Recovery of Rare Earth Elements From Waste Materials: Leaching and Separation Using Supported Liquid Membranes

Rare earth elements (REEs; defined as the stable lanthanides, yttrium, and scandium) are ubiquitous in a wide range of modern technologies; however, their global market supply is dominated by a single source and has been subject to major disruptions in recent years. This has led to an interest in finding alternative, non-traditional sources of REEs. Potential non-traditional sources include secondary waste materials such as coal combustion residual (CCR), acid mine drainage (AMD), and electronic wastes (e-waste), among others.While methods for processing and purification of REEs from traditional ores are reasonably well established, there are not well-established processing and purification methods for low-grade, non-traditional sources. This is in part because non-traditional sources have a complex and diverse chemical composition and relatively low REE concentrations. This work focused on understanding the factors controlling REE recovery in two steps of the recovery process: 1) acid leaching of REEs from solid wastes and 2) separation using a supported liquid membrane (SLM) process. The waste materials used in this research were CCR, coal processing refuse, and AMD. The overall aim of this work was to understand how the diverse chemistry of waste materials affects REE recovery during leaching and purification. Specifically, this research addressed the following aims: 1. Identify the factors that control REE solubility (e.g., pH and major element concentration) in acid leachates of coal fly ashes and coal residue.

2. Quantify the effects of various leachate characteristics such as major metal concentrations and feed pH on REE mass transfer in SLMs using synthetic feeds that mimic real feedstocks from low-grade waste materials.3. Evaluate the effectiveness of SLMs as a method for selective separation of REEs from real AMD feeds from sites distinct water chemistries and identify feed characteristics that are predictive of SLM performance. The aim of Chapter 2 of this research was to understand the connection between REE solubility, pH, and major elemental components of leachates for coal by-products. One of the early steps in recovering REEs from coal by-products is often acid leaching, which can result in low pH leachates with complex aqueous chemistry. To accomplish the aim of this chapter, we investigated the effects of solids concentration (i.e., pulp density) and pH adjustment on REE solubility in acid leachates of coal fly ashes from the Powder River Basin (PRB) and Appalachian Basin in the United States, and a coal processing refuse from the Southwestern U.S. For PRB ashes, the concentrations of soluble REEs generally increased with increasing pulp density; however, at pulp density values above 80-100 g/L, the soluble REE concentrations in the leachates were markedly lower. Similarly, the soluble concentrations of other major solutes (Fe, Al, Si) that leached from PRB fly ashes were also non-linear with pulp density. These major elements tended to reach maximum concentration values at 60-70 g/L pulp density. In contrast, for the Appalachian fly ashes and the coal by-product, soluble concentrations of REE and major elements in leachates increased linearly with pulp density. Chemical equilibria calculations of mineral saturation indices indicated that trends in soluble REE concentrations could be explained by saturation conditions for Fe and Al-(hydr)oxides and possibility sulfate minerals, but not lanthanide hydroxides. Furthermore, pH adjustment of the acid leachates showed that REEs and many major solutes were removed from solution at pH values above 4.5, also consistent with Fe- and Al-(hydr)oxide precipitation. These results highlight the importance of understanding the chemical composition of leachates when designing REE recovery processes for low-grade geologic feedstocks and that precipitation of hydr(oxide) or sulfate minerals of major elements rather than discreet formation of REE mineral phases could be used for process optimization. The aim of Chapter 3 of this work was to quantify the effects of three different competitive metals (Fe(III), Fe(II), and Al) and feed pH on the mass transfer and recovery of two rare earth elements (REE) (Nd and Er) using SLMs. SLMs are a promising alternative to solvent exchange processes that combine two different unit operations (extraction and stripping) into one. To our knowledge, there are no studies that have systematically assessed the effects of individual metals on REE mass transfer using SLMs with feeds that have metals concentrations relevant to low-grade waste materials. Previous work has suggested that competition by non-REE metals for metal binding sites at the membrane interface can decrease REE recovery when using CCR leachates. To accomplish the aim of this chapter, we used simulated feeds that are representative of low-grade feedstocks of REEs, such as coal fly ash and acid mine drainage (AMD). The simulated feeds consisted of either Nd or Er as the REE of interest and either Fe(III), Fe(II), or Al as the competitive metal. The feeds had relatively low REE concentrations (0.01-100 µmol/L) compared to the competitive metals, Fe and Al (0.017-35 mmol/L), and had pH values from 1 to 3.5. The results showed that at sufficiently high concentrations of competitive metals REE mass transfer could be impeded. Further, the concentration at which REE mass transfer was inhibited was lowest for Fe(III) compared to Fe(II) and Al. The results also showed that feed pH is a major driver of REE transfer across the membrane, with feeds at higher pH having higher mass transfer. Additionally, we were able to show that the observed decreases in REE mass transfer were not due to unwanted loss mechanisms in the bulk feed (e.g., coprecipitation with mineral phases formed over the lifetime of the reactor) or due to mineral formation on the membrane surface blocking sorption sites at the feed-membrane interface. The results of this study will help to provide a framework for predicting how REE mass transfer will be affected when using real feeds with complex aqueous chemistries and multiple competitive metals. In the fourth chapter of this research, we evaluated the recovery of REEs from a real potential feedstock, AMD, using SLMs. The major aims for this study were to: 1) assess the effectiveness of SLM-based REE separation from AMD samples representing a spectrum of aqueous compositions; 2) determine the effects of AMD storage and holding time on separation performance; and 3) assess the impact of AMD pre-treatment (e.g., filtration and pH adjustment) on REE recovery. The results showed that relative separation fluxes of REE by SLM correlated with AMD characteristics such as pH and major ions such as Fe and Ca. The purity of acid strippant product, expressed as REE dry weigh content, depended on the initial REE concentrations in the AMD. Additionally, Fe(II) oxidation during the aging of AMD samples significantly decreased REE mass transfer by SLM separation. However, filtration of freshly collected AMD limited Fe(II) oxidation, enabling flexibility in feed stock storage time for separation of AMD. Pre-treatment of AMD samples by pH adjustment did not substantially improve separation performance. Overall, this study provides a framework for applying SLMs for REE recovery from AMD sources by establishing primary water quality parameters that influence separation flux and product purity. Such insights are needed to support a mechanistic understanding of critical metals separation by SLM for complex and nontraditional feedstocks such as AMD wastes. In total, this research provides insight on key feed characteristics that control recovery of REEs from low-grade waste materials during two stages of the recovery process: acid leaching and separation using SLMs. The results from this research can be used to help engineer more efficient recovery processes that could help low-grade waste materials become economically viable sources of REEs.