Graphene (a 2D sheet of carbon atoms arranged in a honeycombed lattice) can be used to detect individual molecules adsorbed on its surface thanks, in particular, to its high surface area of around 2600 m2/g. Moreover, the high mobility of charge carriers in graphene, which can vary between around 2500 to 40,000 cm2V–1s–1, means that it can sense a greater number of chemicals and biomarkers compared to other solid materials.

The researchers led by Ethan Minot of Oregon State University and Paul McEuen at the Kavli Institute at Cornell for Nanoscale Science determined the frequency dependent impedance between an aqueous electrolyte and a graphene sheet for a number of graphene transistor devices. Their measurements show that the graphene-electrolyte interface acts as a dissipative circuit element – a surprising result that will not only have important implications for biosensors made from the carbon material, but also for other systems like graphene-based capacitors that rely on this interface.

The Oregon-Cornell team, reporting its work in Nano Letters DOI: 10.1021/acs.nanolett.5b01788, began by applying an oscillating bias voltage between the graphene and the electrolyte, and then measured the amount of current flowing between the two. “For the most part, this interaction is like that seen in a capacitor,” explains team member Michael Crosser at Linfield College in Oregon, and lead author of the paper. “However, some of the current going in and out of the system dissipates energy. We can quantify the current that dissipates energy by measuring the time delay between the current and the driving voltage.”

Determining the thermal noise in GFETs

So how did the researchers use these measurements to determine the thermal noise in their GFETs? “We made use of the so-called fluctuation-dissipation theorem, which states that any system dissipating energy will also exhibit thermal fluctuations (Brownian motion is a famous example of this principle),” explains Minot. “We used this principle to predict the thermal fluctuations of charge, and hence the noise, across the graphene-electrolyte interface.”

“Our results might help make better graphene biosensors – for example in applications like extracellular neural recording that require noise levels below 10 microVolts”, he tells nanotechweb.org. Here, extracellular voltages cause a change in the resistance of a GFET that can then be measured.

The findings could also be important for supercapacitors made out of graphene. “An energy dissipation mechanism occurs as a graphene supercapacitor charges and discharges and we are eager to find out exactly what is happening here,” says Minot.

Alexander Balandin of the University of California – Riverside, who was not involved in this study says that quantifying the thermal noise limits in graphene devices is indeed important. “Graphene and other 2D materials are promising for chemical and biological sensor applications owing to their ultimate surface-to-volume ratio. The performance of sensors will depend on how well we can control noise in these devices.

“The specific novelty of this work has been to study graphene transistors working in aqueous electrolytes,” he adds.

You can find more research on noise in graphene devices and Biosensing in Nanotechnology.