[Paper Review] Understanding and predicting trends in adsorption energetics on monolayer transition metal dichalcogenides

Brian Lee, Jameela Fatheema+2 more|arXiv (Cornell University)|Oct 9, 2024

2D Materials and ApplicationsMaterials Science3 citations

TL;DR

This study develops predictive models for adsorption energetics of transition metal adatoms on monolayer transition metal dichalcogenides (TMDs), using electronic and atomic structure descriptors such as d-band center and COHP analysis. It demonstrates that adsorption energy trends across MoS₂, MoSe₂, WS₂, and WSe₂ can be accurately predicted using a compressed sensing approach with ionization energy, electronegativity, and atomic radius descriptors, enabling rational design of low-energy, stable resistive switching devices.

ABSTRACT

Two-dimensional (2D) transition metal dichalcogenides (TMDs) have recently been shown to demonstrate non-volatile resistive switching (NVRS), offering significant advantages such as high-density integration and low energy consumption due to their atomic-scale thinness. In this study, we focus on the adsorption and desorption of metal adatoms, which can modulate the electrical resistivity by several orders of magnitude. We develop material-based relationships of the adsorption energy with electronic and atomic structure descriptors by examining the effects of various transition-metal adsorbates on the surface of TMDs. Our results reveal that adsorption energies of transition metals exhibit consistent trends across different TMDs (MoS$_2$, MoSe$_2$, WS$_2$, WSe$_2$) and can be explained using simple descriptors of the atomic and electronic structure. We propose several models to describe this adsorption process, providing a deeper understanding of a crucial step in the resistive switching mechanism based on formation and dissolution of point defects. Finally, we connect our computed adsorption energies to the switching energy. These findings will help guide rational materials selection for the development of NVRS devices using 2D TMDs.

Motivation & Objective

• To understand and predict the adsorption energetics of transition metal adatoms on monolayer TMDs (MoS₂, MoSe₂, WS₂, WSe₂) as a proxy for resistive switching energy.
• To identify material descriptors that correlate with adsorption energy trends across different TMDs and transition metals.
• To develop a predictive model for adsorption energy using atom-based descriptors without requiring full DFT calculations.
• To link computed adsorption energies to switching energy in resistive memory devices for materials selection.

Proposed method

• Density functional theory (DFT) calculations were used to compute adsorption energies of 30 transition metal adatoms on pre-existing anion vacancies in four TMDs.
• The d-band center relative to the Fermi level was used as a key electronic descriptor to explain adsorption energy trends.
• Crystal Orbital Hamilton Populations (COHP) analysis was performed to quantify bonding and anti-bonding interactions between adatoms and S atoms.
• Compressed sensing (SISSO) was applied to derive predictive models using compound descriptors: ionization energy, electronegativity, and atomic radius of adsorbate and adsorption site.
• Adsorption energy trends were correlated with periodic table group position, revealing systematic variation across early, middle, and late transition metals.
• The predictive model was validated against DFT data and shown to capture most of the variation in adsorption energies without requiring additional DFT runs.

Figure 1: (a) Periodic table highlighting the elements studied in this work. Adatoms investigated for adsorption on transition metal dichalcogenides (TMDs) are marked in green. The transition metal (M) and chalcogen (X) elements in the MX 2 configuration are highlighted in purple and yellow, respect

Experimental results

Research questions

• RQ1 How do adsorption energies of transition metal adatoms vary across different monolayer TMDs (MoS₂, MoSe₂, WS₂, WSe₂)?
• RQ2 What electronic and atomic structure descriptors can explain the observed trends in adsorption energy across TMDs?
• RQ3 Can a machine learning-inspired compressed sensing approach accurately predict absolute adsorption energies using only basic atomic properties?
• RQ4 How do the d-band center and COHP analysis inform the nature of bonding at the adsorption site?
• RQ5 What is the relationship between adsorption energy and the switching energy in resistive memory devices?

Key findings

• Adsorption energies vary systematically with the periodic group of the transition metal: early transition metals (Hf, Zr, Ti) exhibit the strongest adsorption (e.g., -7.78 eV for W), while late transition metals (Cu, Au, Ag, Zn, Cd, Hg) show the weakest (e.g., -2.96 eV for Cu).
• The d-band center relative to the Fermi level is a strong descriptor that captures most of the trend in adsorption energies across different TMDs and adatoms.
• COHP analysis reveals that bonding interactions are strongest between early transition metals and sulfur atoms, while anti-bonding character dominates for late transition metals.
• A compressed sensing (SISSO) model using ionization energy, electronegativity, and atomic radius successfully predicts absolute adsorption energies without requiring DFT calculations.
• Copper (Cu) and nickel (Ni) show low adsorption energies (-2.96 eV and -4.01 eV), making them promising for low-energy, reversible resistive switching; tungsten (W) and molybdenum (Mo) have high adsorption energies (-7.78 eV), favoring stable, non-volatile states.
• The study provides a framework for screening defects in high-throughput databases and accelerating materials discovery for resistive switching, catalysis, and sensing.

Figure 2: Relaxed atomic structures of MoS 2 with (a) a single sulfur vacancy, (b) gold adsorbate, (c) platinum adsorbate, (d) zirconium adsorbate. The distances between the nearest neighbor molybdenum atoms in each structure are indicated.