Wafer surface low-damage efficient repairing method based on grinding wheel optimization

Abstract

The invention discloses a wafer surface low-damage efficient repair method based on grinding wheel optimization, and relates to the technical field of wafer regeneration processing, and the method comprises the following steps: S1, wafer placement; s2, multi-stage grinding is conducted; and S3, stability verification. The multi-mesh grinding wheels are arranged in series, the design of coarse-to-fine gradient grinding is adopted, the tedious processes of single-grinding-wheel reciprocating grinding and frequent mesh number switching in the prior art are omitted, seamless connection of procedures is achieved, and the machining efficiency is greatly improved. And meanwhile, real-time water injection cooling of the grinding wheel in the polishing process is achieved through cooperation of the liquid storage cavity and the liquid leakage opening, the machining temperature is effectively controlled, and wafer thermal deformation and subsurface damage caused by local overheating in the prior art are avoided.

Description

Technical Field

[0001] This invention relates to the field of wafer regeneration processing technology, and in particular to a low-damage and high-efficiency repair method for wafer surfaces based on grinding wheel optimization. Background Technology

[0002] In the back-end processes of semiconductor wafer regeneration and chip manufacturing, wafer backside thinning is a core step in the back-end processes of wafer regeneration and chip manufacturing. It is the key to improving device integration and heat dissipation performance. The industry mainstream adopts a biaxial grinding process of coarse grinding and fine grinding. The low-mesh grinding wheel is used to quickly remove material and eliminate macroscopic defects, and then the high-mesh grinding wheel is used to remove the damaged layer and obtain a smooth surface.

[0003] However, existing traditional processes face a core contradiction when dealing with heavily scratched wafers: deep damage layers in rough grinding and low efficiency in fine grinding. This not only leads to low overall processing efficiency but also increases the material removal volume and cost of downstream polishing processes due to residual damage layers. Furthermore, the stringent requirements of advanced packaging for low-damage and high-flatness wafers pose even greater challenges to the damage control capabilities of traditional grinding processes. Therefore, it is urgent to achieve a balance between efficient defect repair, precise damage control, and overall cost optimization through process optimization. Insufficient fine grinding leaves residual damage layers, forcing subsequent polishing processes to increase material removal, leading to increased polishing slurry consumption, longer processing times, and higher costs.

[0004] There are already relevant invention patents concerning the repair of wafer surface defects, as detailed below:

[0005] Chinese Patent Application No. CN202511816710.6, entitled "Wafer Re-scratching Three-Dimensional Morphology Repair Process Based on DFG8560 Single-Axis Fine Grinding Mode," discloses a wafer re-scratching three-dimensional morphology repair process based on the DFG8560 single-axis fine grinding mode. This process aims to solve the problems of deep subsurface damage, low processing efficiency, and significant material waste introduced by coarse grinding in traditional two-stage grinding processes. The process includes: identifying re-scratching defects on the wafer surface using an automated optical inspection system; employing an 8000-grit ceramic-bonded diamond grinding wheel, and performing single-axis fine grinding with ultrapure water cooling at optimized parameters of 1300–2300 rpm spindle speed, 1785–2100 rpm table speed, and a speed ratio of 0.6–1.3, skipping coarse grinding and directly completing the three-dimensional morphology repair; the grinding process involves three stages of dynamic adjustment of speed and pressure, with a total time ≤135 seconds and a removal amount of 6–12 micrometers. By adopting the above-mentioned technical solution, this application can achieve a surface roughness Ra < 0.5 nm, a subsurface damage layer < 50 nm, a repair success rate > 98.5%, significantly shorten the processing time by more than 35%, and reduce the amount of material removed by 40%.

[0006] While the aforementioned existing patents achieve efficient, low-damage, and highly continuous repair of the three-dimensional morphology of heavily scratched wafers, this process skips rough grinding and directly uses an 8000-grit wheel for uniaxial fine grinding. For wafers with deep scratches, the cutting capability of the high-grit wheel is limited, easily leading to insufficient grinding force and incomplete defect removal, thus limiting its adaptability to heavily scratched wafers. Furthermore, uniaxial fine grinding relies on a single wheel to complete the entire repair work, significantly accelerating the wear rate of the abrasive grains. This not only shortens the wheel's lifespan and increases consumable replacement costs but also easily causes fluctuations in repair accuracy in subsequent processing due to wheel wear, affecting mass production stability. Summary of the Invention

[0007] The purpose of this application is to provide a low-damage and high-efficiency wafer surface repair method based on grinding wheel optimization, which can solve the problem of low processing efficiency in the existing technology of wafer grinding and repair.

[0008] This invention provides a method for high-efficiency, low-damage repair of severely scratched wafer surfaces on a multi-axis grinding machine by optimizing different combinations of grit sizes and process parameters of coarse and fine grinding wheels. It aims to overcome the shortcomings of traditional grinding wheel combinations in treating severely scratched wafers, such as deep damage layers, low repair efficiency, and increased polishing burden. By providing a variable combination of grinding wheels and supporting processes, it achieves superior repair of severely scratched wafers on a bi-axis grinding machine, achieving quantifiable goals of increased efficiency, reduced damage, and a clear reduction in burden for downstream polishing processes.

[0009] A low-damage and high-efficiency wafer surface repair method based on grinding wheel optimization includes the following steps:

[0010] S1, Wafer Placement: Place the wafer body at the center of the top of the placement ring;

[0011] S2, Multi-stage polishing: The wafer body is polished by rotating grinding wheels with different grit sizes, and the grinding wheels are used in sequence from low to high grit size;

[0012] S3, Stability Verification: Continuously process multiple defective wafers to verify the stability of the scratch removal rate, TTV, and subsurface damage layer depth.

[0013] As a further improvement of the present invention, in step S1, a plurality of support rods are uniformly and fixedly connected to the bottom end of the placement ring, and a bonding disk is fixedly connected to the bottom end of the support rods. A vacuum pump is fixedly connected to the center of the top end of the bonding disk, and an adsorption disk is fixedly connected to the top end of the vacuum pump. A plurality of vacuum holes are opened through the top end of the adsorption disk. The adsorption disk adsorbs onto the back side of the wafer body through atmospheric pressure difference to form a negative pressure area. The vacuum holes are used to prevent excessive local pressure from causing deformation of the wafer body. The placement ring and support rods achieve the initial positioning and support of the wafer. With the help of the vacuum pump and adsorption disk, a negative pressure area is formed on the back side of the wafer by atmospheric pressure difference to complete the stable adsorption. At the same time, the plurality of vacuum holes on the adsorption disk can effectively disperse the pressure and prevent excessive local pressure from causing deformation of the wafer body. This not only ensures the positioning stability of the wafer in the subsequent polishing process, but also avoids the wafer deformation problem caused by improper clamping from the structural design.

[0014] As a further improvement of the present invention, a rotating ring is fixedly connected to the side of the bonding disk, and a base plate is provided at the bottom of the bonding disk. A positioning groove is formed at the top of the base plate below the bonding disk, and a rotating groove is formed on the side of the positioning groove. The rotating ring is rotatably connected inside the rotating groove. A motor groove is formed at the center of the bottom of the positioning groove, and a rotary motor is fixedly connected inside the motor groove. The bottom of the bonding disk is fixedly connected to the output end of the rotary motor. The rotary motor is used to drive the adsorbed wafer body to rotate in the opposite direction to the polishing end. Through the nested rotational cooperation of the rotating ring and the rotating groove, and the positioning groove limiting the bonding disk, the coaxial and precise positioning of the bonding disk and the adsorbed wafer is achieved, ensuring the concentricity of the wafer rotation during polishing and avoiding eccentric offset that affects the polishing flatness. At the same time, the rotary motor can drive the wafer and the polishing end to rotate in the opposite direction, forming a differential cutting effect, which can improve the cutting efficiency of polishing, reduce cutting resistance, and reduce the probability of scratches on the wafer surface caused by polishing.

[0015] As a further improvement of the present invention, in step S2, a telescopic motor is fixedly connected to the top center of the chassis, a telescopic rod is fixedly connected to the top output end of the telescopic motor, a series disk is rotatably connected to the top end of the telescopic rod, a second rotary motor is provided at the bottom end of the series disk, the output end of the second rotary motor is fixedly connected to the top end of the series disk, and several L-shaped connecting rods are fixedly connected to the side of the series disk. The second rotary motor can drive the series disk and the L-shaped connecting rods on the side to rotate as a whole, which can realize the rapid switching and precise alignment of different grinding wheels, eliminating the need for manual adjustment of the grinding wheel position and making the multi-stage gradient grinding process smoother.

[0016] As a further improvement of the present invention, a rotary motor three is fixedly connected to the end of the L-shaped connecting rod away from the series disk, and a liquid storage chamber is fixedly connected to the bottom output end of the rotary motor three. The rotary motor three rotates in the opposite direction to the rotary motor one. By setting the rotary motor three at the end of the L-shaped connecting rod and connecting it to the liquid storage chamber, an independent rotary drive source is provided for the grinding wheel. Furthermore, the rotary motor three rotates in the opposite direction to the rotary motor one that drives the wafer, creating a differential cutting effect. This significantly improves the cutting contact efficiency between the abrasive grains and the wafer surface, accelerates the grinding process, makes the cutting force more uniform, and reduces scratches on the wafer surface.

[0017] As a further improvement of the present invention, the bottom outer ring of the liquid storage chamber is provided with threaded grooves, and threaded rings are threadedly connected inside the threaded grooves. A thickness ring is fixedly connected to the bottom end of each threaded ring. A grinding wheel placement groove is provided between the thickness ring and the threaded ring. Different grinding wheel placement grooves are used to sequentially place grinding wheels of different grit sizes according to their mesh size. The threaded connection between the threaded grooves and threaded rings at the bottom of the liquid storage chamber facilitates convenient assembly and disassembly, allowing for quick replacement and fixation of grinding wheels, thus improving equipment operating efficiency. Furthermore, the grinding wheel placement grooves between the thickness ring and the threaded ring allow for the orderly placement of grinding wheels of different mesh sizes, adapting to the process requirements of multi-stage gradient grinding, eliminating the need for frequent adjustments to the grinding wheel position, and making the transition from coarse grinding to fine grinding smoother.

[0018] As a further improvement of the present invention, each of the liquid storage cavities is movably connected to a sealing plug at its top and a drain outlet at its bottom. The sealing plug and drain outlet are used to inject water onto the surface of the grinding wheel during the rotary grinding of the wafer body. This water injection prevents localized overheating of the wafer body, which could lead to deformation. The sealing plug at the top of the liquid storage cavity facilitates convenient filling of the coolant and seals the cavity, while the drain outlet at the bottom precisely guides the coolant to the surface of the grinding wheel. Real-time water injection and cooling during grinding effectively removes the heat generated during grinding, preventing wafer deformation due to localized overheating. It also reduces wear on the grinding wheel caused by high temperatures, thereby extending the grinding wheel's service life.

[0019] As a further improvement of the present invention, in step S3, the wafer body used for verification is a 12-inch single-crystal Si wafer. A rapid surface inspection step for the wafer body is set at the process connection between each stage of the multi-stage polishing process. This inspection step uses an online laser thickness gauge to detect whether the wafer body removal amount has reached the polishing target. Simultaneously, the inspection step also uses a high-speed vision camera to quickly determine the initial removal effect of severe scratches. By combining the online laser thickness gauge's precise detection of whether the wafer removal amount meets the standard and the high-speed vision camera's rapid determination of the initial removal effect of severe scratches, real-time quality control of the polishing process is achieved without the need for offline inspection. This allows the processing parameters of each polishing stage to be adjusted in a timely manner based on the inspection results, effectively avoiding over-polishing or incomplete defect removal, while also ensuring smooth process connections and improving overall processing efficiency.

[0020] Compared with the prior art, the beneficial effects of this invention are as follows:

[0021] 1. The design employs a porous, uniform negative pressure adsorption system paired with distributed support rods to effectively prevent wafer deformation caused by localized pressure overload, ensuring uniform stress on the wafer surface. Simultaneously, the nested positioning structure of the rotating ring and positioning groove achieves coaxial positioning of the wafer and grinding mechanism, significantly improving grinding concentricity. Furthermore, the reverse differential design, where the rotary motor drives the wafer and grinding wheel in opposite directions, not only improves grinding efficiency but also reduces cutting resistance, minimizing scratches on the wafer surface. This fundamentally avoids repair quality defects caused by positioning misalignment and uneven stress in existing technologies.

[0022] 2. The design employs a series arrangement of multi-mesh grinding wheels, with a gradient grinding pattern from coarse to fine, eliminating the cumbersome process of reciprocating grinding with a single wheel and frequent grit switching required in existing technologies. This achieves seamless process integration and significantly improves processing efficiency. Simultaneously, the combination of a liquid storage chamber and a drain outlet enables real-time water cooling of the grinding wheels during grinding, effectively controlling the processing temperature and preventing wafer thermal deformation and subsurface damage caused by localized overheating in existing technologies. Furthermore, the counter-rotating differential rotation design between the grinding wheels and the wafer ensures more uniform cutting force and higher repair precision. The threaded quick-release structure simplifies the grinding wheel replacement process, improving operational convenience. The gradient grit cutting method weakens cutting stress and significantly reduces the depth of the subsurface damage layer, effectively solving the core pain points of existing technologies—namely, the lack of transition between coarse and fine grinding and the deep damage layer—thereby improving the wafer repair yield.

[0023] 3. A real-time detection mode combining an online laser thickness gauge and a high-speed vision camera is implemented at the junctions of multi-stage polishing processes. This allows for rapid detection of whether the wafer removal amount has reached the target and assesses the initial removal effect of re-scratch, achieving closed-loop process control, timely correction, and effectively avoiding batch scrap and reducing production costs. Adaptive calibration of process parameters is achieved through continuous processing and verification of multiple defective wafers, ensuring uniform batch repair results. Furthermore, online rapid detection eliminates the need for downtime, further shortening the processing cycle and achieving both high efficiency and stability, perfectly adapting to the needs of industrial mass production.

[0024] 4. The grinding wheel can be precisely matched according to the wear condition detected in the early stage of wafer grinding. Coolant can also be pre-supplyed before grinding to ensure effective heat dissipation at the beginning of grinding, avoiding problems such as wafer surface ablation and grinding wheel thermal deformation caused by high temperatures in the initial grinding stage. Simultaneously, the counter-rotation of the wafer and grinding wheel allows the coolant to cover the grinding contact surface more evenly under centrifugal force, enhancing heat dissipation and cooling uniformity. Furthermore, targeted inspections are set after each grinding process to accurately control the real-time repair status of the wafer surface, avoiding incomplete repair and over-grinding. The grinding wheel grit can also be matched step-by-step according to the repair progress to achieve graded processing. While efficiently eliminating surface wear defects, the depth of the subsurface damage layer on the wafer is precisely controlled, further improving the surface quality and performance of the repaired wafer. Attached Figure Description

[0025] To more clearly illustrate the technical solutions in the embodiments of this application or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0026] Figure 1 This is a flowchart of the steps of the present invention.

[0027] Figure 2 This is a three-dimensional structural diagram of the present invention.

[0028] Figure 3 This is a three-dimensional structural diagram of the chassis and telescopic motor in this invention.

[0029] Figure 4 This is a three-dimensional structural diagram of the bonding plate and the adsorption plate in this invention.

[0030] Figure 5 In this invention Figure 4 A schematic diagram of the three-dimensional structure from another angle.

[0031] Figure 6 This is a three-dimensional structural diagram of the liquid storage cavity and the threaded ring in the separated state in this invention.

[0032] Figure 7 This is a three-dimensional structural diagram of the threaded ring and the thickness ring in this invention.

[0033] Figure 8 This is a cross-sectional three-dimensional structural diagram of the liquid storage cavity in this invention.

[0034] Figure 9 For the present invention Figure 3 A magnified three-dimensional structural diagram at point A in the middle.

[0035] In the diagram: 101, chassis; 102, positioning groove; 103, rotating groove; 104, motor groove; 201, bonding plate; 202, rotating ring; 203, rotary motor one; 204, support rod; 205, placement ring; 206, vacuum pump; 207, adsorption plate; 208, vacuum hole; 301, telescopic motor; 302, telescopic rod; 303, series plate; 304, rotary motor two; 305, L-shaped connecting rod; 306, rotary motor three; 307, liquid storage chamber; 308, sealing plug; 309, leakage port; 310, threaded groove; 311, threaded ring; 312, grinding wheel placement groove; 313, thickness ring; 400, wafer body. Detailed Implementation

[0036] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0037] A low-damage and high-efficiency wafer surface repair method based on grinding wheel optimization, such as Figure 1 As shown, it includes the following steps:

[0038] S1, Wafer Placement: The wafer body 400 is placed at the center of the top of the placement ring 205. Several support rods 204 are evenly fixedly connected to the bottom of the placement ring 205. A bonding disk 201 is fixedly connected to the bottom of the support rods 204. A vacuum pump 206 is fixedly connected to the center of the top of the bonding disk 201. An adsorption disk 207 is fixedly connected to the top of the vacuum pump 206. Several vacuum holes 208 are opened through the top of the adsorption disk 207. The adsorption disk 207 adsorbs onto the back of the wafer body 400 through atmospheric pressure difference to form a negative pressure area. The vacuum holes 208 are used to prevent excessive local pressure from deforming the wafer body 400. A rotating ring 202 is fixedly connected to the side of the bonding disk 201. A base plate 101 is provided at the bottom of the bonding disk 201. A positioning groove 102 is provided at the top of the base plate 101 below the bonding disk 201. A rotating groove 103 is provided on the side of the positioning groove 102. The rotating ring 202 is rotatably connected inside the rotating groove 103. A motor groove 104 is provided at the center of the bottom of the positioning groove 102. A rotary motor 203 is fixedly connected inside the motor groove 104. The bottom of the bonding disk 201 is fixedly connected to the output end of the rotary motor 203. The rotary motor 203 is used to drive the wafer body 400 that has completed adsorption to rotate in the opposite direction to the polishing.

[0039] Using the wafer immobilization and adsorption process as a comparative condition, the advantages of this invention compared to existing technologies are analyzed in detail, as shown in the following comparison table:

[0040] Table 1 Comparison of Wafer Immobilization and Adsorption Processes

[0041]

[0042] Based on Table 1 above, it can be concluded that this invention achieves synergistic optimization of precise positioning, uniform force distribution, and efficient grinding in the wafer fixing and adsorption process. It abandons the drawbacks of single-point or localized vacuum adsorption in existing technologies, employing a porous uniform negative pressure adsorption design, coupled with distributed support rods, effectively avoiding wafer deformation caused by localized pressure overload and ensuring uniform force distribution on the wafer surface. Furthermore, the reverse differential design of the rotary motor and the grinding wheel, rotating in opposite directions, not only improves grinding efficiency but also reduces cutting resistance, minimizing scratches on the wafer surface. Moreover, the nested limiting structure of the rotating ring and positioning groove achieves coaxial positioning of the wafer and the grinding mechanism, significantly improving grinding concentricity and ensuring controllable flatness of the repaired surface, fundamentally avoiding repair quality defects caused by positioning misalignment and uneven force distribution in existing technologies.

[0043] S2, Multi-stage polishing: Different grit grinding wheels are used to rotate and polish the wafer body 400, with the grit of the grinding wheels increasing sequentially. A telescopic motor 301 is fixedly connected to the top center of the chassis 101. A telescopic rod 302 is fixedly connected to the top output end of the telescopic motor 301. A series disk 303 is rotatably connected to the top of the telescopic rod 302. A second rotary motor 304 is set at the bottom of the series disk 303. The output end of the second rotary motor 304 is fixedly connected to the top of the series disk 303. Several L-shaped connecting rods 305 are fixedly connected to the side of the series disk 303. A third rotary motor 306 is fixedly connected to the end of each L-shaped connecting rod 305 away from the series disk 303. A liquid storage chamber 307 is fixedly connected to the bottom output end of each third rotary motor 306. The rotation direction of the third rotary motor 306 is opposite to that of the first rotary motor 203. The bottom outer ring of the liquid storage chamber 307 is provided with threaded grooves 310, and each threaded groove 310 is threaded with a threaded ring 311. The bottom end of each threaded ring 311 is fixedly connected with a thickness ring 313. A grinding wheel placement groove 312 is provided between the thickness ring 313 and the threaded ring 311. Different grinding wheel placement grooves 312 are used to place grinding wheels of different grit sizes in sequence according to their grit size. The top of each liquid storage chamber 307 is movably connected with a sealing plug 308, and the bottom of each liquid storage chamber 307 is provided with a drain port 309. The sealing plug 308 and the drain port 309 are used to inject water into the surface of the grinding wheel when the wafer body 400 is rotated and polished. The water injection is used to prevent the wafer body 400 from being deformed due to local overheating.

[0044] Using a multi-stage polishing process as a comparison, the advantages of this invention compared to existing technologies are analyzed in detail, as shown in the following comparison table:

[0045] Table 2 Comparison of Multi-Stage Grinding Processes

[0046]

[0047] Based on Table 2 above, it can be concluded that this technology achieves multiple breakthroughs in efficiency improvement, damage control, and ease of operation compared to existing technologies. The design employing a series arrangement of multi-mesh grinding wheels and a gradient grinding pattern from coarse to fine eliminates the cumbersome process of single-wheel reciprocating grinding and frequent mesh switching required in existing technologies, achieving seamless process integration and significantly improving processing efficiency. Simultaneously, the combination of a liquid storage chamber and a drain outlet enables real-time water cooling of the grinding wheels during grinding, effectively controlling the processing temperature and avoiding wafer thermal deformation and subsurface damage caused by localized overheating in existing technologies. Furthermore, the grinding drive and grinding wheel replacement utilize a reverse differential rotation design between the grinding wheel and the wafer, resulting in more uniform cutting force and higher repair accuracy. The threaded quick-release structure simplifies the grinding wheel replacement process, improving operational convenience. The gradient mesh cutting method weakens cutting stress, significantly reducing the depth of the subsurface damage layer and effectively improving the wafer repair yield, solving the core pain points of existing technologies such as lack of transition between coarse and fine grinding and deep damage layers.

[0048] S3, Stability Verification: Multiple defective wafers were continuously processed to verify the stability of the re-scratch removal rate, TTV (Total Television Volume), and subsurface damage layer depth. The verification used a 12-inch single-crystal Si wafer body (WBC) 400. A rapid surface inspection step on the WBC 400 was implemented at the connection points between each stage of the multi-stage polishing process. This inspection step used an online laser thickness gauge to check whether the removal amount on the WBC 400 met the polishing target. Simultaneously, a high-speed vision camera quickly determined the initial removal effect of the re-scratch.

[0049] Based on the comparison criteria of processing stability and testing procedures, the advantages of this invention compared to existing technologies are analyzed in detail, as shown in the following comparison table:

[0050] Table 3 Comparison of Processing Stability and Testing Procedures

[0051]

[0052] Based on Table 3 above, it can be concluded that this invention overcomes the drawbacks of the lag in offline sampling inspection of existing technologies. It adopts a real-time detection mode combining an online laser thickness gauge and a high-speed vision camera, enabling rapid detection at the process transitions of multi-stage polishing to determine whether the wafer removal amount has reached the target. Simultaneously, it assesses the initial removal effect of re-scratch, achieving closed-loop process control, timely correction, and effectively avoiding batch scrap, thus reducing production costs. Furthermore, it overcomes the limitations of single-index detection in existing technologies by simultaneously verifying three core indicators: re-scratch removal rate, TTV, and subsurface damage layer depth. This provides comprehensive control over repair quality, significantly improving process stability. Through continuous processing verification of multiple defective wafers, it achieves adaptive calibration of process parameters, solving the problems of large parameter drift and poor batch consistency in existing technologies. This ensures uniform batch repair results, perfectly adapting to the needs of industrial mass production. Simultaneously, online rapid detection eliminates the need for downtime, further shortening the processing cycle and achieving the dual goals of high efficiency and stability.

[0053] In this embodiment, taking the repair of a moderately worn wafer as an example, the wafer body 400 is first aligned with the center of the placement ring 205. Then, the wafer body 400 is adsorbed onto the top of the adsorption plate 207 using a vacuum pump 206. After the wafer body 400 is fixed, the wear condition of the wafer body 400 surface is detected through a testing step. Based on the detected wear condition, a grinding wheel of different grit is used to polish it. Aligning the wafer body 400 with the center of the placement ring and fixing it with the vacuum pump 206 first enables precise positioning and stable clamping of the wafer body 400, effectively avoiding the wafer body 400 from shifting or sliding during the polishing process, ensuring the uniformity and processing accuracy of the polishing, and preventing new surface defects caused by the displacement of the wafer body 400.

[0054] Before polishing, the liquid reservoir 307, containing a suitable grinding wheel, and the thickness ring 313 are moved above the wafer body 400 by starting the second rotary motor 304. At this point, the sealing plug 308 on top of the liquid reservoir 307 is manually removed, and coolant is injected through the hole on top of the liquid reservoir 307. The coolant flows to the surface of the grinding wheel through the drain port 309. Simultaneously, the third rotary motor 306 at the top of the liquid reservoir 307 and the first rotary motor 203 at the bottom of the bonding disk 201 are started, causing the wafer body 400 and the grinding wheel to rotate in opposite directions. The coolant assists in heat dissipation during the polishing process of the wafer body 400. The reverse rotation of the wafer body 400 and the grinding wheel improves polishing efficiency. The second rotary motor 304 precisely moves the adapted grinding wheel liquid reservoir 307 and thickness ring 313 above the wafer body 400, achieving rapid alignment and adaptation of the grinding wheel, thus preparing for efficient subsequent polishing. Simultaneously, coolant is injected manually by removing the plug and flows to the grinding wheel surface through the drain port 309. This pre-supply of coolant before grinding ensures effective heat dissipation from the initial stage, preventing issues such as wafer body 400 surface ablation and grinding wheel thermal deformation caused by high temperatures at the beginning of grinding. Simultaneously, rotary motors 306 and 203 drive the wafer body 400 and grinding wheel to rotate in opposite directions. This significantly improves the cutting contact efficiency between the abrasive grains and the wafer surface, accelerating the grinding process and increasing overall processing efficiency. Furthermore, the centrifugal force of the reverse rotation allows the coolant to more evenly cover the grinding contact surface, enhancing heat dissipation and cooling uniformity, and reducing surface defects on the wafer body 400 caused by high temperatures. In addition, the reverse rotation grinding method results in more uniform grinding marks on the wafer body 400 surface, improving surface smoothness after grinding. Overall, heat dissipation, efficiency, and processing quality are simultaneously guaranteed in wafer grinding.

[0055] After the initial grinding stage is completed, the wear condition of the wafer body 400 surface is further inspected through testing. Then, depending on the wear condition, different grit grinding wheels are used, and the above steps are repeated for secondary grinding. This process is repeated until the wafer body 400 surface is ground to meet usage standards. Through targeted testing after each grinding process, the real-time wear repair status of the wafer body 400 surface is precisely controlled, avoiding incomplete repair and over-grinding problems that can occur with single-wheel grinding, thus ensuring grinding accuracy. Furthermore, the grinding wheel grit can be matched step-by-step according to the repair progress, achieving gradient processing from rough finishing to fine grinding. This efficiently eliminates surface wear defects while precisely controlling the depth of the subsurface damage layer on the wafer body 400, improving the surface quality and performance of the repaired wafer.

[0056] Finally, it should be noted that the above are merely preferred embodiments of the present invention and are not intended to limit the present invention. Although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art can still modify the technical solutions described in the foregoing embodiments or make equivalent substitutions for some of the technical features. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the protection scope of the present invention.

Claims

1. A low-damage, high-efficiency repair method for wafer surfaces based on grinding wheel optimization, characterized in that, Specifically, the steps include the following: S1, wafer placement: Place the wafer body (400) at the center of the top of the placement ring (205); S2, Multi-stage polishing: The wafer body (400) is polished by rotating grinding wheels with different grit sizes, and the grinding wheels are used in sequence from low to high grit size; S3, Stability Verification: Continuously process multiple defective wafers to verify the stability of the scratch removal rate, TTV, and subsurface damage layer depth.

2. The method for low-damage and high-efficiency repair of wafer surfaces based on grinding wheel optimization as described in claim 1, characterized in that: In step S1, several support rods (204) are uniformly fixedly connected to the bottom end of the placement ring (205). A bonding disk (201) is fixedly connected to the bottom end of the support rod (204). A vacuum pump (206) is fixedly connected to the center of the top end of the bonding disk (201). An adsorption disk (207) is fixedly connected to the top end of the vacuum pump (206). Several vacuum holes (208) are opened through the top end of the adsorption disk (207). The adsorption disk (207) is adsorbed on the back of the wafer body (400) by atmospheric pressure difference to form a negative pressure area. The vacuum holes (208) are used to avoid excessive local pressure that could cause deformation of the wafer body (400).

3. The method for low-damage and high-efficiency repair of wafer surfaces based on grinding wheel optimization as described in claim 2, characterized in that: A rotating ring (202) is fixedly connected to the side of the bonding disk (201). A base plate (101) is provided at the bottom of the bonding disk (201). A positioning groove (102) is provided at the top of the base plate (101) below the bonding disk (201). A rotating groove (103) is provided on the side of the positioning groove (102). The rotating ring (202) is rotatably connected inside the rotating groove (103). A motor groove (104) is provided at the center of the bottom of the positioning groove (102). A rotary motor (203) is fixedly connected inside the motor groove (104). The bottom of the bonding disk (201) is fixedly connected to the output end of the rotary motor (203). The rotary motor (203) is used to drive the wafer body (400) that has completed adsorption to rotate in the opposite direction to the polishing.

4. The method for low-damage and high-efficiency repair of wafer surfaces based on grinding wheel optimization as described in claim 3, characterized in that: In step S2, a telescopic motor (301) is fixedly connected to the top center of the chassis (101), a telescopic rod (302) is fixedly connected to the top output end of the telescopic motor (301), a series disk (303) is rotatably connected to the top end of the telescopic rod (302), a second rotary motor (304) is provided at the bottom end of the series disk (303), the output end of the second rotary motor (304) is fixedly connected to the top end of the series disk (303), and several L-shaped connecting rods (305) are fixedly connected to the side of the series disk (303).

5. The method for low-damage and high-efficiency repair of wafer surfaces based on grinding wheel optimization as described in claim 4, characterized in that: The L-shaped connecting rod (305) is fixedly connected to a rotary motor three (306) at the end away from the series disk (303). The bottom output end of the rotary motor three (306) is fixedly connected to a liquid storage chamber (307). The rotary motor three (306) and the rotary motor one (203) rotate in opposite directions.

6. The method for low-damage and high-efficiency repair of wafer surfaces based on grinding wheel optimization as described in claim 5, characterized in that: The bottom outer ring of the liquid storage cavity (307) is provided with a threaded groove (310), and a threaded ring (311) is threadedly connected inside the threaded groove (310). A thickness ring (313) is fixedly connected to the bottom of the threaded ring (311). A grinding wheel placement groove (312) is provided between the thickness ring (313) and the threaded ring (311). Different grinding wheel placement grooves (312) are used to place grinding wheels of different mesh sizes in sequence according to the mesh size.

7. The method for low-damage and high-efficiency repair of wafer surfaces based on grinding wheel optimization as described in claim 6, characterized in that: Each of the liquid storage chambers (307) is movably connected to a sealing plug (308) at its top end, and each of the liquid storage chambers (307) is provided with a drain port (309) at its bottom end. The sealing plug (308) and the drain port (309) are used to inject water into the surface of the grinding wheel when the wafer body (400) is rotated and polished. The water injection is used to prevent the wafer body (400) from overheating locally and causing it to deform.

8. The method for low-damage and high-efficiency repair of wafer surfaces based on grinding wheel optimization as described in claim 1, characterized in that: In step S3, the wafer body (400) used for verification is a 12-inch single-crystal Si wafer. A rapid detection step on the surface of the wafer body (400) is set at the process connection between each stage of the multi-stage polishing. The detection step uses an online laser thickness gauge to detect whether the amount of removal of the wafer body (400) has reached the polishing target. The detection step also uses a high-speed vision camera to quickly determine the preliminary removal effect of the heavy scratches.

Images