An Internal-Electrostatic-Field-Boosted Self-Powered Ultraviolet Photodetector

Self-powered photodetectors are of significance for the development of low-energy-consumption and environment-friendly Internet of Things. The performance of semiconductor-based self-powered photodetectors is limited by the low quality of junctions. Here, a novel strategy was proposed for developing high-performance self-powered photodetectors with boosted electrostatic potential. The proposed self-powered ultraviolet (UV) photodetector consisted of an indium tin oxide and titanium dioxide (ITO/TiO2) heterojunction and an electret film (poly tetra fluoroethylene, PTFE). The PTFE layer introduces a built-in electrostatic field to highly enhance the photovoltaic effect, and its high internal resistance greatly reduces the dark current, and thus remarkable performances were achieved. The self-powered UV photodetector with PTFE demonstrated an extremely high on–off ratio of 2.49 × 105, a responsivity of 76.87 mA/W, a response rise time of 7.44 ms, and a decay time of 3.75 ms. Furthermore, the device exhibited exceptional stability from room temperature to 70 °C. Compared with the conventional ITO/TiO2 heterojunction without the PTFE layer, the photoresponse of the detector improved by 442-fold, and the light–dark ratio was increased by 8.40 × 105 times. In addition, the detector is simple, easy to fabricate, and low cost. Therefore, it can be used on a large scale. The electrostatic modulation effect is universal for various types of semiconductor junctions and is expected to inspire more innovative applications in optoelectronic and microelectronic devices.


Introduction
With the rapid development of the Internet of Things (IoT), ensuring a large-scale distributed power supply has become a challenge because of the numerous sensors and detectors used in IoT systems. Conventional power sources, such as batteries, have limited life and cause environmental pollution. Furthermore, it is difficult to manage, replace, and maintain power sources. The self-powered device without an external power supply is a promising solution for the development of a green and sustainable IoT [1,2], which has attracted considerable attention.
Ultraviolet (UV) photodetectors, as basic units of photoelectric information systems and IoT, are widely used in fields such as fire warning, astronomical exploration, environmental monitoring, chemical/biological sensing, and optoelectronic storage [3][4][5][6][7]. Generally, most UV photodetectors, such as UV phototubes, and semiconductor-based diodes [8,9], require external power sources for operation. For example, the working voltage of the ultraviolet phototube reached as high as one hundred volts. Although a p-n junction device can function without a power source, its dark current is high, and the detection performance is limited. Therefore, a reverse bias is typical for the p-n junction Figure 1a displays the design of the proposed IEFB-SP ultraviolet photodetector. The photodetector was a four-layer film of ITO/TiO 2 /PTFE/Cu. Among them, ITO/TiO 2 functioned as a conventional n-type heterojunction. The PTFE was an electret layer that generated static electricity on the interface and also functioned as a high-resistance layer to reduce the dark current. ITO and Cu were used as the electrodes of the detector. The working principle of the device is displayed in Figure 1b,c. On the one hand, a difference in electron affinities between ITO and TiO 2 produces a built-in electric field in the ITO/TiO 2 junction. On the other hand, the PTFE, owning high electronegativity, gains electrons from TiO 2, and a built-in electrostatic potential generates on the TiO 2 /PTFE interface. Based on the first principles calculation and molecular dynamics simulation, the intermolecular transferred charges on the TiO 2 /PTFE and the TiO 2 /PDMS interface were quantitatively analyzed (see supplementary information in note.1), and the varying contact electrification properties with intermolecular distances are shown in Figures S1 and S2, revealing an existing electrostatic field on the TiO 2 /PTFE and TiO 2 /PDMS interface. Experimentally, based on the coupling of contact electrification and electrostatic induction, a triboelectric nanogenerator (TENG) composed of TiO 2 and PTFE was fabricated, and the charge transfer properties between the TiO 2 and PTFE surfaces were investigated when they were in contact. As shown in Figures S3 and S4, the surface of PTFE is negatively charged, and the surface of TiO 2 is positively charged after contact. Without UV lights, photo-generated carriers were not developed in the TiO 2 layer, and the device outputted a small short-circuit dark current ( Figure 1d) because of a weak contact potential difference of ITO/TiO 2 heterojunction and the high impedance of the PTFE layer. With UV illumination, photo-generated carriers were produced in TiO 2 . Under the simultaneous actions of the built-in electric field of ITO/TiO 2 heterojunction and the electrostatic field of TiO 2 /PTFE interface, photo-generated carriers were accelerated to separate and diffuse to generate a boosted photovoltaic effect. The negative potential of PTFE remarkably promoted the optical response and considerably improved output performances. The equivalent circuit diagrams of the detector in light-off and light-on states are illustrated in Figure 1e,f. triboelectric nanogenerator (TENG) composed of TiO2 and PTFE was fabricated, and the charge transfer properties between the TiO2 and PTFE surfaces were investigated when they were in contact. As shown in Figures S3 and S4, the surface of PTFE is negatively charged, and the surface of TiO2 is positively charged after contact. Without UV lights, photo-generated carriers were not developed in the TiO2 layer, and the device outputted a small short-circuit dark current (Figure 1d) because of a weak contact potential difference of ITO/TiO2 heterojunction and the high impedance of the PTFE layer. With UV illumination, photo-generated carriers were produced in TiO2. Under the simultaneous actions of the built-in electric field of ITO/TiO2 heterojunction and the electrostatic field of TiO2/PTFE interface, photo-generated carriers were accelerated to separate and diffuse to generate a boosted photovoltaic effect. The negative potential of PTFE remarkably promoted the optical response and considerably improved output performances. The equivalent circuit diagrams of the detector in light-off and light-on states are illustrated in Figure 1e,f.

Material Characterizations
Figure 2a displays the scanning electron microscopy (SEM) images of the ITO/TiO2/PTFE/Cu film and the photograph of the fabricated device. The thickness of the TiO2 film was 236 nm, and that of the PTFE film was 150 nm. The optical transmission spectra of ITO with the glass substrate and the sample after the growth of the TiO2 film are displayed in Figure 2d. The cut-off absorption wavelength of ITO was 303 nm. After the growth of the TiO2 film, the cut-off absorption edge was red-shifted at 344 nm. The bandgap of the ITO film was 3.75 eV, and the bandgap of the TiO2 film was 3.44 eV, which was calculated by the optical absorption spectra. Figure 2e displays the X-ray diffraction pattern of the TiO2 film. Taking the XRD standard card of PDF #89-4920 and PDF #21-1272 as a reference, the prepared TiO2 film was a mixture of anatase and rutile. Figure 2b,c display the surface SEM images of the TiO2 and PTFE films. The atomic force microscopy (AFM) images of the TiO2 surface are displayed in Figure 2f. The root mean square roughness of the TiO2 surface, average roughness, and maximum roughness were 2.26, 1.81, and 15.5 nm, respectively. Certain surface roughness of TiO2 was conducive to the generation of the electrostatic field in the interface between TiO2 and PTFE films.  Figure 2a displays the scanning electron microscopy (SEM) images of the ITO/TiO 2 / PTFE/Cu film and the photograph of the fabricated device. The thickness of the TiO 2 film was 236 nm, and that of the PTFE film was 150 nm. The optical transmission spectra of ITO with the glass substrate and the sample after the growth of the TiO 2 film are displayed in Figure 2d. The cut-off absorption wavelength of ITO was 303 nm. After the growth of the TiO 2 film, the cut-off absorption edge was red-shifted at 344 nm. The bandgap of the ITO film was 3.75 eV, and the bandgap of the TiO 2 film was 3.44 eV, which was calculated by the optical absorption spectra. Figure 2e displays the X-ray diffraction pattern of the TiO 2 film. Taking the XRD standard card of PDF #89-4920 and PDF #21-1272 as a reference, the prepared TiO 2 film was a mixture of anatase and rutile. Figure 2b,c display the surface SEM images of the TiO 2 and PTFE films. The atomic force microscopy (AFM) images of the TiO 2 surface are displayed in Figure 2f. The root mean square roughness of the TiO 2 surface, average roughness, and maximum roughness were 2.26, 1.81, and 15.5 nm, respectively. Certain surface roughness of TiO 2 was conducive to the generation of the electrostatic field in the interface between TiO 2 and PTFE films.

Photodetection Performance of IEFB-SP Device
The ultraviolet detection performance of the IEFB-SP device at 0 V is displayed in Figure 3. A UV light-emitting diode (LED) with a wavelength of 365 nm fixed on a linear motor periodically illuminated the detector by linear motion, producing turn-on and turnoff states (see video 1). In the measurements, the humidity of the air environment was between 30 RH and 40 RH, and the UV LED moved to illuminate the device and stayed for 3 s and then moved away at a speed of 1 m/s. The proposed device exhibited excellent self-powered photodetection performances, as displayed in Figure 3. Without UV light, the dark current of the device was extremely low, 8.03 × 10 −12 A. With UV light, due to the high resistance of the PTFE layer, the dark current (Id) of the IEFB-SP device was extremely low, 8.03 × 10 −12 A (see Figure 3a). With UV irradiation, the photocurrent (Ip) was approximately 2.00 × 10 −6 A, and the light-to-dark current ratio reached 2.49 × 10 5 . The currentvoltage (I-V) curves ( Figure S8) show an obvious increase in the magnitude of photocurrent at zero bias in the light state compared with the dark state. As illustrated in Figure  2b. the photovoltage (Vp) and dark voltage (Vd) of the IEFB-SP device were 0.02 V and 8.52 × 10 −5 V, respectively. To figure out the functions of the PTFE layer, we measured the internal resistances for photodetectors using the impedance matching method and the photodetection performance of a conventional ITO/TiO2/Cu heterojunction without a PTFE

Photodetection Performance of IEFB-SP Device
The ultraviolet detection performance of the IEFB-SP device at 0 V is displayed in Figure 3. A UV light-emitting diode (LED) with a wavelength of 365 nm fixed on a linear motor periodically illuminated the detector by linear motion, producing turn-on and turnoff states (see Video S1). In the measurements, the humidity of the air environment was between 30 RH and 40 RH, and the UV LED moved to illuminate the device and stayed for 3 s and then moved away at a speed of 1 m/s. The proposed device exhibited excellent self-powered photodetection performances, as displayed in Figure 3. Without UV light, the dark current of the device was extremely low, 8.03 × 10 −12 A. With UV light, due to the high resistance of the PTFE layer, the dark current (I d ) of the IEFB-SP device was extremely low, 8.03 × 10 −12 A (see Figure 3a). With UV irradiation, the photocurrent (I p ) was approximately 2.00 × 10 −6 A, and the light-to-dark current ratio reached 2.49 × 10 5 . The current-voltage (I-V) curves ( Figure S8) show an obvious increase in the magnitude of photocurrent at zero bias in the light state compared with the dark state. As illustrated in Figure 2b. the photovoltage (V p ) and dark voltage (V d ) of the IEFB-SP device were 0.02 V and 8.52 × 10 −5 V, respectively. To figure out the functions of the PTFE layer, we measured the internal resistances for photodetectors using the impedance matching method and the photodetection performance of a conventional ITO/TiO 2 /Cu heterojunction without a PTFE layer for comparison, as displayed in the supplementary information (see Note S3, and Figures S6 and S7), and the performance comparisons are listed in Table S1. We can see that, without PTFE, the average photocurrent (I p ) and dark current (I d ) of the ITO/TiO 2 /Cu device at 0 V are 4.76 × 10 −6 A and 3.67 × 10 −6 A, respectively, and the on-off ratio is only 1.30. Connecting the TiO 2 and Cu, the I-V curve (see Figure S7c) reveals there is an excellent ohmic contact between the Cu electrode and the TiO 2 , and the main contact potential difference was generated by the ITO/TiO 2 interface. In contrast, the light-dark ratio of the device with PTFE was increased by 8.40 × 10 5 times. The photovoltage increment (V p -V d ) is increased by 665 times, as presented in Note S4. It is interesting that the internal resistance of the device with PTFE increased by~1.3 × 10 5 times, but the photocurrent only decreased by 2.4 times. This is because the strong built-in electrostatic field significantly improved the photocurrent. Therefore, the PTFE has double functions. One is the high internal resistance of PTFE, which greatly reduces the dark current, and the other is the built-in electrostatic field, which remarkably enhances the photovoltaic effect. Thus, the IEFB-SP device achieved high performance of low dark current and large photocurrent.  Figures S6 and S7), and the performance comparisons are listed in Table S1. We can see that, without PTFE, the average photocurrent (Ip) and dark current (Id) of the ITO/TiO2/Cu device at 0 V are 4.76 × 10 −6 A and 3.67 × 10 −6 A, respectively, and the on-off ratio is only 1.30. Connecting the TiO2 and Cu, the I-V curve (see Figure S7c) reveals there is an excellent ohmic contact between the Cu electrode and the TiO2, and the main contact potential difference was generated by the ITO/TiO2 interface. In contrast, the light-dark ratio of the device with PTFE was increased by 8.40 × 10 5 times. The photovoltage increment (Vp-Vd) is increased by 665 times, as presented in Note S4. It is interesting that the internal resistance of the device with PTFE increased by ~ 1.3 × 10 5 times, but the photocurrent only decreased by 2.4 times. This is because the strong built-in electrostatic field significantly improved the photocurrent. Therefore, the PTFE has double functions. One is the high internal resistance of PTFE, which greatly reduces the dark current, and the other is the built-in electrostatic field, which remarkably enhances the photovoltaic effect. Thus, the IEFB-SP device achieved high performance of low dark current and large photocurrent.  To further quantify the photocurrent properties enhanced by the PTFE, the transferred charge was measured, as displayed in Figure 3c. As the UV light periodically irradiated on the device, the transferred charge increased stepwise with the irradiation times, and the amount of transferred charge for each illumination was 5.97 µC. The current calculated by I = q/t was 1.99 × 10 −6 A, which is consistent with the detection current. As for the device without PTFE, the amount of transferred charge in a light switching cycle is 14.08 µC, as shown in Figure S9. When we reversed the positive and negative connections between the ITO and Cu electrodes, the current value remained unchanged, but the current direction was opposed (see Figure S10), indicating that the measured signals were produced by the absorbed lights. Figure 3d displays the photocurrent dependence on the UV intensity in the intensity range of 0.29-15.94 mW/cm 2 . The photocurrent increased gradually with the UV power density. With a low optical power density of 0.29 mW/cm 2 , the device still exhibited a superior response. The responsivity R and the detectivity D* of the self-powered photodetector were evaluated using the following expression [37,38]: where I P is the photocurrent, I D is the dark current of the device, P is the incident optical power, A is the effective detection area, and e is the amount of electronic charge. At 365-nm wavelength, the incident light power was 0.64 mW/cm 2 , and the beam size was 1 mm 2 . The responsibility and the detectivity of the IEFB-SP photodetector were 72.41 mA/W and 4.51 × 10 12 jones, respectively. The photocurrent and responsivity as a function of illumination intensity are displayed in Figure 3e. The photocurrent followed the power law [39,40] of I P ∝ P β , where the β value is 1.03, indicating the photocurrent was almost linearly increased with the incident optical power. The responsivity exhibited excellent stability with the increase in the optical power. Figure 3f displays the responsivity of the IEFB-SP device at various light wavelengths. As the optical wavelengths were 325, 365, 396, 457, 532, and 607 nm, the responsivity of the device was 19.10, 72.24, 28.93, 14.14, 4.20, and 2.35 mA/W, respectively. Without regard to the absorption of the ITO layer in the UV region, the device exhibits a good UV/visible rejection ratio. A photodetector testing system was used to reveal the spectral response characteristics of the device (ITO/TiO 2 /PTFE/Cu) and the conventional heterojunction (ITO/TiO 2 /Cu) in the wavelength range of 250-700 nm (see Figure S11) in the supporting information. At the UV wavelength of 360 nm, the responsivity of our device reached as high as 76.874 mA/W, while that of the device without PTFE was only 0.174 mA/W. The responsivity increased by 442 times.
The external quantum efficiency (EQE) of the photodetector is another key parameter for evaluating the photodetection performance and is expressed as follows [41,42]: where h represents the Planck constant, c is the velocity of the incident light, e is the charge, and λ is the wavelength of the incident light. The EQE was 24.54% for the proposed device at a wavelength of 365 nm. Under the illumination of a xenon lamp of a photodetector testing system, the EQEs of our detector (ITO/TiO 2 /PTFE/Cu) and the conventional heterojunction (ITO/TiO 2 /Cu) in the wavelength range of 250-700 nm are displayed in Figure S12. As the ultraviolet wavelength is 360 nm, the EQE of our device reached as high as 26.48%, whereas that of the conventional heterojunction device without PTFE was 0.06%. We evaluated the response speed of the IEFB-SP device. The rise time (T R90 ) is defined as the time required by the photovoltage to reach 90% of the maximum photovoltage, and the decay time (T D10 ) is the time at which the photovoltage decreases to 10% of the maximum photovoltage. The rise and decay times of this photodetector were 7.44 and 3.75 ms, respectively (see Figure 4a,b), and it had a fast response. The rise and fall times of the conventional ITO/TiO 2 /Cu heterojunction were 0.83 s and 1 s, respectively, as displayed in Figure S13, which is slower by 111 and 261 times, respectively, than those of the device with PTFE. Nanomaterials 2022, 12, x FOR PEER REVIEW 8 of 13 maximum photovoltage. The rise and decay times of this photodetector were 7.44 and 3.75 ms, respectively (see Figure 4a,b), and it had a fast response. The rise and fall times of the conventional ITO/TiO2/Cu heterojunction were 0.83 s and 1 s, respectively, as displayed in Figure S13, which is slower by 111 and 261 times, respectively, than those of the device with PTFE. Furthermore, we assessed the temperature stability and environmental stability of the IEFB-SP device. Figures 4c and S14 exhibited the photocurrent and dark current characteristics at various temperatures from room temperature to 70 °C, respectively. With the increase in the temperature, the photocurrent and dark current both increased slightly, but the light-to-dark ratio remained unchanged. Compared with the conventional photodetector of the semiconductor heterojunction, it exhibited exceptionally superior temperature stability. Furthermore, we irradiated the device continuously for 2 h and tested it repeatedly for 96 circles, and its photocurrent was invariable, as displayed in Figure S15. Furthermore, we assessed the temperature stability and environmental stability of the IEFB-SP device. Figures 4c and S14 exhibited the photocurrent and dark current characteristics at various temperatures from room temperature to 70 • C, respectively. With the increase in the temperature, the photocurrent and dark current both increased slightly, but the light-to-dark ratio remained unchanged. Compared with the conventional photodetector of the semiconductor heterojunction, it exhibited exceptionally superior temperature stability. Furthermore, we irradiated the device continuously for 2 h and tested it repeatedly for 96 circles, and its photocurrent was invariable, as displayed in Figure S15. After placing in an air environment for 60 days, the IEFB-SP photodetector with PTFE maintained its excellent photodetection performance (see Figure 4d).

Influence of Dielectric Materials
We studied the influence of dielectric layer thickness and material on the photodetection performances of IEFB-SP devices. Figure 4e displays the photocurrent properties of the device with PTFE films of thicknesses 50, 150, 300, and 800 nm. The thinner the thickness of PTFE was, the larger photocurrent and dark current were. Thicker PTFE films not only increased the internal resistance but also reduced the effect of electrostatic induction, which resulted in a considerable decrease in the photocurrent. As the thickness increased, the voltage increased ( Figure S16). Among the four thicknesses of PTFE membranes, the detector with a thickness of 150 nm presented the best photodetection performance. Figure 4f displays a comparison of the photocurrent properties with dielectric layer materials PTFE, polydimethylsiloxane (PDMS), and silicone rubber. Under similar UV illumination, the photocurrents of the IEFB-SP devices with PDMS, PTFE, and silicone rubber were 118.30 nA, 1.97 µA, and 0.04 nA, respectively. The response speeds of the devices with PDMS or with silicone rubber were slower than the self-powered photodetector with PTFE. Since the built-in electrostatic field is derived from the charge transfer between TiO 2 and the dielectric materials, the ability to gain electrons from TiO 2 plays a key role in the photodetection performance of the IEFB-SP photodetector. It is closely dependent on the electronegativity of materials. The better the electronegativity of a material is, the higher its ability to gain electrons from other material is, and the stronger the electrostatic field it yields. A stronger electrostatic field can be generated in the interface when a material of high electronegativity is in contact with a material of high electropositivity. In the triboelectric series [43,44], among these three materials, PTFE exhibited the strongest capability for gaining electrons and the highest electronegativity [45]. Therefore, the device with PTFE has the best photodetection performance.

Performance Comparison
Finally, Table 1 lists a comparison of state-of-the-art photodetectors and our proposed device. Our IEFB-SP photodetector exhibited excellent comprehensive detection performance, including high optical responsivity and sensitivity, especially with large photocurrent and ultra-low dark current. It is worth noting that we do not need a high-quality semiconductor junction for to achieve high-performance photodetection. Moreover, universal devices can be devised for other types of junctions to improve function. We provide a novel strategy for the development of high-performance, easy-to-use, and low-cost photodetectors.

Conclusions
In summary, the charge transfer characteristics between the TiO 2 and the dielectric interfaces are quantitatively analyzed both from theoretical simulations and experiments. Furthermore, an electret of PTFE was inducted to remarkably boost the photovoltaic effect of a semiconductor heterojunction to develop self-powered UV photodectors. The proposed IEFB device achieved outstanding photodetection performances without a power source. Under the illumination of a xenon lamp, the optical responsivity of the IEFB device with PTFE was 76.87 mA/W, the specific detection rate was 4.79 × 10 12 jones, and the EQE reached 26.48% at a wavelength of 360 nm. It also exhibited a rapid rise and decay time of 7.44 and 3.75 ms. Compared with a conventional heterojunction device under the same experimental conditions, the IEFB-SP photodetector demonstrated its photoresponse, lightdark ratio, rise speed, and decayed time improved by 442, 8.40 × 10 5 , 111, 267 times those of traditional device. Furthermore, the IEFB-SP device exhibited excellent temperature and circumstance stability. The photodetection performance remained unchanged from room temperature to 70 • C and did not degrade after 60 days of exposure under air surroundings without packaging. Overall, the IEFB-SP device achieves excellent photodetection performance without the requirement of high-quality junctions. It has the advantages of stable high-performance, simplicity, ease-of-fabrication, and low-cost. The strategy of using a built-in electrostatic field on the semiconductor-electret interface to modulate the photovoltaic effect is universal for various types of semiconductor junction devices. It is of significant guidance to improve the performance of self-powered photodetectors. The results in this work deepened the understanding of the electrostatic effect of the dielectric interface in the micro-region and are expected to inspire more applications in the future.
Supplementary Materials: The following supporting information can be downloaded at: https: //www.mdpi.com/article/10.3390/nano12183200/s1, Supplementary Note S1: Quantitative analyse of the intermolecular transferred charges on TiO 2 /PTFE and TiO 2 /PDMS interfaces; Supplementary Note S2: Investigation of charge transfer properties in TiO 2 /PTFE and TiO 2 /PDMS interfaces through triboelectric nanogenerators (TENGs) [57]; Supplementary Note S3: The measurements of internal resistances for photodetectors through impedance matching method; Supplementary Note S4: Calculation of photovoltage increment for the device with PTFE; Figure S1: Results of molecular dynamics simulation and charge properties of TiO 2 and PTFE molecules; Figure S2: Results of molecular dynamics simulation and charge properties of TiO 2 and PDMS molecules; Figure S3: Working mechanism of TENG composed of PTFE and TiO 2 films; Figure S4: Output characteristic curves of triboelectric nanogenerators composed of ITO/TiO 2 and Cu/PTFE and Cu/PDMS, respectively; Figure S5: Tauc's plots analysis image of ITO and TiO 2 . Figure S6: The output power of the internal electrostatic field boosted device with different dielectric layer under different loads; Figure S7: Photodetection performances of conventional ITO/TiO 2 /Cu junction device without PTFE; Figure S8: Current-voltage (I-V) curves for the ESPB-SP photodetector under 365 nm light illumination with intensity of 15.94 mW/cm 2 and dark condition; Figure S9: The amount of transferred charges in each illumination for the photodetector without PTFE; Figure S10: Performance of ESPB-SP device with reversed the positive and negative connections between ITO and Cu electrodes; Figure S11: Comparison of the responsivity of devices with and without a PTFE layer; Figure S12: Comparison of the EQE of devices with and without a PTFE layer; Figure S13: Response speed of conventional device without a PTFE layer; Figure S14: Dark current of the ESPB-SP device from room temperature to 70 • C; Figure S15: Repeatability measurement for the ESPB-SP device with PTFE for 2 h; Figure S16: Photovoltage of the device with different PTFE thicknesses (50, 150, 300, and 800 nm); Table S1: Performance comparison of three PDs (irradiated by a UV light with the wavelength of 365 nm and the optical power density of 15.94 mW/cm 2 ). Ip/Id: the ratio of photocurrent to dark current; Vp/Vd: the ratio of photovoltage to dark voltage.; Video S1: Selfpowered detection performance of An Internal-Electrostatic-Field-Boosted Self-Powered Ultraviolet Photodetector.

Conflicts of Interest:
The authors declare no conflict of interest.