China Nonferrous Metallurgy

Abstract:Pyrometallurgy of non-ferrous metals is the core technology for the industrial extraction of base metals such as copper, lead, zinc and nickel in China. However, its high-temperature, multiphase, and strongly coupled reaction characteristics lead to long-standing issues such as low energy efficiency, high pollutant emissions, and insufficient process control accuracy. As artificial intelligence (AI) technologies continue to penetrate the metallurgical industry, they demonstrate systematic value in improving efficiency, reducing costs, and enhancing safety and environmental performance. This review, from a full-process perspective of “data acquisition-parameter prediction-process optimization-equipment maintenance,” comprehensively explores the implementation of key technologies, including multimodal sensor network construction, critical parameter prediction in high-temperature multiphase systems, multi-objective dynamic optimization control, and intelligent fault diagnosis. Studies show that AI-based methods significantly enhance process energy efficiency and environmental performance by improving the accuracy of online parameter monitoring, strengthening multivariable coordinated control, and enabling full-lifecycle management of critical equipment. Nonetheless, challenges remain, such as fluctuating data quality under extreme thermal conditions, limited model generalization in complex scenarios, and insufficient understanding of multi-scale spatiotemporal coupling mechanisms. Future research should focus on developing physics-informed modeling approaches, cross-scale dynamic optimization algorithms, and industrial-grade intelligent decision systems . Through the deep integration of metallurgical thermodynamics, process systems engineering, and information science, a smart pyrometallurgical technology framework aligned with green and low-carbon goals can be established.

Abstract:Aiming at the problem that the production state of rare earth electrolytic cell is difficult to perceive, which leads to the inability to accurately control it and affects stable production, this study proposes a hybrid neural network diagnosis model based on signal feature extraction to identify the operation state of electrolytic cell. Firstly, the normalized cell voltage signal is obtained by calculating the cell voltage and cell current, and its time-frequency analysis is carried out to extract the energy values and energy ratios of each frequency band with the time-domain and frequency-domain features in different cell states, and to construct the feature data set. Secondly, a cell state diagnosis model (CNN-BiLSTM-SA) combining convolutional neural network (CNN), bidirectional long-term and short-term memory network (BiLSTM) and self-attention mechanism is established. The local spatial features are captured by CNN, the global dependency is modeled by BiLSTM, and the feature relevance is enhanced by the self-attention mechanism to realize the state diagnosis of the electrolytic cell. Simultaneously, improvements were made to the hippo algorithm by incorporating optimal point set initialization alongside crossover and mutation operations, aiming to enhance the algorithm's hyperparameter optimization capabilities. Finally, the improved hippo algorithm is used to optimize the hyperparameters of the model to improve the diagnostic accuracy of the model. Industrial validation experiments demonstrate that the proposed method achieves a diagnostic accuracy of 94.14% and a Macro-F1 score of 91.08% in cell condition diagnosis. Compared to the unoptimized model, these results represent improvements of 8.28% and 14.19%, respectively. It has advantages in diagnosing the state of rare earth molten salt electrolysis cell, and can provide a basis for optimal control and stable operation of rare earth molten salt electrolysis process.

Abstract:Accurate prediction of the outlet temperature of nickel matte during nickel smelting is crucial for process optimization, equipment safety assurance, and product quality stability. However, the process involves multi-variable nonlinear coupling and dynamic parameter fluctuations, which restrict the application effect of traditional methods. To address this issue, this paper proposes a nickel matte temperature prediction method integrating Temporal Convolutional Network (TCN) with XGBoost dynamic error compensation. First, a fusion strategy of Pearson correlation coefficient and F-score is adopted to select key features from multi-dimensional industrial parameters such as furnace feed composition, blast volume, and oxygen concentration, reducing redundant interference. Second, the dilated causal convolution of TCN is used to capture long-and short-term temporal dependencies and nonlinear relationships among features, generating initial temperature predictions. Furthermore, XGBoost is introduced and first-order/second-order time difference features are constructed to learn the evolution law of prediction residuals for dynamic error compensation. Relying on XGBoost quantile regression, interval prediction of nickel matte temperature can be realized. Experimental results show that in terms of point prediction performance, the TCN-XGBoost model significantly improves accuracy compared with the baseline TCN model: MAE decreases from 7.6277 to 7.4941, MAPE reduces from 0.5928 to 0.5842, RMSE optimizes from 9.7913 to 9.5732, and R2 increases to 0.4516. It also outperforms comparative methods such as LSTM and AGCRN. In terms of interval prediction performance, the 90% prediction interval of TCN-XGBoost exhibits a balanced advantage of “reliable coverage and compact width”. Compared with TCN-LightGBM and TCN-KDE, it can not only tightly wrap the true values in stable temperature intervals but also reliably accommodate changes in true values in sharply fluctuating sections, avoiding problems of excessively wide intervals or missing coverage. This method possesses high-precision prediction capability for the outlet temperature of nickel matte and can effectively adapt to dynamic fluctuations in industrial production, providing scientific support for real-time monitoring and process regulation of the nickel smelting process.

Abstract:To mitigate the frequent failures of Isasmelt furnace lance during copper smelting process and enhance the identification accuracy of few-shot fault data, this study introduces a novel approach, integrating exponential linear unit (ELU), global average pooling (GAP) convolutional neural network (EGCNN) and rime-ice optimization algorithm (RIME) optimized support vector machine (SVM). Initially, ELU is employed as the activation function for the convolutional neural network (CNN), enhancing robustness against noise and input variations, thereby expediting model convergence. Subsequently, GAP replaces the full connect (FC) layer to strengthen the correlation between process parameters and fault categories, which reduces the number of model parameters and mitigates the risk of overfitting. Ultimately, SVM is implemented as the final classifier in lieu of the traditional Softmax function. RIME is employed to optimize the penalty factor and kernel parameter of the SVM, thereby further enhancing the accuracy of the model. The results indicate that the proposed method achieves an accuracy of 97.08%, precision of 97.08%, recall of 97.10%, F1-score of 97.07% and Kappa coefficient of 0.9611 in identifying Isasmelt furnace lance faults. The proposed method exhibits superior fault identification performance.

Abstract:Emulsion liquid membranes (ELMs) are extensively utilized for the extraction and separation of target substances due to their large contact area, rapid reaction kinetics, and capacity for simultaneous extraction and stripping. However, industrial implementation is hindered by membrane instability resulting from phenomena such as coalescence, swelling, or leakage during separation and enrichment processes. This review systematically examines the critical factors influencing ELM stability, their microscopic mechanisms, and future research directions. Carrier concentration exhibits a distinct threshold: low concentrations promote metal-carrier complex formation to enhance mass transfer, while high concentrations induce a sharp increase in membrane-phase viscosity, triggering droplet coalescence and osmotic swelling (water migration into the membrane diluting the internal phase), thereby reducing both stability and separation efficiency. In surfactant systems, Span 80 forms elastic monomolecular films via robust interfacial adsorption; β-cyclodextrin polymers establish high-mechanical-strength interfacial barriers against coalescence; and blended systems (e.g., Span 80/Tween 80) improve stability through HLB synergy. Surfactant concentrations exceeding the critical micelle concentration (CMC) compromise stability due to increased micellar viscosity and mass transfer resistance. Membrane additives (e.g., iso-octanol, amphiphilic polymer P(LM-AA), polymer HPAM) significantly reinforce interfacial films by adjusting HLB values, providing steric hindrance, and forming viscous networks. Destabilization mechanisms in operational parameters manifest as follows: emulsification beyond critical energy thresholds causes excessively small droplets and membrane thinning; prolonged extraction exacerbates osmotic swelling, leading to internal phase osmotic pressure accumulation and rupture; elevated temperatures reduce membrane-phase viscosity and interfacial film strength, accelerating coalescence. Microscopic analyses reveal stability originates from dynamic interfacial behavior: dilatational rheology shows the elastic modulus directly characterizing deformation resistance (high-modulus films formed by Span 80 saturation, with D2EHPA blending reducing rigidity via enhanced hydrophilicity of phosphate groups); shear rheology demonstrates that high interfacial viscosity suppresses droplet rupture through shear-thinning energy dissipation. Future research must focus on two advances: developing green, efficient stabilizers combining high interfacial strength, low environmental impact, and minimal mass transfer resistance; and fundamentally elucidating microscopic mechanisms through in situmolecular-level interfacial characterization to enable rational design of specialized stabilizers. This review provides critical insights for addressing ELM instability and advancing industrial applications.

Abstract:Against the backdrop of the growing demand for sustainable energy, lithium-ion batteries (LIBs) have become a global focal point for technological innovation and the transition to a green economy, owing to their prominent advantages such as high energy density, long cycle life, and low self-discharge rate. However, traditional LIB electrode materials are increasingly revealing limitations under stringent performance requirements like high-capacity output and high-rate charging/discharging, making it difficult to meet the development needs of future energy storage scenarios. Based on recent research advancements highlighting the irreplaceable potential of vanadium oxides in energy storage-attributed to the multi-valent reversibility of vanadium atoms enabling multi-electron reactions during lithium ion insertion/extraction, thus granting high theoretical specific capacity-this review first systematically analyzes the electrochemical properties of vanadium oxides. The V-O system phase diagram, accurately calculated using the Phase Diagram module of FactSage 8.3, provides theoretical support for material performance modulation. Subsequently, it details mainstream synthesis methods such as hydrothermal, solid-state, and electrospinning, along with modification strategies like compositing with carbon materials, elemental doping, and nanostructure construction, all aimed at enhancing material conductivity, structural stability, and cycling performance. Finally, the advantages and application potential of vanadium oxides are discussed, alongside the key challenges in practical applications, for which solutions and future prospects are proposed. This review not only offers a clearer understanding of the core characteristics of vanadium oxides-such as rich oxidation states, diverse crystal structures, and good electrochemical performance-and the fundamental rationale for their status as key candidates for LIB materials, but also elucidates how the structural features of different vanadium oxides affect their electrochemical properties, systematically summarizes synthesis methods and their regulatory effects on material performance, and precisely identifies the core challenges in practical applications while proposing potential solutions.

Abstract:Aqueous zinc-ion batteries (AZIBs) have emerged as a promising new generation of electrochemical energy storage devices due to their high specific capacity, excellent safety, and economic advantages. However, the ZnSO4 electrolyte system is plagued by issues such as a narrow voltage window, poor ionic conductivity, high water activity leading to zinc anode corrosion and dendrite growth, which significantly hinders the performance improvement and industrial application of AZIBs. Many researchers have attempted to modify the ZnSO4 electrolyte by introducing additives. The introduction of organic additives (such as chitosan, D-xylose, tetrahydrofurfuryl alcohol, etc.) into the electrolyte can effectively enhance the battery s cycle life and ionic conductivity through mechanisms such as forming a dynamic interface protective layer, reconfiguring the solvation structure, and suppressing side reactions. Some modified batteries can achieve thousands of hours of cycling. The addition of inorganic additives (such as vanadium sulfate, potassium tripolyphosphate, europium chloride, etc.) can achieve uniform zinc deposition and electrode protection by leveraging the ion synergy effect, inducing crystal orientation growth, and in-situ forming a dense SEI layer, significantly improving the battery s capacity and rate performance. The introduction of ionic additives (such as ammonium hexadecyltrimethyl sulfate, ammonium bicarbonate, ammonium salicylate, etc.) can effectively inhibit zinc dendrite growth and hydrogen evolution reactions, and broaden the electrochemical window of the electrolyte through electrostatic shielding, pH buffering, and optimizing the interface charge distribution. The addition of polymer additives (such as polyacrylic acid, polyethyleneimine, carbonized polymers, etc.) can optimize the electrode/electrolyte interface environment through selective adsorption, regulating the interface electric field, and reducing the desolvation barrier, significantly enhancing the battery s cycle stability and reversibility. Although the introduction of additives in the electrolyte has targeted solutions to some key problems of the ZnSO4 electrolyte, there are still issues such as the unclear mechanism of action of various additives and the difficulty of a single additive to simultaneously address all defects of the electrolyte. Moreover, the long-term cycle stability and scalability of AZIBs still need further verification. In the future, in-depth research is needed in areas such as developing multifunctional additives (zinc deposition regulation, dual-interface passivation, solvation structure reconstruction) and composite additives (multi-component hybrid synergy, composition/concentration optimization) to enhance the performance of AZIBs and promote their industrial application.

Abstract:As a research focus in the current iron and steel metallurgy industry, blast furnace gas fine desulfurization technology employs the three-stage dry treatment process of dechlorination-hydrolysis-desulfurization as the mainstream process. Each stage is supplemented by specific reagents, and the cost of these reagents accounts for 60% to 80% of the overall operation and maintenance of the desulfurization process, making them of great significance to the desulfurization technology. This paper introduces three most common key reagents: dechlorinating agents, hydrolyzing agents, and iron oxide desulfurizing agents, systematically elaborating on their composition, preparation methods, reaction mechanisms, and deactivation mechanisms.For dechlorinating agents, the main preparation methods include mechanical mixing and loading methods. The mechanical mixing method is primarily used for preparing calcium oxide-based dechlorinating agents with lower production costs, while the loading method is mainly applied to alumina-based dechlorinating agents, resulting in more uniform dispersion of active components but higher preparation costs. Alumina-based hydrolysis catalysts dominate the blast furnace gas fine desulfurization process; most studies on their deactivation mechanisms have focused on deactivation factors such as O₂, with insufficient in-depth research on the impact of acidic gases like HCl and HCN in blast furnace gas. The activity of iron oxide desulfurizing agents is significantly influenced by their crystal structure, and those with more crystal defects and hydroxyl active sites typically exhibit higher desulfurization activity.In terms of industrial fine desulfurization operating costs, source control processes currently remain at a disadvantage with higher costs compared to traditional end-of-pipe treatment methods. Looking forward, future research should focus on promoting studies on the synergistic mechanism of the three reagents based on actual operating conditions to optimize desulfurization efficiency and improve reagent utilization; advancing the establishment of reagent quality and evaluation standard systems to standardize the reagent market; and conducting research on the regeneration and resource utilization of waste reagents to reduce resource waste and lower operating costs.

Abstract:To address issues such as high impurity content, limited reusability, and environmental hazards in molten salt chloride residues from titanium metallurgy processes, a high-temperature stratified treatment process using sodium sulphate has been proposed. This process generates insoluble sulphates through the reaction of sodium sulphate with calcium and magnesium ions in the residues, subsequently employing stratified deposition to separate and recover the upper and middle layers of molten salt. Thermodynamic studies indicate that solidification reactions proceed spontaneously within the 400~1200℃ temperature range, though the reaction driving force diminishes with increasing temperature. When temperatures fall below below 800℃, both Kp(MgSO₄) and Kp(CaSO₄) are less than 2.01×10^-3; conversely, when temperatures exceed 800℃, Kp(MgSO₄) progressively surpasses Kp(CaSO₄) due to competitive effects between Mg²⁺ and Ca²⁺ ions. Consequently, the optimal reaction temperature should be maintained between 800 and 950℃. Experimental results indicate that under optimised conditions of 900℃ reaction temperature and a molar ratio of sodium sulphate to impurities of 1∶1, the process treatment achieves optimal performance. Calcium removal rates in the upper and middle molten salt layers reached 69% and 88% respectively, while magnesium removal rates attained 93% and 89% respectively, satisfying the 60% calcium-magnesium removal standard for re-melting. Microstructural analysis of the lower layer sample via SEM-EDS confirmed that impurity elements predominantly precipitated as stable, insoluble compounds such as MgSO₄, CaMgSiO4, and CaMg₃(SO₄)₄. This process not only achieves efficient separation of impurity elements from waste slag but also offers significant advantages including straightforward operation and low running costs. It provides a practical technical solution for the resource recovery of titanium metallurgical waste slag, holding considerable industrial application value and environmental significance.

Abstract:The third generation sulfuric acid rare earth separation process uses a lot of magnesium ions, which produces a lot of transformed magnesium sulfate wastewater. The wastewater contains SO^2-₄, saturated calcium sulfate, Mg²⁺, oil and suspended solids, and the water quality is unstable, so it is difficult to treat. Traditional lime neutralization method and traditional evaporation concentration method have some problems, such as high treatment cost and low efficiency, which lead to the treatment of rare earth magnesium sulfate wastewater has been in a blank state. This paper proposes a green process route of “calcium hydroxide neutralization-reuse of rare earth leaching”, and conducts research from three aspects: neutralization and impurity removal of magnesium sulfate wastewater, rare earth recovery rate in the neutralization liquid leaching process of sulfate rare earth, and moisture content of tailings. The following main conclusions are obtained. The process of neutralizing magnesium sulfate wastewater by calcium hydroxide is mainly affected by the pH value of the solution， when the wastewater pH is 9.5, the content of SO^2-₄ is 80.5g/L, the content of Mg²⁺ decreases to 14.7g/L, and the addition of calcium hydroxide is the smallest. The pH value of neutralization solution has a great influence on the water content of leaching tailings. When the pH value is large, the water content is low and the cake forming degree is good, when pH=9.5, the leaching rate of rare earth did not decrease obviously, and the water content of rare earth tailings reached 42.3% (the water content required for production was less than 43.0%), which reached the standard of entering the tailings pond. Conventional stirring is adopted in this process, because when the concentration of sulfate radical is more than 85.0g/L, the soluble complex formed by Ce⁴⁺, SO^2-₄ and Mg²⁺ is destroyed by high-shear stirring, so that tetravalent cerium and magnesium sulfate wastewater form flocs to precipitate, blocking the water flow channel of filter cake and affecting solid-liquid separation.Using this method to neutralize magnesium sulfate wastewater and then reuse it in rare earth calcine leaching process, rare earth leaching is not affected, and magnesium sulfate wastewater can be recycled.

Abstract:The issues of high energy consumption and high pollution in the aluminum electrolysis industry have become a major focus for the international aluminum community. Research suggests that a new electrolysis system, composed of inert anodes coupled with wettable cathodes, holds the potential to fundamentally improve the existing aluminum production process, thereby achieving the goals of energy saving and environmental protection. This study aims to enhance the physical properties of inert and wettable TiB₂-C composite cathodes for aluminum electrolysis. By modifying coal tar pitch binder with organic resins, we investigated the effects of different modified pitches on the properties of TiB₂-C composite cathode materials. The results indicated that as the resin addition amount increased from 0% to 4%, 8%, 12%, and 16%, the incorporation of furan and epoxy resins significantly improved the bulk density, open porosity, electrical resistivity, and compressive strength of the TiB₂-C composites, whereas phenolic resin showed little effect. The optimal improvement was achieved at a 12% addition of both furan and epoxy resins. For the composite prepared with furan-modified pitch, the improvements in the aforementioned physical properties were 1.6%, 2.7%, 20.4%, and 42%, respectively. Corresponding improvements for the epoxy-modified pitch composite were 0.8%, 1%, 14%, and 39.6%, respectively. SEM analysis revealed that the furan-and epoxy-modified pitches exhibited better bonding with TiB₂ particles compared to the unmodified pitch, with the furan-modified pitch providing the most complete coating. The electrolytic corrosion rate of the composite prepared with unmodified pitch was 8.07mm/a. In contrast, the composites prepared with furan-and epoxy-modified pitches showed lower corrosion rates of 4.42mm/a and 4.13mm/a, respectively, demonstrating that the modification process enhances the corrosion resistance of the TiB₂-C composite cathode materials.

Abstract:In view of the technical problems of rolling mill load caused by the large reduction of the first pass required for the preparation of copper/stainless steel composite ultra-thin strip by cold rolling, and the problem that the material properties of the copper electrodeposition method on the surface of the stainless steel substrate are not up to standard, the author 's research team proposed the idea of electrodeposition assisted cold rolling to prepare copper/stainless steel ultra-thin composite strip, that is, heat treatment and cold rolling are used to improve the bonding effect between the stainless steel substrate and the copper deposition layer. In this paper, the influence of different process parameters on the macro and micro characteristics of copper deposition layer and the growth mechanism of copper deposition layer are systematically studied. The following main conclusions are drawn. The growth process of the copper deposition layer is affected by the current density value and the surface state of the substrate. When the surface of the substrate is smooth, the initial bonding strength between the copper layer and the substrate is insufficient, and the copper layer falls off during the deposition process, so that the subsequent further processing cannot be carried out. Therefore, it is necessary to select the appropriate grinding method to keep the surface of the substrate rough. With the increase of current density (20~80mA·cm^-2), the surface roughness of the copper deposition layer increases, and when the current density reaches 80mA·cm^-2, the surface is loose and granular, and the compactness decreases significantly. During the deposition process in the stable plating zone (less than 60mA·cm^-2), the reduced copper ions form an initial copper layer on the surface of the stainless steel strip, and then continue to grow at the trough position on the outer surface of the copper deposition layer, and finally evolve into a flat and dense morphology. In the deposition process of unstable electroplating area (more than 60mA·cm^-2), copper atoms are attached to the peak position of the surface in the form of spherical particles, forming copper ridges. There are deep gullies between adjacent copper ridges, resulting in loose structure and poor compactness of copper deposition layer. The surface of the substrate was polished by 400 mesh sandpaper, and the electrodeposition effect was better under the current density of 20mA·cm^-2. The thickness of the copper deposition layer was approximately linear with the deposition time. The copper deposition layer with the required thickness can be obtained by controlling the deposition time. The results of this study provide a theoretical basis and control strategy for prefabricating the initial copper composite layer during cold rolling of copper/stainless steel ultra-thin composite strips.

Abstract:High-temperature corrosion in the waste heat boiler (WHB) of a copper flash smelting furnace frequently occurs in the ceiling area between the front wall of the radiation section and the baffle, which has consistently constrained the safe and efficient operation of the WHB. This study aims to mitigate high-temperature corrosion and avoid localized overheating of the water-cooled walls by constructing a three-dimensional model of the radiation section, investigating the flow and heat transfer characteristics of the flue gas, and conducting structural optimization of the boiler to reduce corrosion issues. The results reveal that flue gas impingement on the ceiling is the primary cause of high-temperature corrosion in the radiation section. Increasing the inlet size of the boiler and adjusting the structure and position of the baffle can effectively reduce ceiling corrosion by controlling the flue gas velocity and temperature. In contrast, variations in the parameters of the air for sulfation injection exert negligible influence on the flue gas temperature field. Numerical simulation results were further used to optimize the boiler inlet size and baffle configuration. After expanding the front wall inlet cross-section to 4.1m×4.5m (width×height), the average flue gas velocity decreased by 14%, and the maximum velocity decreased by 21%. Moreover, the distance between the corrosive temperature zone (1467~1531K) and the ceiling increased by 2.6m. These findings demonstrate that the optimized boiler structure effectively suppresses the upward impact of flue gas and reduces ceiling temperature, thereby mitigating corrosion.

Abstract:Driven by the dual demands for high energy density and high safety in power batteries, the allowable threshold for magnetic impurities (e.g., Fe, Ni, Cr, Zn) in battery-grade lithium carbonate continues to decrease amid the rapid development of the global new energy industry. The current non-ferrous metals industry standard “Battery Grade Lithium Carbonate” (YS/T 582—2023) specifies that the magnetic impurity content must be ≤300ppb (mass ratio, 10-9), while some companies require this threshold to be reduced to 100ppb. Although most domestic lithium salt producers can meet the current industry standard, issues such as significant fluctuations in magnetic impurity content and occasional non-compliant batches persist. To address this, based on the spodumene lithium extraction process via the sulfuric acid method, this study adopted a full-process quality tracking approach and identified the main sources of magnetic impurities in the lithium carbonate production system: raw and reagents, martensitic phase transformation induced by cold processing of stainless steel equipment, construction contaminants, and particles from environmental corrosion. It was found that solution treatment (quenching at 1050-1100℃) of austenitic stainless steel can effectively reverse martensitic phase transformation, restoring the relative magnetic permeability (μ_r) to approximately 1 and thereby suppressing the release of magnetic impurities from equipment. Based on comprehensive quality tracking and control, a “five-in-one” hierarchical control technology was proposed, encompassing raw material screening, equipment upgrading and demagnetization, process optimization, environmental purification, and staff standardization. Practical application demonstrates that implementing this control technology can stably maintain the magnetic impurity content in battery-grade lithium carbonate products at ≤50ppb, meeting the quality requirements for high-end lithium battery materials. The successful implementation of this technology promotes the green and high-quality upgrade of the sulfuric acid-based lithium extraction process and provides a technical paradigm for quality control of raw materials in the lithium battery industry.

Abstract:In the novel beryllium extraction process, which involves low-temperature dissociation of beryllium ore followed by physical purification of beryllium-containing compounds, the physical purification step is achieved by exploiting differences in the solubility of fluorides, thereby enabling efficient separation of impurities. As one of the main impurities in beryllium ore, manganese and the dissolution behavior of MnF₂ have received relatively little attention in previous studies. This work systematically investigates the dissolution characteristics of MnF₂ in both H₂O and H₂O-HF systems during the removal of manganese-containing fluorides in beryllium ore smelting. The following key conclusions were obtained. MnF₂ exhibits low solubility in water, and its solubility decreases with increasing temperature. In the presence of HF, it reacts to form soluble H₂MnF₄, resulting in a slight increase in solubility. The theoretical solubility of MnF₂ is 0.69 g/100 g, while the experimentally measured value is 0.78 g/100 g. After dissolution, it exists predominantly in ionic form. In the H₂O-HF system containing BeF_2， BeF₂ demonstrates relatively high solubility. However, owing to its small equilibrium constant, both BeF₂ and H₂BeF₄ coexist in the solution, and the concentration of H₂BeF₄ is governed by both temperature and HF concentration. HF and MnF₂ simultaneously supply F-, which coordinate with BeF₂ to form BeF^2-₄. This complex subsequently combines with Mn²⁺ to produce insoluble MnBeF₄, thereby facilitating the removal of impurities. Unbound coordinating ions remain in solution as H₂BeF₄.