First-principles vs. semi-empirical modeling of global and local electronic transport properties of graphene nanopore-based sensors for DNA sequencing

TL;DR

This study compares first-principles and semi-empirical quantum transport models to understand electronic current changes in graphene nanopores caused by DNA nucleobases, informing sensor design for DNA sequencing.

Contribution

It provides a detailed comparison of NEGF+DFT and semi-empirical methods, highlighting discrepancies crucial for designing graphene-based DNA sensors.

Findings

Significant conductance change when nucleobase is inserted

Bias voltage enhances current contrast

Discrepancies between computational methods

Abstract

Using first-principles quantum transport simulations, based on the nonequilibrium Green function formalism combined with density functional theory (NEGF+DFT), we examine changes in the total and local electronic currents within the plane of graphene nanoribbon with zigzag edges (ZGNR) hosting a nanopore which are induced by inserting a DNA nucleobase into the pore. We find a sizable change of the zero-bias conductance of two-terminal ZGNR + nanopore device after the nucleobase is placed into the most probable position (according to molecular dynamics trajectories) inside the nanopore of a small diameter \mbox{$D=1.2$ nm}. Although such effect decreases as the nanopore size is increased to \mbox{$D=1.7$ nm}, the contrast between currents in ZGNR + nanopore and ZGNR + nanopore + nucleobase systems can be enhanced by applying a small bias voltage $V_b \lesssim 0.1$ V. This is explained microscopically as being due to DNA nucleobase-induced modification of spatial profile of local current density around the edges of ZGNR. We repeat the same analysis using NEGF combined with self-consistent charge density functional tight-binding (NEGF+SCC-DFTB) or self-consistent extended H\"{u}ckel (NEGF+SC-EH) semi-empirical methodologies. The large discrepancy we find between the results obtained from NEGF+DFT vs. those obtained from NEGF+SCC-DFTB or NEGF+SC-EH approaches could be of great importance when selecting proper computational algorithms for {\em in silico} design of optimal nanoelectronic sensors for rapid DNA sequencing.