MARIS Dataset: Rare Earth Element Solvent-Extraction Distribution Coefficients (lg D) as a Function of pH

Description

This dataset compiles 764 experimental data points on rare earth element (REE) solvent-extraction distribution coefficients (lg D) as a function of aqueous-phase pH, assembled from peer-reviewed literature using the Multi-Agent Rare Earth Intelligence System (MARIS), a six-agent AI framework automating literature data extraction, cleaning, and theoretical correction. The dataset spans 35 extractant-condition groups across 20 distinct extractant systems, covering 191 individual REE-system entries. Each record includes extractant identity, extractant concentration, REE, pH, and lg D, with a group identifier linking each record to its unique extractant-concentration combination and original literature source. Extractant concentrations range from 0.002 M (H4L) to 1.51 M (D2EHPA), with some binary-mixture systems reported without a specific concentration value. The data has been cleaned for duplicate removal and naming consistency across extractant and REE labels, and is provided as the empirical (pre-correction) dataset used in the accompanying study "A Multi-Agent AI Framework for Machine Learning-Based Prediction of Rare Earth Element Solvent-Extraction Distribution Coefficients Based on pH." This dataset supports reproducibility of the machine learning models trained in the study and may be useful for further research into REE solvent-extraction behavior across a wide range of extractant systems.

Institutions

• University of Saskatchewan

  Saskatchewan, Saskatoon

Categories

Licence