A PEDOT:PSS / IGZO optoelectronic synapse with nociceptor function and a preparation method and application thereof

Abstract

The application provides a PEDOT:PSS / IGZO optoelectronic neurosynaptic memristor with nociceptor function and a preparation method and application thereof. The memristor comprises a bottom electrode, a resistance change function layer and a top electrode arranged in sequence, wherein the resistance change function layer is a double-layer heterojunction structure composed of PEDOT:PSS and an IGZO thin film, the IGZO is indium gallium zinc oxide, and the IGZO thin film is arranged on the bottom electrode. The memristor has the characteristics of simulating a biological neurosynapse and nociceptor function, and can balance the electrical control precision and the light response capability.

Description

Technical Field

[0001] This invention relates to the fields of semiconductor microelectronic devices and artificial intelligence technology, and more specifically, to a PEDOT:PSS / IGZO opto-neural synaptic memristor with nociceptor function, its preparation method, and its application. Background Technology

[0002] As Moore's Law approaches its physical limits, the "memory wall" problem caused by the separation of computing and storage units in the traditional von Neumann architecture is becoming increasingly prominent. The high latency, bandwidth limitations, and significant energy consumption resulting from frequent data transfers between memory and processors are no longer suitable for memory-intensive applications such as artificial intelligence. Against this backdrop, neuromorphic computing, with its core advantage of simulating the synaptic connections and information processing methods of biological neural systems, has become an important direction for breaking through the current computing paradigm. Its distributed storage and parallel processing mechanisms can significantly improve system energy efficiency. Artificial synaptic memristors, as a key functional unit in this field, possess biological synaptic plasticity characteristics such as long-term potentiation (LTP), long-term inhibition (LTD), and double-pulse facilitation (PPF), effectively simulating the brain's information storage and dynamic regulation mechanisms.

[0003] Nociceptors, as key sensing units in the human peripheral nervous system, can accurately detect harmful signals such as mechanical damage, extreme temperatures, and chemical stimuli. They utilize core mechanisms like threshold response, sensitization, and maladaptation to construct the body's active defense system to prevent further damage. However, traditional complementary metal-oxide-semiconductor (CMOS) devices typically require the integration of multiple modules and the construction of high-power circuits to achieve these complex biomimetic functions, making it difficult to meet practical application requirements such as flexibility, low power consumption, and biocompatibility. Therefore, developing novel neural synaptic devices with biomimetic sensing capabilities has become a crucial path to the practical application of biomimetic sensing systems. Memristors, due to their simple structure, adjustable dynamic resistance, and low power consumption, are considered ideal candidates for simulating biological nociceptors. Therefore, developing novel optoelectronic neural synaptic memristors that combine the plasticity of biological synapses with the integrated capabilities of nociceptors has become a core requirement for overcoming current computational efficiency bottlenecks and building a highly efficient and intelligent next-generation hardware system, possessing irreplaceable significance. In summary, traditional single-material memristors struggle to simultaneously achieve precise electrical control, photoresponse capability, and biomimetic sensing functionality. Summary of the Invention

[0004] In view of this, the present invention proposes a PEDOT:PSS / IGZO photoelectric synaptic memristor with nociceptor function, its preparation method and application. The memristor is prepared by using a sol-gel method combined with spin coating technology. The prepared photoelectric memristor has excellent nociceptor function and biomimetic characteristics of neural synapses.

[0005] To achieve the above objectives, this invention proposes a PEDOT:PSS / IGZO photoelectric synaptic memristor with nociceptor function, comprising: A bottom electrode, a resistive switching functional layer, and a top electrode are sequentially arranged, wherein the resistive switching functional layer is a bilayer heterojunction structure composed of PEDOT:PSS and IGZO thin films, wherein IGZO is indium gallium zinc oxide, and the IGZO thin film is disposed on the bottom electrode.

[0006] Optionally, the bottom electrode is FTO.

[0007] Optionally, the top electrode is Au.

[0008] On the other hand, the present invention provides a method for fabricating the above-mentioned memristor, comprising: The conductive substrate is cleaned and dried, wherein the conductive substrate is used as the bottom electrode; The IGZO precursor solution was spin-coated onto a substrate to form a wet film. The wet film is pre-dried by heating and then annealed to produce an IGZO thin film. PEDOT:PSS solution was spin-coated onto IGZO film to obtain PEDOT:PSS / IGZO bilayer film; The PEDOT:PSS / IGZO bilayer film was heated, dried, and then annealed to form a resistive switching functional layer. A memristor is fabricated by preparing a top electrode on the surface of a resistive switching functional layer using an ion sputtering process.

[0009] Optionally, the IGZO precursor solution is prepared using a sol-gel method, the preparation process including: Indium nitrate hydrate, gallium nitrate hydrate, and zinc acetate dihydrate were dissolved in ethylene glycol methyl ether, and ethanolamine was added as a stabilizer to obtain a mixed solution. The mixed solution was heated and stirred, and the stirred solution was aged to obtain the IGZO precursor solution.

[0010] Optionally, the atomic molar ratio of indium, gallium, and zinc in the IGZO precursor solution is 2:1:1, and the molar concentration is 0.2M.

[0011] Optionally, the wet film is pre-dried on a heating table at 150°C for 20 minutes and then annealed in an annealing furnace at 350°C for 1 hour.

[0012] Optionally, the PEDOT:PSS / IGZO bilayer film is dried on a heating stage at 100°C for 10 minutes, and then annealed at 90°C for 30 minutes to form a resistive switching functional layer.

[0013] On the other hand, the present invention provides an application of the above-mentioned memristor, which is used in artificial nociceptors and neuromorphic computing systems.

[0014] This invention proposes a PEDOT:PSS / IGZO photoelectric synaptic memristor with nociceptor function and its fabrication method. Its advantages over existing technologies are: (1) The memristor prepared by the present invention innovatively adopts a double-layer heterojunction structure composed of PEDOT:PSS and IGZO thin films in its resistive switching functional layer; PEDOT:PSS and IGZO are typical representatives of organic-inorganic materials. By utilizing their unique physicochemical properties and heterojunction synergistic effect, a synaptic memristor with both photoelectric dual control capability and damage perception function was prepared. (2) The PEDOT:PSS / IGZO memristor of the present invention can reproduce the core functions of biological synapses under voltage pulse excitation, including double pulse facilitation (PPF), long duration enhancement (LTP), long duration inhibition (LTD), pulse frequency, amplitude and width dependent plasticity; (3) The PEDOT:PSS / IGZO memristor of the present invention exhibits photoinduced biological synaptic plasticity under ultraviolet light irradiation, including photoinduced excitatory postsynaptic current (EPSC), photoinduced double pulse facilitation (PPF) effect generated by paired light pulse stimulation, realizing dynamic conversion from short-term memory (STM) to long-term memory (LTM), learning-forgetting-relearning; (4) The PEDOT:PSS / IGZO memristor of the present invention simulates the important characteristics of biological nociceptors for the first time, including four key functions: threshold sensing, relaxation, maladaptation and pain sensitization.

[0015] (5) The PEDOT:PSS / IGZO memristor of the present invention was used to perform recognition tasks based on convolutional neural network (CNN) for the MNIST handwritten digit dataset and the Fashion-MNIST fashion clothing dataset. After 100 training cycles, the recognition accuracy reached 97.72% and 82.49% respectively. The results show the feasibility and effectiveness of the Au / PEDOT:PSS / IGZO / FTO memristor neuromorphic computing system in image recognition tasks, and provide strong support for its application in high-efficiency low-power artificial intelligence hardware and neuromorphic computing.

[0016] (6) Through innovative selection of resistive switching functional layer materials and simple and low-cost preparation methods, a reliable device solution is provided for the development of artificial nociceptors and high-efficiency neuromorphic computing systems, which has broad application prospects in the fields of brain-like computing, bionic sensing, and smart hardware. Attached Figure Description

[0017] Various other advantages and benefits will become apparent to those skilled in the art upon reading the following detailed description of preferred embodiments. The accompanying drawings are for illustrative purposes only and are not intended to limit the invention. In the drawings: Figure 1 This is an image of the memristor structure in an embodiment of the present invention; in, Figure 1 Image (Ⅰ) is a cross-sectional SEM image of the memristor of this invention; Figure 1 (II) is a schematic diagram of the memristor structure and electrical performance test of the present invention; Figure 1 Image (Ⅲ) is a surface SEM image of the PEDOT:PSS thin film of the memristor resistive switching functional layer of the present invention; Figure 2 This is a characteristic curve diagram of the memristor IV curve in an embodiment of the present invention; Figure 3 This is an explanatory diagram of the memristor resistance state switching mechanism in an embodiment of the present invention; in, Figure 3 (a) in the figure is the double logarithmic fitting curve of the memristor under negative voltage in an embodiment of the present invention; Figure 3 (b) in the figure is the double logarithmic fitted curve of the memristor under positive voltage in the embodiment of the present invention; Figure 3 (c) in the figure is a schematic diagram of the resistance switching mechanism model of the memristor when a voltage is applied in an embodiment of the present invention; Figure 4 This is a test graph of the simulated double-pulse facilitated (PPF) characteristic in an embodiment of the present invention; in, Figure 4 (a) in the figure is a graph showing the change of the memristor voltage (0V→+3V→0V) during 11 consecutive positive scans in an embodiment of the present invention; Figure 4 (b) in the figure is a graph showing the change of the memristor voltage (0V→-3V→0V) during 11 consecutive negative scans in an embodiment of the present invention; Figure 4 (c) in the figure is a graph showing the change of the double-pulse facilitated (PPF) exponent of the memristor in an embodiment of the present invention; Figure 4 (d) in the figure is a graph showing the variation of the long-term enhancement (LTD) / long-term rejection (LTP) characteristics of the memristor in the embodiment of the present invention; Figure 4 (e) in the figure is a graph showing the variation of the long-term duration enhancement (LTD) / long-term duration rejection (LTP) characteristics of the memristor under 5 electrical pulse cycles in an embodiment of the present invention; Figure 4 (f) in the figure is a graph showing the current change of the memristor under voltage pulses at different time intervals in an embodiment of the present invention; Figure 4 (g) in the figure is a current variation diagram of the memristor under voltage pulses of different amplitudes in an embodiment of the present invention; Figure 4 (h) in the figure is a graph showing the current variation of the memristor under voltage pulses of different durations in an embodiment of the present invention; Figure 5 This is a test diagram of simulated light-induced double-pulse facilitated (PPF) behavior in an embodiment of the present invention; in, Figure 5 (a) is a schematic diagram of the optical performance test of the memristor in an embodiment of the present invention; Figure 5 (b) in the figure is the excitatory postsynaptic current (EPSC) of the memristor under a single light pulse in an embodiment of the present invention. Figure 5 (c) in the figure is a graph showing the change of the light-induced double-pulse facilitated (PPF) index of the memristor under light stimulation in an embodiment of the present invention; Figure 5 (d) in the figure is a schematic diagram of the synaptic plasticity of the memristor under different numbers of optical pulses in an embodiment of the present invention; Figure 5 (e) in the figure is a schematic diagram of the synaptic plasticity of the memristor under different optical pulse powers in an embodiment of the present invention; Figure 5 (f) in the figure is a schematic diagram of the synaptic plasticity of the memristor under different optical pulse widths in an embodiment of the present invention; Figure 5 (g) in the figure is a schematic diagram of the transformation process of the memristor from short-term memory (STM) to long-term memory (LTM) in an embodiment of the present invention; Figure 5 (h) in the figure is a schematic diagram of the “learning-forgetting-learning” process of the memristor in the embodiment of the present invention; Figure 6 This is a test diagram showing the threshold characteristics of the simulated artificial nociceptor in an embodiment of the present invention; in, Figure 6 (a) in the figure is a schematic diagram of the threshold characteristics of the memristor in an embodiment of the present invention; Figure 6 (b) in the figure represents the output current of the memristor in the embodiment of the present invention under stimulation with a voltage of 6V and different pulse widths; Figure 6 (c) in the figure is a schematic diagram of the relaxation characteristics of the memristor in an embodiment of the present invention; Figure 6 (d) in the figure is a schematic diagram of the incompatibility characteristics of the memristor in the embodiment of the present invention; Figure 7 This is a schematic diagram of the pain sensitization characteristics of the memristor in an embodiment of the present invention; in, Figure 7 In the example, (a) represents the pain response current of the memristor under different levels of damage voltage pulses in the embodiments of the present invention; Figure 7 In the example, (b) represents the maximum output current of the memristor in the logarithmic coordinate system for different voltage pulses in the embodiment of the invention. Figure 7 In this embodiment of the invention, (c) represents the maximum output current corresponding to different voltage pulses of the memristor in linear coordinates. Figure 7In this embodiment of the invention, (d) represents the pain response current corresponding to the memristor being stimulated with voltage pulses of different degrees of damage at different time intervals under the same initial voltage pulse (6V, 0.5s). Figure 7 In this embodiment of the invention, (e) represents the maximum output current of the memristor under different voltage pulses in logarithmic coordinates. Figure 7 In this embodiment of the invention, (f) represents the maximum output current of the memristor under different voltage pulses in linear coordinates. Figure 8 This is a graph showing the recognition accuracy of the handwritten digits and clothing images of the convolutional neural network (CNN) model of the memristor in this embodiment of the invention. Detailed Implementation

[0018] Exemplary embodiments of the present disclosure will now be described in more detail with reference to the accompanying drawings. While exemplary embodiments of the present disclosure are shown in the drawings, it should be understood that the present disclosure may be implemented in various forms and should not be limited to the embodiments set forth herein. Rather, these embodiments are provided to enable a more thorough understanding of the present disclosure and to fully convey the scope of the disclosure to those skilled in the art. It should be noted that, unless otherwise specified, the embodiments and features described herein can be combined with each other. The present invention will now be described in detail with reference to the accompanying drawings and embodiments.

[0019] This invention belongs to the interdisciplinary field of semiconductor microelectronics and artificial intelligence, specifically relating to a PEDOT:PSS / IGZO opto-synaptic memristor with nociceptor function and its fabrication method. PEDOT:PSS and IGZO, as typical representatives of organic-inorganic material systems, possess unique physicochemical properties and heterojunction synergistic effects. Through the synergistic effect of oxygen vacancy migration and the redox process of PEDOT, they can construct continuous conductive channels, thereby achieving reversible switching between high and low conductivity states during SET and RESET processes. This provides an ideal solution for constructing a PEDOT:PSS / IGZO opto-synaptic memristor with nociceptor function. This invention discloses a PEDOT:PSS / IGZO memristor device fabricated using the sol-gel method. The fabricated memristor exhibits functional characteristics mimicking biological neural synapses and nociceptors. The opto-synaptic memristor fabrication method provided by this invention is simple and feasible, offering a reliable device solution for the development of artificial nociceptors and high-energy-efficiency neuromorphic computing systems. It has broad application prospects in neuromorphic computing, biomimetic sensing, and smart hardware.

[0020] This invention is achieved through the following technical solution: This invention provides a PEDOT:PSS / IGZO photoelectric synaptic memristor with nociceptor function. The PEDOT:PSS / IGZO memristor comprises, from bottom to top, a bottom electrode, a resistive switching functional layer, and a top electrode. The resistive switching functional layer is a bilayer heterojunction structure composed of PEDOT:PSS and IGZO thin films. The IGZO is indium gallium zinc oxide.

[0021] The thickness of the resistive switching functional layer composed of PEDOT:PSS and IGZO film is 300nm-350nm; PEDOT:PSS is poly(3,4-ethylenedioxythiophene):polystyrene sulfonate.

[0022] The bottom electrode is FTO, which is conductive indium tin oxide; The top electrode is gold (Au); A method for fabricating a PEDOT:PSS / IGZO photoelectric synaptic memristor with nociceptor function includes the following steps: (1) The conductive substrate FTO was cleaned and dried sequentially with alcohol and deionized water using an ultrasonic instrument; (2) Spin-coat the IGZO precursor solution onto the substrate at a speed of 3000 rpm to form a wet film; (3) Pre-dry the IGZO film on a heating table at 120~200℃ for 20~30 minutes; (4) Anneal the obtained film in an annealing furnace at 350~400℃ for 1 hour to form a more continuous and dense film structure; (5) Spin-coat the PEDOT:PSS solution onto the IGZO film at 3000 rpm for 50 seconds; (6) After drying the PEDOT:PSS / IGZO bilayer film on a heating stage at 100~120°C for 10 minutes, continue annealing at 90~100°C for 20~30 minutes to form a resistive switching functional layer composed of PEDOT:PSS / IGZO film. It is generally recommended that the maximum temperature when heating the PEDOT:PSS film should not exceed 120°C. Prolonged high temperature may cause excessive oxidation or performance degradation of PEDOT:PSS.

[0023] (7) An Au top electrode was fabricated on the device surface by ion sputtering to obtain a PEDOT:PSS / IGZO photoelectric synaptic memristor with nocturnal receptor function; The IGZO precursor solution in step (2) above was prepared using the sol-gel method, which includes the following steps: (1) Dissolve indium nitrate hydrate (InN3O9), gallium nitrate hydrate (Ga(NO3)3·xH2O) and zinc acetate dihydrate (C4H6O4Zn·2H2O) in ethylene glycol methyl ether (C3H8O); (2) Add ethanolamine (C2H7NO) as a stabilizer, wherein the molar ratio of ethanolamine to zinc atoms is 1:1, 1:1.5 or 1:2, which enables the solvent ethylene glycol methyl ether to dissolve the above solute well; (3) Heat and stir the mixed solution at 60°C for 1.5 to 2 hours to ensure thorough mixing; (4) The obtained solution was aged at room temperature for 24 hours to finally obtain an IGZO precursor solution with an atomic molar ratio of In, Gallium (Ga) and Zinc (Zn) of In:Ga:Zn=2:1:1 and a molar concentration of 0.2M.

[0024] The above technical solution will be described in detail with reference to the accompanying drawings and embodiments: Example 1: A method for fabricating a PEDOT:PSS / IGZO photoelectric synaptic memristor with nociceptor function includes the following steps: (1) Preparation of IGZO precursor solution by sol-gel method: Indium nitrate hydrate (InN3O9), gallium nitrate hydrate (Ga(NO3)3·xH2O) and zinc acetate dihydrate (C4H6O4Zn·2H2O) were dissolved in ethylene glycol methyl ether (C3H8O); ethanolamine (C2H7NO) was added as a stabilizer; the mixed solution was heated and stirred at 60°C for 2 hours; the resulting solution was aged at room temperature for 24 hours to finally obtain an IGZO precursor solution with an atomic molar ratio of indium (In), gallium (Ga) and zinc (Zn) of In:Ga:Zn=2:1:1 and a molar concentration of 0.2M; (2) The resistive switching functional layer was prepared by spin coating process: the conductive substrate FTO was cleaned and dried by ultrasonic instrument with alcohol and deionized water in sequence; the IGZO precursor solution was spin coated on the substrate at 3000 rpm to form a wet film; the IGZO film was pre-dried on a heating stage at 150°C for 20 minutes; the obtained film was annealed in an annealing furnace at 350°C for 1 hour; the PEDOT:PSS solution was spin coated on the IGZO film at 3000 rpm for 50 seconds; the PEDOT:PSS / IGZO bilayer film was dried on a heating stage at 100°C for 10 minutes, and then annealed at 90°C for 30 minutes to form the resistive switching functional layer composed of PEDOT:PSS / IGZO film. (3) Fabrication of the top electrode layer: An Au top electrode is fabricated on the device surface using an ion sputtering process, such as... Figure 1As shown, a PEDOT:PSS / IGZO photoelectric synaptic memristor with nociceptor function was obtained, and its device structure is as follows. Figure 1 As shown in (II), the cross-sectional SEM image of the device is as follows. Figure 1 As shown in (Ⅰ), the surface SEM image of the PEDOT:PSS thin film of the resistive switching functional layer of the device is as follows: Figure 1 As shown in (Ⅲ) of the diagram.

[0025] Example 2: This embodiment uses a Keithley 2400 instrument to test the electrical performance and artificial synaptic characteristics of the PEDOT:PSS / IGZO memristor, such as... Figure 1 As shown in (II), all voltage scans and pulse signals are applied to the Au top electrode, and the FTO bottom electrode is grounded.

[0026] The IV characteristic curve of the memristor under 50 consecutive scan cycles is as follows: Figure 2 As shown, under the scanning voltage of 0V→-3V→0V→+3V→0V, the device exhibits a stable current response and a certain switching ratio, demonstrating good memristor characteristics.

[0027] exist Figure 3 In the process, the fitting of the IV characteristic curve of the memristor is as follows: Figure 3 As shown in (a) and (b), under both positive and negative bias, the slopes of the high-resistivity state (HRS) in the low-voltage region are 0.84 and 1.08 (approximately 1), consistent with ohmic conduction. In the higher-voltage region, the slope is 2.29 (approximately 2), consistent with space charge-limited current conduction (SCLC); while in the low-resistivity state (LRS), the slopes of the IV curve fittings are 1.31 and 1.38 (approximately 1), indicating ohmic conduction; the slopes of the linearly fitted IV curves above indicate that the current conduction of the memristor follows the space charge-limited current conduction mechanism (SCLC).

[0028] like Figure 3 As shown in (c), the resistance state switching of the memristor under applied voltage follows the dynamic adjustment mechanism model of the conductive wire; when a negative bias is applied, oxygen vacancies in the IGZO film migrate to the IGZO / PEDOT:PSS interface; as oxygen vacancies accumulate, discontinuous oxygen vacancy-rich regions are formed in the IGZO layer, and the local resistance decreases; at the same time, the non-conductive PEDOT... 0 PEDOT oxidized to a conductive state + PEDOT in the conductive state +When in contact with the oxygen vacancy-rich region in IGZO, a continuous conductive filament is eventually formed, and the device switches from HRS to LRS (SET process). Conversely, under positive bias, the oxygen vacancies in IGZO move in the opposite direction, simultaneously promoting the PEDOT... + Reduced to the non-conductive state of PEDOT 0 This causes the formed conductive filament to break, and the device switches from LRS back to HRS (RESET process); this illustrates the ion migration mechanism under electric field control, which provides physical support for the bipolar resistance switching characteristics of the memristor.

[0029] exist Figure 4 In, such as Figure 4 As shown in (a), the conductivity of the device gradually increases during 11 consecutive +3V voltage scans; Figure 4 As shown in (b), the conductivity of the device gradually decreases under the same number of continuous scans of -3V voltage; the change in conductivity is closely related to synaptic plasticity, and the above results indicate that the prepared memristor has adjustable synaptic weighting characteristics.

[0030] In order to simulate the dual-pulse facilitated (PPF) characteristics using the memristor of this invention, such as Figure 4 As shown in (c), two voltage pulses of equal width (4V, 0.1s) with a time interval (from 0.1s to 2.0s) were applied to the device, allowing for the measurement of two current spikes and the calculation of the PPF exponent. The second pulse response current of the device was significantly higher than that of the first pulse response current, exhibiting typical short-term plasticity (STP) characteristics, i.e., the double-pulse facilitation (PPF) phenomenon.

[0031] To simulate long-term synaptic plasticity using the memristor of this invention, 50 consecutive pulses with an interval of 0.5 s and an amplitude of ±3 V were applied to the device; as Figure 4 As shown in (d), the device conductance gradually increases with the increase of the number of positive pulses; while under the action of negative pulses, the conductance gradually decreases, exhibiting typical long-term peak (LTP) and long-term peak decline (LTD) behavior; as shown in (d), the device conductance gradually increases with the increase of the number of positive pulses; while under the action of negative pulses, the conductance gradually decreases, exhibiting typical long-term peak (LTP) and long-term peak decline (LTD) behavior. Figure 4 As shown in (e), the long-term boost (LTP) and long-term rejection (LTD) curves of the memristor were further tested in 5 repeated tests. The conductance change trend remained consistent, indicating that the device has good stability and repeatability in terms of synaptic conductance regulation.

[0032] like Figure 4As shown in (f)–(h), during the application of continuous pulse stimulation to the device, the response current shows an upward trend as the pulse interval (0.5 s–0.9 s), pulse amplitude (2 V–4 V), and pulse width (0.5 s–1 s) gradually increase, reflecting the continuous enhancement of synaptic weight. This indicates that the device can exhibit a variety of synaptic plasticity behaviors.

[0033] Example 3: like Figure 5 As shown in the figure, this embodiment uses a Keithley 2400 instrument and a 365 nm ultraviolet light source to test the current response of the PEDOT:PSS / IGZO memristor under light stimulation at a readout voltage of 0.1V, as shown in Figure 4(a).

[0034] like Figure 5 Results (b) show that when the memristor is exposed to continuous ultraviolet light, the current rises rapidly; after 10 seconds, when the light is stopped, the current gradually decreases, indicating that light stimulation can effectively regulate the synaptic weights of the device.

[0035] In order to simulate the light-induced two-pulse facilitated (PPF) behavior using the memristor of this invention, such as Figure 5 As shown in (c), in a pair of strengths of 23.05 mW / cm 2 Under the stimulation of a light pulse with a width of 3s, the double-pulse facilitated (PPF) index of the device gradually decreases as the pulse interval increases, exhibiting typical short-term plasticity of light modulation.

[0036] To test the optical pulse number flexibility, power flexibility, and width flexibility of the memristor of this invention, the number, intensity, and width of the optical pulses are adjusted; such as Figure 5 As shown in (d) in the figure, at a power of 23.05 mW / cm 2 Under the conditions of a pulse width of 10s and an interval of 1s, as the number of optical pulses increases from 1 to 4, the photocurrent of the device gradually increases; for example... Figure 5 As shown in (e), within a fixed illumination time of 10 s, as the light pulse power increases from 2.3 mW / cm², the light pulse power decreases. 2 Gradually increased to 23.05 mW / cm 2 The photocurrent of the device shows a significant upward trend, exhibiting intensity-dependent response characteristics; such as Figure 5 As shown in (f), at a constant power of 23.05 mW / cm 2 As the optical pulse width increases from 1 s to 15 s, the synaptic weights of the device continuously accumulate, while the relaxation time of the photocurrent gradually lengthens, indicating a gradual transition from short-term memory (STM) to long-term memory (LTM). Figure 5(g) in the middle.

[0037] In order to simulate the "learning-forgetting-relearning" process in the human brain using the memristor of this invention, such as Figure 5 As shown in (h), 23.05 mW / cm was used first. 2 A 10-second light pulse simulates the initial learning process; then the light is turned off for 15 seconds to simulate the "forgetting" stage, followed by 5 seconds of light to represent "relearning"; after three consecutive "forgetting-relearning" training sessions, the device's maximum excitatory postsynaptic current (EPSC) is significantly higher than that of the initial learning stage, indicating its potential to simulate the dynamic process of learning and memory.

[0038] Example 4: In the human body, when the skin is subjected to harmful stimulation, the nociceptors located at the terminals of sensory neurons can receive the signal and trigger an action potential when the stimulation intensity exceeds the pain threshold. Ultimately, the injury information is transmitted to the brain through the spinal cord and other central nervous system structures. In order to simulate the sensory function of human nociceptors using the memristor of the present invention, this embodiment uses a voltage pulse as an external stimulation signal to test the memristor.

[0039] like Figure 6 As shown, to simulate the threshold characteristics of an artificial nociceptor using the memristor of this invention, a series of voltage pulses (0.5V to 7V) of varying amplitudes are applied to the memristor; as... Figure 6 As shown in (a), when the pulse voltage is 0.5V, 1V, 1.5V, and 2V, the device only produces a weak current response; however, when the voltage reaches 3V, the device is significantly activated, and the output current increases dramatically. This behavior is similar to the triggering response of a biological nociceptor to a painful stimulus exceeding a threshold. Furthermore, in Figure 6 In (b) of the invention, a 6V voltage pulse higher than the threshold was used, and its duration was varied (from 0.5s to 6.5s). It was observed that the output current increased with the increase of the pulse width, indicating that the memristor with nociceptor function of the present invention can reflect the positive correlation between the duration of stimulation and the intensity of pain.

[0040] To simulate the relaxation characteristics of an artificial nociceptor using the memristor of this invention, a strong 6V voltage pulse is first applied to the device to induce a sensitive state, followed by a weak 2V pulse applied after different time intervals (0.1s to 1s); as... Figure 6 As shown in (c), as the interval between the two pulses lengthens, the postsynaptic current response induced by the weak pulse gradually weakens, indicating that the device gradually recovers to the baseline level after the strong stimulus is removed, while still retaining the ability to remember recent stimuli and responding to stimuli below the threshold.

[0041] The memristor with nociceptor function of the present invention also has an "adaptive" characteristic, that is, the response gradually saturates under continuous strong stimulation, in order to avoid over-responding to repeated noxious signals, thereby playing a physiological protective role; such as Figure 6 As shown in (d), when the device is stimulated by a continuous 6V voltage pulse, the output current gradually increases and eventually tends to saturate. Even if the pulse is continued to be applied, the current no longer changes significantly. This behavior is highly consistent with the maladaptive protective mechanism exhibited by biological nociceptors when faced with continuous pain stimuli.

[0042] Pain sensitization is a key mechanism in the regulation of pain perception in biological nociceptive systems. After nociceptors are sensitized by high-intensity stimulation, their response to subsequent stimuli is significantly enhanced, exacerbating pain from existing noxious stimuli and even inducing pain from slight touch. This is a regulatory mechanism for protecting damaged tissues, primarily manifested in two clinical phenomena: "abnormal pain" (pain induced by harmless stimuli) and "hyperalgesia" (increased pain from noxious stimuli). Figure 7 As shown, to simulate the pain sensitization behavior of an artificial nociceptor using the memristor of this invention, the invention verifies this by applying voltage pulses of different intensities and time intervals to the memristor. First, voltage pulses of 0V, 3V, 5V, and 6V (width 0.5s) are used as initial nociceptive stimuli, followed by secondary stimulation with pulses of 1–6V (width 0.5s). The pain response of the device is evaluated by observing the output current. The results are as follows: Figure 7 As shown in (a) above, under the condition of a fixed initial and secondary stimulus interval, as the initial pulse voltage increases, the device's response to subsequent stimuli is significantly enhanced, manifested as a decrease in the threshold voltage ( Figure 7 (b) and the increase in output current amplitude ( Figure 7 (c) accurately reproduces the biological sensitization principle that "the more severe the initial injury, the more sensitive the subsequent pain sensation"; furthermore, initial tissue injury is simulated with a 6V, 0.5s pulse, and secondary stimulation of 1–6V is applied after different delays (0.3s, 10s, 40s); such as Figure 7 As shown in (d)–(f), as the interval between the two stimuli shortens, the device's response sensitivity to the secondary stimulus gradually increases, effectively simulating the time-dependent sensitization behavior of "being more sensitive to repeated injuries in the short term after trauma" in biological systems. This indicates that the memristor of the present invention can flexibly modulate the degree of pain sensitization by adjusting the stimulation interval. The above experimental results consistently show that the memristor of the present invention can effectively simulate the pain sensitization behavior of biological nociceptors, and its response pattern is highly consistent with the physiological mechanism.

[0043] Example 5: To verify the application potential of the developed device in neuromorphic computing, this invention constructs a pattern recognition system based on a convolutional neural network (CNN). To obtain the conductance state for storing synaptic weights, 50 consecutive electrical pulses were applied to the device, and the long-term duration enhancement (LTP) and long-term duration inhibition (LTD) behaviors were measured. The LTP and LTD results of the fabricated device were fitted and integrated into a neural network model. The device was then used to perform image recognition tasks on the MNIST handwritten digit dataset and the Fashion-MNIST fashion clothing dataset, respectively. Figure 8 As shown, after 100 training iterations, the constructed CNN model achieved recognition accuracies of 97.72% and 82.49% on the two datasets mentioned above, respectively. This result fully verifies the feasibility and effectiveness of the neuromorphic computing system based on the memristor of this invention in image recognition tasks, and provides strong support for its application in high-efficiency and low-power artificial intelligence hardware.

[0044] The above embodiments are merely illustrative of the technical concept and features of the present invention, and are intended to enable those skilled in the art to understand the content of the present invention and implement it accordingly. They should not be construed as limiting the scope of protection of the present invention. All equivalent changes or modifications made based on the essence of the content of the present invention should be covered within the scope of protection of the present invention.

[0045] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and not to limit it. Although the present invention has been described in detail with reference to the above embodiments, those skilled in the art should understand that modifications or equivalent substitutions can still be made to the specific implementation of the present invention. Any modifications or equivalent substitutions that do not depart from the spirit and scope of the present invention should be covered within the scope of protection of the claims of the present invention.

Claims

1. A PEDOT:PSS / IGZO photoelectric synaptic memristor with nociceptor function, characterized in that, include: A bottom electrode, a resistive switching functional layer, and a top electrode are sequentially arranged, wherein the resistive switching functional layer is a bilayer heterojunction structure composed of PEDOT:PSS and IGZO thin films, wherein IGZO is indium gallium zinc oxide, and the IGZO thin film is disposed on the bottom electrode.

2. The memristor according to claim 1, characterized in that, The bottom electrode is FTO.

3. The memristor according to claim 1, characterized in that, The top electrode is Au.

4. A method for fabricating a memristor according to any one of claims 1-3, characterized in that, include: The conductive substrate is cleaned and dried, wherein the conductive substrate is used as the bottom electrode; The IGZO precursor solution was spin-coated onto a substrate to form a wet film. The wet film is pre-dried by heating and then annealed to produce an IGZO thin film. PEDOT:PSS solution was spin-coated onto IGZO film to obtain PEDOT:PSS / IGZO bilayer film; The PEDOT:PSS / IGZO bilayer film was heated, dried, and then annealed to form a resistive switching functional layer. A memristor is fabricated by preparing a top electrode on the surface of a resistive switching functional layer using an ion sputtering process.

5. The preparation method according to claim 4, characterized in that, The IGZO precursor solution was prepared using the sol-gel method, and the preparation process included: Indium nitrate hydrate, gallium nitrate hydrate, and zinc acetate dihydrate were dissolved in ethylene glycol methyl ether, and ethanolamine was added as a stabilizer to obtain a mixed solution. The mixed solution was heated and stirred, and the stirred solution was aged to obtain the IGZO precursor solution.

6. The preparation method according to claim 4, characterized in that, The atomic molar ratio of indium, gallium, and zinc in the IGZO precursor solution is 2:1:1, and the molar concentration is 0.2M.

7. The preparation method according to claim 4, characterized in that, The wet film was pre-dried on a heating table at 150°C for 20 minutes and then annealed in an annealing furnace at 350°C for 1 hour.

8. The preparation method according to claim 4, characterized in that, After drying the PEDOT:PSS / IGZO bilayer film on a heating stage at 100°C for 10 minutes, it is then annealed at 90°C for another 30 minutes to form a resistive switching functional layer.

9. An application of the memristor according to any one of claims 1-3, characterized in that, The memristor is used in artificial nociceptors and neuromorphic computing systems.

Images