2017 Devices for Integrated Circuit (DevIC), 23-24 March, 2017, Kalyani, India 681

Sensitivity study of staggered heterojunctions based

SRG TFET sensor
Avik Chakraborty Debasis Singha
ECE Department, Jalpaiguri Government Engineering Surendranath Evening College
College Kolkata, India
Jalpaiguri, India singharec@yahoo.co.in
chakraborty.avik.ece@gmail.com

Abstract— This paper presents the sensitivity analysis of a compared to HJ counterparts [10-11]. The initial results are
staggered heterojunctions based surrounding gate (SRG) Tunnel highly promising, which sets the need for a in-depth analysis
FET (TFET) based biosensor, for the first time. The sensitivity of of HETJ-based SRG TFET-based biosensor. In this paper, a
the biosensor has been measured using the conventional change simulation study has been performed to demonstrate the
in threshold voltage as well as in terms of ION/IOFF ratio also. The improvement of the sensitivity of the staggered
results obtained using TCAD simulation reveals that the heterojunctions based surrounding gate (SRG) Tunnel FET
proposed biosensor outperforms the homojunction (HJ) TFET- (TFET)-based biosensor as compare to their conventional
based biosensor counterpart. counterpart.
Keywords—biosensor; TFET; Surrounding gate; sensitivity;
heterojunction II. SIMULATION SETUP
The cross-sectional device structure is shown in Fig. 1. For
I. INTRODUCTION gate contact, a mid bandgap material with workfunction equals
Biosensor is used to detect the biomolecules [1]. Among to 4.61eV has been chosen.
many FET-based biosensors, dielectric-modulated biosensors
evinced itself as the most potential candidate for the design of
next generation of biosenosrs. The presence or absence of
biomolecules causes change in the electrical characteristics of
the device. The threshold voltage or current in the device
changes when biomolecules with certain permittivity is
present in the nanogap cavity regions as compared to the
presence of air in the cavity regions. Recently, nanogap-based
dielectric modulated biosesnors are increasingly drawing
attention due to their label-free detection capability and high
sensitivity [2-4].Recently Banerjee et al. demonstrated that
TFET-based biosensors can overcome the fundamental
limitations such as maximum sensitivity and limited detection
time of conventional MOSFET-based biosensor due to their
different carrier injection mechanism [5] in the form of band-
to-band-tunneling. In a recent work, Kanungo et al. proposed
and investigated the use of TFET as a biosensor [6]. In another Fig. 1: Cross-sectional structure of heterojunctions based surrounding
gate (SRG) Tunnel FET (TFET)-based biosensor. TFET is basically a
work, Narang et al. examined the possibility of p-n-p-n TFET reversed-biased p-i-n structure. The nanogap cavity region is intended to
as a biosensor by developing an analytical model [7]. It is capture biomolecules, thus modulating the tunneling barrier and hence the
known that TFET exhibits low subthreshold swing BTBT current.
(<60mv/dec). However, the Si-based TFETs suffers with their
Non-local Band-to-band-tunneling (BTBT) model has
low ON-state current ION due to the high carrier effective
been incorporated to model the tunneling of carriers from
mass. To circumvent this problem, narrow bandgap III-V
source-to-channel region. The doping profile is Gaussian in
based TFET structures was proposed due to its low carrier
nature with a gradiant 2mV/dec and a peak concentration of
effective mass, smaller bandgap and smaller tunneling width
5x1019/cc. To take care of the recombination’s, standard
[8]. Recently, the use of III–V staggered hetero-tunnel-
Shockley–Read–Hall(SRH) and Auger recombination model
junctions in the source-channel junctions exhibits a large
has been used in the simulation. Ferm carrier statistics and
enhancement in the ION/IOFF ratio of the as compared to
Quantum density gradient model has been used to model the
conventional one [9]. Moreover, recent studies of staggered
carrier statistics and carrier transport mechanisms. To model
heterojunctions based surrounding gate (SRG) Tunnel FET
the temperature-dependency and electric field dependency of
(TFET) has exhibited better RF and analog performance as
the carrier mobility, Lombardi (CVT) mobility model has been

978-1-5090-4724-6/17/$31.00 ©2017 IEEE

 2017 Devices for Integrated Circuit (DevIC), 23-24 March, 2017, Kalyani, India 682

incorporated in this study. A combination of decoupled different permittivity in the nanogap region. The plot in Fig. 2
Gummel method with fully-coupled Newton’s method has reveals that as dielectric constant increases, the tunneling
been used to solve the numerical iteration by the TCAD. To barrier width decreases due to the change in effective gate
model the interface trapped charges, a fixed amount of capacitance, which results in steep rise in the potential profile.
interface trapped charge equals to 2x1011 cm-3 to 6x1012 cm-3
with uniform distribution has been considered in this
simulation. The material parameters used for the simulation
has been listed in Table I.
The structural parameters of the device used in this
simulation has been specified in Table II, unless otherwise
mentioned and/or varied.. Device simulations were performed
for the SRG n-type TFET, as sketched in Fig. 1, using the
version 5.11.24.C Silvaco Atlas device simulator [12]. Ni is
used to make the source and the drain contact. Bandgap
narrowing (BGN) model has been enabled in this simulation.
The A and B parameters has been varied simultaneously with
the amount of interface trapped charges Dit in order to
calibrate the simulator as shown in [11].

TABLE I. MATERIAL PARAMETERS OF TFET MATERIALS

Al0.5Ga0.5As0.3Sb0.7 In0.8Ga0.2As
Electron affinity 3.6 4.73

Light Hole mass 0.11 0.034

Fig. 2: Plot of the surface potential along the channel from the source to
Electron 0.09 0.036 the drain for different permittivity of the biomolecules present in the nanogap
effective mass regions.
Lattice constant in Ao 5.98 5.98 Fig. 3 plots the change in energy band diagram for
Bandgap (Eg) in eV 1.36 0.50 different permittivity of the biomolecules present in the
nanogap cavity region. The plot shows that for k=1, the TFET
Dielectric 14.5528
is in OFF state. The tunneling barrier between the valence
constant
band of the sourcenad conduction band of the channel is large
TABLE II. STRUCTURE PARAMETERS OF HETJ SRG-BASED TFET-BASED
and hence, source-channel tunneling is negligible, except only
BIOSENSOR leakage current through p-i-n junction and drain-to-channel
tunneling. However, the situation turns around for k= 10,
Value Unit where the TFET is in ON state. In this case, the valence band
Gate length 30 nm of the source and the conduction band of the channel are
Source and the drain length 20 nm aligned, thus resulting in a reduced tunneling barrier,
corresponds to large carrier injection from the source to the
Length of the nanogap cavity 12 nm channel, and hence in large BTBT current. Therefore, it
regios Lgap indicates that tunneling barrier depends on the permittivity of
Radius of the SRG TFET 10 nm the biomolecules. Therefore, the biomolecules can be detected
by the corresponding change in current.
Equivalent oxide thickness (EOT 1 nm
of t1) Fig. 4 plots the drain current versus gate voltage for
t2 (Thickness of the Al2O3 layer) 11 nm different permittivity of the biomolecules present in the
nanogap cavity regions. It is known that subthershold region is
Doping level of the source (p+) 2x1019 cm-3 the optimal one for detection of biomolecules for conventional
Doping level of the drain (n+) 5x1019 cm-3 MOSFET as well as for the TFET-based one [13]. However,
conventional MOSFET suffers with limited subthreshold
Doping level of the channel (n-) 5x1016 cm-3
swing (>60mv/dec). However, TFET overcomes this by
Gate workfunction 4.61 eV Fermi-tail cutting of the bandgap of the semiconductor.
Moreover, it was reported that staggered HETJ-based TFET
Underlap Length 0 nm
offers lowest achievable subthreshold swing as compared to
other types of III-V based TFETs [9]. As a result, the change
in current for a staggered HETJ-based biosensor is higher than
III. RESULTS AND DISCUSSIONS that of other TFET-based biosensors, mainly owing to the
Fig. 2 shows the plot of the surface potential as a function of lower subthreshold swing. Fig. 4 shows the Id-Vgs
the position of the channel from the source to the drain for characteristics before and after biomolecule conjugation of
 2017 Devices for Integrated Circuit (DevIC), 23-24 March, 2017, Kalyani, India
different permittivities. The change in current clearly indicates
the possibility of detecting biomolecules by sensing the
change in current modulated by the presence of biomolecules.
Fig. 3: Plot of energy band as a function of the position along the channel
from the source to the drain.
Fig. 4: Drain current versus gate-to-source voltage for different
permittivity of the target bio-analytes.
The sensitivity of a biosensor is defined by the difference
of threshold voltage VTh for nanogap filled with air and
nanogap filled with biomolecules.
ΔVTh = VTh( gap = air ) − VTh( gap = filled )
Fig. 5 plots the change in threshold voltage and sensitivity as a
function of dielectric constant of the biomolecules. The
threshold voltage is determined by a constant current method
(VTh = VGS when IDS = 10−7 A/μm). Fig. 5 shows that the
change in threshold voltage and sensitivity is quite large for a
staggered HETJ SRG-based biosensor. Therefore, it is
683
possible to detect the biomolecules by the sensing the changes
in the threshold voltage. Moreover, the same staggered HETJ
SRG-based transistor architecture can be used for sensing and
circuit units using in a stand-alone and inverter configuration
(as a two-component common source amplifier as shown by
Rigante et al. [14]).
Fig. 5: Change in threshold voltage and sensitivity as a function of
dielectric constant of the biomolecules
Fig. 6: ION/IOFF ratio as a function of dielectric constant of the
biomolecules
Fig. 6 plots the ION/IOFF ratio as a function of dielectric
constant of the biomolecules. As compared to a MOSFET-
(1) based biosensor, TFET-based biosensors use another sensing
parameter known as ION/IOFF ratio [7]. This can be attributed to
the difference of basic operation mechanism between TFET
and conventional MOSFET. In a conventional MOSFET-
based biosernsor, presence of biomolecules causes
deformation in surface potential. The minimum of this surface
potential is results in change in threshold voltage. However,
this potential deformation increases with increase in nanogap
 2017 Devices for Integrated Circuit (DevIC), 23-24 March, 2017, Kalyani, India 684

cavity length. This potential deformation causes increase in
Threshold voltage VTh and subsequent decrease in the ION.
However, in a TFET as permittivity k of the biomolecule
changes, ION increases drastically and ION/IOFF ratio increases.
IV. CONCLUSIONS
tunnel-field-effect-transistors." Applied Physics Letters 97, no. 26
(2010): 263109.
[14] Rigante, Sara, Paolo Scarbolo, Mathias Wipf, Ralph L. Stoop, Kristine
Bedner, Elizabeth Buitrago, Antonios Bazigos et al. "Sensing with
Advanced Computing Technology: Fin Field-Effect Transistors with
High-k Gate Stack on Bulk Silicon." ACS nano 9, no. 5 (2015): 4872-
4881.

In summary, an analysis of the sensitivity and biomoleule

detection capability of the staggered heterojunction-based
SRG-TFET-based biosensor has been provided. The results
indicate that the proposed biosensor will evince itself as an
potential alternative to the conventional TFET-based
biosensors primarily due to its high sensitivity. Therefore,
staggered-heterojunction SRG TFET-based biosensors can
facilitate detection of biomolecules of low concentration,
which is crucial to prevent biological accidents or attacks.
Therefore, the proposed biosensor can pave the way for a
paradigm shift in biosensing applications.

REFERENCES
[1] P. Bergveld, “The development and application of FET-based
biosensors,” Biosensors, vol. 2, no. 1, pp. 15–33, 1986
[2] B. Gu, T. J. Park, J.-H. Ahn, X.-J. Huang, S. Y. Lee, and Y.-K. Choi,
“Nanogap field-effect transistor biosensors for electrical detection of
avian influenza”, Small, vol. 5, no. 21, pp. 2407–2412, Nov. 2009
[3] Sarkar, Deblina, and Kaustav Banerjee. "Proposal for tunnel-field-effect-
transistor as ultra-sensitive and label-free biosensors." Applied Physics
Letters 100, no. 14 (2012): 143108.
[4] Abdi, Dawit Burusie, and M. Jagadesh Kumar. "Dielectric modulated
overlapping gate-on-drain tunnel-FET as a label-free
biosensor." Superlattices and Microstructures 86 (2015): 198-202.
[5] Gao, Anran, Na Lu, Yuelin Wang, and Tie Li. "Robust ultrasensitive
tunneling-FET biosensor for point-of-care diagnostics." Scientific
reports 6 (2016).
[6] Kanungo, S.; Gupta, P.S.; Rhaman, H., "Effects of Germanium mole
fraction variation at the source of a dielectrically modulated Tunneling
FET based biosensor," Devices, Circuits and Systems (ICDCS), 2014
2nd International Conference on , vol., no., pp.1,5, 6-8 March 2014
[7] Narang, R.; Reddy, K.V.S.; Saxena, M.; Gupta, R.S.; Gupta, M., "A
Dielectric-Modulated Tunnel-FET-Based Biosensor for Label-Free
Detection: Analytical Modeling Study and Sensitivity Analysis,"
Electron Devices, IEEE Transactions on , vol. 59, no.10, pp. 2809-2817,
Oct. 2012
[8] Baravelli, Emanuele, Elena Gnani, Roberto Grassi, Antonio Gnudi,
Susanna Reggiani, and Giorgio Baccarani. "Complementary n-and p-
type TFETs on the same InAs/Al 0.05 Ga 0.95 Sb platform." In Solid-
State Device Research Conference (ESSDERC), 2013 Proceedings of the
European, pp. 69-72. IEEE, 2013.
[9] Wang, Lingquan, Edward Yu, Yuan Taur, and Peter Asbeck. "Design of
tunneling field-effect transistors based on staggered heterojunctions for
ultralow-power applications." IEEE Electron Device Letters 31, no. 5
(2010): 431-433.
[10] Chakraborty, Avik, and Angsuman Sarkar. "Staggered Heterojunctions-
Based Nanowire Tunneling Field-Effect Transistors for Analog/Mixed-
Signal System-on-Chip Applications." NANO 10, no. 02 (2015):
1550027.
[11] Chakraborty, Avik, and Angsuman Sarkar. "Investigation of analog/RF
performance of staggered heterojunctions based nanowire tunneling
field-effect transistors." Superlattices and Microstructures 80 (2015):
125-135.
[12] Manual, ATLAS User’S. "Silvaco Int." Santa Clara, CA 5 (2008).
[13] Sarkar, Deblina, Michael Krall, and Kaustav Banerjee. "Electron-hole
duality during band-to-band tunneling process in graphene-nanoribbon