Abstract

This work demonstrates a mixed-dimensional piezoelectric-gated transistor in the microscale that could be used as a millinewton force sensor. The force-sensing transistor consists of 1D piezoelectric zinc oxide (ZnO) nanorods (NRs) as the gate control and multilayer tungsten diselenide (WSe2) as the transistor channel. The applied mechanical force on piezoelectric NRs can induce a drain–source current change (ΔIds) on the WSe2 channel. The different doping types of the WSe2 channel have been found to lead to different directions of ΔIds. The pressure from the calibration weight of 5 g has been observed to result in an ∼30% Ids change for ZnO NRs on the p-type doped WSe2 device and an ∼−10% Ids change for the device with an n-type doped WSe2. The outcome of this work would be useful for applications in future human–machine interfaces and smart biomedical tools.
Keywords: piezo-gated field-effect transistors, force sensors, low dimensional materials, ZnO nanorods, 2D materials, piezoelectric effect
1. Introduction
With the rapid development of the Internet of Things (IoT) and complex human–machine interfaces, much attention has been paid to the miniaturization of the force-sensing units in the past few decades.1−4 Apart from providing the possibility for a higher integration level and a higher spatial resolution,1,5,6 because of a smaller effective loading area, the smaller force-sensing unit can be used to detect the smaller tactile force in the millinewton or micronewton range. The small force detection could be helpful in providing feedback during the manipulation of micro-objects.7 One example of the small force sensor application is the assistant tool for precise surgery, such as ophthalmological surgery, when the tactile force is critical for the safety and outcomes of surgery operations.8,9
In the past few decades, much effort has been made to minimize the capacitive, piezoelectric, and piezoresistive components as the loading units of mechanical force sensors.7,10,11 However, currently, most of the force-sensing units are passive. The passive signals depend only on the functional structure or materials and cannot be controlled without an external system. For an active sensing system, one could control the signal by applying an additional signal to provide smarter and more adaptive sensing functions.5,12,13 In the literature, to achieve the adaptive force sensing, the integration of the force-sensing materials or structures onto a field-effect transistor (FET) has been treated as one of the possible solutions.2,3,5,14 With a higher integration level, the transistor-type force sensor could be more compatible with the current CMOS (Complementary Metal Oxide Semiconductor) system, and then the advanced programming algorithm such as machine learning can be implemented,15 allowing more complex functions and more adaptive systems to be achieved.15−17
The development of one-dimensional (1D) piezoelectric nanomaterials such as zinc oxide (ZnO) and gallium nitride (GaN) nanorods (NRs) provides the possibility to achieve the piezoelectric based force sensor in the microscale.18−20 These piezoelectric nanomaterials can be integrated with a FET as the gate control (see Figure 1): when pressure is loaded on the piezoelectric material, there will be a drain–source current change (ΔIds) in the FET channel. The channel of FET can be made by different materials, such as doped silicon (Si),21−23 graphene,24−26 and two-dimensional (2D) molybdenum disulfide (MoS2).14 Among these channel materials, 2D materials have drawn much interest due to a higher carrier mobility in the “short channel” scale, which could offer more potential for the field effect to amplify the electrical signal when the device geometry is scaled down further.27,28 With the 1D–2D integrated structure, the small pressure from the normal direction can induce a large strain on the vertically grown piezoelectric NRs due to the high aspect ratio in the z-axis, leading to a large polarization. The polarization could be amplified by the 2D semiconductor channel with a back-gate voltage. Overall, the mixed-dimensional transistor could be a potential way to sense the small pressure and amplify the piezoelectric signal by itself. The output signal (drain–source current Ids) is more compatible with the CMOS system to achieve more complex and adaptive functions. In addition, with a self-powered piezoelectric gate-control block,29−31 the sensing system based on the piezo-gated transistor could have an improved power efficiency in the overall sensing system. However, the development on these mixed-dimensional transistors is still in its infancy; the study of the mechanism of these novel structures is still limited, and the integration and the fabrication process of these novel mixed-dimensional transistors still require much effort.
Figure 1.

Schematic showing mixed-dimensional transistor as an active sensing unit. The channel materials can be doped Si (MOSFET, metal-oxide-semiconductor FET),21−23 graphene,24−26 and 2D MoS2.14 Normally, the piezoelectric ZnO NRs could be employed as a gate to achieve force/pressure/strain sensing. In addition to force sensing, mixed-dimensional transistors have been reported to have capabilities in photodetection23,25,26 and gas sensing.32
This paper demonstrates a piezo-gated force-sensing transistor that consists of a vertical one-dimensional (1D) ZnO NR array and a multilayer tungsten diselenide (WSe2) channel (1D–2D transistor, see Figure 1). The ZnO NR array has been selected as the piezo-gate of the transistor, mainly because of the relatively high sensitivity to pressure (piezoelectric coefficient d33 up to 9.5 pC/N33) and the relatively simple synthesis methods.34,35 The reasons to select the WSe2 as the channel materials are that (1) the doping level of WSe2 can be modified,36−38 which gives more possibility for tuning the Ids in the channel , and (2) the breaking strain of WSe2 is relatively high, which could be helpful in future flexible substrates.39 In this work, two example microscale devices have been presented that show the different directions of Ids change and different sensitivities (up to 30% Ids change) under pressure loading by calibration weights (equivalent to the millinewton range force). The piezo-gated mechanism and the factors that influence the force-sensing performance have been discussed. The integration of ZnO NRs on both p-type and n-type doped WSe2 transistors and the successful detection of two different direction changes of Ids under pressure could demonstrate the potential of the 1D–2D transistor for smart force-sensing applications in the future, such as human–machine interfaces and biomedical applications.
2. Results and Discussion
2.1. Piezo-Gated Transistor with a p-Type Doped WSe2 Channel (Dev. 1)
2.1.1. Fabricated Device (Dev. 1) and the Testing Setup
An example device (Dev. 1) with ZnO NRs flower grown on the p-typed doped WSe2 transistor has been fabricated (for details see Materials and Methods and Supporting Information Section S1) and is presented in Figure 2. For the example device, mechanical exfoliated multilayer WSe2 (∼128 layers) has been used as the transistor channel, and the patterning of the channel (2 μm × 4 μm) and the metal contact (100 nm thick titanium (Ti)) have been achieved by photolithography. The WSe2 channel layer has been overetched by XeF2 vapor, which has been reported to induce the defect on the etching edge and could possibly lead to a p-type doping behavior.38 The integration of the vertical ZnO NR array (diameter ∼ 1 μm, length ∼ 6 μm, flower cluster shape array on WSe2 channel, Figure 2b) has been achieved by patterned hydrothermal growth on an e-beam evaporated ZnO seed layer.40 On the basis of our previous characterization by electron backscatter diffraction (EBSD), the vertical ZnO nanorod grown hydrothermally on e-beam evaporated ZnO seed layer possesses a single crystalline with (0001) orientation along the z-axis (see the inset of Figure 2b).40
Figure 2.

(a) Device schematic of the ZnO-WSe2 piezo-gated transistor (inset with the device testing setup). (b) SEM images of exampled ZnO-WSe2 piezo-gated transistor (inset with AFM profile showing the thickness of the WSe2 flake, together with the EBSD pole figure showing the crystal (0001) orientation of ZnO NRs grown on the ZnO seed layer, adapted from ref (40) and used under Creative Commons CC-BY 3.0 license). (c) Sheet conductivity of the WSe2 channel before and after ZnO NRs growth. (d) Schematic showing the input force and the stress applied on the ZnO NR array.
The measurement of the transistor has been performed by the semiconductor analyzer Keithley 4200, and the testing setup can be seen in the inset of Figure 2a. To study the influence of the integration process of ZnO NRs on the electric properties of the WSe2 channel, the sheet conductivity of the transistor channel has been measured before and after the integration of ZnO NRs under different back-gate voltages (Vbgs), and the results are presented in Figure 2c. It can be observed that the deposition of the ZnO seed layer and ZnO NRs has decreased the conductivity of the WSe2 transistor by a factor of ∼102 (see Figure 2c), and the back-gate threshold voltage of the overall transistor has been observed to shift from −23 V to −16 V (see Figure 2c). The decrease of the overall conductivity and the shifting of the threshold voltage could be explained by the additional surface scattering on the top side and the possible oxidation of WSe2 (more holes, equivalent to p-type doping)41 due to the deposition of ZnO seed layer/NRs.
The input force on the devices has been performed by the calibration weights, and a polypropylene (PP) plastic plate (1 cm2, 2 mm thick, ∼0.18 g) has been used for encapsulation and mechanical force coupling (Figure 2a, inset). Calibration weights (1 g, 2 g, 5 g) have been used to apply a mechanical load on the 1D–2D transistor, which is equal to the force of ∼11.6 mN, ∼21.4 mN, and 50.7 mN, respectively (F = m × g, taking g = 9.8 N/kg as the standard gravity, including the PP plastic plate). As shown in Figure 2d, the ZnO NR area is close to the center of gravity of the overall load, and the length of ZnO NRs (∼6 μm) is significantly shorter than the width of the PP plastic plate. Therefore, the force applied on ZnO NRs is assumed to be the same as the overall force from the gravity of the weights and the PP plastic plate. The area of ZnO NRs for Dev .1 is around 15 μm2 (∼10 μm × 1.5 μm), the applied forces by calibration weights (1 g, 2 g, 5 g) are equivalent to the stresses of 0.77, 1.43, and 3.39 GPa on ZnO NRs. However, it is worth noting that the exact strain on ZnO NRs could be significantly lower due to the unstable mechanical coupling and strain sharing by the substrate.
2.1.2. Force-Sensing Characterization of Dev. 1
After the integration of ZnO NRs, the drain–source current (Ids) of the overall transistor can be controlled by the back-gate voltage (Vbgs), the drain–source voltage (Vds), and the applied mechanical load (weights). The transfer and output characteristics of the overall transistor after integration of ZnO NRs can be seen in Figure 3a,b.
Figure 3.

(a) Back-gate transfer characteristics of the ZnO-WSe2 piezo-gated transistor as different weights are loaded, inset with the bar diagram that shows the (ΔIds/Ids)% under a 5 g load with Vbgs ranging from −40 V to −25 V. (b) Output characteristics of the ZnO-WSe2 piezo-gated transistor. (c) Real-time sensing measurement of the transistor with 2 and 5 g of weight loading and Ids normalized (the unloaded Ids has been set as 1). (d) Drain–source current change ((ΔIds/Ids)%)) as a function of applied weight and estimated pressure (with data scatter plots); in addition, the bar diagram shows the (ΔIds/Ids)% under Vbgs between −40 V to −30 V.
Without a mechanical load (0 g), the overall transistor is similar to a long channel depletion PMOS transistor. The overall transistor is “closed” when the Vbgs is between ∼−20 and 0 V. When the back-gate voltage is more negative than ∼−20 V, the Ids increases up to a few nanoamperes (see transfer characteristics in Figure 3a). The back-gate control (Vbgs) is based on the field effect. The negative voltage induces more holes (majority carriers) in multilayer WSe2 to the bottom by an electric field through 300 nm thick SiO2. Holes accumulate in the first few bottom layers of WSe2 or the interface between the bottom layers of WSe2 and the back-gate dielectric (300 nm thick SiO2), which is believed to form the p-type channel. For the output characteristic (Figure 3b), the Ids has been observed to be saturated when the Vds values are larger than 3 V, which is similar to the pinch-off mechanism of the MOSFET. The Ids seems relatively unstable compared to the high performance 2D WSe2 FETs in the literature.42 The reason could be the high contact resistance between Ti and p-type doped WSe2 (Schottky contact)43 and the surface scattering between SiO2 and WSe2.44
After the weight loading (1–5 g), a positive shift of 2–5 V in Vbgs has been observed for the overall transfer curve (Figure 3a). As for the output characteristics (Figure 3b), when the Vds value is over 2 V, it can be observed that the loading weights can result in an apparent increase in the level of Ids. Overall, the effects of applied weights to Dev. 1 contribute to an increasing Ids. The sensitivity of weight sensing has been observed to be different when the different Vbgs values have been applied (see inset of Figure 3a). When the applied Vbgs changes from −25 V to −40 V and the load has been fixed at 5 g, the percentage change of Ids ((ΔIds/Ids)%) has been observed with a decreasing trend as Vbgs becomes more negative (see inset of Figure 3a). It is worth noting that the 1 g load seems to have a larger output than the 2 g load, as can be seen in Figure 3a,b, which could be explained by the hysteresis between deformation and recovery of the piezoelectric NRs (loading sequence 2 g, 5 g, and then 1 g; see Supporting Information Section S2).
When the PP plastic plate (0.18 g, ∼1.7 mN) is loaded on the device, a relatively large current change has been observed (from 6 nA to 7.5 nA, Vbgs = −40 V, Vds = 5 V, see Supporting Information Section S2), which may indicate that the minimum detection limit of the transistor could be potentially less than several millinewtons. For the maximum loading limit, it has been observed that ZnO NRs (fraction strength from 2.9 to 8.1 GPa45) can stand for a 5 g load without breaking. During the heavier weights (over 10 g) loading, the device has been observed to be broken. However, based on the optical image after the breaking of the device (see Supporting Information Section S2), apart from overloading, the shear force between the plastic plate and ZnO NRs during loading could be another reason for the breaking since the ZnO-WSe2 has been observed to be peeled from the substrate.
The real-time force-sensing test has also been performed. The Ids has been observed with ∼30% and ∼20% increase under 5 and 2 g of loading, with a relatively high signal-to-noise ratio (Figure 3c). It is worth noting that the Ids has been found to be sensitive to light intensity, which is believed to be one of the reasons for the shifting of the real-time Ids curve (Figure 3c); more details have been provided in Supporting Information S3. In addition, the relationship between the applied weights and the percentage change of Ids ((ΔIds/Ids)%) has been plotted with scatter data points (Figure 3d). The trend for the Ids change has been observed to have a larger increasing slope and a quasi-linear behavior at low pressure (0–1 g) and then saturatation as the applied pressure increases (2–5 g). The overall relationship has been found to be similar to the piezoelectric output curve of the ZnO NRs-MOS capacitor device and ZnO NRs-2D MoS2 devices.14,46 The explanation for the nonlinear relationship between force and the Ids change could be the possible dominant role of contacting/triboelectric charge under low pressure and the saturation of the stress–strain behavior of ZnO NRs within such a small area.14,46
2.1.3. Force-Sensing Mechanism of Dev. 1
The mechanism of the overall transistor has been discussed with the help of the structure diagram (Figure 4a) and the energy diagram (Figure 4b). The overall working mechanism of Dev. 1 is as follows:
-
(1)
Without pressure, the channel of holes is formed in the bottom side of p-type doped WSe2 by the field-effect from the back-gate voltage (see Figure 4a). It is worth noting that even the top sides of the ZnO NRs are floated, and there could be a formation of the semiconductor junction between n-type doped ZnO NRs (due to the oxygen-related defects47) and p-type doped WSe2 (overetched by XeF238). Once ZnO is in contact with WSe2, the majority carriers from both WSe2 (holes) and ZnO (electrons) could diffuse to each other as a result of the concentration gradient. A depletion region (see Figure 4b) could be formed when the thermodynamical equilibrium has been reached such that the inner electrostatic force is the same as the diffusion force. The existence of the depletion region could be one of the reasons for the degradation of the channel conductivity and the decreasing of Ids after the ZnO NRs have been grown on WSe2 (see Figure 2c). However, the study of the depletion region between two different low-dimensional materials is still under investigation.
-
(2)
When pressure is applied on the ZnO NRs, the dipoles of ZnO NRs could be compressed, which is equivalent to a positive piezoelectric polarization (along the z-axis) between the bottom and the top side of the ZnO NR.48 The top sides of ZnO NRs are connected with PP plastic (dielectric) and band-aligned with the iron (calibration weights, virtual grounded) after contacting. The piezoelectric polarization (positive charge) on the bottom side of ZnO could be equivalent to applying a positive voltage on the n-side of the p–n junction, which prevents the majority carriers from flowing through the junction (see Figure 4b). Then, the positive piezoelectric voltage can drive more electrons in WSe2 (minority carrier) accumulated on the top of WSe2 and repel more holes of the WSe2 (majority) carrier to the bottom of WSe2 (the hole conducting channel). Therefore, the overall current Ids in the WSe2 channel increases. In addition, the increasing Ids can also be possibly due to the minority carrier (holes) injection from the bottom of the ZnO NRs to the WSe2 channel driven by piezoelectric voltage.
-
(3)
The observation that, at 5 g, the ((ΔIds/Ids)%) decreases as Vbgs becomes more negative (see inset of Figure 3a) suggests the higher the hole concentration in the channel to begin with at high negative Vbgs, the weaker the modulation of the channel from the effective gate voltage resulting from the 5 g weight.
Figure 4.

(a) Working mechanism schematic of the piezo-gated transistor. (b) Energy band diagram showing the p–n junction control mechanism (the red curves are the conduction band, Fermi level, and valence band after applying pressure on ZnO).
2.2. Piezo-Gated Transistor with an n-Type Doped Channel and a Ti Intermediate Layer (Dev. 2)
The other device has been designed and presented (Dev. 2, see Figure 5), which includes an intermediate Ti metal layer (10 nm thick) between n-type WSe2 and ZnO. There are a few purposes for fabricating and presenting another force-sensing transistor device. First, the pristine WSe2 shows an ambipolar behavior,49 and the mechanically exfoliated WSe2 with different thicknesses can have different carrier types.50 A device with an n-type doped channel could have a different force-sensing behavior under the force loading compared to Dev. 1. The study of the different doping types of the force-sensing transistor channel may help with the future sensing logic system, similar to the “CMOS” system. Second, using Ti as a metal contact with WSe2, the contact type could be both Schottky and ohmic,51 depending on the doping level of WSe2. During our experiments, the Ti contact mostly shows a Schottky contact with WSe2. One purpose of Dev. 2 is to study the piezo-gated behavior with Schottky barrier modulation (see Figure 5a,c) and fabricate a piezo-gated “MESFET” (metal–semiconductor field-effect transistor). Meanwhile, the Ti adhesion layer has been used frequently to improve the adhesion of the seed layer and control the ZnO NRs morphology.52 Therefore, it is worthwhile to investigate the device with a Ti layer between the ZnO and the WSe2 channel. Furthermore, as a force sensor, the critical sensing block is the ZnO NR array. The area of the ZnO NR array and the sizes of ZnO NRs can change the force distribution. On the basis of the study of Dev. 1, the overall system can work under the force applied by less than 5 g of load. To increase the maximum mechanical force that could be applied, a ZnO NR array with a larger area could be integrated on the devices.
Figure 5.
(a) Device structure and working mechanism schematic of the ZnO-Ti-WSe2 piezo-gated transistor. (b) SEM images of exampled ZnO-Ti-WSe2 piezo-gated transistor (inset with AFM profile showing the thickness of WSe2 flake, and the 45° tilted high magnification SEM images of ZnO NRs). (c) Energy diagram showing the piezo-gated Schottky barrier mechanism (red line is the band diagram after applying strain on ZnO NRs). (d) Transfer characteristics and (e) output characteristics of the ZnO-WSe2 piezo-gated transistor at different weight loadings. (f) Drain–source current change ((ΔIds/Ids)%) as a function of applied weights; the scatter points are from both transfer and output characteristics.
For the Dev. 2, a multilayer (∼55 layers) WSe2 transistor with n-type doped behavior has been selected before the growth the ZnO NRs. Then 10 nm Ti and 50 nm ZnO has been deposited as the seed layer. After hydrothermal growth, ZnO NRs of Dev. 2 (see SEM image in Figure 5b) have been observed to have a relatively small size (∼50 nm diameter, ∼700 nm length) and a high density with the same growth parameters as Dev. 1. The reason could be the smaller grains of the seed layer due to the existence of the Ti layer,52 together with the interactions among the less adjacent crystals, the weaker Gibbs–Thomson effect, and the more competitive growth53 due to a larger growth area (4 μm × 20 μm) than that of Dev. 1. With a smaller size of ZnO NRs and a larger growth area, the mechanical pressure on ZnO NRs could be smaller than that on Dev. 1. Therefore, larger weights can be loaded on Dev. 2 without breaking the ZnO NRs (up to 50 g or 0.5 N during the experiment).
For the electrical properties of the WSe2 channel in Dev. 2, the transfer characteristics of the overall Dev. 2 can be seen in Figure 5d. An n-type doped behavior has been observed when the back gate voltage is between −20 to 20 V. It is noticed that a large off-current and a low on–off ratio have been observed, and the Ids has been found to have an almost linear relationship when Vbgs is between 0 to 10 V. The reason for the large off-current could be that the Ti metal layer on the top-side short circuits part of the WSe2 channel. From the output characteristics (Figure 5e), an apparent Schottky contact behavior has been observed where Ids starts to increase until Vds is larger than 2 V. Therefore, it can be confirmed also that Ti forms a Schottky contact with n-type doped WSe2 in Dev. 2.
For the force-sensing performance of Dev. 2, the strain applied by calibration weights has been observed to induce a positive shift of back-gated transfer characteristics to the n-side by 3–5 V (Figure 5d) and a decreasing of Ids by up to ∼15% (Figure 5e). Figure 5f shows the percentage of Ids change ((ΔIds/Ids)%) under the different loading weights. The measurements clearly shows two phases: from 0 to 5 g, the Ids changes linearly with the load increasing, which agrees with the linear relationship between the piezoelectric voltage and the applied strain; when the loaded weight is over 10 g, the Ids change becomes saturated.
Although the main direction of the Ids change in Dev. 2 (negative) is different from that in Dev. 1 (positive), the roles of piezo-gated ZnO NRs are similar. For both devices, the pressure induces a positive piezoelectric voltage on the bottom side of ZnO (top side of WSe2 or Ti) and then results in a positive shift on back-gate transfer characteristics. However, a significant difference is the doping type of the WSe2, which is believed to be the reason for the different direction of the Ids change. The WSe2 of Dev. 2 is pristine n-typed doped, while the WSe2 in Dev 1 is observed to be p-typed doped possibly because of XeF2 overetching. In addition, the heterojunction control mechanism of Dev. 2 is believed to be based on the Schottky barrier modulation of Ti-WSe2, which is different from the reversed biased p–n junction control of Dev. 1. As shown in Figure 5a,c, there will be a depletion region on the WSe2 side of the Ti-WSe2 Schottky junction due to the different work functions. The electron could not flow freely from WSe2 to Ti without a positive voltage. When pressure is applied on ZnO NRs, the positive piezoelectric potential on the bottom side of ZnO could induce the electron flowing from Ti to ZnO (ohmic-contacted), which increases the electric potential of the Ti layer. Simultaneously, the increasing potential on the Ti layer could lower the Schottky barrier between Ti and WSe2, leading to the electron injection from WSe2 to Ti. The majority of the carriers (electrons) flow away from the conducting channel of the FET, resulting in the decreasing of Ids.
2.3. Discussion of Mixed-Dimensional Transistors as Force Sensors
Table 1 lists the different cases of field-effect transistors that have been gated by vertical ZnO NRs from both our experiments and others’ research.14,22,24,46 It is worth noting that, in addition to the presented devices, we have also fabricated one device which has n-type ZnO NRs directly on n-type WSe2 (Supporting Information Section S4).
Table 1. ZnO NR Piezo-gated Field-Effect Transistor.
| device | bottom of ZnO NRs | ZnO NR length, area | piezo-gated interface | top of ZnO NRs | detection range | sensitivity |
|---|---|---|---|---|---|---|
| ref (46) | MOS capacitor | ∼8 μm, 1 mm ×1 mm | ZnO-Ag/Ti (Ohmic) | Kapton tape, metal probe | 1–32 N, 0.1–3.2 MPa | +80 mV/8 N |
| ref (22) | N-MOSFET (p-Si) | ∼800 nm, large area | ZnO-Al-SiO2-Si | Ceramic probe | Pressure not measured | +10% Ids change |
| ref (24) | graphene | ∼2 μm, large area | ZnO-graphene (Schottky) | Polyurethane (PU), metal probe | 40–160 kPa | Dirac point shifts: 1.62 mV/kPa |
| ref (14) | 2D MoS2 (N-type) | ∼1 μm, 5 μm × 30 μm | ZnO-MoS2 (N–N) | Epoxy, metal probe | 0.75–6.25 MPa | –25% Ids change/6.25 MPa |
| our work (main content) | multilayer WSe2 (P-type) | ∼6 μm, 2 μm × 4 μm | ZnO-WSe2 (N–P) | PP plastic plate, calibration weights | 1–5 g, 10–50 mN, 0.1–0.5 kPaa | +30% Ids change/5 g |
| multilayer WSe2 (N-type) | ∼700 nm, 4 μm × 20 μm | ZnO-Ti-WSe2 (Ohmic-Schottky) | 1–50 g, 10–500 mN, 0.1–5 kPaa | –11% Ids change/5 g | ||
| our work in Section S4 | multilayer WSe2 (N-type) | ∼2 μm, 9 μm × 16 μm | ZnO-WSe2 (N–N) | Parylene C, metal probe | 1–10 N, 0.1 MPa–1 MPab | –25% Ids change/10 N |
The pressure of weights on Dev. 1 and Dev. 2 have been estimated by gravity force divided by the area of the PP square (1 cm2). However, the exact stress on the ZnO NRs could be in the scale of GPa, due to the small loading area of the tips of ZnO NRs.
The pressure on the Parylene C coated device in Section S4 has been applied by a metal probe (2 mm × 5 mm).
As a result of the different device designs, encapsulations, and mechanical coupling, the output of the force-sensing transistor can be different. The force/pressure/strain sensing sensitivity of the integrated transistor could depend on the transfer characteristics of the semiconducting channel and the piezoelectric characteristics of the ZnO NRs. Generally, mechanical pressure on ZnO NRs leads to a positive potential on the top side of the interface between ZnO NRs and WSe2, which is equivalent to an additional negative voltage on the back gate. The direction of the Ids change could be dependent on the channel material doping type and the charge transfer between the ZnO NRs and the transistor channel. In our case with pressure on ZnO NRs, an increasing Ids has been observed in the channel p-type doped WSe2, while a decreasing Ids has been observed in the n-type doped WSe2 channel.
The piezoelectric output could relate to the ZnO NRs themselves and the mechanical coupling. On the one hand, the ZnO NRs’ geometries and growth areas could possibly have an impact on the piezoelectric output.46,54 As shown in Table 1, apart from the different directions of the current change, the force sensing range and sensitivity may depend on the ZnO NRs’ areas. In our case, Dev. 2 is designed with a larger area of piezoelectric NRs compared to Dev. 1 and has been observed to possess a larger force sensing range but a lower sensitivity under the same force. On the other hand, the encapsulation layer (such as epoxy, PP plastic, Parylene C) could possibly have an influence by sharing strain and capacitance coupling,46 leading to the different degrees of output signals. For the encapsulation, as shown in Table 1, the device in Supporting Information Section S4 has a larger sensing range (1 to 10 N) by the Parylene C coating than the device with the PP plastic plate (Dev. 1 and Dev. 2).
In addition to the piezoelectric effect of ZnO NRs, the origin of charge redistribution in ZnO NRs could be possibly due to the contacting/friction interface charges between ZnO NRs and PP plastic55 or the electrical coupling between ZnO NRs and calibration weights (steel). Furthermore, 2D transition metal dichalcogenides (including MoS2 and WSe2) have been reported to possess a piezoelectric effect;56−59 however, the piezoelectric response of vertical ZnO NRs with a high aspect ratio is believed to be dominant for the output of the mixed-dimensional device.14 Moreover, the overall hybrid transistor can be sensitive to other environmental factors, such as gas and light (Supporting Information S3 and refs (23, 25, 26, and 34)), due to more complex mechanisms.
This work demonstrates the possibility of the 1D–2D transistors as a force sensing unit with the modification of carrier distribution in the channel by the piezoelectric effect of ZnO NRs. The force sensing performance of the 1D–2D transistors is relatively limited compared to mature force sensors, mainly due to the unstable mechanical/charge coupling between ZnO NRs and encapsulation layer, together with the inconsistent electrical performance of WSe2 after the hydrothermal growth at 90 °C. In addition, the top sides of the ZnO NRs are floated and in contact with the encapsulation layer directly, which could lead to unstable residue charges on the top gate of the transistor. The shifting between the ZnO NRs and the encapsulation layer under mechanical loadings could degrade the consistency of the force sensing, also resulting in relatively large hysteresis and noise. A grounded metal layer and a larger encapsulation layer could be desirable to improve the overall mechanical/charge coupling. In addition, a large-area deposition method of the 2D WSe2 with high quality and uniformity is still required to achieve the high-density integration and the force-mapping system.
Overall, with a more controlled process and systematic parametric study in the future, the hybrid transistor could be designed as a multisensing unit in future smart sensor networks or medical instruments.
3. Conclusions
In conclusion, we have proposed the concept and demonstrated the possibility of the 1D–2D transistor as a force-sensing unit by integrating 1D ZnO piezoelectric gate control on the multilayer WSe2 channel. Two example devices have been fabricated and discussed. The electrical properties of WSe2 FET have been found to play a significant role in the direction of Ids change and the sensitivity of the overall force-sensing transistor. The pressure from a calibration weight of 5 g has been observed to result in ∼30% Ids positive change for ZnO NRs on the p-type doped WSe2 device, and ∼10% Ids negative change for the device with an n-type doped WSe2. The percentage change of Ids ((ΔIds/Ids)%) shows a dependence on the applied back-gate voltage, which provides the possibility of the integrated transistor as an active force sensor. The overall results show the possibility of the 1D–2D transistor as a nanoscale multisensing unit with tunable Ids changing directions and sensitivities. With a more controlled process and systematic parametric study in the future, the hybrid transistor could be designed as a multisensing unit in smart sensor networks or medical instruments.
4. Materials and Methods
4.1. Fabrication of the WSe2 Transistor
Multilayer WSe2 has been exfoliated and transferred onto Si substrates with 300 nm SiO2. The WSe2 flake has been patterned with photolithography and XeF2 vapor etching (25 sccm vapor XeF2 with 100 sccm N2 at 1 Torr for 3–5 min). The RIE system with CHF3 has been used to label the location of the flakes by etching 25 nm thick of SiO2. After stripping the photoresist and cleaning the substrate by solvent NMP 1165 remover, another step of photolithography has been performed to define the area of the electrode. Electron beam evaporation has been used to deposit 100 nm thick Ti as the electrode together with the lift-off technique.
4.2. Integration of ZnO NRs on the WSe2 Transistor
First, photolithography has been performed to define the area for the ZnO seed layer and ZnO NRs. After exposing the area, electron beam evaporation has been performed to deposit 50 nm ZnO (Dev. 1) or 10 nm Ti/50 nm ZnO (Dev. 2) as the seed layer. It is worth noting that the ZnO seed layer has been found to be absent on photoresists without a Ti intermediate seed layer. Before the stripping of photoresist, ZnO NRs have been grown hydrothermally on the seed layer. The equimolar (1:1) zinc nitrate and HMTA have been mixed in the DI water (concentration: 40 mM) as the hydrothermal chemical solution. ZnO NRs have been grown at 90 °C for 3 h in the chemical solution. After the growth, the photoresist has been stripped, and the seed layer and NRs have been lifted off by NMP 1165 remover.
4.3. Characterization
The thickness of WSe2 has been measured by AFM, and the overall structure has been characterized by SEM. The electrical properties of the transistors have been measured before and after the integration of ZnO NRs. Before the force-sensing characterization, a PP plastic plate was put on the device for encapsulation. The steel calibration weights (1–5 g for Dev. 1 and 1–50 g for Dev. 2) have been used to apply the mechanical force on the device. All of the electrical characterizations have been performed by a Keithley 4200 semiconductor analyzer.
Acknowledgments
The authors acknowledge the financial support of the U.K. Engineering and Physical Sciences Research Council (EPSRC) for this work. The China Scholarship Council (CSC) is acknowledged for financial support through the Ph.D. scholarship at the University of Edinburgh. The work was performed in the Scottish Microelectronics Centre and Institute for Integrated Micro and Nano Systems, University of Edinburgh.
Glossary
Abbreviations
- FET
field-effect transistor
- ZnO
zinc oxide
- WSe2
tungsten diselenide
Supporting Information Available
The Supporting Information is available free of charge at https://pubs.acs.org/doi/10.1021/acsami.2c15730.
Section S1 describing a supplementary method for device fabrication, Section S2 providing details on the force-sensing characterization of Dev. 1, Section S3 showing the light intensity sensing measurement of Dev. 1, and Section S4 showing the supplementary device (ZnO–n-type doped WSe2 with Parylene C for encapsulation) (PDF)
Author Contributions
The manuscript was written through contributions of all authors. All authors have given approval to the final version of the manuscript. Y.G. and R.C. conceived and led the project. Y.G. performed the experiments and the data analysis. M.A.B.C.M. provided support during the ZnO NR growth. P.L., E.M., and R.C. provided the experimental tools and supervision during the project. Y.G., J.X., M.M.S., and R.C. contributed to the concept explanation, writing, and proofreading of this paper.
The authors declare no competing financial interest.
Author Status
§ (E.M.) Deceased July 2019.
Supplementary Material
References
- Lee B.; Oh J.-Y.; Cho H.; Joo C. W.; Yoon H.; Jeong S.; Oh E.; Byun J.; Kim H.; Lee S.; Seo J.; Park C. W.; Choi S.; Park N.-M.; Kang S.-Y.; Hwang C.-S.; Ahn S.-D.; Lee J.-I.; Hong Y. Ultraflexible and Transparent Electroluminescent Skin for Real-Time and Super-Resolution Imaging of Pressure Distribution. Nat. Commun. 2020, 11 (1), 663. 10.1038/s41467-020-14485-9. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Shin S.-H.; Ji S.; Choi S.; Pyo K.-H.; Wan An B.; Park J.; Kim J.; Kim J.-Y.; Lee K.-S.; Kwon S.-Y.; Heo J.; Park B.-G.; Park J.-U. Integrated Arrays of Air-Dielectric Graphene Transistors as Transparent Active-Matrix Pressure Sensors for Wide Pressure Ranges. Nat. Commun. 2017, 8 (1), 14950. 10.1038/ncomms14950. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Zang Y.; Zhang F.; Huang D.; Gao X.; Di C.; Zhu D. Flexible Suspended Gate Organic Thin-Film Transistors for Ultra-Sensitive Pressure Detection. Nat. Commun. 2015, 6 (1), 6269. 10.1038/ncomms7269. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Pan C.; Dong L.; Zhu G.; Niu S.; Yu R.; Yang Q.; Liu Y.; Wang Z. L. High-Resolution Electroluminescent Imaging of Pressure Distribution Using a Piezoelectric Nanowire LED Array. Nat. Photonics 2013, 7 (9), 752–758. 10.1038/nphoton.2013.191. [DOI] [Google Scholar]
- Yang Z. W.; Pang Y.; Zhang L.; Lu C.; Chen J.; Zhou T.; Zhang C.; Wang Z. L. Tribotronic Transistor Array as an Active Tactile Sensing System. ACS Nano 2016, 10 (12), 10912–10920. 10.1021/acsnano.6b05507. [DOI] [PubMed] [Google Scholar]
- Wu W.; Wen X.; Wang Z. L. Taxel-Addressable Matrix of Vertical-Nanowire Piezotronic Transistors for Active and Adaptive Tactile Imaging. Science (80-.) 2013, 340 (6135), 952–957. 10.1126/science.1234855. [DOI] [PubMed] [Google Scholar]
- Wei Y.; Xu Q. An Overview of Micro-Force Sensing Techniques. Sensors Actuators A Phys. 2015, 234, 359–374. 10.1016/j.sna.2015.09.028. [DOI] [Google Scholar]
- Kuru I.; Gonenc B.; Balicki M.; Handa J.; Gehlbach P.; Taylor R. H.; Iordachita I.. Force Sensing Micro-Forceps for Robot Assisted Retinal Surgery. In 2012 Annual International Conference of the IEEE Engineering in Medicine and Biology Society; IEEE: 2012; pp 1401–1404; 10.1109/EMBC.2012.6346201. [DOI] [PMC free article] [PubMed]
- Payne C. J.; Rafii-Tari H.; Marcus H. J.; Yang G.-Z.. Hand-Held Microsurgical Forceps with Force-Feedback for Micromanipulation. In 2014 IEEE International Conference on Robotics and Automation (ICRA); IEEE: 2014; pp 284–289, 10.1109/ICRA.2014.6906623. [DOI]
- Eaton W. P.; Smith J. H. Micromachined Pressure Sensors: Review and Recent Developments. Smart Mater. Struct. 1997, 6 (5), 530–539. 10.1088/0964-1726/6/5/004. [DOI] [Google Scholar]
- Song P.; Ma Z.; Ma J.; Yang L.; Wei J.; Zhao Y.; Zhang M.; Yang F.; Wang X. Recent Progress of Miniature MEMS Pressure Sensors. Micromachines 2020, 11 (1), 56. 10.3390/mi11010056. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Liu S.; Qin Y.. Metal Oxide Transistor for Tactile Imaging. Semiconducting Metal Oxide Thin-Film Transistors; IOP Publishing: 2020; pp 8–24, 10.1088/978-0-7503-2556-1ch8. [DOI] [Google Scholar]
- van Doremaele E. R. W.; Gkoupidenis P.; van de Burgt Y. Towards Organic Neuromorphic Devices for Adaptive Sensing and Novel Computing Paradigms in Bioelectronics. J. Mater. Chem. C 2019, 7 (41), 12754–12760. 10.1039/C9TC03247A. [DOI] [Google Scholar]
- Chen L.; Xue F.; Li X.; Huang X.; Wang L.; Kou J.; Wang Z. L. Strain-Gated Field Effect Transistor of a MoS2-ZnO2D-1D Hybrid Structure. ACS Nano 2016, 10 (1), 1546–1551. 10.1021/acsnano.5b07121. [DOI] [PubMed] [Google Scholar]
- Wang H. S.; Hong S. K.; Han J. H.; Jung Y. H.; Jeong H. K.; Im T. H.; Jeong C. K.; Lee B.-Y.; Kim G.; Yoo C. D.; Lee K. J. Biomimetic and Flexible Piezoelectric Mobile Acoustic Sensors with Multiresonant Ultrathin Structures for Machine Learning Biometrics. Sci. Adv. 2021, 7 (7), eabe5683. 10.1126/sciadv.abe5683. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Yeh S. K.; Hsieh M. L.; Fang W. CMOS-Based Tactile Force Sensor: A Review. IEEE Sens. J. 2021, 21 (11), 12563–12577. 10.1109/JSEN.2021.3060539. [DOI] [Google Scholar]
- Chun S.; Kim J. S.; Yoo Y.; Choi Y.; Jung S. J.; Jang D.; Lee G.; Song K. Il; Nam K. S.; Youn I.; Son D.; Pang C.; Jeong Y.; Jung H.; Kim Y. J.; Choi B. D.; Kim J.; Kim S. P.; Park W.; Park S. An Artificial Neural Tactile Sensing System. Nat. Electron. 2021, 4 (6), 429–438. 10.1038/s41928-021-00585-x. [DOI] [Google Scholar]
- Wu W.; Wang Z. L. Piezotronics and Piezo-Phototronics for Adaptive Electronics and Optoelectronics. Nat. Rev. Mater. 2016, 1 (7), 16031. 10.1038/natrevmats.2016.31. [DOI] [Google Scholar]
- Park J. B.; Song M. S.; Ghosh R.; Saroj R. K.; Hwang Y.; Tchoe Y.; Oh H.; Baek H.; Lim Y.; Kim B.; Kim S.-W.; Yi G.-C. Highly Sensitive and Flexible Pressure Sensors Using Position- and Dimension-Controlled ZnO Nanotube Arrays Grown on Graphene Films. NPG Asia Mater. 2021, 13 (1), 57. 10.1038/s41427-021-00324-w. [DOI] [Google Scholar]
- Pan C.; Zhai J.; Wang Z. L. Piezotronics and Piezo-Phototronics of Third Generation Semiconductor Nanowires. Chem. Rev. 2019, 119 (15), 9303–9359. 10.1021/acs.chemrev.8b00599. [DOI] [PubMed] [Google Scholar]
- Soleimanzadeh R.; Kolahdouz M.; Ebrahimi P.; Norouzi M.; Aghababa H.; Radamson H. Ultra-High Efficiency Piezotronic Sensing Using Piezo-Engineered FETs. Sensors Actuators A Phys. 2018, 270, 240–244. 10.1016/j.sna.2018.01.002. [DOI] [Google Scholar]
- Clavijo W.Nanowire Zinc Oxide MOSFET Pressure Sensor; Thesis, Virginia Commonwealth University, 2014; 10.25772/KGP3-CQ84. [DOI]
- Saikumar A. K.; Skaria G.; Sundaram K. B. ZnO Gate Based MOSFETs for Sensor Applications. ECS Trans. 2014, 61 (26), 65–69. 10.1149/06126.0065ecst. [DOI] [Google Scholar]
- Quang Dang V.; Kim D.-I.; Thai Duy L.; Kim B.-Y.; Hwang B.-U.; Jang M.; Shin K.-S.; Kim S.-W.; Lee N.-E. Piezoelectric Coupling in a Field-Effect Transistor with a Nanohybrid Channel of ZnO Nanorods Grown Vertically on Graphene. Nanoscale 2014, 6 (24), 15144–15150. 10.1039/C4NR04713C. [DOI] [PubMed] [Google Scholar]
- Cook B.; Liu Q.; Liu J.; Gong M.; Ewing D.; Casper M.; Stramel A.; Wu J. Facile Zinc Oxide Nanowire Growth on Graphene via a Hydrothermal Floating Method: Towards Debye Length Radius Nanowires for Ultraviolet Photodetection. J. Mater. Chem. C 2017, 5 (38), 10087–10093. 10.1039/C7TC03325G. [DOI] [Google Scholar]
- Dang V. Q.; Trung T. Q.; Kim D.-I.; Duy L. T.; Hwang B.-U.; Lee D.-W.; Kim B.-Y.; Toan L. D.; Lee N.-E. Ultrahigh Responsivity in Graphene–ZnO Nanorod Hybrid UV Photodetector. Small 2015, 11 (25), 3054–3065. 10.1002/smll.201403625. [DOI] [PubMed] [Google Scholar]
- Liu Y.; Duan X.; Shin H.-J.; Park S.; Huang Y.; Duan X. Promises and Prospects of Two-Dimensional Transistors. Nature 2021, 591 (7848), 43–53. 10.1038/s41586-021-03339-z. [DOI] [PubMed] [Google Scholar]
- Chhowalla M.; Jena D.; Zhang H. Two-Dimensional Semiconductors for Transistors. Nat. Rev. Mater. 2016, 1 (11), 16052. 10.1038/natrevmats.2016.52. [DOI] [Google Scholar]
- Ham S. S.; Lee G.-J.; Hyeon D. Y.; Kim Y.; Lim Y.; Lee M.-K.; Park J.-J.; Hwang G.-T.; Yi S.; Jeong C. K.; Park K.-I. Kinetic Motion Sensors Based on Flexible and Lead-Free Hybrid Piezoelectric Composite Energy Harvesters with Nanowires-Embedded Electrodes for Detecting Articular Movements. Compos. Part B Eng. 2021, 212, 108705. 10.1016/j.compositesb.2021.108705. [DOI] [Google Scholar]
- Xu S.; Qin Y.; Xu C.; Wei Y.; Yang R.; Wang Z. L. Self-Powered Nanowire Devices. Nat. Nanotechnol. 2010, 5 (5), 366–373. 10.1038/nnano.2010.46. [DOI] [PubMed] [Google Scholar]
- Wang Z. L.; Yang R.; Zhou J.; Qin Y.; Xu C.; Hu Y.; Xu S. Lateral Nanowire/Nanobelt Based Nanogenerators, Piezotronics and Piezo-Phototronics. Mater. Sci. Eng. R Reports 2010, 70 (3–6), 320–329. 10.1016/j.mser.2010.06.015. [DOI] [Google Scholar]
- Clavijo W. P.; Atkinson G. M.; Castano C. E.; Pestov D. Novel Low-Temperature Fabrication Process for Integrated High-Aspect Ratio Zinc Oxide Nanowire Sensors. J. Vac. Sci. Technol. B 2016, 34 (2), 022203. 10.1116/1.4943041. [DOI] [Google Scholar]
- Scrymgeour D. a; Hsu J. W. P. Correlated Piezoelectric and Electrical Properties in Individual ZnO Nanorods (Sup Info). Nano Lett. 2008, 8 (8), 2204–2209. 10.1021/nl080704n. [DOI] [PubMed] [Google Scholar]
- Vayssieres L. Growth of Arrayed Nanorods and Nanowires of ZnO from Aqueous Solutions. Adv. Mater. 2003, 15 (5), 464–466. 10.1002/adma.200390108. [DOI] [Google Scholar]
- Xu S.; Wang Z. L. One-Dimensional ZnO Nanostructures: Solution Growth and Functional Properties. Nano Res. 2011, 4 (11), 1013–1098. 10.1007/s12274-011-0160-7. [DOI] [Google Scholar]
- Tosun M.; Chan L.; Amani M.; Roy T.; Ahn G. H.; Taheri P.; Carraro C.; Ager J. W.; Maboudian R.; Javey A. Air-Stable n-Doping of WSe2 by Anion Vacancy Formation with Mild Plasma Treatment. ACS Nano 2016, 10 (7), 6853–6860. 10.1021/acsnano.6b02521. [DOI] [PubMed] [Google Scholar]
- Liu B.; Ma Y.; Zhang A.; Chen L.; Abbas A. N.; Liu Y.; Shen C.; Wan H.; Zhou C. High-Performance WSe2 Field-Effect Transistors via Controlled Formation of In-Plane Heterojunctions. ACS Nano 2016, 10 (5), 5153–5160. 10.1021/acsnano.6b00527. [DOI] [PubMed] [Google Scholar]
- Zhang R.; Drysdale D.; Koutsos V.; Cheung R. Controlled Layer Thinning and P-Type Doping of WSe2 by Vapor XeF2. Adv. Funct. Mater. 2017, 27 (41), 1702455. 10.1002/adfm.201702455. [DOI] [Google Scholar]
- Zhang R.; Koutsos V.; Cheung R. Elastic Properties of Suspended Multilayer WSe2. Appl. Phys. Lett. 2016, 108 (4), 042104. 10.1063/1.4940982. [DOI] [Google Scholar]
- Geng Y.; Jeronimo K.; Bin Che Mahzan M. A.; Lomax P.; Mastropaolo E.; Cheung R. Comparison of ZnO Nanowires Grown on E-Beam Evaporated Ag and ZnO Seed Layers. Nanoscale Adv. 2020, 2 (7), 2814–2823. 10.1039/C9NA00553F. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Hoffman A. N.; Stanford M. G.; Zhang C.; Ivanov I. N.; Oyedele A. D.; Sales M. G.; McDonnell S. J.; Koehler M. R.; Mandrus D. G.; Liang L.; Sumpter B. G.; Xiao K.; Rack P. D. Atmospheric and Long-Term Aging Effects on the Electrical Properties of Variable Thickness WSe2 Transistors. ACS Appl. Mater. Interfaces 2018, 10 (42), 36540–36548. 10.1021/acsami.8b12545. [DOI] [PubMed] [Google Scholar]
- Fang H.; Chuang S.; Chang T. C.; Takei K.; Takahashi T.; Javey A. High-Performance Single Layered WSe2 p-FETs with Chemically Doped Contacts. Nano Lett. 2012, 12 (7), 3788–3792. 10.1021/nl301702r. [DOI] [PubMed] [Google Scholar]
- Schulman D. S.; Arnold A. J.; Das S. Contact Engineering for 2D Materials and Devices. Chem. Soc. Rev. 2018, 47 (9), 3037–3058. 10.1039/C7CS00828G. [DOI] [PubMed] [Google Scholar]
- Ma N.; Jena D. Charge Scattering and Mobility in Atomically Thin Semiconductors. Phys. Rev. X 2014, 4 (1), 011043. 10.1103/PhysRevX.4.011043. [DOI] [Google Scholar]
- Roy A.; Mead J.; Wang S.; Huang H. Effects of Surface Defects on the Mechanical Properties of ZnO Nanowires. Sci. Rep. 2017, 7 (1), 9547. 10.1038/s41598-017-09843-5. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Geng Y.; Bin Che Mahzan M. A.; Jeronimo K.; Saleem M. M.; Lomax P.; Mastropaolo E.; Cheung R. Integration of ZnO Nanorods with MOS Capacitor for Self-Powered Force Sensors and Nanogenerators. Nanotechnology 2021, 32 (45), 455502. 10.1088/1361-6528/ac19d7. [DOI] [PubMed] [Google Scholar]
- Janotti A.; Van de Walle C. G. Fundamentals of Zinc Oxide as a Semiconductor. Rep. Prog. Phys. 2009, 72 (12), 126501. 10.1088/0034-4885/72/12/126501. [DOI] [Google Scholar]
- Wang Z. L.; Song J. Piezoelectric Nanogenerators Based on Zinc Oxide Nanowire Arrays. Science (80-.) 2006, 312 (5771), 242–246. 10.1126/science.1124005. [DOI] [PubMed] [Google Scholar]
- Wang Z.; Li Q.; Chen Y.; Cui B.; Li Y.; Besenbacher F.; Dong M. The Ambipolar Transport Behavior of WSe2 Transistors and Its Analogue Circuits. NPG Asia Mater. 2018, 10 (8), 703–712. 10.1038/s41427-018-0062-1. [DOI] [Google Scholar]
- Pudasaini P. R.; Oyedele A.; Zhang C.; Stanford M. G.; Cross N.; Wong A. T.; Hoffman A. N.; Xiao K.; Duscher G.; Mandrus D. G.; Ward T. Z.; Rack P. D. High-Performance Multilayer WSe2 Field-Effect Transistors with Carrier Type Control. Nano Res. 2018, 11 (2), 722–730. 10.1007/s12274-017-1681-5. [DOI] [Google Scholar]
- Liu W.; Kang J.; Sarkar D.; Khatami Y.; Jena D.; Banerjee K. Role of Metal Contacts in Designing High-Performance Monolayer n-Type WSe2 Field Effect Transistors. Nano Lett. 2013, 13 (5), 1983–1990. 10.1021/nl304777e. [DOI] [PubMed] [Google Scholar]
- Geng Y.; Jeronimo K.; Bin Che Mahzan M. A.; Lomax P.; Mastropaolo E.; Cheung R. Simplified Patterning Process for the Selective 1D ZnO Nanorods Growth. J. Vac. Sci. Technol. B 2020, 38 (1), 012204. 10.1116/1.5131363. [DOI] [Google Scholar]
- Lee J. M.; No Y.-S. S.; Kim S.; Park H.-G. G.; Park W. Il. Strong Interactive Growth Behaviours in Solution-Phase Synthesis of Three-Dimensional Metal Oxide Nanostructures. Nat. Commun. 2015, 6 (1), 6325. 10.1038/ncomms7325. [DOI] [PubMed] [Google Scholar]
- Yang D.; Qiu Y.; Jiang Q.; Guo Z.; Song W.; Xu J.; Zong Y.; Feng Q.; Sun X. Patterned Growth of ZnO Nanowires on Flexible Substrates for Enhanced Performance of Flexible Piezoelectric Nanogenerators. Appl. Phys. Lett. 2017, 110 (6), 063901. 10.1063/1.4975477. [DOI] [Google Scholar]
- Venugopal K.; Panchatcharam P.; Chandrasekhar A.; Shanmugasundaram V. Comprehensive Review on Triboelectric Nanogenerator Based Wrist Pulse Measurement: Sensor Fabrication and Diagnosis of Arterial Pressure. ACS Sensors 2021, 6 (5), 1681–1694. 10.1021/acssensors.0c02324. [DOI] [PubMed] [Google Scholar]
- Seo J.; Kim Y.; Park W. Y.; Son J. Y.; Jeong C. K.; Kim H.; Kim W. H. Out-of-Plane Piezoresponse of Monolayer MoS2 on Plastic Substrates Enabled by Highly Uniform and Layer-Controllable CVD. Appl. Surf. Sci. 2019, 487 (May), 1356–1361. 10.1016/j.apsusc.2019.05.140. [DOI] [Google Scholar]
- Yarajena S. S.; Biswas R.; Raghunathan V.; Naik A. K. Quantitative Probe for In-Plane Piezoelectric Coupling in 2D Materials. Sci. Rep. 2021, 11 (1), 1–9. 10.1038/s41598-021-86252-9. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Cui C.; Xue F.; Hu W.-J.; Li L.-J. Two-Dimensional Materials with Piezoelectric and Ferroelectric Functionalities. npj 2D Mater. Appl. 2018, 2 (1), 18. 10.1038/s41699-018-0063-5. [DOI] [Google Scholar]
- Lee J.-H.; Park J. Y.; Cho E. B.; Kim T. Y.; Han S. A.; Kim T.-H.; Liu Y.; Kim S. K.; Roh C. J.; Yoon H.-J.; Ryu H.; Seung W.; Lee J. S.; Lee J.; Kim S.-W. Reliable Piezoelectricity in Bilayer WSe2 for Piezoelectric Nanogenerators. Adv. Mater. 2017, 29 (29), 1606667. 10.1002/adma.201606667. [DOI] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.

