skip to main content
OSTI.GOV title logo U.S. Department of Energy
Office of Scientific and Technical Information

Title: A chemistry-informed hybrid machine learning approach to predict metal adsorption onto mineral surfaces

Journal Article · · Applied Geochemistry
ORCiD logo [1];  [1];  [2];  [3]
  1. Lawrence Livermore National Laboratory (LLNL), Livermore, CA (United States)
  2. Lawrence Berkeley National Laboratory (LBNL), Berkeley, CA (United States)
  3. Lawrence Berkeley National Laboratory (LBNL), Berkeley, CA (United States); University of California, Berkeley, CA (United States)

Historically, surface complexation model (SCM) constants and distribution coefficients (Kd) have been employed to quantify mineral-based retardation effects controlling the fate of metals in subsurface geologic systems. Our recent SCM development workflow, based on the Lawrence Livermore National Laboratory Surface Complexation/Ion Exchange (L-SCIE) database, illustrated a community FAIR data approach to SCM development by predicting uranium(VI)-quartz adsorption for a large number of literature-mined data. Here, we present an alternative hybrid machine learning (ML) approach that shows promise in achieving equivalent high-quality predictions compared to traditional surface complexation models. At its core, the hybrid random forest (RF) ML approach is motivated by the proliferation of incongruent SCMs in the literature that limit their applicability in reactive transport models. Our hybrid ML approach implements PHREEQC-based aqueous speciation calculations; values from these simulations are automatically used as input features for a random forest (RF) algorithm to quantify adsorption and avoid SCM modeling constraints entirely. Named the LLNL Speciation Updated Random Forest (L-SURF) model, this hybrid approach is shown to have applicability to U(VI) sorption cases driven by both ion-exchange and surface complexation, as is shown for quartz and montmorillonite cases. The approach can be applied to reactive transport modeling and may provide an alternative to the costly development of self-consistent SCM reaction databases.

Research Organization:
Lawrence Livermore National Laboratory (LLNL), Livermore, CA (United States); Lawrence Berkeley National Laboratory (LBNL), Berkeley, CA (United States)
Sponsoring Organization:
USDOE Office of Science (SC), Biological and Environmental Research (BER). Earth & Environmental Systems Science (EESS); USDOE Office of Nuclear Energy (NE), Office of Spent Fuel and Waste Disposition. Office of Spent Fuel and Waste Science and Technology
Grant/Contract Number:
SCW1053; AC52-07NA27344; AC02-05CH11231
OSTI ID:
1989507
Alternate ID(s):
OSTI ID: 1992184
Report Number(s):
LLNL-JRNL-838970; S0883292723001762; 105731; PII: S0883292723001762
Journal Information:
Applied Geochemistry, Journal Name: Applied Geochemistry Vol. 155; ISSN 0883-2927
Publisher:
ElsevierCopyright Statement
Country of Publication:
United Kingdom
Language:
English

References (51)

Ligand properties of surface silanol groups. I. surface complex formation with Fe3+, Cu2+, Cd2+, and Pb2+ journal May 1976
Machine learning prediction of biochar yield and carbon contents in biochar based on biomass characteristics and pyrolysis conditions journal September 2019
Towards an understanding of the sorption of U(VI) and Se(IV) on sodium bentonite journal December 1998
Sorption of uranium(VI) onto phyllite journal October 1998
Adsorption of UO2 2+ on natural composite materials journal November 1999
A hybrid machine learning model to predict and visualize nitrate concentration throughout the Central Valley aquifer, California, USA journal December 2017
U(VI) sorption on montmorillonite in the absence and presence of carbonate: A macroscopic and microscopic study journal September 2012
The effect of pH and ionic strength on the adsorption of glyphosate onto ferrihydrite journal May 2019
The Influence of Uranyl Hydrolysis and Multiple Site-Binding Reactions on Adsorption of U(VI) to Montmorillonite journal January 1995
A computational pipeline to generate a synthetic dataset of metal ion sorption to oxides for AI/ML exploration journal September 2022
Acid–Base Properties and Heavy and Alkaline Earth Metal Adsorption on the Oxide–Solution Interface: Non-Electrostatic Model journal October 1998
Modeling groundwater nitrate concentrations in private wells in Iowa journal December 2015
Reactive Transport at the Crossroads journal September 2019
Potential application of machine learning for exploring adsorption mechanisms of pharmaceuticals onto biochars journal January 2022
Effect of phosphate on U(VI) sorption to montmorillonite: Ternary complexation and precipitation barriers journal February 2016
Conceptual Model of Arsenic Mobility in the Shallow Alluvial Aquifers Near Venice (Italy) Elucidated Through Machine Learning and Geochemical Modeling journal September 2020
Modelling the sorption of Mn(II), Co(II), Ni(II), Zn(II), Cd(II), Eu(III), Am(III), Sn(IV), Th(IV), Np(V) and U(VI) on montmorillonite: Linear free energy relationships and estimates of surface binding constants for some selected heavy metals and actinides journal February 2005
Surface complexation modeling of the effects of phosphate on uranium(VI) adsorption journal May 2007
Development of machine learning models of β-cyclodextrin and sulfobutylether-β-cyclodextrin complexation free energies journal October 2011
Sorption of Actinides in Well-Defined Oxidation States on Geologic Media journal January 1981
Cesium sorption reversibility and kinetics on illite, montmorillonite, and kaolinite journal January 2018
Surface complexation modeling of Uranium(VI) sorption on quartz in the presence and absence of alkaline earth metals journal June 2013
machine. journal October 2001
Modeling uranium(VI) adsorption onto montmorillonite under varying carbonate concentrations: A surface complexation model accounting for the spillover effect on surface potential journal January 2018
Uranium(VI) adsorption and surface complexation modeling onto vadose sediments from the Savannah River Site journal February 2018
Intercomparison and Refinement of Surface Complexation Models for U(VI) Adsorption onto Goethite Based on a Metadata Analysis journal June 2021
Sorption of uranium(VI) at the clay mineral–water interface journal October 2010
Additive Surface Complexation Modeling of Uranium(VI) Adsorption onto Quartz-Sand Dominated Sediments journal May 2014
A triple-layer model of the surface electrochemical properties of clay minerals journal February 2004
Deep learning and process understanding for data-driven Earth system science journal February 2019
Community Data Mining Approach for Surface Complexation Database Development journal February 2022
Error propagation by the Monte Carlo method in geochemical calculations journal December 1976
Experimental Investigation and Modeling of Uranium (VI) Transport Under Variable Chemical Conditions journal December 1996
Reactive transport modeling of uranium 238 and radium 226 in groundwater of the Königstein uranium mine, Germany journal October 1999
Surface Complexation of Ferrous Iron and Carbonate on Ferrihydrite and the Mobilization of Arsenic journal July 2002
Use of Spectroscopic Techniques for Uranium(VI)/Montmorillonite Interaction Modeling journal March 2004
A comparison of electrostatic models for the oxide/solution interface journal March 1980
What is an optimal value of k in k-fold cross-validation in discrete Bayesian network analysis? journal June 2020
The effect of calcium on aqueous uranium(VI) speciation and adsorption to ferrihydrite and quartz journal March 2006
Enthalpy of Uranium Adsorption onto Hematite journal November 2020
Sorption mechanism of U(VI) on a reference montmorillonite: Binding to the internal and external surfaces journal January 2001
Fundamentals of the NEA Thermochemical Database and its influence over national nuclear programs on the performance assessment of deep geological repositories journal January 2019
Adsorption of NOM onto activated carbon: Electrostatic and non-electrostatic effects journal January 1997
Sorption of uranyl and arsenate on SiO2, Al2O3, TiO2 and FeOOH journal April 2014
The physical chemistry of uranium (VI) immobilization on manganese oxides journal June 2020
Sorption behavior of U(VI) on phyllite: experiments and modeling journal February 2001
Applied machine learning for predicting the lanthanide-ligand binding affinities journal August 2020
UraniumVI sorption behavior on silicate mineral mixtures journal February 2001
Soil and Aquifer Properties Combine as Predictors of Groundwater Uranium Concentrations within the Central Valley, California journal December 2020
The NEA Thermochemical Database Project:30 years of accomplishments journal July 2015
Multi-component reactive transport modeling of natural attenuation of an acid groundwater plume at a uranium mill tailings site journal November 2001