Entropy-dominated and alkali-enhanced scandium partitioning into clinopyroxene revealed by density functional theory and machine learning

Abstract Scandium holds significant importance in modern technology and is recognized as a strategic or critical metal in many countries. Scandium-rich clinopyroxene is important in forming various Sc resources, whereas the favorable conditions in forming Sc-rich clinopyroxene remain unclear. By compiling available experimental data on Sc partition coefficients between clinopyroxene and melt (DSccpx/melt), we demonstrate that DSccpx/melt values increase exponentially with the alkali contents in melts. Using first-principle calculations combined with a simple thermodynamic model, we propose that this trend arises from simultaneous competitive formation of AlM1AlT, NaM2AlM1 and NaM2ScM1 pairs in clinopyroxene. This mechanism differs from conventional models such as charge-coupled substitutions, which would increase Sc partitioning with Al through ScM1AlT pairing. The incorporation of Sc into clinopyroxene is highly energetically unfavorable compared to Al, suggesting that Sc uptake may be governed mainly by configurational entropy rather than enthalpic factors. Entropy-driven competition model predicts a pronouced temperature dependence of Sc partitioning, with lower temperature and alkali-richer melts tending to yield larger DSccpx/melt values, thereby potentially promoting Sc enrichment in clinopyroxene. Our model demonstrates that Sc partitioning may reflect a complex interplay of different components in both the melt and solid phases, and previous regressions of experimental data may have not fully captured these effects. To complement the thermodynamic approach, we develop a predictive machine-learning model trained on experimental data, achieving a high coefficient of determination (R2 = 0.98). Together, the thermodynamic framework and data-driven model quantify how melt alkalis and temperature regulate Sc partitioning, provide new constraints on the petrogenic conditions and magma compositions that promote Sc-rich clinopyroxene and, by extension, Sc enrichment in alkaline magmatic systems.

Authors

Institutions

Publication Details

Journal
American Mineralogist
Published
2026-09-04
DOI
https://doi.org/10.2138/am-2025-10153
Primary Topic
Bauxite Residue and Utilization
Type
article
Field-Weighted Citation Impact
0.00
Controls
|||
ALL TIME
JAN
FEB
MAR
APR
MAY
JUN
JUL
AUG
SEP
OCT
article

Entropy-dominated and alkali-enhanced scandium partitioning into clinopyroxene revealed by density functional theory and machine learning

Mei‐Fu Zhou, Zhenchao Wang, Guang-Shao Wang, Fei-Wu Zhang et al.
American Mineralogist
Bauxite Residue and Utilization
article

Entropy-dominated and alkali-enhanced scandium partitioning into clinopyroxene revealed by density functional theory and machine learning

Mei‐Fu Zhou, Zhenchao Wang, Guang-Shao Wang, Fei-Wu Zhang, Ping-Ping Liu, Joshua MR Muir
article en

Abstract

Abstract Scandium holds significant importance in modern technology and is recognized as a strategic or critical metal in many countries. Scandium-rich clinopyroxene is important in forming various Sc resources, whereas the favorable conditions in forming Sc-rich clinopyroxene remain unclear. By compiling available experimental data on Sc partition coefficients between clinopyroxene and melt (DSccpx/melt), we demonstrate that DSccpx/melt values increase exponentially with the alkali contents in melts. Using first-principle calculations combined with a simple thermodynamic model, we propose that this trend arises from simultaneous competitive formation of AlM1AlT, NaM2AlM1 and NaM2ScM1 pairs in clinopyroxene. This mechanism differs from conventional models such as charge-coupled substitutions, which would increase Sc partitioning with Al through ScM1AlT pairing. The incorporation of Sc into clinopyroxene is highly energetically unfavorable compared to Al, suggesting that Sc uptake may be governed mainly by configurational entropy rather than enthalpic factors. Entropy-driven competition model predicts a pronouced temperature dependence of Sc partitioning, with lower temperature and alkali-richer melts tending to yield larger DSccpx/melt values, thereby potentially promoting Sc enrichment in clinopyroxene. Our model demonstrates that Sc partitioning may reflect a complex interplay of different components in both the melt and solid phases, and previous regressions of experimental data may have not fully captured these effects. To complement the thermodynamic approach, we develop a predictive machine-learning model trained on experimental data, achieving a high coefficient of determination (R2 = 0.98). Together, the thermodynamic framework and data-driven model quantify how melt alkalis and temperature regulate Sc partitioning, provide new constraints on the petrogenic conditions and magma compositions that promote Sc-rich clinopyroxene and, by extension, Sc enrichment in alkaline magmatic systems.

American Mineralogist
Peking University (CN), China University of Geosciences (CN), Institute of Geochemistry (CN)
Industry, innovation and infrastructure
Openalex Percentile: Top 19%
Bauxite Residue and Utilization
AI Navigator

Ask Laika to Summarize, Analyze, and Connect papers live on the map.

Summarize Papers & Methodologies

Extract key findings, datasets, and comparative methods across publications.

Benchmark Rankings & Visual Analytics

Rank top research institutions, authors, funders, topics, and journals by Field-Weighted Citation Impact (FWCI) and paper volume with instant charts.

Connect Distant Disciplines

Bridge topological clusters on the map to find hidden collaborative intersections.