Graphene as a subnanometre trans-electrode membrane

Abstract

Isolated, atomically thin conducting membranes of graphite, called graphene, have recently been the subject of intense research with the hope that practical applications in fields ranging from electronics to energy science will emerge. The atomic thinness, stability and electrical sensitivity of graphene motivated us to investigate the potential use of graphene membranes and graphene nanopores to characterize single molecules of DNA in ionic solution. Here we show that when immersed in an ionic solution, a layer of graphene becomes a new electrochemical structure that we call a trans-electrode. The trans-electrode's unique properties are the consequence of the atomic-scale proximity of its two opposing liquid-solid interfaces together with graphene's well known in-plane conductivity. We show that several trans-electrode properties are revealed by ionic conductance measurements on a graphene membrane that separates two aqueous ionic solutions. Although our membranes are only one to two atomic layers thick, we find they are remarkable ionic insulators with a very small stable conductance that depends on the ion species in solution. Electrical measurements on graphene membranes in which a single nanopore has been drilled show that the membrane's effective insulating thickness is less than one nanometre. This small effective thickness makes graphene an ideal substrate for very high resolution, high throughput nanopore-based single-molecule detectors. The sensitivity of graphene's in-plane electronic conductivity to its immediate surface environment and trans-membrane solution potentials will offer new insights into atomic surface processes and sensor development opportunities.

Figure 1. Schematic of our experiments

A graphene membrane was mounted over a 200×200 nm² aperture in SiN_x suspended across a Si frame (not to scale). The membrane separates two ionic solutions in contact with Ag/AgCl electrodes. Inset: A graphene membrane into which a nanopore has been drilled.

Figure 2. Trans-electrode I–V curves

Results for an as-grown graphene membrane (dashed line) and a membrane with a 8 nm pore (solid line). The ionic conductance of the pore is quantitatively in agreement with the modeling presented in the text. Applying bias voltages in excess of ~250 mV gradually degraded the insulating properties of the membranes. Insets, top: transmission electron micrograph (TEM) of a mounted graphene membrane; bottom: TEM images of the 8 nm pore.

Figure 3. Graphene nanopore conductance

Closed circles are experimental results with a 1 M KCl solution of conductivity σ = 11 Sm⁻¹. The solid curve shows the modeled conductance of a 0.6 nm thick insulating membrane and is the best fit to the experimentally measured conductances. Error bars represent standard deviation of 4 diameter measurements along different nanopore axes. Modeled conductances for a 2 nm thick membrane (dotted line) and a 10 nm thick membrane (dash-dot line) are presented for comparison.

Figure 4. Average nanopore current blockades vs. blockade duration during DNA translocation

DNA (16 μg/ml) was electrophoretically driven through a 5 nm diameter graphene pore by an applied voltage bias of 160 mV. The graphene membrane separated two fluid cells containing unbuffered 3M KCl solutions, pH 10.4. Insets show typical current-time traces for two translocation events sampled from among those pointed to by the arrows. The hyperbolic curve corresponds to freely translocating events at a fixed ecd. Encircled events are delayed by graphene DNA interactions.

Figure 5. Geometric Resolution

Modeled nanopore conductance as the abrupt diameter decrease of a model molecule (inset) translocates through a 2.4 nm pore. The attainable resolution for two membranes of different insulating thicknesses is assumed to be achieved when the measured current through the nanopore changes from 75% to 25% of the maximum blockade change that would occur as the model molecule translocates through the nanopore.