XRD), Fourier transform infrared (FTIR) spectroscopy, differential scanning calorimetry (DSC), scanning electron microscopy (SEM), and energy-dispersive spectroscopy (EDS), which provided a theoretical basis for the preparation of high value-added building materials with a large number of gold tailings.

2. Materials and Methods

2.1. Preparation of Materials

2.1.1. Foamed Ceramics

On the basis of previous research [17], in this study, gold tailings were used as the main raw materials instead of perlite, talc, kaolin and bentonite. Then, foaming agents (silicon carbide, SiC) and water reducer (polycarboxylic acid (PC)) produced by Beijing Muhu admixture Co., Ltd. (Beijing, China)) were added to prepare foam ceramics (

Table 1. Material composition of the foam ceramic samples.

Sample	Raw Materials (wt%)						Auxiliary Agent (wt% of Raw Materials)
	Gold Tailings	Perlite	Kaolin	Bentonite	Calcite	Talc	Foaming Agent	Water Reducer
S0 (dry)	0	77	6	9	2	6	0.6	0
S0 (wet)	0	77	6	9	2	6	0.6	0.6
S1	30	47	6	9	2	6	0.6	0.6
S2	50	27	6	9	2	6	0.6	0.6
S3	70	70	6	9	2	6	0.6	0.6
S4	77	0	6	9	2	6	0.6	0.6
S5	77	0	6	9	2	6	0.2	0.6
S6	77	0	6	9	2	6	0.9	0.6
S7	77	0	6	9	2	6	1.2	0.6
Price (CNY/t)	0–20	600–1500	1500–2000	300–500	100–500	800

).

The gold tailings used in this study were taken from the Sanshan Island Gold Mine in Shandong Province [18]. The chemical components of the tailings are shown in

Table 2. Chemical composition of the various raw materials used for foam ceramic preparation.

Composition	SiO2	Al2O3	K2O	MgO	Na2O	Fe2O3	CaO	TiO2
Gold tailings	78.75	10.99	4.23	0.48	0.44	2.61	1.30	0.13
Perlite	72.41	11.97	4.71	0.25	2.86	0.85	0.47	0.11
Talc	18.54	0.46	0.01	39.56	0.02	0.43	0.96	0.01
Kaolin	69.94	15.43	4.93	0.20	0.24	1.80	0.15	0.23
Bentonite	69.74	15.40	2.00	1.15	0.48	1.61	1.11	0.20

All values are in weight percent.

, where SiO₂ and Al₂O₃ are at 78.75 and 10.99 wt%, respectively. The gold tailings were analysed by XRD (Figure 1). Results showed that the gold tailings were mainly composed of quartz and potassium (sodium) feldspar.

It has been demonstrated that silicon carbide is a very effective foaming agent. The carbide can form CO₂ and CO gas bubbles, which are formed in a viscous matrix, and the component retains its shape during cooling [15,16].

The chemical composition of the gold tailings in this experiment is mostly similar to that of perlite. As shown in Table 2, perlite is a kind of glassy rock dominated by an amorphous phase. Its SiO₂ and Al₂O₃ contents are 72.41 and 11.97 wt%, respectively, which are suitable for making high-temperature foam ceramics. Previous studies [19] have shown that the melting and softening temperatures of perlite are distributed at around 1500 °C and 1175 °C.

The chemical composition of talc is shown in Table 2, which consisted of 39.56 wt% MgO and 18.54 wt% SiO₂, and kaolin was mainly composed of SiO₂ (69.94%). The main components of bentonite included SiO₂ (69.74%) and Al₂O₃ (15.40%). The main mineral composition of bentonite was montmorillonite.

2.1.2. Glass Granules

Based on previous research, combined with the composition characteristics of gold tailings and the phase diagram of the CaO–Al₂O₃–SiO₂ glass system, the main crystal phase of glass was set to the CS phase region [20], as shown in Figure 2. Furthermore, the principal composition range of the CaO–Al₂O₃–SiO₂ glass system was designed as follows: 45–60% SiO₂, 5–15% Al₂O₃ and 15–30% CaO. The composition of the glass granules is shown in

Table 3. Composition of glass granules.

Gold Tailings	CaO	ZnO	BaO	Na2CO3	Na2B4O7·10H2O
75	15	3	3	2	2

All values are in weight percent.

.

2.2. Methods

2.2.1. Preparation of Foam Ceramics

The preparation process of foam ceramics is shown in Figure 3, which is blending, grinding, mixing, filtration drying, moulding, sintering, and cutting.

Each raw material was ground to less than 200 mesh for 3–6 min. Then the raw materials were divided into two parts: one is dry grinding, and the other is wet grinding. Silicon carbide [21] was selected as the foaming agent. The foaming agent was then added to one of the raw material powders and ground for 3 ~ 5 h by using a laboratory SM500 × 500 small ball mill until its particle size was less than 45 μm. A measured amount of water was added to the foaming agent and raw material powder and stirred evenly to prepare a slurry with a water content of 40–50%. Using a planetary ball mill with Φ1 cm corundum balls as a grinding medium, wet grinding of the slurry was undertaken for 4–6 h until samples were milled to very fine particles (<45 μm). The uniformly mixed raw materials in slurry form were subsequently poured into a vacuum filter to filter out water to obtain a solid block (residue), dried in an electric blast drying box at 50–60 °C for 10–12 h, and kept its temperature at 50 °C. The dried blank was ground into powder to a particle size of less than 2 mm. The blank powder was stored in a square mould surrounded by refractory bricks wrapped around the release paper. The sample was heated from room temperature to 1100–1200 °C at a heat-up rate of 10 °C/min. After 1–1.5 h, it was quenched by air to around 900 °C and then cooled from 900 °C to room temperature using a furnace. The sample was cut into a cube measuring 4 cm × 4 cm× 4 cm. It was then weighed and tested for compressive strength using a YED-200 electronic pressure testing machine. The remaining sample powder was used for other tests.

2.2.2. Preparation of Glass Granules

The raw material was uniformly mixed for 30 min and was then placed into a corundum crucible. Using a high-temperature furnace, the corundum crucible was heated from room temperature to 1500 °C at a heat-up rate of 10 °C/min and at a holding time of 90–120 min at 1500 °C. Then we took it out and quenched it with water, and ground it in a corundum crucible to obtain glass granules. The basic glass granules after water quenching are shown in Figure 4.

2.3. Preparation of Microcrystalline Foam Ceramics

The method for preparing microcrystalline foam ceramics is to add glass granules to the original foam ceramics. The ratio of raw materials for preparing microcrystalline foam ceramics is shown in

Table 4. Preparation parameters for the microcrystalline foam ceramic samples.

Sample Number	Glass Granule Content (wt%)	Original Foam Ceramic Content (wt%)	Foaming Temperature (°C)
1-1 (S6)	0	100	1190
1-2	10	90	1190
1-3	30	70	1190
1-4	50	50	1190
1-5	70	30	1190
1-6	100	0	1190
2-1	0	100	1100
2-2	10	90	1100
2-3	30	70	1100
2-4	50	50	1100
2-5	70	30	1100
2-6	100	0	1100
3-4	50	50	1050
3-5	70	30	1050
3-6	100	0	1050
4-4	50	50	1000
4-5	70	30	1000
4-6	100	0	1000

All values are in weight percent.

. For the samples with glass granules, the crystallisation temperature in the sintering process needs to be reduced to achieve the appearance or mechanical properties similar to those of foam ceramics. The higher the content of glass particles, the lower the crystallisation temperature. None of the samples could be foamed at 1000 °C. In addition, there were 4 samples that could be made with a good foamed state, which were S6(1-1) at 1190 °C, 1-2 at 1150 °C, 2-3 at 1100 °C and 3-4 at 1050 °C.

2.4. Characterisation of Samples

The compressive strength and bulk density measurements were carried out according to the Test Method for Inorganic Hard Thermal Insulation Products (GB/T5486-2008). A YED-200 electronic pressure tester (Yokota Industrial Co., Ltd., Osaka, Japan) was used, and the loading speed was about 100 N/s. The mineral and crystallisation degrees of the raw materials and sintered samples were obtained using an Ultima IV XRD analyser (Rigaku, Tokyo, Japan). The samples were scanned at a scanning speed of 10°/min under a voltage of 40 kV and current of 100 mA, and the scanning angle range was at 5–90°. The material composition was analysed using an XRF-1800 scanning X-ray fluorescence spectrometer (Shimadzu, Kyoto, Japan).

For the differential thermal analysis, the equipment used was Netzsch DSC 404 (Netzsch, Selb, Germany); the five heating rates were set at 2.5 °C/min, 5 °C/min, 10 °C/min, 20 °C/min, and 40 °C/min; and the temperature range was from 50° to 1300 °C. For the FTIR measurements, a Nicolet™ iS10 FTIR spectrometer (Thermo Fischer Scientific, Waltham, MA, USA) was used, and the wavenumber range was 400–4000 cm⁻¹. The micromorphology of the microcrystalline foam ceramics was observed using a scanning electron microscope and energy-dispersive spectrometer that used FEI Quanta 650 (Thermo Fisher, Waltham, MA, USA).

3. Results

3.1. Compressive Strength and Volume Density

Figure 5 and Figure 6a show the comparison of the properties of the foam ceramics prepared by dry grinding and wet grinding. It can be seen that the compressive strength of the S0 sample after wet milling was 7.6 MPa, which was much higher than that after dry milling (3.8 MPa). The volume density of wet- and dry-milled samples were 407.8 and 600.4 kg/m³, respectively, because the raw material distribution was not uniform in the sintering process. It can be seen from Figure 6b and Figure 7 that the compressive strength decreased by 3.7%, from 8.7 to 7.8 MPa, and the bulk density decreased by 23.2%, from 529 to 406.3 kg/m³, as the proportion of gold tailing replacing perlite increased. Considering the economic and environmental benefits of the solid waste utilisation ratio, 77% of gold tailings (100% instead of perlite) was the most suitable amount.

Figure 6c and Figure 8 show the basic properties of the S4, S5, S6, and S7 foam ceramics. It can be seen that the compressive strength and bulk density of the foam ceramics decreased with an increasing amount of the foaming agent. It can be seen that the pore diameter of the S5 sample was relatively small, and the sample height under the same heating curve was only 2 cm. Although the compressive strength was very high under this condition, the volume density of the S5 sample was as high as 517.7 kg/m³, which did not meet the requirements of light materials (400 kg/m³). The foaming height of the S4 and S6 samples could reach more than 4 cm, and the difference in their cross sections was obvious, as shown by the presence of some large and not round and oval pores in the S6 samples, whereas some significantly small pores could be observed in the S4 samples. When the volume densities of the S6 and S7 samples were less than 400 kg/m³, which was our purpose, the compressive strength of S6 was higher. Therefore, 0.9% was determined as the most suitable incorporation amount of foaming agent.

After obtaining the optimal ratio of gold tailings foam ceramics, we continued to study the glass granules addition in foam ceramics. The compressive strength and bulk density of the four groups (1-1, 1-2, 2-3, 3-4) of samples with the best foaming state were measured, and results are shown in Figure 6d. With the addition of an appropriate number of glass granules, the overall compressive strength of the microcrystalline foam ceramics with glass granules had a trend of first increasing and then decreasing compared with the foam ceramics from gold tailings. On the other hand, the bulk density showed a decreasing trend at first and then increased. When the optimum number of glass granules added was about 30%, the compressive strength and bulk density could reach 8.81 MPa and 391.5 kg/m³, which is equivalent to a 16% increase and 1.2% decrease, respectively.

3.2. Differential Calorimetric Analysis of Glass Granules

Through differential thermal analysis of the glass granules, the heat treatment system needed for the micro-crystallisation process can be determined. The crystallisation activation energy can directly reflect the influence of composition change on the energy required for crystallisation. If the crystallisation activation energy is reduced after the composition change, the difficulty of crystallisation is reduced, and the degree of crystallisation can be controlled by reducing the temperature or reducing the holding time. Through the endothermic and exothermic conditions in the curve, the nucleation temperature (T_g) and crystallisation temperature (T_p) can be determined.

It can be seen from Figure 9a that there was an obvious endothermic and exothermic process in the heating process; the glass transition temperature was about 700 °C, the crystallisation temperature was around 1100 °C, and the heating rate had a significant influence on the peak value of the curve. When the heating rate increased, the area of the exothermic peak increased, and the exothermic peak temperature moved to a higher value (T_p increased from 1125.5 to 1153.6 °C). It can be found that the crystallisation peak tended to become sharp. Based on the thermal analysis curve, the glass transition temperature and exothermic peak temperature at different heating rates are shown in

Table 5. T_g and T_p values of the glass granules at different heating rates.

Heating Rate α (K/min)	2.5	5	10	20	40
Tg (°C)	673.3	699.2	689.1	693.8	703.4
Tp (°C)	1125.5	1130.1	1132.8	1139.2	1153.6

and

Table 6. Relationship between crystallisation peak temperature T_p and heating rate α.

Heating Rate α (K/min)	Tp (°C)	Tp (K)	1000/Tp	ln(Tp2/α)
5	1125.5	1398.5	0.715	13.57
10	1130.1	1403.1	0.713	12.88
15	1132.8	1405.8	0.711	12.19
20	1139.2	1412.2	0.708	11.51
40	1153.6	1426.6	0.701	10.84

. The 1000/T_p values were plotted against ln(T_p²/α), and each point is fitted with a straight line. The slope of the straight line is E max R, and the linear fitting result is shown in Figure 9b. According to the slope calculation, the crystallisation activation energy E of the glass granules was 1572 kJ/mol. Therefore, the crystallisation of basic glass required 1572 kJ/mol of energy, which has a certain reference significance for the study of the temperature-control curve.

3.3. Microstructure Analysis of the Foam Ceramics and Microcrystalline Foam Ceramics

It can be seen from Figure 10a that quartz (SiO₂), calcium silicate (CaSiO₃), and aluminium silicate (Al₂SiO₅) were the main crystalline phases of the foam ceramics from gold tailings. With an increase in the gold tailing substitution of perlite, the glass phase and amorphous material decreased, and the diffraction peak of the main crystal phase increased, indicating that the content of the crystal phase increased with increasing gold tailing content. When the content of gold tailings was 77%, the diffraction peak intensity and crystallinity of the S4 samples were the highest, suggesting that gold tailings can fully replace perlite to achieve the maximum utilisation ratio.

Figure 10b shows the XRD patterns of the S6(1-1), 1-2, 2-3, and 3-4 samples with a good foaming state at different glass granule addition and sintering temperatures. When the sintering temperature decreased from 1190 to 1100 °C, S6 to 2-3, the crystal phases of albite (NaAlSi₃O₈) and wollastonite (CaSiO₃) began to appear obviously, indicating that the ceramic foam samples doped with microcrystalline particles appear crystalline when glass granules addition is 30%, and the sintering temperature is about 1100 °C. After adding 30% of glass granules, the crystal diffraction peak in the spectrum of the 2-3 sample was obviously enhanced, and the increase of crystallisation degree could increase the strength of the supporting material. With the addition of 50% of glass particles, the crystal phase increased, but the intensity of the crystal diffraction peak was weaker than that in the spectrum of the 2-3 sample. It can be seen that the main crystal phases of samples 2-3 and 3-4 were Quartz, albite (NaAlSi₃O₈) and wollastonite (CaSiO₃).

As shown in Figure 11a, the absorption peaks of the foam ceramics with different contents of gold tailings in the wavenumber range of 4000–400 cm⁻¹ were almost the same [21,22,23,24,25,26]. The absorption bands at 3442.31 cm⁻¹ are mainly caused by the stretching vibration of the O–H in water. A strong absorption band near 1083.8 cm⁻¹ is mainly caused by the anti-symmetric stretching vibration of the [SiO₄]⁴⁻ tetrahedral Si–O–Si bond, the anti-symmetric stretching vibration of the Si–O bond, and the stretching vibration of the O–Si–O bond. The two weak absorption bands near 796 and 781 cm⁻¹ are mainly caused by the symmetric stretching vibration of the Si–O bond in [SiO₄]⁴⁻ tetrahedron, the stretching vibration of the O–Si–O, and the stretching vibration of the O–Al–O bond in [AlO₄]⁴⁻ tetrahedron. The bending vibration of the Si–O–Si bond in [SiO₄] is characterised by a strong absorption band near 460 cm⁻¹. By comparing the frequency of these characteristic peaks, it can be concluded that the foam ceramics are mainly composed of silicate minerals, which corresponds to the XRD results.

In Figure 11b, S6(1-1), 1-2, 2-3, and 3-4 samples were also selected for infrared spectrum analysis. The absorption band at 1031.05 cm⁻¹ is mainly caused by the anti-symmetric stretching vibration of the silicate Si–O–Si bond, the anti-symmetric stretching vibration of the Si–O bond, and the stretching vibration of the O–Si–O bond. With the increase of the number of glass granules and the decrease of sintering temperature, the absorption peak shifted to a low frequency (1003.87 cm⁻¹), showing the characteristics of SiO₃²⁻ groups, which is consistent with the results of XRD analysis of wollastonite crystals.

The micromorphology and pore structure of foam ceramics with different glass particle contents and sintering temperatures of four samples (1-1, 1-2, 2-3, 3-4) were observed. Combined with EDS spectrum measurement, the crystal growth of the materials can be semi-quantitatively analysed. Figure 12a–d show the fracture surface SEM images of selected samples under 100× magnification, including the S6(1-1) and 1-2 samples with 0% and 10% glass granules content at 1190 °C. It can be seen that in 1-1, there were many pores with a diameter of 0.2–0.7 mm. For the 1-2 sample, some of the tiny pores were relatively uniform, but because the original heating rate was significantly fast and the sintering temperature was extremely high, there were large pores with a diameter of about 3 mm in the picture. This is the main reason for a slight decrease in compressive strength. It was found that the pore structure of the 2-3 samples had excellent characteristics, and the pore size and distribution all below 0.5 mm could be observed. When the foaming temperature was 1050 °C, the 3-4 sample could be foamed. However, it can be found that the pores below 0.5 mm in the 3-4 sample were round. Figure 12e,g show the crystal growth and pore wall in the 2-3 sample. There was a small number of needle-like crystals between the bulk crystals, and clusters of granular crystals were widely grown on the surface. When magnified to 8000×, it can be observed in Figure 12f,h that the bulk crystal was in the shape of a cuboid with a length of about 2–3 μm. Most of the crystals in the 2-3 samples were intermelting. The composition of point 1 was analysed by EDS, and the results are shown in

Table 7. Point 1 of EDS results of the 2-3 sample under 8000× magnification.

Element	C	Na	Mg	Al	Si	Ca	Zn	O
Atomic%	12.37	1.43	0.72	3.29	10.62	8.26	1.36	61.96

and

Table 8. Point 2 of EDS results of the 2-3 sample under 8000× magnification.

Element	O	C	Na	Mg	Al	Si	K	Ca	Fe	Zn	Ba
Atomic%	63.04	9.90	1.44	0.21	4.88	15.13	1.35	2.06	0.46	0.91	0.62

. According to the point scanning results, the Ca-to-Si atomic ratio of the granular crystal phase at point 1 was 1.3. Because the ceramic foam material was porous, the point scanning result was not accurate; thus, combined with the XRD results, the granular crystal may be wollastonite.

4. Discussion

The change in the compressive strength of the foam ceramics may be attributed to the introduction of iron in the gold tailings, resulting in a higher melting point of the raw material powder [27,28], an increase in the amount of liquid in the green body, a decrease in viscosity, a relatively larger pore size, and finally the formation of closed pores. In addition, the aluminium content in gold tailings is slightly lower than that in perlite tailings [29].

By observing the specific sintering process, it is found that increasing the replacement amount of gold tailings will change the heating curve, foaming temperature and holding time. When the replacement amount of gold tailings was 30% (S1) and 50% (S2), the sample could be foamed to the appropriate height (5 cm) at 1175 °C for 60 min. When the replacement amount of gold tailings was 70% (S3), the sample needed to be foamed at 1185 °C for 60 min. When the amount of perlite was completely replaced with gold tailings, it took 60 min to complete the foaming at 1190 °C. Obviously, in order to achieve the same sintering and foaming effect under the same holding time, the required foaming temperature needs to be increased, and the replacement of gold tailings increases the required foaming temperature [16].

Through the new idea of adding glass materials into foam ceramics from gold tailings, we found that the original crystallisation temperature was set to a higher value, and the mutual melting of the crystal phase and glass phase led to a decrease in crystal content. Thus, the intensity of XRD diffraction peaks was reduced. In addition, results also showed that the addition of microcrystalline particles to the original foam ceramics could reduce the foaming temperature required in the preparation process. Thus, the energy consumption was lower than that of the original foam ceramic preparation process. When the temperature was reduced to 1100 °C, and glass granules added more than 30%, the main crystal phase type shown in the figure did not change, but the diffraction peak intensity of the albite phase was enhanced, and wollastonite appeared obviously.

The glass granules prepared by properly adding gold tailings could promote the formation of crystals in the samples, and crystallisation could occur at the same time as foaming. However, the number of glass granules should not be significantly high. The microcrystalline ceramic foam samples with superior crystallinity could be prepared with 30% of glass granules, which is consistent with the preparation conditions that yield excellent mechanical properties. The crystallisation process of microcrystalline particles and the softening process of glass would appear first. Through the test results, we speculated that when the addition amount exceeded 30%, the influence of the softening of the glass matrix (above the glass transition temperature) was greater than the crystallisation strengthening degree of the glass. Therefore, the strength results shown under the combined action of the foaming agent were quite different.

According to the results of EDS point scanning analysis and SEM detection, due to the porous material characteristics of microcrystalline foamed ceramics, there would be some errors in the point scanning results, but the element composition in the crystal could reflect the following rules to a certain extent: with the increase of glass particle content and the decrease of sintering temperature, the crystal of particle cluster growth gradually changes from the internal growth of the sample to the growth of dispersed on the surface of the sample, and the number gradually decreases. The number of bulk crystals increases significantly, and the crystal volume becomes larger. Combined with EDS point scanning analysis, the needle-like and massive crystals should be feldspar minerals, which matches the results of XRD.

5. Conclusions

The optimum sample (S6) foam ceramics consisted of 77 wt% of gold tailings, 6 wt% of kaolin, 6 wt% of talc, 9 wt% of bentonite, 2 wt% of calcite, 0.9 wt% of foaming agent and 0.6% of water reducer. It exhibits a compressive strength and bulk density of 7.4 MPa and 396 kg/m³, respectively. Therefore, the addition of 30% of glass granules increased the compressive strength by 19%. Furthermore, the nucleation temperature of the gold tailings and glass particles, crystallisation temperature, and crystallisation activation energy were 700 °C ± 50 °C, 1100 °C ± 50 °C, and 1572 kJ/mol, respectively.

The XRD analysis results of the foam ceramics from gold tailings showed that their crystal phases were mainly quartz (SiO₂), calcium silicate (CaSiO₃), and aluminium silicate (Al₂SiO₅). In addition, the main crystal phases in the microcrystalline foam ceramics were quartz (SiO₂), albite (NaAlSi₃O₈), and wollastonite (CaSiO₃).

The infrared spectrum analysis results of the microcrystalline foam ceramics showed that with an increasing number of glass granules and decreasing sintering temperature, the absorption peak shifted to a low frequency, showing the characteristics of the SiO₃²⁻ groups, which is consistent with the chemical bond characteristics of silica fume. The two low-frequency absorption peaks at 567.91 and 681.13 cm⁻¹ could reflect the participation of Al₂O₃ and other compounds in the raw materials, showing the stretching vibration of O–Al–O.

The SEM images of the foam ceramics from gold tailings showed that the pore edge of the sintered sample with the best ratio (S6. 1-1), relative to the samples made of ordinary proportions, was smooth; the curve was smooth, and the pore structure was microscopically excellent. Particularly, the difference could be seen in the small pores of 0.1 mm and below. It can be concluded that with increasing glass granules content and decreasing sintering temperature, the crystals growing in clusters gradually change from the internal growth of the sample to dispersed growth on the surface of the sample, and the number of crystals gradually decreases. The number of needle bars and block crystals obviously increases, and the crystal volume becomes larger. Combined with the 2-3 sample results of EDS, it is inferred that the needle bars and block crystals are wollastonite minerals.