Scanning tunneling microscope and artificial intelligence assisted control method thereof

Abstract

The present application relates to a kind of scanning tunneling microscopes and its artificial intelligence auxiliary control method, the scanning tunneling microscope includes sample stage, scanning main body device and shock absorbing frame;The sample stage includes first platform, cross slide and object platform arranged sequentially from bottom to top;The scanning main body device includes second platform, third platform and top platform arranged sequentially from bottom to top, and scanning head main body and driving device, sample stage and scanning main body device are placed on shock absorbing frame as a whole.The present application scanning tunneling microscope has the characteristics of lower cost, easy to build, and the space occupied is small, and its artificial intelligence auxiliary control method, using artificial intelligence is realized Automation control, greatly reduce the difficulty of equipment operation, it is advantageous to the completion of high-throughput scanning.

Description

Technical Field

[0001] This invention belongs to the field of scanning tunneling microscopy technology, and specifically provides a scanning tunneling microscope and its artificial intelligence-assisted control method. Background Technology

[0002] Scanning tunneling microscopy (STM) is an instrument that uses the tunneling effect in quantum mechanics to probe the surface structure of materials. It measures the surface morphology and electronic density of states of a sample using the quantum tunneling effect to study material properties, and has wide applications and important roles in surface science, materials science, and life sciences. Its working principle is as follows: when the distance between the sample and the probe tip is sufficiently close (typically less than 1 nanometer), a tunneling current is formed between the tip and the sample surface under the influence of an applied electric field. The magnitude of the tunneling current is related to the distance between the tip and the sample surface. By observing the minute changes in the tunneling current between the tip and the sample surface, the precise surface morphology of the sample can be obtained.

[0003] However, the operation of scanning probe microscopy is highly dependent on the operator's experience and understanding of the system being measured. Especially in order to obtain a clear signal, a stable testing environment is required, and the operator needs to make continuous adjustments for different situations. Even experienced researchers often need several hours or even days to master this technique. Therefore, scanning probe microscopy presents the problems of high operational difficulty and long operating time. Summary of the Invention

[0004] Purpose of the invention: In order to solve the problems existing in the prior art, the present invention provides a scanning tunneling microscope and its artificial intelligence-assisted control method.

[0005] Technical Solution: This invention provides a scanning tunneling microscope, including a sample stage, a scanning main body device, and a shock-absorbing frame. The sample stage includes a first platform, a cross slide, and a loading platform arranged sequentially from bottom to top. The cross slide is fixed on the first platform, and the loading platform can move horizontally on the cross slide. The scanning main body device includes a second platform, a third platform, and a top platform arranged sequentially from bottom to top, as well as a scanning head body and a driving device. The scanning head body is installed at the bottom of the third platform. The second platform and the third platform are elastically connected. The driving device passes through the top platform and the third platform and can drive the third platform to move the scanning head body up and down. An opening is provided on the second platform directly below the scanning head body. The loading platform is located below the opening. The second platform and the top platform are respectively fixedly connected to the first platform. The sample stage and the scanning main body device are placed on the shock-absorbing frame as a whole.

[0006] As a preferred, improved, or specific implementation plan:

[0007] The cross slide consists of two vertically arranged linear slides. Each linear slide includes a motor, a guide rail, a lead screw, and a slider. The lead screw is connected to the guide rail, and the slider passes through the lead screw and is locked on the guide rail. The motor is fixed on the guide rail to drive the slider to slide in a certain direction. The two linear slides are arranged one above the other. The bottom of the upper linear slide is fixed to the slider of the lower linear slide, and a loading platform is fixed on the slider of the upper linear slide.

[0008] The drive device includes a high-precision threaded pair, a coupling, and a stepper motor. The threaded bushing of the high-precision threaded pair is fixed on the second platform. The high-precision threaded pair passes through the third platform and is exposed, and is connected to the stepper motor through the coupling. The stepper motor is located on the upper part of the top platform. Preferably, the drive device is evenly arranged in 3 sets.

[0009] The second platform and the third platform are elastically connected by springs; preferably, the springs are evenly arranged in three groups.

[0010] The third platform is also equipped with a displacement sensor for detecting the displacement of the slider.

[0011] The scanning head body includes a piezoelectric buzzer, an aluminum cylinder, a sapphire crystal, and a platinum-iridium needle tip arranged sequentially from top to bottom. The circular electrode of the piezoelectric buzzer is cut into four equal quadrants, and the needle tip is controlled by applying different voltages to the four quadrants.

[0012] The shock-absorbing frame includes an elastic suspension device, a buffer pad, an eddy current device, and a support. The elastic suspension device includes an upper top plate and a lower main plate. The top plate is fixed on the support and the top plate and the main plate are elastically connected. The buffer pad is placed on the upper surface of the main plate. The sample stage and the scanning main body are placed on the buffer pad as a whole. The eddy current device is located below the main plate.

[0013] Furthermore, the buffer pad is composed of alternating stacks of thick steel plates and fluororubber rings; the eddy current device includes an aluminum plate and a ring magnet, with the aluminum plate mounted on the lower surface of the main board and the ring magnet placed below the aluminum plate; the top plate and the main board are elastically connected by springs; preferably, three springs are evenly arranged.

[0014] This invention proposes a method for detecting material surface images using a scanning tunneling microscope according to the above-mentioned method, comprising the following steps:

[0015] S1: Select the number of areas to be scanned, and the system will automatically select the scanning areas;

[0016] S2: Scan the selected area:

[0017] (a) Bring the probe close to the sample surface until a tunneling current is detected between the probe tip and the sample surface, then stop approaching further; set the initial scanning parameters: probe tip bias: 1V, tunneling current between the probe tip and the sample surface: 1nA, scanning speed: 5s / 100nm.

[0018] (b) According to the parameters set in step (a), the scanning adopts constant current mode, that is, during the scanning of the sample surface, the height of the needle tip is changed to keep the magnitude of the tunneling current constant. The change of the needle tip height reflects the sample surface structure. After the scanning is completed, the height signal of the needle tip is obtained.

[0019] S3: Check whether the needle tip height signal obtained by scanning contains noise through the first neural network model. If there is noise, perform noise removal and repeat steps S2-S3 until there is no noise.

[0020] S4: After noise elimination, the image obtained in step S2 is input into the second neural network model to obtain the corresponding tip bias voltage and tunnel current;

[0021] S5: Based on the tip bias voltage and tunneling current obtained in step S4, scan the sample surface to obtain the final scan image.

[0022] Furthermore, in step S1, the system automatically selects the region according to the following rules: In the entire scannable range with length X and width Y, the number of scanned regions is N. Let N = A * B, where A and B are two positive integers N. And among all combinations of A * B = N, the combination with the smallest |AB| is selected. Then the coordinates of the center of each scanned region are (X * (2j-1) / 2A, Y * (2k-1) / 2B), j = 1, 2, ..., A, k = 1, 2, ..., B.

[0023] Furthermore, the method for step S3 is as follows:

[0024] The needle tip height signal is input into the first neural network model, and the noise category is output.

[0025] The first neural network model is trained using the needle tip height signal as input and the corresponding noise category as output. The input needle tip height signal and the output classification labels are as follows:

[0026] 0 - The signal contains no noise;

[0027] 1- The signal contains noise, and the source of the noise is the electrical signal of the device;

[0028] 2- The signal contains noise, and the noise source is the sample itself;

[0029] 3- The signal contains noise, and the source of the noise is vibration from the equipment environment;

[0030] 4- The signal contains noise, and the noise source is the needle tip;

[0031] If the noise output is not 0, meaning the signal contains noise, then the noise is processed accordingly, and steps S2-S3 are repeated until the output is 0, meaning the signal does not contain noise.

[0032] Furthermore, the method for step S4 is as follows:

[0033] The second neural network model is obtained by training a scanned image with initial scan parameters as input and corresponding tip bias voltage and tunneling current as output.

[0034] Beneficial effects: Compared with the prior art, the present invention provides a scanning tunneling microscope with the characteristics of low cost, easy construction and small space occupation. Furthermore, the present invention provides an artificial intelligence-assisted control method, which realizes automated control by using artificial intelligence, greatly reduces the difficulty of equipment operation and is conducive to the completion of high-throughput scanning. Attached Figure Description

[0035] Figure 1 This is a schematic diagram of the overall structure of the scanning tunneling microscope of the present invention.

[0036] Figure 2 This is a schematic diagram of the sample stage and scanning main body device in the scanning tunneling microscope of the present invention.

[0037] Figure 3 This is a schematic diagram (top view) of the sample stage in the scanning tunneling microscope of the present invention.

[0038] Figure 4 This is a schematic diagram of the scanning main device in the scanning tunneling microscope of the present invention.

[0039] Figure 5 This is a schematic diagram of the structure of the second platform in the scanning tunneling microscope of the present invention.

[0040] Figure 6 This is a schematic diagram of the structure of the third platform in the scanning tunneling microscope of the present invention.

[0041] Figure 7 This is a schematic diagram of the main body of the scanning head in the scanning tunneling microscope of the present invention.

[0042] Figure 8 This is a schematic diagram of the vibration damping device in the scanning tunneling microscope of the present invention.

[0043] Figure 9 This is a flowchart illustrating the fully automated scanning process according to an embodiment of the present invention.

[0044] Figure 10This is a flowchart illustrating process S2 of the present invention. Detailed Implementation

[0045] The present invention will be further described below.

[0046] In the description of this invention, it should be understood that if terms such as "upper," "lower," "left," "right," "top," "bottom," "inner," and "outer" indicate the orientation or positional relationship based on the orientation or positional relationship shown in the accompanying drawings, they are only for the convenience of describing this invention and simplifying the description, and do not indicate or imply that the device or element referred to must have a specific orientation, or be constructed and operated in a specific orientation. Therefore, the terms used to describe positional relationships in the accompanying drawings are only for illustrative purposes and should not be construed as limiting this patent.

[0047] Example 1

[0048] A scanning tunneling microscope, such as Figure 1 As shown, it includes a sample stage 1, a scanning main body device 2, and a shock-absorbing frame 3;

[0049] like Figure 2 and Figure 3 As shown, the sample stage 1 includes a first platform 11, a cross slide 12, and a loading platform 13 arranged sequentially from bottom to top. The cross slide 12 is fixed on the first platform 11, and the loading platform 13 can move horizontally under the drive of the cross slide 12. The cross slide 12 is composed of two vertically arranged linear slides (combined in the X-axis and Y-axis directions). Each linear slide includes a motor 121, a guide rail 122, a lead screw 123, and a slider 124. The lead screw 123 is connected to the guide rail 122, and the slider 124 passes through the lead screw 123 and is locked on the guide rail 122. The motor 121 is fixed on the guide rail 122 to drive the slider 124 to slide in a certain direction. The two linear slides are arranged vertically, with the bottom of the upper linear slide fixed to the slider 124 of the lower linear slide, and the loading platform 13 is fixed on the slider 124 of the upper linear slide.

[0050] like Figure 2 and Figure 4As shown, the scanning main unit 2 includes a second platform 21, a third platform 22, and a top platform 23 arranged sequentially from bottom to top, as well as a scanning head body 24 and a driving device 25. The scanning head body 24 is installed at the bottom of the third platform 22. The second platform 21 and the third platform 22 are elastically connected by springs 26, and three sets of springs 26 are evenly arranged. The driving device 25 passes through the top platform 23 and the third platform 22 and can drive the third platform 22 to move the scanning head body 24 up and down. An opening is provided on the second platform 21 directly below the scanning head body 24, and the loading platform 13 is located below the opening. The second platform 21 and the top platform 23 are respectively fixedly connected to the first platform 11. The sample stage 1 and the scanning main unit 2 are placed on the shock-absorbing frame 3. The drive unit 25 includes a high-precision threaded pair 251, a coupling 252, and a stepper motor 253. The threaded bushing 2511 of the high-precision threaded pair 251 is fixed on the second platform 21. The high-precision threaded pair 251 passes through the third platform 22 and protrudes, and is connected to the stepper motor 253 through the coupling 252. The stepper motor 253 is located on the upper part of the top platform 23. Three sets of drive units 25 are evenly arranged. The third platform 22 is also equipped with a displacement sensor 27 for detecting the displacement of the slider 12.

[0051] The specific design of the second platform 21 is as follows: Figure 5 As shown, the second platform 21 has a large square hole 213 at the front, the size of which matches the size of the cargo platform 13. The cargo platform 13 is located below the large square hole 213. A lower hole 211 is opened on both sides and the rear end of the large square hole 213. A lower circular groove 212 is opened at each of the four corners of the second platform 21.

[0052] The specific design of the third platform 22 is as follows: Figure 6 As shown, the third platform 22 has a small square hole 221 and a circular hole 222 at the front, and an upper hole 223 on both sides and at the rear end of the circular hole 222; and an upper circular groove 224 at each of the four corners of the third platform 22.

[0053] The scanning head body 24 is fixed in the circular hole 222 of the third platform 22, the displacement sensor 27 is fixed in the small square hole 221 of the third platform 22, the threaded bushing 2511 of the high-precision threaded pair 251 is fixed in the lower hole 211 of the second platform 21, the high-precision threaded pair 21 is exposed through the upper hole 223 of the third platform 22, and is connected to the stepper motor 253 through a coupling 252. The lower end and the upper end of the spring 26 are fixed in the lower circular groove 212 and the upper circular groove 224, respectively, thereby elastically supporting the third platform 22 on the second platform 21.

[0054] like Figure 7As shown, the scanning head body 24 includes a piezoelectric buzzer 241, an aluminum cylinder 242, a sapphire crystal 243, and a platinum-iridium needle tip 244 arranged sequentially from top to bottom. The circular electrode 2411 of the piezoelectric buzzer 241 is cut equally in four quadrants. The needle tip is controlled by applying different voltages to the four quadrants. Specifically, applying a positive voltage to the +X region and an equally large reverse voltage to the -X region will cause the needle tip to move in the positive X-axis direction; the same applies to the Y-axis; for the movement of the needle tip in the Z-axis direction, voltages can be applied to all four quadrants simultaneously.

[0055] like Figure 8 As shown, the shock-absorbing frame 3 includes an elastic suspension device 31, a buffer pad 32, an eddy current device 33, and a support 34. The elastic suspension device 31 includes an upper top plate 311 and a lower main plate 312. The top plate 311 is fixed on the support 34, and the top plate 311 and the main plate 312 are elastically connected. The buffer pad 32 is placed on the upper surface of the main plate 312. The sample stage 1 and the scanning main body device 2 are placed on the buffer pad 32. The eddy current device 33 is located below the main plate 312. The buffer pad 32 is composed of alternating stacks of thick steel plates and fluororubber rings. The eddy current device 33 includes an aluminum plate 331 and a ring magnet 332. The aluminum plate 331 is installed on the lower surface of the main plate 312, and the ring magnet 332 is placed below the aluminum plate 331. The top plate 311 and the main plate 312 are elastically connected by springs 313, and three springs 313 are evenly arranged.

[0056] This invention proposes a method for detecting material surface images using a scanning tunneling microscope according to the above-mentioned method, comprising the following steps:

[0057] S1: Select the number of areas to be scanned, and the system will automatically select the scanning areas;

[0058] S2: Scan the selected area:

[0059] (a) Bring the probe close to the sample surface until a tunneling current is detected between the probe tip and the sample surface, then stop approaching further; set the initial scanning parameters: probe tip bias: 1V, tunneling current between the probe tip and the sample surface: 1nA, scanning speed: 5s / 100nm.

[0060] (b) According to the parameters set in step (a), the scanning adopts constant current mode, that is, during the scanning of the sample surface, the height of the needle tip is changed to keep the magnitude of the tunneling current constant. The change of the needle tip height reflects the sample surface structure. After the scanning is completed, the height signal of the needle tip is obtained.

[0061] S3: Check whether the needle tip height signal obtained by scanning contains noise through the first neural network model. If there is noise, perform noise removal and repeat steps S2-S3 until there is no noise.

[0062] S4: After noise elimination, the image obtained in step S2 is input into the second neural network model to obtain the corresponding tip bias voltage and tunnel current;

[0063] S5: Based on the tip bias voltage and tunneling current obtained in step S4, scan the sample surface to obtain the final scan image.

[0064] Furthermore, in step S1, the system automatically selects the region according to the following rules: In the entire scannable range with length X and width Y, the number of scanned regions is N. Let N = A * B, where A and B are two positive integers N. And among all combinations of A * B = N, the combination with the smallest |AB| is selected. Then the coordinates of the center of each scanned region are (X * (2j-1) / 2A, Y * (2k-1) / 2B), j = 1, 2, ..., A, k = 1, 2, ..., B.

[0065] Furthermore, the method for step S3 is as follows:

[0066] The needle tip height signal is input into the first neural network model, and the noise category is output.

[0067] The first neural network model is trained using the needle tip height signal as input and the corresponding noise category as output. The input needle tip height signal and the output classification labels are as follows:

[0068] 0 - The signal contains no noise;

[0069] 1- The signal contains noise, and the source of the noise is the electrical signal of the device;

[0070] 2- The signal contains noise, and the noise source is the sample itself;

[0071] 3- The signal contains noise, and the source of the noise is vibration from the equipment environment;

[0072] 4- The signal contains noise, and the noise source is the needle tip;

[0073] If the noise output is not 0, meaning the signal contains noise, then the noise is processed accordingly, and steps S2-S3 are repeated until the output is 0, meaning the signal does not contain noise.

[0074] Furthermore, the method for step S4 is as follows:

[0075] The second neural network model is obtained by training a scanned image with initial scan parameters as input and corresponding tip bias voltage and tunneling current as output.

[0076] Although embodiments of the invention have been shown and described, it will be understood by those skilled in the art that various changes, modifications, substitutions and alterations can be made to these embodiments without departing from the principles and spirit of the invention, the scope of which is defined by the appended claims and their equivalents.

Claims

1. A scanning tunneling microscope, characterized in that, The device includes a sample stage (1), a scanning main unit (2), and a shock-absorbing frame (3). The sample stage (1) includes a first platform (11), a cross slide (12), and a loading platform (13) arranged sequentially from bottom to top. The cross slide (12) is fixed on the first platform (11), and the loading platform (13) can move horizontally under the drive of the cross slide (12). The scanning main unit (2) includes a second platform (21), a third platform (22), and a top platform (23) arranged sequentially from bottom to top, as well as a scanning head body (24) and a driving device (25). Installed at the bottom of the third platform (22), the second platform (21) and the third platform (22) are elastically connected. The driving device (25) passes through the top platform (23) and the third platform (22) and can drive the third platform (22) to move the scanning head body (24) up and down. An opening is provided on the second platform (21) directly below the scanning head body (24). The loading platform (13) is located below the opening. The second platform (21) and the top platform (23) are fixedly connected to the first platform (11) respectively. The sample stage (1) and the scanning body device (2) are placed on the shock-absorbing frame (3). The scanning head body (24) includes a piezoelectric buzzer (241), an aluminum cylinder (242), a sapphire crystal (243), and a platinum-iridium needle tip (244) arranged sequentially from top to bottom. The circular electrode (2411) of the piezoelectric buzzer (241) is cut according to the four quadrants. The needle tip is controlled by applying different voltages to the four quadrants.

2. The scanning tunneling microscope according to claim 1, characterized in that, The cross slide (12) consists of two vertically arranged linear slides. Each linear slide includes a motor (121), a guide rail (122), a lead screw (123), and a slider (124). The lead screw (123) is connected to the guide rail (122). The slider (124) passes through the lead screw (123) and is locked on the guide rail (122). The motor (121) is fixed on the guide rail (122) to drive the slider (124) to slide in a certain direction. The two linear slides are arranged above and below each other. The bottom of the upper linear slide is fixed on the slider (124) of the lower linear slide. The loading platform (13) is fixed on the slider (124) of the upper linear slide.

3. The scanning tunneling microscope according to claim 1, characterized in that, The drive device (25) includes a high-precision threaded pair (251), a coupling (252) and a stepper motor (253). The threaded bushing (2511) of the high-precision threaded pair (251) is fixed on the second platform (21). The high-precision threaded pair (251) passes through the third platform (22) and is exposed, and is connected to the stepper motor (253) through the coupling (252). The stepper motor (253) is located on the upper part of the top platform (23).

4. The scanning tunneling microscope according to claim 1, characterized in that, The second platform (21) and the third platform (22) are elastically connected by a spring (26); the third platform (22) is also provided with a displacement sensor (27) for detecting the displacement of the slider (12).

5. The scanning tunneling microscope according to claim 1, characterized in that, The shock-absorbing frame (3) includes an elastic suspension device (31), a buffer pad (32), an eddy current device (33), and a bracket (34). The elastic suspension device (31) includes an upper top plate (311) and a lower main plate (312). The top plate (311) is fixed on the bracket (34). The top plate (311) and the main plate (312) are elastically connected. The buffer pad (32) is placed on the upper surface of the main plate (312). The sample stage (1) and the scanning main body device (2) are placed on the buffer pad (32). The eddy current device (33) is located below the main plate (312).

6. The scanning tunneling microscope according to claim 5, characterized in that, The buffer pad (32) is made of alternating stacks of thick steel plates and fluororubber rings; the eddy current device (33) includes an aluminum plate (331) and an annular magnet (332), the aluminum plate (331) is installed on the lower surface of the main plate (312), and the annular magnet (332) is placed below the aluminum plate (331); the top plate (311) and the main plate (312) are elastically connected by a spring (313).

7. A method for detecting images of a material surface using a scanning tunneling microscope according to any one of claims 1-6, characterized in that, Includes the following steps: S1: Select the number of areas to be scanned, and the system will automatically select the scanning areas; S2: Scan the selected area: (a) Approach the probe close to the sample surface until a tunneling current is detected between the probe tip and the sample surface, then stop approaching; set the initial scanning parameters: probe tip bias: 1 V, tunneling current between the probe tip and the sample surface: 1 nA, scanning speed: 5 s / 100 nm. (b) According to the parameters set in step (a), the scanning adopts constant current mode, that is, during the scanning of the sample surface, the height of the needle tip is changed to keep the magnitude of the tunneling current constant. The change of the needle tip height reflects the sample surface structure. After the scanning is completed, the height signal of the needle tip is obtained. S3: Check whether the needle tip height signal obtained by scanning contains noise through the first neural network model. If there is noise, perform noise removal and repeat steps S2-S3 until there is no noise. S4: After noise elimination, the image obtained in step S2 is input into the second neural network model to obtain the corresponding tip bias voltage and tunnel current; S5: Based on the tip bias voltage and tunneling current obtained in step S4, scan the sample surface to obtain the final scan image.

8. The method according to claim 7, characterized in that, In step S1, the system automatically selects regions according to the following rules: In the entire scannable range with length X and width Y, the number of scanned regions is N. Let N = A * B, where A and B are two positive integers. Among all combinations of A * B = N, the combination with the smallest |AB| is selected. Then the coordinates of the center of each scanned region are [X * (2j-1) / (2A), Y * (2k-1) / (2B)], j = 1, 2, ..., A, k = 1, 2, ..., B.

9. The method according to claim 7, characterized in that, The method for step S3 is as follows: The needle tip height signal is input into the first neural network model, and the noise category is output. The first neural network model is trained using the needle tip height signal as input and the corresponding noise category as output. The input needle tip height signal and the output classification labels are as follows: 0 - The signal contains no noise; 1- The signal contains noise, and the source of the noise is the electrical signal of the device; 2- The signal contains noise, and the noise source is the sample itself; 3- The signal contains noise, and the source of the noise is vibration from the equipment environment; 4- The signal contains noise, and the noise source is the needle tip; If the noise output is not 0, meaning the signal contains noise, then the noise is processed accordingly, and steps S2-S3 are repeated until the output is 0, meaning the signal does not contain noise.

10. The method according to claim 7, characterized in that, The method for step S4 is as follows: The second neural network model is obtained by training a scanned image with initial scan parameters as input and corresponding tip bias voltage and tunneling current as output.

Images