Tom Frömbgen
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Tom Frömbgen has a total of 43 co-authors. Here are the top co-authors by number of shared publications:
- Barbara Kirchner
University of Bonn Beringstraße 4
16 shared publications - Jürgen Dölz
Institute for Numerical Simulation, University of Bonn, Bonn, Germany
5 shared publications - Daniel Blankschtein
Department of Chemical Engineering
4 shared publications - Rahul Prasanna Misra
Department of Chemical Engineering
4 shared publications - Shuang Luo
Department of Chemical Engineering
4 shared publications - Thierry Tassaing
Institut Des Sciences Moléculaires (ISM), UMR5255 CNRS, Bordeaux INP Université De Bordeaux Talence Cedex France
4 shared publications
Latest Publications
- 13 July 2026
- Published by American Chemical Society (ACS) in ACS Nano
- Vol. 20 (29) , 20810-20822
Electronic Polarization Governs Structure–Transport Coupling of Angstrom-Scale Confined Water
Angstrom-scale confinement fundamentally controls the structure and dynamics of water, leading to behaviors that are very different from those observed in the bulk. Using many-body polarizable force fields combined with Grand Canonical Molecular Dynamics (GCMD) simulations, we investigate the behavior of water confined between multilayer graphene channels spanning 5.5 to 20 Å. We find that explicit modeling of the electronic polarization of graphene is essential: it suppresses the artificially solid-like ordering that arises with conventional nonpolarizable pairwise-additive descriptions based on Lennard-Jones interactions and restores liquid-like behavior even at the smallest channel spacings. The equilibrium densities obtained using GCMD simulations exhibit pronounced oscillations with channel spacing, reflecting discrete monolayer, bilayer, and trilayer packing regimes, which in turn give rise to strongly nonmonotonic variations in lateral diffusion. Structure factor analysis reveals that diffusion minima coincide with ordered, highly correlated configurations, whereas diffusion maxima arise from sparsely populated water configurations with lower density, where lateral correlations are also weaker. By integrating the hydration pressure, we construct the confinement-free-energy landscape and show that these oscillations emerge from a balance between energetic stabilization within compact layers and entropic penalties associated with restricted configurational freedom. As additional molecular layers form, confinement-induced structural, dynamical, and thermodynamic signatures progressively weaken and converge smoothly toward bulk-like behavior. Together, these results provide a unified microscopic picture of the behavior of water under extreme confinement, with direct implications for the design of ultrathin membranes, selective transport devices, and next-generation 2D nanofluidic platforms.
- 13 July 2026
- Published by American Chemical Society (ACS) in ACS Nano
- Vol. 20 (29) , 20810-20822
Electronic Polarization Governs Structure–Transport Coupling of Angstrom-Scale Confined Water
Angstrom-scale confinement fundamentally controls the structure and dynamics of water, leading to behaviors that are very different from those observed in the bulk. Using many-body polarizable force fields combined with Grand Canonical Molecular Dynamics (GCMD) simulations, we investigate the behavior of water confined between multilayer graphene channels spanning 5.5 to 20 Å. We find that explicit modeling of the electronic polarization of graphene is essential: it suppresses the artificially solid-like ordering that arises with conventional nonpolarizable pairwise-additive descriptions based on Lennard-Jones interactions and restores liquid-like behavior even at the smallest channel spacings. The equilibrium densities obtained using GCMD simulations exhibit pronounced oscillations with channel spacing, reflecting discrete monolayer, bilayer, and trilayer packing regimes, which in turn give rise to strongly nonmonotonic variations in lateral diffusion. Structure factor analysis reveals that diffusion minima coincide with ordered, highly correlated configurations, whereas diffusion maxima arise from sparsely populated water configurations with lower density, where lateral correlations are also weaker. By integrating the hydration pressure, we construct the confinement-free-energy landscape and show that these oscillations emerge from a balance between energetic stabilization within compact layers and entropic penalties associated with restricted configurational freedom. As additional molecular layers form, confinement-induced structural, dynamical, and thermodynamic signatures progressively weaken and converge smoothly toward bulk-like behavior. Together, these results provide a unified microscopic picture of the behavior of water under extreme confinement, with direct implications for the design of ultrathin membranes, selective transport devices, and next-generation 2D nanofluidic platforms.
- Open Access
- 5 April 2026
- Published by American Chemical Society (ACS) in ChemRxiv
Hydrogen-Bond Networks and Proton Transfer in Choline Geranate: Insights from IR Spectroscopy
Choline–geranate (CAGE) systems have attracted considerable interest due to their remarkable performance in biomedical and drug-delivery applications, yet their microscopic structure remains incompletely understood. In particular, the nature of the interaction between geranate (GE) and geranic acid (AGE) in CAGE formulations with excess acid has been the subject of ongoing debates, with spectroscopic evidence suggesting proton transfer but not resolving whether the proton is symmetrically shared or dynamically exchanged between the carboxylate groups. In this work, we combine experimental attenuated total reflectance Fourier-transform infrared (ATR–FTIR) spectroscopy with theoretical vibrational spectra derived from ab initio molecular dynamics (AIMD) simulations to elucidate the hydrogen-bond network and local structural organization in CAGE at 1:1 and 1:2 choline-to-acid ratios. The simulations are performed at a representative water content of 8 wt%, while the experimental IR spectra are recorded over a range of water contents to assess the effect of hydration. Structural analysis of the AIMD trajectories reveals a dynamic hydrogen-bond network in which proton transfer between AGE and GE occurs intermittently, consistent with a rattling proton rather than a symmetrically shared configuration. The calculated infrared (IR) spectra reflect this behavior through the coexistence of vibrational signatures characteristic of both carboxylate and carboxylic acid groups. Experimental IR spectra corroborate these findings, showing persistent carboxylate and carboxylic acid features across compositions and water contents, with no evidence for a delocalized proton-bound dicarboxylate species. Together, the combined experimental and theoretical results provide a consistent structural picture of CAGE, reconciling previously reported nuclear magnetic resonance observations of proton transfer with the vibrational signatures observed in IR spectroscopy and clarifying the role of proton dynamics in these systems.
- 5 April 2026
- Published by American Chemical Society (ACS) in ChemRxiv
Open AccessHydrogen-Bond Networks and Proton Transfer in Choline Geranate: Insights from IR Spectroscopy
Choline–geranate (CAGE) systems have attracted considerable interest due to their remarkable performance in biomedical and drug-delivery applications, yet their microscopic structure remains incompletely understood. In particular, the nature of the interaction between geranate (GE) and geranic acid (AGE) in CAGE formulations with excess acid has been the subject of ongoing debates, with spectroscopic evidence suggesting proton transfer but not resolving whether the proton is symmetrically shared or dynamically exchanged between the carboxylate groups. In this work, we combine experimental attenuated total reflectance Fourier-transform infrared (ATR–FTIR) spectroscopy with theoretical vibrational spectra derived from ab initio molecular dynamics (AIMD) simulations to elucidate the hydrogen-bond network and local structural organization in CAGE at 1:1 and 1:2 choline-to-acid ratios. The simulations are performed at a representative water content of 8 wt%, while the experimental IR spectra are recorded over a range of water contents to assess the effect of hydration. Structural analysis of the AIMD trajectories reveals a dynamic hydrogen-bond network in which proton transfer between AGE and GE occurs intermittently, consistent with a rattling proton rather than a symmetrically shared configuration. The calculated infrared (IR) spectra reflect this behavior through the coexistence of vibrational signatures characteristic of both carboxylate and carboxylic acid groups. Experimental IR spectra corroborate these findings, showing persistent carboxylate and carboxylic acid features across compositions and water contents, with no evidence for a delocalized proton-bound dicarboxylate species. Together, the combined experimental and theoretical results provide a consistent structural picture of CAGE, reconciling previously reported nuclear magnetic resonance observations of proton transfer with the vibrational signatures observed in IR spectroscopy and clarifying the role of proton dynamics in these systems.
- Top Cited PapersOpen Access
- 4 March 2026
- Published by American Chemical Society (ACS) in Chemical Reviews
- Vol. 126 (7) , 4189-4236
Uncertainty Quantification for In Silico Chemistry
The rapid growth of worldwide computing power has transformed in silico chemistry into a discipline that is integrated into the daily work of many chemists. Nowadays, researchers find it increasingly straightforward to predict a wide range of molecular properties and chemical processes at reasonable computational cost. The resulting abundance of data generated by quantum chemistry, molecular dynamics simulations, and chemical machine learning naturally raises questions about accuracy, precision, and reliability as well as the systematic treatment of errors and uncertainties. Addressing these questions through rigorous mathematical frameworks is at the heart of uncertainty quantification. In the past years, the incorporation of uncertainty quantification into in silico chemistry has gained attention, motivated by its ability to provide deeper insights into chemical phenomena. In this review, we establish a common language for uncertainty quantification with respect to in silico chemistry, introduce the key mathematical formalisms, and survey the growing body of work that applies uncertainty quantification across different areas of in silico chemistry.
- 4 March 2026
- Published by American Chemical Society (ACS) in Chemical Reviews
- Vol. 126 (7) , 4189-4236
Top Cited PapersOpen AccessUncertainty Quantification for In Silico Chemistry
The rapid growth of worldwide computing power has transformed in silico chemistry into a discipline that is integrated into the daily work of many chemists. Nowadays, researchers find it increasingly straightforward to predict a wide range of molecular properties and chemical processes at reasonable computational cost. The resulting abundance of data generated by quantum chemistry, molecular dynamics simulations, and chemical machine learning naturally raises questions about accuracy, precision, and reliability as well as the systematic treatment of errors and uncertainties. Addressing these questions through rigorous mathematical frameworks is at the heart of uncertainty quantification. In the past years, the incorporation of uncertainty quantification into in silico chemistry has gained attention, motivated by its ability to provide deeper insights into chemical phenomena. In this review, we establish a common language for uncertainty quantification with respect to in silico chemistry, introduce the key mathematical formalisms, and survey the growing body of work that applies uncertainty quantification across different areas of in silico chemistry.
- Open Access
- 15 January 2026
- Published by American Chemical Society (ACS) in ChemRxiv
Uncertainty Quantification for In Silico Chemistry
The rapid growth of worldwide computing power has transformed in silico chemistry into a discipline that is integrated into the daily work of many chemists. Nowadays, researchers find it increasingly straightforward to predict a wide range of molecular properties and chemi- cal processes at reasonable computational cost. The resulting abundance of data, generated by quantum chemistry, molecular dynamics sim- ulations, and chemical machine learning natu- rally raises questions about accuracy, precision, and reliability, as well as the systematic treat- ment of errors and uncertainties. Addressing these questions through rigorous mathematical frameworks is at the heart of Uncertainty Quan- tification. In the past years, the incorpora- tion of uncertainty quantification into in silico chemistry has gained attraction, motivated by its ability to provide deeper insights into chem- ical phenomena. In this review, we establish a common language for uncertainty quantifica- tion with respect to in silico chemistry, intro- duce the key mathematical formalisms, and sur- vey the growing body of work that applies un- certainty quantification across different areas of in silico chemistry.
- 15 January 2026
- Published by American Chemical Society (ACS) in ChemRxiv
Open AccessUncertainty Quantification for In Silico Chemistry
The rapid growth of worldwide computing power has transformed in silico chemistry into a discipline that is integrated into the daily work of many chemists. Nowadays, researchers find it increasingly straightforward to predict a wide range of molecular properties and chemi- cal processes at reasonable computational cost. The resulting abundance of data, generated by quantum chemistry, molecular dynamics sim- ulations, and chemical machine learning natu- rally raises questions about accuracy, precision, and reliability, as well as the systematic treat- ment of errors and uncertainties. Addressing these questions through rigorous mathematical frameworks is at the heart of Uncertainty Quan- tification. In the past years, the incorpora- tion of uncertainty quantification into in silico chemistry has gained attraction, motivated by its ability to provide deeper insights into chem- ical phenomena. In this review, we establish a common language for uncertainty quantifica- tion with respect to in silico chemistry, intro- duce the key mathematical formalisms, and sur- vey the growing body of work that applies un- certainty quantification across different areas of in silico chemistry.
- 9 January 2026
- Published by AIP Publishing in The Journal of Chemical Physics
- Vol. 164 (2)
A multifidelity Monte Carlo approach for simulating the diffusion coefficient of water. I. Forward problem
A multifidelity Monte Carlo framework for molecular dynamics simulations of the diffusion coefficient of liquid water is presented. The model hierarchy is constructed based on the size of the simulation box, taking advantage of the well-known size effects that simulations of the diffusion coefficient suffer from. Water is represented as a rigid three-point model, and uncertainties in the non-bonded force field parameters, i.e., the Lennard-Jones parameters and partial charges, are propagated forward through the simulations. A large-scale study exploring the capabilities of the multifidelity Monte Carlo framework is conducted, involving six computational models, three different calibrations, cubic and tetragonal simulation box shapes, and two sets of perturbed force field parameters. Results reveal that significant computational speed-ups of up to one order of magnitude can be achieved if successive models have small differences in correlations to the high-fidelity model. It is also found that the combination of high- and low-fidelity models provides a computational speed-up compared to exclusively relying on a single high-fidelity model in almost all cases.
- 9 January 2026
- Published by AIP Publishing in The Journal of Chemical Physics
- Vol. 164 (2)
A multifidelity Monte Carlo approach for simulating the diffusion coefficient of water. I. Forward problem
A multifidelity Monte Carlo framework for molecular dynamics simulations of the diffusion coefficient of liquid water is presented. The model hierarchy is constructed based on the size of the simulation box, taking advantage of the well-known size effects that simulations of the diffusion coefficient suffer from. Water is represented as a rigid three-point model, and uncertainties in the non-bonded force field parameters, i.e., the Lennard-Jones parameters and partial charges, are propagated forward through the simulations. A large-scale study exploring the capabilities of the multifidelity Monte Carlo framework is conducted, involving six computational models, three different calibrations, cubic and tetragonal simulation box shapes, and two sets of perturbed force field parameters. Results reveal that significant computational speed-ups of up to one order of magnitude can be achieved if successive models have small differences in correlations to the high-fidelity model. It is also found that the combination of high- and low-fidelity models provides a computational speed-up compared to exclusively relying on a single high-fidelity model in almost all cases.
- Open Access
- 1 January 2026
- Published by Royal Society of Chemistry (RSC) in Physical Chemistry Chemical Physics
- Vol. 28 (28) , 17605-17614
Hydrogen-bond networks and proton transfer in choline geranate: insights from IR spectroscopy
Experiment and simulation reveal the molecular origin of proton transfer in choline–geranate, showing that transient proton exchange occurs while preserving distinct carboxylic acid and carboxylate species.
- 1 January 2026
- Published by Royal Society of Chemistry (RSC) in Physical Chemistry Chemical Physics
- Vol. 28 (28) , 17605-17614
Open AccessHydrogen-bond networks and proton transfer in choline geranate: insights from IR spectroscopy
Experiment and simulation reveal the molecular origin of proton transfer in choline–geranate, showing that transient proton exchange occurs while preserving distinct carboxylic acid and carboxylate species.
- Open Access
- 17 December 2025
- Published by American Chemical Society (ACS) in ChemRxiv
A Multifidelity Monte Carlo Approach for Simulating the Diffusion Coefficient of Water. I. Forward Problem
A multi-fidelity Monte Carlo framework for molecular dynamics simulations of the diffusion coefficient of liquid water is presented. The model hierarchy is constructed based on the size of the simulation box, taking advantage of the well-known size effects that simulations of the diffusion coefficient suffer from. Water is represented as a rigid three-point model, and uncertainties in the non-bonded force field parameters i.e., the Lennard-Jones parameters and partial charges, are propagated forward through the simulations. A large-scale study exploring the capabilities of the multifidelity Monte Carlo framework is conducted, involving six computational models, three different calibrations, cubic and tetragonal simulation box shapes, and two sets of perturbed force field parameters. Results reveal that significant computational speed-ups of up to one order of magnitude can be achieved if successive models have small differences in correlations to the high-fidelity model. It is also found that the combination of high- and low-fidelity models provides a computational speed-up compared to exclusively relying on a single high-fidelity model in almost all cases.
- 17 December 2025
- Published by American Chemical Society (ACS) in ChemRxiv
Open AccessA Multifidelity Monte Carlo Approach for Simulating the Diffusion Coefficient of Water. I. Forward Problem
A multi-fidelity Monte Carlo framework for molecular dynamics simulations of the diffusion coefficient of liquid water is presented. The model hierarchy is constructed based on the size of the simulation box, taking advantage of the well-known size effects that simulations of the diffusion coefficient suffer from. Water is represented as a rigid three-point model, and uncertainties in the non-bonded force field parameters i.e., the Lennard-Jones parameters and partial charges, are propagated forward through the simulations. A large-scale study exploring the capabilities of the multifidelity Monte Carlo framework is conducted, involving six computational models, three different calibrations, cubic and tetragonal simulation box shapes, and two sets of perturbed force field parameters. Results reveal that significant computational speed-ups of up to one order of magnitude can be achieved if successive models have small differences in correlations to the high-fidelity model. It is also found that the combination of high- and low-fidelity models provides a computational speed-up compared to exclusively relying on a single high-fidelity model in almost all cases.
- 15 August 2025
- Published by American Chemical Society (ACS) in ACS Applied Materials & Interfaces
- Vol. 17 (34) , 48919-48931
Beyond Pairwise Interactions: How Interfacial Polarization Modulates Water Flow in Graphene Nanochannels
Although incorporating many-body polarization effects is essential for accurately modeling water transport under nanoconfinement, the use of conventional pairwise additive force fields fails to account for polarization-induced charge redistribution at the interface. Using Grand Canonical Molecular Dynamics (GCMD) simulations, we demonstrate that explicitly incorporating many-body polarization effects to model graphene-water interactions as well as graphene flexibility is essential for accurately predicting water interfacial structure and dynamics under nanoconfinement. For narrow channels of 8 Å spacing, graphene polarization is shown to disrupt the confinement-induced ordering of water predicted using Lennard-Jones-based nonpolarizable force fields, resulting in (i) increased out-of-plane orientations of the water molecules, (ii) weakening of hydrogen bonding and hindering of water crystallization, and (iii) faster force decorrelation, resulting in lower interfacial friction. Further, the water densities obtained from the GCMD simulations and accounting for polarization effects increased monotonically with an increase in the channel spacing, whereas the interfacial friction was found to reach a minimum value at 10 Å spacing before attaining a constant value at larger channel spacings. Decomposing friction into static and dynamic contributions reveals that molecular memory effects enhance friction under extreme confinement, while bulk-like interfacial behavior emerges at larger spacings. Importantly, we predict a slip length of ∼200 Å using the polarizable model at a larger channel spacing, in close agreement with previous experimental measurements. These findings demonstrate that many-body polarization, graphene flexibility, and accurate water density predictions are all essential for capturing nanoscale water transport. By providing a predictive framework that links interfacial structure and dynamics, this study advances our fundamental understanding of confined fluid behavior, which is essential for membrane-based applications, including seawater desalination, molecular separation, and energy harvesting.
- 15 August 2025
- Published by American Chemical Society (ACS) in ACS Applied Materials & Interfaces
- Vol. 17 (34) , 48919-48931
Beyond Pairwise Interactions: How Interfacial Polarization Modulates Water Flow in Graphene Nanochannels
Although incorporating many-body polarization effects is essential for accurately modeling water transport under nanoconfinement, the use of conventional pairwise additive force fields fails to account for polarization-induced charge redistribution at the interface. Using Grand Canonical Molecular Dynamics (GCMD) simulations, we demonstrate that explicitly incorporating many-body polarization effects to model graphene-water interactions as well as graphene flexibility is essential for accurately predicting water interfacial structure and dynamics under nanoconfinement. For narrow channels of 8 Å spacing, graphene polarization is shown to disrupt the confinement-induced ordering of water predicted using Lennard-Jones-based nonpolarizable force fields, resulting in (i) increased out-of-plane orientations of the water molecules, (ii) weakening of hydrogen bonding and hindering of water crystallization, and (iii) faster force decorrelation, resulting in lower interfacial friction. Further, the water densities obtained from the GCMD simulations and accounting for polarization effects increased monotonically with an increase in the channel spacing, whereas the interfacial friction was found to reach a minimum value at 10 Å spacing before attaining a constant value at larger channel spacings. Decomposing friction into static and dynamic contributions reveals that molecular memory effects enhance friction under extreme confinement, while bulk-like interfacial behavior emerges at larger spacings. Importantly, we predict a slip length of ∼200 Å using the polarizable model at a larger channel spacing, in close agreement with previous experimental measurements. These findings demonstrate that many-body polarization, graphene flexibility, and accurate water density predictions are all essential for capturing nanoscale water transport. By providing a predictive framework that links interfacial structure and dynamics, this study advances our fundamental understanding of confined fluid behavior, which is essential for membrane-based applications, including seawater desalination, molecular separation, and energy harvesting.
- 4 August 2025
- Published by American Chemical Society (ACS) in The Journal of Physical Chemistry B
Formulation of Polarizable Force Fields to Model Simple Ionic Liquid/Graphite Interfaces
Force fields that are specifically parametrized for performing molecular dynamics simulations of liquid/solid interfaces are in high demand. Such interfaces are broadly present in nature and applied in, for example, batteries, membranes, or catalysts, and therefore, an accurate modeling is required, including the explicit treatment of electronic polarization effects. In molecular dynamics simulations, the liquid/solid interactions are most often modeled by combining existing force fields for both phases using mixing rules. Ionic liquids, a very versatile class of compounds that can be utilized in the aforementioned applications, exert strong electric fields on the surrounding molecules, and therefore, are best described using polarizable force fields. In this work, the derivation of polarizable force fields for the simple, model ionic liquid ([C1C1Im][BF4]) at a graphite interface is presented. In doing so, we explicitly parametrize the ionic liquid/graphite interactions instead of relying on mixing rules. Implementing the derived force fields, we observe that the explicit parametrization significantly affects the structure of the interface, resulting in the formation of a sharply defined contact layer and about 30 to 50 pm reduced ion-graphite distances when compared to mixing rules-based force fields. Upon calculating the work of adhesion of the ionic liquid on graphite, we show that the liquid/solid interactions are dominated by dispersive interactions while induction effects only play a minor role. Additionally, we find that using mixing rules strongly underpredicts the overall work of adhesion, highlighting the demand for and justifying the computational efforts of explicitly parametrizing the ionic liquid/graphite interactions.
- 4 August 2025
- Published by American Chemical Society (ACS) in The Journal of Physical Chemistry B
Formulation of Polarizable Force Fields to Model Simple Ionic Liquid/Graphite Interfaces
Force fields that are specifically parametrized for performing molecular dynamics simulations of liquid/solid interfaces are in high demand. Such interfaces are broadly present in nature and applied in, for example, batteries, membranes, or catalysts, and therefore, an accurate modeling is required, including the explicit treatment of electronic polarization effects. In molecular dynamics simulations, the liquid/solid interactions are most often modeled by combining existing force fields for both phases using mixing rules. Ionic liquids, a very versatile class of compounds that can be utilized in the aforementioned applications, exert strong electric fields on the surrounding molecules, and therefore, are best described using polarizable force fields. In this work, the derivation of polarizable force fields for the simple, model ionic liquid ([C1C1Im][BF4]) at a graphite interface is presented. In doing so, we explicitly parametrize the ionic liquid/graphite interactions instead of relying on mixing rules. Implementing the derived force fields, we observe that the explicit parametrization significantly affects the structure of the interface, resulting in the formation of a sharply defined contact layer and about 30 to 50 pm reduced ion-graphite distances when compared to mixing rules-based force fields. Upon calculating the work of adhesion of the ionic liquid on graphite, we show that the liquid/solid interactions are dominated by dispersive interactions while induction effects only play a minor role. Additionally, we find that using mixing rules strongly underpredicts the overall work of adhesion, highlighting the demand for and justifying the computational efforts of explicitly parametrizing the ionic liquid/graphite interactions.
- 25 July 2025
- Published by American Chemical Society (ACS) in ACS Nano
- Vol. 19 (30) , 27500-27514
Explicit Modeling of Electronic Polarization Reveals Water-Mediated Screening of Ion Adsorption at Hexagonal Boron Nitride Interfaces
Understanding ion adsorption at hexagonal boron nitride (hBN)/water interfaces is essential for applications in nanofluidics, membrane separations, and electrochemical sensing. Water, being polar, and salt ions, being charged, can generate electric fields that strongly polarize a contacting hBN surface. Further, with hBN being heteropolar, there is a possibility of a coupling between the permanent and induced charge distributions, whose implications on the ion adsorption phenomenon remain unclear. In this study, we develop all-atomistic polarizable force fields by incorporating electronic polarization effects via the classical Drude oscillator model to accurately model hBN-water and hBN-ion interactions. By carrying out classical molecular dynamics (MD) simulations, we compute free energy profiles for the adsorption of various ions comprising the well-known Hofmeister series at the hBN/water interface. Our explicit modeling of the electronic polarization of hBN accurately predicts ion-specific effects, which agree well with ab initio MD simulation results. In contrast, implicit modeling of hBN-ion and hBN-water polarization energies using traditional nonpolarizable force fields overestimates adsorption free energies, leading to incorrect predictions of ion adsorption. Furthermore, our findings indicate that interfacial water molecules significantly attenuate hBN-ion polarization interactions, a screening effect that modulates the ion adsorption behavior. Our study demonstrates the need to explicitly model many-body polarization effects to accurately describe ion adsorption thermodynamics at hBN and other 2D material interfaces, offering insights for the design of 2D-material nanodevices.
- 25 July 2025
- Published by American Chemical Society (ACS) in ACS Nano
- Vol. 19 (30) , 27500-27514
Explicit Modeling of Electronic Polarization Reveals Water-Mediated Screening of Ion Adsorption at Hexagonal Boron Nitride Interfaces
Understanding ion adsorption at hexagonal boron nitride (hBN)/water interfaces is essential for applications in nanofluidics, membrane separations, and electrochemical sensing. Water, being polar, and salt ions, being charged, can generate electric fields that strongly polarize a contacting hBN surface. Further, with hBN being heteropolar, there is a possibility of a coupling between the permanent and induced charge distributions, whose implications on the ion adsorption phenomenon remain unclear. In this study, we develop all-atomistic polarizable force fields by incorporating electronic polarization effects via the classical Drude oscillator model to accurately model hBN-water and hBN-ion interactions. By carrying out classical molecular dynamics (MD) simulations, we compute free energy profiles for the adsorption of various ions comprising the well-known Hofmeister series at the hBN/water interface. Our explicit modeling of the electronic polarization of hBN accurately predicts ion-specific effects, which agree well with ab initio MD simulation results. In contrast, implicit modeling of hBN-ion and hBN-water polarization energies using traditional nonpolarizable force fields overestimates adsorption free energies, leading to incorrect predictions of ion adsorption. Furthermore, our findings indicate that interfacial water molecules significantly attenuate hBN-ion polarization interactions, a screening effect that modulates the ion adsorption behavior. Our study demonstrates the need to explicitly model many-body polarization effects to accurately describe ion adsorption thermodynamics at hBN and other 2D material interfaces, offering insights for the design of 2D-material nanodevices.