Effects of roasting and NH4Cl catalysis on the direct electrotransformation products from an AlCl3 solution

Contemporary processes for extracting alumina from fly ash involve various disadvantages, including high-energy consumption, heavy slag discharge, and severe equipment corrosion, so a new method for extracting alumina via electrotransformation was developed in this work. This novel method reduced slag discharge, enabled recycling, and produced no acid or alkali waste. The theoretical solution pH associated with Al(OH)3 precipitation and the actual solution pH during electrotransformation were also compared herein. Differential thermogravimetric analysis was performed for the direct electrotransformation products generated with and without NH4Cl as a catalyst; the activation energy and reaction order of these reactions revealed large fluctuations. The effects of the roasting temperature on the phase, morphology, D(50), and D(90) of the roasting products were also studied. The results indicated that the solution reached the theoretical pH (4.7) of complete Al(OH)3 precipitation in a short time when NH4Cl was implemented as a catalyst. Additionally, the D(50) and D(90) of the roasting products were nearly ten times higher when NH4Cl was used, relative to when it was not, reaching 227.1 and 519.6 μm, respectively. The activation energies of each reaction in the electrotransformation and roasting process were also reduced in the presence of the NH4Cl catalyst. Between 100 °C–250 °C, the physical adsorption of water occurred, and some of the structural water of the electrotransformation products generated with NH4Cl were removed in an endothermic reaction (reaction order, n1 = 0.43; activation energy, E1 = 2.13J mol−1). At temperatures between 250 °C–500 °C, the nitrogen-containing substances were removed in another endothermic reaction (reaction order, n2 = 0.4697; activation energy, E2 = 5.316J mol−1). Increasing the roasting temperature was beneficial for improving the crystal form of the roasting products, and the reduced electrotransformation products appeared more compact after roasting.


Introduction
In recent years, concerns regarding environmental problems have continued to grow around the world [1]. Coal fly ash is an industrial by-product of coal combustion, which is produced in high-temperature coal-fired thermal power plants, and it represents one of the most abundant incombustible solid wastes [2][3][4][5][6][7]. The global coal fly ash emissions reach about 780 million tons annually, and India alone contributes about 23% of that total [8][9][10]. Stockpiling of coal fly ash is irreversible, and storing such a huge quantity of coal fly ash would take up enormous land space. Meanwhile, the demand for electric power continues to increase all over the world, which raises much more serious problems in terms of fly ash disposal and environmental pollution [11,12]. However, it is also considered to be the world's fifth largest raw material resource [13].
Coal fly ash from parts of China's Inner Mongolia and Shanxi provinces contains significant amounts of alumina (up to 40%-50% [14]), which is a promising secondary mineral resource (other than bauxite), and is therefore amenable to metallurgical and chemical recovery processes. As bauxite resources decrease and the demand for alumina increases, fly ash with high alumina content can be used as a substitute for bauxite in alumina production [15,16]. Considerable research efforts are currently aimed at determining potential applications for these solid waste materials; such studies have vital environmental significance in addition to their inherent scientific research and industrial value [17]. The rational and efficient utilization of fly ash not only alleviates environmental pollution, but also plays an important role in scientific advancements and sustainable development [18]. Currently, the extraction of aluminum and other valuable elements from fly ash has become a research hotspot worldwide [19][20][21][22][23].
The traditional Bayer process is suitable for recovering aluminum from bauxite resources with low silica content. Therefore, this process is not suitable for fly ash with high silica content, although alumina recovery from fly ash has been attempted using various approaches, including acid methods [24][25][26][27][28][29] and alkali methods [30,31]. However, there are many shortcomings or deficiencies in these processes. Acid processes are the most commonly applied strategies for extracting alumina, and are accordingly the preferred methods for extracting alumina from fly ash. However, the acid-based process inevitably causes very serious equipment corrosion, so it requires highly acid-resistant and air-tight equipment, which increases investment costs; this process also produces acid smog, among other environmental problems [32]. The main steps of the alkaline process involves intering, dissolution, and separation, and its disadvantages mainly include high-energy consumption, large quantities of slag discharge, and lower alumina extraction rates [33].
Considering the aforementioned aspects and available techniques, developing a new method for extracting alumina from fly ash is crucial to overcome the deficiencies in the conventional fly ash processing treatments and to ease the current environmental burden [33]. In this paper, a novel approach for alumina preparation via electrotransformation of fly ash using an AlCl 3 solution is proposed ( figure 1). This method has various advantages, including slag reduction, recycling, and no acid or alkali waste. The theoretical solution pH value related to Al(OH) 3 precipitation and the actual solution pH value during electrotransformation were compared and analyzed. Additionally, the differential scanning calorimetry thermogravimetric analysis(TG-DSC) of direct electrotransformation products and electrotransformation products catalyzed by NH 4 Cl was studied. The endothermic and exothermic peaks with large fluctuations in their differential scanning calorimetry (DSC) curves were fitted, and the activation energy and reaction order of the corresponding chemical reactions were calculated. The effects of roasting temperature on the phase, morphology, D(50)(the average particle size), and D(90)(the cumulative particle distribution is 90% particle size) of the roasting products were also investigated.

Materials and equipment
Currently, the application of electrolysis for alumina extraction from fly ash is still a new research direction, so such experiments are in the basic research stage. Therefore, fly ash was not used as the raw material in the experiments described in this article; instead, the main reagent used in the experiments was pure AlCl 3 ·6H 2 O(Sinopharm Group Co. Ltd, Shanghai, China; analytical grade, used as received without further processing). The main equipment used for the experiments were a plexiglass electrolyzer (Dongling Plexiglass Factory, Shenyang, China), and a DC power supply (eTM-L305SPL, Dong guan Tong men Electronic Technology Co. Ltd, China), etc. The main testing instrumentation were an x-ray diffractometer (Bruker D8, Bruker Corp., Billerica, MA, USA), a simultaneous thermal analyzer (STA449F3, NETZSCH company, Germany), a scanning electron microscope (SU-8010, Hitachi High-Technologies Corp., Tokyo, Japan), and a laser particle size tester (Dandong Baxter Instrument Co., Ltd, Liaoning, China) etc.

Experimental process
The specific experimental conditions for the electrotransformation process are as follows: current intensity=2A; electrotransformation time=3 h; electrode spacing=2 cm; AlCl 3 solution concentration in the anodic chamber and cathodic chamber=100 and 0 g l −1 , respectively; NH 4 Cl solution concentration in the anodic chamber and cathodic chamber=0 and 60 g l −1 , respectively.
The main steps in this experimental process included weighing reagents, preparing solutions, electrolysis, filtering products, drying, and roasting, among other operations. The specific procedure was as follows: first, the cation exchange membrane (CJMC-3, Hefei Ke jia Polymer Material Technology Co., Ltd, China) was installed, and the power was switched on and the current value set to 2A (according to the experimental scheme for conducting electrotransformation experiments). Simultaneously, the solution pH in the cathodic chamber was recorded (and monitored during the experiment). After the experiment was completed, the power was turned off, and the solution in the cathodic chamber was poured out for filtration. Next, the products that precipitated after filtration were isolated and dried, and the dried products were analyzed and roasted to ultimately obtain the desired alumina products. The schematic diagram illustrating the electrotransformation using AlCl 3 solution is shown in figure 2.

2.3.
Determining the pH value of the Al(OH) 3 solution at the beginning and after completing precipitation Al(OH) 3 is an insoluble electrolyte, and its dissolution and formation are reversible. When the dissolution rate and formation rate of Al(OH) 3 are equal, the concentration of each ion in solution does not change, and the precipitation dissolution equilibrium between solid Al(OH) 3 and hydrated ions in the solution is established. The specific equilibrium is shown in equation (1): Al OH Al aq 3OH aq 1 3 3 The solubility product(K sp ) expression for Al(OH) 3 is shown in equation (2), is the concentration of OH -, C is the standard concentration (generally defined as 1molL −1 , and is used to cancel the units of the K sp calculation), and K sp (Al(OH) 3 ) is the solubility product constant of Al(OH) 3 . The formula for calculating the solution pH is presented in equation (3): According to literature reports, the K sp of amorphous Al(OH) 3 at room temperature, is 1.3×10 -33 . Based on calculations, when the mass concentration of AlCl 3 solution is 100 g l −1 (equivalent to a molar concentration of 0.75mol l −1 ), the solution pH when Al 3+ begins to precipitate is 3.08. When the Al 3+ is completely precipitated (i.e., when the Al 3+ concentration is lower than 1×10 −5 mol l −1 ), the solution pH value is 4.7.

Direct electrotransformation of the AlCl 3 solution
The actual solution pH value during direct electrotransformation was compared with the theoretical pH value of Al(OH) 3 precipitation. The TG-DSC analysis of direct electrotransformation products, and the DSC curve fitting of the absorption and exothermic peaks were investigated. Additionally, the effects of roasting temperature on the phase, morphology, D(50), and D(90) of the roasting products were studied. Figure 3 shows the actual solution pH value during direct electrotransformation and the theoretical pH of Al(OH) 3 precipitation. As the direct electrotransformation proceeded, the actual solution pH value increased gradually as H + ions in the cathodic chamber were continuously reduced to H 2 . Simultaneously, Al(OH) 3 precipitation was observed. In the anodic chamber, Clions were continuously oxidized to Cl 2 . After approx.150 min, the actual solution pH value exceeded 3.08, where Al 3+ began to precipitate; however, in reality, Al(OH) 3 precipitation began before that time. This was mainly because the actual local solution pH value near the electrode had reached the theoretical Ph value of Al(OH) 3 precipitation before 150 min. When the electrotransformation experiment finished, the actual solution pH value was 3.13, which was lower than the theoretical solution pH value(4.7) when Al(OH) 3 was completely precipitated. Figure 4 presents the differential thermogravimetric analysis(TG-DSC) of electrotransformation products obtained via direct electrotransformation of an AlCl 3 solution. The weightlessness temperature of the first stage was between 100°C-350°C. Within this temperature range, the physical adsorption water occurred and some of the structural water for the direct electrotransformation products was removed. The discharge of structural water was an endothermic reaction, so a more intense endothermic peak appeared in this region. The weightlessness temperature of the second stage was between 350°C-800°C. Within this temperature range, a series of crystalline transformations took place, resulting in small endothermic peaks. The weightlessness temperature of the third stage was between 800°C-860°C. Within this temperature range, the alumina underwent a crystal line transformation, thus generating an exothermic peak. As the temperature gradually increased from 1000 to 1200°C, there was no detectable weight loss phenomenon (i.e., exothermic peak), indicating that alumina had good thermal stability in this temperature range.  figure 5(a), a linear fit(slope =-0.15139; intercept=0.73537) could be obtained. This enabled calculation of the reaction order (n1=0.74) and the activation energy (E1=2.90kJ mol −1 ) of the endothermic reaction. According to the fitting data shown in figure 5(b), a linear fit (slope=−1.0033; intercept=0.0744) was obtained, and the reaction order (n2=0.07) and activation energy (E2=19.21 kJ mol −1 ) of the exothermic reaction were calculated. In this linear fitting process, the degree of fitting was not fully clear because of some data distortion, which led to some potential calculation errors. Figure 6 presents the phases of roasting products generated by roasting the direct electrotransformation products at different temperatures. As the roasting temperature increased, the diffraction peak strength of the roasting products also increased. When the roasting temperature was 500°C, the roasting products were all amorphous. In contrast, when the roasting temperature was between 700°C-900°C, the roasting products were 'steamed bread' alumina with a cubic crystal system and a P space group. Then, when the roasting temperature was between 1100°C-1300°C, the roasting products were alumina with a rhombohedral crystal system and an R-3c space group. Therefore, it was concluded that the roasting temperature had a significant impact on the crystallization state of roasting products; the higher the roasting temperature, the better the crystallization state. Figure 7 shows the scanning electron microscopy(SEM) images (i.e., the morphology) of roasting products obtained after roasting the direct electrotransformation products at different temperatures. When the roasting temperature was 1300°C, the roasting products all adopted honeycomb-like appearances, and as the roasting temperatures decreased, their appearance became denser. When the roasting temperature was 300°C, all of the  roasting products appeared flakey. These results indicated that the roasting temperature could influence the morphology of roasting products. Figure 8 shows the D(50) and D(90) of roasting products prepared by roasting direct electrotransformation products at various temperatures. The D(50) and D(90) of the roasting products obtained in these experimental conditions were between 14-17 and 39-47 μm, respectively, and importantly, both were within one order magnitude. This result indicated that the roasting temperature did not have an appreciable effect the on D(50) or D(90) of the roasting products obtained in this study.

Electrotransformation of the AlCl 3 solution catalyzed by NH 4 Cl
The actual solution pH during the electrotransformation of the AlCl 3 solution catalyzed by NH 4 Cl was compared with the theoretical pH value of Al(OH) 3 precipitation. Additionally, the TG-DSC analysis of the electrotransformation products obtained following the electrotransformation of the AlCl 3 solution catalyzed by NH 4 Cl was performed, and the DSC curve fitting of the absorption and exothermic peaks for these products were investigated. Simultaneously, the effects of the roasting temperature on the phase, morphology, D(50), and D(90) of the roasting products were also studied.   Figure 9 shows the actual solution pH values during the electrotransformation of the AlCl 3 solution catalyzed by NH 4 Cl and the theoretical pH of Al(OH) 3 precipitation. Within a short time (approx.5 min), the actual solution pH value reached the theoretical pH(4.7) where Al(OH) 3 was completely precipitated. The solution pH value increased as the electrotransformation process proceeded over time, and the local solution pH reached 9.5 by the end of the experiment. This was mainly due to the continuousoxidation of Clto Cl 2 in the anodic chamber. Accordingly, the H + in the cathodic chamber was continuously reduced to H 2 .To achieve a charge balance of cations and anions in the electrolyte, Al 3+ ions in the anodic chamber were constantly transferred to the cathodic chamber during the electrotransformation process. When the actual solution pH reached the Al 3+ precipitate range, Al 3+ ions combined with OHions to form an Al(OH) 3 precipitate in the cathodic chamber. This occurred simultaneously with NH 3 formation, leading to the high pH value observed in the cathodic chamber. Figure 10 presents the TG-DSC analysis of the electrotransformation products obtained using NH 4 Cl catalytic conditions. The weightlessness temperature of the first stage was between 100°C-250°C. Within this temperature range, the physical adsorption of water occurred and some of the structural water for the  electrotransformation products prepared via NH 4 Cl catalysis was removed. The discharge of structural water was associated with an endothermic reaction, so a more intense endothermic peak appeared in this region. The weightlessness temperature of the second stage was between 250°C-500°C. Within this temperature range, nitrogen-containing substances were removed, and this nitrogen removal reaction was also an endothermic reaction, thus, a very intense endothermic peak was observed. As the temperature gradually increased from 500 to 1000°C, there was no clear weight loss phenomenon (i.e., neither endothermic nor exothermic peaks appeared), indicating that alumina had good thermal stability in this temperature range. Figures 11(a) and (b) respectively show the DSC analysis and curve fitting of the two endothermic peaks, which shoed large fluctuations in the case of electrotransformation products prepared under NH 4 Cl catalytic conditions. According to the fitting data shown in figure 11(a), a linear fit (slope=-0.11115; intercept=0.42512) was obtained. The reaction order (n1=0.43) and the activation energy (E1=2.13 kJ mol −1 ) of the corresponding endothermic reaction were calculated. Based on the fitting data shown in figure 11(b), a linear fit (slope=-0.27766; intercept=0.4697) was obtained. The reaction order (n2=0.4697) and the activation energy (E2=5.316 kJ mol −1 ) of this endothermic reaction were also calculated. As expected, the activation energy was lower in these cases than in experiments without the NH 4 Cl catalyst, thus confirming that the NH 4 Cl catalyst reduced the activation energy of each reaction in the roasting process of electrotransformation. Figure 12 shows the phases of the products obtained via electrotransformation catalyzed by NH 4 Cl and roasting at different roasting temperatures. As the roasting temperature increased, the diffraction peak intensity of the roasting products increased. At roasting temperatures of 500, 700, and 900°C, the roasting products were all 'steamed bread' alumina with a cubic crystal system, an Fd-3m space group, and 227 space group numbers. When the roasting temperature was 1100°C-1300°C, the roasting products were alumina with a rhomboid  crystal system, an R-3c space group, and 167 space group numbers, and the number of diffraction peaks increased substantially. Therefore, the roasting temperature had a significant impact on the crystallization state of the roasting products, such that the crystallization state was better when the roasting temperature was higher than 1100°C. Figure 13 shows the morphology (SEM images) of the roasting products following electrotransformation catalyzed by NH 4 Cl and roasting at various temperatures. Han Xiuxiu et al (2019a; 2019b) found that additives (e.g., HCl, Na 2 CO 3 , NaHCO 3 ), the current intensity, the electrotransformation duration, and the initial electrotransformation temperature, among other parameters, all had no clear effect on the morphology of roasting products, and all products adopted an irregular sheet structure. However, from figure 13, it is clear that in this study, both NH 4 Cl and the roasting temperature influenced the morphology of the roasting products. When the roasting temperature was 1300°C, the roasting products had regular, angular, and smooth block structural morphologies. When the roasting temperature was less than 1300°C, the apparent morphology of the roasting products were all massive structures with rough surfaces, and the microstructure did not change significantly with the roasting temperature decreased further. Therefore, when the roasting temperature was higher than 1300°C, the morphology of the roasting products could effectively be controlled.    90) values of the roasting products obtained following electrotransformation catalyzed by NH 4 Cl and roasting at various temperatures. Han Xiuxiu et al determined that the particle size of the roasting products did not change significantly in the presence of certain additives (e.g., HCl, Na 2 CO 3 , NaHCO 3 ), and they were all small. However, from figure 14, it is clear that D(50) and D(90) of the roasting products obtained in this study under the described conditions were all larger, with values ranging from 101-227.1 μm and 371.5-519.6 μm, respectively. Additionally, the D(50) and D(90) of the roasting products first increased and then decreased as the roasting temperature increased. When the roasting temperature was 900°C, the D(50) and D(90) of the roasting products both reached their maxima(227.1 and 519.6 μm, respectively). These results indicated that it was possible to effectively control the D(50) and D(90) of the roasting products prepared under the applied conditions by controlling the roasting temperature.

Conclusions
The study presented in this paper led to the following key results: (1) The solution reached the theoretical pH value (4.7) associated with complete Al(OH) 3 precipitation in a short time when using a NH 4 Cl catalyst.
(2) For the non-catalyzed reaction, between 100°C-350°C, the physical adsorption of water occurred, and some of the structural water of the direct electrotransformation products were removed in an endothermic reaction with a reaction order of n1=0.74 and an activation energy of E1=2.90J mol −1 . At temperatures between 800°C-860°C, crystal transformation of alumina occurred in an exothermic reaction with a reaction order of n2=0.07 and an activation energy of E2=19.21 kJ mol −1 . The law of fitting was not entirely clear because of some data distortion.
(3) Between 100°C-250°C, the physical adsorption of water occurred, and some of the structural water of the electrotransformation products obtained via NH 4 Cl catalysis were removed in an endothermic reaction with a reaction order of n1=0.43 and an activation energy of E1=2.13J mol −1 . At temperatures between 250°C-500°C, the nitrogen-containing substances were removed in another endothermic reaction with a reaction order of n2=0.4697 and an activation energy of E2=5.316J mol −1 . These results confirmed that the NH 4 Cl catalyst reduced the activation energy of each reaction in the electrotransformation products' roasting process.
(4) The D(50) and D(90) of the roasting products obtained using the catalyst were nearly ten times higher than those generated without NH 4 Cl, reaching 227.1 and 519.6 μm, respectively.
(5) Increasing the roasting temperature helped improve the crystal form of the roasting products, and the reduced products appeared relatively compact after roasting the direct electrotransformation products; however, the roasting temperature had no appreciable effect on the D(50) and D(90) of the roasting products.