A relaxation type memristor and its preparation method and application

Abstract

The present invention belongs to the technical field related to semiconductor devices and discloses a relaxor-type memristor, its preparation method, and application. The memristor comprises a bottom electrode layer, a resistive switching layer, and a top electrode layer arranged from bottom to top. The top electrode layer is made of an inactive metal element or an alloy containing an inactive metal; the resistive switching layer is an inorganic molecular crystal thin film. The memristor uses an inactive metal as the material for the top electrode layer and controls the growth of conductive filaments in the inorganic molecular crystal resistive switching layer to achieve a stable and adjustable conductance state. This further enables precise mapping of analog electrical signals, providing higher precision for memristor-based neuromorphic computing.

Description

Technical Field

[0001] The present invention belongs to the technical field related to semiconductor devices, and more specifically, relates to a relaxation type memristor and a preparation method and application thereof. Background Art

[0002] As a new type of electronic device with resistive memory function, the resistance of the memristor can be reversibly switched with external voltage or current stimulation, thereby storing and processing information in the form of resistance. Thanks to its inherent dynamic characteristics and nonlinear resistive behavior, the memristor has become a strong candidate for realizing artificial neural networks at the hardware level, showing great potential in simulating the plasticity of synapses and performing complex computing tasks.

[0003] Most reported memristors achieve resistance switching using conductive filaments made of active metals (such as Cu and Ag). The sudden change in conductance during the formation and breakage of the filaments limits the number of adjustable conductance states and reduces the linearity of the system. This limited number of conductance states cannot fully reflect the original data information, resulting in erroneous signal output during computational tasks, seriously affecting computing performance.

[0004] Analog encoding based on multiple resistance states can more accurately preserve the original characteristics of the data, which is crucial for improving the recognition accuracy of neural networks, making neuromorphic computing suitable for a wider range of application scenarios and more complex computing tasks. Therefore, current research and development focuses on developing memristors with adjustable conductance states and excellent short-term dynamic characteristics, and ensuring that their internal storage states accurately map the characteristic information of the input data. Summary of the Invention

[0005] In response to the above defects or improvement needs of the prior art, the present invention provides a relaxation-type memristor and a preparation method and application thereof, which aims to solve the problem of limited number of conductance states of existing memristors.

[0006] To achieve the above objectives, according to one aspect of the present invention, a relaxation-type memristor is provided, wherein the memristor includes a bottom electrode layer, a resistive switching layer, and a top electrode layer arranged from bottom to top, wherein the material of the top electrode layer is an inactive metal element or an alloy containing an inactive metal; and the resistive switching layer is an inorganic molecular crystal film.

[0007] Furthermore, the inactive metal element is any one of antimony, gold, ruthenium and platinum.

[0008] Furthermore, the alloy contains two or more of antimony, gold, ruthenium, and platinum, and the atomic content of any one of the doping components is greater than 5%.

[0009] Furthermore, the thickness of the top electrode layer is 5 nm-200 nm.

[0010] Furthermore, the material of the inorganic molecular crystal thin film is one of Sb2O3 and As2O3, and the thickness of the resistive layer is 5 nm-100 nm.

[0011] Furthermore, the material of the bottom electrode layer is any one of Pt, Cr, TiN and graphite.

[0012] Furthermore, the material of the bottom electrode layer is Pt, Cr or TiN, and its thickness is 5 nm-50 nm.

[0013] Furthermore, the bottom electrode layer is made of graphite, and its thickness is 0.3 nm-30 nm.

[0014] The present invention provides a method for preparing a relaxation type memristor, and the method is used to prepare the relaxation type memristor as described above.

[0015] The present invention also provides an application of the relaxation-type memristor described above in neuromorphic computing.

[0016] In general, compared with the prior art, the above technical solutions conceived by the present invention provide a relaxor memristor and its preparation method and application, which have the following beneficial effects:

[0017] 1. This invention uses inactive metals such as antimony (Sb), gold (Au), ruthenium (Ru), or platinum (Pt) as the top electrode layer of the memristor. Based on the electrochemical metallization resistive switching mechanism, an electric field drives the migration and redox reactions of metal ions in the resistive switching layer, resulting in the formation or breakage of conductive filaments, thereby achieving switching between high and low resistance states. Unlike other active metal top electrodes (such as silver, copper, and titanium), inactive metals have higher diffusion activation energies, resulting in slower growth of conductive filaments under an applied electric field. The device conductance gradually changes during the resistive switching process, resulting in a large number of distinguishable conductance states and precise control of the memristor's resistance state.

[0018] 2. The application of this invention will overcome the limitation of neuromorphic computing that it is difficult to directly process analog signals through hardware, greatly improve the accuracy of complex signal processing, and provide a feasible solution for the hardware implementation of brain-like neuromorphic computing. BRIEF DESCRIPTION OF THE DRAWINGS

[0019] Figure 1 Schematic diagram of the structure of the memristor in Example 1 of the present invention.

[0020] Figure 2 2 is a graph showing the IV characteristics of the memristor in Example 1.

[0021] Figure 3 2 is a relaxation characteristic diagram of the memristor in Example 1.

[0022] Figure 4 3 is a pulse response characteristic diagram of the memristor in Example 1.

[0023] Figure 5 5-bit pulse coding diagram of the memristor in Example 1. DETAILED DESCRIPTION

[0024] In order to make the objectives, technical solutions and advantages of the present invention more clearly understood, the present invention is 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 for the purpose of explaining the present invention and are not intended to limit the present invention. In addition, the technical features involved in the various embodiments of the present invention described below may be combined with each other as long as they do not conflict with each other.

[0025] The present invention provides a relaxation-type memristor, which uses an inactive metal as the material of the top electrode layer of the memristor and controls the growth process of conductive filaments in the inorganic molecular crystal resistive layer to obtain a stable and adjustable conductivity state. It will further realize the precise mapping of analog electrical signals and provide higher accuracy for neuromorphic computing based on memristors.

[0026] The memristor comprises a bottom electrode layer, a resistive switching layer and a top electrode layer arranged from bottom to top. The material of the top electrode layer is an inactive metal element or an alloy containing an inactive metal; the resistive switching layer is an inorganic molecular crystal film.

[0027] The inactive metal element is any one of antimony (Sb), gold (Au), ruthenium (Ru), and platinum (Pt); the alloy contains two or more of antimony (Sb), gold (Au), ruthenium (Ru), and platinum (Pt), and the atomic content of any doping component is greater than 5%. The thickness of the top electrode layer is 5 nm-200 nm.

[0028] The material of the inorganic molecular crystal film is one of Sb2O3 and As2O3, and the thickness of the resistive layer is 5 nm-100 nm.

[0029] The material of the bottom electrode layer includes Au, Pt, Cr, TiN and graphite; when the material of the bottom electrode layer is Pt, Cr or TiN, its thickness is 5 nm-50 nm; when the material of the bottom electrode layer is graphite, its thickness is 0.3 nm-30 nm.

[0030] The present invention also provides a method for preparing the relaxation-type memristor as described above, the method comprising the following steps:

[0031] (1) Depositing a bottom electrode layer on a substrate or forming a bottom electrode layer by mechanically exfoliating a few-layer graphene.

[0032] (2) Prepare an inorganic molecular crystal film on the bottom electrode to form a resistive switching layer.

[0033] (3) Depositing an inactive metal element or an alloy containing an inactive metal on the resistive layer to form a top electrode layer.

[0034] Methods for depositing an inactive metal element on the resistive layer include: forming a thin film on a substrate by thermal evaporation deposition, electron beam evaporation deposition, or magnetron sputtering deposition; methods for depositing an alloy containing an inactive metal on the resistive layer include: forming an alloy film on a substrate by multi-source evaporation, evaporation of an alloy metal source, or alternating deposition of different metal films and annealing.

[0035] In one embodiment, the resistive layer is a Sb2O3 or As2O3 thin film grown directly on the bottom electrode layer by thermal evaporation deposition.

[0036] The bottom electrode layer is one of Au, Pt, Cr, TiN deposited by thermal evaporation, electron beam evaporation or magnetron sputtering, or mechanically exfoliated graphene.

[0037] The present invention also provides an application of the relaxation-type memristor described above in neuromorphic computing, including pulse signal processing, time series data prediction and image classification.

[0038] The present invention is further described in detail below with reference to specific embodiments.

[0039] Example 1

[0040] This embodiment 1 provides a memristor device, the structure of which is as follows Figure 1 As shown, the top electrode layer is a metal Sb single-element thin film.

[0041] The method for preparing the memristor device in this embodiment 1 includes the following steps:

[0042] (1) A silicon wafer was used as the substrate and treated with oxygen plasma. The plasma treatment conditions were: oxygen flow rate of 20 sccm, power of 80 W, and cleaning time of 10 min. A tape with mechanically exfoliated graphene was then attached to the substrate surface. The substrate was placed on a hot plate and heated at 100 °C for 2 min. After heating, the tape was removed and the graphene was transferred to the substrate surface. A thinner portion of graphene was selected under a light microscope as the bottom electrode.

[0043] (2) Electron beam lithography was used to prepare a lead pattern on graphene. The photoresist used was PMMA, which was spun on and then baked at 150°C for 5 minutes. After exposure, the sample was immersed in a developer for 5 to 10 seconds, then immersed in a fixer for 10 seconds and dried with a nitrogen gun.

[0044] (3) Cr / Au films were deposited on the lead pattern prepared in step (2) by thermal evaporation for test pinning. The thickness of the Cr and Au films were 10 nm and 50 nm, respectively. The process conditions were: Cr or Au particles were used as the evaporation source, nitrogen was used as the evaporation atmosphere, the evaporation rate was 0.2 Å / s, and the chamber pressure was less than 8×10 -4 After coating, the sample with the bottom electrode was placed in acetone, and the acetone was heated to 60 ℃-70 ℃ and soaked for 20 min to complete the degumming.

[0045] (4) Using thermal evaporation deposition, a layer of Sb2O3 was deposited on the bottom electrode as a resistive switching layer. The thickness of the Sb2O3 film was 10 nm. The process conditions were: Sb2O3 powder was used as the evaporation source, nitrogen was used as the evaporation atmosphere, the evaporation rate was 0.1 Å / s, and the chamber pressure was less than 8×10 -4 Pa.

[0046] (5) Prepare the top electrode pattern on the Sb2O3 resistive switching layer using electron beam lithography, and the process conditions are the same as those in step (2).

[0047] (6) Using thermal evaporation deposition, a layer of Sb film is deposited on the top electrode pattern prepared in step (5) as the top electrode layer. The thickness of the Sb film is 40 nm. The process conditions are: Sb particles are used as the evaporation source, nitrogen is used as the evaporation atmosphere, the evaporation rate is 0.1 Å / s, and the chamber pressure is less than 8×10 -4 Pa.

[0048] (7) Using thermal evaporation deposition, a layer of Au film was deposited on the top electrode prepared in step (6) as a protective layer. The thickness of the Au film was 20 nm. The process conditions were: Au particles as the evaporation source, nitrogen as the evaporation atmosphere, an evaporation rate of 0.2 Å / s, and a chamber pressure of less than 8×10 -4 After the coating, the sample with the top electrode deposited thereon was placed in acetone, and the acetone was heated to 60° C.-70° C. and immersed for 20 minutes to complete the debonding, thereby obtaining the memristor.

[0049] Result Analysis

[0050] The IV characteristic curve of the memristor based on the Sb single-element top electrode layer in this embodiment 1 is as follows: Figure 2 As shown in Figure 2, at a lower current limit (100 nA), the device exhibits typical non-threshold volatile resistive switching characteristics, with conductance gradually changing during set and reset. The set voltage of the device is 2.5 V, and the on / off ratio is about 10. 4 , with distinct high and low resistance states.

[0051] The relaxation curve of the memristor based on the Sb single-element top electrode layer in Example 1 is as follows: Figure 3 As shown in the figure, a 2 V stimulation pulse was applied to the device, and then the device current was continuously read with a voltage of 0.1 V. It was observed that the current gradually decayed with time, and the relaxation time was about 5 μs; this shows that the device has typical timing dependence characteristics and can be used for timing signal processing.

[0052] The pulse response characteristics of the Sb single-element top electrode layer memristor in this embodiment 1 are as follows: Figure 4 As shown in Figure 2. 5 identical pulses (pulse amplitude of 3.5 V, pulse width of 10 μs) were applied continuously with an interval of 2 μs and the dynamic response of the device was detected. The current gradually increased with the accumulation of pulses. Furthermore, a 5-bit pulse sequence was used to modulate the current of the memristor, and 32 distinguishable conductance states were obtained. The results are shown in Figure 2. Figure 5 The above experimental results show that the conductivity of the device can be precisely adjusted by electric pulses.

[0053] Example 2

[0054] This embodiment 2 uses the same steps as embodiment 1 to prepare a memristive device. The difference from embodiment 1 is that the top electrode is an alloy film deposited by a co-sputtering method, wherein the base metal is Au, the doped inactive metal is Sb, and the Sb doping atomic ratio is 10%.

[0055] It will be easily understood by those skilled in the art that the above description is merely a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, and improvements made within the spirit and principles of the present invention should be included in the scope of protection of the present invention.

Claims

1. A relaxation memristor, characterized in that: The memristor comprises a bottom electrode layer, a resistive switching layer and a top electrode layer arranged from bottom to top, wherein the material of the top electrode layer is an inactive metal element or an alloy containing an inactive metal; the resistive switching layer is an inorganic molecular crystal film; The inactive metal element is any one of antimony, gold, ruthenium, and platinum; the alloy contains two or more of antimony, gold, ruthenium, and platinum, and the atoms of any doping component are greater than 5%; the material of the inorganic molecular crystal film is one of Sb2O3 or As2O3.

2. The relaxation memristor according to claim 1, wherein: The thickness of the top electrode layer is 5 nm-200 nm.

3. The relaxor memristor according to claim 1, wherein: The thickness of the resistive switching layer is 5 nm-100 nm.

4. The relaxation memristor according to claim 1, wherein: The material of the bottom electrode layer is any one of Pt, Cr, TiN and graphite.

5. The relaxor memristor according to claim 1, wherein: The bottom electrode layer is made of Pt, Cr or TiN, and has a thickness of 5 nm to 50 nm.

6. The relaxor memristor according to claim 1, wherein: The bottom electrode layer is made of graphite, and its thickness is 0.3 nm to 30 nm.

7. Use of the relaxor memristor according to any one of claims 1 to 6 in neuromorphic computing.

Images