A gas sensing and computing integrated synaptic device and its synaptic response and preparation method

Abstract

The present application provides a gas sensing and computing integrated synaptic device and its preparation method, which belongs to the technical field of nanoelectronic devices. The device includes: a first functional layer, a barrier layer, a second functional layer, a first metal electrode and a second metal electrode; the first functional layer is heavily doped silicon; the second functional layer is CuO x The material is provided with defect energy levels formed by interstitial oxygen and copper vacancies; the barrier layer increases the contact area between the second functional layer and the test gas; a DC voltage sweep is applied between the first metal electrode and the second metal electrode to simulate the resistive switching behavior of the memristor; the second functional layer reacts with the test gas to realize the sensing function of the test gas; a constant voltage signal is applied between the first metal electrode and the second metal electrode, and in a gas pulse environment, the current response state presents a synaptic response; the present application can realize in-situ sensing and processing of gas information, improve information processing efficiency, and the gas information inference method of the gas sensing and computing integrated device is simpler.

Description

Technical Field

[0001] The present application belongs to the field of nanoelectronic device technology, and more specifically, relates to a gas sensing and computing integrated synaptic device, its synaptic response, and preparation method. Background Art

[0002] With the recent development of artificial neural network technology, biomimetic sensing and computing architectures have been recognized as an effective solution to current computer energy efficiency challenges. Bionic optical sensing and computing devices have demonstrated broad development prospects in the field of machine vision. However, the development of biomimetic gas sensing and computing systems remains limited, hindering their widespread adoption for in-situ sensing and computing tasks.

[0003] Traditional gas sensing and computing systems typically require two functional modules: sensing and information processing. After the external gas information is converted into electrical signals by the sensing module, it is filtered, amplified, and integrated for analysis by the information processing module. The sensing module is typically implemented as a gas sensor array, while the information processing and computing module consists of a transistor or memristor array. This architecture, with the sensing and computing modules separated from each other, results in a complex hardware structure and a large system size. There is also a significant time delay between information sensing and computing, and data transmission requires additional power. Therefore, it is necessary to utilize integrated gas sensing and computing devices with in-situ sensing and computing capabilities to simplify the system architecture, reduce the information processing distance, and accommodate more complex task processing functions.

[0004] However, current gas sensing and computing devices are limited to producing resistors in different resistance states in different atmospheres, and are unable to achieve real-time dynamic gas information sensing and processing. They have limited computing capabilities and low computational efficiency, and often use a three-terminal architecture, resulting in complex device fabrication processes. Therefore, exploring integrated gas sensing and computing devices with simple fabrication steps, capable of achieving biomimetic neuromorphic responses to gas signals and efficiently processing dynamic gas information, is of great long-term significance for the development of next-generation integrated gas sensing and computing systems. Summary of the Invention

[0005] In response to the defects of the existing technology, the purpose of this application is to provide a gas sensing and computing integrated synaptic device and its synaptic response and preparation method, aiming to solve the problems that the currently commonly used gas sensing and computing devices cannot provide real-time dynamic sensing and processing of gas signals, have limited information processing functions and poor efficiency, and cannot achieve bionic neuromorphic response to gas pulses and have a relatively complex preparation process.

[0006] To achieve the above objectives, the present application provides a gas sensing and computing integrated synaptic device, comprising: a first functional layer, a barrier layer, a second functional layer, a first metal electrode and a second metal electrode;

[0007] The second functional layer is divided into a first portion of the second functional layer and a second portion of the second functional layer; the barrier layer is located between the first functional layer and the first portion of the second functional layer, and the second metal electrode is located above the first portion of the second functional layer; the second portion of the second functional layer is located above the first functional layer, and the first metal electrode is also located above the first functional layer;

[0008] The first functional layer is heavily doped silicon; the second functional layer is CuO x The material is configured to have defect energy levels formed by interstitial oxygen and copper vacancies; the barrier layer is configured to elevate the second functional layer of the first portion to increase the contact area between the second functional layer and the test gas; the first metal electrode forms an ohmic contact with the first functional layer; the second metal electrode forms an ohmic contact with the second functional layer; a DC voltage scan is applied between the first metal electrode and the second metal electrode to simulate the resistive switching behavior of the memristor; the second functional layer is configured to react with the test gas to realize the sensing function of the test gas; when a constant voltage signal is applied between the first metal electrode and the second metal electrode, the current response state presents a synaptic response in a gas pulse environment; wherein 0.5≤x≤2.

[0009] Further preferably, the doping impurities in the first functional layer are one or more of B, Ga, In, P, As, and Sb;

[0010] Further preferably, the first metal electrode and the second metal electrode are one of Pt, Al, Ti, W, Ag, TiN or Cu; and the barrier layer is one of SiO2, Si3N4, Al2O3, HfO2 or Ta2O5.

[0011] Further preferably, the thickness of the first functional layer is 3nm-10um, the thickness of the barrier layer is 3nm-500nm, the thickness of the second functional layer is 3nm-500nm, and the thickness of the first and second metal electrodes is 3nm-200nm.

[0012] Further preferably, the test gas includes an oxidizing gas or a reducing gas; wherein the oxidizing gas includes O2, SO2, NO2 or H2S; and the reducing gas includes H2 or CO.

[0013] Further preferably, the synaptic response includes excitatory postsynaptic current, pulse amplitude-dependent plasticity, paired-pulse differentiation and pulse frequency-dependent plasticity.

[0014] In a second aspect, based on the gas sensing and computing integrated synaptic device provided in this application, this application provides a corresponding synaptic response method, comprising the following steps:

[0015] The gas-sensing and computing integrated synaptic device is placed in a confined space filled with N2 or an inert gas, a test gas pulse is introduced into the confined space, a constant voltage is applied between the first metal electrode and the second metal electrode, and the current response state of the gas-sensing and computing integrated synaptic device is read;

[0016] The changes in the current response state as the test gas pulse time increases are analyzed to obtain the sensing characteristics of the gas sensing and computing synaptic device for the test gas.

[0017] In a third aspect, based on the gas sensing and computing integrated synaptic device provided in this application, this application provides a corresponding synaptic response method, specifically:

[0018] The gas sensing and computing integrated synaptic device is placed in a closed space filled with N2 or inert gas. When a test gas pulse is introduced into the sealed space, the current response state is analyzed as the test gas concentration increases under the condition of the same test gas pulse time, and the pulse amplitude response characteristics in the neuromorphic response are obtained.

[0019] In a fourth aspect, based on the gas sensing and computing integrated synaptic device provided by this application, this application provides a corresponding synaptic response method, specifically:

[0020] The gas sensing and computing integrated synaptic device is placed in a confined space filled with N2 or inert gas. Multiple continuous test gas pulses are introduced into the confined space. The current response state is analyzed as the interval between test gas pulses shortens, and the pulse frequency response characteristics in the neuromorphic response are obtained.

[0021] In a fifth aspect, based on the gas sensing and computing integrated synaptic device provided in this application, this application provides a corresponding preparation method, which specifically includes the following steps:

[0022] S1: Prepare a doped Si layer and planarize it to obtain the first functional layer;

[0023] S2: preparing a patterned mask layer of a barrier layer on the first functional layer;

[0024] S3: preparing a barrier layer on the patterned mask layer of the barrier layer, and peeling off the mask layer to form a barrier layer;

[0025] S4: preparing a patterned mask layer of a second functional layer on the barrier layer;

[0026] S5: preparing a second functional layer on the patterned mask layer of the second functional layer, and peeling off the mask layer to form the second functional layer;

[0027] S6: preparing patterned mask layers of a first metal electrode and a second metal electrode on the first functional layer and the second functional layer, respectively;

[0028] S7: preparing the first metal electrode and the second metal electrode on the patterned mask layer of the first metal electrode and the second metal electrode, and peeling off the mask layer to form the first metal electrode and the second metal electrode, thereby completing the preparation of the gas sensing and computing integrated device.

[0029] Further preferably, the first metal electrode, the barrier layer, the second functional layer and the second metal electrode are prepared by a method selected from the group consisting of sputtering, physical vapor deposition, chemical vapor deposition, molecular beam epitaxy and electrochemical methods.

[0030] Further preferably, S1 is specifically: preparing a doped Si layer by doping the melt during the growth of single crystal Si by Czochralski method or zone melting method; or growing an intrinsic silicon layer by sputtering, physical vapor deposition or chemical vapor deposition, and then injecting impurities by thermal diffusion or ion implantation to prepare heavily doped n-type Si or p-type Si, cutting it into Si slices after doping is completed, and performing planarization treatment with chemical mechanical planarization equipment to form a first functional layer.

[0031] In general, the above technical solutions conceived by this application have the following beneficial effects compared with the existing technologies:

[0032] The present application provides a gas sensing and computing integrated device and its preparation method, wherein the first functional layer is made of heavily doped p-type silicon or heavily doped n-type silicon; the second functional layer is made of CuO x The material enhances the gas sensing response of the device through the heterojunction structure formed by the first functional layer and the second functional layer. When gas molecules reach the interface between the first functional layer and the second functional layer, the gas molecules will affect the carrier concentration inside the material due to surface adsorption, thereby causing changes in the current response inside the device. Compared with existing gas sensing and computing systems, the CuO based on the above structure x , with a more convenient inference process. Through the neuromorphic response to gas pulse signals, it can achieve in-situ sensing and processing of gas information, improving information processing efficiency. The gas information inference method of the gas sensing and computing device based on neuromorphic response is simpler and has the advantages of simpler device structure and preparation process. BRIEF DESCRIPTION OF THE DRAWINGS

[0033] Figure 1 The embodiment of the present application provides a CuO-based x Schematic diagram of the structure of the gas sensing and computing integrated device; wherein, 1-first functional layer; 2-barrier layer; 3-second functional layer; 4-second metal electrode; 5-first metal electrode;

[0034] Figure 2 It is the CuO provided in the embodiment of this application x Current-voltage curve of the gas sensing and computing device under a single DC voltage scan;

[0035] Figure 3 The CuO provided in the embodiment of the present application x Current-voltage curve of the gas sensing and computing device under three consecutive DC voltage sweeps;

[0036] Figure 4 The CuO provided in the embodiment of the present application x The current response curve of the gas sensing and computing device in a sealed space, read using a +1V voltage, when 50ppm NO2 gas is introduced for 750s.

[0037] Figure 5 The CuO provided in the embodiment of the present application x The current response curve of the gas sensor and calculator device under the conditions of 5ppm, 10ppm, 20ppm, 30ppm, 40ppm and 50ppm NO2 gas for 300s was read using a +1V voltage in a sealed space.

[0038] Figure 6 It is the CuO provided in the embodiment of this application x The current response curve of the gas sensing and computing device is read using a +1V voltage in a sealed space when five consecutive NO2 gas pulses with a concentration of 0ppm, a single pulse duration of 200s, and intervals of 200s and 400s are introduced. DETAILED DESCRIPTION

[0039] The embodiments of the present application are described below in conjunction with the drawings in the embodiments of the present application.

[0040] The present application provides a CuO-based x The gas sensing and computing integrated device uses the gas adsorption process to change the number of carriers inside the material, thereby changing the current response of the device in the constant voltage reading state and realizing a neuromorphic response. The device proposed in this application can realize in-situ processing of gas sensing information, improve information processing efficiency, and make the gas inference method simpler, which helps to complete complex tasks. More specifically, the application provides a gas sensing and computing integrated memristive synaptic device, including a first functional layer, a barrier layer, a second functional layer, a first metal electrode and a second metal electrode; the first functional layer adopts heavily doped silicon; the second functional layer adopts CuO x The barrier layer is located between the first functional layer and the second functional layer, and a portion of the second functional layer is raised to increase the contact area between the second functional layer and the gas; the first metal electrode forms an ohmic contact with the first functional layer; the second metal electrode forms an ohmic contact with the second functional layer; the gas sensing and computing integrated memristive synaptic device of the present application has a simulated conductivity-adjustable memristive characteristic under DC scanning; at the same time, the CuO xThe functional layer can react with oxidizing or reducing gases and can realize the perception function of the gas; the device can produce typical synaptic responses under the stimulation of oxidizing or reducing pulses, such as excitatory postsynaptic current (EPSC), pulse amplitude-dependent plasticity (SADP), paired pulse differentiation (PPF), and pulse frequency-dependent plasticity (SRDP); therefore, the device has the integrated gas sensing and computing function, and can simultaneously complete complex tasks such as the perception of oxidizing or reducing gases and information processing in a single device.

[0041] On the one hand, the present application provides a CuO-based x The gas sensing and computing integrated device comprises a first functional layer 1, a barrier layer 2, a second functional layer 3, a first metal electrode 5 and a second metal electrode 4 stacked in sequence from bottom to top;

[0042] The first functional layer is made of heavily doped p-type silicon or heavily doped n-type silicon. The doping impurities in the first functional layer may be one or more of B, Ga, In, P, As, and Sb. The resistivity is less than 10 ohm-cm.

[0043] The second functional layer uses CuO x material; wherein 0.5≤x≤2, so that there are defect energy levels formed by interstitial oxygen and copper vacancies, forming a P-type conductivity type;

[0044] The first metal electrode is used to read the current response state of the device; a DC voltage scan is applied between the first metal electrode and the second metal electrode, and when the voltage scan value is higher than the device threshold, the device exhibits memristor resistive switching behavior;

[0045] The specific working mechanism of the first functional layer and the second functional layer is as follows: when the test gas molecules reach the interface between the first functional layer and the second functional layer, the gas molecules will affect the carrier concentration inside the material due to surface adsorption; when the test gas is a reducing gas, the gas molecules will transfer electrons to the second functional layer during the adsorption process; since the carriers in the second functional layer are holes, this behavior will cause the carrier concentration in the second functional layer to decrease, the absolute value of the overall current response of the device to decrease, and the overall current response change rate to increase; when the test gas is an oxidizing gas, the gas molecules will take electrons from the second functional layer during the adsorption process; this behavior will cause the carrier concentration in the second functional layer to increase, the absolute value of the overall current response of the device to increase, and the overall current response change rate to increase.

[0046] Further preferably, the first metal electrode and the second metal electrode are one of Pt, Al, Ti, W, Ag, TiN or Cu; and the barrier layer is one of SiO2, Si3N4, Al2O3, HfO2 or Ta2O5.

[0047] Further preferably, the thickness of the first functional layer is 3nm-10um, the thickness of the barrier layer is 3nm-500nm, the thickness of the second functional layer is 3nm-500nm, and the thickness of the first and second metal electrodes is 3nm-200nm.

[0048] Further preferably, the test gas may be an oxidizing gas or a reducing gas; wherein the oxidizing gas includes O2, SO2, NO2 or H2S; and the reducing gas includes H2 or CO.

[0049] Further preferably, the device is placed in a closed space filled with N2 or an inert gas. When a test gas pulse is introduced into the space, a constant voltage is applied between the first metal electrode and the second metal electrode, and the current response state of the device is read. As the test gas pulse duration increases, the amplitude of the device current change also increases accordingly, demonstrating the device's sensing characteristics for the test gas.

[0050] Further preferably, the device is placed in a closed space filled with N2 or an inert gas. When a test gas pulse is introduced into the space, under the condition that the test gas pulse duration is the same, the greater the gas concentration, the greater the amplitude of the device current change, exhibiting the pulse amplitude response characteristic of the neuromorphic response.

[0051] Further preferably, the device is placed in a closed space filled with N2 or an inert gas, and when multiple consecutive test gas pulses are introduced into the space, the shorter the interval between the test gas pulses, the greater the rate of change of the device current, which manifests the pulse frequency response characteristics of the neuromorphic response.

[0052] On the other hand, the present application provides the above-mentioned CuO x A method for preparing a gas sensing and computing integrated device, the method specifically comprising the following steps:

[0053] S1: Prepare a doped Si layer by doping the melt during single crystal Si growth by Czochralski or zone melting; or grow an intrinsic silicon layer by sputtering, physical vapor deposition, or chemical vapor deposition, and then implant impurities by thermal diffusion or ion implantation to prepare heavily doped n-type Si or p-type Si. After doping, the Si wafers are cut into pieces and planarized using chemical mechanical planarization equipment to form a first functional layer; the thickness of the first functional layer is 3nm to 500um;

[0054] S2: preparing a patterned mask layer of a barrier layer on the first functional layer;

[0055] S3: preparing a barrier layer on the patterned mask layer of the barrier layer, and peeling off the mask layer to form a barrier layer;

[0056] S4: preparing a patterned mask layer of a second functional layer on the barrier layer;

[0057] S5: preparing a second functional layer on the patterned mask layer of the second functional layer, and peeling off the mask layer to form the second functional layer;

[0058] S6: preparing patterned mask layers of a first metal electrode and a second metal electrode on the first functional layer and the second functional layer, respectively;

[0059] S7: Prepare a first metal electrode and a second metal electrode on the patterned mask layer, and peel off the mask layer to form the first metal electrode and the second metal electrode, thereby completing the preparation of the gas sensing and computing integrated device.

[0060] Further preferably, the first metal electrode, the barrier layer, the second functional layer and the second metal electrode are prepared by a method selected from the group consisting of sputtering, physical vapor deposition, chemical vapor deposition, molecular beam epitaxy and electrochemical methods.

[0061] In order to further illustrate the gas sensing and computing integrated device and its preparation method provided in the embodiment of the present application, the embodiment of the present application is described below in conjunction with the drawings in the embodiment of the present application.

[0062] Example 1

[0063] like Figure 1 As shown, the embodiment of the present application provides a gas sensing and computing integrated device with a four-layer planar structure, and the specific preparation process is as follows:

[0064] Si with a crystal orientation of (100) is selected as a seed crystal, and a silicon rod is prepared in molten Si during the single crystal Czochralski method and degenerately doped with B elements. A silicon wafer with a resistivity of 0.005 ohm-cm is obtained by mechanical cutting and chemical mechanical planarization process, thereby obtaining a first functional layer 1; the wafer is cut into 1 cm*1 cm size samples, placed in an acetone solution, and cleaned for 10-15 minutes using an ultrasonic cleaner. The sample is removed from the acetone solution, re-prepared in an ethanol solution, and cleaned for 10-15 minutes using an ultrasonic cleaner, wherein the power of the ultrasonic cleaner is 20 W. After cleaning, it is rinsed with deionized water and blown dry with a nitrogen gun;

[0065] A patterned mask layer for a barrier layer is prepared, a SiO2 layer for the barrier layer is prepared by PECVD, and the mask layer is stripped by acetone; wherein the thickness of the barrier layer is 50 nm;

[0066] Prepare the patterned mask layer of the second functional layer and prepare CuO by magnetron sputtering xThe target material is a Cu target, and sputtering is performed using a DC power supply. During the sputtering process, the thickness of the second functional layer can be adjusted by adjusting the sputtering power and sputtering time. In this embodiment, O2:Ar=32:16, the total gas pressure is 0.5Pa, the power is 100W, and the sputtering time is 3000s to prepare a second functional layer 3 with a thickness of 300nm, and the patterned mask layer is stripped using acetone.

[0067] A patterned mask layer for a metal electrode is prepared, and a Pt first metal electrode and a second metal electrode are prepared by magnetron sputtering. More specifically, a Pt target is used and sputtering is performed using a DC power supply to obtain a first metal electrode and a second metal electrode with a thickness of 100 nm. The patterned mask layer is then stripped using acetone.

[0068] After completing the above steps, the CuO x Preparation of gas sensing and computing integrated device, device structure such as Figure 1 shown.

[0069] Example 2

[0070] In Example 2 of the present application, Aglient 1500A and a gas pulse generator were used to perform DC electrical scanning and gas pulse testing on the gas sensing and computing integrated device.

[0071] Figure 2 This is the DC current-voltage relationship curve obtained by continuously scanning the voltage from 0V to +3V for the gas sensing and computing integrated device provided in an embodiment of the present application; during a single forward voltage scan, the device will show a phenomenon of changing from a high-resistance state to a low-resistance state, indicating that the device has basic memristive characteristics.

[0072] Figure 3 This is the DC current-voltage relationship curve obtained by the gas sensing and computing integrated device provided in the embodiment of the present application under three consecutive voltage scans of 0V to +3V; during the continuous positive pressure scan, the device shows a trend of continuous increase in conductivity, indicating that the device has simulated adjustable conductivity characteristics.

[0073] Figure 4 It is the CuO provided in the embodiment of this application x The gas sensing and computing device is placed in a closed test space filled with N2 or other inert gases. When a NO2 gas pulse is introduced into the space, a constant +1V voltage is applied between the first metal electrode and the second metal electrode, and the current response state of the device is read. As the NO2 gas pulse time increases, the device current value also increases accordingly, showing the device's sensing characteristics for NO2 gas. After the NO2 gas is removed, the current of the device shows a slow decrease process. This indicates that the device exhibits one of the important synaptic behaviors in response to external gas stimulation, the excitatory postsynaptic current (EPSC).

[0074] Figure 5 It is the CuO provided in the embodiment of this application x The gas sensing and computing device is placed in a closed space filled with N2 or other inert gases. When NO2 gas pulses are introduced into the space, a constant +1V voltage is applied between the first metal electrode and the second metal electrode to read the current response state of the device; 5ppm, 10ppm, 20ppm, 30ppm, 40ppm and 50ppm of NO2 gas are introduced respectively; when the gas pulse time is the same, the greater the gas concentration, the greater the device current, showing the pulse amplitude response characteristics of the neuromorphic response, which corresponds to the pulse amplitude dependent plasticity (SADP) behavior characteristics in the synaptic device.

[0075] Figure 6 It is the CuO provided in the embodiment of this application x The gas sensing and computing device is placed in a closed space filled with N2 or other inert gases. When multiple consecutive NO2 gas pulses are introduced into the space, a constant +1V voltage is applied between the first metal electrode and the second metal electrode to read the current response state of the device. The gas pulse width is uniformly set to 200s, with intervals of 200s and 400s respectively, and it is introduced continuously for five times. The shorter the interval between the gas pulses, the greater the rate of change of the device current response, showing the pulse frequency response characteristics of the neuromorphic response, which is the same as the double-pulse differentiation characteristic (PPF) of the artificial synaptic device and the pulse frequency-dependent plasticity (SRDP) in the short-term plasticity of the synapse.

[0076] In summary, compared with the prior art, this application has the following advantages:

[0077] The present application provides a gas sensing and computing integrated device and its preparation method, wherein the first functional layer is made of heavily doped p-type silicon or heavily doped n-type silicon; the second functional layer is made of CuO x Materials; the device has a stacked planar two-terminal structure with a simple preparation process; the device can realize electroresistance switching under DC voltage scanning; the device can realize neuromorphic current response to gas pulse signals under gas pulse excitation, including pulse amplitude dependence characteristics and pulse frequency dependence characteristics, and has gas sensing and computing integrated functions; based on the above structural settings, compared with existing gas sensing and computing integrated devices, it has a simpler structure and a simpler preparation process; and because it can realize neuromorphic response to gas pulse signals, the device has a simpler inference process, and is expected to be used in in-situ gas sensing and computing systems to achieve more complex tasks.

[0078] It should be understood that expressions such as “include” and “may include” used in the present application indicate the existence of disclosed functions, operations, or constituent elements, and do not limit one or more additional functions, operations, and constituent elements.

[0079] In the description of the embodiments of the present application, it should be noted that, unless otherwise specified and limited, the term "connection" should be understood in a broad sense.

[0080] The above description is merely a specific embodiment of the present application, but the scope of protection of the present application is not limited thereto. Any changes or substitutions that can be easily conceived by a person skilled in the art within the technical scope disclosed in this application should be included in the scope of protection of this application. Therefore, the scope of protection of this application should be based on the scope of protection of the claims.

Claims

1. A gas sensing and computing integrated synaptic device, characterized in that: include: a first functional layer, a barrier layer, a second functional layer, a first metal electrode, and a second metal electrode; The second functional layer is divided into a first portion of the second functional layer and a second portion of the second functional layer; the barrier layer is located between the first functional layer and the first portion of the second functional layer, and the second metal electrode is located above the first portion of the functional layer; the second portion of the second functional layer is located above the first functional layer, and the first metal electrode is also located above the first functional layer; The first functional layer is heavily doped silicon, and the second functional layer is CuO x The material is provided with defect energy levels formed by interstitial oxygen and copper vacancies; The blocking layer is used to raise the second functional layer of the first part to increase the contact area between the second functional layer and the test gas; the first metal electrode forms an ohmic contact with the first functional layer; the second metal electrode forms an ohmic contact with the second functional layer; a DC voltage scan is applied between the first metal electrode and the second metal electrode to simulate the resistive switching behavior of the memristor; the second functional layer is used to react with the test gas to realize the sensing function of the test gas; when a constant voltage signal is applied between the first metal electrode and the second metal electrode, the current response state presents a synaptic response in a gas pulse environment; wherein 0.5≤x≤2.

2. The gas sensing and computing integrated synaptic device according to claim 1, characterized in that: The doping impurities in the first functional layer are one or more of B, Ga, In, P, As, and Sb.

3. The gas sensing and computing integrated synaptic device according to claim 1 or 2, characterized in that: The first metal electrode and the second metal electrode are one of Pt, Al, Ti, W, Ag, TiN or Cu; the barrier layer is one of SiO2, Si3N4, Al2O3, HfO2 or Ta2O5.

4. The gas sensing and computing integrated synaptic device according to claim 3, characterized in that: The thickness of the first functional layer is 3nm-10um, the thickness of the barrier layer is 3nm-500nm, the thickness of the second functional layer is 3nm-500nm, and the thickness of the first and second metal electrodes is 3nm-200nm.

5. The gas sensing and computing integrated synaptic device according to claim 1, characterized in that: The test gas includes an oxidizing gas or a reducing gas; wherein the oxidizing gas includes O2, SO2, NO2 or H2S; and the reducing gas includes H2 or CO.

6. The gas sensing and computing integrated synaptic device according to claim 1, characterized in that: Synaptic responses include excitatory postsynaptic currents, pulse amplitude-dependent plasticity, paired-pulse differentiation, and pulse frequency-dependent plasticity.

7. A synaptic response method based on the gas sensing and computing integrated synaptic device according to claim 1, characterized in that: The following steps are involved: The gas-sensing and computing integrated synaptic device is placed in a confined space filled with N2 or an inert gas, a test gas pulse is introduced into the confined space, a constant voltage is applied between the first metal electrode and the second metal electrode, and the current response state of the gas-sensing and computing integrated synaptic device is read; The changes in the current response state as the test gas pulse time increases are analyzed to obtain the sensing characteristics of the gas sensing and computing synaptic device for the test gas.

8. A synaptic response method based on the gas sensing and computing integrated synaptic device according to claim 1, characterized in that: Specifically: The gas sensing and computing integrated synaptic device is placed in a closed space filled with N2 or inert gas. When a test gas pulse is introduced into the sealed space, the current response state is analyzed as the test gas concentration increases under the condition of the same test gas pulse time, and the pulse amplitude response characteristics in the neuromorphic response are obtained.

9. A synaptic response method based on the gas sensing and computing integrated synaptic device according to claim 1, characterized in that: Specifically, the gas sensing and computing integrated synaptic device is placed in a closed space filled with N2 or inert gas, and multiple continuous test gas pulses are introduced into the closed space. The current response state is analyzed as the interval between the test gas pulses shortens, and the pulse frequency response characteristics in the neuromorphic response are obtained.

10. A method for preparing the gas sensing and computing integrated synaptic device according to claim 1, characterized in that: The specific steps include: S1: Prepare a doped Si layer and planarize it to obtain the first functional layer; S2: preparing a patterned mask layer of a barrier layer on the first functional layer; S3: preparing a barrier layer on the patterned mask layer of the barrier layer, and peeling off the mask layer to form a barrier layer; S4: preparing a patterned mask layer of a second functional layer on the barrier layer; S5: preparing a second functional layer on the patterned mask layer of the second functional layer, and peeling off the mask layer to form the second functional layer; S6: preparing patterned mask layers of a first metal electrode and a second metal electrode on the first functional layer and the second functional layer, respectively; S7: preparing a first metal electrode and a second metal electrode on the patterned mask layer of the first metal electrode and the second metal electrode, and peeling off the mask layer to form the first metal electrode and the second metal electrode, thereby completing the preparation of the gas sensing and computing integrated device; The first metal electrode, the barrier layer, the second functional layer and the second metal electrode are prepared by a method selected from the group consisting of sputtering, physical vapor deposition, chemical vapor deposition, molecular beam epitaxy and electrochemical methods.

Images