Recovering materials from end-of-life lithium-ion batteries
File(s)
Author(s)
Wei, Xiaochu
Type
Thesis
Abstract
The rapidly growing market of consumer electronics and electric vehicles is driving the global lithium ion battery (LiB) market, the value of which is projected to grow exponentially from USD 41.1 billion in 2021 to USD 116.6 billion by 20301. Classification of different types of LiBs is based mainly on their positive electrode (‘cathode’) chemistries; usually, LIBs contain CoIII/II, NiIII/II, MnIV/III/II, in addition to LiI/0, as well as Al and Cu current collectors. The quantities of these materials are geologically limited, with some being depleted rapidly, therefore recovering materials from end-of-life batteries needs to be accelerated urgently. To this end, the ultimate technological aim of this project was to develop an energy-efficient, environmentally benign and economically viable electrochemical process for recovering metal-containing materials from spent LiBs.
The results reported in this thesis were based on the conceptual design of a closed-loop electrochemical system for the separation and recovery of lithium(I) and other metals from spent LiBs. The process design comprised two electrochemical reactors coupled hydraulically: an electrochemical reactor for the dissolution of electrode materials, and an electrodeposition reactor for the recovery of metal ions from solution. Both reactors notionally incorporate the same membrane separator for transporting LiI ions selectively, and separating anolyte and catholyte, the latter possibly containing organic solvent rather than the aqueous solutions of the former.
The development and characterisation of membranes that can transport LiI ions selectively were the first objectives of this PhD project. Perovskite-type (Li3xLa(2/3) xTiO3, LLTO), garnet-type (Li7La3Zr2O12, LLZO) and NASICON-type (Li1+xAlxTi2−x(PO4)3, LATP and Li1+xAlxGe2−x(PO4)3, LAGP) were the membrane material candidates identified in the literature review. Of these, LLTO and LATP were synthesised successfully and investigated via electrochemical impedance spectroscopy (EIS) measurements in synthetic aqueous solutions containing LiI, MnII, CoII and NiII, mimicking leach solutions from LiB positive electrodes. Behaviours of the synthesised membranes were compared with those of commercial non-porous LAGP pellets, which were the closest analogues to non-porous LATP pellets that could be obtained. EIS demonstrated the feasibility of using ceramic Li-ion conducting membranes in aqueous solutions, but time- and pH-dependent material degradation of both LLTO and LAGP membranes was detected. Their durability needs to be increased to enable adequate longevity of the proposed process.
The second objective was the electrochemical recovery of MnII ions (as well as CoII and NiII ions) in elemental or oxide forms from the anolyte of an electrodeposition reactor, while LiI ions migrate through the membrane, thereby increasing LiI concentrations in the catholyte. MnII oxidation kinetics were determined as a function of the concentration of metal ions (LiI, MnII, NiII and CoII), pH, electrode type and electrode potential. Anodic electrodeposition of MnO2 was determined to be feasible both thermodynamically and kinetically, in parallel with LiI transport through the membrane. Hence, designing an electrochemical reactor to achieve this at an optimised current density became the second focus of the project.
Several electrodeposition reactors incorporating either a ceramic membrane or a polymeric anion permeable membrane were fabricated with different configurations and flow distributions. These reactors were operated in batch recycle mode to determine the time evolution of conversion ratios of MnII ions and charge yields. The best performance was delivered with a reactor incorporating a commercial LAGP membrane, achieving the highest charge yield (ca. 99 %) with a conversion of 97.5 % (from 0.01 mol dm-3 to 1.5 × 10-4 mol dm-3) for the recovery of MnII ions from the anolyte, as well as a ca. 90 % charge yield of lithium(I) enrichment in the catholyte (from 0 mol dm-3 to 0.045 mol dm-3). Furthermore, NiII and CoII ions were found not to co-oxidise with MnII over an electrode potential range of 0.8 - 1.2 V vs. AgCl/Ag. Hence, NiII and CoII would have to be recovered, probably as an alloy, by cathodic electrodeposition in a third reactor, the feasibility of which has been reported in the literature2 but was not investigated in this project.
Overall, the project developed an electrodeposition system based on inorganic LiI ion-conducting membranes to separate and recover MnII ions as MnO2, and LiI ions in a separate, pure aqueous solution, potentially forming the basis of a process for material recovery and recycling from spent LiBs.
The results reported in this thesis were based on the conceptual design of a closed-loop electrochemical system for the separation and recovery of lithium(I) and other metals from spent LiBs. The process design comprised two electrochemical reactors coupled hydraulically: an electrochemical reactor for the dissolution of electrode materials, and an electrodeposition reactor for the recovery of metal ions from solution. Both reactors notionally incorporate the same membrane separator for transporting LiI ions selectively, and separating anolyte and catholyte, the latter possibly containing organic solvent rather than the aqueous solutions of the former.
The development and characterisation of membranes that can transport LiI ions selectively were the first objectives of this PhD project. Perovskite-type (Li3xLa(2/3) xTiO3, LLTO), garnet-type (Li7La3Zr2O12, LLZO) and NASICON-type (Li1+xAlxTi2−x(PO4)3, LATP and Li1+xAlxGe2−x(PO4)3, LAGP) were the membrane material candidates identified in the literature review. Of these, LLTO and LATP were synthesised successfully and investigated via electrochemical impedance spectroscopy (EIS) measurements in synthetic aqueous solutions containing LiI, MnII, CoII and NiII, mimicking leach solutions from LiB positive electrodes. Behaviours of the synthesised membranes were compared with those of commercial non-porous LAGP pellets, which were the closest analogues to non-porous LATP pellets that could be obtained. EIS demonstrated the feasibility of using ceramic Li-ion conducting membranes in aqueous solutions, but time- and pH-dependent material degradation of both LLTO and LAGP membranes was detected. Their durability needs to be increased to enable adequate longevity of the proposed process.
The second objective was the electrochemical recovery of MnII ions (as well as CoII and NiII ions) in elemental or oxide forms from the anolyte of an electrodeposition reactor, while LiI ions migrate through the membrane, thereby increasing LiI concentrations in the catholyte. MnII oxidation kinetics were determined as a function of the concentration of metal ions (LiI, MnII, NiII and CoII), pH, electrode type and electrode potential. Anodic electrodeposition of MnO2 was determined to be feasible both thermodynamically and kinetically, in parallel with LiI transport through the membrane. Hence, designing an electrochemical reactor to achieve this at an optimised current density became the second focus of the project.
Several electrodeposition reactors incorporating either a ceramic membrane or a polymeric anion permeable membrane were fabricated with different configurations and flow distributions. These reactors were operated in batch recycle mode to determine the time evolution of conversion ratios of MnII ions and charge yields. The best performance was delivered with a reactor incorporating a commercial LAGP membrane, achieving the highest charge yield (ca. 99 %) with a conversion of 97.5 % (from 0.01 mol dm-3 to 1.5 × 10-4 mol dm-3) for the recovery of MnII ions from the anolyte, as well as a ca. 90 % charge yield of lithium(I) enrichment in the catholyte (from 0 mol dm-3 to 0.045 mol dm-3). Furthermore, NiII and CoII ions were found not to co-oxidise with MnII over an electrode potential range of 0.8 - 1.2 V vs. AgCl/Ag. Hence, NiII and CoII would have to be recovered, probably as an alloy, by cathodic electrodeposition in a third reactor, the feasibility of which has been reported in the literature2 but was not investigated in this project.
Overall, the project developed an electrodeposition system based on inorganic LiI ion-conducting membranes to separate and recover MnII ions as MnO2, and LiI ions in a separate, pure aqueous solution, potentially forming the basis of a process for material recovery and recycling from spent LiBs.
Version
Open Access
Date Issued
2022-05-01
Date Awarded
01/12/2022
Advisor
Hankin, Anna
Kelsall, Geoff
Publisher Department
Chemical Engineering
Publisher Institution
Imperial College London
Qualification Level
Doctoral
Qualification Name
Doctor of Philosophy (PhD)