Incorporation of metal speciation data into QICAR models and application to data-poor technology-critical elements

Area of Science:

• Environmental Chemistry
• Ecotoxicology
• Computational Chemistry

Background:

• Quantitative ion character-activity relationships (QICARs) previously linked metal properties to aquatic toxicity.
• These models were initially developed for data-rich elements and applied to technology-critical elements (TCEs).

Purpose of the Study:

• To extend QICARs by incorporating calculated metal speciation for improved toxicity prediction.
• To evaluate the predictive power of free metal ion activity versus total dissolved metal concentration.
• To investigate the toxicity of TCEs, particularly those with complex speciation.

Main Methods:

• Utilized linear free energy relationships to estimate thermodynamic constants.
• Calculated metal speciation and expressed toxicity endpoints as free metal activities.
• Performed regression analyses using the covalent index (log-transformed χm2r) as a predictor.

Main Results:

• Incorporating metal speciation slightly improved regression determination coefficients for training metals.
• The covalent index remained the best predictor for training metals' acute toxicity.
• Regressions for TCEs were significantly poorer, especially at very low free metal ion concentrations (< 10-18 M).

Conclusions:

• The poor performance for TCEs is attributed to their distinctive speciation, dominated by neutral polyhydroxo species.
• QICARs are still valuable, but the study highlights that free-ion activity alone may be inadequate for predicting toxicity of metals with complex speciation.
• This suggests limitations in current models for data-poor metals with complex environmental chemistry.