# Delay analysis for evaluation of carrier velocity in graphene field-effect transistors

**Graphene Week 2017, Athens, Greece, 25-29 September, 2017**(2017)

[Konferensbidrag, refereegranskat]

One of the main challenges in the development of graphene field-effect transistors (GFETs) for applications in high frequency electronics is achieving high maximum frequency of oscillation (fmax), which is the power gain parameter. A promising way to achieve higher fmax is drain current saturation via saturation of the charge carrier velocity at high electric fields [1]. Therefore, accurate evaluation of the charge carrier velocity in GFETs, and its field dependence, are of importance. In this work, a method is presented that allows for the evaluation and analysis of the carrier velocity in GFETs via delay time analysis using measured cut-off frequencies. The measured cut-off frequency is inversely proportional to the total delay time, which, in GFETs on Si substrates, can be expressed as the sum of intrinsic and extrinsic delay times [2, 3, 4]. The intrinsic delay is defined by the transit time, i.e. the time taken by the charge carriers to travel across the channel, which is related to the carrier velocity. The extrinsic delays are charging delays, i.e. RC time constants required to charge and discharge the parasitic parts of the GFETs, associated with contact resistance and gate pad capacitance. In order to evaluate the extrinsic delays the contact resistance and gate pad capacitance are found. The contact resistance is found by applying a drain resistance fitting model on the measured GFET transfer characteristics. The gate pad capacitance is calculated using the corresponding delay time, which is found as difference between the total delay and the delay in the GFETs with virtual infinite gate width W (i.e. at 1/W=0), as shown in Fig. 1 [4]. The intrinsic delay time is found by subtracting the extrinsic delay from the total delay and, subsequently, used to calculate the charge carrier velocity (Fig. 2). The advantage of this method, in comparison with the previously used methods based on analysis of the GFET current-voltage characteristics, is that the carrier velocity is calculated directly, using measured cut-off frequency, independently from the carrier concentration, and, thereby, avoiding uncertainties associated with the carrier generation from traps at high fields. This allows for the accurate evaluation of the charge carrier velocity and its field dependence.

Denna post skapades 2018-01-03. Senast ändrad 2018-01-03.

CPL Pubid: 254268