J. Cent. South Univ. (2021) 28: 1637-1651

DOI: https://doi.org/10.1007/s11771-021-4723-9

Sintering behavior and mechanical properties of sintered ceramics based on spodumene tailings

YANG Jie(杨洁)¹, XU Long-hua(徐龙华)^{1, 2}, WU Hou-qin(巫侯琴)¹, WANG Zhou-jie(王周杰)¹,

SHU Kai-qian(舒开倩)¹, XU Yan-bo(徐妍博)¹, LUO Li-ping(罗利萍)¹, TANG Zhen(唐珍)²

1. Key Laboratory of Solid Waste Treatment and Resource Recycle (Ministry of Education), Southwest University of Science and Technology, Mianyang 621010, China;

2. State Key Laboratory for Environment-friendly Energy Materials, Southwest University of Science and Technology, Mianyang 621010, China

Central South University Press and Springer-Verlag GmbH Germany, part of Springer Nature 2021

Abstract: In this study, ceramics was prepared by slip casting (no pressure was used during shaping step) and atmospheric pressure sintering with low-melting point glass (LPG) powder as the binding material to facilitate the transformation of spodumene flotation tailings (SFTs) into ceramics at lower temperatures. The influence of sintering temperature and mass ratio of LPG on the mechanical properties (flexural strength and compressive strength) of ceramic materials was studied by orthogonal test. The results showed that when the mass ratio of LPG powder was higher than or equal to 20 wt% and the sintering temperature was higher than or equal to 550 °C, mutual adhesion between the sample particles was realised and consequently the ceramic materials could be prepared with good mechanical properties (the maximum flexural strength=19.55 MPa, the maximum compressive strength=42.25 MPa, average porosity=24.52%, average apparent density=1.66 g/cm³, and average water absorption=14.79%). The sintered ceramics were characterized by XRF, XRD, optical microscopy analysis, SEM, TGA-DSC and FT-IR. The formation of liquid phase at high temperature may lead to the mutual bonding between particles, which might be the main reason for the improvement of mechanical properties of ceramic materials. Overall, SFTs were successfully sintered at low temperature to prepare ceramic materials with good mechanical properties, which are crucial for energy conservation and environmental preservation.

Key words: spodumene tailings; low-temperature sintering; ceramic materials; mechanical property

Cite this article as: YANG Jie, XU Long-hua, WU Hou-qin, WANG Zhou-jie, SHU Kai-qian, XU Yan-bo, LUO Li-ping, TANGZhen. Sintering behavior and mechanical properties of sintered ceramics based on spodumene tailings [J]. Journal of Central South University, 2021, 28(6): 1637-1651. DOI: https://doi.org/10.1007/s11771-021-4723-9.

1 Introduction

Zero waste approach has garnered considerable interest in recent years due to the continuous increase in the amount of industrial by-products [1-3]. Among the industrial by-products, mine tailings are expected to dramatically increase in the coming years due to the increasing demand of metallic minerals in the global market [4-6]. This can increase the generation of waste water and tailings during the process of beneficiation, thereby increasing the environmental pollution [7-9].Lithium is the least dense metal, which is widely used in aeronautics, energy, and optical domain due to its outstanding physical and chemical properties [10-13]. Spodumene is one of the main sources of lithium. Consequently, the efficient separation of spodumene from gangue minerals and the extraction of lithium has gained significant research attention [14]. With a reported lithium reserve of 0.48 Mt, the Jiajika pegmatite deposit in western Sichuan Province of China is the most important solid lithium deposit in Asia [15]. SHU et al [16] carried out an in-depth study on the enhanced flotation of spodumene. ZHU et al [17] have studied the effects of grinding environment and lattice impurities on spodumene flotation. Due to the relatively low lithium content of approximately 3.0 wt%-7.7 wt% LiO₂ in natural spodumene, the production of lithium from spodumene ore generates high volume of industrial by-products like spodumene tailings [18, 19]. Spodumene tailings are solid wastes that are discharged after beneficiation. These solid wastes are characterized by high contents of alumina and silica [20], and are generally deposited in tailing ponds to form tailing wasteland. This treatment causes air, soil, and water pollution in the surrounding region, which is harmful to human health [21-24].

How to turn flotaion tailings into resources has been a subject of intense investigation [25]. In fact, the effective use of tailings has always attracted considerable research interest [23, 26]. YIN et al [27] studied the feasibility and sustainability, for the use of iron ore tailings as additives in the ceramic industry. For instance, LIU et al [28] prepared ceramics substrate sintering at 1000-1250 °C from tungsten mine tailings. However, the effective use of spodumene tailings has been rarely studied in the field of ceramic materials, especially in the context of low-temperature sintered ceramic materials. LEMOUGNA et al [29, 30] manufactured ceramic materials from spodumene tailings sintering at 700-900 °C and 1050-1200 °C. The demand for ceramic materials is continuously increasing [31], and the potential utilization of mine tailings in the production of ceramics complies with the concept of circular economy, which is an approach for sustainable economic development and environmental preservation.

In this study, we proposed a practical and cost-effective approach for reducing the environmental hazards of spodumene tailings. The ceramics were prepared using spodumene tailings as the principal raw material. The sintering temperature of the ceramics was found to be about 550 °C. Further, we systematically studied the basic mechanism for low-temperature sintering of ceramic materials at approximately 550-650 °C. Furthermore, to widen the practical applications ceramic materials were prepared by optimizing the chemical composition and sintering temperature.

2 Experimental

2.1 Raw materials

Spodumene flotation tailings (SFTs) used in this study were obtained from a concentrator in Ganzi Prefecture, Sichuan Province, China. A commercial kaolin and low-melting point glass (LPG) powder were purchased for the study. The chemical composition of the raw materials is presented in Table 1.

SFTs are raw ores without ball milling and refining treatment. The particle size of SFTs and other raw materials are presented in Table 2.

Table 1 Chemical composition of main raw materials (%)

Table 2 Particle size distribution of raw materials

2.2 Material design

To effectively utilise the solid waste, ceramics were prepared using SFTs as the principal raw materials. Further, to facilitate wide applications of the ceramics, the compositions of raw materials were considered to be as simple as possible (SFT, kaolin, LPG). To save energy, we eliminated some processes including grinding and refining of raw materials and reducing the sintering temperature to a large extent. The influence of sintering temperature and mass ratio of LPG on the physical propertie (flexural strength and compressive strength) of ceramic materials was studied by orthogonal test.

2.3 Experimental procedure

Put the raw materials into NJ-160 cement paste mixer (produced by Wuxi Construction Engineering Test Equipment Co., Ltd., China) and then mix for 4 min. The water/material mass ratio was 0.3:1; three samples were poured in each pot; and the average value was taken as the test result. The ceramics was shaped in rectangular prism alloy moulds (120 mm×20 mm×20 mm) and cylindrical moulds (Φ20 mm×20mm). Subsequently, they were placed in an electric blast dryer and dried at 60 °C for 48 h. After drying, they were demoulded and baked at 120 °C for 24 h. After the samples were cooled to 40 °C, they were sintered using XZWL-14-12Y type medium-temperature test furnace (Sinosteel Luoyang Institute of Refractories Research Co., Ltd., China). The heating rate was 8 °C/min, and the dwell time was 2 h at the maximum temperature (550, 600 and 650 °C). After the sintering process, they were naturally cooled down to 40 °C.

2.4 Characterization methods

2.4.1 Mechanical properties

The flexural strength and compressive strength of the ceramic materials were assessed according to GB/T4741-1999 (Standard test method for bending strength of ceramic materials, 1999) and GB/T4740-1999 (Standard test method for compressive resistance of ceramic materials, 1999), respectively. Three-point flexural strength was determined using a mechanical testing machine (Shenzhen Wance Experimental Equipment Co., Ltd., China) and the compressive strength was tested by the YAW06 microcomputer controlled pressure testing machine produced by Mets Industrial System Co., Ltd., China. The flexural strength was calculated according to the following equation:

δ_f=3FL/(2bd²)                                            (1)

where δ_f is the flexural strength, MPa; F is the maximum load, N; L is the support distance, mm; b is the width of the tested beam, mm; d is the height of the tested beam, mm. Further, the compressive strength was calculated as follows:

δ_c=4P/πD²                                                (2)

where δ_c is the compressive strength, MPa; P is the maximum load, N; D is the diameter of the tested beam, mm. For each composition, at least three specimens were tested, and the average was considered the representative value of the strength.

The shrinkage, Y was calculated as follows:

                          (3)

where L₀ and L are the lengths of the unsintered and sintered samples, mm, respectively.

The apparent porosity and apparent density were assessed according to GB/T1966-1996 (Test method for apparent porosity and bulk density of porous ceramic, 1996). The apparent porosity q (%) was calculated as follows:

                         (4)

where m₁ and m₂ are the masses of the dry and wet (samples immersed in boiling water for 12 h) samples, g, respectively; m₃ is the mass of the sample in water. The apparent density (ρ) was calculated as follows:

                                  (5)

where m (g) and v (cm³) are the mass and volume of the sample, respectively.

The water absorption W_a was tested according with GB/T3299-2011 (Test method for water absorption of domestic ceramicware, 2011). It was calculated as follows:

                        (6)

where m₀ and m₁ are the masses of the dry and wet samples, g, respectively.

2.4.2 Chemical and phase composition

The chemical composition was examined by Axios X-ray fluorescence (XRF) spectrometer (PANalytical B.V). The phase compositions of powdered material were examined by D/max IIIA X-ray diffractometer (XRD) produced by Japan Co., Ltd.

2.4.3 Optical microscopy analysis

The glass slides prepared by the sample were visualized by an optical microscope (DMI3000, Leica, Germany).

2.4.4 Scanning electron microscopy (SEM)

The samples were coated with carbon and examined using scanning electron microscopy (SEM, LEO440, Leica Cambridge Ltd.)

2.4.5 Thermogravimetric and differential scanning calorimetry analysis

Thermogravimetric analysis (TGA) and differential scanning calorimetry (DSC) were simultaneously performed in air using a comprehensive thermal analyser (Jupiter STA449C, Netzsch) at a constant heating rate of 20 °C/min. The samples were heated from room temperature to 700 °C.

2.4.6 Fourier transform infrared spectroscopy analysis

Fourier transform infrared spectroscopy (FT-IR) was used to test the samples with frontier infrared spectrometer produced by Perkin Elmer Instruments Co., Ltd., USA. The medium is potassium bromide (KBr).

2.5 Design of orthogonal experiment

The characteristics of the ceramics were optimized to ensure their wide practical applications.

2.5.1 Test indicators

The flexural strength and compressive strength were considered the optimization parameters.

2.5.2 Optimal levels of parameters

The optimization parameters in Section 2.2 were considered in the orthogonal experiments. In order to explore the general range of material preparation conditions, conditions were explored through pre experiment. The main method is the single factor control variable method. Firstly, the quality ratio of unfired samples can be determined by adjusting the content of kaolin, and the general conditions of successful bonding can be determined by adjusting the quality ratio of LPG and sintering temperature. Figure 1(a) shows that when the sintering temperature was less than 550 °C, the sample exhibited low flexural strength. When the temperature was 550 °C, the strength of the sample was significantly increased, so the sintering temperature was determined to be 550, 600 and 650 °C, respectively. Similarly, when the content of LPG powder was less than 15 wt%, the sample exhibited low strength, so the mass ratio of LPG powder was considered to be 20 wt%, 25 wt% and 30 wt%. When the mass ratio of kaolin was 10 wt%, it facilitated the demoulding of the sample. The formulation design for a series of experiments is presented in Table 3. Figure 1(b) shows that when the common glass powder is used as high temperature binder, its flexural strength is lower than that of LPG under the same conditions.

Figure 1 (a) Effect of proportion of LPG on flexural strength at different sintering temperatures; (b) Comparison of flexural strength between LPG and common glass powder at different sintering temperatures

Table 3 Factors and levels of orthogonal experiment L₉(3⁴)

3 Results and discussion

3.1 Macroscopic appearance of ceramic materials

The images of the prepared specimens at 550-650 °C and the section of sample at each sintering temperature are presented in Figures 2(a) and (b), respectively.

It is evident in Figure 2(a) that the sample was well formed and was uniformly shaped. According to Figure 2(b), when the sintering temperature was 500 °C, the section of sample exhibited rough surface; the adhesion between the particles was weak, and the sample was porous and loose. When the sintering temperature reached 550 °C, the surface of the sample changed significantly; the section was smoother and the adhesion between particles was more compact.

Figure 2 (a) Image of prepared samples of ceramic materials; (b) Section of sample at each sintering temperature

3.2 Orthogonal experimental analysis

3.2.1 Flexural strength and compressive strength

Each level number in the orthogonal columns is replaced by the actual level value, which forms the scheme for the orthogonal experiment, as shown in Table 4.

3.2.2 Range analysis

The effects of two factors on the flexural strength and compressive strength were evaluated by investigating the impact of various levels of each factor. The conditions for every test and corresponding results are shown in Figure 3. D and E represent flexural strength and compressive strength, respectively. Here, there are six crucial parameters: K₁, K₂, K₃, L₁, L₂, and L₃. K_i (i=1, 2, 3) is defined as the average of the evaluation index D (flexural strength, MPa) of three levels in each factor. L_i (i=1, 2, 3) is defined as the average of the evaluation index E (compressive strength, MPa) of three levels in each factor.

Table 4 Results of orthogonal table L₉(3⁴)

Figure 3 Results of flexural strength and compressive strength of samples

The ranges of R₁ and R₂ are the main indexes that can be visually analyzed by orthogonal experiment. R₁ and R₂ are defined as the range between the maximum and minimum K and L values in the column of the corresponding factor. They are expressed as follows:

R₁=max(K₁, K₂, K₃)-min(K₁, K₂, K₃)          (7)

R₂=max(L₁, L₂, L₃)-min(L₁, L₂, L₃)           (8)

The ranges of R₁ and R₂ reflect the degree of influence of each factor on the flexural strength and compressive strength, respectively. Larger range indicates that the impact of indicators is larger [32].

It is clear from Figure 3 that the sample with the maximum flexural strength and compressive strength is S₉, and the corresponding proportion of LPG is 30 wt%. The sintering temperature is 650 °C, and the flexural strength and compressive strength are 19.55 and 42.25 MPa, respectively.

By comparing range of R₁ and range of R₂ in Table 5, we can see that the range value of sintering temperature is the larger (3.76 and 7.78, respectively) than the proportion of LPG (2.99 and 5.62, respectively), while the range value reflects the influence of this factor on the flexural strength and compressive strength of the sample. To recapitulate, the flexural strength and compressive strength of the sample are strongly affected by the sintering temperature, followed by the proportion of LPG.

3.2.3 Analysis of variance

The results of analysis of variance (ANOVA) are shown in Table 6, where the square of deviance (S_{D, E}) of each factor was calculated using Eqs. (9)- (12). Here, T_D and T_E represent the sum of the flexural strength and compressive strength, respectively. Further, S_D and S_E represent the variance corresponding to the specimen when the flexural strength and compressive strength are the test results, respectively.

                         (9)

                        (10)

  (11)

    (12)

where K_i and L_i are the average flexural strength and average compressive strength for the sample number i, respectively. The F ratio of each factor can be expressed as follows:

                           (13)

                            (14)

where S_De is the square deviance of error; f_e and f_i are the degree of freedom of error for factor e and factor i, respectively, which are both equal to 2. For the inspection levels of 0.01, 0.05 and 0.1, the critical values of F_0.01, F_0.05 and F_0.1 could be obtained from the distribution table of F-values, which were 18.00, 6.94 and 4.32, respectively. For the condition of F-ratio≥18.00, 6.94≤F-ratio<18.00, 4.32≤F-ratio<6.94, and F<4.32, the significance of the factor j was marked as “***” (highly significant), “**” (significant), “*” (less significant), and blank(non-significant), respectively.

According to the variance analysis, the proportion of LPG and sintering temperature exerted a less significant (*) influence on the flexural strength of the sample, and a significant influence on the compressive strength. There are two possible explanations for this result: 1) Due to the large size of tailings particles, there is a large gap between the particles after bonding. The LPG powder softs during the sintering process, which facilitates the bonding of the particles with each other and filling of some pores, but the large pores between the particles (including the gaps between the particles, the voids generated during the preparation of the sample and intra-particles pores) are not filled, which may be the reason for the different flexural strength of the samples [33]. This is consistent with the results of high apparent porosity in Table 7. 2) The flexural strength and compressive strength of the sample is lower than that of high-strength material, so the test results are greatly affected by the specimen size, which leads to a larger error in the results than that for high-strength material [34].

Table 6 Table for variance analysis

Table 7 Measured linear shrinkage, apparent porosity, apparent density, and water absorption

3.2.4 Two factor linear regress analysis

In order to further study the change trend of flexural strength and compressive strength with sintering temperature and mass ratio of LPG, and further predict the change trend. SPS (Statistical Product and Service Solutions) software was used for two factor linear regression analysis. The results were Eqs. (15) and (16), with R² values of 0.891 and 0.920, respectively. The closer R² value is to 1, it shows that the trend of change with the factors is more consistent with the linear regression.

δ_f=0.038x+0.299y-13.995                 (15)

δ_c=0.078x+0.562y-24.838                 (16)

where δ_f is the flexural strength, MPa; δ_c is the compressive strength, MPa; x is the sintering temperature, °C; y is the mass ratio of LP, wt%.

According to Eqs. (15) and (16), the linear binary linear regression diagram of the flexural strength and compressive strength of the sample is drawn, as shown in Figure 4. The black balls and red balls in Figures 4(a) and (b) represent the actual test results of flexural strength and compressive strength, respectively, and the color plane represents the simulation and prediction results of binary linear regression. It can be seen from the Figure 4, that the fitting results are very consistent with the actual results, and the change trend of sample strength with the factors can be shown clearly. The flexural strength and compressive strength of the samples are directly proportional to the sintering temperature and the mass ratio of LPG. However, considering the cost of raw materials and heating energy consumption, the practical application of the product needs more exploration.

Figure 4 Linear regression diagram of sample:

3.2.5 Linear shrinkage, apparent porosity, apparent density, and water absorption

The other relevant parameters of the sample are shown in Table 7, which reveals that the average linear shrinkage, apparent porosity, apparent density, and water absorption are 2.08%, 24.52%,1.66 g/cm³ and 14.79%, respectively. The water absorption is more than 10%, which is consistent with the National Standard of the People’s Republic of China GB/T4100-2015 ceramic tiles (earthenware tile). As mentioned in Section 4.2.3, the high porosity of the samples can primarily be attributed to the accumulation of large particles, the incomplete discharge of bubbles during the forming process and the internal pores of particles. But the LPG powder could not fill all the pores between the particles. The measured water absorption and apparent density is consistent with the high porosity, which causes a large water absorption and low apparent density.

3.3 Chemical composition

To further study the sintering mechanism of the ceramic materials, we assessed the SFT by XRF and XRD, and then analyzed their chemical and mineralogical compositions.

Table 8 shows that the main chemical constituent of the samples (S₁, S₂, S₃) after sintering is SiO₂, followed by Al₂O₃, P₂O₅, etc. It is evident from Table 1 (chemical composition of main raw materials) that SiO₂ and Al₂O₃ are primarily derived from SFT, and the P₂O₅ from the LPG powder. The results of these three groups show that the main elemental constituents are Si, Al, P, Na, and K. The results of chemical composition are helpful for the accurate analysis of XRD measurements. There is no significant change in the chemical composition of the sample under different sintering temperatures.

3.4 Phase composition

The phase composition of the samples was tested by XRD as shown in Figure 5. Figure 5(a) shows the mineral phases of the raw materials. XRD of SFT in Figure 5(a) exhibits a multiphase composition consisting of quartz (SiO₂, PDF#46-1045), muscovite (KAl₂(AlSi₃O₁₀)(OH)₂, PDF#07-0042), and albite (Na₂O·Al₂O₃·6SiO₂, PDF#09-0466). The kaolin consists mainly of kaolinite (Al₂Si₂O₅(OH)₄, PDF#14-0164). According to the XRD of LPG powder in Figure 5(a), the diffraction peak is diffused, which is a typical feature of glass diffraction [35]. There is no obvious crystallization peak in the sample, which confirms that the components of the LPG powder are easily miscible, and a weak connection is formed between the groups. Further, it is difficult to rapidly crystallize and precipitate the LPG powder, and it exhibits a good thermal stability.

Figure 5(b) shows the XRD patterns of ceramic materials under different sintering temperatures. It can be seen that the main components of the sintered samples are quartz (SiO₂, PDF#46-1045) with smaller quantities of albite (Na₂O·Al₂O₃·6SiO₂, PDF#09-0466), muscovite-3T (KAl₂(AlSi₃O₁₀)(OH)₂, PDF#07- 0042), and microcline (KalSi₃O₈, PDF#19-0932), When the sintering temperature is 550 °C, the characteristic peaks of feldspar at the two red arrows disappear, which may be attributed to some crystalline variation of albite. The XRD patterns of the three sample (S₁, S₂, S₃) are basically the same. This proves that the samples have similar composition, and they mainly consist of quartz, mica, and feldspar, which is also consistent with the XRF test results of the earlier materials and samples.

The peak of kaolinite disappears in the XRD sample after being sintered, due to the dehydration and decomposition of kaolinite at about 600 °C, forming amorphous meta-kaolin.

To further confirm the above evolution phases, the green box in Figure 5(b) is enlarged in Figure 5(c). The three vertical lines with different colours (red, green, and blue) represent the diffraction peaks of quartz, microcline, and albite, respectively. It is clear from the red line that the diffraction peaks shift to the right when the sintering temperature raised to 600 °C due to the transformation of α-quartz to β-quartz at 573 °C [29]. It is evident from the green vertical line that the diffraction peak of microcline gradually vanishes with the increase in the temperature. Especially, when the sintering temperature is up to 500 °C, the diffraction peak disappears. It is observed that the albite peak vanishes when the temperature increases from 450 to 550 °C, which can be attributed to the phase transformation of albite. When the sintering temperature reaches 600 °C, the albite peak reappears and shifts to the right, which may represent the formation of a new type of albite structure after the phase transformation.

3.5 Optical microscopy analysis

For avoiding the limit of XRD results due to a small amount of certain phase in the sample, and for a more comprehensive understanding of the mineral composition of the sample, we cut the sample into thin slices and then observed them under optical microscope. The results are shown in Figure 6, which confirms a uniform distribution of the sample particles. This indicates that the raw materials were evenly mixed in the preparation process. The samples consist of quartz (Qtz), muscovite (Ms), and plagioclase (Pl) [36, 37]. While analyzing the mineral phase, we observed an obvious change in feldspar. Further, it is difficult to find a clear feldspar phase at 550 °C. The result is consistent with the peak migration result of albite in Figure 6(c).

Table 8 Chemical composition of sample at each sintering temperature (20 wt% LPG+10 wt% kaolin+70 wt% SFT)

Figure 5 XRD patterns of raw and ceramic materials under different sintering temperatures

Figure 6(c) shows that the surface of plagioclase has changed. With increasing sintering temperature, the number of pores on the surface of the sampleas well as the pore size decreases, which is consistent with the previously discussed slightly decreasing trend of apparent porosity with the increase in temperature. When the sintering temperature reaches 550 °C, the samples are closely bonded together, and the bonding material may be LPG powder. This result is consistent with the XRF and XRD measurements.

3.6 SEM analysis

The SEM images of the samples are shown in Figure 7. Here, the sintering temperatures are 500, 550, 600 and 650 °C from top to bottom. The image on the right is the magnified image of the region surrounded by white rectangle in the left images. A clear difference is observed between the images in Figure 7(c), Figures 5(d) and (e). The particles are not tightly bound in Figure 7(c), and there are several interstices between them. However, it is evident in Figures 7(d) and 5(e) that the sample has fused into a unified whole. This proves that some materials have changed and the particles are tightly bound. This may be attributed to the softening of LPG powder. However, these holes are huge, and it is clear from Figure 7(f) that the materials in the holes are not bonded together. When the sintering temperature raised to 600 °C, the sample is further changed, as shown in Figures 7(g)-(i). Two large circular pores with a size of about 100 μm are observed in Figure 7(g). This may be attributed to following two reasons: 1) The circular holes are produced in the process of sample preparation like the holes in Figure 7(b); 2) The circular pores are transformed from the pores between the large particles. The LPG powder softs and fills the edge of the pores, so the pores become circular. It is clear in Figures 7(h) and (i) that the particles are more tightly bonded and the size of most of the pores is reduced to 20-60 μm. Further, the number of fine particles in the pores is reduced in Figure 7(i). Similarly, it is evident in Figures 7(i), (g) and (l) that when the sintering temperature is up to 650 °C, the particles are more tightly bound and the number and size of pores are reduced.

Figure 6 Optical microscopy of samples under different sintering temperatures:

3.7 TGA-DSC tests

To further investigate the sintering mechanism and the phase variation of samples during the sintering process, and to validate the results of XRD analysis, we conducted TGA-DSC test on SFT. It is clear from the TGA curve in Figure 8(a) that the mass of the sample significantly decreases at approximately 100 °C, which is due to the removal of the water molecules adsorbed on particle surface.

When the temperature rises from 400 to 600 °C, the quality of the sample is significantly reduced, which may be attributed to the removal of structural water in the sample, the dehydration of kaolin into metakaolin and dehydrogenation of muscovite in the sample. A sharp exothermic peak appears at about 572 °C, due to the rapid transformation of β-quartz to α-quartz.

The TGA-DSC curve of LPG powder in Figure 8(b) confirms that the maximum mass loss of the LPG powder was only approximately 0.9 wt%. The endothermic peak is about 410 °C which is attributed to the softing of the LPG powder. This implies that when the sintering temperature reaches 410 °C, the softening process of the sample begins [38]. When the temperature reaches about 550 °C, an exothermic peak appears, which may be due to the crystallization of LPG [39]. This may be the reason for the significant increase in the flexural strength and compressive strength of the sample when the sintering temperature reaches 550 °C, which is consistent with the SEM results.

3.8 FT-IR analysis

The FT-IR spectra of samples at different sintering temperatures are shown in Figure. 9. It can be seen from Figure 9 that when the sintering temperatures were 450 and 550 °C, the infrared spectrum peaks of the sample are basically the same. This shows that the main reason for the adhesion between the sample particles is that LPG is crystallized after softing, and there is no obvious chemical change. The larger absorption peak near 3425 cm^-1 should be caused by the water absorption of KBr [40, 41]. When the temperature reached 600 °C, The absorption band near 1100 cm^-1 is attributed to P—O stretching vibration in P—O—Si linkages, whereas an absorption band near 1020 cm^-1 is assigned to Si—O stretching vibration in P—O—Si linkages [42, 43]. This means that when the sintering temperature is higher than 600 °C, the flexural strength and compressive strength of the samples increased might due to the LPG crystallization and chemical reactions (the formation of P—O bond and Si—O bond).

Figure 7 SEM of samples at different sintering temperatures:

Figure 8 (a) TGA-DSC diagram of raw material (20 wt% LPG+10 wt% kaolin+70 wt% SFT); (b) TGA-DSC diagram of LPG powder

Figure 9 FT-IR diagram of samples at different sintering temperatures

3.9 Simulation of ceramic sintering process

In order to have a more intuitive understanding of the sintering process of ceramic materials at low temperature, reasonable speculation on the sintering mechanism is shown in Figure 10. From the above results, the main reason for sintering densification is the softing of LPG, which fills the gap between SFT particles, making the particles adhere each other. Secondly, the appearance of the liquid phase also enhances the mass transfer among particles, and the particles adhere each other, improving its mechanical properties. The round hole in the figure is caused by the incomplete filling of the gap between the particles.

4 Conclusions

Ceramic materials with good mechanical properties were prepared using SFT as raw material and LPG powder as high-temperature binder at a sintering temperature of 550 °C. The maximum flexural strength and compressive strength of the sample was exhibited by S₉, where the proportion of LPG was 30 wt%, the sintering temperature was 650 °C, and the flexural strength and compressive strength were 19.55 and 42.25 MPa, respectively. The average porosity, average apparent density, and average water absorption of the sample were 24.52%, 1.66 g/cm³ and 14.79%, respectively.

The flexural strength and compressive strength of the sample are strongly affected by the sintering temperature, followed by the proportion of LPG. The regression equation has a good fitting effect, that is, the strength is proportional to the sintering temperature and the mass ratio of LPG.

The bonding process of particles with LPG was effectively confirmed by SEM analysis. The chemical composition, XRD, and optical microscopy analysis showed that the primary phase groups were quartz, feldspar, and muscovite. Combining the thermal analysis and SEM, it was further proved that the link of the particles at 550 °C was due to the softing and crystallization of LPG powder. The results of FT-IR shows that P—O bond and Si—O bond were formed when the sintering temperature reaches 600 °C, which may further improve the mechanical properties.

However, the immature preparation process, LPG powder cost and other problems are the reasons to limit its large-scale application. The authors will continue to research efficient utilization of tailings in order to improve economic, environmental and social outcomes.

Figure 10 Simulation of low-temperature sintering process of ceramic materials

Contributors

XU Long-hua provided the concept, conducted the literature reviews and edited the draft of manuscript. YANG Jie wrote the first draft of the manuscript. WU Hou-qin and WANG Zhou-jie analyzed the experimental data. SHU Kai-qian and XU Yan-bo advised some experimental methods. LUO Li-ping and TANG Zhen designed the experimental program.

Conflict of interest

YANG Jie, XU Long-hua, WU Hou-qin, WANG Zhou-jie, SHU Kai-qian, XU Yan-bo, LUO Li-ping and TANG Zhen declare that they have no conflict of interest.

References

[1] ZAMAN A U. A comprehensive review of the development of zero waste management: Lessons learned and guidelines [J]. Journal of Cleaner Production, 2015, 91: 12-25. DOI: 10.1016/j.jclepro.2014.12.013.

[2] BURLAKOVS J, JANI Y, KRIIPSALU M, VINCEVICA- GAILE Z, KACZALA F, CELMA G, OZOLA R, ROZINA L, RUDOVICA V, HOGLAND M, VIKSNA A, PEHME K M, HOGLAND W, KLAVINS M. On the way to “zero waste” management: Recovery potential of elements, including rare earth elements, from fine fraction of waste [J]. Journal of Cleaner Production, 2018, 186: 81-90. DOI: 10.1016/ j.jclepro.2018.03.102.

[3] ZAMAN A U. Identification of key assessment indicators of the zero waste management systems [J]. Ecological Indicators, 2014, 36: 682-693. DOI: https://doi.org/ 10.1016/j.ecolind.2013.09.024.

[4] HENCKENS M, van IERLAND E C, DRIESSEN P, WORRELL E. Mineral resources: Geological scarcity, market price trends, and future generations [J]. Resources Policy, 2016, 49: 102-111. DOI: 10.1016/j.resourpol.2016. 04.012.

[5] WEST J. Extractable global resources and the future availability of metal stocks: “Known Unknowns” for the foreseeable future [J]. Resources Policy, 2020, 65: 101574. DOI: 10.1016/j.resourpol.2019.101574.

[6] ZHANG L, XU Z. A critical review of material flow, recycling technologies, challenges and future strategy for scattered metals from minerals to wastes [J]. Journal of Cleaner Production, 2018, 202: 1001-1025. DOI: 10.1016/j.jclepro.2018.08.073.

[7] GUO Y G, HUANG P, ZHANG W G, YUAN X W, FAN F X, WANG H L, LIU J S, WANG Z H. Leaching of heavy metals from Dexing copper mine tailings pond [J]. Transactions of Nonferrous Metals Society of China, 2013, 23(10): 3068-3075. DOI: 10.1016/S1003-6326(13)62 835-6.

[8] LI X F, YE Y Z, XUE S G, JIANG J, WU C, KONG X F, HARTLEY W, LI Y W. Leaching optimization and dissolution behavior of alkaline anions in bauxite residue [J]. Transactions of Nonferrous Metals Society of China, 2018, 28(6): 1248-1255. DOI: 10.1016/S1003-6326 (18)64763-6.

[9] ZHU X, HAN Y, SUN Y, LI Y, WANG H. Siderite as a novel reductant for clean utilization of refractory iron ore [J]. Journal of Cleaner Production, 2020, 245: 118704. DOI: 10.1016/j.jclepro.2019.118704.

[10] AALDERING L J, SONG C H. Tracing the technological development trajectory in post-lithium-ion battery technologies: A patent-based approach [J]. Journal of Cleaner Production, 2019, 241: 118343. DOI: 10.1016/j.jclepro.2019. 118343.

[11] SCOTT S R, SHAFER M M. A new method for measurement of lithium isotopes with nuclear applications [J]. Progress in Nuclear Energy, 2020, 120: 103234. DOI: 10.1016/j.pnucene.2019.103234.

[12] SHAO L, JIN S. Resilience assessment of the lithium supply chain in China under impact of new energy vehicles and supply interruption [J]. Journal of Cleaner Production, 2020, 252: 119624. DOI: 10.1016/j.jclepro.2019.119624.

[13] ZHANG Y C, BRIAT O, DELETAGE J Y, MARTIN C, CHADOURNE N, VINASSA J M. Efficient state of health estimation of Li-ion battery under several ageing types for aeronautic applications [J]. Microelectronics Reliability, 2018, 88-90: 1231-1235. DOI: 10.1016/j.microrel.2018. 07.038.

[14] TIAN J, XU L, WU H, FANG S, DENG W, PENG T, SUN W, HU Y. A novel approach for flotation recovery of spodumene, mica and feldspar from a lithium pegmatite ore [J]. Journal of Cleaner Production, 2018, 174: 625-633. DOI: 10.1016/j.jclepro.2017.10.331.

[15] WU H, TIAN J, XU L, FANG S, ZHANG Z, CHI R. Flotation and adsorption of a new mixed anionic/cationic collector in the spodumene-feldspar system [J]. Minerals Engineering, 2018, 127: 42-47. DOI: 10.1016/j.mineng. 2018.07.024.

[16] SHU Kai-qian, XU Long-hua, WU Hou-qin, TANG Zhen, LUO Li-ping, YANG Jie, XU Yan-bo, FENG Bo. Selective flotation separation of spodumene from feldspar using sodium alginate as an organic depressant [J]. Separation and Purification Technology, 2020, 248: 117122. DOI: 10.1016/ j.seppur.2020.117122.

[17] ZHU G L, WANG Y H, WANG X M, MILLER J D, LU D F, ZHENG X Y, ZHAO Y H, ZHENG H T. Effects of grinding environment and lattice impurities on spodumene flotation [J]. Transactions of Nonferrous Metals Society of China, 2019, 29(7): 1527-1537. DOI: 10.1016/S1003-6326(19) 65060-0.

[18] GARRETT D E. Lithium [M]// Handbook of Lithium and Natural Calcium Chloride. Amsterdam: Elsevier, 2004. DOI: 10.1016/B978-0-12-276152-2. X5035-X.

[19] MANETA V, BAKER DR. The potential of lithium in alkali feldspars, quartz, and muscovite as a geochemical indicator in the exploration for lithium-rich granitic pegmatites: A case study from the spodumene-rich Moblan pegmatite, Quebec, Canada [J]. Journal of Geochemical Exploration, 2019, 205: 106336. DOI: 10.1016/j.gexplo.2019.106336.

[20] CHOUBEY P K, KIM M S, SRIVASTAVA R R, LEE J C, LEE J Y. Advance review on the exploitation of the prominent energy-storage element: Lithium. Part I: From mineral and brine resources [J]. Minerals Engineering, 2016, 89: 119-137. DOI: 10.1016/j.mineng.2016.01.010.

[21] BURRITT R L, CHRIST K L. Water risk in mining: Analysis of the Samarco dam failure [J]. Journal of Cleaner Production, 2018, 178: 196-205. DOI: 10.1016/j.jclepro. 2018.01.042.

[22] PARK I, TABELIN C B, JEON S, LI X, SENO K, ITO M, HIROYOSHI N. A review of recent strategies for acid mine drainage prevention and mine tailings recycling [J]. Chemosphere, 2019, 219: 588-606. DOI: 10.1016/ j.chemosphere.2018.11.053.

[23] QI C, FOURIE A. Cemented paste backfill for mineral tailings management: Review and future perspectives [J]. Minerals Engineering, 2019, 144: 106025. DOI: 10.1016/ j.mineng.2019.106025.

[24] MENG X, KHOSO S A, KANG J, WU J, GAO J, LIN S, WU M, SUN W, HU Y. A novel scheme for flotation tailings pulp settlement and chemical oxygen demand reduction with polyferric sulfate [J]. Journal of Cleaner Production, 2019, 241: 118371. DOI: 10.1016/j.jclepro.2019.118371.

[25] WANG C, HARBOTTLE D, LIU Q, XU Z. Current state of fine mineral tailings treatment: A critical review on theory and practice [J]. Minerals Engineering, 2014, 58: 113-131. DOI: 10.1016/j.mineng.2014.01.018.

[26] RITCEY G M. Tailings management in gold plants [J]. Hydrometallurgy, 2005, 78(1): 3-20. DOI: 10.1016/ j.hydromet.2005.01.001.

[27] YIN Z, SUN W, HU Y, ZHANG C, GUAN Q, WU K. Evaluation of the possibility of copper recovery from tailings by flotation through bench-scale, commissioning, and industrial tests [J]. Journal of Cleaner Production, 2018, 171: 1039-1048.

[28] LIU W, WU T, LI Z, HAO X, LU A. Preparation and characterization of ceramic substrate from tungsten mine tailings [J]. Construction and Building Materials, 2015, 77: 139-144. DOI: 10.1016/j.conbuildmat.2014.12.094.

[29] LEMOUGNA P N, YLINIEMI J, ISMAILOV A, LEVANEN E, TANSKANEN P, KINNUNEN P, RONING J, ILLIKAINEN M. Recycling lithium mine tailings in the production of low temperature (700-900°C) ceramics: Effect of ladle slag and sodium compounds on the processing and final properties [J]. Construction and Building Materials, 2019, 221: 332-344. DOI: 10.1016/j.conbuildmat.2019. 06.078.

[30] LEMOUGNA P N, YLINIEMI J, ISMAILOV A, LEVANEN E, TANSKANEN P, KINNUNEN P, RONING J, ILLIKAINEN M. Spodumene tailings for porcelain and structural materials: Effect of temperature (1050–1200°C) on the sintering and properties [J]. Minerals Engineering, 2019, 141: 105843. DOI: 10.1016/j.mineng.2019.105843.

[31] AYODE OTITOJU T, UGOCHUKWU OKOYE P, CHEN G, LI Y, ONYEKA OKOYE M, LI S. Advanced ceramic components: Materials, fabrication, and applications [J]. Journal of Industrial and Engineering Chemistry, 2020, 85: 34-65. DOI: 10.1016/j.jiec.2020.02.002.

[32] DENG L, FENG B, ZHANG Y. An optimization method for multi-objective and multi-factor designing of a ceramic slurry: Combining orthogonal experimental design with artificial neural networks [J]. Ceramics International, 2018, 44(13): 15918-15923. DOI: 10.1016/j.ceramint.2018.06.010.

[33] ZHANG S, CAO K, WANG C, WANG X, DENG G, WEI P. Influence of the porosity and pore size on the compressive and splitting strengths of cellular concrete with millimeter-size pores [J]. Construction and Building Materials, 2020, 235: 117508. DOI: 10.1016/ j.conbuildmat.2019.117508.

[34] FLADR J, BILY P. Specimen size effect on compressive and flexural strength of high-strength fibre-reinforced concrete containing coarse aggregate [J]. Composites Part B: Engineering, 2018, 138: 77-86. DOI: 10.1016/ j.compositesb.2017.11.032.

[35] XI C, ZHOU J, ZHENG F, GAO J M, HU P, LI Y, ZHEN Q, BASHIR S, LIU J L. Conversion of extracted titanium tailing and waste glass to value-added porous glass ceramic with improved performances [J]. Journal of Environmental Management, 2020, 261: 110197. DOI: 10.1016/ j.jenvman.2020.110197.

[36] WILLE G, LAHONDERE D, SCHMIDT U, DURON J, BOURRAT X. Coupling SEM-EDS and confocal Raman- in-SEM imaging: A new method for identification and 3D morphology of asbestos-like fibers in a mineral matrix [J]. Journal of Hazardous Materials, 2019, 374: 447-458. DOI: 10.1016/j.jhazmat.2019.04.012.

[37] YANG Z, WANG R, CHE X, YIN R, XIE L, HU H. Formation of columbite and microlite after alteration of Nb- and Ta-bearing biotite from the Lizaizhai pegmatite (Guangning ore district, Guangdong, South China): Identification of a new potential Nb–Ta mineralization type [J]. Journal of Asian Earth Sciences, 2020, 190: 104154. DOI: 10.1016/j.jseaes.2019.104154.

[38] QI L, WENG X, WEI B, YUAN L, HUANG G, DU X, WU X, LIU H. Effects of low-melting glass powder on the thermal stabilities of low infrared emissivity Al/polysiloxane coatings [J]. Progress in Organic Coatings, 2020, 142: 105579. DOI: j.porgcoat.2020.105579.

[39] ZHANG B, CHEN Q, SONG L, LI H, HOU F, ZHANG J. Fabrication and properties of novel low-melting glasses in the ternary system ZnO–Sb₂O₃–P₂O₅ [J]. Journal of Non-Crystalline Solids, 2008, 354(18): 1948-1954. DOI: 10.1016/j.jnoncrysol.2007.11.007.

[40] JIN J, GAO Y, WU Y, LI R, LIU R, WEI H, QIAN G, ZHENG J. Performance evaluation of surface-organic grafting on the palygorskite nanofiber for the modification of asphalt [J]. Construction and Building Materials, 2021， 268: 121072. DOI: 10.1016/j.conbuildmat.2020.121072.

[41] JIN J, GAO Y, WU Y, LIU S, LIU R, WEI H, QIAN G, ZHENG J. Rheological and adhesion properties of nano-organic palygorskite and linear SBS on the composite modified asphalt [J]. Powder Technology, 2021, 377: 212-221. DOI: 10.1016/j.powtec.2020.08. 080.

[42] JIN J, LIU S, GAO Y, LIU R, HUANG W, WANG L, XIAO T, LIN F, XU L, ZHENG J. Fabrication of cooling asphalt pavement by novel material and its thermodynamics model [J]. Construction and Building Materials, 2021, 272: 121930. DOI: 10.1016/j.conbuildmat. 2020.121930.

[43] MENAA B, MIZUNO M, TAKAHASHI M, TOKUDA Y, YOKO T. Polycarboxylic acids as network modifiers for water durability improvement of inorganic–organic hybrid tin-silico-phosphate low-melting glasses [J]. Journal of Solid State Chemistry, 2006, 179(2): 492-499. DOI: 10.1016/ j.jssc.2005.11.001.

(Edited by ZHENG Yu-tong)

中文导读

锂辉石尾矿制备陶瓷材料的烧结行为及其力学性能

摘要：本文以锂辉石浮选尾矿为主要原料，低熔点玻璃粉(LPG)为高温粘结材料，通过湿法注模成型，常压烧结制备陶瓷材料。通过正交试验研究了烧结温度和LPG质量分数对陶瓷材料抗折强度和抗压强度的影响。结果表明，当LPG质量比不低于20 wt%，烧结温度不低于550 °C时，尾矿颗粒间相互粘结紧密，制备出了力学性能优异的陶瓷材料(最大抗折强度=19.55 MPa，最大抗压强度= 42.25 MPa，平均显气孔率24.52%，体积密度1.66 g/cm³，吸水率14.79%)。采用XRF、XRD、光学显微镜、SEM、TGA-DSC和FT-IR等对陶瓷材料低温烧结机理进行了测试分析。结果表明，高温下液相的生成使得颗粒间相互粘结可能为陶瓷材料力学性能提高的主要原因。

关键词：锂辉石尾矿；低温烧结；陶瓷材料；力学性能

Foundation item: Projects(51674207, 51922091) supported by the National Natural Science Foundation of China; Project(2018QNRC001) supported by the Young Elite Scientists Sponsorship Program by CAST, China; Projects(2019YFS0453, 2018JY0148) supported by the Sichuan Science and Technology Program, China

Received date: 2020-07-13; Accepted date: 2021-01-05

Corresponding author: XU Long-hua, PhD, Professor; Tel: +86-18281552389; E-mail: neuxulonghua@163.com, xulonghua@swust.edu.cn; ORCID: https://orcid.org/0000-0003-3715-5148