Mar 28, 2014
Y-function technique sidesteps contact resistance
Contact resistance (Rc) can significantly obscure recorded electron mobilities in field-effect devices, especially those made from 2D semiconducting materials. The Rc effect thus makes it difficult to accurately determine a material’s real mobility, which is important in assessing how useful it will be for future electronics applications. A team of researchers at the University of Texas at Austin in the US has now put forward a simple “Y-function” technique to overcome this problem that allows them to calculate true transistor mobilities and Rc values without having to make complicated device structures.
Contact resistance often limits the performance of nanoelectronic devices and this is particularly true for ultrathin materials, such as the semiconducting 2D dichalcogenides (TMDCs). These up-and-coming technologically important compounds are easily processed semiconducting monolayers that might be used to make circuits for low-power electronics, flexible displays, high-performance sensors and even flexible electronics that can be coated onto a wide variety of surfaces.
The team, led by Deji Akinwande, employed a Y-function (or Ghibaudo) method in their study on MoS2 – one of the most well known TMDCs. The Y-function is defined as the current divided by the square root of the transconductance in a device. It allows researchers to cancel out Rc effects from measured currents, and thus accurately calculate transistor mobilities. What is more, the technique allows other essential device parameters, such as the maximum Rc and threshold voltage, to be determined.
To back up their results, the Texas team also fabricated MoS2 device structures and analysed these using the more conventional transfer length method (TLM).
TLM backs up Y-function technique
“From our results, we found that the extracted mobility from the Y-function method degrades with transistor channel length – possibly because of local variations in the MoS2 crystal,” explained Akinwande. “When we correct the mobility for each channel length, we observe good agreement between the Rc extracted from the Y-function method and that from the TLM analyses on device structures. In other words, the well established TLM technique, which is a much more complicated method in itself, corroborates the results from the simpler Y-function method.”
He adds that the Y-function method is accurate even in the presence of gate-dependent or Schottky-barrier Rc, frequently encountered in semiconducting TMDC transistor devices. “The main advantage of the technique, compared with more complex methods such as the TLM, is that it also works for practical two-terminal devices,” he told nanotechweb.org. “The TLM method, on the other hand, requires many terminals and hence much longer crystals sizes – something that is challenging to achieve at the moment since TMDC crystal flakes are typically quite small.”
The researchers developed their Y-function method mainly to study mobilities and Rc in 2D semiconductors. They now hope to extend the technique to extract saturation velocities in 2D TMDCs. Saturation velocity is a critical parameter that determines ultimate speeds and frequencies achievable in semiconductors.
The current work is detailed in Applied Physics Letters.
About the author
Belle Dumé is contributing editor at nanotechweb.org