MoTe2 thin layer processing and thickness detection method based on visual identification algorithm

Abstract

The invention discloses a MoTe2 thin layer processing and thickness detection method based on a visual identification algorithm. The method comprises the following steps: step 1, thinning MoTe2; 2, constructing a color-thickness database (hereinafter referred to as C-T database) by means of optical contrast changes between different thicknesses of MoTe2; and 3, screening the two-dimensional material by means of a visual identification algorithm. According to the method, the two-dimensional thin-layer material can be effectively prepared, the sample preparation success rate and the thin-layer quality are improved, and meanwhile, the sample searching time in the sample preparation process can be greatly shortened by setting up a C-T database and by means of a visual identification algorithm.

Description

Technical Field

[0001] The present invention relates to the technical field of detection instruments, and in particular to thin-layer processing of MoTe2 and thickness detection and identification screening of two-dimensional materials with optical contrast varying with thickness. Background Art

[0002] In recent years, two-dimensional transition metal dichalcogenides (2MTDs) have become a research hotspot in the field of nanomaterials due to their unique electronic, optical, and mechanical properties. Molybdenum telluride disulfide (MoTe2) exhibits great potential in flexible electronics due to its diverse structural phases (such as 2H and 1T' phases), minimal differences in ground-state energy, and distinct physical properties. It also provides an ideal platform for studying non-thermally induced phase transitions under the precise control of femtosecond lasers. Principle studies often involve the preparation of thin-layer samples.

[0003] Currently, thickness detection mainly relies on atomic force microscopy (AFM) and optical contrast analysis. Although AFM has high precision, its detection time is long and its efficiency is low. Thickness estimation through color contrast using a separate optical microscope is affected by interference from substrate reflectivity, lighting conditions, etc., resulting in large errors.

[0004] In order to improve the efficiency of sample preparation during the experiment, the present invention proposes a MoTe2 thin layer processing and thickness detection method based on a visual recognition algorithm. Summary of the Invention

[0005] The embodiment of the present application provides a MoTe2 thin layer processing and thickness detection method based on a visual recognition algorithm, which thins the MoTe2 material by a specific method and uses a visual recognition algorithm to perform two-dimensional material screening.

[0006] The purpose of the present invention is to provide a MoTe2 thin layer processing and thickness detection method based on a visual recognition algorithm, aiming to solve or improve at least one of the above-mentioned technical problems.

[0007] To achieve the above object, the present invention provides the following solution: The present invention provides a MoTe2 thin layer processing and thickness detection method based on a visual recognition algorithm, comprising the following steps:

[0008] Step 1: First, take a 200mm long tape, fold the two sides of the tape inward by 30mm, take out a small amount of MoTe2 sample (about 1m×1m) and place it on the right side of the center of the tape (as shown in the attached figure). During the folding process, use your fingers to quickly tear the area where the sample is located before and after the folding from the non-adhesive side. By folding the sample four times, the sample is thinned four times and concentrated in a rectangular area of 10mm×10mm. At the same time, cut a 12mm×12mm piece of PDMS material and stick it on a glass slide after ultrasonic cleaning. Place the sample on the tape on the glass slide and scrape it evenly with tweezers. When the optical contrast of the tape in the sample area changes, slowly peel off the tape.

[0009] Step 2: Prepare a standard sample sequence containing 2-10 MoTe2 thin layers according to the method in step 1. Adjust the light source intensity, record the optical intensity parameters, take optical photos of each sample area, record the magnification and scale, perform high-precision AFM scanning to obtain the true thickness, and correspond the optical image, thin layer and background grayscale difference, and thickness one by one. Mark the location of the standard thin layer to obtain the CT database;

[0010] Step 3: Considering that the resolution of the optical image is too high, the image is downsampled to reduce the image size and reduce the computational complexity of image recognition. According to the experimental results, the downsampling factor is set to 4 in this patent. The SegmentAnything instance segmentation algorithm is used to perform visual recognition operations on the image, and the sample is separated from the image to form a bounding box. Considering that the grayscale image only contains brightness information, compared with the RGB image, each pixel only needs one value to represent the brightness, the data volume is smaller, and the processing speed is faster. The image is converted into a grayscale image, and the corresponding grayscale mean is extracted for each bounding box (sample area). Since the optical contrast of MoTe2 of different thicknesses is different, it has different grayscale values compared to the background, and the grayscale values of interference sources such as bubbles are smaller than the background in the grayscale image. At the same time, the background grayscale is widely distributed in the sample image sequence. Therefore, the grayscale with low distribution and large area distribution in the image is determined as the background grayscale in the algorithm, and the points less than or equal to the background grayscale are filtered out. Then, in each bounding box (sample area), the grayscale of the sample is filtered out. The DBSCAN clustering algorithm is used in the box to perform clustering by adjusting eps (the maximum distance between two samples considered as neighbors) and min_sample (the minimum number of samples considered during clustering). At the same time, min_area (the minimum area during clustering) is set by the image scale and the intended area. The mean of each cluster is calculated and stored, and the clustering result with the smallest mean is drawn with a rectangle on the original image. The difference between its grayscale mean μ and the background grayscale is obtained and compared with the standard grayscale sequence result to obtain a coarse layer number estimate. At the same time, the DetectRS visual recognition algorithm is used to uniformly sample the known MoTe2 sample area in the optical image, extract the grayscale mean and standard deviation of each sampling point, and the image features of the center pixel grayscale value, and use the calibrated thickness data to train the regression model to achieve the prediction of 2-10 layer thickness. Finally, the final thickness value is obtained by combining the coarse layer number estimate and the prediction of the DetectRS algorithm.

[0011] The present invention discloses the following technical effects:

[0012] The MoTe2 thin layer processing and thickness detection method based on the visual recognition algorithm of the present invention standardizes the method for preparing two-dimensional material thin materials. At the same time, through the visual recognition technology, it can more accurately find the required two-dimensional material area through the combination of grayscale and DetectRS algorithm, detect the thin layer thickness, shorten the sample preparation time, and improve the sample preparation efficiency. At present, it can realize screening and recognition of optical images under 20x objective lens, 50x objective lens and 100x objective lens. The larger the objective lens magnification of the captured image, the higher the recognition accuracy. BRIEF DESCRIPTION OF THE DRAWINGS

[0013] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the present invention is further described below with reference to the accompanying drawings and embodiments:

[0014] Figure 1 This is the thinning process provided by the embodiment of the present application;

[0015] Figure 2 This is the application interface and examples provided by the embodiments of this application;

[0016] Figure 3 This is the CT database construction and visual recognition algorithm process provided in the embodiment of this application. DETAILED DESCRIPTION

[0017] The following will clearly and completely describe the technical solutions in the embodiments of the present invention in conjunction with the accompanying drawings. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making creative efforts are within the scope of protection of the present invention.

[0018] In order to make the above-mentioned objects, features and advantages of the present invention more obvious and easy to understand, the present invention is further described in detail below with reference to the accompanying drawings and specific embodiments.

[0019] Reference Figure 1-3 The present invention provides a MoTe2 thin layer processing and thickness detection method based on a visual recognition algorithm, comprising the following steps:

[0020] Step 1: Thinning the MoTe2 material;

[0021] Step 2: Prepare standard sample sequences and construct a CT database;

[0022] Step 3: Use visual recognition algorithm to determine the thinnest area and its thickness in the target image;

[0023] To further optimize the solution, in step 2, for materials with slightly smaller optical contrast changes, the sample sequence can be prepared by transferring them to SiO2 coated with a blue-purple oxide film.

[0024] In the description of the present invention, it should be understood that the terms "longitudinal", "transverse", "up", "down", "front", "back", "left", "right", "vertical", "horizontal", "top", "bottom", "inside", "outside", etc., indicating the orientation or position relationship, are based on the orientation or position relationship shown in the accompanying drawings, and are only for the convenience of describing the present invention, rather than indicating or implying that the device or element referred to must have a specific orientation, be constructed and operated in a specific orientation, and therefore cannot be understood as a limitation on the present invention.

[0025] The embodiments described above are merely descriptions of preferred embodiments of the present invention and are not intended to limit the scope of the present invention. Without departing from the spirit of the present invention, various modifications and improvements made to the technical solutions of the present invention by persons skilled in the art should fall within the scope of protection defined by the claims of the present invention.

Claims

1. A MoTe2 thin layer processing and thickness detection method based on visual recognition algorithm, characterized in that: The following steps are involved: Step 1: Use tape to thin the MoTe2 according to a specific method, and then transfer the MoTe2 to a glass slide according to a specified process after thinning; Step 2: After data detection using optical microscopy and atomic force microscopy, CT data is constructed according to the specified method; Step 3: Use specific algorithms to visually identify two-dimensional materials.

2. The MoTe2 thin layer processing and thickness detection method based on the visual recognition algorithm according to claim 1 is characterized in that: After taking out the MoTe2, thinning and transfer are performed on the tape according to the specified operation and specified process parameters.

3. The MoTe2 thin layer processing and thickness detection method based on the visual recognition algorithm according to claim 1 is characterized in that: In step 2, a CT database is established with the help of an optical microscope and an atomic force microscope according to a specified scheme, and an improved process is proposed to improve the contrast between the two-dimensional material and the substrate color when establishing a CT database of materials with optical contrast that varies with thickness.

4. The MoTe2 thin layer processing and thickness detection method based on the visual recognition algorithm according to claim 1 is characterized in that: In step three, MoTe2 is identified and screened using a specific visual recognition algorithm.

Images