A ferroelectric material-based memristor device, a preparation method therefor, and an application thereof

Abstract

The application discloses a kind of memristor devices based on ferroelectric material and preparation method and application thereof, by laying bottom electrode, resistance change layer, top electrode, and its thickness, material is limited, can improve the conductivity and stability of resistance change layer and entire memristor device. Among them, the resistance change layer is ferroelectric material, by different current limiting test to the memristor device, different resistance state can be obtained, so that the resistance state of memristor device is more stable and can be used for ternary storage and ternary logic calculation. The memristor device provided in the application is based on the unpredictability of the conduction of the memristor inside. The conduction voltage of the memristor device is relatively dispersed, showing good randomness, and the resistance value decreases sharply when conducting, which is convenient for detection and can be used to generate random numbers. It can be applied to construct random number generator, and has wide application prospect.

Description

TECHNICAL FIELD

[0001] The present application relates to a ferroelectric material-based memristor device and a preparation method and application thereof, and belongs to the technical field of ferroelectric storage. BACKGROUND

[0002] A memristor is a two-port nonlinear passive electronic device representing the relationship between magnetic flux and electric charge. It is based on the resistance switching effect and can remember the amount of electric charge flowing through it, with advantages such as simple structure, easy integration, fast erase-write speed, low power consumption, and compatibility with CMOS (complementary metal-oxide-semiconductor) hybrid units.

[0003] The research on memristors is mainly divided by the material system used. The most mature system is based on oxide materials, and its working mechanism mainly relies on the migration and aggregation of ions and oxygen vacancies under the action of an electric field. Obviously, the process based on the migration of ions and oxygen vacancies is slower than that based on the migration of electrons, and is affected by thermal effects, resulting in poor resistance state stability and greater randomness, which limits the application and development of memristors in the fields of non-volatile storage and logic computing.

[0004] Ferroelectric materials are widely used materials with excellent resistance state stability, and have become a new research hotspot. The application of ferroelectric materials in memristors can produce high-performance memristor devices with stable resistance states, in order to build a storage and computing integrated architecture. BaFe 12 O 19 BaFe 12 O 19 is a ferroelectric material with hexagonal symmetry, belonging to the hexagonal system. Studies have found that BaFe 12 O 19 has better performance than other ferroelectric materials, showing good memory characteristics and switching characteristics, and the resistance state is more stable, and often has a lower operating voltage, which can greatly reduce the power consumption of the device.

[0005] Compared with the binary logic currently used in computers, ternary logic has many advantages: ternary digital logic can be fully compatible with binary digital logic, ternary logic has higher bit density, can realize more new logic functions, has higher A / D conversion and D / A conversion precision, can build all computer classic modules and can be made smaller, lighter and more powerful. Based on these advantages, ternary logic can greatly improve the speed and efficiency of information processing, and has wide application in quantum computers, optical computers, and fuzzy computing.

[0006] True random number generator (TRNG) has a wide range of applications in statistics, information security and other fields. In these fields, not only is it required that the data sequence is uniformly distributed and independent of each other, but also it is required to have unpredictability. The design of a new type of true random number generator should be based on digital circuits to meet the low cost, low power, miniaturization and portability requirements of true random number generators in industry. SUMMARY

[0007] The purpose of the present application is to overcome the deficiencies in the prior art, provide a ferroelectric material-based memristor device and its preparation method and application, by laying the bottom electrode, resistive switching layer, top electrode, and limiting its thickness and material, the conductivity and stability of the resistive switching layer and the entire memristor device can be improved, making the memristor device more stable and can be used for ternary storage and ternary logic calculation. On the other hand, its turn-on voltage can be used as a random source for a true random number generator.

[0008] To achieve the above purpose, the present application is implemented by using the following technical scheme:

[0009] In a first aspect, the present application provides a ferroelectric material-based memristor device, comprising a bottom electrode arranged on a substrate, a resistive switching layer arranged on the bottom electrode, and a top electrode arranged on the resistive switching layer, wherein the top electrode and the bottom electrode are used to connect with an external power supply.

[0010] The resistive switching layer is barium ferrite, the thickness of the top electrode is 90-110 nm, the thickness of the resistive switching layer is 50-70 nm, and the thickness of the bottom electrode is 70-90 nm.

[0011] Optionally, the top electrode and the bottom electrode are one of aluminum, molybdenum, niobium, copper, gold, palladium, platinum, tantalum, ruthenium, ruthenium oxide, silver, tantalum nitride, titanium nitride, tungsten, and tungsten nitride.

[0012] Optionally, the thickness of the top electrode is 100-105 nm, the thickness of the resistive switching layer is 55-60 nm, and the thickness of the bottom electrode is 80-90 nm.

[0013] Optionally, the bottom electrode is a platinum electrode, the top electrode is a copper electrode, the thickness of the copper electrode is 100 nm, the thickness of the resistive switching layer is 60 nm, and the thickness of the platinum electrode is 80 nm.

[0014] Optionally, the shape and size of the bottom electrode, the resistive switching layer and the substrate are matched one by one.

[0015] Optionally, the size of the bottom electrode and the resistive switching layer is 1-100 um.

[0016] In a second aspect, the present application provides a preparation method of the memristor based on ferroelectric material, comprising the following steps:

[0017] A first mask plate is arranged on the substrate, and the bottom electrode material is sputtered as a first sputtering source to the upper surface of the substrate in a vacuum environment to obtain a bottom electrode with a thickness of 70-90 nm;

[0018] The first mask plate is removed, the second mask plate is covered by aligning the mark, the first sputtering source is replaced by a second sputtering source containing barium ferrite, and the second sputtering source is sputtered to the bottom electrode to obtain a resistance change layer with a thickness of 50-70 nm arranged on the bottom electrode;

[0019] A third mask plate is arranged above the resistance change layer, the top electrode material is selected as a third sputtering source, and the top electrode material is sputtered to the resistance change layer to obtain a top electrode with a thickness of 90-110 nm, thereby obtaining the memristor based on ferroelectric material.

[0020] In a third aspect, the present application further provides an application of the memristor based on ferroelectric material in three-value logic storage and calculation.

[0021] Optionally, an external power supply is connected to the memristor, and the current size of the memristor based on ferroelectric material is limited to obtain different orders of magnitude of resistance.

[0022] In a fourth aspect, the present application further provides an application of the memristor based on ferroelectric material in a true random number generator.

[0023] Compared with the prior art, the present application has the following beneficial effects:

[0024] 1. The present application provides a memristor based on ferroelectric material, comprising a bottom electrode, a resistance change layer and a top electrode, the resistance change layer is barium ferrite, the thickness of the top electrode is 90-110 nm, the thickness of the resistance change layer is 50-70 nm, and the thickness of the bottom electrode is 70-90 nm. A layer of ferroelectric material film is covered on the bottom electrode to form the resistance change layer, which is conducive to the growth of conductive filaments. According to the conductive filament model of the resistance change device, the ferroelectric material improves the conductivity and stability of the resistance change layer and the entire memristor. At the same time, the thicknesses of the top electrode, the resistance change layer and the bottom electrode are limited, which can make the resistance state of the memristor more stable and can be used for three-value storage and three-value logic calculation. In addition, since the growth of conductive filaments is unpredictable, the device can also be used to construct a true random number generator.

[0025] 2. The present application provides a preparation method of a memristor based on ferroelectric material, which can be used for a memristor based on ferroelectric material. The preparation method is simple, efficient and low in cost, and is suitable for popularization and use in industry. BRIEF DESCRIPTION OF DRAWINGS

[0026] Figure 1 is a structural schematic diagram of a ferroelectric material-based memristor device provided by an embodiment of the present application.

[0027] Figure 2 is a Pt / BaFe 12 O 19 Preparation flowchart of a Pt / BaFe

[0028] Figure 3 is a typical I-V curve of a memristor device provided by an embodiment of the present application under a-1.5 V~2 V stimulus, 10 -4 A current-limiting time I- V curve diagram.

[0029] Figure 4 is a typical I-V curve of a memristor device provided by an embodiment of the present application under a-1.5 V~2 V stimulus, 10 -4 A current-limiting time 10 cycles I-V curve diagram.

[0030] Figure 5 is a typical I-V curve of a memristor device provided by an embodiment of the present application under a-4.5 V~5 V stimulus, 10 -1 A current-limiting time I- V curve diagram.

[0031] Figure 6 is a typical I-V curve of a memristor device provided by an embodiment of the present application under a-5 V~5 V stimulus, 10 -1 A current-limiting time 10 cycles I-V curve diagram.

[0032] Figure 7 is a typical resistance state retention diagram of a memristor device provided by an embodiment of the present application.

[0033] Figure 8 is a positive voltage scanning I-V curve of a memristor device provided by an embodiment of the present application under a 10 -2 A current-limiting time I-V curve diagram. DETAILED DESCRIPTION

[0034] The present application will be further described below in conjunction with the accompanying drawings and specific embodiments. The examples of the embodiments are shown in the accompanying drawings, and the specific embodiments described in the following embodiments of the present application are only exemplary descriptions of the specific embodiments of the present application, and are intended to explain the present application, but do not constitute a limitation on the present application.

[0035] The application provides a ferroelectric material-based memristor, which comprises a bottom electrode arranged on a substrate, a resistance change layer arranged on the bottom electrode, and a top electrode arranged on the resistance change layer. Figure 1 、 Figure 2 As shown in the figure, a bottom electrode is arranged on a substrate, and the substrate is a silicon substrate layer; the resistance change layer is a barium ferrite, and the resistance change layer is used to realize the conversion between resistance states; the top electrode comprises a plurality of top electrodes which do not intersect with each other; and the bottom electrode comprises a plurality of bottom electrodes which do not intersect with each other. The top electrode and the bottom electrode are used to be electrically connected with an external power supply, and the materials of the top electrode and the bottom electrode can be the same or different. A person skilled in the art can select one of aluminum, molybdenum, niobium, copper, gold, palladium, platinum, tantalum, ruthenium, ruthenium oxide, silver, tantalum nitride, titanium nitride, tungsten and tungsten nitride as the electrode material.

[0036] In the embodiment, the thickness of the top electrode is 90-110 nm, the thickness of the resistance change layer is 50-70 nm, and the thickness of the bottom electrode is 70-90 nm.

[0037] The top electrode of the application is prepared by a PVD (physical vapor deposition) method. Figure 1 As shown in the figure, the shapes and sizes of the resistance change layer, the bottom electrode and the substrate are matched and corresponded one by one. A person skilled in the art can also design the bottom electrode and the resistance change layer into different sizes according to actual needs. For example, the size of the bottom electrode and the resistance change layer can be 50 um or 80 um. A person skilled in the art can design the size of the bottom electrode and the resistance change layer to be 1-100 um.

[0038] The preparation method of the ferroelectric material-based memristor comprises the following steps.

[0039] Step one: a first mask plate is arranged on the substrate, a bottom electrode material is used as a first sputtering source to be sputtered and deposited on the upper surface of the substrate in a vacuum environment, and a bottom electrode with a thickness of 70-90 nm is prepared;

[0040] Step two: the first mask plate is removed, a second mask plate is covered by aligning the mark, a second sputtering source containing barium ferrite is used to replace the first sputtering source, and the second sputtering source is sputtered onto the bottom electrode to prepare a resistance change layer with a thickness of 50-70 nm arranged on the bottom electrode;

[0041] Step three: a third mask plate is arranged above the resistance change layer, a top electrode material is selected as a third sputtering source, and the top electrode material is sputtered to the resistance change layer to prepare a top electrode with a thickness of 90-110 nm, so as to obtain the ferroelectric material-based memristor.

[0042] As a preferred embodiment, the present invention provides a method for fabricating a memristor device with a structure of copper / ferroelectric material / platinum, wherein the bottom electrode is a platinum electrode, the top electrode is a copper electrode, the resistive switching layer is barium ferrite, and the memristor device is a 3×3 array. The fabrication steps are as follows:

[0043] Step 1: In a vacuum environment, the silicon substrate is fixed on the target gun of the sputtering system. Platinum is selected as the sputtering source, and a platinum electrode with a thickness of 80 nm is deposited by magnetron sputtering. The platinum electrode uniformly and completely covers the upper surface of the silicon substrate.

[0044] Step 2: Under vacuum conditions, platinum is fixed on the target gun of the sputtering system. Barium ferrite is selected as the sputtering source, and the second sputtering source is sputtered onto the platinum electrode to obtain a resistive switching layer with a thickness of 60 nm on the platinum electrode.

[0045] Step 3: Fix the resistive switching layer obtained in Step 2 onto the sputtering target gun, install a mask on top of the resistive switching layer, select a copper sputtering source, and sputter to deposit the top electrode. The thickness of the copper electrode is 100 nm, thereby preparing a memristor device based on ferroelectric materials.

[0046] The copper / ferroelectric / platinum memristor device prepared by this invention has the following technical effects: The performance of the memristor device can be determined by applying a DC voltage to it. Specifically, the top electrode is connected to the positive terminal of the power supply while the bottom electrode is grounded and a voltage is applied to it. When a positive voltage is applied, the voltage sweeps from 0 V to a set positive voltage and then back to 0 V. During this process, the dielectric layer of the memristor device undergoes a resistance transition effect due to voltage excitation, and the memristor device turns on, changing from a high-resistance state to a low-resistance state, i.e., the SET process. The voltage and current data saved at this time can be used to measure the corresponding SET process. I-V Curve; During the RESET process, the voltage is swept from 0 V to a set negative voltage and then back to 0 V. During this process, the memristor is stimulated by the reverse voltage and changes from a low-resistance state to a high-resistance state. The corresponding curve can also be measured. I-V Curve. Memristor device. I-V The more consistent the curve shape, the better its stability.

[0047] Figure 3 The memristor device of this invention, under stimulation of -1.5 V to 2 V, 10 -4 Typical case of A-rate limiting I-V The graph shows the forward scan-stop voltage during the SET process set to 0 V to 2 V with a step size of 0.2 V. The reverse scan-stop voltage during the RESET process is set to 0 V to -1.5 V with a step size of 0.2 V. Both forward and reverse scans are bidirectional. As shown in the graph, the memristor device exhibits typical resistive switching characteristics, with an on / off ratio reaching 10.2 .

[0048] Figure 4 is the I-V curve of the memory device of the present application under the stimulus of-1.5 V ~ 2 V and 10 -4 A current limit I-V The forward scan-stop voltage in the SET process is set to 0 V to 2 V with a step of 0.2 V. The reverse scan-stop voltage in the RESET process is set to 0 V to-1.5 V with a step of 0.2 V. Both the forward and reverse are bidirectional scanning. The above scanning process is cycled 10 times. As can be seen from the figure, the memory device of the present application shows high stability, and the I-V curve of each cycle is roughly the same.

[0049] Figure 5 is the I-V curve of the memory device of the present application under the stimulus of-4.5 V ~ 5 V and 10 -1 A current limit I-V The forward scan-stop voltage in the SET process is set to 0 V to 5 V with a step of 0.2 V. The reverse scan-stop voltage in the RESET process is set to 0 V to-4.5 V with a step of 0.2 V. Both the forward and reverse are bidirectional scanning. As can be seen from the figure, the memory device shows typical resistive switching characteristics, and the on-off ratio reaches 10 5 .

[0050] Figure 6 is the I-V curve of the memory device of the present application under the stimulus of-5 V ~ 5 V and 10 -1 A current limit I-V The forward scan-stop voltage in the SET process is set to 0 V to 5 V with a step of 0.2 V. The reverse scan-stop voltage in the RESET process is set to 0 V to-5 V with a step of 0.2 V. Both the forward and reverse are bidirectional scanning. The above scanning process is cycled 10 times. As can be seen from the figure, the memory device of the present application shows high stability, and the I-V curve of each cycle is roughly the same.

[0051] Figure 7 is the typical resistance state diagram of the memory device of the present application. As can be seen from the figure, the memory device of the present application shows relatively stable DC characteristics. The resistance of the low resistance state of the memory device of the present application is about 10 1 Ω, the resistance of the medium-low resistance state is about 10 3 Ω, and the resistance of the high resistance state is about 10 5 Ω, and the on-off ratio is at least about 10 2 times. At the same time, it also shows that the memory device of the present application has good durability under the current limit, the switching times are large enough to stably store data, and the memory device of the present application can be used as a nonvolatile memory device. 2The three resistance states can be well distinguished by the switch ratio of 10 times, and the accuracy of the memory storage of the memory device is better.

[0052] Figure 8 The memory device of the application has a 10 -2 When A is current limiting, the forward voltage scanning is I-V The curve diagram is shown in the figure. The forward scanning-stop voltage in the SET process is set to 0 V to 5 V with a step of 0.2 V. As shown in the figure, the SET voltage of the memory device of the application is uniformly distributed between 2.5 V and 4.5 V, and good randomness is shown.

[0053] It can be known from the combination of Figure 3 , Figure 4 , Figure 5 , Figure 6 and Figure 7 that the memory device of the application has the typical characteristics of a memory device, has good stability and non-volatility, and can be used for three-value logic calculation.

[0054] It can be known from the combination of Figure 8 that the memory device of the application has good conduction stability, good randomness and uniformity of conduction voltage, and can be used to construct a true random number generator.

[0055] In summary, the memory device based on ferroelectric material provided by the application includes a bottom electrode, a resistance change layer and a top electrode, the resistance change layer is barium ferrite, the thickness of the top electrode is 90-110 nm, the thickness of the resistance change layer is 50-70 nm, and the thickness of the bottom electrode is 70-90 nm. The memory device can be applied to three-value logic storage and calculation and construction of a true random number generator.

[0056] Based on the unpredictability of the conduction of the memory device, the conduction voltage of the memory device is relatively dispersed, good randomness is shown, and the resistance value sharply decreases when conduction, which is convenient for detection, can be used to generate random numbers, can be applied to construct a random number generator, and has a wide application prospect.

[0057] The resistance change layer is formed by covering a layer of ferroelectric material film on the bottom electrode, which is conducive to the growth of conductive filaments. According to the conductive filament model of the resistance change device, the ferroelectric material improves the conductivity and stability of the resistance change layer and the entire memory device. Meanwhile, the thicknesses of the top electrode, the resistance change layer and the bottom electrode are limited, which can make the resistance state of the memory device more stable and can be used for three-value storage and three-value logic calculation. Furthermore, since the growth of the conductive filaments is unpredictable, the device can also be used to construct a true random number generator.

[0058] The above merely describes the preferred embodiments of the present application, and it should be pointed out that, for those skilled in the art, several improvements and modifications can be made without departing from the technical principles of the present application, and these improvements and modifications should also be considered as the protection scope of the present application.

Claims

1. Use of ferroelectric material based memristive devices in ternary logic computation, characterized in that, The ferroelectric material-based memristor device comprises a bottom electrode arranged on a substrate, a resistance change layer arranged on the bottom electrode, and a top electrode arranged on the resistance change layer, wherein the bottom electrode and the top electrode are used to be connected with an external power supply; The substrate is a silicon substrate layer, the resistance change layer is a barium ferrite, the top electrode comprises a plurality of top electrodes which do not cross each other, and the bottom electrode comprises a plurality of bottom electrodes which do not cross each other; the thickness of the top electrode is 90-110 nm, the thickness of the resistance change layer is 50-70 nm, and the thickness of the bottom electrode is 70-90 nm; the thickness of the top electrode is 100-105 nm, the thickness of the resistance change layer is 55-60 nm, and the thickness of the bottom electrode is 80-90 nm; the bottom electrode is a platinum electrode, the top electrode is a copper electrode, the thickness of the copper electrode is 100 nm, the thickness of the resistance change layer is 60 nm, and the thickness of the platinum electrode is 80 nm; the bottom electrode, the resistance change layer and the substrate are matched in shape and size; the size of the bottom electrode and the resistance change layer is 1-100 um.

2. Use according to claim 1, characterized in that, The preparation method of the ferroelectric material-based memristor device comprises the following steps: A first mask plate is arranged on the substrate, a bottom electrode material is used as a first sputtering source to be sputtered and deposited on the upper surface of the substrate in a vacuum environment, and a bottom electrode with a thickness of 70-90 nm is prepared; The first mask plate is removed, a second mask plate is covered by aligning the mark, a second sputtering source containing barium ferrite is used to replace the first sputtering source, and the second sputtering source is sputtered onto the bottom electrode to prepare a resistance change layer with a thickness of 50-70 nm arranged on the bottom electrode; A third mask plate is arranged above the resistance change layer, a top electrode material is selected as a third sputtering source, and the top electrode material is sputtered to the resistance change layer to prepare a top electrode with a thickness of 90-110 nm, thereby obtaining the ferroelectric material-based memristor device.

3. Use according to claim 1, characterized in that, The external power supply is connected to the memristor device, and the current size of the ferroelectric material-based memristor is limited to obtain different orders of magnitude of resistance values.

4. Application of ferroelectric material-based memristor in true random number generator, the ferroelectric material-based memristor comprising a bottom electrode arranged on a substrate, a resistive switching layer arranged on the bottom electrode, and a top electrode arranged on the resistive switching layer, the top electrode and the bottom electrode being used to be connected with an external power source; the substrate is a silicon substrate layer, the resistive switching layer is barium ferrite, the top electrode comprises a plurality of top electrodes that do not cross each other, and the bottom electrode comprises a plurality of bottom electrodes that do not cross each other; the thickness of the top electrode is 90-110 nm, the thickness of the resistive switching layer is 50-70 nm, and the thickness of the bottom electrode is 70-90 nm; the thickness of the top electrode is 100-105 nm, the thickness of the resistive switching layer is 55-60 nm, and the thickness of the bottom electrode is 80-90 nm; the bottom electrode is a platinum electrode, the top electrode is a copper electrode, the thickness of the copper electrode is 100 nm, the thickness of the resistive switching layer is 60 nm, and the thickness of the platinum electrode is 80 nm; the bottom electrode, the resistive switching layer, and the substrate are matched in shape and size; and the size of the bottom electrode and the resistive switching layer is 1-100 um.

Images