Stretching-pain sensation sensing system based on multi-gate synaptic transistor and preparation method and application of stretching-pain sensation sensing system

Abstract

The invention relates to a stretching-pain perception system based on a multi-gate synaptic transistor and a preparation method and application thereof, and belongs to the technical field of artificial sensory neuron and neuromorphic calculation. The system comprises a crack type strain sensor and a multi-gate synaptic transistor, the strain sensor adopts a metal / elastic composite material double-layer conductive structure, and has high sensitivity and a continuous conductive path; the multi-gate synaptic transistor takes an electrolyte material as a gate medium, takes a metal oxide as an electrode and a channel, and is provided with a bottom gate and a side gate. The strain sensor is connected with the bottom grid and the side grid through a voltage division circuit, and dynamic integration and processing of stretching signals are achieved by regulating and controlling different grid voltages. According to the invention, multi-stage mode conversion from conventional stretching perception to pain early warning is realized, the method is suitable for biological nerve simulation systems such as electronic skin, intelligent artificial limbs and bionic robots, and the bottleneck problem that an existing pain perception system is difficult to realize multi-signal fusion and dynamic regulation and control is solved.

Description

Technical Field

[0001] This invention belongs to the field of application technology of artificial sensory neurons, specifically relating to a stretching-pain perception system based on multi-gate synaptic transistors, its preparation method, and its application in simulated biological nervous systems. Background Technology

[0002] Pain perception is a core defense mechanism for biological organisms to avoid injury and ensure safety. When an injury occurs, the human body's pain perception is not solely triggered by a specific nerve transmission, but rather involves the dynamic coordination of excitatory nerves (responsible for initiating and transmitting pain signals) and inhibitory nerves (responsible for filtering, regulating, and blocking pain signals) at the three levels of "periphery-spinal cord-brain": the former ensures that the injury stimulus is not missed (ensuring survival), and the latter avoids excessive pain response (reducing unnecessary suffering). The balance between the two directly determines the intensity, nature, and duration of pain.

[0003] Current mainstream artificial pain perception solutions fall into two main categories: pain processing circuits based on sensors (such as resistive and capacitive sensors) and artificial neurons based on synaptic devices (memristors and transistors). However, traditional sensor electronics struggle to simulate the dynamic response characteristics of biological pain, such as adjustable thresholds, pain sensitization / desensitization, and self-protection mechanisms. Furthermore, conventional two-terminal devices, such as memristors, and three-terminal devices, such as transistors, mean that each synaptic device can only process one input voltage signal, resulting in a limited and singular selection of external stimuli and an inability to effectively simulate the dynamic coordination of excitation and inhibition in the human body when experiencing pain.

[0004] To address the aforementioned technical challenges, this invention, developed through systematic research and development, specifically targets pain perception during stretching. This technology aims to promote the design and practical application of neuromorphic electronic skin in the fields of artificial intelligence robotics and rehabilitation, playing a crucial role in areas such as self-powered AI electronic skin and human-computer interaction. Summary of the Invention

[0005] To address the problem that existing biomimetic pain sensing systems can only achieve sensor signal modulation of a single neural pathway and cannot realistically reproduce the pain regulation and perception judgment during human stretching, this invention provides a stretching-pain sensing system based on multi-gate synaptic transistors, its fabrication method, and its application. The stretching-pain sensing system can adjust the signal output in real time through a front-end strain sensor to achieve human-like pain perception judgment at the device end.

[0006] The technical solution adopted in this invention is: A stretching-pain sensing system based on multi-gate synaptic transistors includes N crack-type strain sensors for sensing external stretching / bending changes and multi-gate synaptic transistors for processing N input signals; N is an even number; the crack-type strain sensors adopt a conductive double-layer structure, ensuring a continuous conductive path throughout while possessing high sensitivity; the multi-gate synaptic transistors are ion field-effect transistors using an electrolyte material as the gate dielectric and a metal oxide material as the electrode and channel, having a vertical bottom gate that dominates the switching of the device and a coplanar side gate that assists in adjusting the conductivity of the device; each pair of strain sensors is connected to the bottom / side gate of the multi-gate transistor through a voltage divider circuit.

[0007] Furthermore, the strain sensor includes a primary conductive layer and a secondary conductive layer. The primary conductive layer is a metal crack conduction layer, and the material is selected from gold, platinum, and silver. The secondary conductive layer is a conductive elastic substrate composite material made of one of the following materials: multi-walled carbon nanotubes, reduced graphene oxide, and carbon black, mixed with polydimethylsiloxane.

[0008] Furthermore, the gate dielectric electrolyte material used in the multi-gate synaptic transistor is one of the following biomaterials: chitosan, a mixture of polyethylene and graphene oxide, or sodium alginate. The metal oxide materials used for the electrodes and channels are semiconductor oxides such as indium tin oxide, indium gallium zinc oxide, or indium tungsten oxide.

[0009] Furthermore, in the voltage divider circuit, the strain sensor connected to the bottom gate receives external tensile stimulation and regulates the positive voltage of the input bottom gate by changing its own resistance, thereby dominating the conductivity at the channel. The strain sensor in the voltage divider circuit that divides the voltage with the side gate receives the same tensile stimulation and regulates the negative voltage of the input side gate, forming a local reverse electric field to assist in regulating the overall conductivity of the device. The absolute value of the total positive voltage of the bottom gate circuit is proportional to the absolute value of the total negative voltage of the side gate circuit.

[0010] Furthermore, when the stretching stimulus is less than the warning threshold of the stretch-pain perception system, the system responds normally with a stretching intensity signal; when the stretching stimulus is greater than the pain warning threshold of the stretch-pain perception system but less than the damage threshold, the system outputs a pain perception response; when the stretching stimulus is greater than the damage threshold of the stretch-pain perception system, the system's pain response output is further enhanced, and the stretch-pain perception system is damaged.

[0011] Furthermore, when the stretching stimulus does not reach the pain warning threshold, the system's on / off time is consistent with the stimulation duration of the strain sensor, and a stable response current is output; when the stretching stimulus is greater than the pain warning threshold of the sensing system but less than the damage threshold, the intensity of the system output signal slowly increases with the extension of the stretching time, and when the stimulus signal disappears, the sensing system can still maintain a conducting state for a period of time, the duration of which is proportional to the intensity of the stretching stimulus; when the stretching stimulus is greater than the damage threshold of the sensing system, the intensity of the system output signal rapidly increases with the extension of the stretching time until it reaches saturation, and when the stimulus signal disappears, a significant signal hold can be observed.

[0012] This invention also proposes a method for fabricating the above-mentioned stretch-pain sensing system based on multi-gate synaptic transistors, comprising the following process steps: S1: Mix one of the following materials, namely multi-walled carbon nanotubes, reduced graphene oxide, and carbon black, with polydimethylsiloxane to obtain a conductive elastic substrate composite material. Use a solvent and a curing agent to fully mix with the elastic substrate material and pour it into a template for curing to obtain a secondary conductive layer. S2: After hydrophilizing the secondary conductive layer obtained in step S1, a metal conductive layer is deposited by magnetron sputtering to obtain the main conductive layer, thereby obtaining a crack strain sensor. S3: Spin-coat the gate dielectric electrolyte material onto the cleaned conductive substrate, let it stand and dry to form a film, and then use it as the insulating gate dielectric layer of the synaptic device. S4: Deposit the source, drain, and side gate electrodes of the synaptic device on the gate dielectric layer obtained in step S3 using magnetron sputtering and a mask; control the distance between the mask and the gate dielectric layer to form a channel using diffraction. S5: The strain sensor and the multi-gate synaptic transistor are connected by wires to form an artificial pain perception system.

[0013] Preferably, the solvent in step S1 is ethyl acetate or isopropanol; step S1 specifically involves: placing 8% to 15% of the conductive composite material by mass in a stirring table at room temperature and stirring at a speed of 400 r / min to 600 r / min for 30 to 40 minutes, then curing it at 60 to 70°C; and finally using magnetron sputtering to uniformly cover the metal of the main conductive layer onto the surface of the secondary conductive layer.

[0014] Preferably, the electrolyte material in step S3 is one of chitosan solution, methylcellulose, and graphene oxide; the spin coating process is set as follows: first, the spin coating speed is set to 400 r / min to 600 r / min and the time is set to 15-20 s, and the prepared electrolyte material is fully drop-coated onto the surface of the conductive substrate. Then the speed is changed to 1800 r / min to 2000 r / min until the electrolyte material uniformly covers the substrate; then it can be placed in a clean room at room temperature for 24 h to 30 h to form a film.

[0015] Preferably, in step S4 above, the material used for depositing the electrode is a semiconductor material such as indium tin oxide, indium gallium oxide, or indium gallium zinc oxide. All of the above electrodes can be fabricated using magnetron sputtering.

[0016] This invention proposes the application of the above-mentioned stretch-pain perception system based on multi-gate synaptic transistors in simulated biological nervous systems, including electronic skin, intelligent prostheses, or bionic robots.

[0017] Invention Principle: The multi-gate transistors fabricated using this method are thin-film field-effect transistors (TFTs). Under the influence of a gate electric field, free ions in the gate dielectric move directionally according to their polarity. When the applied electric field strength is low, the free ions, after their directional movement, accumulate at the electrolyte / channel interface or the electrolyte / gate electrode interface, forming a double-layer phenomenon. This gives the gate dielectric a high specific capacitance, thus affecting the transistor's conductivity. Because the gate dielectric of a double-layer TFT has good ionic conductivity, based on this characteristic, when a forward voltage is applied to the bottom electrode, free ions continuously accumulate at the electrolyte / channel interface, forming a double layer and generating a capacitive coupling effect. Conversely, when a reverse voltage is applied to the planar side electrode, some ions are attracted to move towards the electrolyte / side gate electrode interface, thus achieving a stable dynamic equilibrium at the electrolyte / channel interface. Therefore, the transistor in this state mainly responds to the excitation electric field strength, exhibiting low time sensitivity. Because the distance between the bottom electrode and the channel is much smaller than the distance between the side electrode and the channel, as the electric field strength gradually increases, the cumulative ion effect at the electrolyte / channel interface is much higher than the ion attraction efficiency of the side gate electrode. Furthermore, free ions have a chance to enter defects in the semiconductor material, forming a doped state that further enhances the conductivity at the channel. At this point, the transistor exhibits a significant time dependence. Based on this principle, the channel resistance and output characteristics of a multi-gate transistor are affected by ion migration and charge trapping. This characteristic is similar to the transmission of neurotransmitters in biological nerve conduction, thus this invention is applied to the biomimetic field of simulating human nerve signal transmission. In particular, since the voltage input to the gate electrode forms a point charge electric field, the distance between the field source and the channel interface determines its ability to dominate free ions. Based on this principle, this invention is extended to multi-gate devices with different gate structures to handle different signal weight selection and integration, achieving phased and differentiated processing of stimulus signals at the device end, breaking through the limitation of traditional artificial pain perception systems that can only handle one sensing mode.

[0018] Compared with existing pain perception systems, the advantages of this invention are reflected in: (1) The present invention constructs an artificial pain perception system based on resistive strain sensors and multi-gate field-effect transistors, realizing the effective coupling of multiple sensors with the same synaptic device, with high system integration and the ability to respond to multi-source stimuli and fuse signals.

[0019] (2) By adding a sensor and a reverse voltage divider structure to the side gate, the function of the human intermediate inhibitory neuron is innovatively simulated, and the gate weight distribution and conductivity characteristics of the multi-gate transistor are fully utilized to realize multi-scale analysis and dynamic control of stretching stimulation.

[0020] (3) Apply positive and reverse external voltages to the bottom electrode and the side grid electrode respectively, and stretch the voltage divider sensor. By adjusting the ratio of the external voltage applied to the side grid electrode and the external voltage applied to the bottom grid electrode, the threshold of the pain perception system can be controlled, thereby greatly expanding the application potential of the pain perception system.

[0021] (4) The system can not only reproduce the core plasticity of traditional pain synapses, including short-term plasticity, long-term plasticity, and double-pulse facilitation, but also achieves a breakthrough in the automatic phased conversion of pain threshold calibration and signal processing mode, fully simulating the key behavioral characteristics of biological pain receptors.

[0022] (5) This system can simulate the "excitation-inhibition" synergistic mechanism of biological pain perception, outputting a stable response when the stretching stimulus is below the threshold, and exhibiting pain sensitization and signal enhancement behavior when the stimulus exceeds the threshold. It also possesses neuromorphic characteristics such as short-term / long-term plasticity and double-pulse facilitation. This invention realizes a multi-stage mode conversion from conventional stretching perception to pain warning, and is applicable to biological neural simulation systems such as electronic skin, intelligent prostheses, and bionic robots. It solves the bottleneck problem of existing pain perception systems being unable to achieve multi-signal fusion and dynamic control.

[0023] In summary, this invention solves the problem that traditional artificial pain systems struggle to simultaneously simulate and dynamically regulate multi-dimensional pain sensations, providing a new approach and feasible solution for constructing a perception system with biological neuromorphic characteristics. Attached Figure Description

[0024] Figure 1 This is a schematic diagram of the system structure of the multi-gate synaptic transistor-based stretching-pain perception system of the present invention; Figure 2 This is a circuit design schematic diagram of the multi-gate synaptic transistor-based stretching-pain perception system of the present invention; Figure 3 This is a schematic diagram of the crack strain sensor structure of the present invention. Figure 4 This is a schematic diagram of the multi-gate synaptic transistor structure of the present invention; Figure 5 This is a comparison diagram of the output current of pain perception systems with and without side-gate suppression input modulation. Figure 5 (a) is a comparison of peak current output under instantaneous intensity stimulation. Figure 5 (b) is a comparison chart under continuous stretching; Figure 6 To simulate the relationship between pain output and stretching length during continuous stretching in a pain perception system; Figure 7 To simulate the relationship between pain output and stretching length changes during multiple stretching operations in a pain perception system. Detailed Implementation

[0025] The present invention will be further described in detail below with reference to the accompanying drawings and specific embodiments.

[0026] The accompanying drawings are merely schematic diagrams intended to illustrate the technical solution of the present invention. The thickness of each thin film layer and the shape and size of each region do not reflect the actual proportional relationship of the device.

[0027] Specific embodiments of the present invention are as follows: The present invention provides a structure for a stretch-pain perception system based on a multi-gate synaptic transistor, such as... Figure 1 It includes two sensing elements for detecting external stimuli and a synaptic transistor for signal processing, all three connected by wires. The resulting output current is read and written to the observation interface through the source measurement unit (SMU).

[0028] Among them, the sensing element used to detect external tensile stimuli is a crack strain sensor, whose structure is a sandwich structure of elastic substrate / secondary conductive layer / main conductive layer, such as... Figure 3 As shown. In this embodiment, PDMS is used as the substrate, multi-walled carbon nanotube / PDMS composite material is used as the secondary conductive layer, and platinum metal is used as the main conductive layer. The overall thickness of the device after fabrication is 2 mm.

[0029] The signal processing element used is a field-effect synaptic transistor, and its structure is as follows: Figure 4 As shown, chitosan is used as the gate dielectric material, and indium tin oxide is used as the channel and electrode material.

[0030] The fabrication method of the multi-gate synaptic transistor-based stretch-pain perception system structure in the above example is as follows: (1) Fabrication of crack strain sensor Step 1: At room temperature, weigh 0.5g of multi-walled carbon nanotubes, 5g of PDMS, 0.5g of curing agent, and 10g of ethyl acetate according to the specified ratio. Place them on a stirring table and stir thoroughly at 400r / min for 30min. After ultrasonic vibration treatment, pour the mixture into a pre-set mold and cure at 60℃ for 1h. Then, take 5g of PDMS and evenly coat it on the surface of the cured substrate, place it in a vacuum environment to remove air bubbles, and then place it in a 60℃ environment again for 1h of curing. The curing agent used is methylhydrosiloxane-dimethylsiloxane copolymer.

[0031] Step 2: After the sample has completely solidified, remove it and use magnetron sputtering to uniformly deposit platinum metal onto the surface of the multi-walled carbon nanotube / PDMS composite conductive material to form a metal layer.

[0032] (2) Fabrication of field-effect synaptic transistors Step 3: Spin-coat a 5% (w / w) chitosan ionomer film onto the cleaned conductive substrate. Spin-coat the film at 600 rpm for 15 seconds, then spin-coat it at 1200 rpm for 20 seconds. Finally, let it stand at room temperature in a cleanroom for 24 hours to cure the film.

[0033] Step 4: Using magnetron sputtering and a mask, the channel of the synaptic device, as well as the source, drain, and coplanar gate electrodes, are deposited on the cured chitosan film. Both the electrode and channel materials are indium tin oxide (ITO) semiconductor materials. The resulting channel width is approximately 80 μm, the electrode width is approximately 150 μm, and the length of each electrode is 1000 μm.

[0034] (3) Constructing an artificial stretching-pain perception system Step 5: Take several wires and two fixed resistors to build a voltage divider circuit. The specific connection method is as follows: connect one end of the wire to an external power supply, and connect the other end to the strain sensor through the fixed resistor to form a series closed loop. Then connect the strain sensor and the corresponding gate probe of the field-effect transistor in parallel for voltage division. With the assistance of an optical microscope, accurately attach the probe to the gate of the device to complete the overall construction of the artificial stretching-pain sensing system.

[0035] An artificial tension-pain sensing system (N is an even number) comprising N strain sensors is constructed. The input voltage of each gate is dynamically controlled by adjusting the equivalent resistance of the parallel sensor network. Simultaneously, a microcontroller is used to acquire Rs connected in series in the source-drain circuit. C3 The real-time voltage across the channel is indirectly calculated by measuring the voltage across the fixed resistor. This solution is suitable for closed-loop feedback applications involving multi-sensor fusion, such as smart gloves, and enables adaptive adjustment of sensing signals and execution control.

[0036] Figure 2 This is a circuit design schematic diagram of the multi-gate synaptic transistor-based stretching-pain perception system of the present invention. Explanation of reference numerals: V GS1 The bottom gate constant voltage; V GS2 The side gate constant voltage; V DS The source-drain voltage is constant; R C R is a fixed resistor. sensor This is the resistance corresponding to the strain sensor.

[0037] Under the regulation of the side-gate inhibitory current, the artificial pain sensing system of this invention can automatically switch signal processing modes. As shown in Figure 5, when the system is under low tensile strength conditions, the signal processed by the side-gate inhibitory current is less sensitive to time superposition characteristics compared to the case without side-gate inhibitory current, and can output tensile strength feedback closer to the true value at any time point. As the tensile strength gradually increases, the signal output capability with and without side-gate inhibitory current tends to be consistent. At this time, the system again exhibits significant time-dependent characteristics, and its output signal can reflect both the intensity of the external stimulus and the duration of the stimulus. This invention defines the coincidence point of the signal output with and without side-gate inhibitory current as the pain threshold of the pain sensing system, and the precise control of this pain threshold can be achieved by adjusting the external voltage parameters of the side-gate circuit.

[0038] When the human body exceeds the pain threshold at a specific time point due to continuous or repetitive mechanical loading, it marks a critical shift in tissue homeostasis from physiological adaptation to pathological warning. This process involves the synergistic effect of peripheral sensitization and central regulation: inflammatory mediators released from micro-injuries in tissues lower the activation threshold of nociceptors, while enhanced transmission of afferent signals through the dorsal horn synapses of the spinal cord induces central sensitization. The motor control system then initiates compensatory reorganization, manifested as reflexive inhibition of α-motor neurons at the spinal cord level, alterations in the synergistic activation patterns of antagonistic muscles, and real-time correction of the motor program by the cerebellar-cortical pathway. This change in neuroplasticity driven by nociceptive input essentially constitutes a multi-level negative feedback defense mechanism, which forcibly terminates motor behaviors that may lead to further damage by dynamically reconstructing the excitation-inhibition balance, thereby maintaining the integrity of the biological system. Therefore, in order to fully reproduce the entire process of the human body radiating protective pain signals to the brain in a sub-damage margin state during excessive and continuous exercise, we subjected strain sensors to varying degrees of continuous and repeated stretching, and observed the output signal characteristics of the artificial stretching-pain perception system.

[0039] Figure 6This study presents the signal processing mode switching behavior of an artificial stretch-pain sensing system when strain sensors are subjected to progressively increasing continuous stretching stimulation. In the experiment, two strain sensors were connected to the bottom and side gates of the device, respectively, in a voltage divider configuration, maintaining consistent deformation. The tensile strain was gradually increased from 5% to 50% (each tensile strength lasted 10 seconds, with a strain step size of 5%). During this process, the system's time-dependent characteristics were observed to gradually increase with increasing strain. This is because under relatively small tensile strain, local cracks appear in the conductive metal layer, increasing its resistance and leading to a continuous and relatively stable external input voltage at the gate in the parallel voltage divider structure. This drives a certain number of ions to migrate at the channel-electrode interface, forming a dynamically balanced capacitive coupling behavior. When the tensile strain continues to increase and exceeds a certain deformation threshold, the output current intensity gradually increases with the extension of the stretching time, and its increase intensifies with increasing strain. At this stage, the strong potential obtained by the bottom gate significantly weakens or even completely inhibits the modulating effect of the side gate, causing the system to exhibit behavior similar to "pain sensitization" after exceeding a threshold in biological pain sensing systems. This phenomenon marks a crucial stage in the system's transition from a conventional stretching perception mode to a pain warning mode.

[0040] Figure 7 This refers to the signal output of an artificial stretch-pain perception system under repeated stretching stimulation. The deformation of the two strain sensors is kept consistent (where S...). E and S I The strain sensors (representing the tensile deformation of the bottom and side gate voltages, respectively) were subjected to 10 stretching cycles at the same frequency with intensities of 10%, 30%, and 50%, respectively. The morphological transition mode of the artificial stretch-pain sensing system remained intact under high-frequency conditions. Due to the relaxation characteristics during ion migration, ions require a certain amount of time to return to their initial positions; therefore, high-frequency voltage input leads to facilitated enhancement of the output current signal. Thanks to the unique dynamic signal processing of synaptic devices, it can be observed that the artificial stretch-pain sensing system can also produce a biological-like sensory pattern transition when processing high-frequency stretching stimuli. The described multi-gate synaptic transistor-based stretch-pain sensing system can be applied in simulated biological neural afferent systems, such as electronic skin, tactile receptors, or biomimetic robots.

[0041] The embodiments described above are merely specific implementations of the present invention, used to illustrate the technical solutions of the present invention in detail, and are not intended to limit the scope of protection of the present invention. Any modifications, equivalent substitutions, improvements, etc., made within the framework of the principles of the present invention should be included within the scope of the claims of the present invention.

Claims

1. A multi-gate synapse transistor based stretch-pain perception system, characterized by, The application relates to a stretch-sensing system, which comprises N crack strain sensors for sensing external stretch / bending changes and a multi-gate synaptic transistor for processing N input signals; N is an even number; the crack strain sensor adopts a conductive double-layer structure, which ensures full continuous conductive paths while having high sensitivity; the multi-gate synaptic transistor is an ion field effect transistor with an electrolyte material as a gate dielectric and a metal oxide material as electrodes and a channel, and has a vertical structure bottom gate which dominates the on-off of the device and a coplanar structure side gate which assists in adjusting the conductivity of the device; each group of paired strain sensors is connected to the bottom gate and the side gate of the multi-gate synaptic transistor through a voltage dividing circuit.

2. The multi-gate synapse transistor based stretch-pain perception system of claim 1, wherein, The strain sensor comprises a primary conductive layer and a secondary conductive layer; the primary conductive layer is a metal crack conductive layer, and the material is selected from gold, platinum and silver; and the secondary conductive layer is a conductive elastic base composite material mixed with one of multi-walled carbon nanotubes, reduced graphene oxide and carbon black and polydimethylsiloxane.

3. The multi-gate synapse transistor based stretch-pain perception system of claim 1, wherein, The gate dielectric electrolyte material of the multi-gate synaptic transistor is one of a chitosan, a polyethylene oxide graphene oxide mixture and sodium alginate; and the metal oxide material of the electrode and channel material is selected from indium tin oxide, indium gallium zinc oxide and indium tungsten oxide.

4. The multi-gate synapse transistor based stretch-pain perception system of claim 1, wherein, The strain sensor connected to the bottom gate in the voltage dividing circuit receives external stretch stimulation, changes the resistance thereof to regulate the positive voltage input to the bottom gate, and thus dominates the conductivity at the channel; the strain sensor connected to the side gate in the voltage dividing circuit receives the same stretch stimulation, regulates the negative voltage input to the side gate, forms a local reverse electric field, and assists in regulating the overall conductivity of the device; and the absolute value of the total positive voltage of the bottom gate circuit and the total negative voltage of the side gate circuit is proportional to each other.

5. The multi-gate synapse transistor based stretch-pain perception system of claim 4, wherein, When the stretch stimulation is less than the warning threshold of the stretch-pain perception system, the system normally responds to output a stretch intensity signal; when the stretch stimulation is greater than the pain warning threshold but less than the damage threshold of the stretch-pain perception system, the system makes a pain perception output reaction; when the stretch stimulation is greater than the damage threshold of the stretch-pain perception system, the pain reaction output of the system is further enhanced, and the stretch-pain perception system is damaged.

6. The multi-gate synapse transistor based stretch-pain perception system of claim 5, wherein, When the stretch stimulation does not reach the pain warning threshold, the on-off time of the system is consistent with the stimulated time length of the strain sensor, and a stable response current is output; when the stretch stimulation is greater than the pain warning threshold but less than the damage threshold of the perception system, the intensity of the output signal of the system is slowly enhanced with the extension of the stretch time, and when the stimulation signal disappears, the perception system can still maintain a conduction state for a period of time, and the time length is proportional to the stretch stimulation intensity; when the stretch stimulation is greater than the damage threshold of the perception system, the intensity of the output signal of the system is rapidly enhanced with the extension of the stretch time until saturation is reached, and obvious signal retention can be observed when the stimulation signal disappears.

7. The method for fabricating a stretch-pain sensing system based on a multi-gate synaptic transistor as described in any one of claims 2-6, characterized in that, The application further discloses a stretch-sensing system manufacturing method, which comprises the following process steps: S1: one of the multi-walled carbon nanotubes, reduced graphene oxide, carbon black material is mixed with polydimethylsiloxane to obtain a conductive elastic substrate composite material, a dissolving agent and a curing agent are used to mix with the elastic substrate material and poured into a mold for curing to obtain a secondary conductive layer; S2: after the secondary conductive layer obtained in step S1 is subjected to hydrophilization treatment, a metal conductive layer is deposited by a magnetron sputtering process to obtain a primary conductive layer, thereby obtaining a crack type strain sensor; S3: spin coating a gate dielectric electrolyte material on a cleaned conductive substrate, and after standing and drying to form a film, the material is used as an insulating gate dielectric layer of a synaptic device; S4: depositing source, drain conductive electrodes and side gate electrodes of a synaptic device on the gate dielectric layer obtained in step S3 by magnetron sputtering and a mask plate; controlling the distance between the mask plate and the gate dielectric layer to form a channel by using diffraction phenomenon; S5: connecting the strain sensor and the multi-gate synaptic transistor through a wire and an external fixed resistance to form an artificial pain perception system.

8. The method of claim 7, wherein the method further comprises the step of: The dissolving agent in step S1 is ethyl acetate or isopropyl alcohol; step S1 specifically includes: placing 8-15% by mass of the conductive composite material on a constant temperature stirring table, setting the stirring speed to 400-600 r / min, and stirring for 30-40 min, and then curing and forming at 60-70°C; then a magnetron sputtering process is used to uniformly cover the metal of the primary conductive layer on the surface of the secondary conductive layer.

9. The method of claim 7, wherein the method further comprises the step of: The electrolyte material in step S3 is one of chitosan solution, methyl cellulose and graphene oxide; the spin coating process is set as follows: first, set the spin coating speed to 400-600 r / min for 15-20 s, drop coat the prepared electrolyte material on the surface of the conductive substrate, and then change the speed to 1800-2000 r / min until the electrolyte material uniformly covers the substrate; then it can be placed in a super clean room at room temperature for 24-30 h to form a film. ​ 10. Use of the multi-gate synapse transistor based stretch-pain perception system according to claims 1-6 in simulating a biological nervous system, characterized in that, The biological nervous system includes an electronic skin, an intelligent prosthesis or a bionic robot.

Images