Implementation method of a gate-controllable switch device based on metal tunnel junction

Abstract

The application discloses a realization method of a gate-controllable switch device based on a metal tunnel junction, and belongs to the field of nanometer or atomic scale devices. The method introduces a switch control electrode outside the tunnel junction structure, the switch control electrode is separated from the tunnel junction via a medium or air, and electrical insulation is realized. When a voltage is applied on the switch control electrode, an electric field is also formed between the switch control electrode and the tunnel junction. Due to the introduction of the electric field of the switch control electrode, the electric field distribution of the tunnel junction surface is changed, and an electric field component perpendicular to the channel direction is generated. The electric field component perpendicular to the channel direction also has a Coulomb force effect on the atoms of the tunnel junction surface, and can make the atoms migrate along the direction perpendicular to the channel, so that the atoms originally staying at the tip of the tunnel electrode pair will move to both sides and leave the tip position, thereby realizing the purpose of controlling the tunnel junction current size through the switch control electrode, and realizing the gate-controllable switch.

Description

Technical Field

[0001] This invention relates to the field of nano / atomic device design, specifically a method for realizing gate-controllable switching devices using metal tunnel junctions. Background Technology

[0002] With the continuous development of integrated circuit technology, semiconductor manufacturing processes have advanced to the 5nm and 3nm nodes, and the feature sizes of micro- and nanoelectronic devices (such as transistor gate spacing and metal spacing) have shrunk to the range of 50nm to 30nm. However, according to information released by TSMC, even if semiconductor manufacturing processes advance to the 2nm node, transistor gate spacing and metal spacing will still be 30nm and 20nm, respectively. Therefore, it is extremely difficult to scale silicon-based devices down to below 10nm proportionally. Further shrinking device feature sizes to sub-10nm or even the atomic scale is extremely challenging. For metal-oxide-semiconductor field-effect transistors (MOSFETs), further improving device performance requires simultaneous improvements in materials, device structure, and processing technology. This not only presents significant technical challenges but also increases economic costs. Novel nano / atomic scale devices, such as nano / atomic scale tunneling junction devices, differ from traditional microelectronic devices in their working principles and structures, enabling smaller physical sizes, higher integration density, and higher device performance. Therefore, tunneling junction devices have received considerable attention in recent years.

[0003] A tunneling junction typically consists of a pair of metal electrodes, with the tunnel gap usually filled with vacuum or air, forming a tunneling barrier. When the tunnel gap is very small, if a bias voltage is applied, changing the energy level positions of the two tunneling electrode pairs, electrons on one side of the tunneling electrode can travel from one electrode to the other via quantum tunneling, thus generating a tunneling current. Based on metal tunneling junctions, switching devices with extremely high on / off ratios can be realized, and they can be applied in many fields, such as memory, quantum devices, and optoelectronic devices.

[0004] However, currently, tunneling junction devices can only be used as two-terminal devices, such as switches and non-volatile memories. Tunneling junction devices cannot control the tunneling current through the gate electrode, and therefore cannot be implemented as three-terminal devices like field-effect transistors. Consequently, they are difficult to use in more complex digital logic circuits. The main problem is that, to date, tunneling junctions cannot achieve gate-controllable switching functionality, and gate-controllable switching devices are core components of digital logic functions. Therefore, methods for implementing gate-controllable switching using tunneling junctions are of great value. Summary of the Invention

[0005] The purpose of this invention is to provide a method for implementing gate-controllable switching using a tunnel junction.

[0006] The principle of this invention:

[0007] Metal nanoribbons can be fabricated on substrates using methods such as photolithography or electron beam lithography, metal deposition lift-off processes, or FIB deposition. These nanoribbons can be further processed into metal tunneling junctions with nanoscale / atomic-scale tunneling gaps using methods such as electromigration, gated mechanical pyrolysis, and direct etching. In nanoscale or atomic-scale metal tunneling junctions, the electrostatic field within the tunneling gap significantly affects atomic migration on the junction surface. If an electric field is applied near the tunneling gap, perpendicular to the source-drain electrode connection line, through the gate electrode, the electric field distribution on the junction surface will change, altering the direction of atomic migration and thus the tunneling gap width. This, in turn, changes the tunneling current, ultimately enabling the switching of the tunneling junction via a switching control electrode.

[0008] The physical mechanism is as follows: When a certain voltage is applied to the tunnel junction, an electrostatic field exists in the tunnel gap, generating a tunneling current. The tunneling electrons collide with atoms, providing energy and momentum, forming a current force. The electrostatic field exerts a Coulomb force on the atomic cores. Simultaneously, the tunneling current generates Joule heating, intensifying the thermal motion of atoms, giving them higher kinetic energy, and making it more likely to overcome the migration barrier. Under the combined effect of these factors, atoms on the surface of the tunnel junction can migrate towards the tip of the tunnel junction electrode and remain at the tip. This allows the tunnel electrode to extend forward, thereby reducing the tunnel gap width and increasing the tunneling current. By fabricating gate electrodes at the bottom, side, or top of the tunnel junction, when a gate voltage is applied, the atomic migration direction deviates from the channel direction, thus increasing the tunneling distance and decreasing the tunneling current.

[0009] Based on the above principle, a switch control electrode can be introduced outside the tunnel junction electrode pair structure. The switch control electrode is electrically insulated from the tunnel junction via a dielectric or air. Applying a voltage to the switch control electrode creates an electric field between it and the tunnel junction. The introduction of this electric field alters the electric field distribution on the tunnel junction surface, generating an electric field component perpendicular to the channel direction. This channel-perpendicular electric field component also exerts Coulomb forces on the atoms on the tunnel junction surface, causing them to migrate along this direction. Atoms that were originally stationary at the tips of the tunnel junction electrode pair will move to the sides, leaving their tip positions. This migration increases the tunnel gap width, reducing the tunnel junction current, thus achieving the goal of controlling the tunnel junction current through the switch control electrode, realizing a gate-controlled switch.

[0010] The technical solution provided by this invention is as follows:

[0011] A method for implementing a gate-controllable switching device based on a metal tunnel junction, comprising the following steps:

[0012] 1) Fabrication of metal nanoribbons: Metal nanoribbon structures can be fabricated by photolithography or electron beam exposure, metal deposition and lift-off techniques, or focused ion beam (FIB) local deposition. The metal nanoribbons have a neck structure with a width of less than 100 nm in the middle.

[0013] 2) Fabrication of metal nanowires into tunnel junctions: Metal nanoribbons can be further processed into metal tunnel junctions using methods such as electromigration, gated mechanical pyrolysis, helium ion beam etching, and hard mask-plasma etching. The tunneling gap distance is at the nanometer or atomic scale, ranging from 50 nm to 0.1 nm.

[0014] 3) A switch control electrode is prepared at the bottom, side or top of the tunnel junction, wherein the switch control electrode and the tunnel junction are in a vacuum or insulating medium, and the tunnel junction and the switch control electrode constitute a gate-controlled switch.

[0015] 4) Connect the gate-controllable switch to the controller;

[0016] 5) Adjust the controller, apply a voltage waveform to the tunnel junction, perform IV scanning, and ensure that its tunneling current reaches the value required for the turn-on state. This turn-on state is the initial state of the gate-controllable switch.

[0017] 6) Set the switch control voltage, apply this voltage to the switch control electrode through the controller, and keep it stable;

[0018] 7) When a switching control voltage is applied, a voltage waveform is applied to the tunnel junction, and an IV scan is performed to read the state of the switching control electrode. At this time, depending on the magnitude of the switching control voltage applied in step (5), different sizes of tunneling current can be read: if a higher switching control voltage is applied, the conductance of the tunnel junction will decrease, and thus the tunneling current will be smaller, indicating that the tunnel junction has been switched to the off state; if a lower or 0V switching control voltage is applied, the conductance of the tunnel junction will not change, and the tunneling current will remain at the magnitude of the on state in step (5), so the tunnel junction will remain in the on state.

[0019] This invention enables the tunnel junction to reach different termination states from the same initial state by controlling the voltage of the switch, thereby realizing the function of a gate-controlled switch.

[0020] In this method, the material of the tunnel junction can be a metal (gold, palladium, platinum, aluminum, indium, silver, tantalum, etc.). The switch control electrode structure can use a conductive substrate covered with an insulating dielectric (including but not limited to silicon dioxide-low resistivity silicon substrate, boron nitride-graphene substrate, sapphire-metal substrate, etc.) as the control electrode, or a locally deposited metal on the substrate surface adjacent to but not connected to the tunnel junction, serving as the control electrode at the bottom, side, or top of the tunnel junction. The control electrode and the tunnel junction can be filled or not filled with dielectric. If a dielectric is required, the dielectric can be an insulating oxide layer (silicon dioxide, germanium oxide, etc.) prepared by various methods (including but not limited to wet / dry thermal oxidation, epitaxial growth, evaporation, sputtering, atomic layer deposition, etc.), a high dielectric constant dielectric (i.e., High-K dielectric, including but not limited to hafnium oxide, aluminum oxide, etc.), air dielectric, or vacuum, etc. The connection circuit can be a commercial device, such as a probe station system; or a self-designed and prepared connection circuit, such as a PCB board, or an on-chip circuit prepared by tape-out. The controller can be a commercial circuit instrument or a self-designed and fabricated control circuit. Commercial circuit instruments include, but are not limited to, semiconductor analyzers or other source meters. Self-designed and fabricated control circuits include, but are not limited to, drive control circuits built from voltage sources, current sources, programmable logic circuits (FPGAs, PLCs, etc.), microcontroller systems (such as microcontrollers, Raspberry Pis, etc.), and self-designed and fabricated control driver chips. The controller can be built as discrete components or discrete instruments (such as source meters and probe station systems), or it can be integrated with tunneling junction devices (such as FPGAs or on-chip integrated circuits).

[0021] A gate-controlled switch consists of a tunneling junction and a switch control electrode. The tunneling junctions are separated by a tunneling gap ranging from a few angstroms to tens of nanometers in width, forming a tunneling barrier. When a voltage is applied to the tunneling junction, electrons can tunnel through the barrier from one electrode to the other, forming a tunneling current. Atoms can migrate on the surface of the tunneling junction, altering the width of the tunneling gap. The switch control electrode applies an electric field through a dielectric, changing the electric field distribution on the surface of the tunneling junction and affecting the direction of atomic migration. Connecting circuitry (including probes or wires) connects the tunneling junction device to a controller, ensuring a good electrical connection. The controller, which can be a source meter or a control chip, can apply specific voltage or current waveforms to the control electrode of the tunneling junction as needed to control the gate-controlled switch function and detect the switching state of the tunneling junction.

[0022] The gate-controllable switch based on a metal tunnel junction in this invention differs fundamentally from traditional field-effect transistors (FETs) in terms of materials, device structure, working principle, and fabrication process. In traditional FETs, the channel between the source and drain is made of semiconductor material, and the source and drain electrodes are heavily doped semiconductors. The gate electrode alters the Fermi level within the channel to change the number and distribution of charge carriers, causing an inversion layer to appear or disappear near the gate dielectric, thereby changing the channel's conductance and achieving the switching function. In traditional FETs, the gate only controls the distribution and number of electrons without changing the physical structure of the channel. When the switching control voltage is no longer applied, the electron distribution returns to normal, and the channel conductance automatically returns to the state before the switching control voltage, exhibiting volatility. Because the gate of a traditional FET requires Fermi level modulation, precise control of the impurity concentration in the channel is necessary. This necessitates the introduction of methods such as ion implantation and diffusion during the fabrication process to dope different locations within the device channel. The complexity of the doping process, along with the dramatic increase in leakage current and other short-channel effects at small sizes, are important reasons limiting the reduction in the physical size of traditional field-effect transistors.

[0023] The gate-controllable switching device based on a metal tunneling junction in this invention is made of metal, with a tunneling gap as the channel. This gap can be a vacuum, air, or a solid insulating medium, and it lacks semiconductor channel materials and the source / drain electrode structure found in traditional transistors. The switch control electrode does not modulate the Fermi level in the metal electrode, but rather influences the migration direction of metal atoms by changing the electric field distribution in the tunneling gap. This alters the length of the tunneling electrode, changes the tunneling gap, and consequently affects the magnitude of the tunneling current, thus achieving gate-controllable switching capability. During this modulation process, the physical structure of the tunneling junction device changes, ranging from nanometer to atomic scale. Without external forces, the metal atoms can remain at their migrated positions, preserving the changed physical structure of the tunneling junction. Therefore, even if the switch control voltage is removed, the tunneling gap distance remains at the value applied when the switch control voltage was applied, and the switching state of the tunneling junction does not change automatically, exhibiting non-volatility. Furthermore, since no doping process is required, and combined with the nanometer / atomic scale tunneling gap distance, the characteristic physical size of the gate-controllable switching device based on the metal tunneling junction in this invention can easily reach below 10 nm, which is smaller than that of traditional field-effect transistors.

[0024] The advantages of this invention are as follows:

[0025] The tunnel junction gateable switch of this invention features a simple and reliable structure, small device size, easy high integration, fast control speed, and low power consumption. The fabrication of the switch control electrode can be completed based on the tunnel junction structure, making design and processing relatively quick and convenient. The distance between the switch control electrode and the tunnel junction can be relatively large, and if a dielectric is filled, the dielectric can be relatively thick, thus enabling it to withstand a large switching control voltage and exhibiting good reliability. Because it maintains the structure of the tunnel junction electrode itself, its key physical dimensions are the same as the tunnel junction, which can be less than 10 nm, even approaching the diameter of a single atom, far exceeding the physical dimensions of traditional field-effect transistors. The simple and reliable structure and extremely small physical size allow the tunnel junction gateable switch of this invention to be integrated at extremely high density. In principle, the migration time of a single atom in the tunnel junction can be as short as a few picoseconds, which gives the tunnel junction gateable switch device extremely high operating speed. The operating current of the tunnel junction can be as low as a few nanoamperes to a few picoamperes. When the switching control electrode controls the switching state of the tunnel junction, it is insulated from the tunnel junction and is regulated only by the electric field. There is no direct current conduction between the electrode and the tunnel junction, so no additional power consumption is introduced. Therefore, it is easy to realize low-power devices. Attached Figure Description

[0026] Figure 1 Schematic diagrams of a gateable switching device using a metal tunnel junction with locally deposited metal on the substrate surface as the switch control electrode, showing a side view (Fig. (a)) and a top view (Fig. (b)). The control electrode is located below the tunnel junction, with a dielectric filling between them. Wherein: 1—metal nanoribbon; 2—dielectric layer; 3—switch control electrode; 4—substrate; 5—tunnel junction.

[0027] Figure 2 A schematic diagram of a gate-controllable switching device based on a metal tunnel junction with an insulating substrate having a conductor on the back side as the control electrode, wherein: 5—tunnel junction; 2—dielectric layer; 4—insulating substrate with a conductor on the back side.

[0028] Figure 3 A schematic diagram of a gate-controllable switching device of a dielectric-free metal tunnel junction, wherein the switch control electrode is located near the side of the tunnel junction, and: 1—metal nanoribbon; 5—tunnel junction; 3—switch control electrode; 4—substrate.

[0029] Figure 4 A schematic diagram of a gate-controllable switching device with a metal tunnel junction using an insulating substrate with a conductor on the back side as the control electrode, and the electrical connections and controller for realizing the gate-controllable switching of the tunnel junction. Wherein: 5—tunnel junction; 2—dielectric layer; 4—insulating substrate with a conductor on the back side; 6—connection circuit; 7—controller.

[0030] Figure 5Scanning electron microscope (SEM) images of the gate-controllable switching device based on a metal tunnel junction in Example 1. Figure (a) shows a metal nanoribbon; Figure (b) shows a metal tunnel junction: 1—metal nanoribbon; 8—neck structure of the gold nanoribbon; 9—tunneling gap of the metal tunnel junction; 10—silicon dioxide substrate;

[0031] Figure 6 The results of Example 1 are shown in the figure. The hollow square curve represents the median current of the device in the initial state, and the hollow circle curve represents the median current of the device in the final state after applying a 90V switching control voltage. Detailed Implementation

[0032] The implementation process of the present invention will now be described in detail with reference to the accompanying drawings.

[0033] (1) Fabrication of metal nanoribbons with switch control electrodes

[0034] Metal nanoribbon devices with switching control electrodes are fabricated on a substrate. The substrate can be silicon dioxide, glass, mica, silicon nitride thin film, etc., and the tunnel junction material of the sample is a metal, including but not limited to gold, silver, palladium, platinum, aluminum, tantalum, indium, etc.

[0035] Metal nanoribbons with narrow neck structures and switch-controlled electrode structures can be fabricated on substrates using various methods (see reference). Figure 1 , Figure 2 , Figure 3 Methods available for fabricating switch control electrodes include: (1) mask patterning-metal deposition-lifting process; (2) metal layer etching patterning process; (3) local deposition or self-assembly; and (4) direct use of a substrate with an insulating layer. If a deep ultraviolet lithography / electron beam lithography / electron beam lithography / electron beam lithography-metal deposition-lifting process is used, the mask patterning method includes, but is not limited to, deep ultraviolet lithography / electron beam lithography, electron beam lithography / electron beam lithography, imprinting or pattern transfer processes, etc.; the metal layer fabrication method includes, but is not limited to, electron beam evaporation deposition, thermal evaporation deposition, magnetron sputtering, electroplating, atomic layer deposition, epitaxial growth, etc. If the metal layer is directly patterned, methods such as focused ion beam (FIB) etching and plasma etching can be used to directly etch metal nanoribbon patterns on the metal layer. If local deposition or self-assembly methods are used, methods such as FIB deposition and self-assembly of nano-metal particles in solution can be employed to perform local deposition or growth on the substrate, directly obtaining metal nanoribbons and switch control electrode structures; the surface of conductive substrates covered with insulating layers can also be directly used as switch control electrodes, such as low-resistivity silicon substrates with surface thermally oxidized silicon dioxide layers.

[0036] The switch control electrode can be structured as a metal locally deposited on the substrate surface (with dielectric isolation between it and the tunnel junction). Figure 1 Insulating substrate with conductor on the back () Figure 2 Metal tip structure with no dielectric isolation and adjacent to tunnel junction ( Figure 3 For devices like these, the size of the switch control electrode must cover the tunnel gap. The distance between the switch control electrode and the tunnel gap must be within a suitable range, sufficient to significantly influence the electric field distribution in the tunnel gap (i.e., the electric field strength of the switch control electrode should not differ from the electric field strength along the channel direction in the tunnel junction by more than an order of magnitude), without generating significant leakage current between the switch control electrode and the tunnel junction. Typically, the distance between the switch control electrode and the tunnel gap is in the range of 5–1000 nm. For specific devices, the distance between the switch control electrode and the tunnel junction will vary depending on its structure, dielectric material, and the applied voltage range. Dielectric materials can be selected from dry / wet thermal oxidation grown oxide layers, such as thermally oxidized silicon dioxide layers on silicon substrates; or high-dielectric-constant dielectric layers grown by atomic layer deposition (ALD), such as hafnium oxide (HfO2); vacuum or air dielectrics can also be used. The thickness of the dielectric layer is the distance between the switch control electrode and the tunnel gap. The dielectric layer should be able to withstand the maximum switching control voltage amplitude required for regulation without being broken down or generating significant leakage current. Depending on the specific device structure, the maximum switching control voltage amplitude is +5 to +100V / -5 to -100V, and the leakage current should generally not exceed 10pA.

[0037] The fabrication processes for metal nanoribbons are similar to those for metal nanoribbon fabrication. However, for specific metal tunnel junction gate-controlled switching devices, the fabrication process, sequence of steps, and combination of components will differ. For structures with locally deposited metal on the substrate surface (requiring dielectric isolation between the metal and the tunnel junction)... Figure 1 After the switch control electrode and dielectric layer are fabricated, metal nanoribbons can be prepared on the surface of the dielectric layer using a photolithography / electron beam exposure-metal deposition-lift process; or metal nanoribbons can be obtained using methods such as FIB metal deposition. For devices with an insulating substrate containing a conductor on the back side as the switch control electrode structure... Figure 2 The above-mentioned process can be performed directly on the substrate surface to create metal nanoribbons with switching control electrodes; for devices with metal tips near tunnel junctions without dielectric isolation and serving as switching control electrodes ( Figure 3 Metal nanoribbons can be fabricated together with switch control electrodes, using the same patterning, etching, or lift-off processes to simultaneously obtain both the metal nanoribbons and the switch control electrodes. The neck width of the metal nanoribbons can range from 1 nm to 1000 nm, and the thickness can range from 5 nm to 1000 nm.

[0038] (2) Fabrication of tunnel junction devices with switch control electrodes

[0039] Various methods were employed to fabricate tunneling junctions with switch-controlled electrodes on metal nanoribbons. The tunneling electrodes were separated by nanometer- or atomic-scale tunneling gaps, with the switch-controlled electrodes positioned adjacent to these gaps and insulated from the tunneling junction by a dielectric material. Figure 1 , Figure 2 , Figure 3 ).

[0040] Tunneling junctions can be fabricated from metal nanoribbons using electrical methods or direct etching. Electrical methods include electromigration fusing and stretching under a strong electric field. In the electromigration method, a 0-5V voltage is applied to the tunneling junction, generating a 0.5-5mA current. The voltage can be applied using a triangular wave, square wave, or IV scan. Under the influence of Joule heating, current force, and electric field, metal atoms migrate violently, and the neck of the metal nanoribbon melts first, forming a nanometer or atomic-scale gap, thus obtaining the tunneling junction. In the stretching method under a strong electric field, a relatively large gap should be pre-prepared in the metal nanoribbon. Applying a 10-100V voltage to both ends of such a metal nanoribbon causes metal atoms to migrate towards the electrode tip under the influence of the strong electric field, reducing the gap distance and thus obtaining the tunneling junction.

[0041] When using direct etching methods, local etching methods such as focused ion beam bombardment and helium ion beam etching, or methods such as hard mask combined with plasma etching, can be used to etch nano- or atomic-scale tunneling gaps in the neck of metal nanoribbons, thereby directly obtaining tunneling junctions.

[0042] Regardless of the process chosen to fabricate the tunnel junction, the switch control electrode structure should be kept intact and the dielectric should not be significantly damaged during the fabrication process to avoid affecting the control performance of the switch control electrode.

[0043] (3) The tunnel junction and the switch control electrode constitute a gateable switch. Electrical connection of the gateable switch

[0044] The tunneling junction is placed on a stable plane so that it will not move and affect its electrical connection during measurement. The tunneling junction is then connected to the controller via a connection circuit, as follows: Figure 4 As shown.

[0045] (4) The gate switch is set to the initial state.

[0046] Using a controller, perform an IV scan on the tunnel junction, starting from 0V, with a maximum voltage range of 5–50V, a scan speed range of 1mV / s–10V / s, and a current limiting range of 0.5–100nA. The IV scan should bring the tunnel junction device to its current limit, meaning the device reaches and can maintain its initial controlled state. Scans can be performed once or multiple times until the device reaches its initial state and stabilizes. The initial state of the device is a high-current state of the tunnel junction, i.e., the switch is on.

[0047] (5) Apply a switch control voltage to regulate the switching state of the gateable switch.

[0048] Using a controller, a switching control voltage is applied to the gateable switch according to the control requirements. The switching control voltage can vary depending on the desired effect. If it is desired to keep the device on (maintain a high current state), the applied switching control voltage can be -5 to 5V; if it is desired to turn the device off (transition to a low current state), the applied switching control voltage should be -100 to -6V, or 6 to 100V.

[0049] (6) Read the device termination status

[0050] Under the applied switching control voltage, the tunnel junction is scanned using the controller with the same scan parameters as in step 4 to read the termination state of the tunnel junction device. The termination current of the tunnel junction varies depending on the magnitude of the switching control voltage applied in step 4. If a lower switching control voltage is applied, keeping the device in the on state, the current remains at the current-limiting level; if a higher switching control voltage is applied, the device will switch to the off state, and the current should be 0.1–200 pA. To more accurately measure the effect of the gate-controlled switch, multiple measurements can be taken, and the median current value can be calculated.

[0051] The invention is illustrated by using the fabrication of a gate-controllable switch on a silicon dioxide / silicon substrate as an example.

[0052] (1) Fabrication of gold nanoribbons with switch control electrodes. The device was fabricated on a silicon dioxide / silicon substrate, with each source and drain connected to an electrode plate. The silicon dioxide insulating layer was 300 nm thick. The silicon substrate was a low-resistivity silicon substrate with a sheet resistance of 10 Ω. The silicon substrate was used as the switch control electrode. After electron beam exposure, the photoresist mask was patterned, and then metal was deposited by electron beam evaporation. After acetone immersion and lift-off, gold nanoribbons were obtained. The neck width of the nanoribbons was approximately 100 nm. Figure 5 (a)).

[0053] (2) Fabrication of a gold tunnel junction device with a switch control electrode. Voltage-current scanning (IV scanning) was performed on the aforementioned gold nanoribbon device with a switch control electrode. The scanning voltage was 0–2 V, and the scanning speed was 0.2 V / s. This generated approximately 1.3 mA of current in the gold nanoribbon, causing migration and breakage at the neck of the gold nanoribbon, forming a tunneling gap. This yielded a gold tunnel junction device with a switch control electrode. The switch control electrode of this device has an insulating substrate structure with a conductor on the back side, specifically a silicon dioxide-low resistivity silicon substrate. Figure 5 (b)).

[0054] (3) Place the substrate on the MSI TS150 probe stage. The silicon substrate of the switch control electrode contacts the probe stage base electrode, and the electrode plates of the tunnel junction are connected to one probe respectively. Connect the three electrodes to the KeySight B1500A semiconductor analyzer via coaxial cables to complete the electrical connection.

[0055] (4) Perform an initial IV scan on the device. The scan range is 0–20V, the scan speed is 1V / s, and the current limit is 1nA. Repeat the scan 10 times to ensure the device reaches the current limit. Calculate the median value of the tunneling current obtained from the 10 measurements without a switching control voltage. This is the initial state of the device, as shown below. Figure 6 As shown.

[0056] (5) Apply a switching control voltage to the device. The switching control voltage is 90V.

[0057] (6) Under a 90V switching control voltage, repeat the IV scan in step 3. Calculate the median value of the tunneling current obtained from 10 measurements under a 90V switching control voltage. It can be seen that the tunneling current of the device decreases to less than 100pA at this time, meaning the device has switched to the off state, and its gate controllable characteristics are as follows: Figure 6 As shown.

[0058] The embodiments described above are not intended to limit the present invention. Any person skilled in the art can make various changes and modifications without departing from the spirit and scope of the present invention. Therefore, the scope of protection of the present invention is defined by the scope of the claims.

Claims

1. A method for implementing a gate-controllable switching device based on a metal tunnel junction, characterized in that the steps include... include: 1) Fabrication of metal nanoribbons: Metal nanoribbons are fabricated by photolithography or electron beam exposure, metal deposition and lift-off techniques, or focused ion beam local deposition methods. The metal nanoribbons have a neck structure with a width of less than 100 nm in the middle. 2) Metal nanoribbons are fabricated into metal tunneling junctions, wherein the tunneling gap is at the nanometer or atomic scale. 3) A switch control electrode is provided at the bottom, side or top of the above-mentioned metal tunnel junction, wherein the switch control electrode and the tunnel junction are in a vacuum or insulating medium, and the metal tunnel junction and the switch control electrode constitute a gate-controlled switch. 4) Connect the gate-controllable switch to the controller; 5) Adjust the controller, apply a voltage waveform to the tunnel junction, perform IV scanning, and ensure that its tunneling current reaches the value required for the turn-on state. This turn-on state is the initial state of the gate-controllable switch. 6) Set the control voltage for the gate-controlled switch, apply this voltage to the switch control electrode through the controller, and keep it stable; 7) When a switching control voltage is applied, a voltage waveform is applied to the tunnel junction and an IV scan is performed to read the state of the switching control electrode. If a high switching control voltage is applied, the tunnel junction switches to the off state; if a low or 0V switching control voltage is applied, the tunnel junction remains in the on state.

2. The method for implementing a gate-controllable switching device based on a metal tunnel junction as described in claim 1, characterized in that, The metal nanoribbons are made of metals such as gold, palladium, platinum, aluminum, indium, silver, or tantalum.

3. The method for implementing a gate-controllable switching device based on a metal tunnel junction as described in claim 1, characterized in that, Metal nanoribbons were fabricated into metal tunnel junctions using electromigration, gated mechanical pyrolysis, helium ion beam etching, and hard mask-plasma etching methods.

Images