Biosensors are defined as any device that senses and transmits information about a biological element. The importance of biosensors for health and environment monitoring increase in the past decades, due the global warming, pollution and new biological pathogens. In order to achieve better sensing and to scale down the device dimensions and other advantages (1), the use of field effect transistors (FETs) as biosensors have been studied recently.

The goal of this work is to analyze the effect of gate to drain underlapping on n-type Tunnel-FET (nTFET) devices, filled with different dielectric permittivity material (k) in order to simulate the bio element materials. The gate length (L_G, where L_G = 50nm – L_U) was studied changing the underlap (L_U) for a range of 0nm (self-aligned) to 15 nm of underlap, with step of 5 nm. The total channel length (L_SD, where L_SD = 50nm +L_U) was evaluated fixing the L_G=50 nm and adding the underlap part. In all cases, different dielectric permittivity (k) was considered in these analysis as the bio element. The structure of the Double Gate DG-nTFET biosensor is shown in Fig.1. The parameters used for simulation are: silicon film thickness, t_Si = 10 nm, equivalent oxide thickness EOT = 1 nm, channel length L_SD = 50 nm, thickness of the bio element t_Bio = 10 nm and the length of the drain and source regions L_D = L_S = 100 nm. The gate (Titanium Nitride) work function = 4.7 eV, source doping N_A =1.10²⁰ cm^-3, drain doping N_D= 1.10²⁰ cm^-3 and channel doping N_D = 1.10⁷ cm^-3_. The models used are nonlocal band-to-band tunneling (BTBT), Shockley-Read-Hall (SRH) recombination and band-gap narrowing (BGN) model.

The transfer curve of nTFETs with different underlap values are shown in Fig.2. It is observed that with the increment of L_U the TFET ambipolar current (drain to channel tunneling) decreases, as already reported in (2). However, if the k values increase, the ambipolar currents increase as can be seen in Fig.3. This effect can be explained in Fig.4, where it is possible to see that the tunneling length becomes thinner when K raise due to improvement of the tunneling between drain and channel.

In order to analyze a better configuration for a biosensor device, a sensitivity parameter defined as Sensivity = I_D(K=n)/I_D(k=1), where n is the value of the corresponding k, was used for comparing all different type of configurations studied in this paper. The sensitivity increases with the k value and with different L_U is shown in Fig.5. The best sensitivity is obtained for high K and high underlap region.

In the case of the device with L_G fixed at 50 nm, the Fig.6 shows that the sensitivity increases with k and the highest sensitivity was obtained for lower underlap (5 nm). It can be explained analyzing the band diagram of the Fig.7, where it is possible to notice that the tunnel length decreases for lower underlap region.

An investigation of the impact of biosensor permittivity on a DG-nTFET ambipolar current was done. The bio elements detection using the underlap region showed high sensitivity, of orders of magnitude for different L_U and k values. The highest sensitivity for an nTFET biosensor was observed for a L_u = 15 nm (higher L_u), L_SD = 50nm (smaller L_SD) and higher k bio element, in the range studied in this paper.