# Testing Low-Density Polyethylene Membranes for Lithium Isotope Electromigration System

## TL;DR

This paper explores how low-density polyethylene membranes can be used to separate lithium isotopes using electromigration, finding that voltage and migration time significantly affect the enrichment of 6Li and 7Li.

## Contribution

The study introduces a novel setup and protocol for high-precision lithium isotope measurements and evaluates the performance of impregnated and non-impregnated membranes in electromigration.

## Key findings

- Both impregnated and non-impregnated membranes achieved promising 6Li enrichment under different conditions.
- Lithium-ion mobility increases quasi-linearly with voltage up to 15 V and peaks between 20 and 25 hours.
- The maximum single-stage separation factor for 6Li/7Li was achieved at specific times for both membrane types.

## Abstract

The growing energy demand has emphasized the importance of developing nuclear technologies and high-purity lithium isotopes (6Li and 7Li) as raw materials. This study investigates how voltage and migration time affect two types of low-density polyethylene membranes—one impregnated with ionic liquids and the other non-impregnated—for lithium isotope separation via electromigration from a lithium-loaded organic phase to an aqueous solution. We developed a laboratory-made setup for high-precision lithium isotope measurements (2RSD = ±0.30‰) of natural carbonate samples (LSVEC) and an optimized protocol for isotope ratio measurements using quadrupole ICP-MS with the sample-standard bracketing method (SSB). The results document that both impregnated and non-impregnated membranes can achieve promising 6Li enrichment under different environmental conditions, including ionic liquids and organic solutions in the cathode chamber. Lithium-ion mobility is influenced by voltage in an environment assisted by 0.1 mol/L tetrabutylammonium perchlorate and increases quasi-linearly from 5 to 15 V. Between 20 and 25 h, the lithium-ion concentration had the maximum value, after which the trend declined. In the BayesGLM model, we incorporated all data and systematically eliminated those with a low enrichment factor, either individually or in groups. Our findings indicated that the model was not significantly affected by the exclusion of measurements with low α. This suggests that voltage and migration time are crucial, and achieving a better enrichment factor depends on applying the optimal ratio of ionic liquids, crown ethers, and organic solvents. Ionic liquids used for impregnation sustain enrichment in the first hours, particularly for 7Li; however, after 25 h, 6Li demonstrated a higher enrichment capacity. The maximum single-stage separation factor for 6Li/7Li was achieved at 24 and 48 h for an impregnated membrane M2 (α = 1.021/1.029) and a non-impregnated membrane M5 (α = 1.031/1.038).