High-flux photo-thermal synergistic nano welding method for micro-nano structure casting

Abstract

The invention relates to the field of micro-nano structure casting, in particular to a high-flux photo-thermal synergistic nano welding method for micro-nano structure casting. Specifically, an experimental platform integrated with a high-precision optical element and a quick response heating system is constructed. A double-color or multi-color laser synchronous irradiation strategy is adopted, on one hand, a short-wavelength laser is used for exciting a local plasma surface resonance effect, a high temperature gradient is generated around particles, surrounding media are melted, and therefore a fine three-dimensional structure is engraved; on the other hand, long-wavelength laser is used for providing uniform background heating, and cracks caused by overcooling are prevented. In the whole process, a computer-aided design model is combined with a neural network to guide processing path planning, and the final finished product quality is monitored by means of a microscope. The method has the beneficial effects that high-flux welding among high-precision micro-nano points, lines and surfaces can be realized, and the method can be widely applied to the technical fields of sensing, electronics, microreactors and the like.

Description

1. Technical Field

[0001] The present invention relates to the field of micro-nano structure casting, and in particular to a high-throughput photothermal synergistic nano-welding method for micro-nano structure casting. Specifically, a new type of high-precision nano-welding environment is constructed by using long- and short-wave laser resonant plasma combined with a computer-aided design model and a neural network to guide processing path planning. The present invention can achieve high-throughput welding between high-precision micro-nano points, lines, and surfaces, and can be widely used in technical fields such as sensing, electronics, and microreactors. 2. Background Technology

[0002] Micro-nano casting, molding and welding technology are important research directions in the field of modern manufacturing, and have made significant progress in recent years. Micro-nano casting technology can achieve rapid manufacturing of complex structures by precisely controlling the molding process of materials at the micron to nanometer scale. For example, the precision of micro parts manufactured by micro injection molding technology can reach less than 1 micron, which is widely used in medical devices and micro-electromechanical systems (MEMS). In terms of molding technology, nanoimprinting technology has become an important breakthrough direction for chip manufacturing in the post-Moore era due to its high resolution and low cost advantages, and its resolution can reach less than 10 nanometers. In terms of welding technology, nano-welding uses mechanical micro-motion equipment to splice nano-scale components, and combines heating, laser irradiation and other methods to achieve interface bonding. For example, through scanning tunneling microscope probe operation and in-situ annealing methods, defect-free welding of two-dimensional semiconductor SnSe nanosheets is achieved, and the welding accuracy reaches the subatomic level. In addition, the application of laser welding technology in micro-nano manufacturing is also becoming more and more extensive. For example, femtosecond laser direct writing technology can achieve precise processing of three-dimensional micro-nano structures, and its processing accuracy can reach the nanometer level.

[0003] Although micro-nano welding technology has made many advances, its defects in precision, controllability and high throughput limit its further application. First, the welding accuracy of existing technologies is limited by equipment resolution and material properties. For example, although the precision of electron beam lithography technology can reach 10 nanometers, it is easy to introduce material damage during the welding process, resulting in interface defects. Secondly, the controllability of the welding process is poor. For example, traditional nano welding technology mostly relies on manual operation, which makes it difficult to achieve automated assembly of complex structures and has low welding efficiency. In addition, insufficient high-throughput manufacturing capabilities are another major bottleneck of existing technologies. For example, although femtosecond laser direct writing technology can achieve high-precision processing, its processing speed is slow and it is difficult to meet the needs of large-scale production. At the same time, the problem of heat-affected zone (HAZ) in the welding process is also more prominent. In laser welding, the size of the heat-affected zone can reach several microns, resulting in degradation of material properties. These defects indicate that the existing micro-nano welding technology urgently needs to be improved through technological innovation in terms of precision, controllability and high throughput.

[0004] In order to improve the defects of micro-nano welding technology, research teams at home and abroad have made many progresses. Chang Kai's team at the Beijing Institute of Quantum Information Science has achieved defect-free welding of two-dimensional semiconductor SnSe nanosheets through scanning tunneling microscope probe operation and in-situ annealing methods, and the welding accuracy has reached the subatomic level, providing new ideas for high-precision nano welding; Professor Li Xiaochun's team at the University of California, Los Angeles, has significantly increased the depth of the melting zone of laser welding and reduced the heat-affected zone by adding nano-alumina particles to metal nickel, providing technical support for automated welding; Professor Li Yongfeng's team at China University of Petroleum has developed ceramic nanowire welding technology, which has achieved high-precision connection of ceramic materials through chemical reactions, and the mechanical properties of the welded joints are better than those of the original materials; femtosecond laser two-photon direct writing technology has achieved simultaneous processing of thousands of focal points through digital micromirror devices and microlens arrays, significantly improving the processing throughput. But in general, the existing micro-nano welding technology has too high requirements for the hot melting process, but lacks effective control over the power density, and due to the limitation of the processing scale, it is difficult to achieve large-scale high-throughput welding, so its application scope is subject to certain restrictions. III. Summary of the invention

[0005] The present invention proposes a high-throughput photothermal synergistic nano-welding method for micro-nano structure casting, aiming to overcome the limitations of traditional micro-nano welding technology in terms of precision, speed and complex structure manufacturing. First, the present invention constructs an experimental platform that integrates an ultra-high-resolution microscope objective and a fast-response heating system based on silicon carbide ceramic heaters. Short-wavelength lasers excite localized plasma on the surface of metal or semiconductor materials, generating strong local temperature gradients, melting or evaporating the surrounding medium, and engraving fine three-dimensional structures. Long-wavelength lasers provide uniform background heating, maintain an appropriate temperature field, and prevent cracks. According to the formula:

[0006] E=P / A

[0007] Where E is the energy density, P is the laser power, and A is the spot area, which ensures the formation of a local high temperature gradient.

[0008] Secondly, the present invention introduces a specially designed micro-nano fixture, which adopts electromagnetic adsorption fixation to ensure that the nanostructure to be processed is stable and immobile. The fixture is made of a composite material of polyimide and polydimethylsiloxane, and has excellent insulation, flexibility and creep resistance; the surface is coated with MXene nanosheets, which enhances the electromagnetic adsorption force and reduces the contact resistance. The electromagnetic adsorption system adopts a high-precision electromagnetic coil and controls the current through PWM to achieve precise fixation of the nanostructure. The surface of the fixture is treated with argon plasma to introduce hydroxyl and carboxyl functional groups to enhance the interfacial bonding force. The electromagnetic adsorption force F of the fixture can be calculated by the following formula:

[0009] F=μ0·N2·I 2 / (2·g 2 )

[0010] Where μ0 is the vacuum permeability, N is the number of coil turns, I is the current, and g is the gap distance.

[0011] Finally, the present invention uses deep learning computer-aided design software to generate a high-precision three-dimensional model and converts it into numerical control instructions to guide the laser processing path. The path planning algorithm takes into account the thermal conductivity, thermal expansion coefficient and laser absorptivity of the material and optimizes the processing path. During the processing, an infrared thermal imager and a fiber Bragg grating sensor are used to monitor the temperature distribution in real time, and the laser power and scanning speed are dynamically adjusted through the PID control algorithm. The PID control algorithm follows the following equation:

[0012] u(t)=K p e(t)+K i ∫0 t e(τ)dτ+K d de(t) / dt

[0013] Where u(t) is the control signal, e(t) is the error signal, K p is the proportionality coefficient, Ki is the integration time constant, K d is the differential time constant. After processing, the finished product is imaged at high resolution using a scanning electron microscope to assess structural integrity and dimensional accuracy. Convolutional neural networks automatically identify and classify different defect types, providing a more accurate quality assessment, further optimizing processing parameters, and improving yield.

[0014] This method not only significantly improves the speed and accuracy of nano-welding, but also reduces the risk of thermal damage. It is suitable for the efficient manufacturing of complex three-dimensional structures and has great potential in the field of micro-nano manufacturing.

[0015] In order to achieve the above-mentioned purpose, the technical solution of the present invention is a high-throughput photothermal synergistic nano-welding method for micro-nano structure casting, characterized in that the method specifically comprises the following steps:

[0016] Step 1: First, build an experimental platform that integrates high-precision optical components and a fast-response heating system. The optical component uses a microscope objective with ultra-high resolution (numerical aperture NA ≥ 0.95) that can focus the laser to a sub-micron spot. The fast-response heating system uses a silicon carbide-based ceramic heater with a response time of ≤ 1ms, a temperature range of room temperature to 1200°C, and a heating rate of > 100°C / s. The entire platform is installed on a stable vibration-damping table to ensure mechanical stability during processing.

[0017] Step 2, design a special micro-nano fixture, and use electromagnetic adsorption to fix the fixture to ensure that the nanostructure to be processed remains stable during the processing; the micro-nano fixture uses a composite material of polyimide and polydimethylsiloxane, in which the mass ratio of polyimide to polydimethylsiloxane is 7:3, and a layered skeleton structure is prepared by a two-way freezing and nitrogen annealing process to ensure that the fixture has excellent insulation (dielectric constant ≤3.5), flexibility (elongation at break ≥150%) and creep resistance (load retention rate ≥95%) 610; MXene nanosheets (thickness ≤10nm, conductivity ≥5000S / cm) are coated on the surface of the fixture, and are evenly distributed through an ultrasonic-assisted impregnation process to enhance the electromagnetic adsorption force (adsorption strength ≥0.5N / cm 2 ) and reduce contact resistance (≤0.1Ω)6; the electromagnetic adsorption system adopts a high-precision electromagnetic coil (coil diameter ≤1mm, number of turns ≥100), which is composed of boron iron niobium permanent magnets (magnetic energy product ≥45MGOe) and copper wires (diameter ≤0.1mm), and controls the current (0.1A to 1A, frequency 1kHz to 10kHz) through pulse width modulation (PWM) to achieve precise fixation of nanostructures (displacement error ≤10nm); the system integrates temperature sensors (response time ≤1ms, accuracy ±0.1℃) and vibration sensors The sensor (sensitivity ≥ 0.01g) monitors the fixture state in real time and dynamically adjusts the adsorption force 610; the fixture surface is treated with argon plasma (power 100W, treatment time 10min) to introduce hydroxyl (-OH) and carboxyl (-COOH) functional groups to enhance the interface bonding force with the nanostructure; at the same time, the pearl layer structure design is adopted to form an alternating multilayer structure (layer thickness ≤ 100nm) through a two-way freezing process to further improve the impact resistance (impact force attenuation ≥ 85%) and electromagnetic shielding effectiveness (≥ 50dB);

[0018] Step 3, the laser light source uses a dual-color or multi-color laser, the short-wavelength laser has a wavelength of 405nm, a power range of 10mW to 1W, and a frequency of 1kHz to 1MHz, which is used to excite the localized plasma surface resonance effect; the long-wavelength laser has a wavelength of 1064nm, a power range of 50mW to 5W, and a frequency of 1kHz to 1MHz, which provides uniform background heating; the two lasers are transmitted through the same optical fiber to ensure synchronous irradiation;

[0019] Step 4: The short-wave laser is focused to the target area through a high-precision microscope objective lens, and uses extremely high energy density to excite localized plasma on the surface of metal or semiconductor materials, generating a strong local temperature gradient. This high temperature gradient can melt or evaporate the surrounding medium, thereby carving out a fine three-dimensional structure. The pulse width of the short-wave laser is controlled between 1ns and 10ps to achieve precise energy deposition.

[0020] Step 5: The local plasma substrate is a boron nitride substrate modified with platinum nanoparticles. The size of the platinum nanoparticles ranges from 20 to 80 nm, and the distribution density is 5×10 7 Up to 8×10 22 Pieces / cm 2 , enhancing the plasma resonance effect at low temperatures, and being able to produce significant localized plasma resonance effects at specific wavelengths;

[0021] Step 6: Long-wave laser is incident through the ZnSe infrared transparent window with a thickness of 1 to 3 mm to ensure that the long-wave laser penetrates effectively and is evenly distributed in the processing area, providing a stable background heating environment; the power density of the long-wave laser is controlled at 10 to 100 W / cm 2 , in order to maintain the appropriate temperature field and prevent cracks caused by excessive cooling;

[0022] Step 7, using computer-aided design software based on deep learning, integrating convolutional neural networks and generative adversarial networks, generating high-precision three-dimensional models (resolution ≤ 1nm), and converting them into numerical control instructions to guide the laser processing path; the path planning algorithm is based on the improved A* algorithm, taking into account the thermal conductivity of the material (≥100W / m·K), thermal expansion coefficient (≤20ppm / ℃) and laser absorptivity (≥90%), optimizing the processing path (path error ≤10nm) to ensure the optimal execution of each processing step; during the processing, an infrared thermal imager (resolution ≤1μm, temperature measurement range -50℃ to 1500℃, accuracy ±0.5℃) and a fiber Bragg grating sensor (response time ≤1ms, accuracy ±0.1℃) are used to monitor the temperature distribution of the processing area in real time; the laser power (adjustment accuracy ≤1mW) and scanning speed (adjustment accuracy ≤0.1mm / s) are dynamically adjusted through a PID control algorithm (proportional coefficient Kp=0.8, integral time Ti=0.1s, differential time Td=0.05s);

[0023] Step 8. After processing is completed, use a scanning electron microscope to perform high-resolution imaging of the finished product to evaluate structural integrity and dimensional accuracy. The operating voltage of the scanning electron microscope is set at 5 to 30 kV, and the magnification range is 100 times to 100,000 times to ensure that nano-level details can be observed. Based on the analysis results of the scanning electron microscope, a convolutional neural network is used for image analysis. Through learning and training of a large number of samples, different defect types can be automatically identified and classified, providing more accurate quality assessment, further optimizing processing parameters, and improving the yield rate.

[0024] The present invention is beneficial in that:

[0025] 1) Sub-micron spot focusing is achieved, which significantly improves processing accuracy and complex structure manufacturing capabilities and reduces the risk of thermal damage.

[0026] 2) The synchronous irradiation of long-wave and short-wave lasers ensures the accuracy and consistency of energy deposition, and improves the speed and efficiency of nano-welding.

[0027] 3) Introducing deep learning path planning and real-time temperature monitoring to optimize processing parameters, improve yield, and enhance the reliability and quality control of the manufacturing process. IV. Description of the drawings

[0028] In order to more clearly illustrate the specific embodiments of the present invention, the drawings used in the description of the specific embodiments are briefly described below.

[0029] Figure 1 This is a scanning electron microscope image obtained by welding silver nanowires with an aspect ratio of 25:1 using the present invention.

[0030] Figure 2 The scanning electron microscope image is obtained by welding bismuth nanowires with an aspect ratio of . V. Specific implementation methods

[0031] The present invention is described in detail below with reference to the accompanying drawings and embodiments:

[0032] Embodiment 1:

[0033] A high-throughput photothermal synergistic nano-welding method for micro-nano structure casting, characterized in that the method specifically comprises the following steps:

[0034] Step 1: First, build an experimental platform that integrates high-precision optical components and a fast-response heating system. The optical component uses a microscope objective with ultra-high resolution (numerical aperture NA ≥ 0.95) that can focus the laser to a sub-micron spot. The fast-response heating system uses a silicon carbide-based ceramic heater with a response time of ≤ 1ms, a temperature range of room temperature to 1200°C, and a heating rate of > 100°C / s. The entire platform is installed on a stable vibration-damping table to ensure mechanical stability during processing.

[0035] Step 2, design a special micro-nano fixture, and use electromagnetic adsorption to fix the fixture to ensure that the nanostructure to be processed remains stable during the processing; the micro-nano fixture uses a composite material of polyimide and polydimethylsiloxane, in which the mass ratio of polyimide to polydimethylsiloxane is 7:3, and a layered skeleton structure is prepared by a two-way freezing and nitrogen annealing process to ensure that the fixture has excellent insulation (dielectric constant ≤3.5), flexibility (elongation at break ≥150%) and creep resistance (load retention rate ≥95%) 610; MXene nanosheets (thickness ≤10nm, conductivity ≥5000S / cm) are coated on the surface of the fixture, and are evenly distributed through an ultrasonic-assisted impregnation process to enhance the electromagnetic adsorption force (adsorption strength ≥0.5N / cm2 ) and reduce contact resistance (≤0.1Ω)6; the electromagnetic adsorption system adopts a high-precision electromagnetic coil (coil diameter ≤1mm, number of turns ≥100), which is composed of boron iron niobium permanent magnets (magnetic energy product ≥45MGOe) and copper wires (diameter ≤0.1mm), and controls the current (0.1A to 1A, frequency 1kHz to 10kHz) through pulse width modulation (PWM) to achieve precise fixation of nanostructures (displacement error ≤10nm); the system integrates temperature sensors (response time ≤1ms, accuracy ±0.1℃) and vibration sensors The sensor (sensitivity ≥ 0.01g) monitors the fixture state in real time and dynamically adjusts the adsorption force 610; the fixture surface is treated with argon plasma (power 100W, treatment time 10min) to introduce hydroxyl (-OH) and carboxyl (-COOH) functional groups to enhance the interface bonding force with the nanostructure; at the same time, the pearl layer structure design is adopted to form an alternating multilayer structure (layer thickness ≤ 100nm) through a two-way freezing process to further improve the impact resistance (impact force attenuation ≥ 85%) and electromagnetic shielding effectiveness (≥ 50dB);

[0036] Step 3, the laser light source uses a dual-color or multi-color laser, the short-wavelength laser has a wavelength of 405nm, a power range of 10mW to 1W, and a frequency of 1kHz to 1MHz, which is used to excite the localized plasma surface resonance effect; the long-wavelength laser has a wavelength of 1064nm, a power range of 50mW to 5W, and a frequency of 1kHz to 1MHz, which provides uniform background heating; the two lasers are transmitted through the same optical fiber to ensure synchronous irradiation;

[0037] Step 4: The short-wave laser is focused to the target area through a high-precision microscope objective lens, and uses extremely high energy density to excite localized plasma on the surface of metal or semiconductor materials, generating a strong local temperature gradient. This high temperature gradient can melt or evaporate the surrounding medium, thereby carving out a fine three-dimensional structure. The pulse width of the short-wave laser is controlled between 1ns and 10ps to achieve precise energy deposition.

[0038] Step 5: The local plasma substrate is a boron nitride substrate modified with platinum nanoparticles. The size of the platinum nanoparticles ranges from 20 to 80 nm, and the distribution density is 5×10 7 Up to 8×10 22 Pieces / cm 2 , enhancing the plasma resonance effect at low temperatures, and being able to produce significant localized plasma resonance effects at specific wavelengths;

[0039] Step 6: Long-wave laser is incident through the ZnSe infrared transparent window with a thickness of 1 to 3 mm to ensure that the long-wave laser penetrates effectively and is evenly distributed in the processing area, providing a stable background heating environment; the power density of the long-wave laser is controlled at 10 to 100 W / cm 2, in order to maintain the appropriate temperature field and prevent cracks caused by excessive cooling;

[0040] Step 7, using computer-aided design software based on deep learning, integrating convolutional neural networks and generative adversarial networks, generating high-precision three-dimensional models (resolution ≤ 1nm), and converting them into numerical control instructions to guide the laser processing path; the path planning algorithm is based on the improved A* algorithm, taking into account the thermal conductivity of the material (≥100W / m·K), thermal expansion coefficient (≤20ppm / ℃) and laser absorptivity (≥90%), optimizing the processing path (path error ≤10nm) to ensure the optimal execution of each processing step; during the processing, an infrared thermal imager (resolution ≤1μm, temperature measurement range -50℃ to 1500℃, accuracy ±0.5℃) and a fiber Bragg grating sensor (response time ≤1ms, accuracy ±0.1℃) are used to monitor the temperature distribution of the processing area in real time; the laser power (adjustment accuracy ≤1mW) and scanning speed (adjustment accuracy ≤0.1mm / s) are dynamically adjusted through a PID control algorithm (proportional coefficient Kp=0.8, integral time Ti=0.1s, differential time Td=0.05s);

[0041] Step 8. After processing is completed, use a scanning electron microscope to perform high-resolution imaging of the finished product to evaluate structural integrity and dimensional accuracy. The operating voltage of the scanning electron microscope is set at 5 to 30 kV, and the magnification range is 100 times to 100,000 times to ensure that nano-level details can be observed. Based on the analysis results of the scanning electron microscope, a convolutional neural network is used for image analysis. Through learning and training of a large number of samples, different defect types can be automatically identified and classified, providing more accurate quality assessment, further optimizing processing parameters, and improving the yield rate.

[0042] Embodiment 2:

[0043] The silver nanowires with an aspect ratio of 25:1 were welded using the method described in Example 1. First, an experimental platform integrating an ultra-high resolution microscope objective (numerical aperture NA = 0.98) and a fast response heating system was built. The silver nanowires (diameter 40nm, length 1μm) were fixed using an electromagnetic adsorption micro-nano fixture, and the fixture surface was coated with MXene nanosheets to enhance the electromagnetic adsorption force. The dual-color laser parameters were set as: short-wavelength laser 405nm, power 500mW, frequency 1MHz; long-wavelength laser 1064nm, power 2W, frequency 1MHz. The two lasers irradiated the target area synchronously through the same optical fiber, and the short-wave laser pulse width was set to 5ns.

[0044] During the welding process, the short-wave laser is focused on the intersection of the silver nanowires, exciting the local plasma to generate a local high temperature gradient, melting the surrounding medium to form a welded joint. The long-wave laser provides background heating to prevent cracks. The laser power and scanning speed are adjusted in real time by the PID control algorithm to ensure uniform temperature distribution. After welding, the finished product is imaged with high resolution using a scanning electron microscope to evaluate the structural integrity and dimensional accuracy. The mechanical strength of the solder joint structure on the welding material was characterized by atomic force microscopy (AFM), with an elastic modulus of 120 GPa, a hardness of 4.5 GPa, an elongation at break of 12%, a tensile strength of 2.5 GPa, and a conductivity of 6.3×10 7 S / m, good thermal stability, no obvious deformation or fracture at 800℃.

[0045] Compared with traditional laser welding technology, the method of the present invention significantly improves welding accuracy and efficiency and reduces the risk of thermal damage. Comparison of specific performance indicators shows that the present invention is superior to traditional methods in multiple key performances: elastic modulus increased by 30%, hardness increased by 40.6%, toughness increased by 50%, and tensile strength increased by 38.9%. Intelligent path planning and real-time temperature monitoring optimize processing parameters, improve the mechanical properties and electrical conductivity of welded joints, and enhance the reliability and quality control of the manufacturing process. The present invention demonstrates its superiority in efficient and high-precision manufacturing of complex three-dimensional structures.

[0046] The scanning electron microscope image of the welded silver nanowires obtained in this example is shown in the attached figure. Figure 1 shown.

[0047] Specific welding structure performance parameters are shown in the following table:

[0048]

[0049] Embodiment 3:

[0050] The bismuth nanowires were welded using the method described in Example 1, and the two-color laser parameters were set as follows: short wavelength laser 405nm, power 50mW, frequency 1MHz; long wavelength laser 1064nm, power 0.5W, frequency 1MHz. The laser power and scanning speed were adjusted in real time by the PID control algorithm to ensure uniform temperature distribution. Other welding conditions were the same as in Example 2 and will not be repeated here.

[0051] The mechanical strength of the solder joint structure on the welding material was characterized by atomic force microscopy (AFM), among which the elastic modulus was 18 GPa, the hardness reached 2.1 GPa, the elongation at break was 45%, the tensile strength was 0.6 GPa, and the conductivity was 8.8×10 7 S / m, good thermal stability.

[0052] Compared with traditional laser welding technology, the method of the present invention significantly improves welding accuracy and efficiency and reduces the risk of thermal damage.

[0053] The scanning electron microscope image of the welded bismuth nanowires obtained in this embodiment is shown in the attached figure. Figure 2 shown.

[0054] Specific welding structure performance parameters are shown in the following table:

[0055]

[0056] The specific implementation methods described above are only used to specifically illustrate the spirit of the present invention, and the protection scope of the present invention is not limited thereto. For those skilled in the art, other implementation methods can certainly be easily made by changing, replacing or modifying the technical contents disclosed in this specification, and these other implementation methods should all be covered within the protection scope of the present invention.

Claims

1. A high-throughput photothermal nanowelding method for micro-nanostructure casting, characterized in that The method specifically comprises the following steps: Step 1: First, build an experimental platform that integrates high-precision optical components and a fast-response heating system. The optical component uses a microscope objective with ultra-high resolution (numerical aperture NA ≥ 0.95) that can focus the laser to a sub-micron spot. The fast-response heating system uses a silicon carbide-based ceramic heater with a response time of ≤ 1ms, a temperature range of room temperature to 1200°C, and a heating rate of > 100°C / s. The entire platform is installed on a stable vibration-damping table to ensure mechanical stability during processing. Step 2, design a special micro-nano fixture, and use electromagnetic adsorption to fix the fixture to ensure that the nanostructure to be processed remains stable during the processing; the micro-nano fixture uses a composite material of polyimide and polydimethylsiloxane, in which the mass ratio of polyimide to polydimethylsiloxane is 7:3, and a layered skeleton structure is prepared by a two-way freezing and nitrogen annealing process to ensure that the fixture has excellent insulation (dielectric constant ≤3.5), flexibility (elongation at break ≥150%) and creep resistance (load retention rate ≥95%) 610; MXene nanosheets (thickness ≤10nm, conductivity ≥5000S / cm) are coated on the surface of the fixture, and are evenly distributed through an ultrasonic-assisted impregnation process to enhance the electromagnetic adsorption force (adsorption strength ≥0.5N / cm 2 ) and reduce contact resistance (≤0.1Ω)6; the electromagnetic adsorption system adopts a high-precision electromagnetic coil (coil diameter ≤1mm, number of turns ≥100), which is composed of boron iron niobium permanent magnets (magnetic energy product ≥45MGOe) and copper wires (diameter ≤0.1mm), and controls the current (0.1A to 1A, frequency 1kHz to 10kHz) through pulse width modulation (PWM) to achieve precise fixation of nanostructures (displacement error ≤10nm); the system integrates temperature sensors (response time ≤1ms, accuracy ±0.1℃) and vibration sensors The sensor (sensitivity ≥ 0.01g) monitors the fixture state in real time and dynamically adjusts the adsorption force 610; the fixture surface is treated with argon plasma (power 100W, treatment time 10min) to introduce hydroxyl (-OH) and carboxyl (-COOH) functional groups to enhance the interface bonding force with the nanostructure; at the same time, the pearl layer structure design is adopted to form an alternating multilayer structure (layer thickness ≤ 100nm) through a two-way freezing process to further improve the impact resistance (impact force attenuation ≥ 85%) and electromagnetic shielding effectiveness (≥ 50dB); Step 3, the laser light source uses a dual-color or multi-color laser, the short-wavelength laser has a wavelength of 405nm, a power range of 10mW to 1W, and a frequency of 1kHz to 1MHz, which is used to excite the localized plasma surface resonance effect; the long-wavelength laser has a wavelength of 1064nm, a power range of 50mW to 5W, and a frequency of 1kHz to 1MHz, which provides uniform background heating; the two lasers are transmitted through the same optical fiber to ensure synchronous irradiation; Step 4: The short-wave laser is focused to the target area through a high-precision microscope objective lens, and uses extremely high energy density to excite localized plasma on the surface of metal or semiconductor materials, generating a strong local temperature gradient. This high temperature gradient can melt or evaporate the surrounding medium, thereby carving out a fine three-dimensional structure. The pulse width of the short-wave laser is controlled between 1ns and 10ps to achieve precise energy deposition. Step 5: The local plasma substrate is a boron nitride substrate modified with platinum nanoparticles. The size of the platinum nanoparticles ranges from 20 to 80 nm, and the distribution density is 5×10 7 Up to 8×10 22 Pieces / cm 2 , enhancing the plasma resonance effect at low temperatures, and being able to produce significant localized plasma resonance effects at specific wavelengths; Step 6: Long-wave laser is incident through the ZnSe infrared transparent window with a thickness of 1 to 3 mm to ensure that the long-wave laser penetrates effectively and is evenly distributed in the processing area, providing a stable background heating environment; the power density of the long-wave laser is controlled at 10 to 100 W / cm 2 , in order to maintain the appropriate temperature field and prevent cracks caused by excessive cooling; Step 7, using computer-aided design software based on deep learning, integrating convolutional neural networks and generative adversarial networks, generating high-precision three-dimensional models (resolution ≤ 1nm), and converting them into numerical control instructions to guide the laser processing path; the path planning algorithm is based on the improved A* algorithm, taking into account the thermal conductivity of the material (≥100W / m·K), thermal expansion coefficient (≤20ppm / ℃) and laser absorptivity (≥90%), optimizing the processing path (path error ≤10nm) to ensure the optimal execution of each processing step; during the processing, an infrared thermal imager (resolution ≤1μm, temperature measurement range -50℃ to 1500℃, accuracy ±0.5℃) and a fiber Bragg grating sensor (response time ≤1ms, accuracy ±0.1℃) are used to monitor the temperature distribution of the processing area in real time; the laser power (adjustment accuracy ≤1mW) and scanning speed (adjustment accuracy ≤0.1mm / s) are dynamically adjusted through a PID control algorithm (proportional coefficient Kp=0.8, integral time Ti=0.1s, differential time Td=0.05s); Step 8. After processing is completed, use a scanning electron microscope to perform high-resolution imaging of the finished product to evaluate structural integrity and dimensional accuracy. The operating voltage of the scanning electron microscope is set at 5 to 30 kV, and the magnification range is 100 times to 100,000 times to ensure that nano-level details can be observed. Based on the analysis results of the scanning electron microscope, a convolutional neural network is used for image analysis. Through learning and training of a large number of samples, different defect types can be automatically identified and classified, providing more accurate quality assessment, further optimizing processing parameters, and improving the yield rate.

Images