Gallium is a critical strategic metal with no independently exploitable deposits. It occurs predominantly as a low-grade, dispersed, and associated element in bauxite, lead–zinc ores, and coal seams, and is almost exclusively recovered as a by-product, which constrains the stability of the global gallium supply chain. Conventional coal-based gallium recovery via complete combustion or gasification causes severe gallium volatilization losses at high temperatures, while CO2-based coal gasification has focused exclusively on syngas production, with little attention paid to the recovery of critical metals. This study presents a novel CO2-driven Boudouard reaction strategy for gallium pre-enrichment from low-rank coal. This reaction selectively removes carbon while preserving gallium-bearing aluminosilicate phases, effectively mitigating gallium volatilization and simultaneously achieving gallium enrichment. Under optimized conditions (5 L/min CO2 flow rate, 1100 ℃, 60 min), this strategy achieved 91% decarburization efficiency and a 6.7-fold gallium enrichment (15 to 100 μg/g). Subsequent acid leaching (6 mol/L HCl, 2 h, L/S = 10:1, 90 ℃) yielded 49% gallium leaching efficiency, 7.5 times higher than the 6.5% achieved with raw low-rank coal. Thermokinetic analysis identified the three-dimensional diffusion (Jander) model as the rate-limiting step for decarburization, with an apparent activation energy of 242.47 kJ/mol and a net heat absorption of 12.02 kJ/g during decarburization. This study offers valuable insights into strategic gallium recovery and high-value carbon utilization from low-rank coal.
With the accelerating pace of industrialization, the highly alkaline by-product, red mud, produced in the aluminum industry during the production process, has attracted significant attention due to its increasing accumulation in the environment. However, red mud possesses unique physical and chemical properties, such as relatively high porosity, a substantial content of iron and aluminum oxides, and good adsorption capacity. This makes it a promising low-cost material for environmental remediation. This article summarizes the latest advancements in pollution control technologies that utilize red mud for environmental restoration, including soil improvement, wastewater treatment, and gas purification. Red mud can be used to remediate soil through mechanisms such as alkaline neutralization and co-precipitation, particularly in cadmium-contaminated and saline-alkali soils. Wastewater treatment leverages the high specific surface area, fine particle size, and strong adsorption properties of red mud. Moreover, its adsorption capacity can be enhanced through acid treatment, heat treatment, and other activation methods. The alkaline components in red mud can react with sulfur dioxide and carbon dioxide, thereby significantly improving the removal efficiency of sulfur dioxide in flue gas desulfurization. This paper examines the physicochemical properties of red mud and conducts a risk assessment, identifying several challenges. These include the long-term stability of immobilized pollutants, the potential secondary release of residual heavy metals, and the high energy consumption required during processing. Overall, red mud shows promising application prospects as an environmentally friendly and efficient material for environmental remediation. However, further research is still needed to optimize its comprehensive utilization and reduce potential environmental risks.
In this paper, an algorithmic and simulation-based methodology is proposed to enable closed-loop lithium-ion battery recycling using an intelligent battery identification, degradation prediction, robot disassembly, adaptive material recovery and circular manufacturing optimization decision making pipeline. A multimodal framework is used to integrate data from both the NASA Lithium-Ion Battery Aging Dataset and the RecyBat24 image dataset for battery classification and degradation assessment, and digital-twin and process level simulation models are used to analyze the potential for robotic disassembly, material recovery and optimizing circular battery manufacturing. The average battery degradation prediction error was less than 1% for the Battery Prediction Module and the accuracy of the battery classification module using a Vision Transformer was 98.5% under benchmark testing. The results from the simulation-based validation yielded a 97.6% success rate for robotic disassembly, as well as a 96.2% efficiency of materials recovered, and predicted material purity of more than 98.8% when operating the robot optimally. This was further confirmed through comparative evaluation, ablation analysis and sensitivity studies, which showed the contribution of the proposed multimodal integration and adaptive optimization strategies. The results show the technical viability of the proposed architecture of an intelligent recycling system in a controlled environment of the benchmark and simulations. Laboratory-scale experiments, the validation of recycling at pilot-scale and the deployment at industrial-scale with different battery chemistries and conditions will be focused in future. The experimental testing with open access data sets and controlled simulation environments showed that the developed AI-Integrated Closed-Loop (AICL) approach was able to demonstrate a high level of performance in the identification, degradation prediction, robotic disassembly, and material recovery of batteries.
