Phase transformation kinetics in MoS2 governed by S-S repulsive interactions and defect-interface compatibility

This paper demonstrates that the kinetic arrest of the metastable T' phase in monolayer $\text{MoS}_2$ is driven by repulsive S-S interactions and that phase transformation is governed by the local compatibility between defects and moving interfaces rather than global defect concentration.

The Gist

The "Stubborn Shape-Shifter": Why MoS2 Refuses to Change

Imagine you have a massive collection of LEGO bricks. Most of the time, you build them into a flat, stable floor (this is the H-phase of MoS2). But sometimes, you build them into a slightly more complex, "wobbly" structure (the T’ phase).

In nature, the flat floor is much more stable. If you nudged the wobbly structure, it should instantly snap back into the flat floor. But in the world of MoS2, this "wobbly" structure is incredibly stubborn. Even though it "wants" to change, it stays stuck in its wobbly shape for months.

Scientists have long wondered: Why is it so hard for this material to change its shape? This paper provides the answer.

---

1. The "Social Distancing" Problem (S–S Repulsion)

To change from the wobbly shape to the flat shape, the atoms have to move and rearrange themselves. Think of the sulfur atoms in the material like people at a crowded party.

As the material tries to transform, these sulfur atoms are forced to move into positions where they are uncomfortably close to one another. It’s like being stuck in an elevator with someone who is breathing too loudly in your ear—it’s awkward, high-energy, and everyone wants to get away.

In scientific terms, the sulfur atoms experience "S–S repulsion." Because they hate being that close, they create a "speed bump" (an energy barrier) that stops the transformation in its tracks. Every time the material tries to take a step toward the stable shape, it hits one of these "socially awkward" atomic bumps, and the whole process grinds to a halt.

2. The "Broken Tool" Paradox (The Defect Mystery)

Usually, when something is stuck, we try to "grease the wheels" to make it move. In materials science, we do this by adding defects (like missing atoms, or "sulfur vacancies"). Think of these missing atoms as empty spaces in a crowded room that allow people to move around more easily.

Common wisdom says: "If you add more empty spaces, the transformation should happen faster."

But this paper reveals a twist!

The researchers found that while adding empty spaces can help in some parts of the material, it fails at the most important part: the "Front Line" (the interface where the change is actually happening).

3. The "Unstable Helper" (Defect-Interface Compatibility)

Imagine you are trying to push a heavy heavy door open. You hire a helper to stand at the door and push with you.

However, in MoS2, the "helper" (the sulfur vacancy) is incredibly unstable at the most important part of the door (the ZZ-Mo|- interface). As soon as the helper arrives at the door to help you push, they realize they don't like the environment there, so they immediately run away into the background.

Because the "helpers" (defects) refuse to stay at the front line where the work is being done, they can't help push the material into its new shape. The "Front Line" remains clean, empty, and—most importantly—stuck.

---

The Big Picture: A New Rule for Engineering

This paper changes how we think about building future technology.

Previously, scientists thought that if you wanted to change a material's properties, you just needed to add a certain amount of "stuff" (defects) to it. This paper says: "It’s not about how much stuff you have; it’s about whether that stuff stays where the action is."

If we want to control these 2D materials for next-generation electronics or sensors, we shouldn't just throw defects at them like confetti. Instead, we need to design "helpers" that are specifically compatible with the moving front line of the material. We need to make sure our "pushers" actually stay at the door!

Technical Summary

Technical Summary: Phase Transformation Kinetics in $\text{MoS}_2$

1. Problem Statement

In monolayer $\text{MoS}_2$, there is a significant thermodynamic driving force for the metastable metallic T’ phase to transform into the stable semiconducting H phase. Despite this, the T’ phase exhibits remarkable kinetic persistence, remaining stable at room temperature for months. While previous research suggested that sulfur vacancies ($V_S$) facilitate this transformation, a unified atomistic mechanism linking defect thermodynamics, local atomic interactions, and macroscopic transformation kinetics has remained elusive. Specifically, it was unclear whether defects act as universal catalysts or if their efficacy is limited by their interaction with moving interfaces.

2. Methodology

The study employs a multi-scale computational approach to bridge atomic-scale interactions with macroscopic kinetic observations:

• First-Principles Calculations (DFT): Used to investigate the bonding nature of various coherent grain boundaries (CGBs) and to quantify the energy barriers of sulfur atom hops.
• Crystal Orbital Hamilton Population (COHP) Analysis: Applied to quantify the electronic origin of repulsive interactions (antibonding character) at the interfaces.
• Machine Learning-Accelerated Molecular Dynamics (ML-MD): A high-accuracy Machine Learning Force Field (MLFF) was trained using the DeepMD-kit framework. The MLFF was trained on a massive dataset (37,611 structures) covering diverse configurations (strained, defective, interfacial) to enable large-scale, long-time simulations that are computationally inaccessible via pure DFT.
• Kinetic Modeling: Theoretical derivation of kink-mediated interface propagation rates to confirm whether the process is limited by kink nucleation or migration.

3. Key Contributions

• Identification of the Kinetic Bottleneck: The study identifies S–S repulsive interactions as the primary source of energy barriers during both the nucleation of the H phase and the propagation of the H/T’ grain boundary.
• The "Local Compatibility" Paradigm: The authors propose a new principle: the efficacy of a defect in driving phase transformation is not determined by its global concentration, but by its local thermodynamic compatibility with the advancing interface.
• Re-evaluation of Defect-Mediated Transformation: The work provides a counterintuitive finding that sulfur vacancies do not necessarily accelerate transformation if they are thermodynamically unstable at the most prevalent interface.

4. Results

• S–S Repulsion at Interfaces: DFT and COHP analysis reveal that in S-rich interfaces, sulfur atoms are forced into close proximity ($\sim 1.84\text{ \AA}$), creating strong antibonding (repulsive) interactions. Even after structural relaxation, the S–S distances remain significantly shorter than equilibrium, imposing a high energetic cost.
• Nucleation Kinetics: Nucleation is highly sensitive to the sequence of atomic hops. Every time a sulfur atom hop creates a new S–S pair, a sharp energy spike occurs. While sulfur vacancies reduce the nucleation barrier (from $\sim 5.4\text{ eV}$ to $\sim 2.1\text{ eV}$ for two vacancies) by providing "space," this effect is limited by the tendency of vacancies to cluster.
• Interface Propagation: The study examined several CGBs and found that the ZZ-Mo|– interface is the most thermodynamically stable and the dominant configuration observed in nanostructure simulations. However, this interface is "S-depleted."
• The Vacancy Paradox:
  • At S-rich interfaces (ZZ-S|+), sulfur vacancies are stable and effectively "ride" the interface, significantly accelerating propagation.
  • At the most stable interface (ZZ-Mo|–), sulfur vacancies are thermodynamically unstable. Instead of assisting the interface, vacancies migrate away from the front into the T’ phase. Consequently, the ZZ-Mo|– interface remains "defect-free" and moves extremely slowly due to the high energy barrier ($\sim 1.08\text{ eV}$) required to nucleate a kink via S–S pair formation.
• Direct Nanostructure Observation: Large-scale MD simulations of triangular islands and ribbons confirmed that H-phase growth exclusively proceeds via the ZZ-Mo|– configuration, explaining why the transformation is so slow in practice.

5. Significance

This work provides a fundamental understanding of why certain 2D materials exhibit unexpected kinetic stability. By shifting the focus from global defect concentration to interface-specific defect compatibility, the study offers a new design paradigm for materials engineering. This allows for the precise control of structural transitions in 2D heterostructures by designing interfaces that either trap or repel specific defects to tune the speed of phase transformations.