Self-powered monolithic accelerometer using a photonic gate€¦ · simplified. Nature provides different energy sources that can be exploited to power electronic and sensing devices

Nano Energy 76 (2020) 104950

Available online 2 July 20202211-2855/© 2020 Elsevier Ltd. All rights reserved.

Self-powered monolithic accelerometer using a photonic gate

Thanh Nguyen a,*, Toan Dinh a,b,**, Hoang-Phuong Phan a, Van Thanh Dau c, Tuan-Khoa Nguyen a, Abbin Perunnilathil Joy d, Behraad Bahreyni d, Afzaal Qamar e, Mina Rais-Zadeh f, Debbie G. Senesky g, Nam-Trung Nguyen a, Dzung Viet Dao a,c

a Queensland Micro- and Nanotechnology Centre, Griffith University, Queensland, 4111, Australia b School of Mechanical and Electrical Engineering, University of Southern Queensland, Queensland, 4350, Australia c School of Engineering and Built Environment, Griffith University, Queensland, 4215, Australia d School of Mechatronic Systems Engineering, Simon Fraser University, Canada e Department of Electrical Engineering and Computer Science, University of Michigan, Ann Arbor, MI, 48109, USA f NASA Jet Propulsion Laboratory, California Institute of Technology, Pasadena, CA, 91109, USA g Department of Aeronautics and Astronautics, Stanford University, California, 94305, USA

A R T I C L E I N F O

Keywords: Self-powered sensor Photonic gate Monolithic Light harvesting Silicon carbide Accelerometer

A B S T R A C T

Harvesting sustainable energy resources from surrounding environments to power small electronic devices and systems has attracted massive research attention. Herein, we develop a novel technology to harvest light energy to self-power and simultaneously sense mechanical acceleration in a monolithic structure. When the photonic gate is illuminated the operation mode of the device changes from conventional mode to light harvesting and self-powered operation. The light illumination provides a gradient of majority carrier concentration on the top semiconductor layer, generating a lateral photovoltage, which is the output voltage of the sensor. Under ac-celeration, the mechanical inertial force induces stress in the sensor material leading to the change of mobility of the charge carriers, which shifts their diffusion rate, and hence changes the gradient of the majority carriers and the lateral photovoltage. The sensitivity at 480 lx light illumination was measured to be 107 μV=g, while it was approximately 30 μV=g under the ambient light illumination without any electrical power source. In addition, the acceleration sensitivity is tunable by controlling parameters of the photonic gate such as light power, light spot position and light wavelength. The integration of sensing and powering functions into a monolithic platform proposed in this work eliminates the requirement of external power sources and offers potential solutions for wireless, independent, remote, and battery-free sensing devices and systems.

1. Introduction

Self-powered devices are anticipated to play a critical role in the future development of micro/nano-systems due to numerous advantages such as zero power consumption, miniaturization, wireless communi-cation, and self-sustainability [1–10]. Self-powering is highly desired for wireless and battery-free devices, or sensors operating in harsh envi-ronments (high temperature or corrosive environments), which not only significantly improve the adaptability of devices but also reduce device sizes and maintenance cost [1,11,12]. A self-powered sensor can either directly utilize the energy from detected signals [13] or harvest energy from other source and utilize it [14,15]. With the latter, the two com-mon configurations are (i) sensing module and harvesting module are

two separated structures integrated in one device [2,16] or (ii) both the sensing and harvesting are performed by a monolithic structure [14,15, 17,18]. Using a monolithic structure, the sensor size can be significantly reduced, and fabrication and packaging processes can be substantially simplified.

Nature provides different energy sources that can be exploited to power electronic and sensing devices such as solar power, heat gradient, mechanical vibrations, or acoustic waves [3,19–22]. Mechanical energy sources rely on motions, e.g. from body activities (body movement, muscle stretching, blood vessel contraction) [3,14]. However, these energy sources are random and have a low frequency range, which result in relatively low conversion efficiency in conventional energy trans-duction methods such as electromagnetic generators [19,23]. To

* Corresponding author. ** Corresponding author. Queensland Micro- and Nanotechnology Centre, Griffith University, Queensland, 4111, Australia.

E-mail addresses: [email protected] (T. Nguyen), [email protected] (T. Dinh).

Contents lists available at ScienceDirect

Nano Energy

journal homepage: http://www.elsevier.com/locate/nanoen

https://doi.org/10.1016/j.nanoen.2020.104950 Received 3 March 2020; Received in revised form 14 April 2020; Accepted 8 May 2020

Nano Energy 76 (2020) 104950

2

overcome this drawback, nanogenerators (NGs), which rely on the piezoelectric potential created in nanowires, have been developed [3, 23–27]. By using arrays of zinc oxide nanowires, for example, Wang and Song [24] demonstrated the first piezoelectric NG with an efficiency estimated from 17% to 30%. On the other hand, basing on the tribo-electric effect, triboelectric NGs also have been developed to scavenge energy from vibrations and random displacements/deformation [28–32]. However, one of the greatest disadvantages of these NGs is that they only work with particular materials, resulting in complicated, if not practically impossible, integration challenges with silicon devices. In addition, the power output of these NGs, which includes millions of integrated nanowires, is still too small to power conventional electronic devices [33]. Besides mechanical energy sources, solar power is also considered as the most ubiquitous and abundant energy source. With miniaturization, high-efficient solar cells can be implemented in self-powered micro- and nanosystems. However, these systems are

usually sophisticated with high manufacturing costs, which limits their utilization. Therefore, it is extremely beneficial to develop low cost, reliable, adaptable, simple and energy harvesting micro/nano devices.

In this paper, we introduce a novel technology that harvests light energy to self-power an acceleration sensor in a monolithic structure without any other power source or separate energy conversion devices. A cantilever-shaped 3C–SiC/Si accelerometer with a proof-mass was developed. The sensor was characterized in both conventional and self- powered modes. In the self-powered mode, the accelerometer showed the capability of harvesting light energy with the sensitivity of approx-imately 107 μV=g under a light intensity of 480 lx. Even under ambient light condition, the accelerometer displayed an excellent performance with the sensitivity of approximately 30 μV=g. In addition, the acceler-ation sensitivity can be tuned by controlling parameters of the photonic gate such as changing light wavelength, adjusting light intensity or light position. The ability of self-powering and harvesting light energy, and

Fig. 1. Material characteristics and the accelerometer. (a) Transmission electron microscopy image of p-type 3C– SiC grown on a p-type (100) Si substrate. (b) X- ray diffraction analysis graph. (c) Selected area electron diffraction image. Reprinted from Ref. [34], with the permission of AIP Publishing. (d) Fabrication process of the accelerometer. After growing 3C–SiC on a Si substrate, 3C–SiC piezoresistors were patterned via a dry etching process. Next, aluminum electrodes were formed through two consecutive processes: depositing aluminum on top of the 3C–SiC layer and etching aluminum. Then, 3C–SiC/Si membranes were formed by ablating Si from the back side using laser and cantilever and proof mass structures were built through cutting the material layers by laser from top side. Finally, the acceler-ometers were separated via dicing process. (e) As-fabricated accelerometer and enlarged view of the sensing element.

T. Nguyen et al.

Nano Energy 76 (2020) 104950

3

tunability of the acceleration sensitivity are also analysed and explained.

2. Methods

2.1. Growing 3C–SiC on Si

Using silane and propene as precursor materials to supply silicon and carbon atoms, a 3C–SiC thin film with a thickness of 300 nm was grown on a p-type single crystalline Si (100) substrate by LPCVD (low pressure chemical vapour disposition) in a hot-wall chamber at a temperature of 1150�C. Simultaneously, through an in situ process, aluminum atoms were doped into the 3C–SiC film using trimethyl-aluminum as a pre-cursor compound to form p-type 3C–SiC film. Thickness of the 3C–SiC was controlled by controlling the growth time. Doping concentrations in the 3C–SiC and Si layers were 1018 cm-3 and 1014 cm-3, respectively, evaluated by the hot probe and Hall effect techniques. The transmission electron microscopy image (Fig. 1a) exhibited that 3C–SiC was epitax-ially grown on the Si substrate and there are no grain boundaries, and the X-ray diffraction analysis (Fig. 1b) confirmed the epitaxial growth of 3C–SiC on Si substrate. The selected area electron diffraction image is shown in Fig. 1c. Patterns of diffraction spots are parallel lines and there were not any weak diffraction spots around bright spots. These confirmed that 3C–SiC grown on the Si substrate was in single crystalline structure without crystal defects.

2.2. Fabrication process

We fabricated a monolithic accelerometer through a process depic-ted in Fig. 1d. After the 3C–SiC growing process, a 3C–SiC piezoresistor was created by patterning 3C–SiC layer through conventional photoli-thography and reactive ion etching. Then, SiC/Si membranes with a thickness of 100 μm were formed using laser ablation from the back side. Finally, the cantilever was created using laser cutting on the front side, simultaneously forming a proof-mass at the free end of the cantilever. Fig. 1e shows a photograph of the as-fabricated accelerometer along with the SEM (Scanning electron microscope) image of the SiC piezor-esistor. Ohmic contacts between the aluminum electrodes and the 3C–SiC layer was confirmed by the linearity of the I–V (current-voltage) characteristic, as shown in Supplementary Fig. S1.

2.3. Sensor characterisation

Fig. 2a shows a schematic of the experimental setup used to evaluate the dynamic characteristics of the accelerometer. The device was mounted onto the surface of a high-frequency shaker (SmartShaker model K2004E01). Vibration amplitude and frequency of the shaker were accurately controlled via a closed-loop system including a Spider-

81B vibration controller and a reference accelerometer. Feedback signal from the reference accelerometer was used by the vibration controller to calculate sinusoidal output for driving the shaker. Output voltage across the sensing element of the accelerometer was continuously recorded by a lock-in amplifier. The output signal of the lock-in amplifier indicated output signal of the accelerometer after filtering noises. Under experi-ments using photonic gate, the incident light with intensity of 480 lx was used to illuminate the sensing element and the device was also tested under an ambient light condition with intensity of approximately 213 lx.

3. Results and discussions

Fig. 2b and c present the performance of the accelerometer under conventional mode, in which an electrical current was supplied to the sensing element and the output voltage was measured to evaluate the applied acceleration. As shown in Fig. 2b, the output voltage increased linearly with the amplitude of the external acceleration, and the accel-eration sensitivity was approximately 145 mV/g, 77 mV/g, and 17 mV/g when the supplied current was 100 μA, 50 μA and 10 μA respectively, which are comparable with results from previous reports [35–38]. In term of power consumption, to reach the sensitivity of 145 mV/g, 77 mV/g, 17 mV/g, the applied powers were 939 μW, 240 μW, and 9.78 μW, respectively. Even with a supply power as small as 11.4 nW (0.1 μA supplied current), our devices still can operate excellently with a sensitivity of around 163 μV/g. More details of the operation of the accelerometer under conventional mode is presented in section 3 of the Supplementary.

We then investigated performance of the accelerometer under light illumination (i.e. self-powered mode) without electrical power supply. As shown in Fig. 3a, under 480 lx illumination without any supplied currents, the device was used to measure the acceleration ranging from 0.5g to 8g (m=s2), demonstrating a sensitivity of 107 μV/g. Even under room light (intensity of 213 lx), the sensor can detect the same range of acceleration with a sensitivity of approximately 30 μV/g. The results demonstrated that instead of using electrical source to power the de-vices, the accelerometer can function with a photonic gate (optical illumination), which may have application in wireless systems and miniaturized devices. The devices with the photonic gate are expected to work well in harsh environments, as it was developed based on SiC along with the elimination of battery. In addition, the sensitivity can be tuned by changing illumination conditions as shown in Fig. 3a. Under ambient light condition, the output voltage increased linearly with the external acceleration at a sensitivity of approximately 30 μV/g. By using an external light source with an intensity of 480 lx, the acceleration sensitivity increased by more than three times, reaching 107 μV/g. As such, the light illumination played a role as a gate (i.e. photonic gate) that can change the operation modes of the accelerometer from the

Fig. 2. Dynamic experiment and operation of the accelerometer under conventional mode. (a) Schematic experiment. The accelerometer was firmly mounted on a shaker whose vibration was accurately controlled by a vibration controller. The output signal of the accelerometer was filtered and amplified by a lock-in amplifier. To demonstrate operation of the accelerometer under the self-powered and light-harvesting mode, an external white light with intensity of approxi-mately 480 lx illuminated the sensing element. (b) Characteristics of the output voltage versus acceleration under conventional mode. (c) The relationship between sensitivity and supply power with supplied current.

T. Nguyen et al.

Nano Energy 76 (2020) 104950

4

conventional mode (no light illumination, i.e. photonic gate is OFF) to the self-powered and light-harvesting mode (light illumination, i.e. photonic gate is ON) as well as tune the performance of the acceler-ometer by changing parameters of the photonic gate. The possibility of harvesting light energy for sensing acceleration, powering the acceler-ometer as well as tuning the sensitivity could result from the generation of a lateral photovoltaic voltage (i.e. lateral photovoltage) on the 3C–SiC thin film across the sensing element under illumination. We observed that under the illumination, and even under room light condition, there was a generated lateral photovoltage dropping across the sensing element. Fig. 3b shows repeatability of the photovoltage dropping across the sensing element when the light source with intensity of 480 lx was repeatedly switched ON and OFF. Under 480 lx light the generated photovoltage was approximately 10.5 mV, while that value under room light condition was approximately 1.85 mV, and there was no voltage generated in dark conditions. This was also confirmed by I–V (current- voltage) characteristics of the accelerometer under the illumination and dark condition (Fig. 3c). The I–V curve does not go through the origin under the illumination.

The generation of lateral photovoltage on the top material layer of a heterojunction structure under illumination can be explained based on the lateral photovoltaic effect [39–41,43]. As shown in Fig. 4a–1, when a beam of the incident light was locally illuminated on the top layer of the heterojunction structure at position C, there were plenty of photons injected into material layers. Because the 3C–SiC layer was very thin and transparent, most of the photons were absorbed within the junction region (depletion region) or in the Si substrate [39,40]. Electrons absorbing energy from photons with energies higher than the band gap energy become free electrons which jumped to the conduction bands and leaved holes in the valence bands, generating electron-hole pairs (EHPs). The EHPs generated in the junction were immediately separated by the built-in electric field, which had direction from the Si substrate to the 3C–SiC thin film. Therefore, the photogenerated holes drifted to the 3C–SiC layer while the electrons drifted in the opposite direction to the Si substrate. A number of holes generated in the Si substrate were injected to 3C–SiC by the tunneling mechanism, while the remaining holes diffused within the Si substrate and then recombined with the electrons. There were excess holes (major carriers) injected into 3C–SiC and excess electrons in Si. The excess major carriers injected into 3C–SiC diffused away from the illumination spot radically due to hole gradient.

Part of these holes diffused and were collected at the two electrode positions, A and B. There was a hole gradient within 3C–SiC from the illumination spot to the two electrode positions. If the diffusion from illumination spot C to the two electrodes A and B was different, the hole

concentrations at positions A and B were distinctive, resulting in the lateral photovoltage dropping across the two electrodes. This voltage, which was proportional with gradient of hole concentration between A and B, depended on the number of holes injected to 3C–SiC layer at position C and the number of holes diffusing to positions A and B. While the number of holes injected to the 3C–SiC layer at position C depended on the light source such as light wavelength and light power, the numbers of holes diffusing to positions A to B depended on the initial gradient of hole concentration and mobility of holes within 3C–SiC. In other words, the lateral photovoltage depended on the light source, relative positions between the illumination spot and the two electrodes, and the mobility of major carriers, (Supplementary Fig. S5). At different light wavelengths or light intensities, there was a difference in the number of holes injected to 3C–SiC layer which resulted in the difference in hole concentration and lateral photovoltage. The lateral photovoltage between the two points depended on their relative positions from the illumination spot. For example, when the light spot moved from one electrode to the other electrode, the lateral photovoltage linearly increased from a negative maximum value to a positive maximum value [40].

Besides depending on the light source, the diffusion of holes (Γh) in the 3C–SiC layer significantly affected the generation of lateral photo-voltage. The diffusion flux of holes in the SiC layer is calculated by Ref. [42]:

Γh¼ � Dhdpdx

(1)

where, Dh is the hole diffusion coefficient and dp=dx is the hole con-centration gradient. The hole diffusion coefficient, which is determined through Einstein relation Dh ¼ μh

kBTe , is proportional to the hole mobility

μh. In other words, increasing hole mobility increases the diffusion flux, which in turn decreases the gradient of hole concentration, leading to a diminution in the generated lateral photovoltage. Assume that illumi-nation spot was at electrode A and illumination condition was kept consistently, the hole concentration at electrode A remained constant. Because of the difference in hole mobility such as under compressive and tensile conditions, the number of hole diffusing to the electrode B po-sition was higher for higher hole mobility, hence decreasing the slope of hole concentration profile, which in turn generated smaller lateral photovoltage, Fig. 4b.

The light harvesting and self-powered feasibility of our accelerom-eter can be explained as following. (i) Under nonuniform illumination and zero acceleration, Fig. 6a–1, which was no strain and μh@0 hole mobility condition (Fig. 6b–1), there was a lateral photovoltage V0

Fig. 3. Operation of the accelerometer under self-powered mode. (a) Light harvesting and sensitivity tunability by photonic gate. The accelerometer harvested light energy supplied by a photonic gate to self-power and synchronously sense acceleration in a monolithic structure. Under 480 lx illumination, the accelerometer could work independently without supplying electrical power with a sensitivity of 107 μV=g. Even under room light condition (light intensity of 213 lx), the accelerometer could operate excellent with sensitivity of approximately 30 μV=g. (b) Repeatability of the generated lateral photovoltage when the light was peri-odically turned ON/OFF. When a white light with intensity of 480 lx illuminated the sensing element, a lateral photovoltage of approximately 10.5 mV was generated. (c) Current-voltage (I–V)) characteristics of the accelerometer under illumination and dark condition.

T. Nguyen et al.

Nano Energy 76 (2020) 104950

5

across the two electrodes, which was proportional with the hole gradient or the difference of Femi levels between position A and B in the 3C–SiC layer (Fig. 6c–1). Therefore, under zero acceleration (Fig. 6a–1), the output voltage was constant at the value of V0 (mV) (Fig. 6d–1). (ii) Positive acceleration (Fig. 6a–2) induced tensile strain (as simulatedly demonstrated in Fig. 5a) and resulted in a reduction in hole mobility (μh@þ ¼ μh@0 � δμh@þ) (Fig. 6b–2), and an increase in hole gradient (Fig. 6c–2) or Femi level difference, which caused an increase in the generated lateral photovoltage as aforementioned. As the acceleration increased from zero to A0, the lateral voltage respectively increased from V0 to Vmax as shown in Fig. 6a–2 and Fig. 6d–2, the relationship between positive applied acceleration and output voltage. (iii) On the other hand, under negative acceleration (Fig. 6a–3), the induced compressive strain (as simulatedly demonstrated in Fig. 5b) and an in-crease in hole mobility.

(μh@� ¼ μh@0þ δμh@� ) resulted in reductions in slope of the hole profile and the Femi level gap, hence reduced the lateral photovoltage.

Therefore, when the acceleration decreased from zero to � A0 (Fig. 6a–3), the lateral voltage decreased respectively from V0 to Vmin (Fig. 6d–3). Consequently, as the device vibrated in a sinusoidal shape, the output voltage changed sinusoidally.

Sensitivity of the accelerometer can be calculated:

S¼Vmax � Vmin

ffiffiffi2p

12A0¼

ΔV0

2ffiffiffi2p

A0(2)

ΔV0 is the change of the lateral photovoltage due to applying acceler-ation, which is proportional with change of the hole diffusion flux. Therefore, from equation (2), ΔV0 is proportional with the hole mobility change Δμh and hole concentration gradient dp=dx in the 3C–SiC thin film. While the hole mobility change depends on properties of the ma-terial and applied acceleration, the hole concentration gradient depends on illumination condition. In other words, the sensitivity of the accel-erometer can be tuned by controlling photonic gate (changing

Fig. 4. Generation of the lateral photovoltage. (a) Generation mechanism of lateral photovoltage on pþ-type SiC/p-type Si heterostructure. When the light illuminated onto the top surface of the device at position C, electron-hole pairs were generated in the depletion and substrate areas. A number of these holes then migrated to the 3C–SiC layer around the illuminated area (position C) by the built-in electric field and quantum tunneling mechanism. These excess holes diffused away from position C due to hole gradient. A number of these holes were collected at the electrode areas (A and B). If the diffusion from illumination area C to the two electrodes A and B was different, a lateral photovoltage was formed between electrodes A and B. (b) Hole concentration profiles in the 3C–SiC layer under compressive and tensile strains. Assume that illumination spot was coincide at electrode A, under the same illumination condition (light intensity, light wavelength), magnitude of the lateral photovoltage between electrodes A and B depended on hole mobility. Hole mobility was higher under compressive strain compared to that under tensile strain. Therefore, more holes diffused to electrode B under compressive strain which resulted in smaller generated lateral photovoltage.

Fig. 5. Relationship between induced stress and external acceleration. (a) Induced stress under positive acceleration. Under positive acceleration (upward acceleration or downward deceleration), there was tensile stress induced in the sensing element. (b) Induced stress under negative acceleration. Under negative acceleration (downward acceleration or upward deceleration), there was compressive stress induced in the sensing element.

T. Nguyen et al.

Nano Energy 76 (2020) 104950

6

illumination conditions).

4. Conclusion

In conclusion, we successfully demonstrated an innovative technol-ogy to harvest light energy and self-power accelerometer in a monolithic structure. We also experimentally demonstrated and theoretically dis-cussed the tunability of the acceleration sensitivity by controlling the photonic gate. Under 480 lx light illumination of the photonic gate, the

accelerometer can function without electrical power, offering a sensi-tivity of 107 μV=g. The accelerometer also operated excellently at room light condition with a sensitivity of approximately 30 μV=g. In addition, the sensitivity of the sensor can be tuned by adjusting the light param-eters such as light power, light spot position, as well as light wavelength. The feasibility of working without electrical power source was attrib-uted to the generations of the lateral photovoltage within the 3C–SiC thin film due to an induced hole concentration gradient. The accelera-tion is determined via the changes of the lateral photovoltage, which

Fig. 6. Operation principle of the light harvesting accelerometer. (a) Applied external acceleration. Zero (a-1) acceleration, positive (a-2) and negative (a-3) accelerations. (b) Diffusion of the photogenerated holes under different accelerations. Under the same illumination condition, the diffusion of photogenerated holes depends on their mobility. (b-1) Zero acceleration or no train, the mobility of holes is μh@0 and diffusion flux is Γh@0. (b-2) Under positive acceleration, a tensile strain is induced in the 3C–SiC layer, resulting in decreases in the hole mobility and diffusion speed. (b-3) Under negative acceleration, a compressive strain is induced in the 3C–SiC layer, resulting in increases in the hole mobility and diffusion speed. (c) The shift of energy bands under different accelerations. Under zero acceleration (c-1), the difference in energy bands at the two electrode positions is indicated by a generated lateral photovoltage of V0. There are less holes diffusing to electrode B under positive acceleration (c-2). The energy bands shift down at the electrode A area and up at the electrode B position, a larger lateral photovoltage ðV0þΔVÞ is generated. In contrast (c-3), a smaller lateral photovoltage ðV0 � ΔVÞ is generated under negative acceleration. (d) Changes of the output voltage versus the external acceleration. The lateral photovoltage is V0 under zero acceleration (d-1). This voltage increases when a positive acceleration is applied (d-2). In contrast, the lateral photovoltage decreases from V0 under negative acceleration (d-3).

T. Nguyen et al.

Nano Energy 76 (2020) 104950

7

depends on changes in the hole mobility due to external acceleration and changes in light illumination conditions. This outstanding novel tech-nology will propel forward its practical applications to a new paradigm.

Declaration of competing interest

The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.

CRediT authorship contribution statement

Thanh Nguyen: Conceptualization, Investigation, Formal analysis, Writing - original draft, Methodology. Toan Dinh: Conceptualization, Investigation, Data curation, Writing - original draft. Hoang-Phuong Phan: Methodology, Writing - review & editing. Van Thanh Dau: Investigation. Abbin Perunnilathil Joy: Data curation. Behraad Bah-reyni: Supervision, Writing - review & editing. Afzaal Qamar: Meth-odology. Mina Rais-Zadeh: Supervision. Debbie G. Senesky: Supervision. Nam-Trung Nguyen: Supervision, Writing - review & editing, Project administration, Funding acquisition. Dzung Viet Dao: Supervision, Writing - review & editing, Project administration, Funding acquisition.

Acknowledgements

The 3C–SiC material was developed and supplied by Leonie Hold and Alan Iacopi of the Queensland Microtechnology Facility, part of the Queensland node - Griffith - of the Australian National Fabrication Fa-cility. A company established under the National Collaborative Research Infrastructure Strategy to provide nano and microfabrication facilities for Australia’s researchers. The epitaxial SiC deposition was developed as part of Griffith Universities Joint Development Agreement with SPT Microtechnology, the manufacturer of the Epiflx production reactor. This work has been partially supported by Australian Research Council grants LP160101553 and DE200100238. The project is supported by the Foundation for Australia-Japan Studies under the Rio Tinto Australia- Japan Collaboration Project. T.D. is grateful for the support from Grif-fith University/Simon Fraser University Collaborative Travel Grant 2017 and Griffith University New Research Grant 2019.

Appendix A. Supplementary data

Supplementary data to this article can be found online at https://doi. org/10.1016/j.nanoen.2020.104950.

References

[1] Lin Peng, Linfeng Hu, Xiaosheng Fang, Energy harvesting for nanostructured self- powered photodetectors, Adv. Funct. Mater. 24 (18) (2014) 2591–2610.

[2] S. Xu, Y. Qin, C. Xu, Y. Wei, R. Yang, Z.L. Wang, Self-powered nanowire devices, Nat. Nanotechnol. 5 (5) (2010) 366.

[3] Y. Qin, X. Wang, Z.L. Wang, Microfibre–nanowire hybrid structure for energy scavenging, Nature 451 (7180) (2008) 809–813.

[4] Y. Zhang, W. Xu, X. Xu, J. Cai, W. Yang, X. Fang, Self-powered dual-color uv–green photodetectors based on sno2 millimeter wire and microwires/cspbbr3 particle heterojunctions, J. Phys. Chem. Lett. 10 (4) (2019) 836–841.

[5] C. Sa, X. Xu, X. Wu, J. Chen, C. Zuo, X. Fang, A wearable helical organic–inorganic photodetector with thermoelectric generators as the power source, J. Mater. Chem. C 7 (42) (2019) 13097–13103.

[6] Y. Ning, Z. Zhang, F. Teng, X. Fang, Novel transparent and self-powered uv photodetector based on crossed zno nanofiber array homojunction, Small 14 (13) (2018) 1703754.

[7] B. Zhao, F. Wang, H. Chen, L. Zheng, L. Su, D. Zhao, X. Fang, An ultrahigh responsivity (9.7 ma w- 1) self-powered solar-blind photodetector based on individual zno–ga2o3 heterostructures, Adv. Funct. Mater. 27 (17) (2017) 1700264.

[8] P. Yu, K. Hu, H. Chen, L. Zheng, X. Fang, Novel p–p heterojunctions self-powered broadband photodetectors with ultrafast speed and high responsivity, Adv. Funct. Mater. 27 (38) (2017) 1703166.

[9] S. Das, M.J. Hossain, S.-F. Leung, A. Lenox, Y. Jung, K. Davis, J.-H. He, T. Roy, A leaf-inspired photon management scheme using optically tuned bilayer nanoparticles for ultra-thin and highly efficient photovoltaic devices, Nano Energy 58 (2019) 47–56.

[10] H.-C. Fu, P. Varadhan, M.-L. Tsai, W. Li, Q. Ding, C.-H. Lin, M. Bonifazi, A. Fratalocchi, S. Jin, H. He Jr., Improved performance and stability of photoelectrochemical water-splitting si system using a bifacial design to decouple light harvesting and electrocatalysis, Nano Energy (2020) 104478.

[11] Z.L. Wang, Triboelectric nanogenerators as new energy technology for self- powered systems and as active mechanical and chemical sensors, ACS Nano 7 (11) (2013) 9533–9557.

[12] D. Zhang, K. Zhang, Y. Wang, Y. Wang, Y. Yang, Thermoelectric effect induced electricity in stretchable graphene-polymer nanocomposites for ultrasensitive self- powered strain sensor system, Nano Energy 56 (2019) 25–32.

[13] A.R.M. Foisal, T. Dinh, V.T. Nguyen, P. Tanner, H.-P. Phan, T.-K. Nguyen, B. Haylock, E.W. Streed, M. Lobino, D.V. Dao, Self-powered broadband (uv-nir) photodetector based on 3c-sic/si heterojunction, IEEE Trans. Electron. Dev. 66 (4) (2019) 1804–1809.

[14] Y. Zi, L. Lin, J. Wang, S. Wang, J. Chen, X. Fan, P.-K. Yang, F. Yi, Z.L. Wang, Triboelectric–pyroelectric–piezoelectric hybrid cell for high-efficiency energy- harvesting and self-powered sensing, Adv. Mater. 27 (14) (2015) 2340–2347.

[15] R. Saraf, L. Pu, V. Maheshwari, A light harvesting, self-powered monolith tactile sensor based on electric field induced effects in mapbi3 perovskite, Adv. Mater. 30 (9) (2018) 1705778.

[16] J. Han, F. Fan, C. Xu, S. Lin, M. Wei, X. Duan, Z.L. Wang, Zno nanotube-based dye- sensitized solar cell and its application in self-powered devices, Nanotechnology 21 (40) (2010) 405203.

[17] K. Zhao, B. Ouyang, C.R. Bowen, Z.L. Wang, Y. Yang, One-structure-based multi- effects coupled nanogenerators for flexible and self-powered multi-functional coupled sensor systems, Nano Energy 71 (2020) 104632.

[18] Y. Wang, Y. Wang, Y. Yang, Graphene–polymer nanocomposite-based redox- induced electricity for flexible self-powered strain sensors, Adv. Energy Mater. 8 (22) (2018) 1800961.

[19] Z.L. Wang, Catch wave power in floating nets, Nature 542 (7640) (2017) 159–160. [20] S. Wang, X. Mu, Y. Yang, C. Sun, A.Y. Gu, Z.L. Wang, Flow-driven triboelectric

generator for directly powering a wireless sensor node, Adv. Mater. 27 (2) (2015) 240–248.

[21] Y. Yang, W. Guo, K.C. Pradel, G. Zhu, Y. Zhou, Y. Zhang, Y. Hu, L. Lin, Z.L. Wang, Pyroelectric nanogenerators for harvesting thermoelectric energy, Nano Lett. 12 (6) (2012) 2833–2838.

[22] Y. Yang, Y. Zhou, J.M. Wu, Z.L. Wang, Single micro/nanowire pyroelectric nanogenerators as self-powered temperature sensors, ACS Nano 6 (9) (2012) 8456–8461.

[23] Z.L. Wang, Self-powered nanosensors and nanosystems, Adv. Mater. 24 (2) (2012) 280–285.

[24] Z.L. Wang, J. Song, Piezoelectric nanogenerators based on zinc oxide nanowire arrays, Science 312 (5771) (2006) 242–246.

[25] X. Wang, J. Song, J. Liu, Z.L. Wang, Direct-current nanogenerator driven by ultrasonic waves, Science 316 (5821) (2007) 102–105.

[26] M.-Y. Choi, D. Choi, M.-J. Jin, I. Kim, S.-H. Kim, J.-Y. Choi, S.Y. Lee, J.M. Kim, S.- W. Kim, Mechanically powered transparent flexible charge-generating nanodevices with piezoelectric zno nanorods, Adv. Mater. 21 (21) (2009) 2185–2189.

[27] C. Chang, V.H. Tran, J. Wang, Y.-K. Fuh, L. Lin, Direct-write piezoelectric polymeric nanogenerator with high energy conversion efficiency, Nano Lett. 10 (2) (2010) 726–731.

[28] G. Zhu, C. Pan, W. Guo, C.-Y. Chen, Y. Zhou, R. Yu, Z.L. Wang, Triboelectric- generator-driven pulse electrodeposition for micropatterning, Nano Lett. 12 (9) (2012) 4960–4965.

[29] F.-R. Fan, Z.-Q. Tian, Z.L. Wang, Flexible triboelectric generator, Nano Energy 1 (2) (2012) 328–334.

[30] R. Zhang, M. Hummelgard, J. €Ortegren, M. Olsen, H. Andersson, Y. Yang, H. Olin, Human body constituted triboelectric nanogenerators as energy harvesters, code transmitters, and motion sensors, ACS Appl. Energy Mater. 1 (6) (2018) 2955–2960.

[31] Y. Yang, H. Zhang, J. Chen, S. Lee, T.-C. Hou, Z.L. Wang, Simultaneously harvesting mechanical and chemical energies by a hybrid cell for self-powered biosensors and personal electronics, Energy Environ. Sci. 6 (6) (2013) 1744–1749.

[32] Y. Yang, L. Lin, Y. Zhang, Q. Jing, T.-C. Hou, Z.L. Wang, Self-powered magnetic sensor based on a triboelectric nanogenerator, ACS Nano 6 (11) (2012) 10378–10383.

[33] S. Xu, B.J. Hansen, Z.L. Wang, Piezoelectric-nanowire-enabled power source for driving wireless microelectronics, Nat. Commun. 1 (1) (2010) 1–5.

[34] H.-P. Phan, D. Viet Dao, P. Tanner, L. Wang, N.-T. Nguyen, Y. Zhu, S. Dimitrijev, Fundamental piezoresistive coefficients of p-type single crystalline 3c-sic, Appl. Phys. Lett. 104 (11) (2014) 111905.

[35] N. Yazdi, F. Ayazi, K. Najafi, Micromachined inertial sensors, Proc. IEEE 86 (8) (1998) 1640–1659.

[36] X. Fan, F. Forsberg, A.D. Smith, S. Schr€oder, S. Wagner, M. €ostling, M.C. Lemme, F. Niklaus, Suspended graphene membranes with attached silicon proof masses as piezoresistive nanoelectromechanical systems accelerometers, Nano Lett. 19 (10) (2019) 6788–6799.

[37] R. Amarasinghe, D.V. Dao, T. Toriyama, S. Sugiyama, A silicon micromachined six- degree of freedom piezoresistive accelerometer, in: Sensors, IEEE, 2004, pp. 852–855. IEEE, 2004.

[38] L. Zhang, J. Lu, Y. Kurashima, H. Takagi, R. Maeda, Development and application of planar piezoresistive vibration sensor, Microelectron. Eng. 119 (2014) 70–74.

T. Nguyen et al.

Nano Energy 76 (2020) 104950

8

[39] T. Nguyen, T. Dinh, A.R.M. Foisal, H.-P. Phan, T.-K. Nguyen, N.-T. Nguyen, D. V. Dao, Giant piezoresistive effect by optoelectronic coupling in a heterojunction, Nat. Commun. 10 (1) (2019) 1–8.

[40] A.R.M. Foisal, T. Nguyen, T. Dinh, T.K. Nguyen, P. Tanner, E.W. Streed, D.V. Dao, 3c-sic/si heterostructure: an excellent platform for position-sensitive detectors based on photovoltaic effect, ACS Appl. Mater. Interfaces 11 (43) (2019) 40980–40987.

[41] T.V. Nguyen, T. Dinh, H.-P. Phan, T.-K. Nguyen, A.R.M. Foisal, N.-T. Nguyen, D. V. Dao, Opto-electronic coupling in semiconductors: towards ultrasensitive pressure sensing, J. Mater. Chem. 8 (14) (2020) 4713–4721.

[42] S.O. Kasap, Principles of Electronic Materials and Devices, Tata McGraw-Hill, 2006.

[43] T Dinh, T Nguyen, A.R.M Foisal, H.-P Phan, T.-K Nguyen, N.-T Nguyen, D.V Dao, Optothermotronic effect as an ultrasensitive thermal sensing technology for solid- state electronics, ScienceAdvances 6 (22) (2020) eaay2671, https://doi.org/ 10.1126/sciadv.aay2671.

Toan Dinh is currently a lecturer at School of Mechanical and Electrical Engineering, University of Southern Queensland (USQ). He is also an Adjunct Research Fellow at Queensland Micro- and Nanotechnology Centre, Griffith University. Dr Dinh received his PhD in Materials Engineering from Griffith Uni-versity (Australia) in 2017. His research interests include micro/nano-electromechanical systems (MEMS/NEMS), sensing technologies, sensors for harsh environments, and advanced materials for flexible and wearable applications.

Thanh Nguyen received the B.E degrees in Electrical Engineering and Mechanical Engineering, Hanoi University of Science and Technology, Vietnam, in 2009 and 2012, respectively, and the M.Sc degree from Chulalongkorn University, Thailand, in 2015. He is currently pursuing the Ph.D degree from the Queensland Micro-and Nanotechnology Centre, Griffith University, Australia under super-vision of Prof. Dzung Viet Dao and Prof. Nam-Trung Nguyen. His research interests focus on MEMS sensors and actuators, physics of semiconductors, flexible electronics and silicon carbide MEMS/ NEMS for applications in harsh environments.

T. Nguyen et al.

Nano Energy 76 (2020) 104950

9

Hoang-Phuong Phan (B.E, M.E. The University of Tokyo; Ph.D Griffith University) is an ARC DECRA Fellow at Griffith Uni-versity. His main research interests include Silicon/Silicon carbide MEMS/NEMS, integrated sensors, flexible electronics, and bio-sensing applications. Dr. Phan was a visiting scholar at Stanford University, 2017, and Northwestern University, 2019. He has published over 80 journal articles, two US patents, and two book/book-chapters. Dr. Phan was honoured with the MEXT scholarships, GU publication award, the GGRS-IEIS travel grant, the Springer outstanding theses award, the Australian Nanotechnology Network Overseas Fellowship, GU Postdoctoral Fellowship, ARC DECRA, and Pro Vice Chancellor Excellent Early Career Researcher.

Van Thanh Dau received the B.S. degree in aerospace engi-neering from Hochiminh City University of Technology, Viet-nam, in 2002, and the M.S. and Ph.D. degrees in micro- mechatronics from Ritsumeikan University, Japan, in 2004 and 2007, respectively. From 2007 to 2009, he was a Post-doctoral Fellow with Japan Society for the Promotion of Sci-ence (JSPS) at Micro Nano Integrated Devices Laboratory, Ritsumeikan University. From 2010 to 2018 he was a research scientist at Sumitomo Chemical Co., Japan, where he worked on integrated micro electrospray and atomization methods. Currently, he is a Lecturer at School of Engineering and Built Environment, Griffith University, Australia. His research sub-jects are microfluidics, electrofluidodynamics and micro-mechatronics. He is the author and co-author of more than 100 scientific articles and 25 inventions.

Tuan-Khoa Nguyen is current a research fellow at Griffith University, Australia. His research interests are focused on advanced micro/nano fabrications and characterizations, semiconductor-based electronics, silicon/silicon carbide MEMS/NEMS, miniaturized sensors for manufacturing in-dustries and bio applications. He completed his BEng and MSc degrees from Hanoi University of Science and Technology, Vietnam in 2009 and 2011, respectively. He was awarded Griffith University Postgraduate Research Scholarship (GUPRS) and Griffith University International Postgraduate Research Scholarship (GUIPRS). In 2018, he earned his PhD degree at Griffith University. He has authored and co-authored over 55 high-impact factor journal papers, several papers have been either featured in the covers of renowned journals or selected as Editor’s picks. Dr Nguyen also serves as a reviewer for a number of top-ranked journals.

Abbin Perunnilathil Joy received the B.E. degree in elec-tronics and communication engineering from Anna University, Chennai, India, and the MSc. and MASc. degrees in electrical engineering and mechatronic systems engineering from Na-tional University of Singapore and Simon Fraser University, in 2012, 2014, and 2019 respectively. He has worked in Global-Foundries, Singapore for two years as a process module engi-neer specialized in metrology inspection. He is currently working for 4D LABS at Simon Fraser University as a fabrication specialist. His research interests include development of micro/ nano sensors, 2D materials, and microelectronics.

Behraad Bahreyni is an Associate Professor and the founding Director of the Intelligent Sensing Laboratory (ISL) at the School of Mechatronic Systems Engineering at Simon Fraser University, BC, Canada. He received his BSc in electronics en-gineering from Sharif University of Technology, Iran, and MSc and Ph.D. degrees in electrical engineering from the University of Manitoba, Canada, in 1999, 2001, and 2006, respectively. He was a post-doctoral researcher with the NanoSicence Centre at Cambridge University, UK, where he conducted research on interface circuit design for microresonators. He joined Simon Fraser University in 2008 after a one-year tenure in the industry as a MEMS design engineer. His research activities are focused on the design and fabrication sensing systems comprising micro/nano sensors from silicon, polymers, or nanocomposites, their interface electronics, and the required signal processing algorithms. Dr. Bahreyni is the author of more than 150 tech-nical publications including a book on the fabrication and design of resonant microdevices and severn patents.

Afzaal Qamar received the M.Phil. degree in physics from Pakistan Institute of Engineering and Applied Sciences (PIEAS) in 2006, on a national competitive fellowship, and the Ph.D. degree from Queensland Micro- and Nanotechnology Centre, Griffith University, Australia, on a competitive international scholarship (GUIPRS and GUPRS). Prior to his Ph.D., Dr. Qamar served as the Manager Technical for III-V semiconductor device fabrication at the Institute of Applied Technologies from 2008 to 2014. He received best oral presentation award at ICNNE- 2016 held in Paris for presenting his research paper. Dr. Qamar is currently a postdoctoral research fellow with the Department of Electrical Engineering and Computer Science (EECS), University of Michigan, Ann Arbor, USA. His research interests are MEMS/NEMS devices, piezoelectric and piezor-esistive properties of wide bandgap (SiC, AlN, GaN) semi-conductors, SAW/BAW devices and other acoustic MEMS devices for timing, RF and sensing applications.

Mina Rais-Zadeh received the B.S. degree in electrical engi-neering from Sharif University of Technology and M.S. and Ph. D. degrees both in Electrical and Computer Engineering from Georgia Institute of Technology in 2005 and 2008, respectively. From 2008 to 2009, she was a Postdoctoral Fellow at Georgia Institute of Technology. In 2009, she joined University of Michigan as an Assistant Professor of Electrical Engineering and Computer Science (EECS) and had been a tenured Associate Professor from 2014 to 2018. She is currently leading MEMS and micro-instrument development activities at NASA JPL as a supervisor for the Advanced Optical and Electromechanical Microsystems Group.

Nam-Trung Nguyen received his Dip-Ing, Dr Ing and Dr Ing Habil degrees from Chemnitz University of Technology, Ger-many, in 1993, 1997 and 2004, respectively. Currently, he is a professor and the director of Queensland Micro and Nano-technology Centre at Griffith University. He is a Fellow of ASME and a Member of IEEE. Nguyen’s research is focused on microfluidics, nanofluidics, micro/nanomachining technolo-gies, micro/nanoscale science, and instrumentation for biomedical applications. He published over 400 journal papers and 3 granted US patents. Among the books he has written, the first, second and third editions of the bestseller “Fundamentals and Applications of Microfluidics” were published in 2002, 2006 and 2019, respectively.

T. Nguyen et al.

Nano Energy 76 (2020) 104950

10

Dzung Viet Dao received his Ph.D. degree from Ritsumeikan University, Japan in 2003. He then served as a Postdoctoral Research Fellow from 2003 to 2006, a Lecturer from 2006 to 2007, and a Chair Professor from 2007 to 2011, all at Ritsu-meikan University. From 2011 A/Prof Dao joined Griffith University, Australia, where he has been teaching in Mecha-tronics and Mechanical Engineering. His current research in-terests include advanced manufacturing, MEMS sensors & actuators, transducers for harsh environments, and mecha-tronics/robotics.

T. Nguyen et al.