Researchers at the Swiss Federal Institute of Technology in Lausanne have shown proof of principle for detecting DNA going through a nanopore using graphene nanoribbon sensors rather than ionic current measurements. Meanwhile, scientists at the University of Pennsylvania in Philadelphia have improved methods for fabricating graphene nanoribbon-nanopore devices.
Up until now, most nanopore sequencing approaches have used ion current changes that happen when DNA travels through the pore to determine the order of the bases. However, this type of measurement requires the nanopores to be electrically isolated from each other and might be difficult to scale up to large numbers. A number of groups have thus been working on alternative detection methods, including tunneling electrodes, nanogaps, nanowire field-effect transistors, and graphene nanoribbons.
In a paper published last month in Nature Nanotechnology, the Swiss group, led by Aleksandra Radenovic, studied nanopores drilled through a graphene nanoribbon layer and an underlying silicone nitride membrane. They simultaneously measured the ion current through the pore and the current through the graphene nanoribbon to record DNA translocation and found the two signals to be correlated somewhat.
"This is the first time that translocation events of single DNA molecules have been detected by electrical means other than the ionic current itself, using a graphene-based device integrated on a solid-state nanopore," the authors wrote.
According to Radenovic, the signal from the nanoribbons is local to each pore, so measurements could be parallelized in the future. Ion current, on the other hand, is measured by bulk electrodes and cannot discriminate between signals from different pores unless they are isolated from each other, she explained.
To make their device, Radenovic and her colleagues transferred a graphene monolayer onto a 20-nanometer-thick SiNx membrane and defined graphene nanoribbons using electron beam lithography and oxygen reactive ion etching. They fabricated electrical contacts, added a layer of Al2O3 on top, and drilled a nanopore using transmission electron microscopy.
They then measured the current through the graphene nanoribbon and the ionic current through the pore at different salt conditions, both in the presence and absence of DNA. To study DNA translocation, they used a circular 2.7-kilobase plasmid and looked for corresponding changes in current.
In one DNA experiment, the researchers recorded 125 drops in ion current, for example, of which 70 were correlated with spikes in the graphene current.
According to Radenovic, the ultimate goal is to measure DNA translocation using only the graphene current, but the ion current was a good control for their experiments because it is a well-characterized signal.
Cees Dekker, a professor at the Kavli Institute of Nanoscience at Delft University of Technology in the Netherlands, said the work is "impressive and important."
"It is a beautiful example of combining the special electronic properties of graphene for probing DNA with nanopore detection," he told In Sequence in an email message, noting that it appears to be the first example of using graphene for detection rather than just as a membrane.
He said the signal-to-noise ratio for the graphene signal seems to be better than for the ion current signal, though not by much at this stage. "It will be interesting to see the full potential of this technique, as various groups, including my own group, are pursuing [it] now."
Marija Drndic, a professor at the University of Pennsylvania whose group has also been working on DNA sequencing with graphene nanoribbons, said that this method could enable "faster and cheaper DNA sequencing in the future, and therefore, the potential validation of this approach is an immense step forward."
But while the Swiss group's article is "a glimpse into a working device," further validation is needed to demonstrate reliable device measurements, she said. For example, in the paper, there is "not always a clear correlation (or anti-correlation) between ionic signal and ribbon signal," which could be due to damage to the graphene. Also, the ion current signal sometimes goes up and sometimes goes down, she said, and the device appears to work only for a brief time.
Radenovic acknowledged that for DNA sequencing, the signal-to-noise ratio for the graphene sensor needs to be improved, and the speed of the DNA must be reduced. Her group has already found a way to slow down the DNA, she noted, but cannot provide further details at this time.
Another goal of her group is to remove the thick silicone nitride layer underneath the graphene nanoribbon. This would decrease the time required to drill the nanopores, leading to less damage in the ribbons from the electron beam. "One would like to preserve the best properties of the graphene ribbon, and by removing this nitride below, in the future, it will be hopefully less invasive on the ribbon," she said.
In the meantime, Drndic's team has come up with a way to prevent such damage in the first place.
In a paper published in ACS Nano last month, the researchers fabricated nanopores in silicon nitride membranes with graphene nanoribbons on top. By using the transmission electron microscope in scanning mode, they reduced exposure of the nanoribbons to the electron beam before and during drilling, thus preventing damage to the ribbons.
As a result, the nanoribbons remain highly conducting after drilling, which should result in better DNA translocation signals, Drndic told In Sequence.