Adjustment feedback method for improving flatness stability of silicon wafer

Abstract

The invention relates to an adjustment feedback method for improving flatness stability of a silicon wafer, which belongs to the technical field of silicon wafer processing, and comprises the following operation steps of: 1, measuring the silicon wafer through equipment before polishing, screening out the silicon wafer which does not meet the requirement, sequencing the thickness, and then adopting a double-sided polishing mode; secondly, slight fluctuation of the polishing solution temperature, the polishing solution Ph, the polishing solution flow, the removal rate and the fixed disc cooling temperature needs to be controlled in the polishing process. And 3, measurement feedback adjustment is carried out, and flatness horizontal fluctuation and silicon wafer thickness morphology are monitored and adjusted. And 4, when the silicon wafer is processed and flatness measurement is carried out, the equipment is adjusted by setting a target value. And 5, through software analysis, thickness and morphology monitoring is carried out on the measured silicon wafer, and it is ensured that the thickness and morphology of the processed silicon wafer are consistent. The method has the characteristics of convenient operation and good operation stability. The problem that the flatness fluctuation stability is poor in the machining process is solved.

Description

Technical Field

[0001] The present invention relates to the technical field of silicon wafer processing, and in particular to an adjustment feedback method for improving the flatness stability of a silicon wafer. Background Art

[0002] With the continuous development of information technology, the integration of chips is getting higher and higher, and the requirements for the process are becoming more and more stringent; the chip process can only develop in a narrower direction, so that more devices can be integrated while reducing power consumption; this places extremely strict requirements on the flatness of the photolithography process, and the slightest fluctuation will cause defects in the processing process. Chemical mechanical polishing uses a chemical solution mixed with extremely small abrasive particles to react chemically with the processing surface to change the chemical bonds on its surface to generate products that are easily removed mechanically, and then remove the chemical reactants through mechanical friction to obtain an ultra-smooth and damage-free flat surface. However, since this process uses a combination of chemical and mechanical effects for polishing, there are defects such as many process parameters and unstable processing in the polishing process.

[0003] It is difficult to ensure complete consistency during the polishing process of silicon wafers, and they will fluctuate within a certain range. How to solve the fluctuations in processing and improve the consistency of silicon wafer flatness is crucial to improving device quality. The current means of improving flatness mainly involve the transformation of the polishing equipment itself, such as precise control systems, electrical control, temperature control, etc., the selection and use of Slurry (polishing liquid), and the optimization of the repair process. However, the above methods are costly, complex to control, and have large fluctuations in flatness levels. Therefore, it is necessary to find a method to improve the stability of wafer polishing flatness levels to solve the above technical problems in the prior art. Summary of the invention

[0004] The present invention mainly solves the deficiencies in the prior art and provides an adjustment feedback method for improving the stability of silicon wafer flatness, which has the characteristics of convenient operation and good running stability. The problem of poor stability of flatness fluctuation during processing is solved. The stability of the silicon wafer flatness process is ensured by checking the warpage and thickness of the silicon wafer before polishing, monitoring important parameters of the processing process, and real-time monitoring of the flatness level after polishing.

[0005] The above technical problems of the present invention are mainly solved by the following technical solutions:

[0006] An adjustment feedback method for improving the flatness stability of a silicon wafer comprises the following steps:

[0007] Step 1: Warp control and thickness control before polishing. Before polishing, the silicon wafers are measured by equipment to screen out silicon wafers that do not meet the requirements and sort the thickness to avoid fluctuations in the flatness level after polishing. Then double-sided polishing is used.

[0008] During polishing, a chemical solution mixed with extremely small abrasive particles is used to react chemically with the processed surface to change the chemical bonds on the surface to generate products that are easily removed mechanically. The chemical reactants are then removed by mechanical friction to obtain an ultra-smooth, damage-free, flat surface.

[0009] Step 2: During the polishing process, it is necessary to control slight fluctuations in the polishing liquid temperature, polishing liquid Ph, polishing liquid flow, removal rate and platen cooling temperature.

[0010] Step 3: Perform measurement feedback adjustment to monitor and adjust the flatness level fluctuation and silicon wafer thickness morphology.

[0011] Step 4: When the wafer is processed and flatness is measured, the equipment is adjusted by setting the target value.

[0012] Step 5: Through software analysis, the thickness and morphology of the measured silicon wafer are monitored to ensure that the thickness and morphology of the processed silicon wafer are consistent.

[0013] Preferably, the warping of the silicon wafer has a great influence on the flatness, especially the morphology after wire cutting; the average value of the warping deformation after wire cutting is required to be less than 8um.

[0014] As a preference, the requirements for the line cutting edge are: V-shaped edge, the edge angle is greater than 90°; L-shaped edge, the edge angle is greater than 45°.

[0015] As a preferred processing method, 15 silicon wafers are processed simultaneously in each batch, and the processing removal amount is controlled by laser thickness measurement. If the thickness uniformity of the 15 silicon wafers is Δ>0.8μm, there is a large fluctuation in the flatness of the 15 silicon wafers.

[0016] As a preference, the FDC system is used to monitor and alarm the parameters of the polishing liquid temperature, polishing liquid Ph, polishing liquid flow, removal rate and platen cooling temperature, with the polishing liquid temperature fluctuation being less than 0.5°C, PH being less than 0.2, flow being less than 0.4L / min, removal rate being greater than 0.5μm / min, and platen cooling temperature being Δ less than 0.2°C. This ensures the stability of the equipment operation during the processing and the stability of the flatness level of the silicon wafer.

[0017] Preferably, when the TTV measurement changes by more than 40nm compared to the previous batch, the machine will make appropriate corrections. If the change is +40μm, the corresponding method is to reduce the target thickness by 0.1μm, otherwise add 0.1μm.

[0018] As a preferred method, the silicon wafer is divided into different radial regions with R1 = 80, R2 = 100, and R3 = 140. Each region is divided into four quadrants, and the thickness of each quadrant is calculated separately; when the thickness morphology has a large deviation, the regional thickness value will fluctuate.

[0019] As a preference, radial calculation must satisfy ATR1>ATR2>ATR3>0; ATR1 is AVTHKR1, the average thickness of the R1 circle; ATR2 is the average thickness within the R2-R1 graph, and ATR3ATR3 is the average thickness within the R3-R2 graph.

[0020] As a preferred option, regional calculations must satisfy X1 is equal to the average thickness of quadrant A, X2 is the average thickness of quadrant B, and so on. u is the average thickness of the entire silicon wafer. When the thickness morphology meets the above measurement, the processing morphology is stable. When it does not meet the requirements, it indicates that the thickness morphology has a large deviation and the parameters need to be adjusted.

[0021] The present invention can achieve the following effects:

[0022] The present invention provides an adjustment feedback method for improving the flatness stability of silicon wafers. Compared with the prior art, the method has the characteristics of convenient operation and good running stability. The problem of poor flatness fluctuation stability during processing is solved. The stability of the silicon wafer flatness process is ensured by checking the warpage and thickness of the silicon wafer before polishing, monitoring important parameters of the processing process, and real-time monitoring of the flatness level after polishing. Through the feedback adjustment method, the TTV level reaches below 100nm and the standard deviation is less than 3. DETAILED DESCRIPTION

[0023] The technical solution of the invention is further specifically described below through embodiments.

[0024] Embodiment: An adjustment feedback method for improving the flatness stability of a silicon wafer comprises the following steps:

[0025] Step 1: Warp control and thickness control before polishing. Before polishing, the silicon wafers are measured by equipment to screen out silicon wafers that do not meet the requirements and sort the thicknesses. Then double-sided polishing is used.

[0026] The warpage of silicon wafer has a great influence on flatness, especially the morphology after wire cutting. The average warpage deformation after wire cutting is required to be less than 8um. At the same time, the requirements for wire cutting blades are: V-shaped blades, blade angle greater than 90°; L-shaped blades, blade angle greater than 45°.

[0027] Processing method: 15 silicon wafers are processed at the same time in each batch, and the processing removal amount is controlled by laser thickness measurement. If the thickness uniformity of 15 silicon wafers is Δ>0.8μm, there is a large fluctuation in the flatness of the 15 silicon wafers.

[0028] Step 2: During the polishing process, it is necessary to control the slight fluctuations of the polishing liquid temperature, polishing liquid Ph, polishing liquid flow, removal rate and platen cooling temperature. The FDC system monitors and alarms the parameters of polishing liquid temperature, polishing liquid Ph, polishing liquid flow, removal rate and platen cooling temperature. The polishing liquid temperature fluctuation is less than 0.5℃, PH is less than 0.2, flow is less than 0.4L / min, removal rate is greater than 0.5μm / min, and platen cooling temperature Δ is less than 0.2℃.

[0029] Step 3: Perform measurement feedback adjustment to monitor and adjust the flatness level fluctuation and silicon wafer thickness morphology.

[0030] Step 4: When the wafer is processed and the flatness is measured, the equipment is adjusted by setting the target value. When the TTV measurement changes by more than 40nm compared to the previous batch, the machine will make appropriate corrections. If the change is +40μm, the corresponding method is to subtract 0.1μm from the target thickness, otherwise add 0.1μm.

[0031] Step 5: Through software analysis, the thickness and morphology of the measured silicon wafer are monitored to ensure that the thickness and morphology of the processed silicon wafer are consistent. The silicon wafer is divided into different radial regions with R1=80, R2=100, and R3=140. Each region is divided into four quadrants, and the thickness of each quadrant is calculated separately; when the thickness morphology has a large deviation, the regional thickness value will fluctuate. Monitoring the thickness morphology ensures the stability of the flatness level.

[0032] For radial calculation, ATR1>ATR2>ATR3>0 must be satisfied; ATR1 is AVTHKR1, the average thickness of the R1 circle; ATR2 is the average thickness within the R2-R1 graph, and ATR3ATR3 is the average thickness within the R3-R2 graph.

[0033] Regional calculation, must meet X1 is equal to the average thickness of quadrant A, X2 is the average thickness of quadrant B, and so on. u is the average thickness of the entire silicon wafer. When the thickness morphology meets the above measurement, the processing morphology is stable. When it does not meet the requirements, it indicates that the thickness morphology has a large deviation and the parameters need to be adjusted.

[0034] In summary, the adjustment feedback method for improving the flatness stability of silicon wafers has the characteristics of convenient operation and good operation stability. It solves the problem of poor flatness fluctuation stability during processing. The stability of the silicon wafer flatness process is ensured by checking the warpage and thickness of the silicon wafer before polishing, monitoring important parameters of the processing process, and real-time monitoring of the flatness level after polishing. Through the feedback adjustment method, the TTV level reaches below 100nm and the standard deviation is less than 3.

[0035] The above description is only a specific embodiment of the present invention, but the structural features of the present invention are not limited thereto. Any changes or modifications made by any technician in the field of the present invention are included in the patent scope of the present invention.

Claims

1. A feedback adjustment method for improving the flatness stability of a silicon wafer, characterized in that The steps are as follows: Step 1: Warpage control and thickness control before polishing. Before polishing, the silicon wafers are measured by equipment, and the silicon wafers that do not meet the requirements are screened out and sorted by thickness. Then double-sided polishing is adopted; Step 2: During the polishing process, it is necessary to control the slight fluctuations of the polishing liquid temperature, polishing liquid Ph, polishing liquid flow rate, removal rate and platen cooling temperature; Step 3: Perform measurement feedback adjustment to monitor and adjust the flatness level fluctuation and silicon wafer thickness morphology; Step 4: When the wafer is processed and the flatness is measured, the equipment is adjusted by setting the target value; Step 5: Monitor the thickness and morphology of the measured silicon wafer through software analysis to ensure that the thickness and morphology of the processed silicon wafer are consistent.

2. The adjustment feedback method for improving the flatness stability of a silicon wafer according to claim 1, characterized in that: The warpage of silicon wafers has a great influence on flatness, especially the morphology after wire cutting; the average warpage deformation after wire cutting is required to be less than 8um.

3. The adjustment feedback method for improving the flatness stability of a silicon wafer according to claim 2, characterized in that: The requirements for the cutting edge are as follows: for V-shaped cutting edge, the cutting edge angle is greater than 90°; for L-shaped cutting edge, the cutting edge angle is greater than 45°.

4. The adjustment feedback method for improving the flatness stability of a silicon wafer according to claim 1, characterized in that: In the processing method, 15 silicon wafers are processed simultaneously in each batch, and the processing removal amount is controlled by laser thickness measurement; if the thickness uniformity of the 15 silicon wafers is Δ>0.8μm, there is a large fluctuation in the flatness of the 15 silicon wafers.

5. The adjustment feedback method for improving the flatness stability of a silicon wafer according to claim 1, characterized in that: The FDC system monitors and alarms the parameters of polishing liquid temperature, polishing liquid Ph, polishing liquid flow, removal rate and platen cooling temperature. The polishing liquid temperature fluctuation is <0.5℃, PH<0.2, flow rate<0.4L / min, removal rate>0.5μm / min, and platen cooling temperatureΔ<0.2℃.

6. The adjustment feedback method for improving the flatness stability of a silicon wafer according to claim 1, characterized in that: When the TTV measurement changes by more than 40nm compared to the previous batch, the machine will make appropriate corrections; if the change is +40μm, the corresponding method is to subtract 0.1μm from the target thickness, otherwise add 0.1μm.

7. The adjustment feedback method for improving the flatness stability of a silicon wafer according to claim 1, characterized in that: The silicon wafer is divided into different radial regions with distributions of R1=80, R2=100, and R3=140; each region is divided into four quadrants, and the thickness of each quadrant is calculated separately; when there is a large deviation in the thickness morphology, the regional thickness value will fluctuate.

8. The adjustment feedback method for improving the flatness stability of a silicon wafer according to claim 7, characterized in that: For radial calculation, ATR1>ATR2>ATR3>0 must be satisfied; ATR1 is AVTHKR1, the average thickness of the R1 circle; ATR2 is the average thickness within the R2-R1 graph, and ATR3 is the average thickness within the R3-R2 graph.

9. The adjustment feedback method for improving the flatness stability of a silicon wafer according to claim 7, characterized in that: Regional calculation, must meet X1 is equal to the average thickness of quadrant A, X2 is the average thickness of quadrant B, and so on. u is the average thickness of the entire silicon wafer. When the thickness morphology meets the above measurement, the processing morphology is stable. When it does not meet the requirements, it indicates that the thickness morphology has a large deviation and the parameters need to be adjusted.

Images