Researchers at the University of Delaware have conducted high-performance computer modeling to investigate a new approach for ultrafast DNA sequencing based on tiny holes, called nanopores, drilled into a sheet of graphene.
A tiny hole a few nanometers in diameter is drilled into a sheet of graphene and DNA is threaded through that nanopore. Then, a current of ions flowing vertically through the pore or an electronic current flowing transversely through the graphene is used to detect the presence of different DNA bases within the nanopore. “Since graphene is only one atom thick, the nanopore through which DNA is threaded has contact with only a single DNA base,” Nikolic said.
In 2010, three experimental teams—led by Jene Golovchenko of Harvard, Cees Dekker of Delft and Drndić—demonstrated DNA detection using nanopores in large-area graphene. However, Nikolic said, the process moved too quickly for the existing electronics to detect single DNA bases.
The new device concept proposed by the UD researchers uses graphene nanoribbons—thin strips of graphene that are less than 10 nanometers wide—with a nanopore drilled in their interior. Chemists, engineers, materials scientists and physicists have devised various methods over the past three years to fabricate nanoribbons with a specific zigzag pattern of carbon atoms along their edges, Nikolic said. Nanoribbons could enable fast and low-cost (less than $1,000) DNA sequencing, he said, because of the quantum-mechanically generated electronic currents that flow along those edges.
Simulated graphene nanopore design
We study two-terminal devices for DNA sequencing that consist of a metallic graphene nanoribbon with zigzag edges (ZGNR) and a nanopore in its interior through which the DNA molecule is translocated. Using the nonequilibrium Green functions combined with density functional theory, we demonstrate that each of the four DNA nucleobases inserted into the nanopore, whose edge carbon atoms are passivated by either hydrogen or nitrogen, will lead to a unique change in the device conductance. Unlike other recent biosensors based on transverse electronic transport through translocated DNA, which utilize small (of the order of pA) tunneling current across a nanogap or a nanopore yielding a poor signal-to-noise ratio, our device concept relies on the fact that in ZGNRs local current density is peaked around the edges so that drilling a nanopore away from the edges will not diminish the conductance. Inserting a nucleobase into the nanopore affects the charge density in the surrounding area, thereby modulating edge conduction currents whose magnitude is of the order of microampere at bias voltage 0.1 V. The proposed biosensors are not limited to ZGNRs and they could be realized with other nanowires supporting transverse edge currents, such as chiral GNRs or wires made of two-dimensional topological insulators.
“We used the knowledge acquired from several years of theoretical and computational research on the electronic transport in graphene to increase the magnitude of the detection current in our biosensor by a thousand to million times when compared to other recently considered devices,” Nikolic said. “Two years ago, scientists would have told me our device was impossible, but there are so many people working on graphene that nothing is impossible anymore.
“Every time physicists think something is impossible, materials scientists or chemists come to the rescue—and vice versa.”
Nikolic said he and postdoctoral researcher Kamal Saha have employed their home-grown massively parallel computational codes to simulate the operation of the proposed nanoelectronic biosensor from first principles, using the supercomputer Chimera that UD acquired with support from a National Science Foundation grant.
“This project has to run on 500-1,000 processors for several months continuously,” he said. “We couldn’t have done it without UD Chimera becoming fully operational in early 2011.”
The Chimera Supercomputer was used for the simulation of ultrafast DNA graphene nanopore sequencing