Cof-based memristor and preparation method thereof, and data security memory

Abstract

This invention discloses a COF-based memristor and its fabrication method, as well as a secure data storage device. The COF-based memristor comprises: a bottom electrode layer, a composite active dielectric layer, and a top electrode layer stacked sequentially; the composite active dielectric layer includes a dielectric matrix and a covalent organic framework material; the covalent organic framework material is uniformly dispersed in the dielectric matrix, and the pores of the molecular framework in the dielectric matrix and the covalent organic framework material form an interfacial barrier; the covalent organic framework material has charge trapping centers and forms trap energy levels of different depths. By applying a pulsed voltage, the device can be programmed to different conductance states by utilizing the filling and releasing mechanism of charge carriers inside the composite active dielectric layer in the trap energy levels. These conductance states exhibit a "time-window-limited" stability characteristic, that is, the data remains stable within the time window for read authentication, and then irreversible conductance decay automatically occurs under the drive of a physical mechanism until complete decorrelation.

Description

Technical Field

[0001] This invention relates to the field of memristor technology, and more particularly to a COF-based memristor, its fabrication method, and a data security memory. Background Technology

[0002] With the widespread application of IoT, edge computing, and AI technologies, hardware security has become a crucial cornerstone for ensuring the trustworthiness of information systems. Physically Unclonable Functions (PUFs), a technique that utilizes unavoidable random physical differences in semiconductor manufacturing processes to generate unique "digital fingerprints," are widely used in device authentication and key generation. Among various PUF implementation schemes, memristor-based PUFs have become a research hotspot in both academia and industry due to their advantages such as simple structure, high integration density, low power consumption, and good compatibility with CMOS (Complementary Metal Oxide Semiconductor) processes. Existing memristor PUF technologies typically utilize the random formation and breakage mechanisms of conductive filaments in materials such as metal oxides. By applying a specific voltage stimulus, the resistance state of the device is read, discretized into binary data as a response, thereby constructing a unique challenge-response pair (CRP).

[0003] However, existing memristor-based PUF technology has significant drawbacks in practical applications, especially in high-security or zero-trust environments. First, traditional technologies have long been designed to maximize the retention and stability of the device, aiming to keep the key constant throughout the device's lifespan. This "permanent memory" characteristic translates into a serious static security vulnerability when the device falls into a hostile environment or is physically captured (e.g., a crashed drone or lost sensor): attackers have an unlimited window of time to conduct non-intrusive probing, reverse engineering, or side-channel attacks, thereby extracting the fixed key information and compromising historical communication data. Second, most existing memristor PUFs can only exhibit binary (high / low impedance states) or a very limited number of intermediate states. The random information entropy provided by a single device is limited, and generating high-strength, long keys often requires a huge array area, resulting in a waste of hardware resources.

[0004] Furthermore, to address the security risks posed by physical capture, existing technologies typically rely on external control circuitry for active erasure or complex reset operations to destroy the key. This approach not only increases system power consumption and design complexity, but also means that in the event of power failure, malfunction, or a bypass attack, the remaining physical state can often still be recovered using highly sensitive probes, lacking a "self-forgetting" capability based on the intrinsic physical mechanisms of materials that requires no external energy.

[0005] Therefore, existing technologies still need improvement and development. Summary of the Invention

[0006] The technical problem to be solved by the present invention is to provide a COF-based memristor and its preparation method, as well as a data security memory, in order to address the above-mentioned deficiencies of the prior art and solve the problem of data security risks existing in the prior art.

[0007] The technical solution adopted by this invention to solve the technical problem is as follows: A COF-based memristor, comprising: a bottom electrode layer, a composite active dielectric layer, and a top electrode layer stacked sequentially; The composite active dielectric layer comprises a dielectric matrix and a covalent organic framework material; The covalent organic framework material is uniformly dispersed in the dielectric matrix, and the dielectric matrix forms an interfacial barrier with the pores of the molecular skeleton in the covalent organic framework material. The covalent organic framework material has charge trapping centers and forms trap energy levels at different depths.

[0008] The COF-based memristor wherein the covalent organic framework material is selected from at least one of B12C4TAPP-COF, TpPa-1, COF-LZU1, COF-1, or CTF-1.

[0009] The COF-based memristor wherein the dielectric material of the dielectric matrix is ​​selected from at least one of polymethyl methacrylate, polystyrene, polyvinylpyrrolidone, or polyimide; The pore diameter of the molecular skeleton in the covalent organic framework material is on the nanometer scale.

[0010] The COF-based memristor, wherein the time window of the COF-based memristor is determined according to the molecular weight or penetration ratio of the dielectric material.

[0011] The COF-based memristor has a bottom electrode layer with a thickness of 100nm~300nm, a top electrode layer with a thickness of 20nm~100nm, and a composite active dielectric layer with a thickness of 300nm~600nm.

[0012] A method for fabricating a COF-based memristor as described in any of the above claims, comprising the steps of: We provide bottom electrode layers, dielectric materials, covalent organic framework materials, solvents, and electrode materials. The dielectric material is dissolved in a solvent to obtain a dielectric material solution; the covalent organic framework material is dispersed in a solvent to obtain a covalent organic framework material solution. The dielectric material solution and the covalent organic framework material solution are mixed, spin-coated onto the bottom electrode layer, and annealed to form a composite active dielectric layer; Based on the electrode material, a top electrode layer is formed on the composite active dielectric layer to obtain a COF-based memristor.

[0013] The method for fabricating a COF-based memristor, wherein the mass ratio of the dielectric material to the covalent organic framework material is 30~15:1.

[0014] The method for fabricating a COF-based memristor, wherein the annealing temperature is 80℃~120℃ and the annealing time is 0.5h~2h.

[0015] The method for preparing a COF-based memristor, wherein dispersing the covalent organic framework material in a solvent to obtain a covalent organic framework material solution includes: The covalent organic framework material is placed in a solvent and sonicated to obtain a covalent organic framework material solution; wherein the sonication time is 12h~48h.

[0016] A data security memory, comprising: a COF-based memristor as described in any of the preceding claims.

[0017] Beneficial effects: By applying a pulsed voltage and utilizing the filling and releasing mechanism of charge carriers within the composite active dielectric layer in the trap energy level, the device can be programmed to different conductance states. These conductance states possess a "time-window-limited" stability characteristic, meaning that data remains stable within the time window for read authentication, and then irreversible conductance decay automatically occurs under physical mechanism until complete decorrelation. Based on the "self-forgetting" mechanism of the intrinsic physical relaxation of the composite active dielectric layer, a key self-destruction function is achieved without the need for external energy maintenance or active erasure circuitry. Attached Figure Description

[0018] Figure 1 This is a schematic diagram of the COF-based memristor structure in an embodiment of the present invention.

[0019] Figure 2 These are the evolution curves and box plots of the resolvable conductance energy levels of the memristor based on COF in this embodiment of the invention.

[0020] Figure 3 This is a schematic diagram of challenge-response pair generation and key conversion based on COF in an embodiment of the present invention.

[0021] Figure 4 This is a flowchart of the method for fabricating a COF-based memristor in an embodiment of the present invention.

[0022] Explanation of reference numerals in the attached figures: 1. Top electrode layer; 2. Composite active dielectric layer; 3. Bottom electrode layer; 4. Substrate layer. Detailed Implementation

[0023] To make the objectives, technical solutions, and advantages of this invention clearer and more explicit, the invention will be further described in detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative of the invention and are not intended to limit the invention.

[0024] Please also refer to Figures 1-3 This invention provides some embodiments of a COF-based memristor.

[0025] like Figure 1 As shown, the COF-based memristor of the present invention includes: a bottom electrode layer 3, a composite active dielectric layer 2, and a top electrode layer 1 stacked sequentially; the composite active dielectric layer 2 includes a dielectric matrix and a covalent organic framework (COF); the covalent organic framework is uniformly dispersed in the dielectric matrix, and the dielectric matrix and the pores of the molecular skeleton in the covalent organic framework form an interfacial barrier; the covalent organic framework has charge trapping centers and forms trap levels of different depths.

[0026] Specifically, in the composite active dielectric layer 2, the covalent organic framework material possesses a stable framework connected by covalent bonds and a precisely tunable pore structure. The lowest unoccupied molecular orbital (LUMO) and highest occupied molecular orbital (HOMO) of the covalent organic framework material form significant energy barriers with the bottom electrode layer 3, and also with the top electrode layer 1. The dielectric matrix exhibits insulating properties, and the covalent organic framework material is uniformly dispersed within it. The amorphous COF molecular domains act as isolated "trap islands" within the dielectric matrix. Specifically, the covalent organic framework material possesses dense deep-level charge trapping centers and forms trap levels of varying depths.

[0027] By applying pulsed voltages to the bottom electrode layer 3 and top electrode layer 1 of a COF-based memristor, and utilizing the filling and releasing mechanism of charge carriers within the material of the composite active dielectric layer 2 in the trap energy level, the device can be programmed to different conductance states. These conductance states possess a "time-window-limited" stability characteristic, meaning that data remains stable for authentication within a time window (e.g., more than 250 seconds), and then irreversibly decays until complete decorrelation occurs under physical mechanism. Based on the "self-forgetting" mechanism of the intrinsic physical relaxation of the composite active dielectric layer 2, a key self-destruction function is achieved without external energy maintenance or active erasure circuitry. This "use-and-burn" characteristic ensures that the key automatically expires after immediate authentication, and attackers cannot trace historical keys through physical detection even after the device is captured. This fundamentally solves the reverse engineering risks faced by static storage, making it particularly suitable for security scenarios with extremely high requirements for zero trust and physical capture prevention, such as satellite communications and drone networks.

[0028] The multi-conductance self-forgetting memristor architecture based on a composite material system of covalent organic framework materials and dielectric matrix is ​​fundamentally different from traditional metal oxide memristors in existing technologies that aim for long-term data retention. The internal conductivity mechanism no longer relies on permanent metal conductive filaments, but is based on the dynamic filling and thermodynamic release of charge carriers in the trap energy level. It is based on this physical mechanism that the device exhibits "time-bounded stability," meaning that within a specific time window after programming (e.g., 250 to 800 seconds), the device can maintain a high signal-to-noise ratio and high-stability conductivity state for retrieval. Once this window is exceeded, the stored information automatically undergoes irreversible physical relaxation and decorrelation, thus achieving a hardware-level "self-forgetting" characteristic.

[0029] Furthermore, traditional memristor PUFs (Physical Unclonable Functions) typically only utilize the high / low resistance states of the device for encoding, resulting in low information density and large array area overhead. This invention disperses covalent organic framework materials within a dielectric matrix, forming trap levels at different depths. Utilizing the inherent microscopic randomness in the device fabrication process, under uniform pulsed voltage excitation, each unit in the device array randomly falls into one of several distinct conductance levels. This directly maps multiple physical states to binary number encoding (such as...). Figure 3 As shown in the figure, this method increases the information entropy density of a single hardware unit by several times, enabling miniaturized arrays to generate high-strength long key sequences, significantly breaking through the application bottleneck of existing PUF technology in hardware resource-constrained scenarios.

[0030] Covalent organic framework (COF)-based memristors form trap levels of varying depths, trapping charge carriers through cumulative defects. When a low-amplitude or short-pulse voltage is applied to the device, thermally excited free electrons are injected from the bottom electrode layer 3 into the composite active dielectric layer 2, where they hop into the molecular domains of the covalent organic framework material within the dielectric matrix. The injected electrons are rapidly captured by charge trapping centers on the COF framework. This defect trapping is a cumulative process, not an instantaneous filling. Under continuous pulse stimulation, the trap sites are gradually filled piece by piece. As the number of trapped electrons increases, the quasi-Fermi level inside the device is gradually raised, and the carrier mobility changes continuously. Each pulse of a specific amplitude or width brings the trap filling to a new equilibrium state. This partial filling of traps macroscopically manifests as a continuous, stepwise increase in conductivity, thus precisely controlling dozens of distinct stable conductivity states (e.g., ...). Figure 2 (As shown).

[0031] Different signal modulation methods can be used in the excitation and encoding logic of key generation. For example, a voltage pulse with a fixed amplitude can be used as the excitation (challenge) or current pulse as the drive, or a combination of pulse sequences with different amplitudes and widths can be used to excite the device to produce different conductance distribution characteristics. In terms of encoding, in addition to mapping the steady-state conductance value to a binary number, the device conductance decay rate or the relaxation slope at a specific time point can also be used as a random entropy source for encoding.

[0032] In a preferred embodiment of the present invention, the covalent organic framework material is selected from at least one of COF materials with different pore sizes, such as B12C4TAPP-COF, TpPa-1, COF-LZU1, COF-1, or CTF-1.

[0033] Specifically, the structural formula of B12C4TAPP-COF is: .

[0034] In B12C4TAPP-COF, the electron-deficient central pore, the porphyrin ring center, and the structural defect sites of the crown-ether can all serve as charge-trapping centers. B12C4TAPP-COF is synthesized via the reaction of B12C4-CHO and TAPP. The specific reaction process is as follows: .

[0035] The structural formula of TpPa-1 is: .

[0036] The electron-deficient channels in TpPa-1 can serve as charge trapping centers. TpPa-1 is synthesized by reacting 1,3,5-trialdehyde phloroglucinol with p-phenylenediamine.

[0037] The structural formula of COF-LZU1 is as follows: .

[0038] The electron-deficient channels in COF-LZU1 can serve as charge trapping centers. COF-LZU1 is synthesized by reacting 1,3,5-trialdehydebenzene with p-phenylenediamine.

[0039] The structural formula of COF-1 is: .

[0040] The electron-deficient channels in COF-1 can serve as charge trapping centers. COF-1 is synthesized from terephthalic acid.

[0041] The structural formula of CTF-1 is: .

[0042] The electron-deficient channels in CTF-1 can serve as charge trapping centers. CTF-1 is synthesized from terephthalonitrile.

[0043] In a preferred embodiment of the present invention, the dielectric material of the dielectric matrix is ​​selected from at least one of polymethyl methacrylate, polystyrene, polyvinylpyrrolidone, or polyimide.

[0044] Specifically, the dielectric material of the dielectric matrix is ​​a polymer material with good insulation properties that forms an interfacial barrier with the pores of the molecular backbone in the covalent organic framework material, such as polymethyl methacrylate, polystyrene, polyvinylpyrrolidone, or polyimide.

[0045] In a preferred embodiment of the present invention, the pore diameter of the molecular skeleton in the covalent organic framework material is on the nanometer scale. For example, the pore diameter is 1 nm to 10 nm.

[0046] In a preferred embodiment of the present invention, the time window of the COF-based memristor is determined according to the molecular weight or penetration ratio of the dielectric material.

[0047] Specifically, by adjusting the molecular weight or penetration ratio of the dielectric material, the “forget time” window of the device can be finely adjusted (i.e., from hundreds of seconds to several seconds or hours) to adapt to different safety application scenarios.

[0048] In a preferred embodiment of the present invention, the mass ratio of the dielectric material to the covalent organic framework material is 30 to 15:1. For example, the mass ratio of the dielectric material to the covalent organic framework material is 20:1.

[0049] In a preferred implementation of this invention, such as Figure 1 As shown, the bottom electrode layer 3 includes at least one of an ITO (indium tin oxide) layer, a doped silicon layer, a titanium nitride layer, or a platinum (Pt) layer.

[0050] In a preferred implementation of this invention, such as Figure 1 As shown, the top electrode layer 1 includes at least one of a gold layer, a silver layer, an aluminum layer, or a graphene layer.

[0051] Specifically, both the bottom electrode layer 3 and the top electrode layer 1 employ strip electrodes, and the strip electrodes of the bottom electrode layer 3 and the top electrode layer 1 are arranged orthogonally. Effective working units are formed at the intersection points.

[0052] In a preferred implementation of this invention, such as Figure 1 As shown, the thickness of the bottom electrode layer 3 is 100nm~300nm, the thickness of the top electrode layer 1 is 20nm~100nm, and the thickness of the composite active dielectric layer 2 is 30nm~100nm. For example, the bottom electrode layer 3 is 185nm thick and has a linewidth of 100μm; the top electrode layer 1 is 50nm thick and has a linewidth of 100μm; and the composite active dielectric layer 2 has a thickness of 49nm.

[0053] In a preferred implementation of this invention, such as Figure 1 As shown, COF-based memristors can be fabricated on substrate 4. Substrate 4 can be a rigid substrate such as glass, or a flexible substrate such as polyethylene terephthalate (PET) or polyimide. Using a flexible substrate for substrate 4 allows for applications in wearable devices or flexible electronic skin.

[0054] Based on the COF-based memristor described in any of the above embodiments, the present invention also provides a preferred embodiment of a method for fabricating a COF-based memristor.

[0055] like Figure 4 As shown, the method for fabricating a COF-based memristor according to an embodiment of the present invention includes the following steps: Step S100: Provide a bottom electrode layer, a dielectric material, a covalent organic framework material, a solvent, and an electrode material; Step S200: Dissolve the dielectric material in a solvent to obtain a dielectric material solution; disperse the covalent organic framework material in a solvent to obtain a covalent organic framework material solution; Step S300: After mixing the dielectric material solution and the covalent organic framework material solution, spin-coat the mixture onto the bottom electrode layer and anneal it to form a composite active dielectric layer; Step S400: Based on the electrode material, a top electrode layer is formed on the composite active dielectric layer to obtain a COF-based memristor.

[0056] Specifically, a bottom electrode layer, a composite active dielectric layer, and a top electrode layer are prepared sequentially. The composite active dielectric layer is prepared by spin coating, resulting in a uniform thickness and absence of pinholes. The mass ratio of the dielectric material to the covalent organic framework material is 30~15:1. The annealing temperature is 80℃~120℃, and the annealing time is 0.5h~2h.

[0057] Step S200 specifically includes: Step S210: Place the dielectric material in a solvent and heat and stir to obtain a dielectric material solution.

[0058] Specifically, the dielectric material is dissolved by heating and stirring. The solvent can be chlorobenzene, and the solvent is selected according to the dielectric material.

[0059] Step S200 specifically includes: Step S220: Place the covalent organic framework material in a solvent and sonicate to obtain a covalent organic framework material solution; wherein the sonication time is 12h~48h.

[0060] Specifically, the covalent organic framework material is dissolved and exfoliated using ultrasound. The solvent can be chlorobenzene, and the solvent is selected based on the specific covalent organic framework material. Specific Implementation Example 1 (1) A commercial glass plate with a size of 2×2cm² was selected as the substrate. The substrate was placed in a beaker containing anhydrous ethanol, sealed, and ultrasonically cleaned for 15 minutes. After removal, it was rinsed with ultrapure water. Then, it was ultrasonically treated in ultrapure water and fresh anhydrous ethanol for 15 minutes each. The surface was dried with high-purity nitrogen and placed in a vacuum drying oven at 120°C for 60 minutes.

[0062] (2) Using magnetron sputtering technology, inert argon gas is introduced into a vacuum environment to generate plasma, and a customized metal mask (100 μm stripe width) is covered on the substrate surface. The sputtering time and power are precisely controlled to deposit an ITO conductive layer with a thickness of about 185 nm.

[0063] (3) Preparation of COF dispersion: Weigh the covalent organic framework material B12C4TAPP-COF powder and place it in chlorobenzene solvent (concentration of 1.2 mg / mL), and perform continuous ultrasonic treatment for 24 hours to exfoliate the layered structure. Preparation of PMMA matrix solution: Weigh the dielectric material polymethyl methacrylate PMMA particles and dissolve them in chlorobenzene (concentration of 28 mg / mL), heat and stir at 50℃ for 6 hours, and filter through a microporous membrane. Preparation of composite solution: Mix the above two solutions at a volume ratio of 1:1 (effective concentration of COF is 0.6 mg / mL, effective concentration of PMMA is 14 mg / mL), and perform ultrasonic treatment for 30 minutes before spin coating to eliminate phase separation. Spin coating film: Treat the substrate with ITO with ultraviolet ozone (UVO) for 30 minutes to improve wettability, then add 150 μL of the mixed solution dropwise, and spin coat at 3000 rpm for 40 seconds. Annealing treatment: The sample was transferred to a 100℃ heating stage for annealing for 1 hour to remove residual solvent and release stress, resulting in a composite active medium layer with a thickness of 49nm.

[0064] (4) Vacuum thermal evaporation technology is used instead of magnetron sputtering to avoid damage to the organic composite film by high-energy particles. A metal mask with the same specifications as the bottom electrode is used to fix its stripe direction perpendicular to the bottom electrode. In a chamber that maintains a high vacuum, a tungsten boat is heated by resistance to melt and evaporate high-purity gold, depositing a top electrode with a thickness of about 50 nm.

[0065] This invention utilizes the inherent randomness (physical non-cloning property) of memristor arrays to transform the analog conductance distribution of the device into a high-dimensional digital key sequence through a standardized "challenge-response-analog-digital conversion" process. The specific generation process can be divided into the following four core steps: Step 1: Array Excitation Injection (Challenge Stage). First, a 28×28 memristor crossbar array (containing 784 physical units) is selected as the hardware entropy source. Due to unavoidable microscopic physical fluctuations in the internal COF pore structure, PMMA polymer permeation, and active layer thickness during the micro / nano fabrication process, this provides the material basis for the randomness of the key. During operation, the peripheral circuitry applies a uniform programming voltage pulse (Challenge) synchronously or row-by-row to all devices in the array through the array's word lines (WL0 to WL27) and bit lines (BL0 to BL27). The pulse amplitude is 1.2V, and the pulse width is 1ms.

[0066] Step 2: Simulated Response Readout (Response Stage). After the programming stimulus is completed, the charge carriers inside the device are trapped in trap levels at different depths. Subsequently, the system switches to readout mode, applying an extremely weak and non-destructive readout voltage pulse (amplitude 0.01V, pulse width 10ms) to obtain the transient conductance state of each cell. Although all 784 devices receive the exact same programming stimulus, due to the random fluctuations in the aforementioned microfabrication, there are subtle differences in the ion migration and charge binding capabilities of each device. Therefore, the array exhibits a highly random conductance distribution in space (as shown in the 28×28 heatmap in the figure, with conductance values ​​ranging from 1×10⁻⁶). 5 -300×10 5 (Disordered distribution among S). Physically, each device will randomly and stably fall into one of 50 pre-characterized discrete conductance states.

[0067] Step 3: Signal Extraction and Analog-to-Digital Conversion (A to D Conversion Stage). The analog response currents (or conductivities) of the 784 devices in the array are read out sequentially, forming a vector sequence containing 784 analog values ​​(i.e., Figure 3 The analog signal sequence is then input into a high-precision analog-to-digital converter (ADC) for quantization. The ADC accurately maps and identifies the continuous analog currents as corresponding specific conductance states (State Index 1-50) based on pre-set threshold levels.

[0068] Step 4: High-Dimensional Digital Encoding and Final Key Generation (Generated Keys Stage). To convert the 50 physical states into standard digital ciphertext, the system employs a multi-bit binary encoding scheme. Since 50 states are greater than 2^35, the number of possible combinations is limited. 5 (32) and less than 2 6 (64) Therefore, 6 bits (6-bit) of binary data must be allocated to fully represent these 50 conductance levels (e.g., state 1 is encoded as 000000, state 2 as 000001, and so on). After encoding, each analog current value In in the array is converted into a 6-bit "0 / 1" bit stream. Finally, the entire 28×28 array (784 devices) will output a binary data matrix of size 6×784. Expanding this matrix yields a very long, high-entropy true random key sequence of length 4704 bits (i.e., 784×6).

[0069] like Figure 2As shown, the device can exhibit 50 different conductance states under pulse voltage stimulation. These states maintain high signal-to-noise ratio (SNR~50dB) and stability (correlation coefficient>0.998) within a finite time window of 250 to 800 seconds. Beyond this window, the conductance states undergo irreversible spontaneous decay and decorrelation.

[0070] like Figure 3 As shown, a 28×28 array (784 cells in total) is used to apply a uniform 1.2V, 1ms pulse excitation (Challenge) to all cells. The response current of each cell is read and converted into a conductance value. Due to microscopic fluctuations in the device fabrication process, each cell falls into one of 50 conductance states. The 50 conductance states are encoded into 6-bit binary numbers, and a 784×6-bit binary random sequence is finally output as the key. Specific Implementation Example 2 Unlike Specific Example 1, the covalent organic framework material used is TpPa-1, and the dielectric material is polystyrene. The concentration of TpPa-1 is 0.9 mg / mL, and the concentration of polystyrene is 28 mg / mL. The annealing temperature is 80°C, and the annealing time is 0.5 h. The ultrasonication time is 12 h. The thickness of the bottom electrode layer is 100 nm, the thickness of the top electrode layer is 20 nm, and the thickness of the composite active dielectric layer is 20 nm. Specific Implementation Example 3 Unlike Specific Example 1, the covalent organic framework material used is COF-LZU1, and the dielectric material is polyvinylpyrrolidone. The concentration of COF-LZU1 is 1.3 mg / mL, and the concentration of polyvinylpyrrolidone is 40 mg / mL. The annealing temperature is 120°C, and the annealing time is 2 hours. The ultrasonication time is 48 hours. The thickness of the bottom electrode layer is 300 nm, the thickness of the top electrode layer is 100 nm, and the thickness of the composite active dielectric layer is 100 nm. Specific Implementation Example 4 Unlike Specific Example 1, the covalent organic framework material used is COF-1, and the dielectric material is polyimide. The concentration of COF-1 is 1.5 mg / mL, and the concentration of polyimide is 22.5 mg / mL. Specific Implementation Example 5 Unlike Specific Embodiment 1, the covalent organic framework material used is CTF-1, and the dielectric material used is polyimide.

[0075] It should be understood that the application of the present invention is not limited to the examples above. Those skilled in the art can make improvements or modifications based on the above description, and all such improvements and modifications should fall within the protection scope of the appended claims.

Claims

1. A memristor based on COF, characterized in that, include: A bottom electrode layer, a composite active dielectric layer, and a top electrode layer are stacked sequentially. The composite active dielectric layer comprises a dielectric matrix and a covalent organic framework material; The covalent organic framework material is uniformly dispersed in the dielectric matrix, and the dielectric matrix forms an interfacial barrier with the pores of the molecular skeleton in the covalent organic framework material. The covalent organic framework material has charge trapping centers and forms trap energy levels at different depths.

2. The COF-based memristor according to claim 1, characterized in that, The covalent organic framework material is selected from at least one of B12C4TAPP-COF, TpPa-1, COF-LZU1, COF-1, or CTF-1.

3. The COF-based memristor according to claim 1, characterized in that, The dielectric material of the dielectric matrix is ​​selected from at least one of polymethyl methacrylate, polystyrene, polyvinylpyrrolidone, or polyimide; The pore diameter of the molecular skeleton in the covalent organic framework material is on the nanometer scale.

4. The COF-based memristor according to claim 3, characterized in that, The time window of the COF-based memristor is determined based on the molecular weight or permeation ratio of the dielectric material.

5. The COF-based memristor according to any one of claims 1 to 4, characterized in that, The thickness of the bottom electrode layer is 100nm~300nm, the thickness of the top electrode layer is 20nm~100nm, and the thickness of the composite active medium layer is 300nm~600nm.

6. A method for fabricating a COF-based memristor as described in any one of claims 1 to 5, characterized in that, Including the following steps: We provide bottom electrode layers, dielectric materials, covalent organic framework materials, solvents, and electrode materials. The dielectric material is dissolved in a solvent to obtain a dielectric material solution; the covalent organic framework material is dispersed in a solvent to obtain a covalent organic framework material solution. The dielectric material solution and the covalent organic framework material solution are mixed, spin-coated onto the bottom electrode layer, and annealed to form a composite active dielectric layer; Based on the electrode material, a top electrode layer is formed on the composite active dielectric layer to obtain a COF-based memristor.

7. The method for fabricating a COF-based memristor according to claim 6, characterized in that, The mass ratio of the dielectric material to the covalent organic framework material is 30~15:

1.

8. The method for fabricating a COF-based memristor according to claim 6, characterized in that, The annealing temperature is 80℃~120℃, and the annealing time is 0.5h~2h.

9. The method for fabricating a COF-based memristor according to claim 6, characterized in that, The step of dispersing the covalent organic framework material in a solvent to obtain a covalent organic framework material solution includes: The covalent organic framework material is placed in a solvent and sonicated to obtain a covalent organic framework material solution; wherein the sonication time is 12h~48h.

10. A data security storage device, characterized in that, include: The COF-based memristor as described in any one of claims 1 to 5.

Images