A method and apparatus for stripping an ingot using dual laser beams

Abstract

This application relates to the field of laser ingot lift-off, specifically providing a method and apparatus for dual-laser beam ingot lift-off. The method includes the following steps: Step 1, fixing the ingot on a moving platform; Step 2, turning on the laser, with the laser beam irradiating the interior of the ingot; Step 3, the moving platform moving the ingot so that the laser-modified layer covers the entire ingot; Step 4, turning off the laser and removing the laser-modified ingot from the moving platform; The laser beams include a first laser beam and a second laser beam, with a fixed distance between them. The dual beams scan synchronously at a fixed interval, reducing laser energy density and processing time, weakening the objective lens thermal effect, and ensuring uniform thickness of the modified layer; this facilitates crack propagation, reduces edge chipping, and improves wafer yield; simultaneously, it reduces ingot material waste and lowers the unit cost of high-value wafers.

Description

Technical Field

[0001] This application relates to the field of laser stripping of crystal ingots, and more specifically, to a method and apparatus for stripping crystal ingots using dual laser beams. Background Technology

[0002] The basic process of laser ablation involves first inducing a continuous modified layer within the ingot using a laser. Then, controlled cleavage along the modified layer is achieved through ultrasonic vibration, mechanical ablation, thermal stress separation, and cold stress separation, resulting in separated wafers. If the modified layer thickness is uneven, cracks tend to propagate in thicker areas and deflect in thinner areas during separation, potentially leading to skewed wafers or edge breakage and reduced wafer yield. Conversely, a uniform modified layer allows for the separation of more wafers from an ingot of the same thickness, improving material utilization, especially for high-value crystals like SiC, thus reducing the material cost per wafer. Therefore, forming a uniform modified layer within the ingot is crucial.

[0003] Currently used laser ablation methods typically employ a single-beam laser, focusing it at a predetermined depth within the crystal using a high numerical aperture objective lens. Utilizing nonlinear effects such as multiphoton absorption and avalanche ionization, microcracks, voids, or stress concentration regions are formed at a fixed height, thus creating a continuous modified layer. However, the objective lens used for the emitted laser experiences thermal effects under prolonged high-power transmission. Absorption of laser energy in the objective lens material and coating causes temperature increases, leading to thermal expansion and changes in refractive index, resulting in equivalent focal length drift and altered focusing depth. This causes the thickness of the modified layer to vary with laser scanning, resulting in uneven thickness. Because the energy density of a single beam is more concentrated, the thermal effect of the objective lens is more intense; furthermore, the longer scanning time along the preset path leads to a longer accumulation time of thermal effects, resulting in more pronounced variations in the modified layer thickness and greater waste of crystal material. Existing technologies typically employ a cooling sleeve on the outside of the objective lens to cool it and reduce focal drift caused by thermal effects. However, relying solely on cooling measures often introduces new problems, such as thermal stress caused by temperature gradients, environmental condensation risks, and the contradiction between cooling efficiency and processing cycle time, making it difficult to fundamentally solve the problem of uneven thickness of the modified layer.

[0004] In summary, the existing single-beam ingot stripping method results in uneven thickness of the modified layer, which reduces the yield of separated wafers and thus increases the material cost per wafer. Summary of the Invention

[0005] The purpose of this invention is to address the shortcomings of the prior art by providing a method and apparatus for dual-laser beam stripping of a crystal ingot. A method for dual-laser beam stripping of a crystal ingot includes the following steps: Step 1: Fix the crystal ingot on the moving platform; Step 2: Turn on the laser, and the laser beam will irradiate the inside of the crystal ingot; Step 3: The moving platform moves the crystal ingot so that the laser-modified layer covers the entire crystal ingot; Step 4: Turn off the laser and remove the laser-modified ingot from the moving platform; The laser beam includes a first laser beam and a second laser beam, and the distance between the first laser beam and the second laser beam is fixed.

[0006] In steps 2 and 3 of this application, the first laser beam and the second laser beam synchronously irradiate the ingot at a fixed interval and scan along a preset path with the moving platform. The two laser beams participate in the remodeling process in parallel during the same scan. Compared to the method of forming the remodeling layer point by point with a single beam, the scanning time required to form the remodeling layer is shortened, thereby reducing the continuous heat input and heat accumulation time of the laser on the objective lens, and reducing the slow process effect of thermal drift. Simultaneously, the total energy required to form the remodeling layer is provided by both laser beams. Therefore, under the conditions of multiphoton absorption, avalanche ionization, and other nonlinear threshold triggering conditions, the energy density of a single laser beam is reduced, the heat absorbed by the objective lens and its coating is reduced, the thermal effect intensity of the objective lens is weakened, and thus the change in focusing depth is reduced. This application simultaneously reduces the intensity and duration of the thermal effect, reduces the drift amplitude of the focus in the axial direction, and ensures that the remodeling layer is always formed near the predetermined depth within the scanning path range, resulting in a more uniform remodeling layer thickness.

[0007] Furthermore, the distance between the first and second laser beams is 10-150 μm. The target of this application is SiC. The modified core region of SiC under ultrashort pulse laser irradiation is typically triggered by multiphoton absorption and avalanche ionization, with the spatial scale of the strongly nonlinear absorption region generally ranging from several micrometers to tens of micrometers. Due to SiC's high thermal conductivity and large elastic modulus, thermal diffusion and stress redistribution around the modified core region create a larger stress-affected region, which can extend to the tens of micrometers. This application limits the distance between the two beams to 10-150 μm, allowing the stress-affected regions formed by the two laser beams to be spatially adjacent and partially superimposed. When the beam spacing is on the same order of magnitude as the scale of the SiC stress-affected region, the distance between the two beams is comparable to the scale of the stress-affected region, and the stress fields overlap and continuously expand, thus forming a stress bridge between the two beams. This avoids the appearance of a weak stress zone in the middle, making the modified layer more uniform in axial thickness and more continuous in the transverse direction. Furthermore, the 10-150μm spacing is larger than the scale of the modified core region, preventing excessive energy superposition of the two beams in the core region, which could lead to excessive local damage and sudden thickness increases. This spacing ensures that the core region does not experience excessive superposition and that the stress zone is fully coupled, resulting in a more uniform SiC modified layer thickness.

[0008] Furthermore, in step 3, the line connecting the center of the first laser beam and the center of the second laser beam is perpendicular or parallel to the direction of movement of the ingot.

[0009] During the scanning process, the two laser beams act simultaneously on adjacent spatial locations within the ingot, rather than sequentially at the same location. This results in the formation of laterally parallel energy deposition regions, leading to the continuous generation of a stable-width modified band along the scanning direction within the ingot. This ensures the modified layer remains uniform throughout the scanning path. In contrast, the case where the moving direction is parallel to the line connecting the two laser beams, the latter involves the same spatial location being acted upon by two laser beams at different times. This latter approach is more susceptible to the effects of time intervals, thermal diffusion, and focus drift, leading to fluctuations in localized modification levels and even excessive damage. In the vertical arrangement employed in this invention, the two laser beams act in spatial parallelism. The formation of the modified layer relies primarily on spatial superposition rather than temporal superposition, effectively reducing the impact of heat accumulation and temporal instability on the modification process. This ensures a continuous spatial distribution of stress concentration areas, preventing discontinuities or uneven intensity in the modified layer along the scanning direction. The thickness uniformity of the modified layer is also improved.

[0010] Simultaneously, it also helps improve the straightness of the laser-induced modification path and the consistency of repeated processing. When the ingot moves along a fixed direction, the two laser beams form parallel action positions in space. In actual processes, when multiple laser actions are required on the same path to enhance the strength of the modified layer, vertical movement ensures that the modified structure is formed on the same spatial trajectory each time, avoiding lateral deviation of the modification path due to thermal diffusion, time delay, or focus drift. If the movement direction is parallel to the line connecting the two laser beams, it is easier to introduce positional deviations and path bends, thereby reducing the straightness and consistency of the modified layer. Therefore, the movement method of this invention helps to obtain a modified layer with high positional overlap and good straightness under multiple scanning conditions.

[0011] Furthermore, the wavelengths, pulse widths, and intensities of the first and second laser beams may be the same or different.

[0012] Furthermore, in step 3, the ingot moves in a straight line at a speed of 100 mm / s to 1200 mm / s.

[0013] This application also proposes a device for stripping crystal ingots using dual laser beams. The device includes a moving stage, a laser module, and a fixed frame. The laser module is fixed on the fixed frame and positioned above the moving stage. In use, the crystal ingot is fixed on the moving stage with the side to be stripped facing the laser module. The laser module emits a first laser beam and a second laser beam, with a fixed distance between the first laser beam and the second laser beam.

[0014] Furthermore, the laser module includes a laser source, a beam expander, a beam modulation and separation module, a beam splitter, and multiple lenses and mirrors. The laser emitted from the laser source passes sequentially through the beam expander, the beam modulation and separation module, and the beam splitter. The transmitted beam from the beam splitter is collimated by the lens and mirror before being emitted as the first laser beam. The reflected beam from the beam splitter is collimated by the lens and mirror before being emitted as the second laser beam.

[0015] Furthermore, the beam modulation and separation module includes a spatial light modulator and an asymmetric triangular reflector.

[0016] Furthermore, the laser module includes a laser source, a beam expander, a beam modulation and splitting module, a 4f relay system, a bifocal objective lens, and multiple mirrors. The laser emitted from the laser source passes sequentially through the beam expander, beam modulation and splitting module, 4f relay system, and bifocal objective lens. The output light from the bifocal objective lens consists of a first laser beam and a second laser beam. This application employs a bifocal objective lens to simultaneously output the first and second laser beams. Through internal component phase design, the bifocal objective lens enables the incident laser to form two laterally separated and closely spaced focused beams on the same focal plane. This allows the two laser beams to simultaneously act on adjacent spatial regions within the ingot, thereby forming a close and synergistic modified structure within the ingot. This facilitates the rapid spatial penetration and continuity of the modified layer. The bifocal objective lens disperses the energy originally concentrated at a single focal point to two focal points, allowing different regions within the objective lens to bear energy transmission separately. This reduces the thermal load on local optical elements, thereby mitigating focal length drift caused by thermally induced refractive index changes and thermal expansion. Furthermore, since bifocal points typically correspond to different optical path regions within the objective lens aperture, energy propagates closer to the objective lens edge, which facilitates heat conduction and dissipation to the surrounding structure, further suppressing the heat accumulation effect. This provides favorable conditions for obtaining high-quality, repeatable wafer dicing results.

[0017] Furthermore, a bifocal objective lens consists of a lens and diffractive optical elements.

[0018] Compared with the prior art, the beneficial effects of the present invention are as follows: The first and second laser beams maintain a fixed distance and synchronously irradiate the ingot, scanning together along a preset path with the moving platform. This allows both laser beams to participate in the formation of the modified layer in parallel during the same scanning process. On one hand, compared to the point-by-point peeling method using a single laser, the parallel action of the two beams can complete the construction of the modified layer in a shorter scanning time, thereby reducing the continuous laser action time on the objective lens, decreasing the accumulation of heat in the objective lens, and suppressing the slow focus drift caused by heat accumulation. On the other hand, the total energy required for the formation of the modified layer is provided by the two laser beams, correspondingly reducing the energy density of a single laser beam, thus reducing the heat absorbed by the objective lens material and weakening the intensity of the objective lens thermal effect. By simultaneously reducing the action time and intensity of the thermal effect, this application effectively reduces the focus drift amplitude in the axial direction, keeping the laser focusing depth relatively stable throughout the scanning process, thereby ensuring a uniform thickness of the modified layer formed inside the ingot.

[0019] In terms of improving wafer yield: the distribution of microcracks and stress concentration areas in the axial direction of the uniform thickness of the modified layer is consistent, and the stress field is continuous and balanced along the plane of the modified layer. When ultrasonic vibration, mechanical peeling or thermal stress separation is applied, cracks can be synchronously initiated and stably propagated along the predetermined plane throughout the entire modified layer. It is not easy for stress mismatch to occur in local thick or thin areas, thereby effectively avoiding defects such as crack deflection, crack tilting or edge chipping, and improving the yield of separated wafers.

[0020] Regarding reducing wafer material costs: A uniformly thick modified layer allows for precise control of the spacing between adjacent splitting interfaces. More effective modified layers can be arranged within an ingot of the same thickness, resulting in a greater number of wafers and improved ingot material utilization. For high-value crystal materials like SiC, which are costly and have long preparation cycles, this directly reduces material consumption and manufacturing costs per wafer, significantly lowering overall costs. Attached Figure Description

[0021] Figure 1 A schematic diagram of a method for removing a crystal ingot using a dual laser beam provided by the present invention; Figure 2 This is a schematic diagram of the two laser beams and the scanning direction in a dual-laser beam stripping method for a crystal ingot provided by the present invention (the arrows in the diagram indicate the direction of movement of the crystal ingot). Figure 3 This is a schematic diagram of the optical path of the laser module in a dual-laser beam stripping device for crystal ingots provided in this application.

[0022] Icons: 1-Beam expander; 2-Beam splitter; 3-Lens; 4-Reflector; 5-SLM; 6-Asymmetric triangular reflector. Detailed Implementation

[0023] To make the implementation process of this invention clearer, a detailed description will be provided below in conjunction with the accompanying drawings. Example 1:

[0024] This invention provides a method for separating a crystal ingot using a dual-laser beam, such as... Figure 1 As shown, the method includes the following steps: Step 1: Fix the crystal ingot on the moving platform.

[0025] The ingot to be processed is fixed on the moving platform, which can be achieved by vacuum adsorption or mechanical clamping to prevent movement during scanning. The side of the ingot to be cleaved faces the laser incident direction, and the ingot surface is perpendicular to the laser incident direction. The moving platform can move along the plane perpendicular to the laser under the action of motors, lead screws, etc., and generally has two degrees of freedom in the X and Y dimensions, with at least a linear motion degree of freedom along the scanning direction; furthermore, a Z-axis fine-tuning mechanism can be configured to set the peeling depth, for example, to focus the laser at a predetermined depth position inside the ingot. The platform controller can use closed-loop encoder feedback to ensure the speed stability and path straightness of uniform scanning. The specific structure and encoding of the moving platform are all prior art. It should be noted that the laser module is generally not easy to move; therefore, the laser scanning is achieved by moving the ingot through the moving platform. The crystal to be peeled in this application can be Si, SiC, etc.

[0026] Step 2: Turn on the laser and let the laser beam irradiate the inside of the ingot.

[0027] The laser is activated, allowing the laser beam to perpendicularly irradiate the surface of the crystal ingot and form a modified layer inside the ingot. The laser beam includes a first laser beam and a second laser beam, which do not move relative to each other and maintain a fixed spatial position, both incident perpendicularly to the crystal ingot surface. Further, the distance between the first and second laser beams is set to 10-150 μm (achievable in Embodiment 3 of this application); this allows the two laser beams to form adjacent or partially overlapping nonlinear absorption regions and stress-affected regions inside the crystal ingot, thereby facilitating the formation of a continuously thick, uniform modified layer. The wavelength, pulse width, and intensity of the first and second laser beams are identical to ensure that the conditions for inducing modification inside the crystal ingot are consistent, exhibiting the same energy deposition characteristics and nonlinear action conditions. With the same wavelength, the absorption characteristics and focusing behavior of the two laser beams in the crystal material are consistent, avoiding inconsistencies in modification depth and morphology due to differences in material dispersion or absorption. With identical pulse widths, the triggering timescales of nonlinear processes such as multiphoton absorption and avalanche ionization are consistent, resulting in similar spatial and evolutionary characteristics of microcracks and stress concentration regions induced by the two laser beams. Under conditions of equal intensity, the energy deposition within the ingot is balanced by the two laser beams, avoiding excessively strong or weak local modification, thereby reducing the generation of abrupt intensity changes or discontinuous regions in the modified layer. If the intensities are different, the energy deposition produced by the two laser beams within the ingot will differ. The higher-intensity laser beam will induce excessive damage or non-uniform cracks, while the lower-intensity laser beam will be unable to stably trigger nonlinear absorption, thus forming alternating or discontinuous regions in the modified layer. This unbalanced modified structure weakens the continuity and mechanical consistency of the modified layer, making it unfavorable for forming a modified layer of uniform thickness.

[0028] In this embodiment, the laser can be an ultrashort pulse laser, such as a femtosecond laser or a picosecond laser. Ultrashort pulse lasers have extremely high peak power and extremely short duration, enabling localized modification within the crystal through nonlinear effects such as multiphoton absorption and avalanche ionization, avoiding surface ablation and thermal diffusion caused by linear absorption. The laser wavelength is in the near-infrared band, such as 1030 nm or 1060 nm; the near-infrared band has good material transmittance, facilitating the transmission of laser energy to deeper locations within the ingot to form a modified layer.

[0029] Step 3: The moving platform moves the ingot so that the laser-modified layer covers the entire ingot.

[0030] The moving platform propels the crystal ingot in a linear motion along a predetermined direction, causing the first and second laser beams to sequentially irradiate different positions on the ingot surface, thereby forming a continuous modified layer within the ingot along the scanning path. The predetermined direction is generally parallel or perpendicular to the crystal phase direction; in practice, the direction is marked on the crystal. Typically, the scanning path consists of multiple parallel lines parallel to the predetermined direction, with a large distance between adjacent parallel paths to minimize thermal interference. The ingot's moving speed is set to 100mm / s-1200mm / s to ensure both effective modification and processing efficiency. If the moving speed is too fast, insufficient energy deposition per unit length can lead to discontinuous modified layers; if the moving speed is too slow, heat accumulation can intensify, causing excessive damage and uneven modification, thus reducing the quality of the cleavage.

[0031] Furthermore, such as Figure 2 As shown, the line connecting the centers of the first and second laser beams is perpendicular to the direction of ingot movement, ensuring that the two laser beams act simultaneously on adjacent spatial locations within the ingot during scanning, rather than acting sequentially at the same location. This results in the formation of laterally parallel energy deposition regions within the ingot, continuously generating a stable-width modified band along the ingot's movement direction. This is beneficial for maintaining a uniform and continuous structural characteristic of the modified layer throughout the entire scanning path. In practical processes, the same path can be scanned once or multiple times according to the required modified layer strength. Since there is no relative movement between the two laser beams, and the connecting line is perpendicular to the movement direction, it is easier to ensure overlap in the spatial trajectory during multiple scans, improving the straightness of the modified path and thus providing favorable conditions for stable crack propagation along the modified layer during subsequent cleavage.

[0032] Step 4: Turn off the laser and remove the laser-modified ingot from the mobile platform.

[0033] After completing the predetermined scanning path, the laser is turned off, the platform movement is stopped, and the stripped ingot is removed from the moving platform. At this point, a continuous modified layer has formed at a predetermined depth inside the ingot, allowing it to proceed to the cleaving process. Methods such as ultrasonic vibration, mechanical peeling, or thermal stress separation can be used to induce controlled cleavage along the modified layer, resulting in separated wafers. To ensure cleaving quality, visual inspections can be performed before and after removing the ingot, such as observing the consistency of the scanning trajectory and checking for any abnormal ablation on the surface.

[0034] Furthermore, during the scanning process, the exiting objective lens in the laser source can be thermally homogenized simultaneously. This is achieved by setting a sealed chamber (which can be a ring-shaped barrel structure) on the outside of the objective lens and filling the chamber with liquid metal to improve the objective lens's thermal stability. Liquid metal has a thermal conductivity much higher than that of gases or conventional liquids, enabling it to rapidly transfer and diffuse heat when the objective lens is locally heated, thus making the temperature rise of various parts of the objective lens more uniform and reducing the temperature gradient inside the objective lens. The rapid and uniform distribution of heat achieved through liquid metal suppresses non-uniform thermal effects, stabilizing the propagation conditions of the laser within the objective lens and reducing focal length changes and focus drift. Compared to simply relying on cooling methods to lower the absolute temperature of the objective lens, excessive or uneven cooling is more likely to introduce new temperature gradients and thermal stresses within the objective lens, while thermal field homogenization is more beneficial for maintaining the optical performance stability of the objective lens. Liquid metals are materials that are liquid at room temperature, chemically stable, and have excellent thermal conductivity, such as gallium-based alloys and gallium-indium-tin alloys, enabling efficient and uniform thermal management. A drive coil is coaxially wound on the outer side of the sleeve, with an insulating layer between the coil and the outer wall to ensure a tight fit. When alternating current is applied to the coil, eddy currents are induced within the cavity, creating an induced current loop inside the liquid metal. This loop, under the influence of an external magnetic field, generates a force that drives the liquid metal to circulate. Through electromagnetic drive, the liquid metal transitions from a static, conduction-based heat transfer to enhanced convection heat transfer, increasing the rate of heat transfer and homogenization within the cavity, further reducing the temperature gradient, and improving temperature uniformity.

[0035] To further enhance heat equalization, the inner wall of the thermal shaping sleeve remains circular to conform to the outer surface of the exit lens or lens barrel, ensuring good contact heat conduction conditions. The outer wall is elliptical, with its minor axis parallel to the line connecting the focal points of the first and second laser beams. This creates an annular cylindrical sealed chamber between the inner and outer walls with varying radial gaps at different circumferential positions. This narrows the flow channel and increases the flow velocity of the liquid metal in the hot spot region corresponding to the minor axis, resulting in stronger convective heat transfer in areas of concentrated heat and maintaining a lower flow velocity in areas of less heat variation, thus reducing the overall temperature gradient within the exit lens and its barrel. Simultaneously, a drive coil is positioned along the contour of the elliptical outer wall and energized with alternating current, creating a non-axisymmetric alternating magnetic field within the sealed chamber. This provides a stronger electromagnetic driving force to the liquid metal in the minor axis direction, further promoting its directional flow along the bifocal line and achieving directional enhanced heat transfer in the hot spot region of the bifocal objective lens. When using it, first turn on the coil to establish a stable thermal field, and then start the laser processing; after the process is completed, delay the power-on to remove the residual heat, thereby suppressing thermal shock and thermal hysteresis, so that the focal length, spacing and energy distribution of the dual beams remain stable during long-term processing, and improve the thickness uniformity of the modified layer.

[0036] Steps 1-4 of this application need to be performed in a controlled clean environment, such as a cleanroom or a localized ultra-clean working chamber, to reduce the impact of airborne particulate matter, dust, and moisture on the laser processing. During laser ablation, the laser needs to act precisely on the surface and interior of the ingot. Particulate contamination in the environment may adhere to the ingot surface or optical components, causing laser scattering, energy fluctuations, or localized overheating, thereby affecting the uniformity of the modified layer thickness. In addition, a controlled environment helps maintain constant temperature and humidity, reduces the risk of thermal drift and condensation, and improves the operational stability of the optical system and the moving platform. Example 2:

[0037] This application also proposes a dual-laser-beam stripping device for ingots, used to implement the method described in Example 1. The device includes a moving stage, a laser module, and a mounting frame. The mounting frame is used to fix the laser module and can be stably installed on a wall, floor, or optical platform to ensure the positional stability of the laser module during operation. The laser module is fixedly mounted on the mounting frame and positioned above the moving stage, with the laser emission direction facing the ingot on the moving stage. The ingot is fixed to the moving stage by a vacuum suction clamp, mechanical clamp, or other means, with the side of the ingot to be stripped facing the laser module. The moving stage can achieve precise linear movement in at least one direction under the drive of a control system, and can be set to move in two or three dimensions as needed; used for position adjustment, alignment, or multiple scans. The laser module can simultaneously emit a first laser beam and a second laser beam, with a fixed distance between the two beams, and both beams perpendicularly irradiate the surface of the ingot, forming a modified layer at a specific depth inside the ingot. The entire device operates within a cleanroom or ultra-clean chamber to reduce the impact of airborne particulate matter, dust, and humidity variations on the optical system and ingot surface. It also helps maintain a constant ambient temperature and humidity, thereby further improving the repeatability of the laser ablation process and the quality of the cleavage. All optical components in this application are of equal height and coaxial, and it also includes necessary auxiliary devices such as lens mounts, which can be selected and used by those skilled in the art based on common sense.

[0038] like Figure 3 As shown, the laser module includes a laser source, a beam expander 1, a beam modulation and splitting module, a beam splitter 2, and multiple lenses 3 and reflectors 4. The laser emitted from the laser source passes sequentially through the beam expander 1, the beam modulation and splitting module, and the beam splitter 2. The transmitted beam from the beam splitter 2 is collimated by the lenses 3 and reflectors 4 before exiting as the first laser beam. The reflected beam from the beam splitter 2 is collimated by the lenses 3 and reflectors 4 before exiting as the second laser beam. The beam modulation and splitting module includes an SLM5 (spatial light modulator) and an asymmetric triangular reflector 6. The overall optical path of the laser module is used to stably separate and shape a single laser source into two laser beams with consistent parameters and a fixed spatial relationship.

[0039] The laser emitted from the laser source is first expanded and collimated by beam expander 1 to improve beam quality and match the effective aperture of subsequent modulation elements. The expanded laser beam is reflected by asymmetric triangular reflector 6 and incident on the surface of SLM5 at a preset small angle (the angle between the incident and outgoing light is less than 15°). By loading a predetermined phase distribution onto SLM5, the incident laser beam is phase-modulated, resulting in different diffraction orders in the outgoing light. Using a small-angle illumination method facilitates off-axis modulation, spatially separating different diffraction orders and avoiding overlap between the zero-order beam and the target diffracted beam, thereby improving energy utilization efficiency and modulation stability. The laser beam modulated by SLM5 is then incident on asymmetric triangular reflector 6. Utilizing the geometric characteristics of the multi-reflective surface of asymmetric triangular reflector 6, beams from different directions or different diffraction orders are spatially separated and directionally redirected, guiding the selected beam to a predetermined propagation path, thus achieving effective beam separation and shaping. Subsequently, the separated laser beam is incident on beam splitter 2, which further splits the laser beam into a transmitted beam and a reflected beam. The transmitted beam is collimated and oriented by a corresponding lens 3 and a reflector 4 to form the first laser beam. The reflected beam is collimated by another set of lenses 3 and a reflector 4 to form the second laser beam. Lenses 3 for collimation and focusing are respectively set in the paths of the transmitted and reflected beams of beam splitter 2. The focal lengths of the lenses 3 can be the same to ensure that the first and second laser beams have the same focusing conditions. Through the above optical path structure, the first and second laser beams maintain consistency in parameters such as wavelength, pulse width, and intensity, while also possessing stable and controllable spatial spacing and propagation direction. This optical path not only achieves stable output of the dual beams but also avoids the parameter inconsistency problems common in multi-source systems, which is beneficial for forming a uniform and continuous modified layer during laser ablation. Example 3:

[0040] In this embodiment, the laser module includes a laser source, a beam expander 1, a beam modulation and separation module, a 4f relay system, a bifocal objective lens, and multiple reflectors 4. The laser emitted from the laser source passes sequentially through the beam expander 1, the beam modulation and separation module, the 4f relay system, and the bifocal objective lens. The output light from the bifocal objective lens consists of a first laser beam and a second laser beam. That is, the difference between this embodiment and Embodiment 2 is that the laser output from the asymmetric triangular reflector 6 enters the 4f relay system and the bifocal objective lens, and two laser beams are emitted through the bifocal objective lens.

[0041] Specifically, the 4f relay system consists of a first relay lens L1 (focal length f1) and a second relay lens L2 (focal length f2, f2=f1), with the distance between L1 and L2 being f1+f2, making the exit surface of the asymmetric triangular reflector 6 conjugate with the entrance pupil surface of the bifocal objective. The emitted laser first undergoes a Fourier transform through the first relay lens L1, forming a spatial spectrum plane on the back focal plane of L1. Different angular spectral components or different diffraction orders are spatially separated on this plane. Optionally, a spatial filtering structure, such as an aperture or baffle, can be set on this Fourier plane to suppress zero-order light and stray orders, thereby improving the purity of the light field entering the objective. Subsequently, the laser undergoes an inverse Fourier transform through the second relay lens L2, relaying the light field to the entrance pupil of the bifocal objective at a certain magnification, ensuring that the beam diameter, divergence angle, and entrance pupil filling of the beam entering the bifocal objective remain consistent, thus providing stable incident conditions for the bifocal beam.

[0042] The laser beam, relayed via a 4f relay, is incident on the bifocal objective. A bifocal objective is a focusing component capable of simultaneously forming two focusing positions under the same incident light conditions; in this embodiment, the two focal lengths correspond to the same focal length. While maintaining a high numerical aperture, the bifocal objective distributes the energy of the incident light to the two focal positions in a preset ratio (5:5), causing the exit light from the objective to form a first laser beam and a second laser beam. The separation of the two focal points in a bifocal objective is determined by its internal phase design; bifocal objectives typically consist of lenses and diffractive optical elements (DOEs). A specific phase function is superimposed on the entrance pupil surface or equivalent aperture surface of the objective, causing the incident light to be distributed into two wavefronts after passing through the objective, coherently focusing at different positions. That is, through precise modulation of the incident light wavefront, the energy originally concentrated at a single focal point is separated and redistributed in the axial direction, thereby forming two high-intensity focusing regions separated along the optical axis within the crystal.

[0043] In this embodiment, the DOE is used to generate laterally separated laser light. The corresponding phase can be a laterally linear phase gradient or an equivalent grating-type phase distribution, so that the laser light is distributed to different diffraction orders after passing through the DOE. First-order diffracted light propagates at the same amplitude but opposite tilt angles, thus forming two quasi-parallel beams with different incident angles but identical other parameters. The DOE structure can be directly fabricated on the incident light side of the exit objective lens, or it can be placed at the object plane of the 4f relay system or the conjugate plane of the objective lens entrance pupil. Preferably, the laser first incidents and passes through the DOE, and after phase modulation by the DOE, forms a light field containing multiple angular spectral components. It then enters the first lens L1 in the 4f relay system, forming a Fourier plane at its back focal plane, where different diffraction orders are spatially separated. Optionally, a spatial filter structure can be set in this Fourier plane to suppress the zero-order light, retaining only the first-order light. First-order diffracted light. Then, the beam undergoes an inverse Fourier transform via the second lens L2, making... The first-order diffracted light exits as two collimated beams with symmetrical directions and the same divergence angle, and is relayed to the exit lens. The exit lens can be a common high numerical aperture objective lens, mapping the parallel beams with different incident angles to laterally shifted focal points on the same axial focal plane, thus forming two spatially separated but identically focal-length laser beams in a direction perpendicular to the optical axis. To achieve an approximately 5:5 energy distribution, the phase depth, step height, or blaze angle of the DOE need to be symmetrically designed. The first-order diffraction efficiencies are basically the same, thus ensuring that the two laser beams have similar intensity, numerical aperture and focusing characteristics after passing through the 4f system and objective lens.

[0044] Compared to the method in Example 2, which uses a beam splitter to split the beams and then collimate them separately, in this example, the relative position of the two laser beams is uniformly generated and constrained by the same objective lens, resulting in better coaxiality. Furthermore, the distance between the two laser beams can be set from 10 to 150 μm using a bifocal objective lens. Simultaneously, since the two laser beams travel through the same relay and objective lens optical path, beam parameters such as wavefront quality, divergence angle, focal spot size, and depth of focus are more easily kept consistent, which is beneficial for obtaining a more uniform and continuous modified layer during subsequent stripping. In addition, the entrance pupil imaging and spatial filtering capabilities provided by the 4f relay system can reduce the impact of stray light and order interference on the objective lens's thermal load and focal stability, making the bifocal output more stable during long-term scanning and facilitating the formation of a uniformly thick modified layer. Example 4:

[0045] In step 2 of Example 1, the spot shapes of the first and second laser beams are elliptical, and the major axis directions of the spots of the first and second laser beams are perpendicular to each other. Figure 2 The direction of movement shown is parallel, and correspondingly, the direction of the minor axis is parallel to... Figure 2 The direction of movement shown is perpendicular.

[0046] When the long axis of the laser spot is parallel to the direction of movement of the ingot, the effective working length of the laser in the scanning direction increases, transforming the energy deposition per unit scanning length from an instantaneous concentrated input to a path-distributed input. A laser spot with its long axis parallel to the direction of movement allows for more complete overlap and averaging of the same spatial position between adjacent pulses and adjacent scanning positions. Even with slight focus drift caused by the thermal effect of the objective lens, the effective energy input per unit length remains relatively stable, thus reducing the axial size variation of the nonlinear absorption region. Simultaneously, when the long axes of two elliptical laser beams are arranged along the scanning direction, the stress-affected regions outside the nonlinear absorption regions induced by the two lasers within the crystal, caused by thermal diffusion and elastic response, are elongated and brought closer together in the scanning direction. During scanning, the stress-affected regions of the two beams continuously superimpose and transition, resulting in a larger area of ​​the smooth, continuous stress field formed between the two beams. Local areas are less prone to instantaneous stress concentration or rapid attenuation, avoiding over- or under-modification caused by abrupt changes in stress peaks. This results in a more uniform thickness distribution of the modified layer.

[0047] Based on Example 3, a spot-shaping element is introduced to make the first and second laser beams appear as elliptical spots with their major axis along the direction of ingot movement after focusing. Specifically, a DOE (different from the one in Example 3) or an equivalent phase modulation element for spot shaping can be set at the object plane position of the 4f relay system, the Fourier plane, or the plane conjugate with the entrance pupil of the bifocal objective. The introduced DOE superimposes a one-dimensional quadratic phase term or an equivalent cylindrical lens-type phase structure in its phase distribution, so that the incident light has an effective numerical aperture in the direction parallel to the direction of ingot movement that is different from that in the vertical direction, thereby forming an elliptical focal spot with its major axis parallel to the direction of movement after focusing by the bifocal objective. Since the DOE is set on the conjugate plane of the entrance pupil of the bifocal objective, the phase modulation will take effect simultaneously on the two beams split by the bifocal objective, so that the first and second laser beams both obtain elliptical spots with the same shape and orientation, while the distance between the two beams is still determined by the system of Example 3 and maintained in the range of 10-150 μm.

[0048] The above are merely preferred embodiments of the present invention and are not intended to limit the present invention. Various modifications and variations can be made to the present invention by those skilled in the art. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the scope of protection of the present invention.

Claims

1. A method for exfoliating a crystal ingot using a dual laser beam, the method comprising the following steps: Step 1: Fix the crystal ingot on the moving platform; Step 2: Turn on the laser and the laser beam irradiates the inside of the crystal ingot; Step 3: The moving platform moves the crystal ingot so that the laser-modified layer covers the entire crystal ingot; Step 4: Turn off the laser and remove the laser-modified ingot from the moving platform; Its features are: The laser beam includes a first laser beam and a second laser beam, and the distance between the first laser beam and the second laser beam is fixed.

2. The method for stripping a crystal ingot using dual laser beams according to claim 1, characterized in that: The distance between the first laser beam and the second laser beam is 10-150 μm.

3. The method for removing a crystal ingot using a dual laser beam according to claim 2, characterized in that: In step 3, the line connecting the beam center of the first laser beam and the beam center of the second laser beam is perpendicular or parallel to the movement direction of the ingot.

4. The method for stripping a crystal ingot using a dual laser beam according to claim 3, characterized in that: The first laser beam and the second laser beam may have the same or different wavelengths, pulse widths, and intensities.

5. The method for removing a crystal ingot using a dual laser beam according to claim 4, characterized in that: In step 3, the ingot moves in a straight line at a speed of 100mm / s-1200mm / s.

6. A device for peeling crystal ingots using dual laser beams, the device comprising a moving stage, a laser module, and a fixing frame, wherein the laser module is fixed on the fixing frame and positioned above the moving stage, and in use, the crystal ingot is fixed on the moving stage with the side to be peeled facing the laser module, characterized in that: The laser module emits a first laser beam and a second laser beam, with a fixed distance between the first laser beam and the second laser beam.

7. The apparatus for stripping a crystal ingot using a dual laser beam according to claim 6, characterized in that: The laser module includes a laser source, a beam expander, a beam modulation and separation module, a beam splitter, and multiple lenses and mirrors. The laser emitted by the laser source passes sequentially through the beam expander, the beam modulation and separation module, and the beam splitter. The transmitted beam from the beam splitter is collimated by the lenses and mirrors before exiting as the first laser beam. The reflected beam from the beam splitter is collimated by the lenses and mirrors before exiting as the second laser beam.

8. The apparatus for lifting a crystal ingot using a dual laser beam according to claim 7, characterized in that: The beam modulation and separation module includes a spatial light modulator and an asymmetric triangular reflector.

9. The apparatus for stripping a crystal ingot using a dual laser beam according to claim 6, characterized in that: The laser module includes a laser source, a beam expander, a beam modulation and separation module, a 4f relay system, a bifocal objective lens, and multiple mirrors. The laser emitted from the laser source passes sequentially through the beam expander, the beam modulation and separation module, the 4f relay system, and the bifocal objective lens. The output light from the bifocal objective lens consists of the first laser beam and the second laser beam.

10. The apparatus for stripping a crystal ingot using a dual laser beam according to claim 9, characterized in that: The bifocal objective lens consists of a lens and a diffractive optical element.

Images