*7.1.3. Sensitivities of DM CG TFET and MOSFET*

The comparisons of sensitivities of CG TFET and MOSFET are shown in **Figure 5a** and **b**. At *Nbio* <sup>=</sup> <sup>±</sup><sup>5</sup> <sup>×</sup> <sup>10</sup><sup>11</sup> cm−2 , the values are extracted for *k* = 5, 7, 10, and 12. The MOSFET with exactly equal nanogap height and length as that of CG TFET exhibits extremely inferior sensitivity as compared to CG TFET. Earlier works on dielectric-modulated MOSFET have reported similar poor sensitivity of the device as compared to TFET [38, 39].

> Not only does a shorter channel length affect the electrical characteristics of a MOSFET, but also its principle of thermionic emission by which its operation contributes to such low sensitivity. On the contrary, TFETs which operate by band-to-band tunneling perform well even when scaled. This advantage of TFET is suitable to be exploited for its biosensing capabilities.

> **Figure 5.** (a) Comparison of sensitivities of dielectric-modulated CG TFET and MOSFET biosensors at gate voltage 1.2 V,

dielectric-modulated CG TFET and MOSFET biosensors at gate voltage 1.2 V, drain voltage 1 V and *Nbio* = 5 × 10<sup>11</sup> cm−2

for dielectric constant, *k* = 5, 7, 10 and 12; (b) comparison of sensitivities of

Dielectric-Modulated TFETs as Label-Free Biosensors http://dx.doi.org/10.5772/intechopen.76000 29

The comparison of sensitivity for the four different profiles of partially filled nanogap shown in **Figure 1b**–**e** is shown in **Figure 6**. Only the decreasing and concave step profiles of biomolecule immobilization respond well to the change in dielectric constant. Contrary to this, the

**Figure 6.** Comparison of sensitivities of step profiles of partially filled nanogap in CG TFET biosensor for dielectric

**7.2. Partially filled Nanogap**

constant, *k* = 5, 7, 10 and 12.

drain voltage 1 V and *Nbio* = −5 × 10<sup>11</sup> cm−2

for dielectric constant, *k* = 5, 7, 10 and 12.


**Table 1.** Different reported works on biosensors numbered along the horizontal axis of **Figure 7**.

**Figure 5.** (a) Comparison of sensitivities of dielectric-modulated CG TFET and MOSFET biosensors at gate voltage 1.2 V, drain voltage 1 V and *Nbio* = −5 × 10<sup>11</sup> cm−2 for dielectric constant, *k* = 5, 7, 10 and 12; (b) comparison of sensitivities of dielectric-modulated CG TFET and MOSFET biosensors at gate voltage 1.2 V, drain voltage 1 V and *Nbio* = 5 × 10<sup>11</sup> cm−2 for dielectric constant, *k* = 5, 7, 10 and 12.

Not only does a shorter channel length affect the electrical characteristics of a MOSFET, but also its principle of thermionic emission by which its operation contributes to such low sensitivity. On the contrary, TFETs which operate by band-to-band tunneling perform well even when scaled. This advantage of TFET is suitable to be exploited for its biosensing capabilities.

#### **7.2. Partially filled Nanogap**

**Figure 4b** depicts the transfer characteristics of the CG TFET for *k* = 5, 7, 10, and 12 at

In **Figure 4c**, for *k* = 5, the threshold voltage decreases by 4.03% when the charge changes from

The comparisons of sensitivities of CG TFET and MOSFET are shown in **Figure 5a** and **b**.

equal nanogap height and length as that of CG TFET exhibits extremely inferior sensitivity as compared to CG TFET. Earlier works on dielectric-modulated MOSFET have reported similar

1 Conventional FET [42]

3 DM FET (LGAP = 200 nm, HGAP = 15 nm, *k* = 2.1) [36]

5 Full Gate DMTFET (LGAP = 10 nm, LGATE = 42 nm, HGAP = 5 nm, *k* = 4) [40]

7 DM FET (LGAP = 30 nm, LGATE = 100 nm, HGAP = 9 nm, *k* = 10) [39]

11 DM STS I-MOS (LGAP = 50 nm, LGATE = 120 nm, HGAP = 15 nm, *k* = 10) [55]

15 CG TFET (LGAP = 25 nm, LGATE = 40 nm, HGAP = 11 nm, *k* = 10) Proposed Work

6 Short Gate DMTFET (LGAP = 10 nm, LGATE = 20 nm, HGAP = 5 nm, k = 4)

9 DM PNPN TFET (LGAP = 30 nm, LGATE = 100 nm, HGAP = 9 nm, *k* = 10) 10 DM PNPN TFET (LGAP = 75 nm, LGATE = 250 nm, HGAP = 9 nm, *k* = 10)

12 SiGe Source DM PNPN TFET, Ge composition = 0% (LGAP = 15 nm, LGATE = 100 nm, HGAP = 9 nm, *k* = 2.1)

13 SiGe Source DM PNPN TFET, Ge composition = 10% (LGAP = 15 nm, LGATE = 100 nm, HGAP = 9 nm, *k* = 2.1)

14 SiGe Source DM PNPN TFET, Ge composition = 20% (LGAP = 15 nm, LGATE = 100 nm, HGAP = 9 nm, *k* = 2.1)

16 CG TFET (decreasing step) (LGAP = 25 nm, LGATE = 40 nm, HGAP = 11 nm,

17 CG TFET (concave step) (LGAP = 25 nm, LGATE = 40 nm, HGAP = 11 nm, *k* = 10)

**Table 1.** Different reported works on biosensors numbered along the horizontal axis of **Figure 7**.

*k* = 10)

8 DM FET (LGAP = 75 nm, LGATE = 250 nm, HGAP = 9 nm, *k* = 10)

of negatively charged biomolecules as evident from **Table 1**.

poor sensitivity of the device as compared to TFET [38, 39].

4 DM FET (LGAP = 100 nm, HGAP = 15 nm, *k* = 2.1)

*7.1.3. Sensitivities of DM CG TFET and MOSFET*

28 Design, Simulation and Construction of Field Effect Transistors

, whereas for k = 12, the drop is 2.68%.

. The on current in case of positive charged biomolecules is higher than that in case

, the values are extracted for *k* = 5, 7, 10, and 12. The MOSFET with exactly

[41]

**Biosensors Reference**

*Nbio* = 10<sup>11</sup> cm−2

neutral to 1012 cm−2

At *Nbio* <sup>=</sup> <sup>±</sup><sup>5</sup> <sup>×</sup> <sup>10</sup><sup>11</sup> cm−2

**(horizontal axis of Figure 7)**

2 Nanowire TFET

**Sl. No.**

The comparison of sensitivity for the four different profiles of partially filled nanogap shown in **Figure 1b**–**e** is shown in **Figure 6**. Only the decreasing and concave step profiles of biomolecule immobilization respond well to the change in dielectric constant. Contrary to this, the

**Figure 6.** Comparison of sensitivities of step profiles of partially filled nanogap in CG TFET biosensor for dielectric constant, *k* = 5, 7, 10 and 12.

increasing and convex step profiles demonstrate poor sensitivity. The reason for this is the proximity of the highest step with the source-channel tunnel junction. In case of decreasing and concave step profiles, the higher steps are present near to the tunnel junction as shown in **Figure 1b**, **d** respectively. As the value of *k* increases, the gate-channel coupling increases in the region of higher steps closer to the tunnel junction. So, the response of the biosensor is better than that of the increasing and convex step profiles where the higher steps are located away from the source-channel tunnel junction.

**8. Conclusion**

**Author details**

Rupam Goswami<sup>1</sup>

**References**

This chapter has presented an overview on Tunnel Field Effect Transistors (TFETs) as dielectric-modulated biosensors. Tunnel Field Effect Transistors have emerged as one of the most significant devices for low power applications due to their ability to withstand the effects of scaling. With the interests gathering around FET-based biosensors, research on TFETs as biosensors has recently brought new focus. This chapter has discussed the various aspects of dielectric-modulated TFET as biosensor with emphasis on the design and development through simulation analyses. Practical implications of the biosensors are presented. A Circular Gate TFET as a dielectric-modulated biosensor is presented and analyzed at lesser channel length. The CG TFET is observed to offer an impressive sensitivity as compared to other biosensors. The different challenges in implementing a TFET-based dielectric-modulated biosensor are varied, ranging from the problems of steric hindrance, fabrication issues and uncertainty of probe placement. Simulation and modeling may enable one to predict the various effects.

Dielectric-Modulated TFETs as Label-Free Biosensors http://dx.doi.org/10.5772/intechopen.76000 31

Appropriate physics-based models are necessary to validate the results on TCAD tool.

[1] Sarangi S, Bhushan S, Santra A, Dubey S, Jit S, Tiwari PK. A rigorous simulation based study of gate misalignment effects in gate engineered double-gate (DG) MOSFETs. Superlattices and Microstructures. 2013;**60**:263-279. DOI: 10.1016/j.spmi.2013.05.009 [2] Choi WY, Park B-G, Lee JD, Liu T-JK. Tunneling field-effect transistors (TFETs) with subthreshold swing (SS) less than 60 mV/dec. IEEE Electron Device Letters. 2007;**28**(8):743-

[3] Royer CL, Mayer F. Exhaustive experimental study of tunnel field effect transistors (TFETs): From materials to architecture. In: 10th International Conference on Ultimate Integration of Silicon; March 18-20, 2009. Aachen: IEEE; 2009. pp. 53-56. DOI: 10.1109/

[4] International Technology Roadmap for Semiconductors. http://public.itrs.net/. 2015

\* and Brinda Bhowmick2 \*Address all correspondence to: rupam.goswamifet@kiit.ac.in

2 National Institute of Technology Silchar, Silchar, India

745. DOI: 10.1109/led.2007.901273

Edition. [Accessed: January 06, 2018]

ulis.2009.4897537

1 Kalinga Institute of Industrial Technology, Bhubaneswar, India

#### **7.3. Status map of biosensors**

There are a number of important simulated and modeled works reported on dielectric-modulated TFET and FET. This section presents a map of the sensitivities of such biosensors proposed till date along with sensitivity of the proposed CG TFET.

Although the status map of **Figure 7** mentions the maximum or best sensitivities of each work, yet the architectural specifications under which the biosensors have been reported vary from one to another, and hence, drawing comparisons among them through **Figure 7** is not justified. However, there are a few conclusions that can be derived from the status map. Dielectric-modulated TFETs are more sensitive to the presence of biomolecules than MOSFETs due to the difference in their current transport mechanisms. In MOSFETs, sensitivities reduce at lesser channel lengths. The CG TFET, with a channel length of 40 nm, shows significant sensitivity; a fully filled nanogap in CG TFET for *k* = 10 has sensitivity closer to that of DM PNPN TFET for *k* = 10 possessing a channel length of 250 nm and nanogap length of 75 nm. However, the partially filled nanogaps (decreasing and concave step profiles) have lesser sensitivities than the fully filled case as explained in Section 7.2.

**Figure 7.** Sensitivities of FET-based biosensors of reported works and those of CG TFET. The sensitivities are extracted from the published works, and due to the possible tolerances in extraction, the vertical axis is named as 'approximate sensitivity'. The various biosensors are referred by using serial numbers from 1 to 17, the details of which are listed in **Table 1**.
