Dielectric Modulated Negative Capacitance

Heterojunction TFET as Biosensor: Proposal and

Analysis
Varun Mishra
graphic era deemed university
Lucky Agarwal (  [email protected] )
University of Allahabad
Chandni Tiwari
graphic era deemed university
Vikas Rathi
graphic era deemed university

Research Article

Keywords: Biosensors, tunnel FETs, charge-plasma, sensitivity, negative-capacitance

Posted Date: October 4th, 2023

DOI: https://doi.org/10.21203/rs.3.rs-3365037/v1

License:   This work is licensed under a Creative Commons Attribution 4.0 International License.
Read Full License

Additional Declarations: No competing interests reported.

 Dielectric Modulated Negative Capacitance
Heterojunction TFET as Biosensor: Proposal
and Analysis
Varun Mishra1,a, Lucky Agarwal2,b, Chandni Tiwari3,a, Vikas Rathi4,a
aGraphic Era (Deemed to be University), Dehradun, Uttarakhand, India
bUniversity of Allahabad, Prayagraj, Uttar Pradesh, India
[email protected]¸ [email protected]*, [email protected], [email protected],

Abstract— In this article, a label-free biosensor with a and charge density [5]. Moreover, dielectrically modified FET
single cavity that uses a negative capacitance biosensors having nanocavity are promising candidates owing
heterojunction charge-plasma-based tunnel FET (NC-HJ- to its accuracy, flexibility, and compatibility, which can
CP-TFET) is presented and examined. To increase ON-
state current, ferroelectric material (Si:HfO2) has been produce useful information about bio-analytes. The
added to a stack with a gate insulator and low energy performance parameters of FET biosensors are selectivity and
bandgap material (Si0.5Ge0.5). In terms of biosensing sensitivity which is known as sensing metrics. Sensitivity
properties, comparisons have been made between NC-HJ- detects the biomolecule’s presence or absence, whereas
CP-TFET and Si-based NC-CP-TFET. The different bio- selectivity differentiates among different bio-analytes.
analytes employed in this work are Streptavidin (K = 2.1), Sensitivity is defined by ratio of ON state and OFF-state
APTES (K = 3.57), Bacteriophage (K = 6.3), Protein (K = 8),
and Gelatin (K = 12). Benchmarking is done to compare current, change in threshold voltage, transconductance (gm),
prospective biosensors to literature that has already been and ratio of transconductance and drain current (g m/ID)[6]. The
reported. The maximum ON-state current sensitivity (SION), FETs are employed as biosensor by using DM and gating
transconductance-based sensitivity (Sgm), ION/IOFF, and effect. In gating effect, a layer of receptors re used to modified
subthreshold swing (SS) obtained are 2×108, 4×108, gate metal that reflects variation in the channel energy band,
2.3×1011, and 17.9 mV/decade, respectively, for NC-HJ-CP- and thus the currents, upon conjugation of the charged
TFET with permittivity of K = 12 with fully filled nanogap of
neutral bio-analyte.. biomolecules. In DM, nanogaps are formed by etching a part
of the gate dielectric material. The permittivity of the cavity
Index Terms— Biosensors, tunnel FETs, charge-plasma, changes once biomolecules are trapped in the cavity and the
sensitivity, negative-capacitance. same is reflected in the drive current.

I. INTRODUCTION
Because of recent advances in microbiology, accurate
identification and analysis of nano-biomolecules is becoming
a necessity. Identifying and investigating different proteins,
like DNA, amino acids, biotin-streptavidin, and various types
of bacteria and viruses, is critical for understanding any
unusual behavior within living cells. Quick testing and
identification are crucial steps in preventing humankind and (a)
other earthly living beings from exterminating due to
unidentified biohazards. In this regard, utilization of Field
Effect Transistors (FETs) as biosensors leads to novel research
directions in design, performance metrics studies, and
analytical modelling [1]. Biosensors with FET configuration
have gained tremendous attention owing to its smaller size,
sensing without labels, low cost, lighter weight and
compatibility with CMOS technology [2]–[4]. Furthermore,
the methodology of Dielectric Modulation (DM) is widely (b)
utilized in biosensing because it allows the use of nano
cavities for identifying bio-analytes by observing the change Fig. 1(a) Schematic of NC-JH-CP-TFET, (b) Process flow of NC-HJ-CP-
in electrical quantities caused by changes in dielectric constant TFET.
 Tunnel field effect transistors (TFETs) has shown to enhance the tunnelling of charge carriers, while reducing
outstanding results in past few years for obtaining the energy bandgap. In order to obtain best resluts, mole
subthreshold swing below 60 mV/decade, low OFF-state composition (x) of SixGe1-x is chosen to be 0.5 [23]. However,
current, and better performance below 1 V [7]–[11]. drain regions are kept identical for the proposed biosensors in
However, improving ON state current of TFET is one of the order to maintain low OFF-state current. The total channel
major challenges which still needs considerable research length is of 22 nm, out of which 11 nm is cavity length (Lc),
efforts. There are several reports on attaining high ON state whereas cavity thickness (Tc) is of 4 nm. However, the
current such as use of low band gap source material [8], [12], remaining 11 nm of channel is a ferroelectric material acting
[13]; inclusion of ferroelectric material as gate-insulator [14]– as a gate insulator, whereas channel thickness is of 5 nm (TSi).
The nanogap region is created by etching out the ferroelectric
[17], and so on. TFETs have bypassed the MOSFET as a
insulator existing in the particular area. The insulation gap
biosensing device owing to its better sensitivity, low power
(Tgap) is of 2 nm in between gate and source electrode. The
consumption and better integration in system on chip [18]–
channel, drain, and source regions are intrinsically doped with
[21]. Furthermore, higher sensitivity is not reported for lower concentration of 1015 cm-3. The doping is induced in source
operating voltages for common bio-analytes (APTES, Biotin, and drain regions with the help of suitable electrodes by
and Streptavidin). To acquire higher sensitivity and selectivity employing charge plasma technique that is valid only when
of all types of charged as well as neutral biomolecules, the the channel thickness is smaller than Debye length. Debye
optimization of TFETs and enhanced device analysis are length depicts the average distance over which mobile charge
needed. carriers are separated [24]. To make p-type and n-type doping,
In the proposed architecture for creating trivalent (P+) and the workfunction of source and drain electrodes are set to 5.93
pentavalent (N+) doped regions utilizes the charge plasma eV and 3.9 eV, respectively. Si:HfO2 is used as ferroelectric
(CP) approach. This approach relies on two fundamental material in stack with SiO2 (interface layer) with a thickness of
requirements based on scientific principles.[22]: 4 (tFE) nm and 1 nm (tox), respectively. Interface layer prevents
(1) Debye length and silicon channel thickness: Channel interdiffusion between Si:HfO2 and channel [25] and it also
thickness (TSi) should be less than Debye length (LD) to ensure behaves as immobilizing layer for biomolecules [26]. Table 1
that charge carrier distribution within T Si is controlled by presents the parameters and their values of the ferroelectric
induced charge carriers. LD defines the length over which material employed during simulation.
distribution of charge carriers gets affected by induced
charges. By satisfying this requirement, the induces carriers TABLE I
FERROELECTRIC MATERIAL PARAMETERS [17].
becomes more dominant and the the electrostatic screening in
plasma becomes significant. Parameters Values
(2) Work function of metal electrodes and doping: The
Spontaneous 11.37 µC/cm2
electrode work function at drain (source) needs to be less polarization
(more) in comparison to intrinsic silicon in order to create N+ Remanent polarization 10.75 µC/cm2
(P+) doped regions. This condition allows for the formation of Coercive field 1.15 MV/cm
an excess of electrons or holes, respectively, in the Relative dielectric 32
semiconductor. By carefully selecting the electrode work constant
function, the CP technique can induce the desired type of
doping (P+ or N+) by manipulating the distribution of charge
carriers. Numerical calculations were conducted using the
Accordingly in this work, a ferroelectric material i.e., SILVACO ATLAS TCAD tool. The simulations incorporated
silicon doped hafnium oxide (Si:HfO2) is employed in stack several models to accurately represent the device physics and
with gate insulator alongside a lower bandgap source material phenomena observed; Non-local BTBT model was employed
(Si0.5Ge0.5) to propose a device namely negative capacitance to accurately capture the non-local nature of carrier generation
hetero-junction charge-plasma based tunnel FET (NC-HJ-CP- and recombination, Shockley-Read Hall recombination model
TFET) for investigating it’s biosensing applications. A was employed to accurately account for the recombination of
comparative detailed analysis is carried out with its silicon-
charge carriers through trap states within the semiconductor
based NC-CP-TFET counterpart to obtain transfer
material, Lombardy mobility model characterizes the carrier
characteristics, sensitivity by utilizing positive, negative and
neutral biomolecules. mobility in the device, Band gap narrowing model considers
energy bandgap reduction due to heterojunction formation,
II. DEVICE STRUCTURE AND SIMULATION METHOD Phonon-assisted and Nonlocal trap-assisted tunnelling model.
Figure 1(a) presents the schematic of intended device i.e., In addition, the Landau-Khalatnikov (L-K) ferroelectric
NC-HJ-CP-TFET as biosensor. The preliminary detailed interface model was used in the simulation to account for the
fabrication steps for NC-HJ-CP-TFET biosensor is displayed influence of the ferroelectric material. The L-K equation was
in Fig 1(b). The dimensions and materials employed in both used for calculation of potential drop across the FE layer
the biosensors are identical except for a source material. In sandwiched between a SiO2 layer and channel. This model
NC-HJ-CP-TFET biosensor, a low bandgap material Si0.5Ge0.5 allows for an accurate assessment of the ferroelectric
is employed to form a hetero-junction at source-channel area
material's behavior and properties within the device. The
expression of L-K equation is expressed as:
VFE = aoQ + boQ3 + coQ5 (1)
where Q and VFE are charge and potential drop across
ferroelectric layer. a0, b0, and c0 are the coefficients related to
the thickness of FE layer and Landau parameters m, n, and o
of ferroelectric layer: a0 = 2tfem, b0 = 4tfen, and c0 = 6tfeo. To
employ FE layer into TCAD calculation, Landau-Khalatnikov (a) (b)
Fig. 3 Variation of different biomolecules on NC-HJ-CP-TFET
Nonlinear Capacitance (LKNC) and LKFERRO model [27] biosensor in terms of (a) BTBT rate of electrons, (b) Electron
was utilized that provides full characterization of the concentration.
ferroelectric layer's electrical properties. The L-K equation
that is employed in the TCAD simulation is expressed as: III. RESULTS AND DISCUSSION
EFE = 2 P + 4 P3 + EOFF (2) This section discusses and analyses the results obtained
through numerical calculation. To analyze the biosensor's
where P and EFE is polarization induced and electric field response to biomolecule immobilization, neutral, positive and
across FE layer, α, β are Landau parameters and EOFF is the negative charges of biomolecules are considered. In this work,
offset field. The relationship between polarization and electric five different biomolecules are considered with their different
field is typically represented by an S-shaped graph and dielectric constants; Streptavidin (K = 2.1), APTES (K=3.57),
characterized by LKCOERCIVE, LKREMANENT, and Bacteriophage (K = 6.3), Protein (K =8), and Gelatin (K =12).
However, this section is further divided into three subsections;
LKOFFSET, defined in the interface statement of a model to
a) Neutral biomolecule, b) Positively charge biomolecule, and
set the characteristics of the polarization versus electric field c) Negatively charge biomolecule. The performance of
graph for the ferroelectric material under investigation. biosensors is evaluated with respect to sensitivity. Biosensor's
Figure 2 in the study displays the validation curve obtained sensitivity is calculated using a benchmark value. The
from the numerical calculations performed using the TCAD benchmark is selected in such a way that it can replicate the
tool. The numerical calculation models were tuned by sensor's response to the sensing sample. Benchmark reflects
comparing their results with reference data obtained from the instance when the cavity is filled with air or the condition
experimental research, as documented in reference [28] . before immobilization of biomolecules takes place into cavity,
To fine-tune the non-local BTBT model, the tunnelling i.e., K = 1. Sensitivity can be calculated either in terms of ON-
masses were tuned from their default values to specific state current (ION), transconductance (gm), threshold voltage
regulated values. Specifically, the tunnelling masses were set (Vth) and many more electrical parameters. In this work,
as mh.tunnel = 0.51mo and me.tunnel = 0.22mo, where mo sensitivity is observed in terms of ION and gm whose
represents the electron rest mass. By achieving a decent expressions are mentioned below:
agreement between the numerically calculated results and the I ON ( K = 2.1,3.57, 6.3,8,12)
experimental data, the validity of the numerical calculations is S ION = (3)
confirmed. I ON ( K = 1)
g m,max ( K = 2.1,3.57, 6.3,8,12)
S gm = (4)
g m,max ( K = 1)
where K shows different dielectric constants of biomolecules.

(a) (b)

Fig. 2 Benchmarking of numerical simulation [28].

 region enhances the energy band-bending; suppresses the
tunneling width and thereby increasing the drain current with
respect to Si-based NC-CP-TFET. Inclusion of Si:HfO2
further enahnces the current driving capability due to voltage-
amplification phenomenon for both the proposed biosensors.
Increase in dielectric constant increases the drain current due
to increase in gate-channel capacitance i.e., better electrostatic
control over the channel; that increases the electrostatic field
(c) (d) and consequently BTBT rate at source-channel junction and
Fig. 4 Variation of different biomoelcules in terms of ID-VGS (a) NC- drain current also increases.
HJ-CP-TFET at VDS = 0.1 V, (b) Si-based NC-CP-TFET at VDS = 0.1
V, (c) NC-HJ-CP-TFET at VDS = 1 V, (d) Si-based NC-CP-TFET at
VDS = 1 V.
A. Neutral Biomolecules
This subsection analyses the effect of neutral biomolecules
on the DC characteristics of NC-HJ-CP-TFET and Si-based
NC-CP-TFET biosensors. Fig. 3(a) and 3(b) shows the
variation of different biomolecules on band-to-band tunnelling
(BTBT) rate of electrons and electron concentration HJ-CP- (a) (b)
TFET biosensor. The increase in dielectric constant of
different biomolecules present in the cavity region enhances
the capacitance between gate and channel. Furthermore,
increase in capacitance leads to enhance the energy band-
bending. Consequently, it is observed from Fig. 3 that charge
(electron) concentration increases at source-channel junction
that further enhances the BTBT rate as well.

(c) (d)

(a) (b)

(e) (f)

(c) (d)
Fig. 5 Variation of different biomoelcules in terms of sensitivity (a) (g) (h)
SION of NC-HJ-CP-TFET, (b) SION of Si-based NC-CP-TFET, (c) Sgm Fig. 6 ID-VGS variation of different biomolecules for different
of NC-HJ-CP-TFET, (d) Sgm of Si-based NC-CP-TFET. positively charge biomolecules for NC-HJ-CP-TFET (a) 2.5×1011
cm-2, VDS = 0.1 V, (b) 2.5×1011 cm-2, VDS = 1 V, (c) 5×1011 cm-2, VDS
Fig. 4 (a) and (b) presents the tranfer charcteristics of NC- = 0.1 V, (d) 5×1011 cm-2, VDS = 1 V, (e) 7.5×1011 cm-2, VDS = 0.1 V,
HJ-CP-TFET and Si-based NC-CP-TFET biosensor for (f) 7.5×1011 cm-2, VDS = 1 V, (g) 1×1012 cm-2, VDS = 0.1 V, (h)
different biomolecules at VDS = 0.1 V; whereas Fig. 4(c) and 1×1012 cm-2, VDS = 1 V.
4(d) shows the ID-VGS variation of both the proposed
biosensors at VDS = 1 V for different biomolecules. In TFETs, Furthermore, performance parameter of biosensor i.e.,
drain current depends on probability of tunneling, which is sensitivity, is calculated with respect to ION and
dependent on energy bandgap, effective mass and tunneling transconductance. Higher sensitivity is preferred for
width [29]. It is observed from Fig. 4(a) and 4(c) that biosensors because it indicates a higher probability of
incorporation of low bandgap material (Si0.5Ge0.5) at source detecting target species. Fig. 5(a) and (b) presents the effect of
several biomolecules on the sensitivity of NC-HJ-CP-TFET Fig. 6 shows the variation of different positively charges
and Si-based NC-CP-TFET biosensor, respectively, in terms biomolecules plotted along with different biomolecules with
of ON-state current (ION) for VDS = 0.1 V and VDS = 1V. It is respect to transfer characteristics for different VDS. It can be
evident from Fig. 5(a) and (b) that sensitivity of NC-HJ-CP- observed from Fig. 6 that on increasing the positively charged
TFET is more in compare to Si-based NC-CP-TFET biomolecule i.e., from 2.5×1011 cm-2 to 1×1012 cm-2, drain
biosensor. The reason is quite obvious due to the fact that current increase for particular biomolecules. The reason is
higher dielectric constant provides enhanced gate-channel quite obvious as explained earlier that increase in positive
capacitive coupling, which increases the drain current with interface charge/biomolecule, suppresses the flatband voltage
respect to least dielectric constant i.e., K = 1. Higher the drain which, further increases the effective gate voltage. Increase in
current, higher will be the sensitivity. Moreover, the effective gate voltage reflects better electrostatic control over
sensitivity of NC-HJ-CP-TFET biosensor for VDS = 1 V, is the channel and thus improves the energy band-bending,
less by an order of 2×102 in contrast with Si-based NC-CP- which leads to increase in drain current.
TFET; this reflects that NC-HJ-CP-TFET biosensor performs Moreover, positively charged biomolecule depletes the
better at lower supply voltages, hence it can be used to sense intrinsic channel, thereby causing increase in tunnelling of
the biomolecules at low voltages. Furthermore, benchmarking electrons at source due to increase in ϕs, therefore drain
of proposed biosensor i.e., NC-HJ-CP-TFET, is done with current increases. The positively charged biomolecule of the
respect to previously reported biosensors in terms of fully filled cavity influences the I D-VGS curve with respect to
sensitivity that are mentioned in table 2. ‘k’; transfer characteristics curve shifts towards left as ’k’
increases. The reason is increase in gate-channel capacitance
B. Positive Biomolecules
that leads to decrease in threshold voltage.
Fig. 7 shows the ON-state current sensitivity plotted for
In this subsection, investigation has been carrried out to see different values of VDS. It could be observed that for lower
the impact of positively chraged biomlolecules on the ID-VGS
value of VDS (i.e., Fig 7(a)), higher sensitivity is obtained in
characteristics of NC-HJ-CP-TFET and Si-based NC-CP-
compare to higher VDS (i.e., Fig 7(b)) value, therefore it
TFET biosensors. For analysis, different biomolecules are
justifies that NC-HJ-CP-TFET can act as better biosensor for
considered along with different charges. Biomolecules that are low power supply, i.e., having low power dissipation as well.
considered to be present in cavity region holds charge also, Moreover, one can also observe from Fig. 7(a) that for higher
which acts as interface charges. A MOS structure's voltage
value of dielectric constant, variation in sensitivity is not
balance equation is expressed as:
significant. However, for VDS = 1 V, sensitivity increases with
positively charged biomolecule.
VGB = ox + s + VFB (5)
TABLE II
BENCHMARKING OF PROPOSED AND PREVIOUSLY REPORTED BIOSENSORS.
q( Nbio )
VFB = MS − (6) FET based Biosensors Values
Cox
DM FET (Lgap=200nm, Hgap=15nm, K=2.1 [30] 1×104
where VGB is gate-to-body voltage, ϕox is voltage drop across
Full Gate DM TFET (Lgap=10nm, Lgate=42nm, 1×105
oxide, ϕs is surface potential, VFB is flatband voltage, φMS is Hgap=5nm, K=4) [31]
contact potential between gate metal and semiconductor, Nbio Conventional TFET [20] 50
is the interface charge (either it can be positive or negative),
Lateral DM TFET [32] 10
and Cox is the oxide capacitance. Accordingly, for positively
charge biomolecule, flatband voltage reduces that further Buried strained Si1-x Gex DGTFET (Lgap=20nm, 2.6×105
Lgate=40nm Hgap=5nm, K=8) [33]
enhances the effective gate-to-body voltage (VGB-eff = VGB -
VFB) or gate-to-source voltage (VGS-eff = VGS - VFB). Increase SiGe Source DM PNPN TFET with 20% Ge 5×103
Composition [21]
in effective gate-to-source voltage increases the band-to-band
DM PNPN TFET (Lgap=30nm, Lgate=100nm 3×106
tunnelling, BTBT rate, and consequently drain current. Hgap=9nm, K=10) [34]
DC-DM HTEFT (Lgap1=20nm, Lgap2=10nm, 5×106
Hgap=5nm, Lgate1=25nm, Lgate2=15nm, K=12)
[35]
Si-based NC-CP-TFET (Lc=11nm, Tc=4nm, k=12) 2×107
(This work)
NC-HJ-CP-TFET (Lc=11nm, Tc=4nm, k=12) (This 2×108
work)

(a) (b) C. Negative Biomolecules

Fig. 7 Variation of SION of positively charge bimolecule for NC-HJ- This subsection discusses the effect of negatively charged
CP-TFET at (a) VDS = 0.1 V, (b) VDS = 1 V. different biomolecules on transfer characteristics and
sensitivity of NC-HJ-CP-TFET. Fig. 8 shows the variation of
different negatively charged biomolecules plotted along with
different biomolecules with respect to transfer characteristics Furthermore, sensitivity curve is plotted for negatively
for different VDS. On increasing the values of negatively charged different biomolecules at different VDS, as shown in
charged biomolecules in the cavity region, leads to increase in Fig. 9. One could again observe similar variation that for
flatband voltage as could be observed from equation (4); that lower value of VDS (i.e., Fig 9(a)), higher sensitivity is
consequently suppresses the effective gate voltage, energy obtained in compare to higher VDS (i.e., Fig 9(b)). Sensitivity
band-bending and thereby reducing the tunnelling at source- of the biosensors increases with increase in dielectric constant
channel junction. Therefore, one could easily observe form for both the values of VDS.
Fig. 8, that on increasing the magnitude of negatively charged
biomolecule for a particular dielectric constant, the drain
current decreases.

(a) (b)
Fig. 9 Variation of SION of positively charge bimolecule for NC-HJ-
CP-TFET at (a) VDS = 0.1 V, (b) VDS = 1 V.
(a) (b)
IV. CONCLUSION
A detailed comparative analysis has been made between NC-
HJ-CP-TFET and silicon-based NC-CP-TFET in terms of
transfer characteristics, sensitivity, ION/IOFF, and SS. It is
observed that NC-HJ-CP-TFET outperforms Si-based NC-CP-
TFET in terms of electrical parameters and sensitivity.
Inclusion of low bandgap material and ferroelectric layer in
gate stack enhances the performance of NC-HJ-CP-TFET in
(c) (d)
contrast with Si-based NC-CP-TFET. ON-state current
sensitivity obtained for NC-HJ-CP-TFET and Si-based NC-
CP-TFET with fully filled cavity having dielectric constant of
k = 12 are 2×108 and 2×107, respectively, i.e., sensitivity
increases by an order of 10 times. It is also observed that both
the biosensors perform better for lower values of VDS, thereby
providing high parametric value of sensitivity. Increase in
dielectric constant of bio-analyte increases the gate-channel
capacitance that provides tighter control over the channel
(e) (f)
consequently leads to increase in drain current. Performance
of biosensors is evaluated for different neutral and charged
bio-analytes. Therefore, the higher sensitivity of NC-HJ-CP-
TFET reflects that it has the prospects to function as a robust,
low-cost, low-power, ultra-sensitive dielectrically modulated
biosensor with single cavity.

DECLARATIONS
(g) (h)
Fig. 8 ID-VGS variation of different biomolecules for different
Author Contributions: All authors contributed to the study
negatively charge biomolecules for NC-HJ-CP-TFET (a) -2.5×1011
cm-2, VDS = 0.1 V, (b) - 2.5×1011 cm-2, VDS = 1 V, (c) -5×1011 cm-2, conception and design. Simulations and data analysis were
VDS = 0.1 V, (d) -5×1011 cm-2, VDS = 1 V, (e) -7.5×1011 cm-2, VDS = performed by Varun Mishra and Chandni Tiwari. The draft of
0.1 V, (f) -7.5×1011 cm-2, VDS = 1 V, (g) -1×1012 cm-2, VDS = 0.1 V, the manuscript was written by Lucky Agarwal. Vikas Rathi
(h) -1×1012 cm-2, VDS = 1 V. prepared the final draft.
Conflict of interests: The authors have no competing
Moreover, while keeping negatively charged biomolecule financial or non-financial interests.
and VGS as constant, and on increasing the dielectric constant, Availability of Data and Material Not applicable
surface potential increases due to increase in gate-to-channel Consent to Participate: Not applicable.
capacitance, that leads to increase in BTBT rate at source- Consent for Publication: Not applicable
channel junction which consequently increases the drain Funding: Not applicable
current and decreases threshold voltage, as can be observed
Ethics approval: Not applicable
from Fig. 8.
Acknowledgement: vol. 33, no. 4, pp. 1–11, 2020, doi: 10.1002/jnm.2726.
[18] M. S. Parihar and A. Kranti, “Enhanced sensitivity of double gate
The authors are thankful to MNNIT Allahabad for accessing junctionless transistor architecture for biosensing applications,”
the SILVACO Atlas software Nanotechnology, vol. 26, no. 14, p. 145201, 2015, doi:
10.1088/0957-4484/26/14/145201.
[19] S. Singh, P. N. Kondekar, and N. K. Jaiswal, “Label-free biosensor
REFERENCES using nanogap embedded dielectric modulated schottky tunneling
source impact ionization MOS,” Microelectron. Eng., vol. 149, pp.
[1] R. Goswami and B. Bhowmick, “Comparative analyses of circular 129–134, 2016, doi: 10.1016/j.mee.2015.10.005.
gate TFET and heterojunction TFET for dielectric-modulated label- [20] D. Sarkar and K. Banerjee, “Proposal for tunnel-field-effect-
free biosensing,” IEEE Sens. J., vol. 19, no. 21, pp. 9600–9609, transistor as ultra-sensitive and label-free biosensors,” Appl. Phys.
2019, doi: 10.1109/JSEN.2019.2928182. Lett., vol. 100, no. 14, 2012, doi: 10.1063/1.3698093.
[2] E. Souteyrand et al., “Direct detection of the hybridization of [21] S. Kanungo, S. Chattopadhyay, P. S. Gupta, K. Sinha, and H.
synthetic homo-oligomer DNA sequences by field effect,” J. Phys. Rahaman, “Study and Analysis of the Effects of SiGe Source and
Chem. B, vol. 101, no. 15, pp. 2980–2985, 1997, doi: Pocket-Doped Channel on Sensing Performance of Dielectrically
10.1021/jp963056h. Modulated Tunnel FET-Based Biosensors,” IEEE Trans. Electron
[3] X. P. A. Gao, G. Zheng, and C. M. Lieber, “Subthreshold regime Devices, vol. 63, no. 6, pp. 2589–2596, 2016, doi:
has the optimal sensitivity for nanowire FET biosensors,” Nano 10.1109/TED.2016.2556081.
Lett., vol. 10, no. 2, pp. 547–552, 2010, doi: 10.1021/nl9034219. [22] P. K. Verma, Y. K. Verma, V. Mishra, and S. K. Gupta, “A charge-
[4] and C. Z. Chang, Hsiao-Kang, Fumiaki N. Ishikawa, Rui Zhang, plasma-based dual-metal-gate recessed-source/drain dopingless
Ram Datar, Richard J. Cote, Mark E. Thompson, “Rapid, Label- junctionless transistor with enhanced analog and RF performance,”
Free, Electrical Whole Blood Bioassay Based on Nanobiosensor J. Comput. Electron., vol. 19, no. 3, pp. 1085–1099, 2020, doi:
Systems,” vol. 5, no. 12, pp. 9883–9891, 2011. 10.1007/s10825-020-01528-z.
[5] K. Shoorideh and C. O. Chui, “On the origin of enhanced sensitivity [23] A. K. Jain, S. Sahay, and M. J. Kumar, “Controlling L-BTBT in
in nanoscale FET-based biosensors,” Proc. Natl. Acad. Sci. U. S. A., Emerging Nanotube FETs Using Dual-Material Gate,” IEEE J.
vol. 111, no. 14, pp. 5111–5116, 2014, doi: Electron Devices Soc., vol. 6, no. January, pp. 611–621, 2018, doi:
10.1073/pnas.1315485111. 10.1109/JEDS.2018.2829633.
[6] P. Dwivedi and A. Kranti, “Applicability of Transconductance-to- [24] A. Lahgere, M. Panchore, and J. Singh, “Dopingless ferroelectric
Current Ratio (gm/Ids) as a Sensing Metric for Tunnel FET tunnel FET architecture for the improvement of performance of
Biosensors,” IEEE Sens. J., vol. 17, no. 4, pp. 1030–1036, 2017, dopingless n-channel tunnel FETs,” Superlattices Microstruct., vol.
doi: 10.1109/JSEN.2016.2640192. 96, pp. 16–25, 2016, doi: 10.1016/j.spmi.2016.05.004.
[7] D. Keighobadi, S. Mohammadi, and M. Mohtaram, “Switching [25] S. Hfo et al., “Downscaling ferroelectric field effect transistors by
Performance Investigation of a Gate ‑ All ‑ Around Core ‑ Source using ferroelectric,” Solid State Electron., vol. 88, pp. 65–68, 2013,
InGaAs / InP TFET,” Trans. Electr. Electron. Mater., vol. 22, no. doi: 10.1016/j.sse.2013.04.013.
0123456789, pp. 502–508, 2021, doi: 10.1007/s42341-020-00257-1. [26] R. Narang, K. V. S. Reddy, M. Saxena, R. S. Gupta, and M. Gupta,
[8] S. L. Tripathi, S. K. Sinha, and G. S. Patel, “Low-Power Efficient “A dielectric-modulated tunnel-FET-based biosensor for label-free
p+ Si0.7Ge0.3 Pocket Junctionless SGTFET with Varying detection: Analytical modeling study and sensitivity analysis,” IEEE
Operating Conditions,” J. Electron. Mater., vol. 49, no. 7, pp. 4291– Trans. Electron Devices, vol. 59, no. 10, pp. 2809–2817, 2012, doi:
4299, 2020, doi: 10.1007/s11664-020-08145-3. 10.1109/TED.2012.2208115.
[9] S. Ahish, D. Sharma, Y. B. N. Kumar, and M. H. Vasantha, [27] T. Silvaco and C. A. D. Version, “ATLAS 5.19. 20.” R, 2020.
“Performance enhancement of Novel InAs/Si hetero double-gate [28] Z. X. Chen et al., “Demonstration of tunneling FETs based on
tunnel fet using Gaussian doping,” IEEE Trans. Electron Devices, highly scalable vertical silicon nanowires,” IEEE Electron Device
vol. 63, no. 1, pp. 288–295, 2016, doi: 10.1109/TED.2015.2503141. Lett., vol. 30, no. 7, pp. 754–756, 2009, doi:
[10] C. Sandow, J. Knoch, C. Urban, Q. T. Zhao, and S. Mantl, “Impact 10.1109/LED.2009.2021079.
of electrostatics and doping concentration on the performance of [29] S. Kumar, Y. Singh, B. Singh, and P. K. Tiwari, “Simulation study
silicon tunnel field-effect transistors,” Solid. State. Electron., vol. of dielectric modulated dual channel trench gate TFET-Based
53, no. 10, pp. 1126–1129, 2009, doi: 10.1016/j.sse.2009.05.009. biosensor,” IEEE Sens. J., vol. 20, no. 21, pp. 12565–12573, 2020,
[11] V. Mishra, Y. K. Verma, S. K. Gupta, and V. Rathi, “A SiGe- doi: 10.1109/JSEN.2020.3001300.
Source Doping-Less Double-Gate Tunnel FET: Design and [30] H. Im, X. J. Huang, B. Gu, and Y. K. Choi, “A dielectric-modulated
Analysis Based on Charge Plasma Technique with Enhanced field-effect transistor for biosensing,” Nat. Nanotechnol., vol. 2, no.
Performance,” Silicon, 2021, doi: 10.1007/s12633-021-01030-6. 7, pp. 430–434, 2007, doi: 10.1038/nnano.2007.180.
[12] A. Saeidi et al., “Nanowire Tunnel FET with Simultaneously [31] S. Kanungo, S. Chattopadhyay, P. S. Gupta, and H. Rahaman,
Reduced Subthermionic Subthreshold Swing and Off-Current due to “Comparative performance analysis of the dielectrically modulated
Negative Capacitance and Voltage Pinning Effects,” Nano Lett., vol. full- gate and short-gate tunnel FET-based biosensors,” IEEE Trans.
20, no. 5, pp. 3255–3262, 2020, doi: 10.1021/acs.nanolett.9b05356. Electron Devices, vol. 62, no. 3, pp. 994–1001, 2015, doi:
[13] K. Tomioka and T. Fukui, “Current increment of tunnel field-effect 10.1109/TED.2015.2390774.
transistor using InGaAs nanowire/Si heterojunction by scaling of [32] M. Verma, S. Tirkey, S. Yadav, D. Sharma, and D. S. Yadav,
channel length,” Appl. Phys. Lett., vol. 104, no. 7, 2014, doi: “Performance Assessment of A Novel Vertical Dielectrically
10.1063/1.4865921. Modulated TFET-Based Biosensor,” IEEE Trans. Electron Devices,
[14] C. Jiang, R. Liang, and J. Xu, “Investigation of Negative vol. 64, no. 9, pp. 3841–3848, 2017, doi:
Capacitance Gate-all-Around Tunnel FETs Combining Numerical 10.1109/TED.2017.2732820.
Simulation and Analytical Modeling,” IEEE Trans. Nanotechnol., [33] A. Anam, S. Anand, and S. I. Amin, “Design and Performance
vol. 16, no. 1, pp. 58–67, 2017, doi: Analysis of Tunnel Field Effect Transistor with Buried Strained Si1-
10.1109/TNANO.2016.2627808. xGexSource Structure Based Biosensor for Sensitivity
[15] L. Agarwal, GL Priya, E Papnassam, BP Kumar, “A novel metal Enhancement,” IEEE Sens. J., vol. 20, no. 22, pp. 13178–13185,
dielectric metal based GAA-junction-less TFET structure for low 2020, doi: 10.1109/JSEN.2020.3004050.
loss sram design,” silicon, vol. 15, 2023, doi.org/10.1007/s12633- [34] R. Narang, M. Saxena, and M. Gupta, “Comparative Analysis of
022-02218-0 Dielectric-Modulated FET and TFET-Based Biosensor,” IEEE
[16] S. Salahuddin and S. Datta, “Can the subthreshold swing in a Trans. Nanotechnol., vol. 14, no. 3, pp. 427–435, 2015, doi:
classical FET be lowered below 60 mV/decade?,” Tech. Dig. - Int. 10.1109/TNANO.2015.2396899.
Electron Devices Meet. IEDM, no. 1, pp. 2–5, 2008, doi: [35] S. Mukhopadhyay, D. Sen, B. Goswami, and S. K. Sarkar,
10.1109/IEDM.2008.4796789. “Performance Evaluation of Dielectrically Modulated Extended
[17] V. Mishra, Y. K. Verma, and S. K. Gupta, “Surface potential–based Gate Single Cavity InGaAs/Si HTFET Based Label-Free Biosensor
analysis of ferroelectric dual material gate all around (FE-DMGAA) Considering Non-Ideal Issues,” IEEE Sens. J., vol. 21, no. 4, pp.
TFETs,” Int. J. Numer. Model. Electron. Networks, Devices Fields, 4739–4746, 2021, doi: 10.1109/JSEN.2020.3033576.