Method for improving performance of two-dimensional MXene artificial synapse

Abstract

The invention discloses a method for improving the performance of a two-dimensional MXene artificial synapse, relates to the technical field of semiconductor devices, and aims to solve the problem of how to improve the performance of the artificial synapse in the aspects of information processing capability, learning efficiency and adaptability. The method specifically comprises the following steps: S1, accurately preparing two-dimensional MXene by a liquid phase etching method; s2, preparing different ionic electrolytes; and S3, constructing a multi-gate two-dimensional MXene artificial synapse.

Description

TECHNICAL FIELD

[0001] The application relates to the technical field of semiconductor devices, and in particular to a method for improving the performance of a two-dimensional MXene artificial synapse. BACKGROUND

[0002] Artificial intelligence is increasingly becoming a core technology leading a new round of technological revolution and industrial change, and has wide penetration, and has a revolutionary influence on human science and technology, economy and social development. With the explosive increase in data volume and the continuous rise in computing demand, the transmission computing architecture is facing the severe challenge of high energy consumption and low computing power, mainly because the computing architecture adopts the von Neumann system of separation of storage and calculation, and more than 70% of energy consumption is consumed in data transfer.

[0003] The human brain is the most complex computing system in nature, with tens of billions of synapses, and can complete complex cognitive tasks at 20W extremely low power consumption, and has high parallel processing capability, plasticity, nonlinear computing characteristics and flexible learning mechanism. Unlike the von Neumann system, the human brain transmits information through synaptic connections between neurons, and can efficiently process information without a central processing unit. Therefore, developing brain-like bionic artificial synapse devices has become an important direction of current scientific research.

[0004] Artificial synapses are the best choice for realizing neuromorphic computing due to their low power consumption, high density and non-volatility, and are expected to overcome the limitations of the von Neumann architecture. The activity of biological synapses mainly depends on the change of ion concentration, and the migration of ions plays an important role in information storage and processing. By simulating the ion flow characteristics in biological synapses, the learning ability and adaptability of artificial neural networks can be enhanced, making them more efficient in processing complex tasks.

[0005] MXene is a new member of the two-dimensional material family, which can be represented by the formula Mn+1XnTx(n=1-3), where M is a transition metal (Ti, V, Cr, etc.), X is carbon or nitrogen, and Tx represents surface functional groups (-O or -OH, etc.). MXene has a layered structure and can provide channels for ion migration, and can accommodate the embedding and extraction of various ions. At the same time, the abundant functional groups on the surface provide a large number of active sites, which can significantly improve the adsorption and reaction efficiency of ions, and show great potential for application in ion migration type artificial synapses.

[0006] Currently, the working mechanism of ion migration type artificial synapse mainly relies on single ion migration, and the adjustment ability provided by single ion migration is limited, which cannot meet the demand of complex tasks. Multiple ion migration can provide more abundant signal processing mechanism through the synergistic effect between different ions, enhance the nonlinear response ability of the system, and show higher efficiency and accuracy, thereby improving the performance of artificial synapse in information processing ability, learning efficiency and adaptability, and promoting the further development of neuromorphic computing. SUMMARY

[0007] The present application aims to solve the problem of how to improve the performance of artificial synapse in information processing ability, learning efficiency and adaptability, and proposes a method for improving the performance of two-dimensional MXene artificial synapse, which specifically comprises the following steps: S1: precisely preparing two-dimensional MXene by liquid etching method; S2: preparing different ion electrolytes; S3: constructing multi-gate two-dimensional MXene artificial synapse.

[0008] The present application further provides that step S1 specifically comprises the following steps: S1-1: adding 5ml of deionized water, 15ml of hydrochloric acid and 1-3g of lithium fluoride in a polytetrafluoroethylene beaker, stirring for half an hour, then adding 1g of MAX phase, and etching at 30-60℃ for 24-72h; S1-2: centrifugal washing with deionized water for 4-6 times until the PH is neutral, and then ultrasonic treatment with an ultrasonic machine with a power of 100-300W for 20-60min; S1-3: centrifuging at a speed of 2000rpm-5000rpm for 20-60min, and taking the supernatant as the two-dimensional MXene solution.

[0009] The present application further provides that the ion electrolyte in step S2 comprises proton electrolyte and alkali metal ion electrolyte.

[0010] The present application further provides that the alkali metal ion electrolyte is one of lithium ion electrolyte, sodium ion electrolyte and potassium ion electrolyte.

[0011] The present application further provides that step S2 specifically comprises the following steps: S2-1: preparation of proton electrolyte: dissolving 1g of polyvinyl alcohol in 10ml of deionized water, stirring the obtained solution at 80-95℃ for 10-24h until it becomes transparent, and cooling to room temperature. Then slowly add 0.1-1g of sulfuric acid; S2-2: Preparation of lithium ion electrolyte or sodium ion electrolyte or potassium ion electrolyte: 10-150 mg of lithium perchlorate or sodium perchlorate or potassium perchlorate is weighed and mixed with 250 mg of polyethylene oxide (PEO) in a glass container, then 10 ml of acetonitrile solvent is injected into the container, and continuous stirring is carried out on a hot plate at 80-95℃ for 12-48h until all solid particles are completely dissolved in acetonitrile.

[0012] Further provided by the application is that step S3 specifically comprises the following steps: S3-1: patterning of the second gate: photoetching of the silicon wafer by a photoetching machine and a mask plate, patterning of the second gate by spin-coating negative photoresist, pre-baking, post-baking and developing, hot evaporation of 5-15nm Cr and 50-80nm Au, soaking in 100℃ acetone solution for 30min for photoresist removal, then spin-coating or drop-coating proton electrolyte on the second gate; S3-2: patterning of the channel material: photoetching of the silicon wafer by a photoetching machine and a mask plate, hollowing out the channel area by spin-coating positive photoresist, pre-baking, post-baking and developing, then spin-coating a solution with a concentration of 5-15mg / ml at 95℃, soaking in acetone solution for 30min for photoresist removal, and drop-coating lithium ion electrolyte on the channel material; S3-3: patterning of the source electrode, the drain electrode and the first gate: photoetching of the silicon wafer by a photoetching machine and a mask plate, patterning of the source electrode and the drain electrode by spin-coating negative photoresist, pre-baking, post-baking and developing, evaporation of 10-40nm inert metal as the drain electrode, then spin-coating or drop-coating lithium ion electrolyte or sodium ion electrolyte or potassium ion electrolyte on the channel material MXene, and finally patterning of the first gate by photoetching, and evaporation of 10-40nm inert metal as the first gate.

[0013] Further provided by the application is that the pre-baking temperature in steps S3-1 and S3-3 is 85℃, the time is 120s, the post-baking temperature is 95℃, the time is 80s, the developing temperature is 25℃, and the time is 25s; the pre-baking temperature in step S3-2 is 95℃, the time is 120s, the post-baking temperature is 105℃, the time is 80s, the developing temperature is 25℃, and the time is 25s.

[0014] The beneficial effects of the present application are: the present application breaks through the limitation of traditional two-dimensional MXene artificial synapse relying on single ion migration through innovative multi-gate structure design. The multi-gate configuration can synchronously introduce multiple alkali metal ions (such as protons, potassium ions, sodium ions, etc.) to participate in the migration process, realize the dynamic switching of long / short-term synaptic plasticity of the device, and endow the device with more abundant example chemical regulation ability, which can be quickly switched in the face of different application scenarios (such as calculation or memory). This multi-ion synergistic effect not only can enhance the biological authenticity of synaptic simulation, but also can realize the precise regulation of dynamic range and memory window according to the embedding / extraction characteristics of different ions, providing a more flexible physical basis for complex information processing of artificial neural networks. BRIEF DESCRIPTION OF DRAWINGS

[0015] Figure 1 A two-dimensional MXene multi-gate artificial synapse structure constructed by the present application is shown.

[0016] Figure 2 A two-dimensional MXene artificial synapse long / short-term synaptic plasticity performance graph is shown: the electrolytes are protons and lithium ions, and the mass fractions are 30% and 30%, respectively.

[0017] Figure 3 A two-dimensional MXene artificial synapse long / short-term synaptic plasticity performance graph is shown: the electrolytes are protons and lithium ions, and the mass fractions are 10% and 30%, respectively.

[0018] Figure 4 A two-dimensional MXene artificial synapse long / short-term synaptic plasticity performance graph is shown: the electrolytes are protons and lithium ions, and the mass fractions are 30% and 20%, respectively.

[0019] Figure 5 A two-dimensional MXene artificial synapse long / short-term synaptic plasticity performance graph is shown: the electrolytes are protons and sodium ions.

[0020] Figure 6 A two-dimensional MXene artificial synapse long / short-term synaptic plasticity performance graph is shown: the electrolytes are protons and potassium ions. DETAILED DESCRIPTION

[0021] Those skilled in the art can refer to the content herein to appropriately improve the process parameters. In particular, it should be pointed out that all similar substitutions and changes are obvious to those skilled in the art, and they are considered to be included in the present application. The method and application of the present application have been described by the preferred embodiments, and the relevant personnel can modify or appropriately change and combine the methods and applications described herein without departing from the content, spirit and scope of the present application, to realize and apply the present application technology.

[0022] Example 1

[0023] The application provides a method for improving the performance of a two-dimensional MXene artificial synapse, and specifically comprises the following steps.

[0024] S1: precisely preparing a two-dimensional MXene by a liquid etching method.

[0025] S2: preparing different ion electrolytes.

[0026] S3: constructing a multi-gate two-dimensional MXene artificial synapse.

[0027] The step S1 specifically comprises the following steps.

[0028] S1-1: adding 5 ml of deionized water, 15 ml of hydrochloric acid and 1-3 g of lithium fluoride into a polytetrafluoroethylene beaker, stirring for half an hour, then adding 1 g of a MAX phase, and etching at 30-60 DEG C for 24-72 h.

[0029] S1-2: washing by centrifugation 4-6 times with deionized water until the pH is neutral, and then ultrasonicating for 20-60 min by using an ultrasonic machine with a power of 100-300 W.

[0030] S1-3: centrifuging at a speed of 2000 rpm-5000 rpm for 20-60 min, and taking the supernatant as a two-dimensional MXene solution.

[0031] The ion electrolyte in the step S2 comprises a proton electrolyte and an alkali metal ion electrolyte.

[0032] The alkali metal ion electrolyte is one of a lithium ion electrolyte, a sodium ion electrolyte and a potassium ion electrolyte.

[0033] The step S2 specifically comprises the following steps.

[0034] S2-1: preparing the proton electrolyte: dissolving 1 g of polyvinyl alcohol in 10 ml of deionized water, stirring the obtained solution at 80-95 DEG C for 10-24 h until it becomes transparent, and cooling to room temperature. Then, 0.1-1 g of sulfuric acid is slowly added.

[0035] S2-2: preparing the lithium ion electrolyte or the sodium ion electrolyte or the potassium ion electrolyte: weighing 10-150 mg of lithium perchlorate or sodium perchlorate or potassium perchlorate and mixing with 250 mg of polyethylene oxide (PEO) in a glass container, then injecting 10 ml of acetonitrile solvent into the container, and continuously stirring on a hot plate at 80-95 DEG C for 12-48 h until all solid particles are completely dissolved in the acetonitrile.

[0036] The step S3 specifically comprises the following steps.

[0037] S3-1: patterning of the second gate: photolithography is performed on the silicon wafer by using a photoetching machine and a mask plate, and the patterning of the second gate is realized by spin-coating a negative photoresist, pre-baking, post-baking and developing, and then 5-15 nm of Cr and 50-80 nm of Au are evaporated by thermal evaporation, and the photoresist is removed by immersing in an acetone solution at 100 DEG C for 30 min, and then the proton electrolyte is combined with the second gate by spin-coating or drop-coating.

[0038] S3-2: patterning of the channel material: photolithography is performed on the silicon wafer by using a photoetching machine and a mask plate, and the channel region is hollowed out by spin-coating a positive photoresist, pre-baking, post-baking and developing, and then a solution with a concentration of 5-15 mg / ml is spin-coated, and the photoresist is removed by immersing in an acetone solution at 95 DEG C for 30 min, and then the lithium ion electrolyte is dropped onto the channel material.

[0039] S3-3: patterning of the source electrode, the drain electrode and the first gate: photolithography is performed on the silicon wafer by using a photoetching machine and a mask plate, and the patterning of the source electrode and the drain electrode is realized by spin-coating a negative photoresist, pre-baking, post-baking and developing, and then 10-40 nm of an inert metal is evaporated as the drain electrode, and then the lithium ion electrolyte or the sodium ion electrolyte or the potassium ion electrolyte is combined with the channel material MXene by spin-coating or drop-coating, and finally the patterning of the first gate is realized by the direction of photolithography, and 10-40 nm of an inert metal is evaporated as the first gate.

[0040] In step S3-1 and step S3-3, the pre-baking temperature is 85 DEG C, the time is 120 s, the post-baking temperature is 95 DEG C, the time is 80 s, the developing temperature is 25 DEG C, and the time is 25 s; in step S3-2, the pre-baking temperature is 95 DEG C, the time is 120 s, the post-baking temperature is 105 DEG C, the time is 80 s, the developing temperature is 25 DEG C, and the time is 25 s.

[0041] Example 2

[0042] This embodiment uses the method disclosed in Example 1 to prepare a two-dimensional MXene artificial synapse, which specifically includes the following steps:

[0043] S1: precise preparation of two-dimensional MXene by liquid etching method;

[0044] In a Teflon beaker, 5 ml of deionized water, 15 ml of hydrochloric acid and 1 g of lithium fluoride are added, stirred for half an hour, then 1 g of MAX phase is added, and etched at 40 DEG C for 48 h. Centrifugal washing is performed 4-6 times with deionized water until the PH is neutral, and then ultrasonic treatment is performed for 40 min using an ultrasonic machine with a power of 200 W. Centrifugal separation is performed at a speed of 3500 rpm for 30 min, and the supernatant is taken out, which is a two-dimensional MXene solution.

[0045] S2: preparation of different ion electrolytes;

[0046] Proton electrolyte preparation: 1 g of polyvinyl alcohol was dissolved in 10 ml of deionized water, and the resulting solution was stirred at 85 °C for 24 h until it became transparent, and cooled to room temperature. Then 0.3 g of sulfuric acid was slowly added. Lithium ion electrolyte or sodium ion electrolyte or potassium ion electrolyte preparation: 75 mg of lithium perchlorate or sodium perchlorate or potassium perchlorate was weighed and mixed with 250 mg of polyethylene oxide (PEO) in a glass container, then 10 ml of acetonitrile solvent was injected into the container, and continuous stirring was carried out on a hot plate at 95 °C for 24 h until all solid particles were completely dissolved in acetonitrile.

[0047] S3: Multi-gate two-dimensional MXene artificial synapse construction;

[0048] Patternization of gate two: photolithography was performed on the silicon wafer by a photolithography machine and a mask plate, and the patternization of gate two was realized by spin-coating negative photoresist, pre-baking at 85 °C for 120 s, post-baking at 95 °C for 80 s, and developing at 25 °C for 25 s, and then 5 nm of Cr and 50 nm of Au were evaporated by thermal evaporation, and the photoresist was removed by soaking in 100 °C acetone solution for 30 min, and then the proton electrolyte was combined with gate two by spin-coating or drop-coating on gate two. Patternization of channel material: photolithography was performed on the silicon wafer by a photolithography machine and a mask plate, and the channel region was hollowed out to realize patternization by spin-coating positive photoresist, pre-baking at 95 °C for 120 s, post-baking at 105 °C for 80 s, and developing at 25 °C for 25 s, and then a solution with a concentration of 8 mg / ml was selected for spin-coating, and the photoresist was removed by soaking in 95 °C acetone solution for 30 min, and then lithium ion electrolyte was dropped onto the channel material. Patternization of source electrode, drain electrode and gate one: photolithography was performed on the silicon wafer by a photolithography machine and a mask plate, and the patternization of source electrode and drain electrode was realized by spin-coating negative photoresist, pre-baking at 85 °C for 120 s, post-baking at 95 °C for 80 s, and developing at 25 °C for 25 s, and then 30 nm of inert metal was evaporated as the drain electrode, and then lithium ion electrolyte or sodium ion electrolyte or potassium ion electrolyte was combined with the channel material MXene by spin-coating or drop-coating, and finally the patternization of gate one was realized by the direction of photolithography, and 30 nm of inert metal was evaporated as gate one.

[0049] Example 3

[0050] The difference between this example and example 2 is only that 0.1 g of sulfuric acid is weighed in step S2, and the other step conditions are the same as those of example 2.

[0051] The specific performance is shown in Table 1. Figure 3

[0052] Example 4

[0053] ​The only difference between this embodiment and embodiment 2 is that 50 g of lithium perchlorate is weighed in step S2, and the prepared alkali metal ion electrolyte is a lithium ion electrolyte. The other step conditions are the same as those in embodiment 2.

[0054] Specific performance such as Figure 4 shown.

[0055] Example 5

[0056] The only difference between this embodiment and embodiment 2 is that 75 g of sodium perchlorate is weighed in step S2, and the prepared alkali metal ion electrolyte is a sodium ion electrolyte. The other step conditions are the same as those in embodiment 2.

[0057] Specific performance such as Figure 5 shown.

[0058] Example 6

[0059] The only difference between this embodiment and embodiment 2 is that 75 g of potassium perchlorate is weighed in step S2, and the prepared alkali metal ion electrolyte is a potassium ion electrolyte. The other step conditions are the same as those in embodiment 2.

[0060] Specific performance such as Figure 6 shown.

[0061] The long-term / short-term synaptic plasticity performance of the constructed two-dimensional MXene artificial synapse was analyzed using the B1500 semiconductor parameter analyzer.

[0062] Figure 1 A schematic structural diagram of a two-dimensional MXene multi-ion migration artificial synaptic device is shown.

[0063] Figure 2 The long-term / short-term synaptic plasticity performance of two-dimensional MXene artificial synapses is demonstrated when the electrolytes are lithium ions and protons, respectively. When the gate 1 is positively / negatively applied, lithium ions are embedded in / out of the channel material MXene. When the gate 2 is positively / negatively applied, protons are embedded in / out of the channel material MXene. The source and drain electrodes detect the changes in the current of the channel material MXene, such as Figure 2 As shown in (a), when pulses with a pulse amplitude of 3V and a pulse duration of 500ms are continuously applied, protons are embedded in and out of MXene, causing the MXene current to increase from 162nA to 589nA, with a dynamic range of 363.6%. About 4s after the pulse is stopped, the current returns to the initial current position. The device has volatility and a large dynamic range.

[0064] For lithium ions, such as Figure 2(b) as shown, a pulse with a pulse amplitude of 3 V and a pulse duration of 500 ms is also applied, at which time the MXene current increases from 255 nA to 445 nA, with a dynamic range of 174.5%, and after 300 s of stopping the application of the pulse, the current is 297.3 nA, which is still higher than the initial current of 255 nA, and a change ratio of 22.3% is still retained, so the device has non-volatility at this time, and thus the device can control different ions to enter the channel material through different gate voltages, thereby realizing controllable conversion of volatility / non-volatility.

[0065] Figure 3 The long / short-term synaptic plasticity performance of the two-dimensional MXene artificial synapse when the mass fraction of the proton electrolyte is 10% is shown, at which time when a pulse is applied to the proton electrolyte, the channel material current increases from 203 nA to 656 nA, with a dynamic range of 323%, and after about 4 s of stopping the application of the pulse, the current returns to the initial current position. Compared with other alkali metal ions, protons have the smallest atomic radius, so they have the largest dynamic range and volatility. The radii of other alkali metal ions increase, so they have non-volatility, but the dynamic range is reduced.

[0066] Figure 4 The long / short-term synaptic plasticity performance of the two-dimensional MXene artificial synapse when the mass fraction of the lithium ion electrolyte is 20% is shown, at which time when a pulse is applied to the lithium ion electrolyte, the channel material current increases from 262 nA to 497 nA, with a dynamic range of 189.7%, and after 300 s of stopping the application of the pulse, the current is 295 nA, which is still higher than the initial current of 262 nA, and a change ratio of 14% is retained, so the device has non-volatility.

[0067] Figure 5 The long / short-term synaptic plasticity performance of the two-dimensional MXene artificial synapse when the lithium ion electrolyte is replaced with a sodium ion electrolyte is shown, when a pulse is applied, sodium ions are inserted into and extracted from the MXene, causing the current to increase from 409 nA to 461 nA, with a dynamic range of 112.7%, and after 300 s of stopping the application of the pulse, the current is 425.6 nA, which is still higher than the initial current of 409 nA, and a change ratio of 31.9% is retained, so the device has non-volatility at this time.

[0068] Figure 6 The long / short-term synaptic plasticity performance of the two-dimensional MXene artificial synapse when the lithium ion electrolyte is replaced with a potassium ion electrolyte is shown, when a pulse is applied, potassium ions are inserted into and extracted from the MXene, causing the current to increase from 438 nA to 471 nA, with a dynamic range of 107.5%, and after 300 s of stopping the application of the pulse, the current is 445.3 nA, which is still higher than the initial current of 438 nA, and a change ratio of 22.1% is retained, so the device has non-volatility at this time.

[0069] To sum up, the application breaks through the limitation of traditional two-dimensional MXene artificial synapse relying on single ion migration by innovative multi-gate structure design. The multi-gate configuration can synchronously introduce multiple alkali metal ions (such as protons, potassium ions, sodium ions, etc.) to participate in the migration process, realize dynamic switching of long / short-term synaptic plasticity of the device, and endow the device with richer example chemical regulation ability, which can be quickly switched in the face of different application scenarios (such as calculation or memory). The synergistic effect of multiple ions not only can enhance the biological authenticity of synaptic simulation, but also can realize precise regulation of dynamic range and memory window according to the embedding / extraction characteristics of different ions, providing a more flexible physical basis for complex information processing of artificial neural networks.

[0070] The above only describes the preferred embodiments of the present application, and it should be pointed out that for ordinary skilled in the art, without departing from the principles of the present application, a number of improvements and refinements can be made, and these improvements and refinements should also be within the scope of protection of the present application.

Claims

1. A method for improving the performance of two-dimensional MXene artificial synapses, characterized in that: The specific steps include: S1: Precision preparation of two-dimensional MXene by liquid phase etching; S2: Preparation of electrolytes with different ions; S3: Construction of multi-gate two-dimensional MXene artificial synapses.

2. The method for improving the performance of two-dimensional MXene artificial synapses according to claim 1, characterized in that: Step S1 specifically includes the following steps: S1-1: Add 5 ml of deionized water, 15 ml of hydrochloric acid, and 1-3 g of lithium fluoride to a polytetrafluoroethylene beaker. Stir for half an hour, then add 1 g of MAX phase and etch at 30-60°C for 24-72 hours. S1-2: Use deionized water for centrifugal washing 4-6 times until the pH is neutral, and then use an ultrasonic machine with a power of 100-300W for 20-60 minutes; S1-3: Centrifuge at 2000 rpm-5000 rpm for 20-60 min and remove the supernatant, which is the two-dimensional MXene solution.

3. The method for improving the performance of two-dimensional MXene artificial synapses according to claim 1, characterized in that: The ionic electrolyte in step S2 includes a proton electrolyte and an alkali metal ion electrolyte.

4. The method for improving the performance of two-dimensional MXene artificial synapses according to claim 3, characterized in that: The alkali metal ion electrolyte is one of a lithium ion electrolyte, a sodium ion electrolyte and a potassium ion electrolyte.

5. The method for improving the performance of two-dimensional MXene artificial synapses according to claim 4, characterized in that: Step S2 specifically includes the following steps: S2-1: Proton electrolyte preparation: Dissolve 1 g of polyvinyl alcohol in 10 ml of deionized water. Stir the resulting solution at 80-95°C for 10-24 hours until it becomes transparent, and then cool to room temperature. Then slowly add 0.1-1 g of sulfuric acid. S2-2: Preparation of lithium ion electrolyte, sodium ion electrolyte or potassium ion electrolyte: Weigh 10-150 mg of lithium perchlorate, sodium perchlorate or potassium perchlorate and mix with 250 mg of polyethylene oxide (PEO) in a glass container. Then, inject 10 ml of acetonitrile solvent into the container and continue stirring on a hot plate at 80-95°C for 12-48 hours until all solid particles are completely dissolved in acetonitrile.

6. The method for improving the performance of two-dimensional MXene artificial synapses according to claim 5, characterized in that: Step S3 specifically includes the following steps: S3-1: Patterning of Gate 2: Photolithography is performed on the silicon wafer using a photolithography machine and a mask. The patterning of Gate 2 is achieved by spin coating negative photoresist, pre-baking, post-baking, and developing. 5-15 nm Cr and 50-80 nm Au are deposited by thermal evaporation. The film is then stripped by soaking in a 100°C acetone solution for 30 minutes. Subsequently, a proton electrolyte is bonded to Gate 2 by spin coating or drop coating. S3-2: Channel material patterning: The silicon wafer is photolithographically patterned using a photolithography machine and a mask. The channel area is hollowed out by spin-coating positive photoresist, pre-baking, post-baking, and developing to achieve patterning. A solution with a concentration of 5-15 mg / ml is then spin-coated. The wafer is then soaked in a 95°C acetone solution for 30 minutes to remove the resist. Lithium-ion electrolyte is then dripped onto the channel material. S3-3: Patterning of the source, drain and gate 1. The silicon wafer is photolithographically patterned using a photolithography machine and a mask. The source and drain are patterned by spin-coating negative photoresist, pre-baking, post-baking and developing. 10-40nm of inert metal is evaporated as the drain electrode. Subsequently, the lithium ion electrolyte, sodium ion electrolyte or potassium ion electrolyte is bonded to the channel material MXene by spin coating or drop coating. Finally, the gate 1 is patterned by photolithography direction, and 10-40nm of inert metal is evaporated as the gate 1.

7. The method for improving the performance of two-dimensional MXene artificial protrusions according to claim 6, characterized in that: In step S3-1 and step S3-3, the pre-baking temperature is 85°C, the time is 120s, the post-baking temperature is 95°C, the time is 80s, and the developing temperature is 25°C, the time is 25s; in step S3-2, the pre-baking temperature is 95°C, the time is 120s, the post-baking temperature is 105°C, the time is 80s, and the developing temperature is 25°C, the time is 25s.

Images