Superlubric polycrystalline graphene interfaces

Reducing energy dissipation and wear is crucial across various fields, including mechanics, electronics, and biology. Structural superlubricity, characterized by extremely low friction due to the cancellation of lateral forces at incommensurate crystalline interfaces, offers a promising alternative to traditional lubrication methods. While superlubricity has been demonstrated in nano- and micro-sized monocrystalline samples of layered materials, scaling up to macroscopic dimensions necessitates considering polycrystalline interfaces. Polycrystalline graphene (PolyGr), composed of randomly oriented single-crystalline graphene patches separated by grain boundaries (GBs) with out-of-plane corrugation, presents a relevant model system. These GB corrugations, often characterized by chains of lattice dislocations, can significantly impact friction and wear. This study aims to understand the energy dissipation mechanisms at these elongated graphene GBs, focusing on the collective effects of topological defects under shear to determine how these effects influence friction.

Previous research has extensively explored structural superlubricity in nano- and micro-sized monocrystalline layered materials, showcasing the role of misoriented lattices or mismatched lattice constants in achieving incommensurability. Studies have demonstrated superlubricity in graphite, molybdenum disulfide (MoS2), and graphene/hexagonal boron nitride (h-BN) heterojunctions. However, the impact of polycrystalline grain boundaries, particularly their corrugation, on superlubricity in macroscopic systems remains less understood. Existing studies suggest that grain boundary corrugation can introduce significant friction and enhance wear, highlighting the need for a deeper understanding of energy dissipation mechanisms at these interfaces.

The researchers employed molecular dynamics simulations to model the frictional behavior of extended graphene GBs. The model system comprised a slider (three layers of pristine graphene) and a substrate (a PolyGr layer with two patches of different orientations and two layers of pristine graphene). The PolyGr layer contained two GBs composed of pentagon-heptagon dislocation pairs, introducing out-of-plane corrugation. Simulations were conducted with varying normal loads (up to ~2.3 GPa) and temperatures (0–300 K), applying a constant sliding velocity (5 m/s) to the top slider layer. Energy dissipation was analyzed by examining the steady-state dissipation power through the damped layers in different directions. The vertical motion of atoms with maximal root-mean-square (RMS) corrugation in each dislocation was tracked during sliding to understand the energy dissipation mechanisms. To analyze the dislocation buckling, equilibrium molecular dynamics simulations without shear were performed at room temperature to extract the Helmholtz free-energy profile and estimate transition energy barriers (TEBs). A phenomenological two-state model was developed to capture the essential features of the complex out-of-plane buckling dynamics, simplifying the analysis and prediction of frictional behavior. This model considers two competing effects: (i) the increase of the number of dislocations that undergo buckling with increasing load or temperature and (ii) decrease of the dissipated energy per buckling event under increased load or temperature. The model accounts for the temperature dependence of the buckling probability and solves a first order rate equation to account for the survival probability of dislocations not buckling.

The simulations revealed a non-monotonic relationship between friction and normal load/temperature. At low temperatures, friction initially increased with load, reaching a maximum before decreasing at higher loads. Increasing the temperature led to a monotonic decrease of friction with load, exhibiting negative differential friction coefficients. The analysis of energy dissipation demonstrated that out-of-plane atomic motion at the GBs dominated energy dissipation. Dynamic snap-through buckling of dislocations was identified as the primary mechanism of energy dissipation, with a bi-stable behavior between upward and downward protruding states. The transition energy barriers (TEBs) separating these states decreased with increasing normal load, leading to a decrease in friction at high loads. Similarly, increasing temperature initially increased friction due to thermally assisted buckling events, but at higher temperatures, the increased thermal energy led to more frequent spontaneous buckling, causing a decrease in friction. The phenomenological two-state model accurately captured both the non-monotonic friction-load and friction-temperature relationships observed in the simulations. The model successfully predicted the transition from non-monotonic to monotonic friction behavior with increasing temperature. This model suggests that frictional dissipation in corrugated GBs scales linearly with the overall GB length, implying that superlubricity remains achievable even in large-scale polycrystalline interfaces at high normal loads due to negative differential friction coefficients.

The findings highlight the importance of dynamic snap-through buckling of GB dislocations in determining the frictional properties of polycrystalline graphene interfaces. The non-monotonic friction behavior, characterized by negative differential friction coefficients at high loads, contrasts with the sublinear scaling observed in pristine incommensurate layered contacts. The phenomenological two-state model successfully accounts for the observed behavior, suggesting that collective effects between dislocations are less significant than initially anticipated. This work provides a pathway for achieving large-scale superlubricity by focusing on mitigating the excessive friction associated with individual GBs. The model's success in capturing the frictional behavior of the complex interfaces implies that it can be applied to other layered materials with corrugated GBs.

This research demonstrates that the friction in polycrystalline graphene interfaces is dominated by the dynamic snap-through buckling of grain boundary dislocations, leading to non-monotonic behavior with load and temperature. The developed two-state model successfully predicts this behavior and suggests that superlubricity can be achieved at large length scales despite the presence of grain boundaries, provided the negative differential friction coefficients at high normal loads can be utilized. Future research could explore the applicability of this model to other polycrystalline layered materials and investigate strategies to further reduce friction in such interfaces.

The simulations were based on a specific model system and interatomic potentials. While efforts were made to ensure the model's realism, the results may not fully capture the complexity of real-world polycrystalline graphene interfaces. The phenomenological two-state model assumes independent buckling events, which could be an oversimplification. Further investigation into the influence of collective effects and the applicability of the model to a wider range of GB geometries and materials is warranted.