A Ga2O3 / ITZO heterojunction-based photomemristor, its fabrication method and its application

Abstract

The application discloses a Ga2O3 / ITZO heterojunction photoelectric memristor, a preparation method and application, the photoelectric memristor comprises a bottom electrode layer, a Ga2O3 layer, an IZTO layer and a top electrode layer arranged in sequence from bottom to top, wherein the Ga2O3 layer and the IZTO layer are used as a heterostructure layer. Finally, a photoelectric memristor device with a sandwich structure is formed. The application has the advantages of simple preparation process, large-scale preparation, excellent and stable performance, obvious memristor effect under electric signals and multi-wavelength optical signals, and the resistance value can change according to different voltage values or light intensities, and the resistance value change is highly similar to human brain nerve synapses. The characteristics enable the photoelectric memristor to simulate the behavior of biological nerve synapses and have the ability of a visual bionic system integrating perception and memory.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of semiconductor electronic devices, in particular to a Ga2O3 / ITZO heterojunction photoelectric memristor, a preparation method and applications thereof. BACKGROUND

[0002] Currently, complementary metal-oxide-semiconductor (CMOS) technology is still the core support for neural network algorithms, but its computing efficiency lags far behind the human brain, mainly due to the limitations of the von Neumann architecture. With the advent of the big data era and the rapid development of the Internet of Things technology, traditional computer systems are facing the dual challenges of low efficiency and high energy consumption. In order to address these problems, architectures and operating principles that simulate the human brain have become an important strategy to achieve parallel processing with strong capabilities, high efficiency and low energy consumption. Since the human brain achieves learning and memory through a large number of neurons and their complex synaptic connections, developing artificial synapse devices that can simulate the functional characteristics and operating mechanisms of biological synapses has become a key measure to promote brain-like neural network computing from the hardware level.

[0003] In addition, in nature, most information is transmitted through optical signals, and responding to optical signals is the basis for decoding optical information. The effect of light on devices is mainly through the influence of light on the resistance change layer material to regulate the device, and finding new materials that can acutely perceive and respond to optical signals is a popular research direction in materials science today. Currently, optical response materials are mainly divided into two categories: inorganic materials and organic materials. Inorganic materials include metal oxides, and organic materials include perovskites. However, existing optical response materials have many problems in preparation or storage. For example, perovskite materials have strict requirements for storage conditions, such as changes in environmental factors such as humidity and temperature, which can irreversibly affect their performance, which undoubtedly increases the difficulty of material storage and transportation. These problems limit the promotion of photoelectric memristors and hinder the development of the optoelectronic industry. Therefore, it is of great significance to use simple materials to prepare stable and easily stored optoelectronic devices through simple processing techniques. The realization of this goal not only hopes to reduce production costs and improve production efficiency, but also greatly broadens the application range of optical response materials, injecting new vitality into the vigorous development of optoelectronic devices.

[0004] Therefore, in order to overcome the above challenges, the prior art still needs to be improved and developed. SUMMARY

[0005] The present application relates to the technical field of semiconductor electronic devices, in particular to a Ga2O3 / ITZO heterojunction photoelectric memristor, a preparation method and applications thereof.

[0006] To achieve the above object, the technical scheme of the present application is:

[0007] In a first aspect, the application provides a Ga2O3 / ITZO heterojunction photoelectric memristor, comprising, from bottom to top, a bottom electrode layer, a Ga2O3 layer, an ITZO layer, and a top electrode layer; wherein the Ga2O3 layer and the ITZO layer are heterostructure layers.

[0008] Optionally, the Ga2O3 layer is in the form of a nanoscale thin film layer.

[0009] Alternatively, the ITZO layer is in the form of a nanoscale thin film attached to the Ga2O3 layer.

[0010] Optionally, the bottom electrode layer is made of one of ITO conductive glass, fluorine-doped tin dioxide conductive substrate, or aluminum-doped zinc oxide conductive substrate.

[0011] Optionally, the top electrode layer is made of any one or several of gold, platinum, aluminum, copper, silver, and titanium.

[0012] Optionally, the thickness of the Ga2O3 layer and the ITZO heterojunction thin film layer is 170-200 nm.

[0013] Optionally, the thickness of the top electrode layer is 70-100 nm, and the thickness of the bottom electrode layer is 150 nm.

[0014] In a second aspect, the application provides an application of the Ga2O3 / ITZO heterojunction photoelectric memristor, which realizes visual allogenic synapse bionics under two wavelengths of light signals.

[0015] Or,

[0016] The Ga2O3 / ITZO heterojunction photoelectric memristor is used for handwriting digit recognition based on a neural network.

[0017] In a third aspect, the application provides a preparation method of the Ga2O3 / ITZO heterojunction photoelectric memristor, comprising the following steps:

[0018] Step 1: the bottom electrode layer is made of ITO conductive glass and is pretreated.

[0019] Step 2: a Ga2O3 thin film is prepared by a magnetron sputtering method, and the Ga2O3 thin film layer is attached to the bottom electrode layer to obtain a Ga2O3 thin film grown in situ on the surface of the bottom electrode layer.

[0020] Step 3: an ITZO thin film is prepared by a magnetron sputtering method, and the ITZO thin film is attached to the Ga2O3 thin film to obtain a heterojunction thin film layer composed of Ga2O3 and ITZO.

[0021] Step 4, magnetron sputtering a metal top electrode on the heterojunction thin film layer to form a multi-wavelength performance regulated photoelectric memristor.

[0022] Optionally, the pretreatment comprises: sequentially performing ultrasonic cleaning on the bottom electrode layer with acetone, deionized water and anhydrous ethanol, and then drying.

[0023] Optionally, the magnetron sputtering in step 2 is performed under the condition of a working gas pressure of 0.5 Pa, a sputtering power of 90-110 W, and a deposition time of 10-20 min.

[0024] The magnetron sputtering in step 3 is performed under the condition of a working gas pressure of 0.5 Pa, a sputtering power of 70-100 W, and a deposition time of 5-7 min.

[0025] The magnetron sputtering in step 4 is performed under the condition of a working gas pressure of 1 Pa, a sputtering power of 50-60 W, and a deposition time of 10-15 min.

[0026] Compared with the prior art, the present application has the following beneficial effects:

[0027] Ga2O3 is a wide-bandgap semiconductor, which makes it effective in absorbing ultraviolet light. Due to its wide-bandgap property, Ga2O3 has a high absorption coefficient in the ultraviolet waveband, so it has been widely researched and applied in ultraviolet photodetectors, ultraviolet photodiodes, solar cells and other applications. In addition, Ga2O3 is relatively chemically stable and can maintain its semiconductor properties at high temperatures, which makes it have advantages in extreme environmental conditions. IZTO (Indium Zinc Tin Oxide) as a transparent conductive oxide material, can adjust its photoelectric performance by changing the ratio of indium, zinc and tin. This adjustability allows IZTO to customize its conductivity and transmittance according to specific application requirements to optimize device performance. In addition, IZTO has high transparency and good conductivity, and using transparent IZTO electrodes in memristors can make the device easier to integrate into transparent electronic devices. Moreover, the high mobility of IZTO helps to achieve memristor switching at low current, which enables the memristor to work at low power consumption, thereby prolonging the service life of the device. Ga2O3 and IZTO have multiple advantages such as low cost, high carrier mobility, excellent light absorption ability and stable thermal performance. Therefore, they have wide application prospects in memory, optoelectronic devices and neuromorphic computing devices, and are expected to realize large-scale industrial application and further promote the development of related technologies.

[0028] Based on the Ga2O3 / ITZO heterojunction photoelectric memristor, it can realize the adjustable photoelectric synapse function and the visual bionic system with perception and memory. The proposed device successfully obtains the electrical synapse function by changing the pulse interval time of the electrical stimulus. In addition, the device has good light-induced response to deep ultraviolet, ultraviolet, blue, green and other wavelengths, realizing the light-induced synapse function. Due to the synergistic effect of multiple light signals, the two-terminal device can exhibit a function similar to a three-terminal synapse (i.e. the promotion of heterosynapse), based on which, by simultaneously applying a mixed signal of ultraviolet light (365nm) and blue light (470nm), the classical Pavlovian conditioned reflex experiment is simulated and the visual heterosynapse function is realized. Heterosynapse promotion helps to improve the learning and memory ability of the device, which has great potential in various aspects of neural networks and neuromorphic computing. Moreover, an innovative architecture of a photoelectric memristor cross array is proposed, which combines the advantages of electrical modulation and 275nm / 365nm / 470nm / 550nm light modulation memristors, and uses the subtle difference in the forward conductance of the two types of memristors to accurately represent the neural network weights. This method expands the weight adjustment range, enhances the anti-interference ability, and improves the fault tolerance in the weight mapping process, even in the presence of hardware imperfections and random noise, still maintaining a high recognition rate, indicating its good robustness and potential in neuromorphic computing applications.

[0029] Finally, the material used in the photoelectric memristor is environmentally friendly and simple to prepare, easy to store, and suitable for large-area promotion, which can be widely used in the field of information technology. BRIEF DESCRIPTION OF DRAWINGS

[0030] Figure 1 is a structural schematic diagram of the photoelectric memristor of the present application;

[0031] Figure 2 is a cross-sectional SEM characterization diagram of the photoelectric memristor of Example 2;

[0032] Figure 3 is a plot of the opening and erasing curves of the photoelectric memristor in Example 2;

[0033] Figure 4 is an effect diagram of the double-pulse facilitation or inhibition performance of the electrical signal of the photoelectric memristor in Example 2 when used for synapse simulation;

[0034] Figure 5 is the light response of the photoelectric memristor in Example 2 to deep ultraviolet (275nm), ultraviolet light (365nm), blue light (470nm) and green light (550nm) under different light intensities, respectively;

[0035] Figure 6Figure 2 is a diagram of a dual-pulse facilitation effect of four light signals of the optoelectronic memristor in Example 2 when used for synapse simulation;

[0036] Figure 7 Figure 3 is an application effect diagram of the optoelectronic memristor in Example 2 when used for visual heterosynapse simulation.

[0037] Figure 8 Figure 4 is a diagram of the optoelectronic memristor in Example 2 when used for handwriting digit recognition based on a neural network. DETAILED DESCRIPTION

[0038] The technical solutions of the present application will be further described below in combination with the drawings and examples.

[0039] The present disclosure will be further described below in combination with the drawings and examples. It should be noted that the following detailed description is exemplary and is intended to provide further description of the present disclosure. Unless otherwise specified, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which the present disclosure belongs.

[0040] It should be noted that the terms used herein are only for the purpose of describing specific embodiments and are not intended to limit the exemplary embodiments according to the present disclosure. As used herein, the singular form is intended to include the plural form unless the context clearly indicates otherwise, and it should also be understood that when the terms "comprise" and / or "include" are used in the specification, there is a presence of a feature, step, operation, device, component and / or combinations thereof. It should be noted that the various embodiments in the present disclosure and the features in the embodiments can be combined with each other without conflict. The embodiments will be described in detail below in combination with the drawings.

[0041] In the technical solutions disclosed in one or more embodiments, as shown in Figure 1 The optoelectronic memristor comprises, from bottom to top, a bottom electrode layer of ITO conductive glass, a Ga2O3 layer, an IZTO layer, and a top electrode layer of Ag; wherein the Ga2O3 layer and the IZTO layer between the top electrode and the bottom electrode serve as a heterostructure layer.

[0042] Ga2O3 is a wide-bandgap semiconductor, which makes it effective in absorbing ultraviolet light. Due to its wide-bandgap property, Ga2O3 has a high absorption coefficient in the ultraviolet waveband, making it widely studied and applied in applications such as ultraviolet photodetectors, ultraviolet photodiodes, and solar cells. In addition, Ga2O3 is relatively chemically stable and can maintain its semiconductor properties at high temperatures, making it advantageous for use in extreme environmental conditions. IZTO (Indium Zinc Tin Oxide) is a transparent conductive oxide material that can be adjusted in its photoelectric properties by changing the ratio of indium, zinc, and tin. This adjustability allows IZTO to tailor its electrical conductivity and transmittance to optimize device performance according to specific application requirements. In addition, IZTO has high transparency and good electrical conductivity, and using transparent IZTO electrodes in memristors makes it easier to integrate the device into transparent electronic devices. The high mobility of IZTO helps to achieve memristive switching at low current, which enables the memristor to work at low power consumption, thereby prolonging the service life of the device. Ga2O3 and IZTO materials have multiple advantages such as low cost, high carrier mobility, excellent light absorption ability, and stable thermal performance, so they have wide application prospects in memory, optoelectronic devices, and neuromorphic computing devices, and are expected to realize large-scale industrialization and further promote the development of related technologies.

[0043] In this way, by adopting Figure 1The Ga2O3 / ITZO heterojunction optoelectronic memristor shown in the structure can realize adjustable optoelectronic synaptic function and has a visual bionic system with perception and memory. The proposed device successfully obtains the electrical synaptic function by changing the pulse interval time of the electrical stimulus. In addition, the device has good light-induced response to deep ultraviolet, ultraviolet, blue, green and other wavelengths, realizing the light-induced synaptic function. Due to the synergistic effect of multiple light signals, the two-terminal device can exhibit a function similar to a three-terminal synapse (i.e., the promotion of heterosynapse). Based on this, the Pavlov classical conditioned reflex experiment is simulated by simultaneously applying a mixed signal of ultraviolet light (365 nm) and blue light (470 nm), and the visual heterosynapse function is realized. The promotion of heterosynapse helps to improve the learning and memory ability of the device, and has great potential in various aspects of neural networks and neuromorphic computing. Moreover, an innovative architecture of a cross array of optoelectronic memristors is proposed, which combines the advantages of electrical modulation and 275nm / 365nm / 470nm / 550nm light modulation memristors. The subtle differences in the forward conductance of the two types of memristors are used to accurately represent the neural network weights. This method expands the weight adjustment range, enhances the anti-interference ability, and improves the fault tolerance in the weight mapping process. Even in the presence of hardware imperfections and random noise, it still maintains a high recognition rate, indicating its good robustness and potential in neuromorphic computing applications. Finally, the materials used in the proposed optoelectronic memristor are environmentally friendly and simple to prepare, easy to store, and suitable for large-scale promotion, which can be widely applied in the field of information technology.

[0044] In some embodiments, the Ga2O3 layer 3 is in the form of a nanoscale thin film layer;

[0045] In some embodiments, the IZTO layer 2 is in the form of a nanoscale thin film; the IZTO layer is attached to the Ga2O3 thin film layer in the form of a nanoscale thin film. The thickness of the Ga2O3 layer and the IZTO heterojunction thin film layer is 170nm-200nm

[0046] Optionally, the bottom electrode layer can use one of ITO conductive glass, fluorine-doped tin dioxide (FTO) conductive substrate, or aluminum-doped zinc oxide (AZO) conductive substrate.

[0047] In this embodiment, ITO conductive glass is preferably used as the substrate because Ga2O3 grown on the ITO thin film is more complete.

[0048] In some embodiments, the top electrode can use any one or several of gold, platinum, aluminum, copper, silver, titanium. Preferably, noble metals such as gold, silver, platinum, etc. can be used, which are not easily oxidized during use. The thickness of the Ga2O3 layer and the IZTO heterojunction thin film layer is 170nm-200nm.

[0049] In some embodiments, the Ga2O3 / ITZO heterojunction optoelectronic memristor can be applied to the following aspects:

[0050] Application one: visual heterosynaptic bionics:

[0051] Heterosynapses are widely present in various parts of the visual system, such as the retina and the visual cortex, and play a key role in regulating human visual function. The process promoted by heterosynapses in synaptic devices is verified by a classic Pavlovian conditioning experiment. In the experiment, "food vision" is used as the unconditioned stimulus (US), and "bell sound" is used as the neutral stimulus (NS). The "bell sound" and "food vision" are simulated by light pulses with wavelengths of 365 nm (as pre-signals) and 275 nm (as modulation signals), respectively. The successful transformation of NS into a conditioned stimulus (CS) through the training process indicates that the heterosynaptic promotion has achieved visual heterosynaptic bionics. Visual heterosynaptic bionics shows great potential in improving the learning and memory capabilities of neural networks, and is expected to promote the development of high-performance artificial visual memory systems.

[0052] Application two: neural network-based digital handwritten recognition

[0053] In traditional computer architectures, achieving high-precision handwritten digital recognition usually requires a large amount of computing resources and storage capacity. However, the unique optoelectronic synaptic plasticity of memristors opens up new possibilities for implementing neural networks at the hardware level. To illustrate this capability, a handwritten digital recognition system based on a convolutional neural network is developed, which implements neuromorphic computing functions in a memristor array, using the conductance values of the memristors to replace the weight parameters in the CNN. During the training process, the conductance of the memristor is dynamically adjusted by external voltage or optoelectronic signals to achieve learning and updating of network weights. By normalizing the conductance values, they can be accurately mapped to the weights in the neural network, which is conducive to efficient hardware storage and updating of weights. This method not only simulates the synaptic plasticity of biological neural networks, but also significantly reduces energy consumption and computing resource requirements. Moreover, an innovative architecture of a photoelectric memristor cross array is proposed, which combines the advantages of electrically modulated and 275 nm / 365 nm / 470 nm / 550 nm optically modulated memristors, and uses the subtle differences in the forward conductance of the two types of memristors to accurately represent neural network weights. This method expands the weight adjustment range, enhances the anti-interference ability, and improves the fault tolerance in the weight mapping process. Even in the presence of hardware imperfections and random noise, the use of photoelectric modulation still maintains a high recognition rate, indicating its good robustness and potential in neuromorphic computing applications.

[0054] Embodiment 1

[0055] The preparation method of the photoelectric memristor provided in the embodiment comprises the following steps: first, a Ga2O3 thin film layer is grown on an ITO conductive glass substrate by magnetron sputtering; then, an IZTO nanometer thin film layer is formed on the Ga2O3 thin film layer by magnetron sputtering; and finally, an Ag electrode is formed on the IZTO nanometer thin film layer by magnetron sputtering, so as to form an ITO / Ga2O3 / IZTO / Ag structure photoelectric memristor.

[0056] Step 1: The ITO conductive glass is pretreated. The specific operation of the pretreatment is as follows: first, the bottom electrode layer is ultrasonically cleaned with acetone, deionized water and anhydrous ethanol in sequence, and then dried.

[0057] Specifically, the ITO conductive glass has a size of 2*2 cm, the ultrasonic cleaning time can be 5 min to 10 min, and the drying is performed at a temperature of 55-65 DEG C for 1-2 h.

[0058] Step 2: The Ga2O3 thin film is prepared by magnetron sputtering, and the Ga2O3 thin film layer is grown on the bottom electrode layer.

[0059] Specifically, the preset target material is a Ga2O3 ceramic target with a purity of 99.99%.

[0060] The Ga2O3 thin film is prepared by radio frequency magnetron sputtering process. The parameters of the magnetron sputtering are as follows: the working pressure is 0.5 Pa, the gas flow rate is 40 sccm, the sputtering power is 100 W, and the sputtering time is 15 min.

[0061] Step 3: The IZTO thin film is prepared by magnetron sputtering, and the IZTO thin film is attached to the Ga2O3 thin film to obtain a Ga2O3 / IZTO heterojunction thin film layer.

[0062] Specifically, the preset target material is a ZnO ceramic target and an ITO ceramic target with a purity of 99.99%.

[0063] The IZTO thin film with a thickness of 100 nm is deposited by magnetron co-sputtering of the ZnO ceramic target and the ITO ceramic target. The parameters of the magnetron sputtering are as follows: the working pressure is 0.5 Pa, the gas flow rate is 34 sccm, the ZnO sputtering power is 100 W, the ITO sputtering power is 70 W, and the sputtering time is 10 min.

[0064] Step 4: The Ag top electrode is prepared on the IZTO thin film by direct current magnetron sputtering.

[0065] The Ag top electrode with a thickness of 40 nm is deposited by direct current sputtering through a metal mask. The parameters of the magnetron sputtering are as follows: the working pressure is 0.5 Pa, the gas flow rate is 30 sccm, the sputtering energy is 60 W, and the sputtering time is 10 min.

[0066] Example 2

[0067] The preparation method of the photoelectric memristor provided in this embodiment is completely the same as that of Example 1, except that the sputtering time of the IZTO thin film prepared by the magnetron sputtering method in step 2 is adjusted from “10 min” to “6 min” during preparation. The specific steps are as follows:

[0068] Step 1, pretreat the ITO conductive glass. The specific operation of pretreatment is: first, ultrasonic clean the bottom electrode layer with acetone, deionized water and anhydrous ethanol in sequence, and then dry.

[0069] Step 2, prepare Ga2O3 thin film by magnetron sputtering method, and grow Ga2O3 thin film layer on the bottom electrode layer:

[0070] Specifically, the preset target material is a Ga2O3 ceramic target with a purity of 99.99%;

[0071] Prepare Ga2O3 thin film by radio frequency magnetron sputtering process. The parameters of magnetron sputtering are as follows: working pressure is 0.5 Pa; gas flow rate is 40 sccm; sputtering power is 100 W; sputtering time is 15 min.

[0072] Step 3, prepare IZTO thin film by magnetron sputtering method, and attach IZTO thin film on Ga2O3 thin film to obtain Ga2O3 / IZTO heterojunction thin film layer:

[0073] Specifically, the preset target material is a ZnO ceramic target and an ITO ceramic target with a purity of 99.99%;

[0074] IZTO thin film with a thickness of 80 nm is deposited by magnetron co-sputtering of ZnO ceramic target and ITO ceramic target. The parameters of magnetron sputtering are as follows: working pressure is 0.5 Pa; gas flow rate is 34 sccm; ZnO sputtering power is 100 W, ITO sputtering power is 70 W; sputtering time is 6 min.

[0075] Step 4, prepare Ag top electrode on IZTO thin film by direct current magnetron sputtering method.

[0076] IZTO thin film with a thickness of 80 nm is deposited by magnetron co-sputtering of ZnO ceramic target and ITO ceramic target. The parameters of magnetron sputtering are as follows: working pressure is 0.5 Pa; gas flow rate is 34 sccm; ZnO sputtering power is 100 W, ITO sputtering power is 70 W; sputtering time is 6 min.

[0077] Figure 2 The cross-sectional SEM characterization diagram of the photoelectric memristor in Example 2. Among them, the thickness of IZTO is about 80 nm, and the thickness of Ga2O3 is about 100 nm.

[0078] Figure 3 I-V curves of the photoelectric memristor in Example 2 are shown. When testing the device, the ITO conductive glass layer is grounded, and the voltage is applied on the Ag electrode. In the figure, "1" refers to the curve of the scanning process from 0-1.5V, "2" refers to the curve of the scanning process from 1.5-0V; "3" refers to the curve appearing from the scanning process from 0-1.5V, and "4" refers to the curve appearing from the scanning process from -1.5-0V. Among the "1"-"4" curves, "1" and "2" are the writing process of the memristor, which corresponds to the information storage state; "3" and "4" are the erasing process of the memristor, which corresponds to the information erasing state. Figure 4 It is shown that the photoelectric memristor with "ITO / Ga2O3 / IZTO / Ag" structure prepared in Example 2 exhibits typical non-volatile resistance switching characteristics and self-rectifying characteristics.

[0079] Figure 4 A pair of continuous electrical signal pulses are applied to the two ends of the ITO / Ga2O3 / IZTO / Ag device prepared in Example 2, and the corresponding current value of each stimulation is recorded, successfully simulating the biological synapse PPF / PPD function.

[0080] Figure 5 The photoresponse dependence of the ITO / Ga2O3 / IZTO / Ag photoelectric memristor prepared in Example 2 on deep ultraviolet (275nm), ultraviolet light (365nm), blue light (470nm) and green light (550nm) is shown. It can be concluded that as the light intensity continues to increase, the response current of the device will also gradually increase.

[0081] Figure 6 A pair of continuous light signal pulse stimuli are applied to the two ends of the ITO / Ga2O3 / IZTO / Ag device in Example 2, the light signal pulse is respectively deep ultraviolet (275nm), ultraviolet light (365nm), blue light (470nm) and green light (550nm) excitation, and the corresponding current value of each stimulation is recorded, successfully simulating the biological synapse PPF function.

[0082] Figure 7is the application effect diagram of the optoelectronic memristor in Example 2 for visual heterosynaptic emulation. In neural networks, heterosynapses have a profound impact on high-level cognitive functions such as learning and memory, sensory information processing, and emotional regulation by regulating the interaction between multiple synapses. Focusing on the complete behavior of individual photonic synapses, the synergistic effect of multiple light signals can enable a two-terminal device to exhibit functions similar to a three-terminal synapse (i.e., heterosynaptic facilitation). In biology, heterosynapses refer to the connections between neurons, which do not directly participate in the transmission of visual information but can significantly affect the activity of visual neurons. These synapses exist in various components of the visual system, such as the retina and visual cortex, and have a significant impact on the regulation of human visual function (e.g. Figure 7 a). To verify the process of heterosynaptic facilitation in the synaptic device, a classic Pavlovian conditioning experiment was conducted. In the experiment, "food vision" was used as the unconditioned stimulus (US), and "bell sound" was used as the neutral stimulus (NS). The ability of the memristor to achieve heterosynaptic facilitation under different light stimuli was thoroughly investigated. As shown in Figure 7 b, light pulses with a wavelength of 460 nm (as a pre-signal, pulse width of 1 ms, pulse intensity of 120 μW / cm 2 ) and 365 nm (as a modulating signal, pulse width of 1 ms, pulse intensity of 120 μW / cm 2 ) were used to simulate "bell sound" (NS) and "food vision" (US), respectively. Figure 7 c shows that the photocurrent caused by the neutral stimulus alone is 16 nA, which is set as the threshold current; while the photocurrent caused by the unconditioned stimulus is 26.7 nA, which is significantly higher than that of the neutral stimulus. During the training phase, the simultaneous application of NS and US caused a significant change in the photocurrent. After the training was completed, NS alone could induce a photocurrent of 18.3 nA, exceeding the threshold current. Therefore, NS successfully transformed into a conditioned stimulus (CS), achieving visual heterosynaptic emulation through heterosynaptic facilitation. The above experimental results show that the ITO / Ga2O3 / IZTO / Ag structure successfully simulates the classic Pavlovian conditioning experiment and achieves visual heterosynaptic emulation. Visual heterosynaptic emulation has great potential in improving neural network and computing capabilities, and is expected to significantly improve the learning and memory capabilities of devices, paving the way for the application of powerful artificial visual memory systems in the future.

[0083] Figure 8is a handwritten digit recognition graph based on neural network using the optoelectronic memristor in Example 2. In traditional computer architecture, high-precision handwritten digit recognition usually requires a large amount of computing resources and storage capacity. However, the unique optoelectronic synaptic plasticity of the memristor opens up new possibilities for implementing neural networks at the hardware level, demonstrating the huge computing potential of handwritten digit recognition. To illustrate this capability, a handwritten digit recognition system based on a convolutional neural network (CNN) was developed, as shown in Figure 8 a. The CNN model consists of two convolutional layers, two pooling layers, and a fully connected layer. The first layer has 32 channels and a size of 28x28; the second layer has 64 channels and a size of 14x14. The fully connected layer of the neural network contains 3136 input nodes, 1024 hidden nodes, and 10 output nodes, with the input nodes derived from handwritten digit images in the U.S. National Institute of Standards and Technology (MNIST) dataset. To implement neuromorphic computing functions in the memristor array, the conductance values of the memristors are used instead of the weight parameters in the CNN. During the training process, the conductance of the memristor is dynamically adjusted by external voltage or optoelectronic signals to achieve learning and updating of network weights. By normalizing the conductance values, they can be accurately mapped to the weights in the neural network, facilitating efficient hardware storage and updating of weights. This method not only simulates the synaptic plasticity of biological neural networks, but also significantly reduces energy consumption and computing resource requirements. As shown in Figure 8 b, an innovative architecture of an optoelectronic memristor cross array is proposed, which combines the advantages of electrical modulation and 275nm / 365nm / 470nm / 550nm light modulation memristors. This design takes advantage of the subtle differences in the forward conductance of the two types of memristors to accurately represent neural network weights. By combining electrical and optical modulation, this method expands the weight adjustment range, enhances the anti-interference ability, and improves the fault tolerance during the weight mapping process. Figure 8 c shows the relationship between the number of training iterations and the recognition rate under ideal modulation, electrical modulation, and 275nm / 365nm / 470nm / 550nm optoelectronic co-modulation conditions. It can be seen that the recognition rate under the six conditions increases with the number of iterations. The results show that the memristor neural network maintains a high recognition rate in the presence of hardware imperfections and random noise, indicating its good robustness and potential in neuromorphic computing applications. Figure 8 d is the highest recognition rate in 100 iterations, which is 98.40% under ideal conditions, 95.69% / 97.13% / 95.08% / 91.72% under 275nm / 365nm / 470nm / 550nm optoelectronic modulation, and 85.08% under pure electricity. Figure 8e-i further demonstrates that there is a strong correspondence between the predicted output values and the actual output values of the confusion matrix after 100 training cycles under electrical modulation and 275nm / 365nm / 470nm / 550nm optoelectronic co-modulation. The training process of the confusion matrix is intuitively represented by the color gradient along its diagonal. Under pure electrical modulation, the matrix appears chaotic. However, after the addition of 275nm / 365nm / 470nm / 550nm optoelectronic co-modulation, the matrix turns into a consistent color, indicating the successful identification of all patterns (0-9). This transformation reflects the reduction of recognition errors and the gradual learning of pattern recognition, highlighting the enhancement of synaptic learning, achieving higher image recognition capability through optoelectronic co-modulation, and highlighting its potential for image recognition applications in neuromorphic computing systems.

[0084] The above examples are only to illustrate the technical concepts and characteristics of the present application, and the purpose is to enable those skilled in the art to understand the content of the present application and to implement it, and cannot limit the protection scope of the present application. Any equivalent changes or modifications made according to the essence of the present application shall be covered within the protection scope of the present application.

Claims

1. A Ga2O3 / ITZO heterojunction photoelectromemristor, characterized in that, The Ga2O3 / ITZO heterojunction photoelectric memristor comprises, from bottom to top, a bottom electrode layer, a Ga2O3 layer, an ITZO layer, and a top electrode layer; wherein the Ga2O3 layer and the ITZO layer serve as a heterostructure layer; The Ga2O3 layer is in the form of a nanoscale thin film layer; Alternatively, the ITZO layer is in the form of a nanoscale thin film attached to the Ga2O3 layer; The thickness of the Ga2O3 layer and the IZTO heterojunction thin film layer is 170 nm to 200 nm; The thickness of the top electrode layer is 70 nm to 100 nm, and the thickness of the bottom electrode layer is 150 nm.

2. The Ga2O3 / ITZO heterojunction photoelectromemristor of claim 1, wherein, The bottom electrode layer is made of one of ITO conductive glass, fluorine-doped tin dioxide conductive substrate, or aluminum-doped zinc oxide conductive substrate.

3. The Ga2O3 / ITZO heterojunction photoelectromemristive device of claim 1, wherein, The top electrode is made of any one or several of gold, platinum, aluminum, copper, silver, and titanium.

4. Use of a Ga2O3 / ITZO heterojunction photoelectromemristive device according to any one of claims 1-3, wherein The Ga2O3 / ITZO heterojunction photoelectric memristor realizes visual allogenic synapse bionics under two wavelengths of light signals. Alternatively, The Ga2O3 / ITZO heterojunction photoelectric memristor is used for handwriting digit recognition based on a neural network.

5. A method for fabricating a Ga2O3 / ITZO heterojunction photomemristor as described in claim 1, characterized in that, The method comprises the following steps: Step 1: The bottom electrode layer is made of ITO conductive glass and is pretreated; Step 2: A Ga2O3 thin film is prepared by a magnetron sputtering method, and the Ga2O3 thin film layer is attached to the bottom electrode layer to obtain a Ga2O3 thin film grown in situ on the surface of the bottom electrode layer; Step 3: An IZTO thin film is prepared by a magnetron sputtering method, and the IZTO thin film is attached to the Ga2O3 thin film to obtain a heterojunction thin film layer composed of Ga2O3 and IZTO; Step 4: A metal top electrode is magnetron-sputtered on the heterojunction thin film layer to form a photoelectric memristor with a multi-wavelength regulation performance.

6. The production method according to claim 5, wherein The pretreatment comprises ultrasonic cleaning of the bottom electrode layer with acetone, deionized water, and anhydrous ethanol in sequence, followed by drying.

7. The production method according to claim 5, wherein The magnetron sputtering in step 2 is performed under a working pressure of 0.5 Pa, a sputtering power of 90-110 W, and a deposition time of 10-20 min; The magnetron sputtering in step 3 is performed under a working pressure of 0.5 Pa, a sputtering power of 70-100 W, and a deposition time of 5-7 min; The magnetron sputtering in step 4 is performed under a working pressure of 1 Pa, a sputtering power of 50-60 W, and a deposition time of 10-15 min.

Images