A dual-capacitor phototransistor and its fabrication method

By designing a dual-capacitor phototransistor and utilizing the charge coupling effect, a response to static or dynamic light signals is achieved, solving the problems of complex pixel structure and high power consumption in dynamic vision sensors, and providing a low-power photoelectric detection and dynamic vision sensing solution.

CN117913169BActive Publication Date: 2026-11-17INST OF METAL RESEARCH - CHINESE ACAD OF SCI
View PDF 2 Cites 0 Cited by

Patent Information

Application Number
CN202311731761.X
Authority / Receiving Office
CN · China
Patent Type
Patents(China)
Current Assignee / Owner
Filing Date
2023-12-15
Publication Date
2026-11-17
Estimated Expiration
2043-12-15

AI Technical Summary

Technical Problem

Existing dynamic vision sensors have complex pixel structures, resulting in high power consumption and making it difficult to realize low-power dynamic vision sensors.

Method used

Design a dual-capacitor phototransistor comprising a photosensitive capacitor and an electrostatic modulation capacitor connected in series. By adjusting the thickness of the gate insulating layer, it can respond to static or dynamic optical signals and realize photoelectric detection or dynamic visual sensing functions by utilizing the charge coupling effect.

Benefits of technology

The dual-capacitor phototransistor has a photosensitive sensitivity exceeding 10⁷ and a response speed close to 10 μs in conventional photodetector mode; in dynamic vision sensor mode, it can detect dynamic optical signals in the 405–1065 nm band and the 1–1000 Hz band, with a static power consumption close to 40 pW.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN117913169B_ABST
    Figure CN117913169B_ABST
Patent Text Reader

Abstract

This invention relates to the research and application of photodetectors made of nanomaterial semiconductors, specifically a dual-capacitor phototransistor and its fabrication method. The dual-capacitor phototransistor consists of a photosensitive capacitor and an electrostatic modulation capacitor connected in series. The photosensitive capacitor, from bottom to top, comprises a photosensitive semiconductor, an oxide layer, and a metal disk; the electrostatic modulation capacitor, from bottom to top, comprises a gate, a gate insulating layer, and a channel layer. The metal disk on top of the photosensitive capacitor is connected to the gate on the bottom of the electrostatic modulation capacitor via a metal wire. This invention utilizes the charge-coupled effect, where photogenerated carriers generated by the photosensitive semiconductor in the photosensitive capacitor induce the same type of carriers in the gate of the electrostatic modulation capacitor, altering the conductivity of the channel layer. Depending on the thickness of the gate insulating layer in the electrostatic modulation capacitor, the device can detect static or dynamic optical signals.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] This invention relates to the research and application of photodetectors made of nano-semiconductor materials, specifically to a dual-capacitor phototransistor and its fabrication method. Background Technology

[0002] Vision is a fundamental function of intelligent beings and agents, used to perceive the external world. Designing powerful visual perception systems for application in computer vision and artificial intelligence technologies has always been a research hotspot in the scientific and industrial communities. Traditional dynamic visual sensing systems based on the von Neumann architecture consist of a photosensitive unit, an image processing unit, and a data storage unit [References 1-3]. Due to the physical separation of these three components, data transmission delays exist. Furthermore, traditional image sensors acquire images in a series of "frames," which are unrelated to dynamic information in the real world, leading to the potential loss of important information between "frames." Simultaneously, each "frame" originates from all pixels, resulting in a massive amount of data, generating a large amount of redundant data, requiring significant storage space, and causing substantial power consumption.

[0003] Dynamic vision sensors are electronic devices that respond only to changes in light intensity [References 4-5]. By abstracting the functions of the three-layered structure of photoreceptor cells, bipolar cells, and ganglion cells surrounding the biological retina, dynamic vision sensor pixels consist of a photoelectric conversion circuit, a dynamic detection circuit, and a comparator output circuit [Reference 6]. Compared to traditional cameras, dynamic vision sensors possess high temporal resolution, high dynamic range, and high pixel bandwidth, making them a promising candidate for applications in machine vision, including autonomous driving, drones, robotics, and industrial inspection, particularly in scenarios involving high-speed photography, high-speed motion, and complex lighting.

[0004] Although dynamic vision sensors have optimized their architecture compared to traditional cameras, their pixel composition remains very complex, resulting in high power consumption [References 6-15]. For example, Brandli et al. reported a 240×180 pixel dynamic vision sensor with a dynamic range of up to 130dB; however, the pixel unit of the device consisted of 47 transistors, 3 capacitors, and 1 photodiode, resulting in a power consumption of up to 115nW per pixel [Reference 7]. Posch et al. reported a more integrated dynamic vision sensor, but its pixel unit contained 77 transistors, 3 capacitors, and 2 photodiodes, resulting in a static power consumption of up to 685nW [Reference 8]. Overall, developing dynamic vision sensor units with simple pixel structures to achieve lower power consumption is one of the main challenges in this field.

[0005] 1.Roy,K.,Jaiswal,A.&Panda,P.Towards spike-basedmachine intelligencewith neuromorphic computing.Nature 575,607-617(2019).

[0006] 2.Zhou,F.&Chai,Y.Near-sensor and in-sensor computing.Nat.Electron.3,664-671(2020).

[0007] 3.Li,X.et al.Power-eficient neural network with artificialdendrites.Nat.Nanotechnol.15,776-782(2020).

[0008] 4.Posch,C.,Serrano-Gotarredona,T.,Linares-Barranco,B.&Delbruck,T.Retinomorphic event-basedvision sensors:bioinspired cameras with spikingoutput.Proc.IEEE Inst.Electr.Electron.Eng.102,1470-1484(2014).

[0009] 5.Gallego,G.et al.Event-basedvision:asurvey.IEEE Trans.PatternAnal.Mach.Intell.44,154-180(2022).

[0010] 6.Lichtsteiner,P.,Posch,C.&Delbruck,T.A 128×128 120 dB 15μs latencyasynchronous temporal contrastvisionsensor.IEEE J.Solid-State Circuit.43,566-576(2008).

[0011] 7.Brandli,C.,Berner,R.,Yang,M.,Liu,S.-C.&Delbruck,T.A 240×180 130 dB3μs latency global shutter spatiotemporal vision sensor.IEEE J.Solid-StateCircuit.49,2333-2341(2014).

[0012] 8.Posch,C.,Matolin,D.&Wohlgenannt,R.A QVGA 143 dB dynamic rangeframe-free PWM image sensor with lossless pixel-level video compression andtime-domain CDS.IEEE J.Solid-State Circuit.46,259-275(2011).

[0013] 9.Finateu,T.et al.A 1280×720 back-illuminated stacked temporalcontrast event-based vision sensor with 4.86μmpixels,1.066 GEPS readout,programmable event-rate controller and compressive data-formattingpipeline.IEEE Int.Solid-State Circ.Conf.112-114(2020).

[0014] 10.Son,B.et al.4.1 A 640×480 dynamic vision sensor with a9μm pixeland 300Meps address-event representation.In Proc.2017 IEEE InternationalSolid-State Circuits Conference(ISSCC)66-67(IEEE,2017).

[0015] 11.Suh,Y.et al.A 1280×960 dynamic vision sensor with a 4.95-μm pixelpitch and motion artifact minimization.In Proc.2020 IEEE InternationalSymposium on Circuits and Systems(ISCAS)1-5(IEEE,2020).

[0016] 12.Chen,S.&Guo,M.Live demonstration:CeleX-V:A 1M pixel multi-modeevent-based sensor.2019 IEEE / CVF Conference on Computer Vision and PatternRecognition Workshops(CVPRW)1682-1683(IEEE,2019).

[0017] 13.Insightness event-based sensor modules.http: / / www.insightness.com / technology / ,(2020).

[0018] 14.Lenero-Bardallo,J.A.,Serrano-Gotarredona,T.&Linares-Barranco,B.A3.6μs latency asynchronous frame-free event-driven dynamic-vision-sensor.IEEEJ.Solid-State Circuit.46,1443-1455(2011).

[0019] 15.Serrano-Gotarredona,T.&Linares-Barranco,B.A 128×128 1.5%contrastsensitivity 0.9% FPN 3μs latency 4mW asynchronous frame-free dynamic visionsensor using transimpedance preamplifiers.IEEE J.Solid-State Circuit.48,827-838(2013). Summary of the Invention:

[0020] The purpose of this invention is to provide a dual-capacitor phototransistor and its fabrication method. This device has a photosensitive capacitor and an electrostatic modulation capacitor connected in series. When the gate insulating layer in the electrostatic modulation capacitor is thick, the device can detect static optical signals; when the gate insulating layer is thin, the device will only respond to dynamic optical signals.

[0021] The technical solution of this invention:

[0022] A dual-capacitor phototransistor, comprising a photosensitive capacitor and an electrostatic modulation capacitor connected in series, with the specific structure as follows:

[0023] The photosensitive capacitor is composed of a photosensitive semiconductor, an oxide layer, and a metal disk stacked from bottom to top. The electrostatic modulation capacitor is composed of a gate, a gate insulating layer, and a channel layer stacked from bottom to top. The metal disk on the upper layer of the photosensitive capacitor is connected to the gate on the lower layer of the electrostatic modulation capacitor through a metal wire.

[0024] The dual-capacitor phototransistor described herein has a photosensitive semiconductor in the bottom layer of the photosensitive capacitor, which includes silicon (Si), gallium nitride (GaN), or germanium (Ge); an oxide in the middle layer of the photosensitive capacitor, which includes silicon dioxide (SiO2), aluminum oxide (Al2O3), or hafnium dioxide (HfO2); and a metal disk in the upper layer of the photosensitive capacitor, which includes titanium (Ti), gold (Au), palladium (Pd), aluminum (Al), silver (Ag), chromium (Cr), or alloys thereof.

[0025] The dual-capacitor phototransistor described above has a gate in the electrostatic modulation capacitor that is either a metal or a half-metal. The metals include Ti, Au, Pd, Al, Ag, Cr, or their alloys, while the half-metals include monolayer graphene or few-layer graphene, with the few-layer graphene having 2 to 30 layers. The gate insulating layer in the electrostatic modulation capacitor is hexagonal boron nitride (h-BN), with a thickness ranging from 1 to 20 nm. The channel layer in the electrostatic modulation capacitor can be either a semiconductor or a half-metal. The semiconductor is a thin layer of transition metal sulfide (TMDCs), and the half-metal is graphene, with a channel layer thickness ranging from 1 to 20 nm.

[0026] The dual-capacitor phototransistor described herein uses metal wires, including Ti, Au, Pd, Al, Ag, Cr, or their alloys, to connect the metal disk in the photosensitive capacitor to the gate in the electrostatic modulation capacitor.

[0027] The method for fabricating the dual-capacitor phototransistor includes the following steps:

[0028] (1) An oxide is deposited on the surface of a photosensitive semiconductor by atomic layer deposition (AlD) or a substrate with an oxide / photosensitive semiconductor structure is formed by thermal oxidation.

[0029] (2) Metal gates are deposited on oxide / photosensitive semiconductor substrates by electron beam exposure (EBL), electron beam evaporation (EBV) and lift-off processes, or single-layer graphene or few-layer graphene is transferred to the substrate as a graphene gate by micromechanical lift-off.

[0030] (3) Single-layer graphene or few-layer graphene, TMDCs, and h-BN thin layers are peeled off from their bulk bodies by micromechanical peeling onto a Si substrate with a SiO2 insulating layer with a thickness of 50 to 1000 nm.

[0031] (4) Using a heterojunction transfer platform, the h-BN encapsulation layer is lifted with propylene carbonate (PPC), and then the channel layer and gate insulating layer are lifted in sequence to form a heterojunction. Then, the temperature is raised and it is deposited on an oxide / photosensitive semiconductor substrate with a gate structure.

[0032] (5) Place the stacked heterojunction into a vacuum annealing furnace for annealing treatment;

[0033] (6) The source and drain electrodes connected to the channel, the metal disk in the photosensitive capacitor, and the metal line connecting the metal disk to the gate are constructed by EBL, reactive ion etching (RIE), EBV and stripping processes.

[0034] In the fabrication method of the dual-capacitor phototransistor, in step (1), the thickness of the photosensitive semiconductor ranges from 50 to 1000 μm, and its doping type is intrinsic, p-type or n-type doping. The thickness of the oxide prepared on the surface of the photosensitive semiconductor using the ALD method or thermal oxidation method ranges from 50 to 1000 nm.

[0035] In the fabrication method of the dual-capacitor phototransistor, in step (2), the thickness of the vapor-deposited metal gate ranges from 2 to 60 nm, and the thickness of the transferred graphene gate ranges from 0.3 to 10 nm.

[0036] In the fabrication method of the dual-capacitor phototransistor, in step (4), the encapsulation layer h-BN, the channel layer and the gate insulating layer h-BN are stacked in sequence to form a sandwich structure, and the channel layer is completely covered by the encapsulation layer h-BN and the gate insulating layer h-BN; the heterojunction drop temperature is 90 to 140°C, and the gate needs to cover the channel in the vertical direction.

[0037] In the fabrication method of the dual-capacitor phototransistor, in step (5), the PPC residue on the surface of the heterojunction is removed by annealing, the vacuum annealing temperature is controlled at 350-450℃, the heating time is 0.5-2h, the holding time is 0.5-2h, and the furnace is cooled to room temperature.

[0038] In the fabrication method of the dual-capacitor phototransistor, in step (6), the lead electrode is a composite of Ti layer and Au layer. First, a Ti layer with a thickness range of 4-6 nm is deposited on the substrate and graphene contact electrode, and then an Au layer with a thickness range of 50-60 nm is deposited. The metal disk and metal wire in the photosensitive capacitor are a composite of Ti layer and Au layer or Au layer, with the Ti layer having a thickness range of 1-6 nm and the Au layer having a thickness range of 1-60 nm.

[0039] The design concept of this invention:

[0040] Dynamic vision sensors are electronic devices that respond only to changes in light intensity, and they have a huge market potential in scenarios involving high-speed photography, high-speed motion, and complex lighting. However, the pixels of dynamic vision sensors are generally composed of circuits consisting of dozens of components, leading to complex circuit design and high power consumption. This invention proposes a dual-capacitor phototransistor, which includes a photosensitive capacitor and an electrostatic modulation capacitor in series. Utilizing the charge coupling effect between the two capacitors, the photogenerated carriers generated by the photosensitive semiconductor in the photosensitive capacitor will induce the same type of carriers in the gate of the electrostatic modulation capacitor, changing the conductivity of the channel layer. Depending on the thickness of the gate insulating layer in the electrostatic modulation capacitor, the device can respond to static or dynamic light signals, thus enabling the device to possess the functions of a traditional photodetector or an emerging dynamic vision sensor, and greatly improving the problems of complex unit structure and high power consumption in existing dynamic vision sensors.

[0041] The advantages and beneficial effects of this invention are:

[0042] 1. The novel optoelectronic device dual-capacitor phototransistor design structure proposed in this invention, by adjusting the thickness of the gate insulating layer, allows the device to function as a traditional photodetector or an emerging dynamic visual sensor pixel unit, laying the foundation for the ultimate realization of low-power optoelectronic systems in the future.

[0043] 2. The dual-capacitor phototransistor obtained by this invention exhibits excellent photoelectric performance: when the device is used as a conventional photodetector, its photosensitivity exceeds 10. 7 The device's response speed is close to 10μs. When used as a dynamic vision sensor, it can detect dynamic optical signals in the 405–1065nm wavelength band and the 1–1000Hz frequency band, with a static power consumption of close to 40pW. Attached Figure Description

[0044] Figure 1 This is a schematic diagram of a dual-capacitor phototransistor. The core structure consists of a photosensitive capacitor and an electrostatic modulation capacitor connected in series. The photosensitive capacitors, from bottom to top, are p... +-Si, SiO2, and Ti-Au metal disks; the electrostatic modulation capacitor, from bottom to top, consists of a graphene gate, an h-BN insulating layer, and a MoS2 channel; the graphene gate of the lower layer of the electrostatic modulation capacitor is connected to the metal disk of the upper layer of the photosensitive capacitor via metal lines. Graphene is used as the source and drain contact electrodes of the transistor and connected to the channel, with h-BN on the channel serving as the encapsulation layer. The inset shows the equivalent circuit diagram of the device.

[0045] Figure 2 This is a process flow diagram for the fabrication of a dual-capacitor phototransistor. (a) Material preparation; (b) Heterojunction preparation; (c) Source and drain lead electrode preparation; (d) Metal disk and gate-to-metal disk connection preparation.

[0046] Figure 3 Optical photographs of a dual-capacitor phototransistor as a conventional photodetector. (a) Overall optical photograph of the device, scale bar 300 μm, with the metal disk measuring 100 × 100 μm; (b) Magnified partial photograph of the device, scale bar 10 μm; (c) Atomic force microscopy (AFM) photograph of the gate insulating layer h-BN, scale bar 1 μm; (d) Characterization of the gate insulating layer thickness, 12 nm.

[0047] Figure 4 This diagram illustrates the principle of a dual-capacitor phototransistor as a traditional photodetector and its photoelectric performance under 1065nm illumination. (a) Band structure diagram of the device under illumination, E C E V E F (a) hv represents the conduction band bottom, valence band top, Fermi level and photon, respectively; circle +, h and e represent positive charge, photogenerated hole and photogenerated electron, respectively; (b) Transfer characteristic curves of the device under darkness and different light intensities, with source-drain voltage of 0.5V; (c) Photosensitivity of the device as a function of incident light power, defined as the ratio of dark current to current under illumination; (d) Characterization of the device's response speed.

[0048] Figure 5 Optical photographs of a dual-capacitor phototransistor used as a dynamic vision sensor. (a) Overall optical photograph of the device, scale bar 300 μm, with the metal disk measuring 100 × 100 μm; (b) Magnified partial photograph of the device, scale bar 10 μm; (c) Atomic force microscopy (AFM) photograph of the gate insulating layer h-BN, scale bar 1 μm; (d) Thickness characterization of the gate insulating layer, with a thickness of 4 nm.

[0049] Figure 6This diagram illustrates the principle and photoelectric performance of a dual-capacitor phototransistor as a dynamic vision sensor. (a) Band structure diagram when dynamic light is on; (b) Band structure diagram when dynamic light is off; (c) Current-time curves of the device under static and dynamic illumination, with an incident light wavelength of 516 nm, a source-drain voltage of 10 mV, and a gate voltage of 0 mV; (d) Photosensitivity under 516 nm illumination as a function of incident light power; (e) Photosensitivity of the device at different wavelengths; (f) Photosensitivity of the device at different frequencies. Detailed Implementation

[0050] In specific implementation, the method for fabricating dual-capacitor phototransistors using mechanical stripping, dry transfer, vacuum annealing, electron beam lithography (EBL), reactive ion etching (RIE), and electron beam evaporation (EBV) is as follows:

[0051] (1) An oxide is deposited on the surface of a photosensitive semiconductor by atomic layer deposition (AlD) or a substrate with an oxide / photosensitive semiconductor structure is formed by thermal oxidation.

[0052] (2) Metal gates are deposited on oxide / photosensitive semiconductor substrates by electron beam exposure (EBL), electron beam evaporation (EBV) and lift-off processes, or monolayer or few-layer graphene is transferred to the substrate as gates by micromechanical lift-off.

[0053] (3) Single-layer graphene or few-layer graphene, TMDCs, h-BN thin layers, etc. are peeled off from their bulk and onto a Si substrate with a 300nm thick SiO2 insulating layer on the surface by micromechanical peeling.

[0054] (4) Using a heterojunction transfer platform, the h-BN encapsulation layer is lifted with propylene carbonate (PPC), and then the channel layer and gate insulating layer are lifted in sequence to form a heterojunction. Then, the temperature is raised and it is deposited on an oxide / photosensitive semiconductor substrate with a gate structure.

[0055] (5) Place the stacked heterojunction into a vacuum annealing furnace for annealing treatment;

[0056] (6) The source and drain electrodes connected to the channel, the metal disk in the photosensitive capacitor, and the metal line connecting the metal disk to the gate are constructed by EBL, reactive ion etching (RIE), EBV and stripping processes.

[0057] The feasibility of the present invention will be further demonstrated below through examples.

[0058] Example

[0059] like Figure 1 As shown, the core structure of a dual-capacitor phototransistor consists of a photosensitive capacitor and an electrostatic modulation capacitor connected in series. The photosensitive capacitor, from bottom to top, consists of p... +The electrostatic modulation capacitor is composed of a Si, SiO2, and Ti-Au metal disk; from bottom to top, it consists of a graphene gate, an h-BN insulating layer, and a MoS2 channel; the graphene gate of the lower layer of the electrostatic modulation capacitor is connected to the Ti-Au metal disk of the upper layer of the photosensitive capacitor by a metal line. Simultaneously, graphene is used as a contact electrode connected to the MoS2 channel, and h-BN above the MoS2 channel serves as an encapsulation layer.

[0060] like Figure 2 As shown, the dual-capacitor phototransistor is constructed through the following three steps:

[0061] (a) Material preparation: Two-dimensional materials were exfoliated from the bulk material using a micromechanical exfoliation method, as described in the literature: Novoselov, KS, Geim, AK, Morozov, SV, et al. (2004) Electric Field Effect in Atomically Thin Carbon Films. Science, 306, 666-669. Graphene, h-BN, and MoS2 were exfoliated from the bulk material onto commercially available SiO2 / p + -Si substrate (SiO2 layer is p) + The silicon surface has an oxide layer with a thickness of 50–1000 nm, and the silicon doping dielectric is boron with a doping concentration of 0.2–0.05 Ωcm. -1 ).

[0062] (b) Heterojunction fabrication: A sufficiently thick h-BN encapsulation layer is raised using propylene carbonate (PPC) attached to the bottom of polydimethylsiloxane (PDMS), and then the graphene source, drain electrode, MoS2 channel and gate insulating layer are raised in sequence; the above stacked structure is placed on the graphene gate on the SiO2 / Si substrate at a high temperature of 90-140°C; the PPC residue on the surface of the heterojunction is removed by vacuum annealing, the vacuum annealing temperature is controlled at 350°C, the heating time is 0.5h, the holding time is 1h, and the furnace is cooled to room temperature.

[0063] (c) The lead electrodes of the graphene source and drain electrodes are constructed by electron beam lithography (EBL), reactive ion etching (RIE), electron beam evaporation (EBV) and lift-off processes; then the metal disk and the connection line between the metal disk and the graphene gate are constructed by the same process.

[0064] The lead electrode is a composite of titanium (Ti) layer and gold (Au) layer. First, a Ti layer with a thickness range of 4-6 nm is deposited by vapor deposition, and then an Au layer with a thickness range of 50-60 nm is deposited by vapor deposition.

[0065] The metal disk and connecting wires are composites of Ti and Au layers. First, a Ti layer with a thickness ranging from 1 to 6 nm is deposited by evaporation, and then an Au layer with a thickness ranging from 1 to 60 nm is deposited by evaporation.

[0066] The thickness of the h-BN package layer is 20–50 nm, the thickness of the h-BN gate insulating layer is 1–20 nm, the thickness of the graphene gate and contact electrode layers is 0.3–10 nm, and the thickness of the MoS2 channel layer is 1–20 nm.

[0067] like Figure 3 As shown, the dual-capacitor phototransistor serves as the optical characterization of a traditional photodetector. (a) An overall optical photograph of the device, including an electrostatic modulation capacitor and a photosensitive capacitor connected by a metal wire; the metal disk in the photosensitive capacitor has a size of 100 × 100 μm; (b) A photograph of the electrostatic modulation capacitor in the device, which consists of a graphene gate, an h-BN insulating layer, and a MoS2 channel from bottom to top. The graphene gate of the lower layer of the electrostatic modulation capacitor is connected to the metal disk of the upper layer of the photosensitive capacitor by a metal wire. Graphene is used to connect to the MoS2 channel as the source and drain contacts of the transistor, and h-BN above the channel serves as the encapsulation layer; (c) An atomic force microscope (AFM) photograph of the h-BN gate insulating layer shows that the material surface is clean and smooth; (d) The gate insulating layer is 12 nm thick.

[0068] like Figure 4 As shown, the principle of the dual-capacitor phototransistor as a traditional photodetector and its photoelectric performance characterization under 1065nm illumination are illustrated. (a) When using a 12nm thick h-BN as the gate insulating layer, photogenerated carriers are generated in Si when light shines on the metal disk of the photosensitive capacitor. Under the action of the built-in electric field, photogenerated holes are pushed into the substrate, while photogenerated electrons accumulate at the SiO2 / Si interface below the metal disk and induce positive charges in the Ti-Au metal disk through charge coupling. Since the connected Ti-Au metal disk and graphene gate need to maintain electrical neutrality, negative charges will be pushed towards the graphene gate, which is equivalent to applying a negative gate voltage to MoS2 and generating a negative photoelectric response. (b) Transfer characteristic curves of the device under darkness and different light intensities. As the light intensity increases, the threshold voltage shifts to the right. (c) As the light intensity increases, the photosensitivity gradually increases and exceeds 10. 7 (d) The device's descent speed is 10 μs and its rise speed is 500 μs.

[0069] like Figure 5As shown, the optical characterization of the dual-capacitor phototransistor as an emerging dynamic vision sensor is illustrated. (a) An overall optical photograph of the device, including an electrostatic modulation capacitor and a photosensitive capacitor connected by a metal wire; the metal disk in the photosensitive capacitor has a size of 100×100μm; (b) A photograph of the electrostatic modulation capacitor in the device, which consists of a graphene gate, an h-BN insulating layer, and a MoS2 channel from bottom to top; the lower graphene gate of the electrostatic modulation capacitor is connected to the upper metal disk of the photosensitive capacitor by a metal wire. Graphene is used to connect to the MoS2 channel as the source and drain contacts of the transistor, and h-BN above the channel serves as the encapsulation layer; (c) An atomic force microscope (AFM) photograph of the h-BN gate insulating layer shows that the material surface is clean and smooth; (d) The gate insulating layer is 4nm thick.

[0070] like Figure 6 As shown, the principle and photoelectric performance characterization of the dual-capacitor phototransistor as a dynamic vision sensor are illustrated. (a) When the dynamic light is on, electrons at the graphene gate in the electrostatic modulation capacitor cannot remain but pass through the thin h-BN, causing the negative photocurrent to gradually decay. (b) When the dynamic light is off, the number of electrons accumulated at the SiO2 / Si interface decreases, and the positive charge at the Ti-Au metal disk also decreases. This will recall electrons from the graphene gate, leaving a positive charge there, thus generating a positive photoresponse. However, these positive charges will attract MoS2 channel electrons to tunnel through the thin h-BN, thus attenuating the positive photocurrent; (c) the device only responds to dynamic light; the static power consumption of the device can be calculated to be approximately 40 pW by power equaling the product of voltage and current; (d) the photosensitivity gradually increases with increasing light intensity; (e) at the same incident light power, the photosensitivity gradually decreases with increasing incident light wavelength; the device has a significant photoelectric response to light in the 405 to 1065 nm wavelength range; (f) at the same incident light wavelength and power, the device photosensitivity decreases with increasing light modulation frequency; the device has a significant photoelectric response to light in the 1 to 1000 Hz range.

[0071] The examples demonstrate that the novel optoelectronic device proposed in this invention, the dual-capacitor phototransistor, utilizes the charge coupling effect between the two capacitors to achieve a response to static or dynamic optical signals, thus enabling the device to function as a traditional photodetector or an emerging dynamic sensor. Simultaneously, the dual-capacitor phototransistor exhibits excellent optoelectronic performance: when used as a traditional photodetector, its photosensitivity exceeds 10-1. 7 The response speed is close to 10μs. When used as a dynamic vision sensor, the device can detect dynamic optical signals in the 405–1065nm wavelength band and the 1–1000Hz frequency band, with a static power consumption of close to 40pW. This invention provides new technologies and ideas for realizing high-performance, low-power dynamic vision sensors in the future.

Claims

1. A dual capacitance phototransistor, characterized by, A dual-capacitor phototransistor consists of a photosensitive capacitor and an electrostatic modulation capacitor connected in series, as shown in the following structure: The photosensitive capacitor is composed of a photosensitive semiconductor, an oxide layer and a metal disk stacked from bottom to top. The electrostatic modulation capacitor is composed of a gate, a gate insulating layer and a channel layer stacked from bottom to top. The metal disk on the upper layer of the photosensitive capacitor is connected to the gate on the lower layer of the electrostatic modulation capacitor through a metal wire. The gate insulating layer in the electrostatic modulation capacitor is hexagonal boron nitride (h-BN) with a thickness ranging from 1 to 20 nm. When the gate insulating layer in the electrostatic modulation capacitor is thick, the dual-capacitor phototransistor can detect static light signals. When the gate insulating layer is thin, the dual-capacitor phototransistor will only respond to dynamic light signals.

2. The dual capacitance phototransistor according to claim 1, wherein The photosensitive semiconductors at the bottom of the photosensitive capacitor include silicon (Si), gallium nitride (GaN), or germanium (Ge). The oxides in the middle layer of the photosensitive capacitor include silicon dioxide (SiO2), aluminum oxide (Al2O3), or hafnium dioxide (HfO2). The metal disks on the top layer of the photosensitive capacitor include titanium (Ti), gold (Au), palladium (Pd), aluminum (Al), silver (Ag), chromium (Cr), or their alloys.

3. The dual capacitance phototransistor according to claim 1, wherein The gate of the electrostatic modulation capacitor is a metal or a half-metal. The metals include Ti, Au, Pd, Al, Ag, Cr or their alloys, and the half-metals include monolayer graphene or few-layer graphene, with the number of few-layer graphene layers ranging from 2 to 30. The channel layer of the electrostatic modulation capacitor is a semiconductor or a half-metal. The semiconductor is a thin layer of transition metal sulfide (TMDCs), and the half-metal is graphene. The thickness of the channel layer is 1 to 20 nm.

4. The dual-capacitor phototransistor according to claim 1, characterized in that, The metal wires connecting the metal disk in a photosensitive capacitor to the gate in an electrostatic modulation capacitor include Ti, Au, Pd, Al, Ag, Cr, or their alloys.

5. A method of fabricating a dual capacitance phototransistor according to any one of claims 1 to 4, characterized by, Includes the following steps: (1) An oxide is deposited on the surface of a photosensitive semiconductor by atomic layer deposition (AlD) or a substrate with an oxide / photosensitive semiconductor structure is formed by thermal oxidation. (2) Metal gates are deposited on oxide / photosensitive semiconductor substrates by electron beam exposure (EBL), electron beam evaporation (EBV) and lift-off processes, or single-layer graphene or few-layer graphene is transferred to the substrate as a graphene gate by micromechanical lift-off. The number of layers of few-layer graphene is 2 to 30. (3) Single-layer graphene or few-layer graphene, TMDCs, and h-BN thin layers are peeled off from their bulk material by micromechanical peeling onto a Si substrate with a SiO2 insulating layer of 50~1000 nm thickness on the surface. (4) Using a heterojunction transfer platform, the h-BN encapsulation layer is lifted with propylene carbonate PPC, and then the channel layer and gate insulating layer are lifted in sequence to form a heterojunction. Then the temperature is raised and it is deposited on an oxide / photosensitive semiconductor substrate with a gate structure. (5) Place the stacked heterojunction into a vacuum annealing furnace for annealing treatment; (6) The source and drain electrodes connected to the channel, the metal disk in the photosensitive capacitor, and the metal line connecting the metal disk to the gate are constructed by EBL, reactive ion etching (RIE), EBV and stripping processes.

6. The method of fabricating a dual capacitance phototransistor according to claim 5, wherein In step (1), the thickness of the photosensitive semiconductor ranges from 50 to 1000 μm, and its doping type is intrinsic, p-type or n-type doping. The thickness of the oxide prepared on the surface of the photosensitive semiconductor using the ALD method or thermal oxidation method ranges from 50 to 1000 nm.

7. The method of fabricating a dual capacitance phototransistor according to claim 5, wherein In step (2), the thickness of the vapor-deposited metal gate ranges from 2 to 60 nm, and the thickness of the transferred graphene gate ranges from 0.3 to 10 nm.

8. The method of fabricating a dual capacitance phototransistor according to claim 5, wherein, In step (4), the encapsulation layer h-BN, the channel layer and the gate insulating layer h-BN are stacked in sequence to form a sandwich structure. The channel layer is completely covered by the encapsulation layer h-BN and the gate insulating layer h-BN. The heterojunction drop temperature is 90~140℃, and the gate needs to cover the channel in the vertical direction.

9. The method of fabricating a dual capacitance phototransistor according to claim 5, wherein, In step (5), the PPC residue on the surface of the heterojunction is removed by annealing. The vacuum annealing temperature is controlled at 350~450℃, the heating time is 0.5~2 h, the holding time is 0.5~2 h, and the furnace is cooled to room temperature.

10. The method for fabricating a dual-capacitor phototransistor according to claim 5, characterized in that, In step (6), the lead electrode is a composite of Ti and Au layers. First, a Ti layer with a thickness range of 4~6 nm is deposited on the substrate and graphene contact electrode, and then an Au layer with a thickness range of 50~60 nm is deposited. The metal disk and metal wire in the photosensitive capacitor are a composite of Ti and Au layers or an Au layer. The thickness of the Ti layer ranges from 1~6 nm, and the thickness of the Au layer ranges from 1~60 nm.

Citation Information

Patent Citations

  • Pixel structure, instant motion detection image sensor and method

    CN115332275A

  • Tunable photocapacitive optical radiation sensor enabled radio transmitter and applications thereof

    US20160041030A1