Intelligent statistical method and system for wafer-level characterization of laser-processed carbon nanotubes

Abstract

The invention provides an intelligent statistical method and system for wafer-level characterization of laser-processed carbon nanotubes. The method comprises the following steps: S1, growing a carbon nanotube array or a network structure on a silicon wafer; s2, performing laser processing on the carbon nanotube array or the network structure by using femtosecond laser; s3, acquiring surface data of the silicon wafer after laser processing by adopting a multi-scale characterization technology; s4, constructing a multi-modal deep learning model according to the surface data of the silicon wafer; s5, performing intelligent statistical analysis according to an output result of the multi-modal deep learning model, and generating a wafer-level representation report; and S6, laser processing parameters are optimized according to the wafer level characterization report. According to the method, accurate processing and comprehensive characterization of the carbon nano tube on the wafer level are realized, and the manufacturing precision and the yield of the carbon nano tube device can be improved.

Description

Technical Field

[0001] The present invention relates to the field of new generation information technology, and in particular to an intelligent statistical method and system for wafer-level characterization of laser-processed carbon nanotubes. Background Art

[0002] With the continuous development of semiconductor technology, traditional silicon-based devices are approaching their physical limits. Carbon nanotubes (CNTs), as a nanomaterial with unique electrical and mechanical properties, show great potential in future electronic devices, sensors and energy storage. However, there are still many challenges in successfully integrating carbon nanotubes into large-scale semiconductor manufacturing processes, the most critical of which is how to accurately control and characterize the growth and processing of carbon nanotubes at the wafer level.

[0003] Currently, growing 2-5nm thick carbon nanotube arrays or grid films on 4-8 inch silicon wafers poses great challenges to their characterization and processing. Traditional processing methods are difficult to achieve nanometer-level precision, and conventional characterization techniques cannot evaluate the structure and performance of individual nanotubes. Laser processing, especially femtosecond laser technology, has become an ideal choice for processing carbon nanotubes due to its high precision and low heat-affected zone; however, to achieve uniform and controllable processing at the wafer level, it is still necessary to have a deep understanding of the interaction mechanism between lasers and carbon nanotubes and to precisely control laser parameters. In addition, defects and structural changes may be introduced during the processing, which need to be evaluated and quantified by advanced characterization techniques; however, how to effectively integrate multi-scale and multi-modal data and extract meaningful statistical information from them has become a problem that needs to be solved urgently.

[0004] In view of this, the present invention is proposed. Summary of the invention

[0005] The purpose of the present invention is to provide an intelligent statistical method and system for wafer-level characterization of laser-processed carbon nanotubes, which realizes accurate processing and comprehensive characterization of carbon nanotubes at the wafer level, and is conducive to improving the manufacturing accuracy and yield of carbon nanotube devices.

[0006] The present invention provides an intelligent statistical method for wafer-level characterization of laser-processed carbon nanotubes, comprising the following steps:

[0007] S1: growing carbon nanotube arrays or network structures on silicon wafers;

[0008] S2: Laser processing of carbon nanotube arrays or network structures using femtosecond lasers;

[0009] S3: Use multi-scale characterization technology to obtain surface data of silicon wafers after laser processing;

[0010] S4: Building a multimodal deep learning model based on silicon wafer surface data;

[0011] S5: Perform intelligent statistical analysis based on the output results of the multimodal deep learning model and generate wafer-level characterization reports;

[0012] S6: Optimize laser processing parameters based on wafer-level characterization report.

[0013] In step S1, the size of the silicon wafer is 4-8 inches; the surface of the silicon wafer is provided with a SiO2 layer with a thickness of 200-300 nm; the thickness of the carbon nanotube array or network structure is 2-5 nm; the density of the carbon nanotube array or network structure is 100-1000 tubes / μm 2 ; The average diameter of the carbon nanotube array or network structure is 1-2nm.

[0014] In step S2, the wavelength of the femtosecond laser is 400-800 nm; the pulse width is less than 500 fs; and the power density is 5×10 12 -8×10 12 W / cm 2 ; The repetition frequency is adjustable from 1Hz to 1MHz; the spot diameter is adjustable from 1-100μm.

[0015] In step S3, using multi-scale characterization technology to obtain silicon wafer surface data includes:

[0016] a) Use field emission scanning electron microscopy (FE-SEM) with a resolution better than 1 nm to obtain large-area morphological information;

[0017] b) Analyzing the atomic structure of a single carbon nanotube using a transmission electron microscope (TEM) with a resolution better than 0.5 nm;

[0018] c) evaluating the defect density and lattice changes of carbon nanotubes using confocal Raman spectroscopy with a spatial resolution of less than 1 μm;

[0019] d) The height and surface roughness of carbon nanotubes were measured using atomic force microscopy (AFM) with a vertical resolution better than 0.5 nm.

[0020] In step S4, the multimodal deep learning model has the following functions:

[0021] a) Identify and classify the structural features of carbon nanotubes, including diameter (accuracy ±0.1nm), length (accuracy ±1nm), chirality and defect types;

[0022] b) Evaluate the effect of laser processing on the carbon nanotube crystal structure and identify defect density changes (accuracy ±1%);

[0023] c) Analyze the uniformity and consistency of carbon nanotube distribution on the surface of the silicon wafer and generate a distribution map with a spatial resolution of 1 μm×1 μm.

[0024] The training data set of the multimodal deep learning model contains at least 5,000 samples; the accuracy of the multimodal deep learning model is not less than 95%.

[0025] In step S5, the intelligent statistical analysis step includes:

[0026] a) Evaluation of the growth density of nanotubes in different regions of the silicon wafer (accuracy ±1 nanotube / μm) 2 ) and orientation consistency (accuracy ±1°);

[0027] b) Analysis of the effect of laser processing on the structural integrity of carbon nanotubes, including defect type identification and density estimation (accuracy ±0.1 / nm);

[0028] c) Calculate the length and diameter distribution of carbon nanotubes after laser processing and generate a histogram (bin width is 0.1 nm);

[0029] d) Estimate the surface defect density and type of silicon wafers with a spatial resolution of 1mm×1mm.

[0030] In step S6, the laser processing parameters are optimized according to the wafer-level characterization report to implement closed-loop control so that the processing accuracy reaches ±1nm.

[0031] The present invention also provides an intelligent statistical system for laser processed carbon nanotube wafer-level characterization for implementing the above method, comprising:

[0032] Carbon nanotube growth module: grows carbon nanotube arrays or network structures on silicon wafers;

[0033] Femtosecond laser processing module: Use femtosecond laser to process carbon nanotube arrays or network structures;

[0034] Multi-scale characterization module: uses multi-scale characterization technology to obtain the surface data of silicon wafers after laser processing;

[0035] Multimodal deep learning module: processes the surface data of silicon wafers after laser processing and outputs the results;

[0036] Intelligent statistical analysis module: performs intelligent statistical analysis based on the output results of the multimodal deep learning model and generates wafer-level characterization reports;

[0037] Parameter Optimization Module: Optimize laser processing parameters based on wafer-level characterization reports.

[0038] Specifically, the carbon nanotube growth module can grow carbon nanotubes at a density of 100-1000 tubes / μm on a 4-8 inch wafer.2 The femtosecond laser processing module has an adjustable laser repetition frequency range of 1Hz-1MHz and a spot diameter range of 1-100μm. The multi-scale characterization module includes a FE-SEM with a resolution better than 1nm, a TEM with a resolution better than 0.1nm, a Raman spectrometer with a spatial resolution less than 1μm, and an AFM with a vertical resolution better than 0.5nm. The multimodal deep learning module has a processing speed of no less than 1GB / s, can process data from different characterization devices at the same time, and can identify and classify atomic-scale carbon nanotube structural defects as small as 0.1nm. The intelligent statistical analysis module can complete data analysis and report generation for the entire wafer within 1 hour, and can generate a heat map of nanotube distribution on the wafer surface with a spatial resolution of 10μm×10μm. The parameter optimization module can optimize the laser processing parameters to the optimal state within 10 iterations.

[0039] The laser processed carbon nanotube wafer-level characterization intelligent statistical system of the present invention also includes an automatic sample processing module, which can automatically complete the loading, unloading and transmission of wafers between different characterization devices, and the processing time of a single wafer does not exceed 30 minutes.

[0040] The laser processed carbon nanotube wafer-level characterization intelligent statistical system of the present invention also includes an in-situ monitoring module, which uses a high-speed camera (frame rate not less than 10,000fps) to monitor the structural changes of the carbon nanotubes in real time during the laser processing.

[0041] The laser processed carbon nanotube wafer-level characterization intelligent statistical system of the present invention also includes a database module for storing nano-nanotube characterization data under different growth conditions and laser processing parameters, and the storage capacity is not less than 10TB.

[0042] The laser processed carbon nanotube wafer-level characterization intelligent statistical system of the present invention also includes a human-computer interaction interface, which can display the laser processing and characterization process in real time and allow the operator to adjust the laser processing parameters according to the intermediate results, and the response time is less than 100ms.

[0043] The implementation of the present invention has at least the following advantages:

[0044] 1) The present invention can realize accurate laser processing of 2-5nm thick carbon nanotube layers on 4-8 inch wafers, with a processing accuracy of ±1nm;

[0045] 2) The present invention provides multi-scale, all-round characterization capabilities from atomic to wafer level, which is conducive to a deep understanding of the structure and properties of carbon nanotubes;

[0046] 3) The present invention realizes the rapid processing and analysis of large-scale data through artificial intelligence algorithms, which is conducive to generating high-precision wafer-level statistical reports;

[0047] 4) The present invention significantly improves the manufacturing efficiency and yield of carbon nanotube devices, accelerating their large-scale application in the semiconductor industry;

[0048] 5) The present invention provides a comprehensive platform for the research of carbon nanotubes and other nanomaterials, and promotes the development of nanotechnology. BRIEF DESCRIPTION OF THE DRAWINGS

[0049] In order to more clearly illustrate the specific implementation methods of the present invention or the technical solutions in the prior art, the drawings required for use in the specific implementation methods or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are some implementation methods of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0050] Figure 1 Schematic diagram of the structure of the intelligent statistical system for wafer-level characterization of laser-processed carbon nanotubes.

[0051] Description of reference numerals:

[0052] 1: Carbon nanotube growth module; 2: Femtosecond laser processing module; 3: Multi-scale characterization module; 4: Multimodal deep learning module; 5: Intelligent statistical analysis module; 6: Parameter optimization module; 7: In-situ monitoring module; 8: Database module; 9: Automated sample processing module; 10: Human-computer interaction interface. DETAILED DESCRIPTION

[0053] It should be noted that the following detailed descriptions are illustrative and are intended to provide further explanation of the present application. Unless otherwise specified, all technical and scientific terms used herein have the same meanings as those commonly understood by those skilled in the art to which the present application belongs.

[0054] It should be noted that the terms used herein are only for describing specific embodiments and are not intended to limit the exemplary embodiments according to the present application. As used herein, unless the context clearly indicates otherwise, the singular form also includes the plural form. In addition, it should be understood that when the terms "comprise" and / or "include" are used in this description, it indicates the presence of features, steps, operations, devices, components and / or combinations thereof.

[0055] The technical solution of the present invention will be clearly and completely described below in conjunction with the embodiments. Obviously, the described embodiments are part of the embodiments of the present invention, rather than all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0056] Example 1

[0057] The intelligent statistical method for wafer-level characterization of laser-processed carbon nanotubes in this embodiment is based on the network structure of single-walled carbon nanotubes (SWCNTs) grown on a 6-inch Si wafer, and the steps are as follows:

[0058] 1. Sample preparation:

[0059] First, a high-purity (99.999%) 6-inch p-type Si wafer was selected as the substrate.

[0060] A 280nm thick SiO2 layer is grown on the surface of the Si wafer by thermal oxidation method, which serves as an insulating layer and buffer layer for the subsequent growth of carbon nanotubes; the uniformity of the SiO2 layer thickness is controlled within the range of ±2nm to ensure the consistency of the subsequent carbon nanotube growth.

[0061] Subsequently, a 2-3 nm thick SWCNT network structure is grown on the surface of the SiO2 layer using an improved chemical vapor deposition (CVD) method; the specific steps are as follows:

[0062] a) Spin coating 0.01 mol / L Fe(NO3)3 ethanol solution on the surface of SiO2 layer as catalyst precursor;

[0063] b) placing the sample in a CVD furnace and introducing Ar gas (500 sccm) at 850°C for 30 minutes to form nano-sized Fe catalyst particles;

[0064] c) Switch the gas source to a mixed gas of CH4 (200 sccm) and H2 (50 sccm), and grow at 900°C for 15 minutes;

[0065] d) Cooling to room temperature to obtain a uniformly distributed SWCNT network structure.

[0066] Through the above optimized growth parameters, a SWCNT network structure with an average diameter of 1.2±0.2nm was obtained, and the density of the network structure reached 200±20 roots / μm 2 .

[0067] Raman spectroscopy confirmed that more than 90% of SWCNTs are semiconductor-type, laying the foundation for subsequent device applications.

[0068] 2. Laser processing:

[0069] A Ti:Sapphire femtosecond laser system is used for precise processing of SWCNT network structures. The main parameters of the system are as follows:

[0070] Central wavelength: 800nm;

[0071] Pulse width: 100fs;

[0072] Repetition frequency: adjustable range 1Hz-1MHz, this embodiment is set to 1kHz;

[0073] Maximum output power: 5W;

[0074] Spot diameter: adjustable range 1-100 μm, in this embodiment it is adjusted to 10 μm.

[0075] During the laser processing, the power density is precisely controlled within 5×10 12 -8×10 12 W / cm 2 range, which can achieve selective cutting and modification of SWCNT network structure without introducing too many defects.

[0076] The laser processing mechanism is mainly based on the synergistic effect of photothermal effect and photochemical effect:

[0077] a) Photothermal effect: Femtosecond laser pulses transfer energy to SWCNTs in a very short time, causing the local temperature to rise rapidly (estimated to reach 2000-3000K), causing CC bonds to break;

[0078] b) Photochemical effect: High-energy photons directly cause the breaking and recombination of chemical bonds in SWCNTs, especially at defect locations and tube ends.

[0079] By precisely controlling the above two effects, it is possible to achieve local modification of the network structure while maintaining the integrity of most of the SWCNTs. For example, metallic SWCNTs can be selectively cut off to improve the semiconducting properties of the network; or controlled defects can be introduced at specific locations to adjust the carrier transport properties.

[0080] In order to achieve uniform processing over a large area, a raster scanning strategy was adopted. The 6-inch wafer was divided into 1 cm × 1 cm areas, and a serpentine scan was performed in each area. The overlap rate between adjacent scan lines was set to 30% to ensure uniform processing. The processing time for the entire wafer was about 2 hours.

[0081] 3. Multi-scale representation:

[0082] Comprehensive analysis from atomic to wafer scale is achieved through a variety of high-resolution characterization techniques:

[0083] a) Field Emission Scanning Electron Microscopy (FE-SEM):

[0084] The whole wafer was scanned using a FE-SEM with a resolution of 0.8nm. A low accelerating voltage (1kV) was used to reduce damage to the SWCNTs. High-resolution imaging of the entire wafer was completed within 12 hours using automated stages and image stitching techniques. These images were used to analyze the large-area morphology, density distribution, and connectivity of the SWCNT network.

[0085] b) Transmission Electron Microscopy (TEM):

[0086] The atomic structure of a single SWCNT was analyzed using a TEM with a resolution of 0.2 nm. The sample was prepared using a direct transfer method: a layer of PMMA was applied to the wafer surface, and then the SiO2 layer was etched with HF to transfer the SWCNT network to a TEM copper grid. The SWCNT near the laser processed area was analyzed to observe its wall structure, defect type and distribution.

[0087] c) Confocal Raman Spectrometer:

[0088] A confocal Raman spectrometer with a spatial resolution of 0.5 μm was used with 532 nm laser excitation. The intensity ratio (IG / ID) of the G peak (~1590 cm-1) and the D peak (~1350 cm-1) was analyzed to evaluate the defect density change of SWCNT. -1 ) Analyze the diameter distribution of SWCNT. 100 points were selected on the wafer for measurement to obtain statistically significant changes in the Raman spectrum.

[0089] d) Atomic Force Microscopy (AFM):

[0090] The height distribution and surface roughness of the SWCNT network were measured using an AFM with a vertical resolution of 0.1 nm. Tapping mode was used to reduce the impact on the sample. Five 100 μm × 100 μm scanning areas were randomly selected in each 1 cm × 1 cm area to construct a three-dimensional topographic map of the SWCNT network.

[0091] 4. Multimodal deep learning analysis:

[0092] Build a complex multimodal deep learning model consisting of the following components:

[0093] Convolutional neural networks (CNNs) for SEM image analysis;

[0094] Graph Neural Network (GNN) for TEM atomic structure recognition;

[0095] Recurrent Neural Network (RNN) for Raman spectroscopy data processing;

[0096] 3D-CNN for AFM data analysis;

[0097] Multimodal fusion network,integrates data from different sources.

[0098] The model was trained using 8,000 pre-annotated samples, including SWCNT samples under various growth conditions and laser processing parameters. The training set was further expanded through data augmentation techniques such as rotation, flipping, and adding noise. The final model achieved a recognition accuracy of 96.5% on the validation set.

[0099] The above multimodal deep learning model has the following functions:

[0100] a) Identify the diameter (accuracy ±0.1nm), length (accuracy ±1nm) and chirality of SWCNTs; chirality identification is based on the combined analysis of TEM images and Raman spectroscopy;

[0101] b) Evaluate the effect of laser processing on the SWCNT network structure, including the change in defect density (accuracy ±1%) and defect type (e.g., five-membered ring, seven-membered ring, vacancy, etc.);

[0102] c) Analyze the uniformity and connectivity of the SWCNT network and generate a distribution map with a resolution of 1μm×1μm; connectivity analysis is crucial for predicting the performance of electronic devices.

[0103] 5. Intelligent statistical analysis:

[0104] Based on the output of the multimodal deep learning model, the data analysis of the entire 6-inch wafer is completed in 45 minutes, and a comprehensive characterization report is generated; the report content includes:

[0105] a) SWCNT network density distribution (accuracy ±1 root / μm) 2 ), presented in the form of a heat map;

[0106] b) SWCNT defect type and density distribution in the laser processed area (accuracy ±0.1 defect / nm), including a pie chart of defect type and a gradient map of density;

[0107] c) SWCNT diameter and length distribution histogram (bin width 0.1 nm), and the mean and standard deviation are calculated;

[0108] d) Evaluation of the uniformity of the SWCNT network on the wafer surface with a spatial resolution of 1 mm × 1 mm, quantified using the coefficient of variation (CV).

[0109] e) Prediction of network conductivity based on a comprehensive analysis of SWCNT density, orientation, and defect distribution.

[0110] 6. Closed-loop control optimization:

[0111] The laser processing parameters are automatically adjusted according to the characterization results; the optimization goal is set to maximize electrical power. The optimization process uses the Bayesian optimization algorithm to consider the mutual influence of multiple parameters. After 7 iterations, the following optimization parameters are obtained:

[0112] Laser power density: 6.5×1012W / cm 2 ;

[0113] Scanning speed: 2mm / s;

[0114] Spot overlap rate: 30%;

[0115] Pulse repetition frequency: 2kHz.

[0116] After optimization, the conductivity of the SWCNT network structure increased by 20% while maintaining a low defect density (IG / ID>100). A reduction in the contact resistance between SWCNTs was also observed through a local annealing effect.

[0117] 7. In-situ monitoring:

[0118] The laser processing process is monitored in real time using a high-speed CMOS camera with a frame rate of 20,000fps. The camera is triggered synchronously with the laser to capture the transient changes caused by each laser pulse. Through real-time image processing algorithms, the system is able to detect structural changes in the SWCNT network during processing, such as sudden brightness changes or morphological changes. This information is used to adjust the laser parameters in real time.

[0119] In addition, an integrated photoconductivity measurement module monitors the changes in the conductivity of the SWCNT network during processing by applying a microelectrode array on the sample surface, thereby providing direct feedback on changes in material properties.

[0120] 8. Data storage and management:

[0121] All experimental data and analysis results are stored in a 15TB high-performance solid-state hard disk array. A distributed database system is used to ensure fast data reading, writing and retrieval. Data organization adopts a multi-level structure:

[0122] First layer: raw data (SEM, TEM, Raman, AFM, etc.);

[0123] The second layer: processed data (image enhancement, spectral analysis results, etc.);

[0124] The third layer: deep learning model output;

[0125] Fourth level: statistical analysis reports and optimization suggestions.

[0126] The system automatically classifies and indexes data and implements data version control. Through machine learning algorithms, it is possible to mine potential process laws from historical data and provide guidance for future experimental design.

[0127] Example 2

[0128] Combination Figure 1 As shown, this embodiment provides a laser processed carbon nanotube wafer-level characterization intelligent statistical system for implementing the method of embodiment 1, including:

[0129] 1) Carbon nanotube growth module:

[0130] Carbon nanotube growth module 1 is used to grow carbon nanotubes with a density of 100-1000 tubes / μm on 4-8 inch wafers. 2 Nanotube arrays or network structures.

[0131] 2) Femtosecond laser processing module:

[0132] Femtosecond laser processing module 2 uses femtosecond laser to process carbon nanotube arrays or network structures. It uses a femtosecond laser system with a wavelength of 400-800nm ​​and a pulse width of less than 500fs. The power density can be between 5×10 12 -8×10 12 W / cm 2 Precise adjustment within the range.

[0133] Femtosecond laser processing module 2 can selectively process 2-5nm thick nanotube arrays or network structures to achieve nanometer-level precision cutting, modification or functionalization. The laser repetition frequency can be adjusted in the range of 1Hz-1MHz, and the spot diameter can be adjusted in the range of 1-100μm to meet different processing requirements.

[0134] 3) Multi-scale characterization module:

[0135] The multi-scale characterization module 3 uses multi-scale characterization technology to obtain the surface data of the silicon wafer after laser processing; the multi-scale characterization module 3 integrates a variety of high-resolution characterization equipment, including:

[0136] Field emission scanning electron microscopy (FE-SEM) with a resolution better than 1 nm;

[0137] Transmission electron microscopy (TEM) with a resolution better than 0.1 nm;

[0138] Confocal Raman spectrometer with a spatial resolution of less than 1 μm;

[0139] Atomic force microscopy (AFM) with vertical resolution better than 0.1 nm.

[0140] The combination of the above-mentioned high-resolution characterization equipment enables the system to comprehensively characterize the structure and properties of carbon nanotubes from the atomic to the wafer level.

[0141] 4) Multimodal deep learning module:

[0142] The multimodal deep learning module 4 processes the surface data of the silicon wafer after laser processing and outputs the results. It can simultaneously process data from different characterization devices with a processing speed of no less than 1GB / s.

[0143] Multimodal deep learning module 4 has the following functions:

[0144] Identify and classify the structural features of carbon nanotubes, including diameter (accuracy ±0.1nm), length (accuracy ±1nm), chirality and defect types;

[0145] Evaluate the effect of laser processing on the crystal structure of carbon nanotubes and identify changes in defect density (accuracy ±1%);

[0146] The uniformity and consistency of carbon nanotube distribution on the silicon wafer surface were analyzed, and a distribution map with a spatial resolution of 1μm×1μm was generated.

[0147] 5) Intelligent statistical analysis module:

[0148] The intelligent statistical analysis module 5 performs intelligent statistical analysis based on the output results of the multimodal deep learning model and generates a wafer-level characterization report. It can complete the data analysis and report generation of the entire wafer within 1 hour, including:

[0149] Evaluate the growth density of carbon nanotubes in different areas of silicon wafer (accuracy ±1 tube / μm) 2 ) and orientation consistency (accuracy ±1°);

[0150] Analyze the effect of laser processing on the structural integrity of carbon nanotubes, including defect type identification and density estimation (accuracy ±0.1 / nm).

[0151] Calculate the length and diameter distribution of carbon nanotubes after laser processing and generate a histogram (bin width is 0.1nm);

[0152] Estimate the density and type of defects on the silicon wafer surface with a spatial resolution of 1mm×1mm.

[0153] 6) Parameter optimization module:

[0154] The parameter optimization module 6 performs closed-loop control and parameter optimization on the laser processing parameters according to the wafer-level characterization report, and can optimize the laser processing parameters to the best state within 10 iterations, so that the processing accuracy reaches ±1nm.

[0155] 7) In-situ monitoring module:

[0156] The in-situ monitoring module 7 uses a high-speed camera (frame rate not less than 10,000 fps) to monitor in real time the structural changes of the nanotubes during the laser processing process, providing real-time feedback for process control.

[0157] 8) Database module:

[0158] The storage capacity of the database module 8 is not less than 10 TB, and it stores carbon nanotube characterization data under different growth conditions and processing parameters to provide data support for long-term process optimization.

[0159] 9) Automated sample processing module:

[0160] The automated sample processing module 9 can automatically complete the loading, unloading and transfer of wafers between different characterization devices. The processing time of a single wafer does not exceed 4 hours, which greatly improves work efficiency.

[0161] 10) Human-computer interaction interface:

[0162] The response time of the human-machine interface 10 is less than 100 ms, and it can display the processing and characterization process in real time, and allow the operator to adjust parameters according to the intermediate results.

[0163] The system of this embodiment integrates modules such as carbon nanotube growth, femtosecond laser processing, multi-scale characterization, multi-modal deep learning, intelligent statistical analysis and parameter optimization, and realizes precise laser processing of 2-5nm thick carbon nanotube layers on 4-8 inch wafers. It provides multi-scale and all-round characterization capabilities from atomic to wafer level, and realizes rapid processing and analysis of large-scale data through artificial intelligence algorithms, thereby improving the manufacturing accuracy and yield of carbon nanotube devices and laying the foundation for their application in large-scale integrated circuits.

[0164] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit it. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the aforementioned embodiments, or replace some or all of the technical features therein with equivalents. However, these modifications or replacements do not cause the essence of the corresponding technical solutions to deviate from the scope of the technical solutions of the embodiments of the present invention.

Claims

1. An intelligent statistical method for wafer-level characterization of laser-processed carbon nanotubes, characterized in that: The steps include: S1: growing carbon nanotube arrays or network structures on silicon wafers; S2: Laser processing of carbon nanotube arrays or network structures using femtosecond lasers; S3: Use multi-scale characterization technology to obtain surface data of silicon wafers after laser processing; S4: Building a multimodal deep learning model based on silicon wafer surface data; S5: Perform intelligent statistical analysis based on the output results of the multimodal deep learning model and generate wafer-level characterization reports; S6: Optimize laser processing parameters based on wafer-level characterization report.

2. The intelligent statistical method for wafer-level characterization of laser-processed carbon nanotubes according to claim 1, characterized in that: In step S1, the size of the silicon wafer is 4-8 inches; the surface of the silicon wafer is provided with a SiO2 layer with a thickness of 200-300 nm; the thickness of the carbon nanotube array or network structure is 2-5 nm; the density of the carbon nanotube array or network structure is 100-1000 tubes / μm 2 .

3. The intelligent statistical method for wafer-level characterization of laser-processed carbon nanotubes according to claim 1, characterized in that: In step S2, the wavelength of the femtosecond laser is 400-800 nm; the pulse width is less than 500 fs; and the power density is 5×10 12 -8×10 12 W / cm 2 ; The repetition frequency is adjustable from 1Hz to 1MHz; the spot diameter is adjustable from 1-100μm.

4. The intelligent statistical method for wafer-level characterization of laser-processed carbon nanotubes according to claim 1, characterized in that: In step S3, using multi-scale characterization technology to obtain silicon wafer surface data includes: a) Use field emission scanning electron microscopy with a resolution better than 1 nm to obtain large-area morphological information; b) analyzing the atomic structure of a single carbon nanotube using a transmission electron microscope with a resolution better than 0.5 nm; c) evaluating the defect density and lattice changes of carbon nanotubes using confocal Raman spectroscopy with a spatial resolution of less than 1 μm; d) The height and surface roughness of carbon nanotubes were measured using an atomic force microscope with a vertical resolution better than 0.5 nm.

5. The intelligent statistical method for wafer-level characterization of laser-processed carbon nanotubes according to claim 1, characterized in that: In step S4, the multimodal deep learning model has the following functions: a) Identify and classify the structural characteristics of carbon nanotubes, including diameter, length, chirality, and defect types; b) Evaluate the effect of laser processing on the carbon nanotube crystal structure and identify changes in defect density; c) Analyze the uniformity and consistency of carbon nanotube distribution on the surface of the silicon wafer and generate a distribution map with a spatial resolution of 1 μm×1 μm.

6. The intelligent statistical method for wafer-level characterization of laser-processed carbon nanotubes according to claim 1, characterized in that: In step S5, the intelligent statistical analysis step includes: a) Evaluate the growth density and orientation consistency of carbon nanotubes in different regions of the silicon wafer; b) Analyze the effect of laser processing on the structural integrity of carbon nanotubes, including defect type identification and density estimation; c) calculating the length and diameter distribution of the carbon nanotubes after laser processing and generating a histogram; d) Estimate the surface defect density and type of silicon wafers with a spatial resolution of 1mm×1mm.

7. An intelligent statistical system for wafer-level characterization of laser-processed carbon nanotubes for implementing any of the methods described in claims 1-6, characterized in that: include: Carbon nanotube growth module: grows carbon nanotube arrays or network structures on silicon wafers; Femtosecond laser processing module: Use femtosecond laser to process carbon nanotube arrays or network structures; Multi-scale characterization module: uses multi-scale characterization technology to obtain the surface data of silicon wafers after laser processing; Multimodal deep learning module: processes the surface data of silicon wafers after laser processing and outputs the results; Intelligent statistical analysis module: performs intelligent statistical analysis based on the output results of the multimodal deep learning model and generates wafer-level characterization reports; Parameter Optimization Module: Optimize laser processing parameters based on wafer-level characterization reports.

8. The laser processed carbon nanotube wafer level characterization intelligent statistical system according to claim 7, characterized in that: It also includes an in-situ monitoring module that uses a high-speed camera to monitor the structural changes of carbon nanotubes during laser processing in real time.

9. The laser processed carbon nanotube wafer level characterization intelligent statistical system according to claim 7, characterized in that: The invention also includes a database module for storing the nano-nanotube characterization data under different growth conditions and laser processing parameters.

10. The laser processed carbon nanotube wafer level characterization intelligent statistical system according to claim 7, characterized in that: It also includes a human-machine interface that displays the laser processing and characterization process in real time and allows the operator to adjust the laser processing parameters based on the intermediate results.

Images