- Email: [email protected]
Quantum ballistic analysis of transition metal dichalcogenides based double gate junctionless field effect transistor and its application in nano-biosensor
Accepted Manuscript Quantum ballistic analysis of transition metal dichalcogenides based double gate junctionless field effect transistor and its application in nano-biosensor Abir Shadman, Ehsanur Rahman, Quazi D.M. Khosru PII:
S0749-6036(17)30805-4
DOI:
10.1016/j.spmi.2017.06.055
Reference:
YSPMI 5104
To appear in:
Superlattices and Microstructures
Received Date: 1 April 2017 Revised Date:
21 June 2017
Accepted Date: 22 June 2017
Please cite this article as: A. Shadman, E. Rahman, Q.D.M. Khosru, Quantum ballistic analysis of transition metal dichalcogenides based double gate junctionless field effect transistor and its application in nano-biosensor, Superlattices and Microstructures (2017), doi: 10.1016/j.spmi.2017.06.055. This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
ACCEPTED MANUSCRIPT
Quantum Ballistic Analysis of Transition Metal Dichalcogenides Based Double Gate Junctionless Field Effect Transistor and its Application in Nano-biosensor
Dept. of Electrical and Electronic Engineering, Bangladesh University of Engineering and Technology,
Dhaka-Bangladesh. ¶
These authors contributed equally to this work.
†
Corresponding Author email: [email protected]
SC
1
RI PT
Abir Shadman1¶, †, Ehsanur Rahman1¶, and Quazi D. M. Khosru1
M AN U
Abstract: To reduce the thermal budget and the short channel effects in state of the art CMOS
AC C
EP
TE D
technology, Junctionless field effect transistor (JLFET) has been proposed in the literature. Numerous experimental, modeling, and simulation based works have been done on this new FET with bulk materials for various geometries until now. On the other hand, the two-dimensional layered material is considered as an alternative to current Si technology because of its ultra-thin body and high mobility. Very recently few simulation based works have been done on monolayer molybdenum disulfide based JLFET mainly to show the advantage of JLFET over conventional FET. However, no comprehensive simulation-based work has been done for double gate JLFET keeping in mind the prominent transition metal dichalcogenides (TMDC) to the authors’ best knowledge. In this work, we have studied quantum ballistic drain current-gate voltage characteristics of such FETs within non-equilibrium Green function (NEGF) framework. Our simulation results reveal that all these TMDC materials are viable options for implementing state of the art Junctionless MOSFET with emphasis on their performance at short gate lengths. Besides evaluating the prospect of TMDC materials in the digital logic application, the performance of Junctionless Double Gate trilayer TMDC heterostructure FET for the label-free electrical detection of biomolecules in dry environment has been investigated for the first time to the authors’ best knowledge. The impact of charge neutral biomolecules on the electrical characteristics of the biosensor has been analyzed under dry environment situation. Our study shows that these materials could provide high sensitivity in the sub-threshold region as a channel material in nano-biosensor, a trend demonstrated by silicon on insulator FET sensor in the literature. Thus, going by the trend of replacing silicon with these novel materials in device level, TMDC heterostructure could be a viable alternative to silicon for potentiometric biosensing.
Keywords: 2D Material, NEGF, Junctionless FET, Quantum Ballistic Simulation, Short Channel Effects, Debye -Screening, Potentiometric Biosensor.
ACCEPTED MANUSCRIPT 1. Introduction
TE D
M AN U
SC
RI PT
With the MOSFET dimensions shrinking, we no longer can consider long channel behavior due to undesirable side effects [1]. The depletion widths of source and drain junctions are becoming comparable to the channel length because of the shortening of channel length. This readily causes the punch-through effect [2]. Increasing channel doping can come to the rescue. However, it will increase the threshold voltage. Scaling down of gate oxide thickness can be a solution to the problem. Again, tunneling and other quantum mechanical effects narrow down the chances of ideal scaling rule. Hence, with best scaling control, optimization of the scaled down devices is not possible with the unwarranted so-called short channel effects (SCE). Many techniques such as Gaussian doping profile, gate stack, source/drain design, FinFET architectures with multi-gates, and alternative channel material beside silicon have been proposed to overcome various SCEs and enhance device performance [2-4]. However, there are manufacturing problems related to some of the alternative techniques. Junctionless Field Effect Transistors (JLFETs) is thought to overcome fabrication challenges related to abrupt p-n junctions in state-of-the-art CMOS technology which might open a new era in the technology of nanoscale MOSFETs [5-7]. With the seminal work done by J. P. Colinge et al. [8], much more research on this new concept was done [5, 9-10]. JLFET is reported to show excellent subthreshold slope (SS), small drain induced barrier lowering (DIBL), low leakage current, high ION/IOFF ratio, etc. [5, 8-12]. For inversion mode device, inversion channel at the interface between channel and gate oxide is formed above the threshold voltage, and flat band voltage is below the threshold voltage. In contrast, for JLFET which is an accumulation-mode device, flat-band voltage is above the threshold voltage, and the device is fully depleted below threshold voltage. Numerous analytical models and simulations have been done for JLFETs with various geometries for bulk materials.
AC C
EP
On the other hand,because of highly scaled thickness up to an atomic plane and pristine surface without dangling bond [13-14], 2-D semiconducting transition metal dichalcogenides (TMDC), such as MoS2 and WSe2, have been considered as prospective channel materials for low power CMOS devices [15]. Monolayer TMDCs such as MoS2 and WSe2 have an appreciable band gap (1.8 and 1.6 eV respectively [6]) resulting in higher ION/IOFF ratio than the zero band gap graphene [7]. This property makes these materials suitable for low-power logic applications. In this work, we have done a quantum ballistic evaluation of some widely used monolayer TMDC (MoS2, MoSe2, WS2, and WSe2) based JLFETs. Although some assessments of MoS2 JLFET can be found in recent literatures [16-17], it was done mainly to show the superiority of JLFET over conventional FET while the analysis with the other TMDC materials is still missing. Hence a detailed comparative study of different TMDC JLFETs would provide better understanding and optimization of their performance. Besides its application in logic circuit, the excellent performance of TMDC JLFET has led to the further investigation of this structure in ‘More than Moore Technology’ in this work. Integrating the task of biochemical sensing with microelectronics and integrated circuits has been an extensive field of research in recent past. Specifically, the early detection of biomarkers indicative of lethal diseases like cancers has necessitated the need for the design of robust, selective and highly sensitive biosensors. Electronic detection of biomolecule at a very low concentration with little or no sample preparation has been an extensive field of research in recent literature. Potentiometric biosensors
ACCEPTED MANUSCRIPT have been found to detect analytes like vascular endothelial growth factor (VEGF) molecules even in femtomolar limits in a dry environment [18]. Moreover, potentiometric biosensing in the dry environment has shown better sensitivity over a wet environment where the sensitivity is limited by Debye screening caused by electrolyte ions [19]. Given the prospects of a potentiometric biosensor in detecting bio-analytes, there has been a highly prospective field of ongoing research regarding the performance improvement of this type of sensor.
AC C
EP
TE D
M AN U
SC
RI PT
Layered van der Waals materials, such as transition metal dichalcogenides, few layers thick or exfoliated down to a single layer, have become the subject of extensive research in recent times [20-21]. Stacking multiple layers on top of each other of these materials also leads to interesting changes in electronic properties [22-23]. Besides bilayer 2D heterostructures, trilayer TMDC heterostructures based on MoS2 have also been studied using first principle simulations [24]. Though monolayer and trilayer TMDC materials have been investigated [25-26] for potentiometric biosensing in a microfluidic environment to detect charged biomolecules, such biosensors cannot detect charge neutral biomolecules and suffer from sensitivity reduction caused by Debye screening. Dielectric Modulated FET (DMFET) approach employed in this work for biomolecules detection is different from other familiar methods found in recent literature regarding detection technique and device structures. For example, A. Gao [27] et al. presented a novel biosensor based on a silicon nanowire tunneling field-effect transistor (SiNW-TFET). It was experimentally demonstrated for point of care diagnostics. It functionalizes the surface of SiNW to capture biomolecule, not like conventional DMFET. In another experimental work [28], F. Puppo et al. showed the capability of SiNW-FET to detect antigen in brain tumor extract with high sensitivity specifically. Graphene-based FET used as an aptasensor has been reported as effective in lead detection in children blood [29]. Gold nanoparticle-decorated graphene FET biosensor [30] is reported to have the capability of achieving label-free, ultrasensitive and highly selective detection of miRNA with high sensitivity as well as a high selectivity. Sub-femtomolar cancer biomarker detection is also achieved with these G-FETs [31]. However, these transistor-based sensors work on the watery environment. In an aqueous environment, the charge of the bioparticle to be detected is partially screened by the different ions presents in the electrolyte solution [19] which is well known as Debye screening. Moreover, the charge of such biomolecules in an aqueous environment is highly dependent upon the pH of the solution which requires a very well controlled pH based solution so that biomolecules contain detectable charges. Hence a new detection technique based on the dielectric permittivity of biomolecules has been found in recent literature which is free from the problems of aqueous solution based bio-detection. Specifically, R. Narang et al. [32-33] has proposed Dielectric Modulated Tunnel FET as a biosensor and studied analytical modeling and sensitivity analysis for label-free detection. Tunnel FET sensor is expected to result in better sensitivity compared to FET-based sensor [33]. Various metrics like I on , SS , ∆Vth etc. have been investigated [33] to measure sensitivity. Being inspired by their works, we have further investigated the prospect of such biomolecule detection technique in a different device structure with an entirely different channel made of 2D materials. In this work, we focused on ∆I , a metric which has been I
found to provide the best sensitivity among all.
ACCEPTED MANUSCRIPT Hence, the ultimate objective of this work is to analyze the feasibility of 2D material FET for both logic and sensor applications. 2D materials are highly sought today for its superior performance as a replacement of Si in current MOSFET technology. With the promise of replacing Si in FET, will these TMD be able to work as sensor material also? This work is focused on finding that answer too.
2. Device Structure and Simulation Approach
M AN U
SC
RI PT
2.1 As MOSFET
TE D
Figure 1. Schematic of the device used in this work as MOSFET. Monolayer TMDCs are MoS2, MoSe2, WS2, and WSe2.
AC C
EP
Quantum transport analysis is necessary for these type of materials (MX2) and dimensions of interest. Self-consistent solution of Poisson and Schrodinger equation in an NEGF framework with tight binding (TB) Hamiltonian is needed to capture the phenomena of ballistic transport in these emerging material devices. The electrical characteristics of Junction-Less TMDC FETs are simulated using two-band tight binding (TB) Hamiltonian with an open-source quantum transport simulation framework [34-35]. Simulation procedure is same as those of [17] [36] using the same context. We have benchmarked the Id-Vg characteristics of MoS2 JLFET with that of [17] (not shown) for 30 nm gate length using the same parameter values given in that paper. However, this benchmarking is not surprising since both ref. [17] and this work have used the same open source platform, NanoTcad ViDES [34-35], which is an established platform to simulate 2D material FET. For Fig. 1, Channel is monolayer transition metal dichalcogenides (MoS2, MoSe2, WS2, and WSe2) with a thickness of ~0.7 nm. Channel length L is varied from 5 to 30 nm. Top and bottom oxide (both HfO2) thicknesses are kept 4.9 nm. The doping used is 5 x 10-3 as a molar fraction, which is defined as the ratio between the numbers of doping atoms over the number of atoms [37]. Parameters for monolayer TMDCs have been taken from various sources [38-42] as listed in Table 1. In this work, top gate and bottom gate are connected. Therefore, VTG and VBG are same and named as gate voltage, VG. Source and drain contact are assumed Ohmic which is possible by proper contact engineering [43].
ACCEPTED MANUSCRIPT Table 1. Monolayer TMDC parameters Material
Electron Effective
Band gap
Mass
Dielectric Constant
0.56
1.89
4.8
WSe2
0.35
1.61
4.5
MoSe2
0.62
1.58
6.9
WS2
0.33
2.05
4.4
2.2 For Biosensor
RI PT
MoS2
TE D
M AN U
SC
For biosensor application tri-layer TMDC material has been considered for the channel. The motivation to study tri-layer TMDC JFET for bio-sensor demands some explanations. We have previously investigated the prospects of multilayer and monolayer TMDC based biosensor in an aqueous environment [25-26]. From those study it has been found that the sensitivity of layered TMDC biosensors is considerably higher [26] than that of the monolayer TMDC biosensor [25] while detecting identical biomolecules under similar aqueous environment. A possible reason for such higher sensitivity can be attributed to the relatively small bandgap of trilayer TMDC material compared to that of monolayer which might have resulted in a larger change in sub-threshold current in response to the capture of a biomolecule. Besides, the multilayer MoS2 structure had shown better noise immunity than a single-layer case in the air [44], an additional benefit of using multi-layer structures for sensor applications. It is very easy to mask the performance of a sensor with noise. Even for gas sensor [45], single-layer (1L) MoS2 device showed an unstable response. In our proposed device, both top and bottom layer are connected to receptors for biomolecules and gate metal respectively. Hence, the middle layer is considered free from external environment suitable for capturing the effect of attaching biomolecules on the other layer.
AC C
EP
Fig. 2 shows the device structure proposed in this work for biomolecule sensing. The dimensions and material parameters of various layers are described in detail in Table 2. To sense biomolecule, a part of the top oxide layer has been etched to form a cavity where proper bio-receptors can be functionalized to provide selective binding of the biomolecule of interest. In the absence of target biomolecules in the nanogap cavity region, the holes are filled with air, so dielectric permittivity of the cavity region is different from that of top oxide and the threshold voltage changes from its initial values. When target biomolecules (like Streptavidin, Biotin, Avidin, enzyme, cell, DNA, APTES) are present and immobilized at the binding site, the dielectric permittivity changes and the gate capacitance of the device also changes. Consequently, electrical characteristics such as threshold voltage and current of the device changes per the dielectric constant or charge of the target biomolecules. To investigate the effect of biomolecule on device’s electrical response, quantum transport analysis is done. Same self-consistent simulation approach in an NEGF framework with tight binding (TB) Hamiltonian has been used to capture the phenomena of ballistic transport in these emerging material devices using open source platform, NanoTcad ViDES. Material parameters have been taken from literature [46].
RI PT
ACCEPTED MANUSCRIPT
SC
Fig.2. The DMFET structure under consideration. It has a trilayer TMDC material channel sandwiched between the top and bottom oxides and corresponding upper and lower gates. The channel, the source, and drain are highly n-doped
M AN U
regions of the same 2D material. Top oxide is etched on one side to simulate the effect of the biomolecule on the conductivity of the FET.
Table 2. Parameters used in this work for trilayer TMDC DMFET Parameter
TE D
Channel Thickness [18]
Value 2 nm 20 nm
Top Oxide Length (L2)
20 nm
Channel Length
40 nm
Channel, Source and Drain Doping
2.2 x 1016 m-2
EP
Cavity Length (L1)
Top Oxide Thickness (ZrO2)
9 nm
AC C
(Biomolecule height)
Bottom Oxide Thickness (SiO2)
10 nm
Trilayer TMDC Electron Effective Mass [46]
0.52 x 9.1 x 10-31 kg
Trilayer TMDC Dielectric Permittivity
5.2 x 8.854 x 10-12 Fm-1
ZrO2 Dielectric Permittivity
12.5 x 8.854 x 10-12 Fm-1
SiO2 Dielectric Permittivity
3.9 x 8.854 x 10-12 Fm-1
Biomolecule dielectric constant
(3 to 9) x 8.854 x 10-12 Fm-1
ACCEPTED MANUSCRIPT To predict the sensitivity of trilayer TMDC FET, we have used the NEGF current simulator [34-35]. For measuring sensitivity, we have used the following formulae [47] S=
∆I d
=
Id
I d ( bio ) − I d ( air )
min ( I d ( bio ) , I d ( air ) )
3.1 Ballistic Simulation of Monolayer TMDC MOSFET
RI PT
3. Result and Discussion
d
10
10
-5
-0.2 0 Gate Voltage, Vg(V)
WSe 2
Vd=0.05V Vd=0.5V
-10
-0.4
-0.2 0 Gate Voltage, Vg(V)
SC
-5
10
MoSe 2
-0.4
0.2
10
Vd=0.05V Vd=0.5V
-10
0.2
TE D
-0.4
Drain Current, I (A/µ m)
d
-10
10
M AN U
Vd=0.05V Vd=0.5V 10
Drain Current, I (A/µ m)
MoS2
d
-5
Drain Current, I (A/µ m)
10
EP
d
Drain Current, I (A/µ m)
3.1.1 Drain Current (Id) - Gate Voltage (Vg) Characteristics
-5
-0.2 0 Gate Voltage, Vg(V)
0.2
WS2
Vd=0.05V Vd=0.5V 10
-10
-0.4
-0.2 0 Gate Voltage, Vg(V)
0.2
AC C
Figure 3. Id-Vg characteristics of various monolayer TMDC Junction less FETs for two different drain voltages. In this case, the channel length is 30 nm for all the devices.
Fig. 3 presents the drain current-gate voltage characteristics of four monolayer FETs considered for two different drain voltages. All the devices show high “ON Current” while providing very small subthreshold current. The physics of interband tunneling has been exploited in MOSFETs to design low power switching devices. Although MOSFET is a barrier-controlled device, interband tunneling could still affect device performance in the subthreshold regime when carrier effective mass and band gap are lowered in the channel [48] which is beyond the scope of the present work.
ACCEPTED MANUSCRIPT
3.1.2 Short Channel Effects Id-Vg curve is critical in evaluating short channel characteristics i.e. Drain induced barrier lowering (DIBL), Subthreshold Swing (SS), Threshold voltage roll-off (VTRO), etc.
RI PT
3.1.2.1 Drain Induced Barrier Lowering (DIBL) DIBL is the reduction of threshold voltage with the increase of the drain voltage. From Fig. 3, we extract the values of DIBL for all materials in Table 3. DIBL is the lowest for a FET when top of the barrier from source to drain is least affected by drain voltage. Semiconductor FETs is usually defined
t s tb ε s
εb
, where ts is the
SC
by a particular term called ‘natural scaling length’ given by λ =
M AN U
semiconductor thickness, tb is the gate dielectric thickness, εs and εb are the semiconductor and gate insulator dielectric constants respectively. Small natural length predicts less short channel effects like DIBL. Because of the atomic layer thickness of these materials which results in small natural length, these FETs show strong immunity from the effect of drain to source voltages compared to bulk MOSFET. The threshold voltage is measured using the constant current method. Here reference current used is 1*10-7 A/µm. Table 3. Extracted DIBL values
TE D
Material
30 nm Gate Length DIBL (mV/V) 18
MoSe2
10.88
WS2
11.55
WSe2
14.88
AC C
EP
MoS2
3.1.2.2 Subthreshold Swing (SS) The subthreshold swing is defined as the gate voltage required to change the drain current by one order of magnitude when the gate voltage is below the threshold voltage. In the MOSFET, the subthreshold swing is limited to 60 mV/Dec at room temperature. Since one carrier takes part in conduction in FET (except Tunnel FET where both hole and electrons involve in conduction, and SS can be below 60 mV/Dec for TFET), it is not possible to breach this barrier. However, JLFET is better suited in this context. In our calculations, for 15 and 30 nm gate lengths, we get SS close to ideal ~60 mV/Dec. However, as we shorten the gate length, we get >60 mV/Dec SS for 5 nm gate length as presented in Table 4. SS is measured from Fig 4 for 5 nm gate length. All devices show SS in the region of 76~78 mV/Dec for 5nm gate length.
ACCEPTED MANUSCRIPT
Table 4. Extracted SS values 5 nm Gate Length
Material
SS (mV/Dec) 77.1
MoSe2
77.7
RI PT
MoS2 WS2
76.3
WSe2
77.6
M AN U
SC
The range of SS values reported in this work for various materials is very small indicating all these materials are viable options as channel material in a MOSFET. It has been reported in the recent literature [49-50] that for various FETs with 2D material as channel with an effective mass of meff >=0.3 m0, it becomes difficult to differentiate Id-Vg curves in the subthreshold region. On the other hand, for meff ≥ 0.2×m0 [50] excellent subthreshold behavior and a good suppression of source-drain tunneling and short channel effects are obtained for a gate length of 5 nm. Our simulation result of comparable SS for various materials where effective mass is bigger than 0.3 m0 for all materials proves that the trend is similar for Junction less FET too.
3.1.2.3 Threshold Voltage Roll Off (VTRO)
TE D
In this work threshold voltage roll-off is defined as VTRO = Threshold voltage (VT) at LG = LREF Threshold voltage (VT) at LG = LNEW. Here new gate lengths considered are 15 nm and 5 nm while the reference length is 30 nm. Table 5 shows the corresponding values measured from Fig 4.
EP
Table 5. Extracted VTRO (mV) values 15 nm Gate Length
5 nm Gate Length
MoS2
42.4
308
MoSe2
42.6
306
WS2
43.95
310
WSe2
43.2
307.5
AC C
Material
WSe 2
10
-5
-0.6
Lg=5nm Lg=15nm Lg=30nm -0.4 -0.2 0 0.2 Gate Voltage, Vg(V)
10
-5
-0.6
WS2
10
-5
Lg=5nm Lg=15nm Lg=30nm -0.4 -0.2 0 0.2 Gate Voltage, Vg(V)
RI PT
Drain Current, Id (A/µ m)
-0.6
Lg=5nm Lg=15nm Lg=30nm -0.4 -0.2 0 0.2 Gate Voltage, Vg(V)
MoSe 2
SC
-5
Drain Current, Id (A/µ m)
10
Drain Current, Id (A/µ m)
MoS2
Lg=5nm Lg=15nm Lg=30nm -0.4 -0.2 0 0.2 Gate Voltage, Vg(V)
M AN U
Drain Current, Id (A/µ m)
ACCEPTED MANUSCRIPT
-0.6
Figure 4. Id-Vg characteristics of various monolayer TMDC Junctionless FETs for multiple gate lengths.
AC C
EP
TE D
From the short channel characteristics presented in this work, it can be concluded that transition metal dichalcogenides based FETs have much steeper sub-threshold slopes compared to Si devices, essentially because of their much smaller thicknesses leading to stronger gate controllability and higher immunity to short-channel effects. These results indicate that TMDC monolayer junction-less FET holds the promise of the shorter transistor with less non-ideal effects. Reported DIBL from 10 mV/V to 18 mV/V, SS in the region of 76~77 mV/Dec are encouraging values for next generation MOSFETs. Considering the range over which SS, DIBL, and VTRO of different monolayer TMDC materials used in this work vary, we can consider all of them to be a potential candidate to cope up with well-known short channel effects.
3.2 Application in Biosensor Fig. 5 shows how the channel barrier from source to drain is modulated by the dielectric permittivity of the biomolecules in the cavity. As the biomolecule’s dielectric permittivity increases, the channel potential barrier also increases since the device under study is an accumulation mode JLFET. Thus, current decreases with increase in biomolecule dielectric constant. The bio sensitivity of the proposed device can be monitored from device current as shown in Fig. 6. A detectable change in device current is found for variation in biomolecule dielectric permittivity specifically in the subthreshold and linear region. The change in device current can be seen more clearly in the zoomed portion of Fig. 6 for different biomolecule permittivity. Fig. 7 shows the sensitivity of the proposed dielectric modulated TMDC JLFET. We found that the highest sensitivity is obtained when the device is in subthreshold region whereas the sensitivity decreases as the device gradually moves into
ACCEPTED MANUSCRIPT the linear regime. Hence, to achieve maximum sensitivity, the proposed devices should be operated in the subthreshold regime which is also consistent with other biosensors [51].
= ε0
ε bio
= 3ε 0
ε bio
= 5ε 0
ε bio
= 7ε 0
RI PT
0
ε bio
VGS=-0.3V
-0.05
-0.15 0
1
SC
-0.1
2 3 Channel Position (nm)
4
M AN U
Channel Potential Barrier (eV)
0.05
x 10
-8
Fig. 5. Channel potential barrier of dielectric modulated trilayer TMDC FET biosensor obtained from NEGF current
ε =ε bio 0 ε = 3ε bio 0 -6
ε = 5ε bio 0 ε = 9ε bio 0
10
10
10
10
-8
-10
-12
-8
EP
ε = 7ε bio 0
AC C
Drain Current, Id (A/µ m)
10
-0.5
-0.4
-0.3
=ε
bio
= 3ε
0
ε
bio
= 5ε
0
ε
bio
= 7ε
0
ε
bio
= 9ε
0
0
-12
10
-13
10 -0.65
10
-0.6 -0.55 Gate Voltage, Vg(V)
-8
10
-12
-14
10 -0.2
bio
ε
-10
10
-14
-0.6
10
ε
Drain Current, Id (A/µ m )
-6
10
Drain Current, Id (A/µ m )
-4
Drain Current, Id (A/µ m )
10
TE D
simulator [34-35] for the different dielectric permittivity of the biomolecule immobilized in the cavity.
-9
10
-0.6 -0.5 -0.4 -0.3 Gate Voltage, Vg(V)
-10
10
-0.5
-0.45 -0.4 Gate Voltage, Vg(V)
Gate Voltage, Vg(V)
Fig. 6. Id- Vg characteristics of dielectric modulated trilayer TMDC FET. The left figure shows Id-Vg characteristics for a broad range of gate voltages while data on the right are zoomed in for low gate voltages to demonstrate the difference in currents in the subthreshold region for biomolecules with different dielectric permittivity.
ACCEPTED MANUSCRIPT 80
ε
70
= 3ε
0
ε
bio
= 5ε
0
ε
bio
= 7ε
0
bio
= 9ε
0
ε
60 50 40
RI PT
Sensitivity,S
bio
30 20 10 -0.6
-0.55 -0.5 -0.45 Gate Voltage, VG (V)
-0.4
-0.35
-0.3
SC
-0.65
M AN U
Fig. 7. Sensitivity in the subthreshold region for various biomolecule dielectric permittivity. Sensitivity decreases with increase in gate voltage. A higher dielectric constant of biomolecule results in higher sensitivity.
4. Conclusion
AC C
EP
TE D
In this work, we have performed a ballistic simulation study on the performance of some prominent monolayer TMDC as channel materials in MOSFETs from 30 nm to sub-10 nm region operation. For gate length below 10 nm, SS values are close to the ideal limit for all the FETs incorporating these materials. Another important short channel effect, threshold voltage roll off is also measured. The extracted values will be helpful to designers in choosing the appropriate material for various applications. Moreover, we have also performed a quantum ballistic simulation study on the performance of Van Der Waals trilayer heterostructure as a channel material in biosensing operation using the principle of Dielectric Modulation. Our research shows that these materials could be viable options in implementing FET-based biosensor in a dry environment because of their high sensitivities (up to 80 percent), especially in the subthreshold regime. Moreover, the sensitive nature of the proposed biosensor can be maximized through operation in the subthreshold region which follows the trend of biosensing found in recent literature. We have chosen a trilayer hetero structure as a channel material for DMFET analysis. It would be interesting following work to do a comparative analysis of dielectric modulated biosensor in a TFET structure [33] with varying no. of layers. For more detailed analysis on the sensitivity of biosensor, the effect of biomolecule position and fill in factor in the cavity region on the sensitivity can be studied [52].
Conflict of interest The authors declare that there is no conflict of interest regarding the publication of this paper.
ACCEPTED MANUSCRIPT
References S. M. Sze and K. K. Ng, “Physics of semiconductor devices,” Wiley, pp. 328-338, 2006. R. S. Muller, T. I. Kamins and M. Chan, “Device Electronics for Integrated Circuits,” John Wiley and Sons, Inc., pp. 420-439, 2003. [3] International Technology Roadmap for Semiconductors, 2012: http://www.itrs.net. [4] M. V. Haartman and M. Ostling, “Low-frequency noise in advanced MOS devices,” Springer, 2007. [5] C. W. Lee, A. Afzalian, N. D. Akhavan, R. Yan, I. Ferain, and J. P. Colinge, “Junctionless multigate field-effect transistor,” Applied Physics Letters, vol. 94, no. 5, pp. 97–103, 2009. [6] A. M. Ionescu, “Electronic devices: nanowire transistors made easy,” Nature Nanotechnology, vol. 5, no. 3, pp. 178-179, 2010. [7] J. P. Colinge, F. Balestra, J. P. Raskin, F. Gamiz and V. S. Lysenko, “Semiconductor-On-Insulator Materials for Nanoelectronics Applications,” Springer, pp. 187-200, 2011. [8] J. P. Colinge, C. W. Lee, A. Afzalian, et al., “Nanowire transistors without junctions,” Nature Nanotechnology, vol. 5, no. 3, pp. 225-229, 2010. [9] C. W. Lee, I. Ferain, A. Afzalian, et al., “Performance estimation of junctionless multigate transistors,” Solid-State Electronics, vol. 54, pp. 97–103, 2010. [10] J. P. Colinge, A. Kranti, R. Yan, et al., “Junctionless Nanowire Transistor (JNT): Properties and design guidelines,” Solid-State Electronics, vol. 65-66, pp. 33–37, 2011. [11] R. Doria, M. Pavanello, R. Trevisoli et al., “Junctionless Multiple-Gate Transistors for Analog Applications,” IEEE Transactions on Electron Devices, vol. 58, no. 8, pp. 2511–2519, 2011. [12] J. P. Colinge, C. W. Lee, I. Ferain et al., “Reduced electric field in junctionless transistors,” Applied Physics Letters, vol. 96, no. 7, p. 073510, 2010. [13] W. Cao, J. Kang, S. Bertolazzi, A. Kis, and K. Banerjee, “Can 2D nanocrystals extend the lifetime of floating-gate transistor based nonvolatile memory?” IEEE Trans. Electron Devices, vol. 61, no. 10, pp. 3456–3464, Oct. 2014. [14] W. Cao, J. Kang, W. Liu Y. Khatami, D. Sarkar, and K. Banerjee, “2D Electronics: Graphene and Beyond,” in Proc. ESSDERC, Sep. 2013, pp. 37–44. [15] W. Cao, J. Kang, W. Liu, and K. Banerjee, “A compact current voltage model for 2D semiconductor based field-effect transistors considering interface traps, mobility degradation, and inefficient doping effect,” IEEE Trans. Electron Devices, vol. 61, no. 12, pp. 4282–4290, Dec. 2014. [16] Wei Cao et al., “Junction-Less Monolayer MoS2 FETs”, arXiv:1509.00561. [17] Tarun Agarwal et al., “Comparison of short-channel effects in monolayer MoS2 based junctionless and inversion-mode field-effect transistors”, Applied Physics Letters, 108, 023506 (2016); DOI: 10.1063/1.4939933. [18] F. Puppo, M.-A. Doucey, T. S. Y. Moh, G. Pandraud, P. M. Sarro, G. De Micheli, and S. Carrara, “Femto-molar sensitive field effect transistor biosensors based on silicon nanowires and antibodies,” IEEE Sensors, pp. 1–4, 2013. [19] T. J. Park, J. Kim, J. Ahn, S. Choi, M. Im, S. Kim, J. P. Duarte, and C. Kim, “An Underlap Channel-Embedded Field-Effect Transistor for Biosensor Application in Watery and Dry Environment,” IEEE Trans. Nanotechnology, vol. 11, pp. 390–394, 2012.
AC C
EP
TE D
M AN U
SC
RI PT
[1] [2]
ACCEPTED MANUSCRIPT
AC C
EP
TE D
M AN U
SC
RI PT
[20] J. A. G. Sheneve Z. Butler, Shawna M. Hollen, Linyou Cao, and Yi Cui, “Progress, Challenges, and Opportunities in Two-Dimensional Materials Beyond Graphene,” ACS Nano, vol. 7, no. 4, pp. 2898–2926, 2013. [21] C. N. R. Rao, K. Gopalakrishnan, and U. Maitra, "A Comparative Study of Potential Applications of Graphene, MoS2, and Other Two-Dimensional Materials in Energy Devices, Sensors, and Related Areas", ACS Applied Materials and Interfaces, vol. 7, no. 15. 2015. [22] H. Terrones, F. López-Urías, and M. Terrones, “Novel hetero-layered materials with tunable direct band gaps by sandwiching different metal disulfides and diselenides,” Sci. Rep., vol. 3, p. 1549, 2013. [23] N. Lu, H. Guo, L. Li, J. Dai, L. Wang, W.-N. Mei, X. Wu, and X. C. Zeng, “MoS2/MX2 hetero layers: bandgap engineering via tensile strain or external electrical field.,” Nanoscale, vol. 6, no. 5, pp. 2879–86, 2014. [24] N. Lu, H. Guo, L. Wang, X. Wu, and X. C. Zeng, “Van der Waals trilayers and superlattices: modification of electronic structures of MoS2 by intercalation,” Nanoscale, pp. 4566–4571, 2014. [25] Abir Shadman, Ehsanur Rahman, and Quazi D. M. Khosru, 2016, “Monolayer MoS2 and WSe2 Double Gate Field Effect Transistor as Super Nernst pH sensor and Nanobiosensor,” Sensing and Bio-Sensing Research, vol. 11, no. 1, pp. 45–51, 2016. [26] Kanak Datta, Abir Shadman, Ehsanur Rahman, and Quazi D. M. Khosru, 2017, “Trilayer TMDC Heterostructures for MOSFETs and Nanobiosensors,” Journal of Electronic Materials, vol. 46, no. 2, pp. 1248–1260, 2017. [27] Anran Gao, Na Lu, Yuelin Wang, and Tie Li., ‘Robust ultrasensitive tunneling FET biosensor for point-of-care diagnostics,' Nature Scientific Reports 6, Article number: 22554 (2016). DOI: 10.1038/srep22554 [28] F. Puppo et al., ‘SiNW-FET in-Air Biosensors for High Sensitive and Specific Detection in Breast Tumor Extract,' IEEE Sensors Journal, Vol. 16, No. 10, pp 3374 – 3381, May 15, 2016. [29] Wang, C. et al. ‘A label-free and portable graphene FET aptasensor for children blood lead detection,' Nature Scientific Reports 6, 21711; DOI: 10.1038/srep21711. [30] B, Cai. et al., ‘Gold nanoparticles-decorated graphene field-effect transistor biosensor for femtomolar MicroRNA detection,' Biosensors and Bioelectronics, Volume 74, pp 329–334, 15 December 2015. [31] C-H Yeh et al., ‘High-performance and high-sensitivity applications of graphene transistors with self-assembled monolayers,' Biosensors and Bioelectronics, Volume 77, pp 1008–1015, 15 March 2016. [32] Rakhi Narang, Manoj Saxena, R. S. Gupta, and Mridula Gupta, “Dielectric Modulated Tunnel Field Effect Transistor – A Biomolecule Sensor,” IEEE Electron Device Letters, Vol. 33, no. 2, pp. 266-268, February 2012. [33] R. Narang, K. V. S. Reddy, M. Saxena, R. S. Gupta and M. Gupta, "A Dielectric-Modulated Tunnel-FET-Based Biosensor for Label-Free Detection: Analytical Modeling Study and Sensitivity Analysis," in IEEE Transactions on Electron Devices, vol. 59, no. 10, pp. 2809-2817, Oct. 2012. [34] “NanoTCAD ViDES.’ [Online]. Available: http://vides.nanotcad.com/. [Accessed: 01Aug-2013]”. [35] “NanoTCAD ViDES.’ [Online]. Available: http://vides.nanotcad.com/vides/. [Accessed: 01-Sep-2015]”.
ACCEPTED MANUSCRIPT
AC C
EP
TE D
M AN U
SC
RI PT
[36] Tarun Agarwal et al., “Benchmarking of MoS2 FETs With Multigate Si-FET Options for 5 nm and Beyond,” IEEE Trans. Electron Devices, vol. 62, no. 12, pp. 4051–4057, Dec. 2015. [37] User Manual, NanoTCAD ViDES, 2008 (http://vides. nanotcad. com/vides/documentation/commands-5 /dope reservoir) [38] Ram Krishna Ghosh and Santanu Mahapatra, “Monolayer Transition Metal Dichalcogenide Channel-Based Tunnel Transistor,” IEEE Journal of the Electron Device Society, Vol. 1, No. 10, pp. 175-181, October 2013. [39] Cheng Gong et al., “Band alignment of two-dimensional transition metal dichalcogenides: application in tunnel field effect transistors,” DOI: 10.1039/c5nr06782k. [40] S. U. Z. Khan and Quazi D. M. Khosru, “Quantum Mechanical Electrostatics and Transport Simulation and Performance Evaluation of Short Channel Monolayer WSe2 Field Effect Transistor”, ECS Transaction, Vol. 66 (14), pp. 11-18, 2015. [41] W. Liu, W. Cao, J. Kang, and K. Banerjee, “High-Performance Field-Effect-Transistors on Monolayer WSe2”, Vol. 58 (7), pp. 281-285, 2013. [42] Jiwon Chang, Leonard F. Register and Sanjay K. Banerjee, “Comparison of Ballistic Transport Characteristics of Monolayer Transition Metal Dichalcogenides (TMDs) MX2 (M = Mo, W; X = S, Se, Te) n-MOSFETs”, Proc. of the IEEE SISPAD, pp. 408-411, 2013. [43] W. Liu, J. Kang, W. Cao, D. Sarkar, Y. Khatami, D. Jena, and K. Banerjee, “High-performance few-layer-MoS2 field-effect-transistor with record low contact-resistance,” in IEEE Int. Elect. Dev. Meeting, pp. 499−502, Dec. 2013. [44] H-J Kwon et al., ‘Analysis of flicker noise in two-dimensional multilayer MoS2 transistors’ Applied Physics Letters 104, 083110 (2014). [45] Ha Li et al., ‘Fabrication of Single- and Multilayer MoS2 Film Based Field-Effect Transistors for Sensing NO at Room Temperature’, Small, vol. 8, No. 1, pp 63–67, 2012. [46] K. Datta and Q. D. M. Khosru, “Electronic Properties of MoS2 / MX2 /MoS2 Trilayer Heterostructures: A First Principle Study,” JSS vol. 5, no. 11, pp. 3001–3007, 2016. [47] D. Sarkar, W. Liu, X. Xie, A. C. Anselmo, S. Mitragotri, and K. Banerjee, “MoS2 Field-Effect Transistor for Next- Generation Label-Free Biosensors,” ACS Nano, 2014. [48] N. Ma and D. Jena, “Interband Tunneling in two-dimensional crystal semiconductors,” Appl. Phys. Lett., vol. 102, no. 13, 2013. [49] Wei Cao et al., 2D Semiconductor FETs—Projections and Design for Sub-10 nm VLSI, IEEE Transaction on Electron Devices, pp. 3459-3469, Vol. 62, No. 11, Nov 2015. [50] S. Thiele et al., The Prospects of Two-Dimensional Materials for Ultimately Scaled CMOS. Available Online: http://eurosoi-ulis2017.inn.demokritos.gr/files/abstracts/abstracts_Session9.pdf [51] J. Lee, P. Dak, Y. Lee, H. Park, W. Choi, M. A. Alam, and S. Kim, “Two-dimensional Layered MoS2 Biosensors Enable Highly Sensitive Detection of Biomolecules,” Sci. Rep., pp. 1–7, 2014. [52] Ehsanur Rahman, Abir Shadman, and Quazi D. M. Khosru, 2017, “Effect of biomolecule position and fill in factor on sensitivity of a Dielectric Modulated Double Gate Junctionless MOSFET biosensor,” Sensing and Bio-Sensing Research, vol. 13, pp. 49–54, 2017.
ACCEPTED MANUSCRIPT
AC C
EP
TE D
M AN U
SC
RI PT
Junctionless Field Effect Transistors (JLFETs) is thought to overcome fabrication challenges related to abrupt p-n junctions due to channel length scaling in state-of-the-art CMOS technology which might open a new era in the technology of nanoscale MOSFETs. Because of highly scaled thickness up to an atomic plane and dangling bond free pristine surface, 2-D semiconducting transition metal dichalcogenides (TMDC), such as MoS2 and WSe2, have been considered as prospective channel materials for low power CMOS devices. Due to these advantages, in this work, we have performed a simulation based study of Transition Metal Dichalcogenides Based Double Gate Junctionless Field Effect Transistor to evaluate their performance in short channel regime scaled down to 5nm.We have also investigated the prospect of these TMDC JLFETs in Debye screening free potentiometric biosensing using the technique of dielectric permittivity modulation of the cavity region formed in the oxide for biosensing. Our study shows considerable biomolecule detection capability at dry environment even without any charge with optimal sensor operation in the subthreshold region.
S0749-6036(17)30805-4
DOI:
10.1016/j.spmi.2017.06.055
Reference:
YSPMI 5104
To appear in:
Superlattices and Microstructures
Received Date: 1 April 2017 Revised Date:
21 June 2017
Accepted Date: 22 June 2017
Please cite this article as: A. Shadman, E. Rahman, Q.D.M. Khosru, Quantum ballistic analysis of transition metal dichalcogenides based double gate junctionless field effect transistor and its application in nano-biosensor, Superlattices and Microstructures (2017), doi: 10.1016/j.spmi.2017.06.055. This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
ACCEPTED MANUSCRIPT
Quantum Ballistic Analysis of Transition Metal Dichalcogenides Based Double Gate Junctionless Field Effect Transistor and its Application in Nano-biosensor
Dept. of Electrical and Electronic Engineering, Bangladesh University of Engineering and Technology,
Dhaka-Bangladesh. ¶
These authors contributed equally to this work.
†
Corresponding Author email: [email protected]
SC
1
RI PT
Abir Shadman1¶, †, Ehsanur Rahman1¶, and Quazi D. M. Khosru1
M AN U
Abstract: To reduce the thermal budget and the short channel effects in state of the art CMOS
AC C
EP
TE D
technology, Junctionless field effect transistor (JLFET) has been proposed in the literature. Numerous experimental, modeling, and simulation based works have been done on this new FET with bulk materials for various geometries until now. On the other hand, the two-dimensional layered material is considered as an alternative to current Si technology because of its ultra-thin body and high mobility. Very recently few simulation based works have been done on monolayer molybdenum disulfide based JLFET mainly to show the advantage of JLFET over conventional FET. However, no comprehensive simulation-based work has been done for double gate JLFET keeping in mind the prominent transition metal dichalcogenides (TMDC) to the authors’ best knowledge. In this work, we have studied quantum ballistic drain current-gate voltage characteristics of such FETs within non-equilibrium Green function (NEGF) framework. Our simulation results reveal that all these TMDC materials are viable options for implementing state of the art Junctionless MOSFET with emphasis on their performance at short gate lengths. Besides evaluating the prospect of TMDC materials in the digital logic application, the performance of Junctionless Double Gate trilayer TMDC heterostructure FET for the label-free electrical detection of biomolecules in dry environment has been investigated for the first time to the authors’ best knowledge. The impact of charge neutral biomolecules on the electrical characteristics of the biosensor has been analyzed under dry environment situation. Our study shows that these materials could provide high sensitivity in the sub-threshold region as a channel material in nano-biosensor, a trend demonstrated by silicon on insulator FET sensor in the literature. Thus, going by the trend of replacing silicon with these novel materials in device level, TMDC heterostructure could be a viable alternative to silicon for potentiometric biosensing.
Keywords: 2D Material, NEGF, Junctionless FET, Quantum Ballistic Simulation, Short Channel Effects, Debye -Screening, Potentiometric Biosensor.
ACCEPTED MANUSCRIPT 1. Introduction
TE D
M AN U
SC
RI PT
With the MOSFET dimensions shrinking, we no longer can consider long channel behavior due to undesirable side effects [1]. The depletion widths of source and drain junctions are becoming comparable to the channel length because of the shortening of channel length. This readily causes the punch-through effect [2]. Increasing channel doping can come to the rescue. However, it will increase the threshold voltage. Scaling down of gate oxide thickness can be a solution to the problem. Again, tunneling and other quantum mechanical effects narrow down the chances of ideal scaling rule. Hence, with best scaling control, optimization of the scaled down devices is not possible with the unwarranted so-called short channel effects (SCE). Many techniques such as Gaussian doping profile, gate stack, source/drain design, FinFET architectures with multi-gates, and alternative channel material beside silicon have been proposed to overcome various SCEs and enhance device performance [2-4]. However, there are manufacturing problems related to some of the alternative techniques. Junctionless Field Effect Transistors (JLFETs) is thought to overcome fabrication challenges related to abrupt p-n junctions in state-of-the-art CMOS technology which might open a new era in the technology of nanoscale MOSFETs [5-7]. With the seminal work done by J. P. Colinge et al. [8], much more research on this new concept was done [5, 9-10]. JLFET is reported to show excellent subthreshold slope (SS), small drain induced barrier lowering (DIBL), low leakage current, high ION/IOFF ratio, etc. [5, 8-12]. For inversion mode device, inversion channel at the interface between channel and gate oxide is formed above the threshold voltage, and flat band voltage is below the threshold voltage. In contrast, for JLFET which is an accumulation-mode device, flat-band voltage is above the threshold voltage, and the device is fully depleted below threshold voltage. Numerous analytical models and simulations have been done for JLFETs with various geometries for bulk materials.
AC C
EP
On the other hand,because of highly scaled thickness up to an atomic plane and pristine surface without dangling bond [13-14], 2-D semiconducting transition metal dichalcogenides (TMDC), such as MoS2 and WSe2, have been considered as prospective channel materials for low power CMOS devices [15]. Monolayer TMDCs such as MoS2 and WSe2 have an appreciable band gap (1.8 and 1.6 eV respectively [6]) resulting in higher ION/IOFF ratio than the zero band gap graphene [7]. This property makes these materials suitable for low-power logic applications. In this work, we have done a quantum ballistic evaluation of some widely used monolayer TMDC (MoS2, MoSe2, WS2, and WSe2) based JLFETs. Although some assessments of MoS2 JLFET can be found in recent literatures [16-17], it was done mainly to show the superiority of JLFET over conventional FET while the analysis with the other TMDC materials is still missing. Hence a detailed comparative study of different TMDC JLFETs would provide better understanding and optimization of their performance. Besides its application in logic circuit, the excellent performance of TMDC JLFET has led to the further investigation of this structure in ‘More than Moore Technology’ in this work. Integrating the task of biochemical sensing with microelectronics and integrated circuits has been an extensive field of research in recent past. Specifically, the early detection of biomarkers indicative of lethal diseases like cancers has necessitated the need for the design of robust, selective and highly sensitive biosensors. Electronic detection of biomolecule at a very low concentration with little or no sample preparation has been an extensive field of research in recent literature. Potentiometric biosensors
ACCEPTED MANUSCRIPT have been found to detect analytes like vascular endothelial growth factor (VEGF) molecules even in femtomolar limits in a dry environment [18]. Moreover, potentiometric biosensing in the dry environment has shown better sensitivity over a wet environment where the sensitivity is limited by Debye screening caused by electrolyte ions [19]. Given the prospects of a potentiometric biosensor in detecting bio-analytes, there has been a highly prospective field of ongoing research regarding the performance improvement of this type of sensor.
AC C
EP
TE D
M AN U
SC
RI PT
Layered van der Waals materials, such as transition metal dichalcogenides, few layers thick or exfoliated down to a single layer, have become the subject of extensive research in recent times [20-21]. Stacking multiple layers on top of each other of these materials also leads to interesting changes in electronic properties [22-23]. Besides bilayer 2D heterostructures, trilayer TMDC heterostructures based on MoS2 have also been studied using first principle simulations [24]. Though monolayer and trilayer TMDC materials have been investigated [25-26] for potentiometric biosensing in a microfluidic environment to detect charged biomolecules, such biosensors cannot detect charge neutral biomolecules and suffer from sensitivity reduction caused by Debye screening. Dielectric Modulated FET (DMFET) approach employed in this work for biomolecules detection is different from other familiar methods found in recent literature regarding detection technique and device structures. For example, A. Gao [27] et al. presented a novel biosensor based on a silicon nanowire tunneling field-effect transistor (SiNW-TFET). It was experimentally demonstrated for point of care diagnostics. It functionalizes the surface of SiNW to capture biomolecule, not like conventional DMFET. In another experimental work [28], F. Puppo et al. showed the capability of SiNW-FET to detect antigen in brain tumor extract with high sensitivity specifically. Graphene-based FET used as an aptasensor has been reported as effective in lead detection in children blood [29]. Gold nanoparticle-decorated graphene FET biosensor [30] is reported to have the capability of achieving label-free, ultrasensitive and highly selective detection of miRNA with high sensitivity as well as a high selectivity. Sub-femtomolar cancer biomarker detection is also achieved with these G-FETs [31]. However, these transistor-based sensors work on the watery environment. In an aqueous environment, the charge of the bioparticle to be detected is partially screened by the different ions presents in the electrolyte solution [19] which is well known as Debye screening. Moreover, the charge of such biomolecules in an aqueous environment is highly dependent upon the pH of the solution which requires a very well controlled pH based solution so that biomolecules contain detectable charges. Hence a new detection technique based on the dielectric permittivity of biomolecules has been found in recent literature which is free from the problems of aqueous solution based bio-detection. Specifically, R. Narang et al. [32-33] has proposed Dielectric Modulated Tunnel FET as a biosensor and studied analytical modeling and sensitivity analysis for label-free detection. Tunnel FET sensor is expected to result in better sensitivity compared to FET-based sensor [33]. Various metrics like I on , SS , ∆Vth etc. have been investigated [33] to measure sensitivity. Being inspired by their works, we have further investigated the prospect of such biomolecule detection technique in a different device structure with an entirely different channel made of 2D materials. In this work, we focused on ∆I , a metric which has been I
found to provide the best sensitivity among all.
ACCEPTED MANUSCRIPT Hence, the ultimate objective of this work is to analyze the feasibility of 2D material FET for both logic and sensor applications. 2D materials are highly sought today for its superior performance as a replacement of Si in current MOSFET technology. With the promise of replacing Si in FET, will these TMD be able to work as sensor material also? This work is focused on finding that answer too.
2. Device Structure and Simulation Approach
M AN U
SC
RI PT
2.1 As MOSFET
TE D
Figure 1. Schematic of the device used in this work as MOSFET. Monolayer TMDCs are MoS2, MoSe2, WS2, and WSe2.
AC C
EP
Quantum transport analysis is necessary for these type of materials (MX2) and dimensions of interest. Self-consistent solution of Poisson and Schrodinger equation in an NEGF framework with tight binding (TB) Hamiltonian is needed to capture the phenomena of ballistic transport in these emerging material devices. The electrical characteristics of Junction-Less TMDC FETs are simulated using two-band tight binding (TB) Hamiltonian with an open-source quantum transport simulation framework [34-35]. Simulation procedure is same as those of [17] [36] using the same context. We have benchmarked the Id-Vg characteristics of MoS2 JLFET with that of [17] (not shown) for 30 nm gate length using the same parameter values given in that paper. However, this benchmarking is not surprising since both ref. [17] and this work have used the same open source platform, NanoTcad ViDES [34-35], which is an established platform to simulate 2D material FET. For Fig. 1, Channel is monolayer transition metal dichalcogenides (MoS2, MoSe2, WS2, and WSe2) with a thickness of ~0.7 nm. Channel length L is varied from 5 to 30 nm. Top and bottom oxide (both HfO2) thicknesses are kept 4.9 nm. The doping used is 5 x 10-3 as a molar fraction, which is defined as the ratio between the numbers of doping atoms over the number of atoms [37]. Parameters for monolayer TMDCs have been taken from various sources [38-42] as listed in Table 1. In this work, top gate and bottom gate are connected. Therefore, VTG and VBG are same and named as gate voltage, VG. Source and drain contact are assumed Ohmic which is possible by proper contact engineering [43].
ACCEPTED MANUSCRIPT Table 1. Monolayer TMDC parameters Material
Electron Effective
Band gap
Mass
Dielectric Constant
0.56
1.89
4.8
WSe2
0.35
1.61
4.5
MoSe2
0.62
1.58
6.9
WS2
0.33
2.05
4.4
2.2 For Biosensor
RI PT
MoS2
TE D
M AN U
SC
For biosensor application tri-layer TMDC material has been considered for the channel. The motivation to study tri-layer TMDC JFET for bio-sensor demands some explanations. We have previously investigated the prospects of multilayer and monolayer TMDC based biosensor in an aqueous environment [25-26]. From those study it has been found that the sensitivity of layered TMDC biosensors is considerably higher [26] than that of the monolayer TMDC biosensor [25] while detecting identical biomolecules under similar aqueous environment. A possible reason for such higher sensitivity can be attributed to the relatively small bandgap of trilayer TMDC material compared to that of monolayer which might have resulted in a larger change in sub-threshold current in response to the capture of a biomolecule. Besides, the multilayer MoS2 structure had shown better noise immunity than a single-layer case in the air [44], an additional benefit of using multi-layer structures for sensor applications. It is very easy to mask the performance of a sensor with noise. Even for gas sensor [45], single-layer (1L) MoS2 device showed an unstable response. In our proposed device, both top and bottom layer are connected to receptors for biomolecules and gate metal respectively. Hence, the middle layer is considered free from external environment suitable for capturing the effect of attaching biomolecules on the other layer.
AC C
EP
Fig. 2 shows the device structure proposed in this work for biomolecule sensing. The dimensions and material parameters of various layers are described in detail in Table 2. To sense biomolecule, a part of the top oxide layer has been etched to form a cavity where proper bio-receptors can be functionalized to provide selective binding of the biomolecule of interest. In the absence of target biomolecules in the nanogap cavity region, the holes are filled with air, so dielectric permittivity of the cavity region is different from that of top oxide and the threshold voltage changes from its initial values. When target biomolecules (like Streptavidin, Biotin, Avidin, enzyme, cell, DNA, APTES) are present and immobilized at the binding site, the dielectric permittivity changes and the gate capacitance of the device also changes. Consequently, electrical characteristics such as threshold voltage and current of the device changes per the dielectric constant or charge of the target biomolecules. To investigate the effect of biomolecule on device’s electrical response, quantum transport analysis is done. Same self-consistent simulation approach in an NEGF framework with tight binding (TB) Hamiltonian has been used to capture the phenomena of ballistic transport in these emerging material devices using open source platform, NanoTcad ViDES. Material parameters have been taken from literature [46].
RI PT
ACCEPTED MANUSCRIPT
SC
Fig.2. The DMFET structure under consideration. It has a trilayer TMDC material channel sandwiched between the top and bottom oxides and corresponding upper and lower gates. The channel, the source, and drain are highly n-doped
M AN U
regions of the same 2D material. Top oxide is etched on one side to simulate the effect of the biomolecule on the conductivity of the FET.
Table 2. Parameters used in this work for trilayer TMDC DMFET Parameter
TE D
Channel Thickness [18]
Value 2 nm 20 nm
Top Oxide Length (L2)
20 nm
Channel Length
40 nm
Channel, Source and Drain Doping
2.2 x 1016 m-2
EP
Cavity Length (L1)
Top Oxide Thickness (ZrO2)
9 nm
AC C
(Biomolecule height)
Bottom Oxide Thickness (SiO2)
10 nm
Trilayer TMDC Electron Effective Mass [46]
0.52 x 9.1 x 10-31 kg
Trilayer TMDC Dielectric Permittivity
5.2 x 8.854 x 10-12 Fm-1
ZrO2 Dielectric Permittivity
12.5 x 8.854 x 10-12 Fm-1
SiO2 Dielectric Permittivity
3.9 x 8.854 x 10-12 Fm-1
Biomolecule dielectric constant
(3 to 9) x 8.854 x 10-12 Fm-1
ACCEPTED MANUSCRIPT To predict the sensitivity of trilayer TMDC FET, we have used the NEGF current simulator [34-35]. For measuring sensitivity, we have used the following formulae [47] S=
∆I d
=
Id
I d ( bio ) − I d ( air )
min ( I d ( bio ) , I d ( air ) )
3.1 Ballistic Simulation of Monolayer TMDC MOSFET
RI PT
3. Result and Discussion
d
10
10
-5
-0.2 0 Gate Voltage, Vg(V)
WSe 2
Vd=0.05V Vd=0.5V
-10
-0.4
-0.2 0 Gate Voltage, Vg(V)
SC
-5
10
MoSe 2
-0.4
0.2
10
Vd=0.05V Vd=0.5V
-10
0.2
TE D
-0.4
Drain Current, I (A/µ m)
d
-10
10
M AN U
Vd=0.05V Vd=0.5V 10
Drain Current, I (A/µ m)
MoS2
d
-5
Drain Current, I (A/µ m)
10
EP
d
Drain Current, I (A/µ m)
3.1.1 Drain Current (Id) - Gate Voltage (Vg) Characteristics
-5
-0.2 0 Gate Voltage, Vg(V)
0.2
WS2
Vd=0.05V Vd=0.5V 10
-10
-0.4
-0.2 0 Gate Voltage, Vg(V)
0.2
AC C
Figure 3. Id-Vg characteristics of various monolayer TMDC Junction less FETs for two different drain voltages. In this case, the channel length is 30 nm for all the devices.
Fig. 3 presents the drain current-gate voltage characteristics of four monolayer FETs considered for two different drain voltages. All the devices show high “ON Current” while providing very small subthreshold current. The physics of interband tunneling has been exploited in MOSFETs to design low power switching devices. Although MOSFET is a barrier-controlled device, interband tunneling could still affect device performance in the subthreshold regime when carrier effective mass and band gap are lowered in the channel [48] which is beyond the scope of the present work.
ACCEPTED MANUSCRIPT
3.1.2 Short Channel Effects Id-Vg curve is critical in evaluating short channel characteristics i.e. Drain induced barrier lowering (DIBL), Subthreshold Swing (SS), Threshold voltage roll-off (VTRO), etc.
RI PT
3.1.2.1 Drain Induced Barrier Lowering (DIBL) DIBL is the reduction of threshold voltage with the increase of the drain voltage. From Fig. 3, we extract the values of DIBL for all materials in Table 3. DIBL is the lowest for a FET when top of the barrier from source to drain is least affected by drain voltage. Semiconductor FETs is usually defined
t s tb ε s
εb
, where ts is the
SC
by a particular term called ‘natural scaling length’ given by λ =
M AN U
semiconductor thickness, tb is the gate dielectric thickness, εs and εb are the semiconductor and gate insulator dielectric constants respectively. Small natural length predicts less short channel effects like DIBL. Because of the atomic layer thickness of these materials which results in small natural length, these FETs show strong immunity from the effect of drain to source voltages compared to bulk MOSFET. The threshold voltage is measured using the constant current method. Here reference current used is 1*10-7 A/µm. Table 3. Extracted DIBL values
TE D
Material
30 nm Gate Length DIBL (mV/V) 18
MoSe2
10.88
WS2
11.55
WSe2
14.88
AC C
EP
MoS2
3.1.2.2 Subthreshold Swing (SS) The subthreshold swing is defined as the gate voltage required to change the drain current by one order of magnitude when the gate voltage is below the threshold voltage. In the MOSFET, the subthreshold swing is limited to 60 mV/Dec at room temperature. Since one carrier takes part in conduction in FET (except Tunnel FET where both hole and electrons involve in conduction, and SS can be below 60 mV/Dec for TFET), it is not possible to breach this barrier. However, JLFET is better suited in this context. In our calculations, for 15 and 30 nm gate lengths, we get SS close to ideal ~60 mV/Dec. However, as we shorten the gate length, we get >60 mV/Dec SS for 5 nm gate length as presented in Table 4. SS is measured from Fig 4 for 5 nm gate length. All devices show SS in the region of 76~78 mV/Dec for 5nm gate length.
ACCEPTED MANUSCRIPT
Table 4. Extracted SS values 5 nm Gate Length
Material
SS (mV/Dec) 77.1
MoSe2
77.7
RI PT
MoS2 WS2
76.3
WSe2
77.6
M AN U
SC
The range of SS values reported in this work for various materials is very small indicating all these materials are viable options as channel material in a MOSFET. It has been reported in the recent literature [49-50] that for various FETs with 2D material as channel with an effective mass of meff >=0.3 m0, it becomes difficult to differentiate Id-Vg curves in the subthreshold region. On the other hand, for meff ≥ 0.2×m0 [50] excellent subthreshold behavior and a good suppression of source-drain tunneling and short channel effects are obtained for a gate length of 5 nm. Our simulation result of comparable SS for various materials where effective mass is bigger than 0.3 m0 for all materials proves that the trend is similar for Junction less FET too.
3.1.2.3 Threshold Voltage Roll Off (VTRO)
TE D
In this work threshold voltage roll-off is defined as VTRO = Threshold voltage (VT) at LG = LREF Threshold voltage (VT) at LG = LNEW. Here new gate lengths considered are 15 nm and 5 nm while the reference length is 30 nm. Table 5 shows the corresponding values measured from Fig 4.
EP
Table 5. Extracted VTRO (mV) values 15 nm Gate Length
5 nm Gate Length
MoS2
42.4
308
MoSe2
42.6
306
WS2
43.95
310
WSe2
43.2
307.5
AC C
Material
WSe 2
10
-5
-0.6
Lg=5nm Lg=15nm Lg=30nm -0.4 -0.2 0 0.2 Gate Voltage, Vg(V)
10
-5
-0.6
WS2
10
-5
Lg=5nm Lg=15nm Lg=30nm -0.4 -0.2 0 0.2 Gate Voltage, Vg(V)
RI PT
Drain Current, Id (A/µ m)
-0.6
Lg=5nm Lg=15nm Lg=30nm -0.4 -0.2 0 0.2 Gate Voltage, Vg(V)
MoSe 2
SC
-5
Drain Current, Id (A/µ m)
10
Drain Current, Id (A/µ m)
MoS2
Lg=5nm Lg=15nm Lg=30nm -0.4 -0.2 0 0.2 Gate Voltage, Vg(V)
M AN U
Drain Current, Id (A/µ m)
ACCEPTED MANUSCRIPT
-0.6
Figure 4. Id-Vg characteristics of various monolayer TMDC Junctionless FETs for multiple gate lengths.
AC C
EP
TE D
From the short channel characteristics presented in this work, it can be concluded that transition metal dichalcogenides based FETs have much steeper sub-threshold slopes compared to Si devices, essentially because of their much smaller thicknesses leading to stronger gate controllability and higher immunity to short-channel effects. These results indicate that TMDC monolayer junction-less FET holds the promise of the shorter transistor with less non-ideal effects. Reported DIBL from 10 mV/V to 18 mV/V, SS in the region of 76~77 mV/Dec are encouraging values for next generation MOSFETs. Considering the range over which SS, DIBL, and VTRO of different monolayer TMDC materials used in this work vary, we can consider all of them to be a potential candidate to cope up with well-known short channel effects.
3.2 Application in Biosensor Fig. 5 shows how the channel barrier from source to drain is modulated by the dielectric permittivity of the biomolecules in the cavity. As the biomolecule’s dielectric permittivity increases, the channel potential barrier also increases since the device under study is an accumulation mode JLFET. Thus, current decreases with increase in biomolecule dielectric constant. The bio sensitivity of the proposed device can be monitored from device current as shown in Fig. 6. A detectable change in device current is found for variation in biomolecule dielectric permittivity specifically in the subthreshold and linear region. The change in device current can be seen more clearly in the zoomed portion of Fig. 6 for different biomolecule permittivity. Fig. 7 shows the sensitivity of the proposed dielectric modulated TMDC JLFET. We found that the highest sensitivity is obtained when the device is in subthreshold region whereas the sensitivity decreases as the device gradually moves into
ACCEPTED MANUSCRIPT the linear regime. Hence, to achieve maximum sensitivity, the proposed devices should be operated in the subthreshold regime which is also consistent with other biosensors [51].
= ε0
ε bio
= 3ε 0
ε bio
= 5ε 0
ε bio
= 7ε 0
RI PT
0
ε bio
VGS=-0.3V
-0.05
-0.15 0
1
SC
-0.1
2 3 Channel Position (nm)
4
M AN U
Channel Potential Barrier (eV)
0.05
x 10
-8
Fig. 5. Channel potential barrier of dielectric modulated trilayer TMDC FET biosensor obtained from NEGF current
ε =ε bio 0 ε = 3ε bio 0 -6
ε = 5ε bio 0 ε = 9ε bio 0
10
10
10
10
-8
-10
-12
-8
EP
ε = 7ε bio 0
AC C
Drain Current, Id (A/µ m)
10
-0.5
-0.4
-0.3
=ε
bio
= 3ε
0
ε
bio
= 5ε
0
ε
bio
= 7ε
0
ε
bio
= 9ε
0
0
-12
10
-13
10 -0.65
10
-0.6 -0.55 Gate Voltage, Vg(V)
-8
10
-12
-14
10 -0.2
bio
ε
-10
10
-14
-0.6
10
ε
Drain Current, Id (A/µ m )
-6
10
Drain Current, Id (A/µ m )
-4
Drain Current, Id (A/µ m )
10
TE D
simulator [34-35] for the different dielectric permittivity of the biomolecule immobilized in the cavity.
-9
10
-0.6 -0.5 -0.4 -0.3 Gate Voltage, Vg(V)
-10
10
-0.5
-0.45 -0.4 Gate Voltage, Vg(V)
Gate Voltage, Vg(V)
Fig. 6. Id- Vg characteristics of dielectric modulated trilayer TMDC FET. The left figure shows Id-Vg characteristics for a broad range of gate voltages while data on the right are zoomed in for low gate voltages to demonstrate the difference in currents in the subthreshold region for biomolecules with different dielectric permittivity.
ACCEPTED MANUSCRIPT 80
ε
70
= 3ε
0
ε
bio
= 5ε
0
ε
bio
= 7ε
0
bio
= 9ε
0
ε
60 50 40
RI PT
Sensitivity,S
bio
30 20 10 -0.6
-0.55 -0.5 -0.45 Gate Voltage, VG (V)
-0.4
-0.35
-0.3
SC
-0.65
M AN U
Fig. 7. Sensitivity in the subthreshold region for various biomolecule dielectric permittivity. Sensitivity decreases with increase in gate voltage. A higher dielectric constant of biomolecule results in higher sensitivity.
4. Conclusion
AC C
EP
TE D
In this work, we have performed a ballistic simulation study on the performance of some prominent monolayer TMDC as channel materials in MOSFETs from 30 nm to sub-10 nm region operation. For gate length below 10 nm, SS values are close to the ideal limit for all the FETs incorporating these materials. Another important short channel effect, threshold voltage roll off is also measured. The extracted values will be helpful to designers in choosing the appropriate material for various applications. Moreover, we have also performed a quantum ballistic simulation study on the performance of Van Der Waals trilayer heterostructure as a channel material in biosensing operation using the principle of Dielectric Modulation. Our research shows that these materials could be viable options in implementing FET-based biosensor in a dry environment because of their high sensitivities (up to 80 percent), especially in the subthreshold regime. Moreover, the sensitive nature of the proposed biosensor can be maximized through operation in the subthreshold region which follows the trend of biosensing found in recent literature. We have chosen a trilayer hetero structure as a channel material for DMFET analysis. It would be interesting following work to do a comparative analysis of dielectric modulated biosensor in a TFET structure [33] with varying no. of layers. For more detailed analysis on the sensitivity of biosensor, the effect of biomolecule position and fill in factor in the cavity region on the sensitivity can be studied [52].
Conflict of interest The authors declare that there is no conflict of interest regarding the publication of this paper.
ACCEPTED MANUSCRIPT
References S. M. Sze and K. K. Ng, “Physics of semiconductor devices,” Wiley, pp. 328-338, 2006. R. S. Muller, T. I. Kamins and M. Chan, “Device Electronics for Integrated Circuits,” John Wiley and Sons, Inc., pp. 420-439, 2003. [3] International Technology Roadmap for Semiconductors, 2012: http://www.itrs.net. [4] M. V. Haartman and M. Ostling, “Low-frequency noise in advanced MOS devices,” Springer, 2007. [5] C. W. Lee, A. Afzalian, N. D. Akhavan, R. Yan, I. Ferain, and J. P. Colinge, “Junctionless multigate field-effect transistor,” Applied Physics Letters, vol. 94, no. 5, pp. 97–103, 2009. [6] A. M. Ionescu, “Electronic devices: nanowire transistors made easy,” Nature Nanotechnology, vol. 5, no. 3, pp. 178-179, 2010. [7] J. P. Colinge, F. Balestra, J. P. Raskin, F. Gamiz and V. S. Lysenko, “Semiconductor-On-Insulator Materials for Nanoelectronics Applications,” Springer, pp. 187-200, 2011. [8] J. P. Colinge, C. W. Lee, A. Afzalian, et al., “Nanowire transistors without junctions,” Nature Nanotechnology, vol. 5, no. 3, pp. 225-229, 2010. [9] C. W. Lee, I. Ferain, A. Afzalian, et al., “Performance estimation of junctionless multigate transistors,” Solid-State Electronics, vol. 54, pp. 97–103, 2010. [10] J. P. Colinge, A. Kranti, R. Yan, et al., “Junctionless Nanowire Transistor (JNT): Properties and design guidelines,” Solid-State Electronics, vol. 65-66, pp. 33–37, 2011. [11] R. Doria, M. Pavanello, R. Trevisoli et al., “Junctionless Multiple-Gate Transistors for Analog Applications,” IEEE Transactions on Electron Devices, vol. 58, no. 8, pp. 2511–2519, 2011. [12] J. P. Colinge, C. W. Lee, I. Ferain et al., “Reduced electric field in junctionless transistors,” Applied Physics Letters, vol. 96, no. 7, p. 073510, 2010. [13] W. Cao, J. Kang, S. Bertolazzi, A. Kis, and K. Banerjee, “Can 2D nanocrystals extend the lifetime of floating-gate transistor based nonvolatile memory?” IEEE Trans. Electron Devices, vol. 61, no. 10, pp. 3456–3464, Oct. 2014. [14] W. Cao, J. Kang, W. Liu Y. Khatami, D. Sarkar, and K. Banerjee, “2D Electronics: Graphene and Beyond,” in Proc. ESSDERC, Sep. 2013, pp. 37–44. [15] W. Cao, J. Kang, W. Liu, and K. Banerjee, “A compact current voltage model for 2D semiconductor based field-effect transistors considering interface traps, mobility degradation, and inefficient doping effect,” IEEE Trans. Electron Devices, vol. 61, no. 12, pp. 4282–4290, Dec. 2014. [16] Wei Cao et al., “Junction-Less Monolayer MoS2 FETs”, arXiv:1509.00561. [17] Tarun Agarwal et al., “Comparison of short-channel effects in monolayer MoS2 based junctionless and inversion-mode field-effect transistors”, Applied Physics Letters, 108, 023506 (2016); DOI: 10.1063/1.4939933. [18] F. Puppo, M.-A. Doucey, T. S. Y. Moh, G. Pandraud, P. M. Sarro, G. De Micheli, and S. Carrara, “Femto-molar sensitive field effect transistor biosensors based on silicon nanowires and antibodies,” IEEE Sensors, pp. 1–4, 2013. [19] T. J. Park, J. Kim, J. Ahn, S. Choi, M. Im, S. Kim, J. P. Duarte, and C. Kim, “An Underlap Channel-Embedded Field-Effect Transistor for Biosensor Application in Watery and Dry Environment,” IEEE Trans. Nanotechnology, vol. 11, pp. 390–394, 2012.
AC C
EP
TE D
M AN U
SC
RI PT
[1] [2]
ACCEPTED MANUSCRIPT
AC C
EP
TE D
M AN U
SC
RI PT
[20] J. A. G. Sheneve Z. Butler, Shawna M. Hollen, Linyou Cao, and Yi Cui, “Progress, Challenges, and Opportunities in Two-Dimensional Materials Beyond Graphene,” ACS Nano, vol. 7, no. 4, pp. 2898–2926, 2013. [21] C. N. R. Rao, K. Gopalakrishnan, and U. Maitra, "A Comparative Study of Potential Applications of Graphene, MoS2, and Other Two-Dimensional Materials in Energy Devices, Sensors, and Related Areas", ACS Applied Materials and Interfaces, vol. 7, no. 15. 2015. [22] H. Terrones, F. López-Urías, and M. Terrones, “Novel hetero-layered materials with tunable direct band gaps by sandwiching different metal disulfides and diselenides,” Sci. Rep., vol. 3, p. 1549, 2013. [23] N. Lu, H. Guo, L. Li, J. Dai, L. Wang, W.-N. Mei, X. Wu, and X. C. Zeng, “MoS2/MX2 hetero layers: bandgap engineering via tensile strain or external electrical field.,” Nanoscale, vol. 6, no. 5, pp. 2879–86, 2014. [24] N. Lu, H. Guo, L. Wang, X. Wu, and X. C. Zeng, “Van der Waals trilayers and superlattices: modification of electronic structures of MoS2 by intercalation,” Nanoscale, pp. 4566–4571, 2014. [25] Abir Shadman, Ehsanur Rahman, and Quazi D. M. Khosru, 2016, “Monolayer MoS2 and WSe2 Double Gate Field Effect Transistor as Super Nernst pH sensor and Nanobiosensor,” Sensing and Bio-Sensing Research, vol. 11, no. 1, pp. 45–51, 2016. [26] Kanak Datta, Abir Shadman, Ehsanur Rahman, and Quazi D. M. Khosru, 2017, “Trilayer TMDC Heterostructures for MOSFETs and Nanobiosensors,” Journal of Electronic Materials, vol. 46, no. 2, pp. 1248–1260, 2017. [27] Anran Gao, Na Lu, Yuelin Wang, and Tie Li., ‘Robust ultrasensitive tunneling FET biosensor for point-of-care diagnostics,' Nature Scientific Reports 6, Article number: 22554 (2016). DOI: 10.1038/srep22554 [28] F. Puppo et al., ‘SiNW-FET in-Air Biosensors for High Sensitive and Specific Detection in Breast Tumor Extract,' IEEE Sensors Journal, Vol. 16, No. 10, pp 3374 – 3381, May 15, 2016. [29] Wang, C. et al. ‘A label-free and portable graphene FET aptasensor for children blood lead detection,' Nature Scientific Reports 6, 21711; DOI: 10.1038/srep21711. [30] B, Cai. et al., ‘Gold nanoparticles-decorated graphene field-effect transistor biosensor for femtomolar MicroRNA detection,' Biosensors and Bioelectronics, Volume 74, pp 329–334, 15 December 2015. [31] C-H Yeh et al., ‘High-performance and high-sensitivity applications of graphene transistors with self-assembled monolayers,' Biosensors and Bioelectronics, Volume 77, pp 1008–1015, 15 March 2016. [32] Rakhi Narang, Manoj Saxena, R. S. Gupta, and Mridula Gupta, “Dielectric Modulated Tunnel Field Effect Transistor – A Biomolecule Sensor,” IEEE Electron Device Letters, Vol. 33, no. 2, pp. 266-268, February 2012. [33] R. Narang, K. V. S. Reddy, M. Saxena, R. S. Gupta and M. Gupta, "A Dielectric-Modulated Tunnel-FET-Based Biosensor for Label-Free Detection: Analytical Modeling Study and Sensitivity Analysis," in IEEE Transactions on Electron Devices, vol. 59, no. 10, pp. 2809-2817, Oct. 2012. [34] “NanoTCAD ViDES.’ [Online]. Available: http://vides.nanotcad.com/. [Accessed: 01Aug-2013]”. [35] “NanoTCAD ViDES.’ [Online]. Available: http://vides.nanotcad.com/vides/. [Accessed: 01-Sep-2015]”.
ACCEPTED MANUSCRIPT
AC C
EP
TE D
M AN U
SC
RI PT
[36] Tarun Agarwal et al., “Benchmarking of MoS2 FETs With Multigate Si-FET Options for 5 nm and Beyond,” IEEE Trans. Electron Devices, vol. 62, no. 12, pp. 4051–4057, Dec. 2015. [37] User Manual, NanoTCAD ViDES, 2008 (http://vides. nanotcad. com/vides/documentation/commands-5 /dope reservoir) [38] Ram Krishna Ghosh and Santanu Mahapatra, “Monolayer Transition Metal Dichalcogenide Channel-Based Tunnel Transistor,” IEEE Journal of the Electron Device Society, Vol. 1, No. 10, pp. 175-181, October 2013. [39] Cheng Gong et al., “Band alignment of two-dimensional transition metal dichalcogenides: application in tunnel field effect transistors,” DOI: 10.1039/c5nr06782k. [40] S. U. Z. Khan and Quazi D. M. Khosru, “Quantum Mechanical Electrostatics and Transport Simulation and Performance Evaluation of Short Channel Monolayer WSe2 Field Effect Transistor”, ECS Transaction, Vol. 66 (14), pp. 11-18, 2015. [41] W. Liu, W. Cao, J. Kang, and K. Banerjee, “High-Performance Field-Effect-Transistors on Monolayer WSe2”, Vol. 58 (7), pp. 281-285, 2013. [42] Jiwon Chang, Leonard F. Register and Sanjay K. Banerjee, “Comparison of Ballistic Transport Characteristics of Monolayer Transition Metal Dichalcogenides (TMDs) MX2 (M = Mo, W; X = S, Se, Te) n-MOSFETs”, Proc. of the IEEE SISPAD, pp. 408-411, 2013. [43] W. Liu, J. Kang, W. Cao, D. Sarkar, Y. Khatami, D. Jena, and K. Banerjee, “High-performance few-layer-MoS2 field-effect-transistor with record low contact-resistance,” in IEEE Int. Elect. Dev. Meeting, pp. 499−502, Dec. 2013. [44] H-J Kwon et al., ‘Analysis of flicker noise in two-dimensional multilayer MoS2 transistors’ Applied Physics Letters 104, 083110 (2014). [45] Ha Li et al., ‘Fabrication of Single- and Multilayer MoS2 Film Based Field-Effect Transistors for Sensing NO at Room Temperature’, Small, vol. 8, No. 1, pp 63–67, 2012. [46] K. Datta and Q. D. M. Khosru, “Electronic Properties of MoS2 / MX2 /MoS2 Trilayer Heterostructures: A First Principle Study,” JSS vol. 5, no. 11, pp. 3001–3007, 2016. [47] D. Sarkar, W. Liu, X. Xie, A. C. Anselmo, S. Mitragotri, and K. Banerjee, “MoS2 Field-Effect Transistor for Next- Generation Label-Free Biosensors,” ACS Nano, 2014. [48] N. Ma and D. Jena, “Interband Tunneling in two-dimensional crystal semiconductors,” Appl. Phys. Lett., vol. 102, no. 13, 2013. [49] Wei Cao et al., 2D Semiconductor FETs—Projections and Design for Sub-10 nm VLSI, IEEE Transaction on Electron Devices, pp. 3459-3469, Vol. 62, No. 11, Nov 2015. [50] S. Thiele et al., The Prospects of Two-Dimensional Materials for Ultimately Scaled CMOS. Available Online: http://eurosoi-ulis2017.inn.demokritos.gr/files/abstracts/abstracts_Session9.pdf [51] J. Lee, P. Dak, Y. Lee, H. Park, W. Choi, M. A. Alam, and S. Kim, “Two-dimensional Layered MoS2 Biosensors Enable Highly Sensitive Detection of Biomolecules,” Sci. Rep., pp. 1–7, 2014. [52] Ehsanur Rahman, Abir Shadman, and Quazi D. M. Khosru, 2017, “Effect of biomolecule position and fill in factor on sensitivity of a Dielectric Modulated Double Gate Junctionless MOSFET biosensor,” Sensing and Bio-Sensing Research, vol. 13, pp. 49–54, 2017.
ACCEPTED MANUSCRIPT
AC C
EP
TE D
M AN U
SC
RI PT
Junctionless Field Effect Transistors (JLFETs) is thought to overcome fabrication challenges related to abrupt p-n junctions due to channel length scaling in state-of-the-art CMOS technology which might open a new era in the technology of nanoscale MOSFETs. Because of highly scaled thickness up to an atomic plane and dangling bond free pristine surface, 2-D semiconducting transition metal dichalcogenides (TMDC), such as MoS2 and WSe2, have been considered as prospective channel materials for low power CMOS devices. Due to these advantages, in this work, we have performed a simulation based study of Transition Metal Dichalcogenides Based Double Gate Junctionless Field Effect Transistor to evaluate their performance in short channel regime scaled down to 5nm.We have also investigated the prospect of these TMDC JLFETs in Debye screening free potentiometric biosensing using the technique of dielectric permittivity modulation of the cavity region formed in the oxide for biosensing. Our study shows considerable biomolecule detection capability at dry environment even without any charge with optimal sensor operation in the subthreshold region.