CN111525029B - Pd conductive filament-based bionic synaptic device and preparation method and application thereof - Google Patents
Pd conductive filament-based bionic synaptic device and preparation method and application thereof Download PDFInfo
- Publication number
- CN111525029B CN111525029B CN202010498788.9A CN202010498788A CN111525029B CN 111525029 B CN111525029 B CN 111525029B CN 202010498788 A CN202010498788 A CN 202010498788A CN 111525029 B CN111525029 B CN 111525029B
- Authority
- CN
- China
- Prior art keywords
- snse
- nsto
- substrate
- bionic
- functional layer
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
Classifications
-
- H—ELECTRICITY
- H10—SEMICONDUCTOR DEVICES; ELECTRIC SOLID-STATE DEVICES NOT OTHERWISE PROVIDED FOR
- H10N—ELECTRIC SOLID-STATE DEVICES NOT OTHERWISE PROVIDED FOR
- H10N70/00—Solid-state devices without a potential-jump barrier or surface barrier, and specially adapted for rectifying, amplifying, oscillating or switching
- H10N70/20—Multistable switching devices, e.g. memristors
- H10N70/24—Multistable switching devices, e.g. memristors based on migration or redistribution of ionic species, e.g. anions, vacancies
- H10N70/245—Multistable switching devices, e.g. memristors based on migration or redistribution of ionic species, e.g. anions, vacancies the species being metal cations, e.g. programmable metallization cells
-
- H—ELECTRICITY
- H10—SEMICONDUCTOR DEVICES; ELECTRIC SOLID-STATE DEVICES NOT OTHERWISE PROVIDED FOR
- H10N—ELECTRIC SOLID-STATE DEVICES NOT OTHERWISE PROVIDED FOR
- H10N70/00—Solid-state devices without a potential-jump barrier or surface barrier, and specially adapted for rectifying, amplifying, oscillating or switching
- H10N70/011—Manufacture or treatment of multistable switching devices
-
- H—ELECTRICITY
- H10—SEMICONDUCTOR DEVICES; ELECTRIC SOLID-STATE DEVICES NOT OTHERWISE PROVIDED FOR
- H10N—ELECTRIC SOLID-STATE DEVICES NOT OTHERWISE PROVIDED FOR
- H10N70/00—Solid-state devices without a potential-jump barrier or surface barrier, and specially adapted for rectifying, amplifying, oscillating or switching
- H10N70/801—Constructional details of multistable switching devices
- H10N70/881—Switching materials
- H10N70/882—Compounds of sulfur, selenium or tellurium, e.g. chalcogenides
- H10N70/8825—Selenides, e.g. GeSe
Abstract
The invention provides a bionic synapse device based on Pd conductive filaments, a preparation method and application thereof. The bionic device is composed of a strontium niobate titanate (NSTO) substrate, a tin selenide (SnSe) functional layer and a Pd top electrode layer. The high-quality SnSe layer and the Pd layer are prepared by applying pulse laser deposition and magnetron sputtering technology respectively. The bionic synapse device provided by the invention adopts the SnSe film as the functional layer, and the SnSe film is easier to form Sn and Se defects when an external electric field acts, so that the device can form a relatively stable Pd conductive filament. Thus, the device of the present invention not only better mimics the neural synapse enhancing and inhibiting properties in the "RESET" and "SET" regions, but also reduces the variability of its switching voltages.
Description
Technical Field
The invention relates to the technical field of electronics and materials, in particular to a bionic synapse device based on Pd conductive filaments, a preparation method and application thereof.
Background
The human brain, which is efficient, fast and low in power consumption, has long been known as a "supercomputer" for processing information. Neuromorphic calculations are a new computational model that simulates neurobiological processes by building computational architectures similar to the human brain. Simulating synapses is an important step in achieving a highly efficient artificial neuromorphic system. The artificial nerve morphology calculation structure built by the existing electronic device is difficult to be comparable with the perception, learning and memory of the human brain. There are up to 10 in the human brain 11 Individual neurons and 10 15 The connections between individual synapses, in particular neurons, are very complex, which makes it difficult for common electronics to be applied to artificial neural networks.
Conventional CMOS-based electronic synapses are composed of components such as transistors, capacitors, and comparators, which make it difficult to overcome their size and power consumption. The memristor is used as a device at two ends capable of memorizing the quantity of charges flowing through the memristor and presenting the time memory characteristic of resistance, is an electronic element which is most likely to simulate the processes of human brain learning and memory and the like to realize brain-like artificial cognition, and provides unprecedented possibility for high-efficiency processing of large-scale data operation and storage. Researchers have attempted to incorporate related methods of human brain information transmission, processing and handling, and designed memristive devices for simulating biological neurons and synapses to apply them to the field of neuromorphic computation. Although some reports on biomimetic synaptic devices exist, the device performance and switch discreteness are not ideal.
Disclosure of Invention
The invention aims to provide a bionic synaptic device based on Pd conductive filaments, and a preparation method and application thereof, so as to improve the performance difference and the switch discreteness of the bionic synaptic device, thereby better simulating the enhancement and inhibition actions of the biological synapses.
The purpose of the invention is realized in the following way: a bionic synaptic device based on Pd conductive filament is prepared by preparing a two-dimensional tin selenide functional layer with controllable thickness on a strontium niobate doped titanate substrate, preparing a palladium top electrode layer on the tin selenide functional layer, and finally forming a device which is expressed as Pd/SnSe/NSTO structure and can internally form Pd conductive filament when an external electric field acts on the device.
The doped niobium concentration of the strontium titanate doped substrate is 0.3% -0.9%.
The thickness of the tin selenide functional layer is 5 nm-100 nm.
The thickness of the palladium top electrode layer is 50 nm-200 nm.
A preparation method of a bionic synapse device based on Pd conductive filaments comprises the following steps:
a. mixing Sn particles and Se particles, ball milling into uniform powder, placing the uniform powder in a muffle furnace, calcining at 500-1000 ℃ for 12-48 hours, cooling, placing the uniform powder in a plasma sintering furnace, and calcining for 3-20 minutes under the conditions of 20-60 MPa and 300-600 ℃ to finally obtain the SnSe target;
b. respectively ultrasonically cleaning an NSTO substrate in acetone, alcohol and deionized water in sequence, and then taking out the NSTO substrate and drying the NSTO substrate by nitrogen for standby;
c. vacuum is small at the back bottom by utilizing pulse laser deposition technologyAt 10 -3 Pa, deposition parameters of 0.01-5 Pa, and laser density of 0.5-10J cm -2 Preparing a two-dimensional SnSe film as a functional layer on an NSTO substrate under the conditions that the laser frequency is 1-10 Hz and the substrate temperature is 25-500 ℃;
d. c, pressing the mask plate on the SnSe/NSTO prepared in the step c, and then adopting a magnetron sputtering technology to realize vacuum of less than 10 on the back -3 And preparing the palladium top electrode layer under the conditions of Pa, gas pressure of 0.01-5 Pa, sputtering power of 1-30W and argon flow of 5-50 sccm.
The purity of both Sn and Se particles was 99.999%.
The use of the device described above for simulating biological synaptic behaviour.
The bionic synapse device provided by the invention adopts the SnSe film as the functional layer, and the SnSe film is easier to form Sn and Se defects when an external electric field acts, so that the device can form a relatively stable Pd conductive filament. Thus, the device of the present invention not only better mimics the neural synapse enhancing and inhibiting properties in the "RESET" and "SET" regions, but also reduces the variability of its switching voltages. Most devices simulating the function of nerve biological synapses have non-centralized switch voltage distribution, and the performance difference among unit devices is large. The invention has small difference between devices, more concentrated switch voltage distribution, more than 90% of SET/RESET voltage floating values of about 0.12V and 0.14V, and can better simulate synapse enhancement and inhibition in a switch area.
The biological synapse device provided by the invention can be used for simulating the behavior of biological synapses, and further can be used for preparing a nerve synapse bionic device. In particular, the neuromorphic device of the present invention may be used to simulate the short-term plasticity, long-term plasticity of a biological synapse.
The nerve bionic device can simulate the learning and memory functions of biological synapses, so that the nerve bionic device can be used for preparing miniaturized, high-density and low-power-consumption nerve morphology calculation chips. The devices so fabricated can mimic the basic functions of a single artificial synapse, including enhancement and suppression, pulse time dependent plasticity (STDP), double pulse facilitation (PPF). These results demonstrate the feasibility of artificial synapses of the neuromorphic system.
Drawings
FIG. 1 shows the current-voltage (I-V) characteristics and the probability of switching voltage distribution of a Pd/SnSe/NSTO structure device according to the invention. Wherein FIG. 1a is an I-V curve before the Pd/SnSe/NSTO device does not reach the electroforming voltage, and the inset is a corresponding resistance-voltage curve (R-V); FIG. 1b is an I-V curve after the electroforming voltage is reached, the inset shows the electroforming voltage; FIG. 1c is a statistical graph of SET voltage distribution probability; FIG. 1d is a statistical graph of probability of RESET voltage distribution.
FIG. 2 is an X-ray spectroscopy chart of a cross-section of a biomimetic synaptic device in an on state. Wherein, figures a-f are profiles of six elements, strontium (Sr), titanium (Ti), oxygen (O), palladium (Pd), selenium (Se) and tin (Sn), respectively.
Fig. 3 is a schematic view of the structure of the device of the present invention. In the figure, 1 is a substrate, 2 is a bionic layer, and 3 is a Pd electrode membrane layer.
FIG. 4 is a graph of the results of synaptic enhanced inhibition of the device of the present invention. Wherein, fig. 4a and fig. 4b are respectively enlarged views of the corresponding illustration of the local area with different conductivities in 100 continuous scans under the action of positive and negative direct voltages; FIG. 4c is a graph of conductance as a function of pulse number for successive application of enhancement and suppression pulses; fig. 4d is a graph of the real-time conductance versus initial conductance as a function of pulse number.
FIG. 5 is a graph of Pd/SnSe/NSTO electronic synapse simulation biological synapse double pulse differentiation (PPF) and pulse time dependent plasticity (STDP) behavior in accordance with the present disclosure; wherein fig. 5a is PPF: a memristor conductance variation graph with interval time; fig. 5b is STDP: memristor conductance variation versus time interval plot.
Detailed Description
In order to better illustrate the biomimetic synaptic device of the present embodiment, the structural function of the device and the technical scheme of the present design are described in detail below with reference to the accompanying drawings.
The structure of the bionic synapse device provided by the invention comprises a substrate NSTO, wherein a two-dimensional tin selenide (SnSe) film is manufactured on the substrate NSTO and used as a functional layer, and a Pd top electrode layer is grown on the SnSe film, so that the device with a Pd/SnSe/NSTO structure is finally formed. The SnSe and Pd layers are prepared by pulse laser and magnetron sputtering technology respectively.
The following describes the preparation process of Pd/SnSe/NSTO biomimetic synaptic device:
1. preparing 99.999% of Sn particles and 99.999% of Se particles, mixing, ball-milling for 1 hour by a ball mill to obtain uniform powder, placing in a muffle furnace, calcining for 24 hours at 950 ℃, cooling for 24 hours, calcining for 7 minutes at 50 MPa and 525 ℃ by a plasma sintering furnace for the second time, and finally obtaining the target material required by SnSe pulse laser deposition;
2. sequentially ultrasonic cleaning strontium titanate (NSTO) single plane substrate doped with 0.7% niobium in acetone, ethanol and deionized water, respectively, and taking out to use N 2 Blow-drying for standby;
3. placing the NSTO substrate and the prepared target material into a vacuum cavity of pulse laser deposition equipment at the same time, and vacuumizing to 10 -5 The formal deposition parameters of Pa, snSe are that the pressure is 0.1 and Pa and the laser density is 1.5J cm -2 Laser frequency 5 Hz, substrate temperature 300 ℃ and 3 minutes deposition time;
4. pressing a mask on the obtained SnSe/NSTO, placing the mask into a cavity of a magnetron sputtering device, and vacuumizing to 10 -4 Pa prepares Pd top electrode with deposition parameters of 1 Pa, 10W, ar: 25 sccm, deposition time of 10 minutes.
In the embodiments of the present invention, a part of the functions of simulating biological synapses using the prepared device of Pd/SnSe/NSTO structure are described in detail with reference to the accompanying drawings:
the invention relates to a nerve bionic device, in particular to a device for simulating electronic synaptic function connected between neurons. The synaptic function to be realized, in particular, the activity time sequence dependent synaptic plasticity function, is defined as a top electrode in the invention being a presynaptic membrane, a SnSe functional layer being a synaptic cleft, and a substrate being a postsynaptic membrane (as shown in FIG. 3). By applying pulses in different directions, the nerve bionic conductivity change can be controlled, and the nerve bionic conductivity change has stability and repeatability.
FIG. 1a is an I-V curve obtained when a scan voltage of + -1.5V is applied when the Pd/SnSe/NSTO device does not reach the electroforming voltage of 2.3V, showing that the device memory window is small. The inset is the corresponding resistance-voltage curve (R-V) showing a smaller ratio of high to low resistance. However, the window becomes larger for the I-V curve (FIG. 1 b) after reaching the electroforming voltage of 2.3V. The inset shows that the device was first electroformed by applying a voltage of 3V. FIG. 1c is a statistical graph of probability of SET voltage distribution for the device; FIG. 1d is a statistical plot of the probability of the RESET voltage distribution, showing that more than 90% of the SET/RESET voltage float values are about 0.12V and 0.14V, with a more concentrated switching voltage distribution. Filaments mainly derived from Pd formation were relatively stable, as we confirmed by X-ray spectroscopy (as shown in fig. 2 d). The X-ray energy spectrum analysis diagrams of the section of the bionic synapse device in the on state of FIGS. 2a-f are six element profiles of strontium (Sr), titanium (Ti), oxygen (O), palladium (Pd), selenium (Se) and tin (Sn), respectively. As shown in fig. 3, the nerve bionic device manufactured by the invention structurally comprises a substrate 1 at the bottom layer, a bionic layer 2 on the substrate 1, and a Pd electrode membrane layer 3 at the top.
Fig. 4a and fig. 4b are respectively enlarged local views of 100 successive scans with different conductivities under the action of positive and negative direct voltages, and it can be seen from the figures that the conductivities of the device are linearly adjustable, and also a better memory line is shown. Fig. 4c is a graph of conductance as a function of pulse number for successive application of enhancement and suppression pulses, wherein enhancement and suppression of synaptic weights can be achieved. Fig. 4d is a plot of the real-time conductance versus initial conductance as a function of pulse number. The positive boost pulse causes accumulation of Pd in the SnSe, where Pd conductive filaments are formed. The negative pressure pulse moves Pd in the opposite direction causing partial Pd filament breakage.
In the present invention, focus is mainly on modeling pulse time dependent plasticity (STDP) as the principle synaptic learning rule. STDP is tightly dependent on the relative time difference (Δt) between an action potential or pulse and input (presynaptic) and output (postsynaptic) neurons. When the presynaptic peak occurs before the postsynaptic peak, Δt>0, long Term Potentiation (LTP) occurs. Conversely, Δt<0, long term inhibition (LTD) occurs. FIG. 5b shows STDP synaptic weights versus interval time "STDP index "is defined as the percent change in device conductance after STDP has occurred, STDP index= (G) 2 -G 1 ）/ G 1 X 100%. If the presynaptic peak precedes the postsynaptic peak (Δt> 0），STDP index>0, a long-term enhancement occurs. If the post-synaptic peak precedes the pre-synaptic peak (Δt<0) Then STDP index < 0, long-term suppression occurs. Furthermore, it can be seen from the graph that the smaller Δt, the greater the change in synaptic weight. This example perfectly reproduces STDP function in biological synapses. Double pulse facilitation (PPF) in biological synapses mainly comprises the temporal sum of inputs, e.g. synaptic weights can be enhanced by reducing the time interval of two sequential potentiating pulses. Fig. 5a shows PPF measurements in a Pd/SnSe/NSTO device, it is evident that the smaller the time interval, the larger the PPF ratio, i.e. the larger the effect, which is consistent with the reflection of biological synapses. The inset in fig. 5a and 5b are the measured pulse shape and parameters, respectively.
Claims (4)
1. A bionic synaptic device based on Pd conductive filaments is characterized in that a two-dimensional tin selenide functional layer with controllable thickness is prepared on a strontium niobate doped substrate, a palladium top electrode layer is prepared on the tin selenide functional layer, and finally a device which is expressed as Pd/SnSe/NSTO structure is formed, and the device can internally form Pd conductive filaments when an external electric field acts on the device;
the doped niobium concentration of the strontium titanate doped niobium substrate is 0.3% -0.9%; the thickness of the thin film of the tin selenide functional layer is 5 nm-100 nm; the thickness of the palladium top electrode layer is 50 nm-200 nm;
the preparation method comprises the following steps:
a. mixing Sn particles and Se particles, ball milling into uniform powder, placing the uniform powder in a muffle furnace, calcining at 500-1000 ℃ for 12-48 hours, cooling, placing the uniform powder in a plasma sintering furnace, and calcining at 20-60 MPa and 300-600 ℃ for 3-20 minutes to finally obtain a SnSe target;
b. sequentially cleaning NSTO substrate with ultrasonic wave in acetone, alcohol and deionized water, respectively, and taking out to obtain N 2 Blow-drying for standby;
c. deposition using pulsed laserTechnology, vacuum is less than 10 at the back -3 Pa, deposition parameters of 0.01-5 Pa, and laser density of 0.5-10J cm -2 Preparing a two-dimensional SnSe film as a functional layer on an NSTO substrate under the conditions that the laser frequency is 1-10 Hz and the substrate temperature is 25-500 ℃;
d. c, pressing the mask plate on the SnSe/NSTO prepared in the step c, and then adopting a magnetron sputtering technology to realize vacuum of less than 10 on the back -3 And preparing the palladium top electrode layer under the conditions of Pa, gas pressure of 0.01-5 Pa, sputtering power of 1-30W and argon flow of 5-50 sccm.
2. The preparation method of the bionic synapse device based on Pd conductive filaments is characterized by comprising the following steps:
a. mixing Sn particles and Se particles, ball milling into uniform powder, placing the uniform powder in a muffle furnace, calcining at 500-1000 ℃ for 12-48 hours, cooling, placing the uniform powder in a plasma sintering furnace, and calcining at 20-60 MPa and 300-600 ℃ for 3-20 minutes to finally obtain a SnSe target;
b. sequentially cleaning NSTO substrate with ultrasonic wave in acetone, alcohol and deionized water, respectively, and taking out to obtain N 2 Blow-drying for standby;
c. vacuum at the back of the wafer is less than 10 by using pulse laser deposition technique -3 Pa, deposition parameters of 0.01-5 Pa, and laser density of 0.5-10J cm -2 Preparing a two-dimensional SnSe film as a functional layer on an NSTO substrate under the conditions that the laser frequency is 1-10 Hz and the substrate temperature is 25-500 ℃;
d. c, pressing the mask plate on the SnSe/NSTO prepared in the step c, and then adopting a magnetron sputtering technology to realize vacuum of less than 10 on the back -3 And preparing the palladium top electrode layer under the conditions of Pa, gas pressure of 0.01-5 Pa, sputtering power of 1-30W and argon flow of 5-50 sccm.
3. The method of claim 2, wherein the Sn particles and Se particles each have a purity of 99.999%.
4. Use of the device of claim 1 for simulating biological synaptic behaviour.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202010498788.9A CN111525029B (en) | 2020-06-04 | 2020-06-04 | Pd conductive filament-based bionic synaptic device and preparation method and application thereof |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202010498788.9A CN111525029B (en) | 2020-06-04 | 2020-06-04 | Pd conductive filament-based bionic synaptic device and preparation method and application thereof |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN111525029A CN111525029A (en) | 2020-08-11 |
| CN111525029B true CN111525029B (en) | 2023-09-22 |
Family
ID=71910039
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202010498788.9A Active CN111525029B (en) | 2020-06-04 | 2020-06-04 | Pd conductive filament-based bionic synaptic device and preparation method and application thereof |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN111525029B (en) |
Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103346256A (en) * | 2013-07-03 | 2013-10-09 | 南京大学 | Memristor based on ferroelectric tunnel junction |
| CN106654006A (en) * | 2016-11-10 | 2017-05-10 | 河北大学 | SiO2 neurobionics layer-based neurobionics device and preparation method thereof |
| CN110970554A (en) * | 2019-12-10 | 2020-04-07 | 河南大学 | Barium titanate single crystal epitaxial film threshold switch device and preparation method thereof |
Family Cites Families (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20190131384A1 (en) * | 2017-10-27 | 2019-05-02 | National University Of Singapore | Ferroelectric tunnel junction structure and method of fabricating the same |
-
2020
- 2020-06-04 CN CN202010498788.9A patent/CN111525029B/en active Active
Patent Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103346256A (en) * | 2013-07-03 | 2013-10-09 | 南京大学 | Memristor based on ferroelectric tunnel junction |
| CN106654006A (en) * | 2016-11-10 | 2017-05-10 | 河北大学 | SiO2 neurobionics layer-based neurobionics device and preparation method thereof |
| CN110970554A (en) * | 2019-12-10 | 2020-04-07 | 河南大学 | Barium titanate single crystal epitaxial film threshold switch device and preparation method thereof |
Non-Patent Citations (2)
| Title |
|---|
| 《Ideal strength and deformation mechanism in high-efficiency thermoelectric SnSe》;Li G, Aydemir U, Wood M, et al.;《Chemistry of Materials》;第29卷(第5期);第2382-2389页 * |
| 《Optimized growth conditions of epitaxial SnSe films grown by pulsed laser deposition》;Hara T, Fujishiro H, Naito T, et al.;《Japanese Journal of Applied Physics》;第56卷(第12期);第1-6页 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN111525029A (en) | 2020-08-11 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Sun et al. | The future of memristors: Materials engineering and neural networks | |
| CN110739393B (en) | Bionic synapse device and manufacturing method and application thereof | |
| CN110416312B (en) | Low-power-consumption neurosynaptic thin film transistor and preparation method thereof | |
| Cao et al. | Memristor-based neural networks: a bridge from device to artificial intelligence | |
| Xiao et al. | Oxygen vacancy migration/diffusion induced synaptic plasticity in a single titanate nanobelt | |
| Wang et al. | Room-temperature developed flexible biomemristor with ultralow switching voltage for array learning | |
| CN106098932A (en) | A kind of linear gradual memristor and preparation method thereof | |
| KR102117658B1 (en) | Memristor Applicable to Neuromorphic System and Method of Manufacturing Memristor Consisting of Metal-oxide Films Fabricated by Solution Process | |
| CN111739935A (en) | Inorganic synapse transistor structure and method of fabrication | |
| Li et al. | Electronic synaptic characteristics and simulation application of Ag/CeO2/Pt memristor | |
| Gupta et al. | Perovskite based low power synaptic memristor device for neuromorphic application | |
| Ye et al. | Overview of memristor-based neural network design and applications | |
| CN111525029B (en) | Pd conductive filament-based bionic synaptic device and preparation method and application thereof | |
| Kim et al. | Artificial synaptic and self-rectifying properties of crystalline (Na1-xKx) NbO3 thin films grown on Sr2Nb3O10 nanosheet seed layers | |
| CN111048582B (en) | Method for adjusting synapse plasticity of metal oxide synapse transistor | |
| CN109585650B (en) | Glial cell-like neuromorphic device and preparation method thereof | |
| Khot et al. | Self‐assembled vapor-transport-deposited SnS nanoflake-based memory devices with synaptic learning properties | |
| CN110289350B (en) | Metal porphyrin heterojunction-based memristor and preparation method and application thereof | |
| CN109978019B (en) | Image mode recognition analog and digital mixed memristor equipment and preparation thereof, and STDP learning rule and image mode recognition method are realized | |
| CN112563414B (en) | Analog all-carbon-based memristor synapse device and preparation method thereof | |
| CN112018236A (en) | PZT-based memristor device, and preparation method and application thereof | |
| Meng et al. | Fiber-shaped cu-ion diffusive memristor for neuromorphic computing | |
| CN111769194A (en) | Flexible photoelectric sensing memristor based on sawtooth structure nanowire | |
| CN112420922B (en) | Low-power consumption CBRAM device based on titanium-silver alloy and preparation method and application thereof | |
| KR102198721B1 (en) | Transparent Photonic Artificial Synapse and Manufacturing Method thereof |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |