Semiconductor trimming integration optimization system and method

Abstract

The invention belongs to the technical field of semiconductor processing, and particularly relates to a semiconductor trimming integration optimization system and method.According to the semiconductor trimming integration optimization system and method, sensitive points are avoided by dynamically adjusting cutting parameters and planning an avoiding route, and the influence on adjacent structures in the cutting process is minimized; a cutting environment is kept stable by applying a multi-stage cooling strategy, and material damage caused by overheating is prevented; in the cooling process, temperature changes are monitored in real time, the cutting speed is corrected in a self-adaptive mode, and the flexibility and accuracy of the cutting process are ensured; and finally, the product consistency is verified through strict quality control inspection. Through the combined action of the measures, the cutting quality and the yield are remarkably improved, defects are reduced, and the more stable production process and the higher semiconductor chip quality are achieved.

Description

Technical Field

[0001] The present invention belongs to the field of semiconductor processing technology, and in particular relates to a semiconductor cutting rib integration optimization system and method. Background Art

[0002] In the semiconductor manufacturing process, the dicing process is a key step in dividing large-sized wafers into individual chips. Traditional dicing methods usually rely on fixed cutting paths and parameter settings. Although these methods can achieve basic dicing tasks, they have significant limitations when dealing with semiconductor materials with complex structures or high-density integration. In particular, during the dicing process, the impact on adjacent structures may cause defects, reduce yields, and increase production costs.

[0003] In the prior art, it is common practice to cut on a preset fixed path, with less consideration of the specific impact of material properties or potentially sensitive areas. In order to reduce thermal damage during the cutting process, coolant is usually used to control the temperature, but this cooling method is often a single stage, and the cooling parameters are not adjusted according to real-time temperature changes. In addition, when encountering microcracks or other defects inside the material, the traditional method lacks an effective avoidance mechanism, which may lead to a decrease in cutting quality. Therefore, how to minimize the impact on adjacent structures during the cutting process to reduce defects and improve yield is an urgent problem to be solved. Summary of the invention

[0004] The purpose of the present invention is to provide a semiconductor cutting rib integration optimization system and method, which not only solves the problems existing in the prior art, but also achieves higher cutting accuracy, lower defect rate and more stable production process through a series of innovative improvements, bringing significant technological progress to the semiconductor manufacturing industry to solve the problems raised in the above background technology.

[0005] To achieve the above object, the present invention proposes a semiconductor cutting rib integration optimization method, comprising the following steps: Step 1: preparing the semiconductor material to be processed and setting an initial cutting path, and performing a pre-scan based on the initial cutting path to evaluate the characteristics of the affected area; Step 2: Adjust cutting parameters based on the obtained evaluation information, plan dynamic avoidance routes based on parameter adjustments, avoid sensitive points, and apply multi-stage cooling strategies to maintain a stable cutting environment; Step 3: Monitor temperature changes in real time during the cooling process and adjust the cutting speed, while implementing adaptive corrections; Step 4: Perform a quality control check after cutting to verify product consistency.

[0006] Preferably, the setting of the initial cutting path includes: Determine the thickness D of the semiconductor material and measure the vertical distance H from the material surface to the predetermined cutting line; According to the obtained D and H values, calculate the initial cutting path offset O1, where O1=sqrt(D^2+H^2)-D; Based on the calculated offset O1, adjust the cutting starting point to keep a safe distance S1 from the edge of the material at the beginning of cutting: S1=O1+5mm; Use the set safety interval S1 to draw the final cutting path diagram.

[0007] Preferably, performing a pre-scan based on the initial cutting path to evaluate the characteristics of the affected area comprises: Pre-scan along the set initial cutting path at a constant speed V1, and record the material density M1 at each point on the path; According to the recorded material density M1, calculate the anti-interference coefficient C1 of each point, where C1=M1 / (M1+10); Based on the obtained anti-interference coefficient C1, a distribution map of the affected area is constructed, marking all locations where the C1 value exceeds the threshold T1; Using the marked influence area, the pre-scan path is adjusted to ensure that the new path avoids the high C1 value area while maintaining the maximum similarity S2 with the original path. Here S2 is measured by calculating the average deviation D1 between the two paths, that is, S2=1-(D1 / L1), where L1 represents the total length of the path.

[0008] Preferably, adjusting the cutting parameters according to the acquired evaluation information includes: According to the constructed influence area distribution map, the anti-interference coefficient C1 value of each cutting point is determined. For each cutting point, the adjusted cutting power Pw1 is calculated, where Pw1=P0*(1-C1), and P0 is the originally set cutting power; Based on the adjusted Pw1, re-evaluate the cutting speed Vc1, so that Vc1=V0*sqrt(Pw1 / P0), V0 is the originally set cutting speed; Using the adjusted speed Vc1 and power Pw1, a set of cutting parameters that minimize interference is planned, while ensuring that all points on the cutting path meet the quality standard Qs1, that is, Qs1=min(Qi1), where Qi1 represents the quality score of each cutting point, which is determined by comparing the material properties before and after cutting.

[0009] Preferably, the step of adjusting parameters and planning a dynamic avoidance route to avoid sensitive points includes: Based on the adjusted cutting parameter set, identify all locations on the cutting path where the C1 value exceeds the threshold T1 and mark them as sensitive points. For each marked sensitive point, calculate the avoidance distance R1, where R1=2*sqrt(C1-T1); According to the avoidance distance R1, a temporary avoidance path segment Ls1 is constructed. The temporary avoidance path segment is based on the original path and is offset outward by a distance R1 at the sensitive point to form a new cutting trajectory; Combined with the temporary avoidance path segment Ls1, the overall cutting path is optimized to ensure that the new path not only avoids all sensitive points but also keeps the cutting efficiency E1 unchanged, that is, E1 = L1 / (ΣLs1), where L1 is the original cutting path length and ΣLs1 represents the sum of all avoidance path segments.

[0010] Preferably, the multi-stage cooling strategy is applied to maintain a stable cutting environment, including: Based on the optimized cutting path, the estimated cutting time Tc1 of each cutting section is determined, and the position of the sensitive point is recorded. For each cutting section, the cooling requirement coefficient K1 is calculated, where K1=(Tc1 / 10)+C1; According to the obtained cooling demand coefficient K1, a multi-stage cooling scheme is planned, including pre-cooling, main cooling and post-cooling. The cooling intensity Ipre1 of the pre-cooling stage is set to K1*0.5, the cooling intensity Imain1 of the main cooling stage is set to K1*1.2, and the cooling intensity Ipost1 of the post-cooling stage is set to K1*0.3. Use the cooling intensity setting to dynamically adjust the coolant flow Q1, monitor the temperature change ΔT1 in real time during the cutting process, and adjust the flow through the formula Q1=Qbase1+ΔT1*K1. Qbase1 is the basic cooling flow to ensure a stable cutting environment.

[0011] Preferably, the real-time monitoring of temperature changes and adjustment of cutting speed during the cooling process includes: As the cooling process proceeds, a temperature sensor is used to monitor the temperature change ΔT1 of the cutting point in real time, and the temperature data within each time interval t1 is recorded; Based on the obtained ΔT1 value, calculate the temperature fluctuation coefficient F1, where F1=|ΔT1| / T01, T01 is the initially set stable temperature; According to the temperature fluctuation coefficient F1, adjust the cutting speed Vc1 so that Vc1=Vbase1*(1-F1), where Vbase1 is the originally set cutting speed; Using the adjusted cutting speed Vc1, re-evaluate the timing of the entire cutting process to ensure that the total cutting time Ttotal1 does not exceed the preset maximum allowable time Tmax1, that is, Ttotal1=Σ(Vc1*Ls1)≤Tmax1.

[0012] Preferably, the implementing of the adaptive correction comprises: Based on the adjusted cutting speed Vc1 and the real-time temperature change ΔT1, the correction coefficient G1 of each cutting section is calculated, where G1=(Vc1 / Vbase1)*(1+ΔT1 / T01); According to the correction coefficient G1, the cutting parameter set is updated, and the new cutting parameter Pnew1=Pold1*G1 is set for each sensitive point position, where Pold1 is the originally set cutting parameter; Using the updated cutting parameters Pnew1, re-plan the avoidance strategy on the cutting path to ensure that the new path not only avoids the high anti-interference coefficient area, but also adapts to the revised parameters to keep the cutting efficiency E1 unchanged, that is, E1=L1 / (ΣLs1); Combined with the optimization path and updated parameters, adaptive correction is implemented. By monitoring the dynamic feedback during the cutting process, parameters are continuously adjusted to ensure that the product quality Qs1 reaches the highest standard, that is, Qs1=max(Qi1).

[0013] Preferably, after the cutting is completed, a quality control check is performed to verify the consistency of the product, including: Based on the implemented adaptive corrections, perform a preliminary visual inspection of the finished cut products, record the location and type of all visible defects, and calculate the defect density Dd1; According to the defect density Dd1, the quality assessment formula Qa1=1-(Dd1 / Dmax1) is applied, where Dmax1 is the maximum acceptable defect density; For products whose quality score Qa1 is lower than the threshold Qt1, perform secondary precision measurement, record the defect area Af1, and update the total defect area At1=At1+Af1; Combined with the total defect area At1, it is finally determined whether the product has passed the quality control. If At1 is less than or equal to the preset maximum allowable defect area Am1, the product is considered qualified; otherwise, it is marked as unqualified.

[0014] In another aspect, the present invention provides a semiconductor cutting integration optimization system, comprising: A material preparation and pre-scan evaluation module, for preparing semiconductor materials to be processed and setting an initial cutting path, and performing a pre-scan based on the initial cutting path to evaluate the characteristics of the affected area; The parameter adjustment and dynamic avoidance planning module is used to adjust the cutting parameters according to the acquired evaluation information, plan the dynamic avoidance route in combination with the parameter adjustment, avoid sensitive points, and apply the multi-stage cooling strategy to maintain the stability of the cutting environment; Real-time monitoring and adaptive correction module, used to monitor temperature changes in real time during the cooling process and adjust the cutting speed, and at the same time, implement adaptive correction; Quality control and product verification module for quality control checks after cutting to verify product consistency.

[0015] Technical effects and advantages of the present invention: The semiconductor cutting rib integration optimization system and method proposed by the present invention have the following advantages compared with the prior art: The present invention minimizes the impact on adjacent structures during the cutting process by dynamically adjusting cutting parameters and planning avoidance routes to avoid sensitive points; applies a multi-stage cooling strategy to maintain a stable cutting environment to prevent material damage caused by overheating; monitors temperature changes in real time during the cooling process and adaptively corrects the cutting speed to ensure the flexibility and accuracy of the cutting process; and finally verifies product consistency through strict quality control inspections. These measures work together to significantly improve cutting quality and yield, reduce defects, and achieve a more stable production process and higher semiconductor chip quality. BRIEF DESCRIPTION OF THE DRAWINGS

[0016] Figure 1 A flow chart of the semiconductor cutting rib integration optimization method of the present invention; Figure 2 FIG. 1 is a block diagram of a semiconductor cutting integration optimization system according to the present invention. DETAILED DESCRIPTION

[0017] The following will be combined with the accompanying drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, rather than all the embodiments. The specific embodiments described herein are only used to explain the present invention and are not used to limit the present invention. Based on the embodiments in the present invention, all other embodiments obtained by ordinary technicians in this field without making creative work are within the scope of protection of the present invention.

[0018] The present invention provides Figure 1 The semiconductor cutting rib integration optimization method shown includes the following steps: Step 1: Prepare the semiconductor material to be processed and set an initial cutting path. Based on the initial cutting path, perform a pre-scan to evaluate the characteristics of the affected area; specifically as follows: Setting the initial cutting path includes: Determine the thickness D of the semiconductor material and measure the vertical distance H from the material surface to the predetermined cutting line; ensure accurate understanding of material properties, provide basic data for subsequent calculations, and help plan a more accurate and safe cutting path.

[0019] Based on the obtained D and H values, calculate the initial cutting path offset O1, where O1=sqrt(D^2+H^2)-D; quantify the safe distance between the cutting starting point and the edge of the material through a mathematical formula to ensure that the cutting process avoids potential sensitive areas and reduces the risk of damage.

[0020] Based on the calculated offset O1, the cutting starting point is adjusted to keep a safe distance S1 from the edge of the material when cutting starts: S1=O1+5mm; an additional safety margin (5mm) is provided to further reduce the possibility of accidental damage to the edge of the material during cutting.

[0021] Use the set safety interval S1 to draw the final cutting path diagram; integrate all the preliminary calculation results into a detailed cutting path diagram to guide the actual cutting operation and ensure that the cutting path is both safe and efficient.

[0022] For example, assuming that the thickness D of a piece of semiconductor material is 0.5 mm and the vertical distance H from the surface of the material to the predetermined cutting line is 1.5 mm, then: The initial cutting path offset O1=sqrt(0.5^2+1.5^2)-0.5≈1.32mm.

[0023] Safety interval S1=1.32mm+5mm=6.32mm.

[0024] Perform a pre-scan at a constant speed V1 along the set initial cutting path, and record the material density M1 at each point on the path; obtain material property information on the cutting path to provide data support for subsequent analysis and help identify key points that may affect cutting quality.

[0025] According to the recorded material density M1, the anti-interference coefficient C1 of each point is calculated, where C1=M1 / (M1+10); quantifying the influence of the material on the cutting process helps to optimize the cutting parameters in a targeted manner and improve the cutting accuracy and yield rate.

[0026] Based on the obtained anti-interference coefficient C1, a distribution map of the affected area is constructed, marking all locations where the C1 value exceeds the threshold T1; the high anti-interference coefficient areas are intuitively displayed to facilitate the planning of avoidance routes and avoid unnecessary damage to these sensitive areas during the cutting process.

[0027] Using the marked impact area, adjust the pre-scan path to ensure that the new path avoids the high C1 value area while maintaining the maximum similarity S2 with the original path. Here, S2 is measured by calculating the average deviation D1 between the two paths, that is, S2=1-(D1 / L1), where L1 represents the total length of the path. Under the premise of ensuring cutting efficiency, avoid sensitive areas as much as possible, reduce defects generated during the cutting process, and improve product quality.

[0028] For example, assuming that after pre-scanning, it is found that the material density M1 on a certain path is 8 units, then: Anti-interference coefficient C1=8 / (8+10)≈0.44.

[0029] If the threshold T1 is set to 0.7, then the C1 value is lower than the threshold and this point does not need to be avoided.

[0030] Assume that on another path, the material density M1 is higher, reaching 20 units, then: The anti-interference coefficient C1=20 / (20+10)≈0.67, which is close to or exceeds the threshold T1.

[0031] The pre-scan path needs to be adjusted to avoid this high anti-interference coefficient area.

[0032] For the entire path, if the original path length L1 is 100 mm and the average deviation D1 between the adjusted path and the original path is 2 mm, then: The path similarity S2 = 1-(2 / 100) = 0.98, indicating that the new path is highly similar to the original path, but successfully avoids the high anti-interference coefficient area.

[0033] Step 2: Adjust cutting parameters based on the obtained evaluation information, plan dynamic avoidance routes based on parameter adjustments, avoid sensitive points, and apply multi-stage cooling strategies to maintain a stable cutting environment; details are as follows: According to the constructed influence area distribution map, determine the anti-interference coefficient C1 value of each cutting point. For each cutting point, calculate the adjusted cutting power Pw1, where Pw1=P0*(1-C1), P0 is the originally set cutting power; ensure a detailed understanding of the material properties to provide data support for subsequent parameter adjustments.

[0034] Based on the adjusted Pw1, the cutting speed Vc1 is re-evaluated to make Vc1=V0*sqrt(Pw1 / P0), where V0 is the originally set cutting speed; by adjusting the cutting power, the impact on the high anti-interference coefficient area is reduced, the sensitive points are protected, and the cutting quality is improved.

[0035] Using the adjusted speed Vc1 and power Pw1, plan a set of cutting parameters that minimize interference, while ensuring that all points on the cutting path meet the quality standard Qs1, that is, Qs1=min(Qi1), where Qi1 represents the quality score of each cutting point, which is determined by comparing the material properties before and after cutting. Ensure that the cutting parameters on the entire cutting path are optimized so that each cutting point can meet the highest quality standards and improve the yield rate of the final product.

[0036] Based on the adjusted cutting parameter set, all locations on the cutting path where the C1 value exceeds the threshold T1 are identified and marked as sensitive points. For each marked sensitive point, the avoidance distance R1 is calculated, where R1=2*sqrt(C1-T1); sensitive areas that require special treatment are accurately identified to provide a basis for subsequent avoidance path planning.

[0037] According to the avoidance distance R1, a temporary avoidance path segment Ls1 is constructed. The temporary avoidance path segment is based on the original path and is offset outward by a distance R1 at the sensitive point to form a new cutting trajectory. The avoidance requirements are quantified to ensure that the avoidance path can effectively avoid the area with high anti-interference coefficient and reduce potential risks.

[0038] Combined with the temporary avoidance path segment Ls1, the overall cutting path is optimized to ensure that the new path not only avoids all sensitive points, but also keeps the cutting efficiency E1 unchanged, that is, E1=L1 / (ΣLs1), where L1 is the original cutting path length and ΣLs1 represents the sum of all avoidance path segments. Plan a reasonable avoidance path to ensure safety and efficiency during the cutting process.

[0039] Based on the optimized cutting path, the estimated cutting time Tc1 of each cutting section is determined, and the position of the sensitive point is recorded. For each cutting section, the cooling requirement coefficient K1 is calculated, where K1=(Tc1 / 10)+C1; the cutting path is optimized to ensure that the cutting efficiency is not sacrificed while avoiding sensitive points and maintaining the production rate.

[0040] According to the cooling demand coefficient K1, a multi-stage cooling scheme is planned, including pre-cooling, main cooling and post-cooling. The cooling intensity Ipre1 of the pre-cooling stage is set to K1*0.5, the cooling intensity Imain1 of the main cooling stage is set to K1*1.2, and the cooling intensity Ipost1 of the post-cooling stage is set to K1*0.3. The specific cooling requirements of each cutting section are evaluated to ensure the reasonable allocation of cooling resources to avoid overheating damage. A phased cooling strategy is designed to ensure the stability of the cutting environment temperature and prevent material damage caused by temperature fluctuations.

[0041] Use the cooling intensity setting to dynamically adjust the coolant flow Q1, monitor the temperature change ΔT1 in real time during the cutting process, and adjust the flow through the formula Q1=Qbase1+ΔT1*K1. Qbase1 is the basic cooling flow to ensure a stable cutting environment. Monitor and adjust the coolant flow in real time to ensure that the temperature is always within a safe range during the cutting process to maintain high-quality cutting.

[0042] For example, assuming that the initial cutting power P0 is 100 W, the cutting speed V0 is 5 mm / s, and the quality score threshold T1 is 0.7: For a cut point with an anti-interference coefficient C1=0.6: Adjusted cutting power Pw1=100*(1-0.6)=40W; New cutting speed Vc1=5*sqrt(40 / 100)≈3.16mm / s; Calculate the quality score Qi1 of this point, assuming it is 0.9 (determined by comparing the material properties before and after cutting), then Qs1=min(Qi1)=0.9.

[0043] For a cut point with an anti-interference coefficient C1=0.8, exceeding the threshold T1: Avoidance distance R1=2*sqrt(0.8-0.7)≈0.63mm; A temporary avoidance path segment Ls1 is constructed to offset the path outward by 0.63 mm.

[0044] Assuming that in the entire cutting path, the total length of the avoidance path segment ΣLs1 is 10 mm, and the original path length L1 is 100 mm, the cutting efficiency E1=100 / 10=10; For a cutting section, the estimated cutting time Tc1 is 15 seconds, including the sensitive point C1=0.8: Cooling demand coefficient K1 = (15 / 10) + 0.8 = 2.3; The cooling intensity in the precooling stage is Ipre1=2.3*0.5=1.15; The cooling intensity of the main cooling stage is Imain1=2.3*1.2=2.76; The cooling intensity of the post-cooling stage is Ipost1=2.3*0.3=0.69; In the actual cutting process, assuming that the temperature change ΔT1 is 5°C and the basic cooling flow Qbase1 is 10L / min: Coolant flow rate Q1=10+5*2.3=21.5L / min.

[0045] Through the above steps and technical effects, the present invention realizes the precise adjustment of cutting parameters, the planning of dynamic avoidance paths and the application of multi-stage cooling strategies, thereby significantly improving the cutting quality and yield rate, reducing defects and ensuring the safety and stability of the cutting process.

[0046] Step 3: Monitor the temperature changes in real time during the cooling process and adjust the cutting speed. At the same time, implement adaptive corrections; the details are as follows: As the cooling process proceeds, a temperature sensor is used to monitor the temperature change ΔT1 at the cutting point in real time, and the temperature data within each time interval t1 is recorded; the temperature change information during the cutting process is obtained in real time to provide accurate data support for subsequent adjustments.

[0047] Based on the obtained ΔT1 value, the temperature fluctuation coefficient F1 is calculated, where F1=|ΔT1| / T01, and T01 is the initially set stable temperature; quantifying the degree of temperature fluctuation helps evaluate the impact of temperature changes on the cutting process and ensure that timely countermeasures are taken.

[0048] According to the temperature fluctuation coefficient F1, the cutting speed Vc1 is adjusted so that Vc1=Vbase1*(1-F1), where Vbase1 is the originally set cutting speed; the cutting speed is dynamically adjusted to cope with temperature fluctuations, reduce the impact of high temperature on materials and cutting quality, and protect cutting equipment.

[0049] Using the adjusted cutting speed Vc1, re-evaluate the time schedule of the entire cutting process to ensure that the total cutting time Ttotal1 does not exceed the preset maximum allowable time Tmax1, that is, Ttotal1=Σ(Vc1*Ls1)≤Tmax1. Ensure that after adjusting the speed, the overall cutting time still meets the production plan and does not affect the efficiency of the production line.

[0050] Based on the adjusted cutting speed Vc1 and the real-time temperature change ΔT1, the correction coefficient G1 of each cutting section is calculated, where G1=(Vc1 / Vbase1)*(1+ΔT1 / T01); combined with the speed adjustment and temperature change, the cutting parameters are further optimized to ensure the flexibility and adaptability of the cutting process.

[0051] According to the correction coefficient G1, the cutting parameter set is updated, and for each sensitive point position, the new cutting parameter Pnew1=Pold1*G1 is set, where Pold1 is the originally set cutting parameter; the cutting parameters of sensitive areas are accurately adjusted to ensure that these areas are specially protected and improve product quality.

[0052] Using the updated cutting parameters Pnew1, the avoidance strategy on the cutting path is replanned to ensure that the new path not only avoids the area with high anti-interference coefficient, but also adapts to the revised parameters to keep the cutting efficiency E1 unchanged, that is, E1=L1 / (ΣLs1); while ensuring the cutting quality, the path is optimized to avoid sensitive areas and maintain efficient cutting.

[0053] Combined with the optimization path and updated parameters, adaptive correction is implemented. By monitoring the dynamic feedback during the cutting process, parameters are continuously adjusted to ensure that the product quality Qs1 reaches the highest standard, that is, Qs1=max(Qi1). Continuous monitoring and adjustment ensure that the final product meets the highest quality requirements, improves the yield rate and customer satisfaction.

[0054] For example, it is assumed that the initially set stable temperature T01 is 25° C., the originally set cutting speed Vbase1 is 5 mm / s, and the maximum allowed time Tmax1 is 30 seconds; In a certain time interval t1, the temperature sensor records a temperature change ΔT1 of +5°C at the cutting point: Temperature fluctuation coefficient F1=|5| / 25=0.2; Adjust the cutting speed Vc1=5*(1-0.2)=4mm / s according to the temperature fluctuation coefficient; Assuming that the total length of the temporary avoidance path segment ΣLs1 is 10 mm and the original path length L1 is 100 mm, then: The total cutting time Ttotal1 = Σ(4*Ls1) = 4*10 = 40 mm / s. In order to ensure that Tmax1 is not exceeded, it is necessary to check whether the condition Ttotal1 ≤ Tmax1 (30 seconds) is met; if it is exceeded, the speed or other parameters need to be further adjusted.

[0055] For a specific cutting section, assuming that the adjusted cutting speed Vc1 is still 4 mm / s and the temperature change ΔT1 is +5°C; Correction coefficient G1=(4 / 5)*(1+5 / 25)=0.8*1.2=0.96; For a certain sensitive point, assume that the original cutting parameter Pold1 is 100W; New cutting parameters Pnew1=100*0.96=96W; Using the updated cutting parameter Pnew1, re-plan the cutting path to ensure that the new path avoids the high anti-interference coefficient area while keeping the cutting efficiency E1 unchanged, that is, E1=L1 / (ΣLs1)=100 / 10=10; Finally, by continuously monitoring the dynamic feedback during the cutting process, we can ensure that the product quality Qs1 reaches the highest standard, that is, Qs1=max(Qi1). Assuming that the highest value of the quality score Qi1 of all cutting points is 0.95, then Qs1=0.95.

[0056] Through the above steps and technical effects, the present invention realizes real-time monitoring of temperature changes and dynamic adjustment of cutting speed during the cooling process, while implementing adaptive correction, thereby significantly improving the cutting quality and yield rate, reducing defects, and ensuring the safety and stability of the cutting process.

[0057] Step 4: After cutting, perform a quality control check to verify product consistency; details are as follows: Based on the implemented adaptive correction, a preliminary visual inspection is performed on the finished cut products, the location and type of all visible defects are recorded, and the defect density Dd1 is calculated; surface defects are quickly identified through preliminary visual inspection to ensure that any problems that may affect product quality can be discovered and recorded in time.

[0058] Based on the defect density Dd1, the quality assessment formula Qa1=1-(Dd1 / Dmax1) is applied, where Dmax1 is the maximum acceptable defect density; quantifying product quality and providing an objective quality score to help determine whether further inspection or processing is needed.

[0059] For products whose quality score Qa1 is lower than the threshold Qt1, perform secondary precision measurement, record the defect area Af1, and update the total defect area At1=At1+Af1; conduct more detailed inspections on potential problem products to ensure that subtle but important defects are not missed, thereby improving the reliability of the final product.

[0060] Combined with the total defect area At1, it is finally determined whether the product has passed the quality control. If At1 is less than or equal to the preset maximum allowable defect area Am1, the product is considered qualified; otherwise, it is marked as unqualified. Ensure that each product meets strict quality standards, ensure the high quality and consistency of factory products, and enhance customer satisfaction.

[0061] For example, it is assumed that the maximum defect density Dmax1 is set to 0.5 defects / cm², the quality score threshold Qt1 is 0.8, and the maximum allowable defect area Am1 is 10 cm²: During the initial visual inspection, 3 visible defects were found on a semiconductor chip with a total area of ​​6cm². Assuming the chip area is 100cm², then: Defect density Dd1 = 3 / 100 = 0.03 defects / cm²; Application quality assessment formula Qa1=1-(0.03 / 0.5)=1-0.06=0.94; The quality score Qa1 is higher than the threshold Qt1 (0.8), so the product is preliminarily judged to be qualified.

[0062] Assume that another chip has more defects in the initial visual inspection, and the quality score Qa1 is 0.75, which is lower than the threshold Qt1 (0.8): A second precision measurement was performed to find additional microscopic defects with a total area of ​​4cm²; Update the total defect area At1=At1+4cm². Assuming that the previous defect area At1 is 2cm², the new total defect area At1=2+4=6cm².

[0063] Final determination of whether the product passes quality control: If the total defect area At1=6cm²≤Am1(10cm²), the product is considered qualified.

[0064] If the total defect area At1 exceeds Am1, the product is marked as unqualified.

[0065] Through the above steps and technical effects, the present invention realizes a comprehensive quality control inspection of semiconductor products after cutting, ensuring that each product is strictly evaluated and verified, thereby improving the consistency and reliability of product quality and meeting the requirements of high-precision manufacturing.

[0066] On the other hand, the present invention provides a semiconductor cutting integration optimization system, such as Figure 2 As shown, including: A material preparation and pre-scan evaluation module, for preparing semiconductor materials to be processed and setting an initial cutting path, and performing a pre-scan based on the initial cutting path to evaluate the characteristics of the affected area; The parameter adjustment and dynamic avoidance planning module is used to adjust the cutting parameters according to the acquired evaluation information, plan the dynamic avoidance route in combination with the parameter adjustment, avoid sensitive points, and apply the multi-stage cooling strategy to maintain the stability of the cutting environment; Real-time monitoring and adaptive correction module, used to monitor temperature changes in real time during the cooling process and adjust the cutting speed, and at the same time, implement adaptive correction; Quality control and product verification module for quality control checks after cutting to verify product consistency.

[0067] In addition, the above-mentioned material preparation and pre-scanning evaluation module, parameter adjustment and dynamic avoidance planning module, real-time monitoring and adaptive correction module, quality control and product verification module are also used to implement other steps of the above-mentioned semiconductor rib cutting integration optimization method during execution, which will not be described one by one here.

[0068] In summary, the present invention minimizes the impact on adjacent structures during the cutting process by dynamically adjusting cutting parameters and planning avoidance routes to avoid sensitive points; applies a multi-stage cooling strategy to maintain a stable cutting environment to prevent material damage caused by overheating; monitors temperature changes in real time during the cooling process and adaptively corrects the cutting speed to ensure the flexibility and accuracy of the cutting process; and finally verifies product consistency through strict quality control inspections. These measures work together to significantly improve cutting quality and yield, reduce defects, and achieve a more stable production process and higher semiconductor chip quality.

[0069] Finally, it should be noted that the above is only a preferred embodiment of the present invention and is not intended to limit the present invention. Although the present invention has been described in detail with reference to the aforementioned embodiments, it is still possible for those skilled in the art to modify the technical solutions described in the aforementioned embodiments or to make equivalent substitutions for some of the technical features therein. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principles of the present invention should be included in the protection scope of the present invention.

Claims

1. A semiconductor cutting rib integration optimization method, characterized in that: The following steps are involved: Step 1: preparing the semiconductor material to be processed and setting an initial cutting path, and performing a pre-scan based on the initial cutting path to evaluate the characteristics of the affected area; Step 2: Adjust cutting parameters based on the obtained evaluation information, plan dynamic avoidance routes based on parameter adjustments, avoid sensitive points, and apply multi-stage cooling strategies to maintain a stable cutting environment; Step 3: Monitor temperature changes in real time during the cooling process and adjust the cutting speed, while implementing adaptive corrections; Step 4: Perform a quality control check after cutting to verify product consistency.

2. The semiconductor cutting rib integration optimization method according to claim 1, characterized in that: The setting of the initial cutting path comprises: Determine the thickness D of the semiconductor material and measure the vertical distance H from the material surface to the predetermined cutting line; According to the obtained D and H values, calculate the initial cutting path offset O1, where O1=sqrt(D^2+H^2)-D; Based on the calculated offset O1, adjust the cutting starting point to keep a safe distance S1 from the edge of the material at the beginning of cutting: S1=O1+5mm; Use the set safety interval S1 to draw the final cutting path diagram.

3. The semiconductor cutting rib integration optimization method according to claim 2, characterized in that: Based on the initial cutting path, performing a pre-scan to evaluate the characteristics of the affected area includes: Pre-scan along the set initial cutting path at a constant speed V1, and record the material density M1 at each point on the path; According to the recorded material density M1, calculate the anti-interference coefficient C1 of each point, where C1=M1 / (M1+10); Based on the obtained anti-interference coefficient C1, a distribution map of the affected area is constructed, marking all locations where the C1 value exceeds the threshold T1; Using the marked influence area, the pre-scan path is adjusted to ensure that the new path avoids the high C1 value area while maintaining the maximum similarity S2 with the original path. Here S2 is measured by calculating the average deviation D1 between the two paths, that is, S2=1-(D1 / L1), where L1 represents the total length of the path.

4. The semiconductor cutting rib integration optimization method according to claim 3, characterized in that: The step of adjusting the cutting parameters according to the acquired evaluation information includes: According to the constructed influence area distribution map, the anti-interference coefficient C1 value of each cutting point is determined. For each cutting point, the adjusted cutting power Pw1 is calculated, where Pw1=P0*(1-C1), and P0 is the originally set cutting power; Based on the adjusted Pw1, re-evaluate the cutting speed Vc1, so that Vc1=V0*sqrt(Pw1 / P0), V0 is the originally set cutting speed; Using the adjusted speed Vc1 and power Pw1, a set of cutting parameters that minimize interference is planned, while ensuring that all points on the cutting path meet the quality standard Qs1, that is, Qs1=min(Qi1), where Qi1 represents the quality score of each cutting point, which is determined by comparing the material properties before and after cutting.

5. The semiconductor cutting rib integration optimization method according to claim 4, characterized in that: The combined parameter adjustment and planning of dynamic avoidance routes to avoid sensitive points include: Based on the adjusted cutting parameter set, identify all locations on the cutting path where the C1 value exceeds the threshold T1 and mark them as sensitive points. For each marked sensitive point, calculate the avoidance distance R1, where R1=2*sqrt(C1-T1); According to the avoidance distance R1, a temporary avoidance path segment Ls1 is constructed. The temporary avoidance path segment is based on the original path and is offset outward by a distance R1 at the sensitive point to form a new cutting trajectory; Combined with the temporary avoidance path segment Ls1, the overall cutting path is optimized to ensure that the new path not only avoids all sensitive points but also keeps the cutting efficiency E1 unchanged, that is, E1 = L1 / (ΣLs1), where L1 is the original cutting path length and ΣLs1 represents the sum of all avoidance path segments.

6. The semiconductor cutting rib integration optimization method according to claim 5, characterized in that: The multi-stage cooling strategy is applied to maintain a stable cutting environment, including: Based on the optimized cutting path, the estimated cutting time Tc1 of each cutting section is determined, and the position of the sensitive point is recorded. For each cutting section, the cooling requirement coefficient K1 is calculated, where K1=(Tc1 / 10)+C1; According to the obtained cooling demand coefficient K1, a multi-stage cooling scheme is planned, including pre-cooling, main cooling and post-cooling. The cooling intensity Ipre1 of the pre-cooling stage is set to K1*0.5, the cooling intensity Imain1 of the main cooling stage is set to K1*1.2, and the cooling intensity Ipost1 of the post-cooling stage is set to K1*0.

3. Use the cooling intensity setting to dynamically adjust the coolant flow Q1, monitor the temperature change ΔT1 in real time during the cutting process, and adjust the flow through the formula Q1=Qbase1+ΔT1*K1. Qbase1 is the basic cooling flow to ensure a stable cutting environment.

7. The semiconductor cutting rib integration optimization method according to claim 6, characterized in that: The real-time monitoring of temperature changes and adjustment of cutting speed during the cooling process include: As the cooling process proceeds, a temperature sensor is used to monitor the temperature change ΔT1 of the cutting point in real time, and the temperature data within each time interval t1 is recorded; Based on the obtained ΔT1 value, calculate the temperature fluctuation coefficient F1, where F1=|ΔT1| / T01, T01 is the initially set stable temperature; According to the temperature fluctuation coefficient F1, adjust the cutting speed Vc1 so that Vc1=Vbase1*(1-F1), where Vbase1 is the originally set cutting speed; Using the adjusted cutting speed Vc1, re-evaluate the timing of the entire cutting process to ensure that the total cutting time Ttotal1 does not exceed the preset maximum allowable time Tmax1, that is, Ttotal1=Σ(Vc1*Ls1)≤Tmax1.

8. The semiconductor cutting rib integration optimization method according to claim 7, characterized in that: The implementation of adaptive correction includes: Based on the adjusted cutting speed Vc1 and the real-time temperature change ΔT1, the correction coefficient G1 of each cutting section is calculated, where G1=(Vc1 / Vbase1)*(1+ΔT1 / T01); According to the correction coefficient G1, the cutting parameter set is updated, and the new cutting parameter Pnew1=Pold1*G1 is set for each sensitive point position, where Pold1 is the originally set cutting parameter; Using the updated cutting parameters Pnew1, re-plan the avoidance strategy on the cutting path to ensure that the new path not only avoids the high anti-interference coefficient area, but also adapts to the revised parameters to keep the cutting efficiency E1 unchanged, that is, E1=L1 / (ΣLs1); Combined with the optimization path and updated parameters, adaptive correction is implemented. By monitoring the dynamic feedback during the cutting process, parameters are continuously adjusted to ensure that the product quality Qs1 reaches the highest standard, that is, Qs1=max(Qi1).

9. The semiconductor cutting rib integration optimization method according to claim 8, characterized in that: After the cutting is completed, a quality control check is performed to verify product consistency, including: Based on the implemented adaptive corrections, perform a preliminary visual inspection of the finished cut products, record the location and type of all visible defects, and calculate the defect density Dd1; According to the defect density Dd1, the quality assessment formula Qa1=1-(Dd1 / Dmax1) is applied, where Dmax1 is the maximum acceptable defect density; For products whose quality score Qa1 is lower than the threshold Qt1, perform secondary precision measurement, record the defect area Af1, and update the total defect area At1=At1+Af1; Combined with the total defect area At1, it is finally determined whether the product has passed the quality control. If At1 is less than or equal to the preset maximum allowable defect area Am1, the product is considered qualified; otherwise, it is marked as unqualified.

10. A semiconductor cutting rib integration optimization system for implementing the method according to any one of claims 19, characterized in that: include: A material preparation and pre-scan evaluation module, for preparing semiconductor materials to be processed and setting an initial cutting path, and performing a pre-scan based on the initial cutting path to evaluate the characteristics of the affected area; The parameter adjustment and dynamic avoidance planning module is used to adjust the cutting parameters according to the acquired evaluation information, plan the dynamic avoidance route in combination with the parameter adjustment, avoid sensitive points, and apply the multi-stage cooling strategy to maintain the stability of the cutting environment; Real-time monitoring and adaptive correction module, used to monitor temperature changes in real time during the cooling process and adjust the cutting speed, and at the same time, implement adaptive correction; Quality control and product verification module for quality control checks after cutting to verify product consistency.

Images