Abstract
In this paper we propose an analytical method to calculate the band structures of graphene-like nanoribbons of the armchair type with arbitrary line defects or uniaxial strains. The model is based on the tight-binding model and the standing wave assumption for the armchair nanoribbons. It gives accurate band results for large supercell systems. Within this method, we analyze different local states near the line defect in the graphene and boron-nitride nanoribbons. We also derive the analytical expression for the local states in a semi-infinite graphene nanoribbon by this method and the transfer matrix technique. The criteria condition for the local states in the semi-infinite nanoribbons is also given.
Introduction
Graphene nanoribbons (GNRs) are quasi-one-dimensional materials with electronic properties that depend heavily on their edge geometry. While zigzag nanoribbons (zGNRs) are characterized by localized edge states, armchair nanoribbons (aGNRs) exhibit a variety of behaviors—ranging from metallic to semiconducting—based on their width. However, real-world applications often involve imperfections such as line defects or physical strain, which significantly alter these electronic states.
Modeling these systems using standard 2D tight-binding (TB) models or first-principles density functional theory (DFT) becomes computationally prohibitive as the ribbon width or the size of the defect supercell increases. Researchers at Chongqing University have addressed this challenge by developing a streamlined analytical approach that reduces the complex 2D problem into a manageable 1D model, providing a "brief and useful" tool for large-scale nanostructure analysis.
Methodology: The Standing Wave Assumption
The core innovation of this work is the standing wave assumption. By assuming that the transverse wavefunction of electrons across the ribbon's width takes the form of a standing wave, the researchers mapped the 2D lattice onto a 1D chain of sub-unit cells.
This dimensionality reduction is a significant breakthrough in computational efficiency. Instead of solving a massive $(2MN) \times (2MN)$ Hamiltonian matrix (where $N$ is width and $M$ is length), this method solves multiple much smaller $4M \times 4M$ matrices. The model employs the $p_z$ orbital tight-binding approximation, where defects and strain are simulated by modulating the "hopping integrals" ($t$)—parameters representing electron movement between adjacent atoms. It also utilizes a transfer matrix technique to derive analytical expressions for localized states in semi-infinite ribbons.
Key Insights and Theoretical Analysis
The 1D analytical approach revealed several critical insights into the behavior of aGNRs:
- Internal "Edge" States: A line defect within an armchair ribbon can effectively act as an internal zigzag boundary. This creates localized "edge-like" states near the defect, even in a ribbon that lacks physical zigzag edges.
- Localization Criteria: The authors derived a specific mathematical condition, $p > (N+1)/3$, for the existence of localized states in semi-infinite ribbons, where $p$ represents the standing-wave mode and $N$ is the width.
- Boron-Nitride Asymmetry: In boron-nitride (BN) ribbons, the different on-site energies of Boron and Nitrogen atoms lead to asymmetric local states, where electron density "leans" toward one side of the defect.
Results and Validation
The model's accuracy was validated against established 2D-TB calculations and existing literature. For standard aGNRs (e.g., $N=6, 8$), the band structures produced by the 1D method perfectly matched known results.
The model also successfully reproduced the periodic oscillations of the band gap observed under uniaxial strain, matching results from more complex perturbation theories. Furthermore, the visualization of electron density distributions confirmed clear localization near defects, consistent with 2D-TB overhead views.
Conclusion and Future Outlook
While highly efficient, the model does have limitations. For ribbons with an odd width ($N$), the 1D method generates "extra" bands that must be manually deduced to maintain consistency with the 2D model. Additionally, as a tight-binding simplification, it may overlook subtle electronic effects captured only by high-level first-principles calculations.
Despite these constraints, this 1D approach offers a powerful framework for studying graphene-like materials. Future work may extend this methodology to more complex defect geometries and other 2D materials beyond graphene and boron-nitride.
Summary updated