A method for optimizing chemical mechanical polishing process parameters for wafers and a chemical mechanical polishing method.

Abstract

This invention relates to a method for optimizing chemical mechanical polishing (CMP) process parameters for wafers and a polishing method thereof. The method involves polishing a wafer using a CMP machine and measuring polishing result 1. At least one parameter of the polishing process is changed, and the wafer is polished again using the CMP machine, measuring polishing result 1'. Polishing result 1 and polishing result 1' are compared. If polishing result 1' is not better than polishing result 1, the parameters are changed further until polishing result 1' is better than polishing result 1. If polishing result 1' is better than polishing result 1, subsequent steps are performed directly. Based on this, other parameters in the polishing process are changed to obtain a polishing process with multiple optimized parameters. The above operations are repeated until a polishing process with all target parameters optimized is obtained. This invention systematically optimizes the parameters of the CMP process, improving the scientific and rational nature of the polishing process parameter settings and enhancing polishing quality.

Description

Technical Field

[0001] This invention relates to a method for optimizing chemical mechanical polishing (CMP) process parameters and a CMP method for wafers, and more particularly to a method for optimizing CMP process parameters and a CMP method for silicon carbide wafers, belonging to the field of semiconductor manufacturing technology. Background Technology

[0002] Silicon carbide (SiC) is a material with extremely high hardness, good thermal conductivity, high temperature resistance, and corrosion resistance, widely used in high-power semiconductors, optoelectronic devices, aerospace, and the automotive industry. Its superior properties make it an ideal material for fabricating high-performance electronic devices, especially under high temperature, high frequency, and high voltage environments. To achieve optimal performance in these devices, the surface of silicon carbide wafers must have extremely low roughness and high flatness, requiring advanced polishing techniques. Because the hardness of silicon carbide is very close to that of diamond, polishing it presents considerable challenges.

[0003] The key challenges of silicon carbide polishing include the following: First, the high hardness and wear resistance of silicon carbide make it difficult to remove the material in a short time using traditional mechanical polishing methods, and the polishing tools and materials suffer severe wear. Second, silicon carbide has very stable chemical inertness, making it difficult to effectively remove the material through chemical reactions during chemical polishing. Finally, silicon carbide typically has multiple crystal structures, such as 4H-SiC and 6H-SiC, which exhibit different polishing reactions and require special handling.

[0004] To overcome the aforementioned challenges, the main polishing techniques for silicon carbide currently include mechanical polishing, chemical mechanical polishing, ion beam polishing, magnetohydrodynamic polishing, and laser-assisted polishing. Mechanical polishing works by using abrasive particles on a polishing pad to physically grind the silicon carbide surface, thereby removing uneven material.

[0005] Mechanical polishing is a simple process with low equipment requirements and is easy to achieve for large areas. However, it is inefficient and prone to surface defects such as scratches and cracks. It is usually used as a rough polish and requires further fine processing to obtain a more precise surface.

[0006] Chemical mechanical polishing (CMP) works by using chemical agents in a polishing slurry to soften the material surface, which is then mechanically removed by abrasive particles, resulting in a more efficient polishing effect. It effectively reduces surface roughness, quickly removes material, and has good compatibility with different crystal structures to obtain high-quality surfaces. However, its process is complex, chemical polishing is costly, and the polishing time is relatively long.

[0007] Ion beam polishing works by bombarding the surface of silicon carbide with a high-energy ion beam to remove tiny uneven areas, and is mainly used for ultra-precision surface processing. It can achieve atomic-level flatness and produce extremely low surface roughness, but the equipment is very expensive, the processing speed is slow, and there may be modification or damage to the surface atomic layer.

[0008] Magnetohydrodynamic (MHD) polishing works by using abrasive particles contained in a magnetic fluid to polish materials under the control of a magnetic field. The shape and flow of the fluid can be precisely controlled by the magnetic field to achieve a uniform surface finish. It can precisely control the distribution of polishing pressure, reducing scratches and wear, and producing a high degree of surface smoothness after polishing. However, the equipment is expensive, the process is complex, and it requires high precision in the preparation of the polishing fluid and the control of the magnetic field.

[0009] Laser-assisted polishing works by heating the surface of silicon carbide with a laser, causing it to soften or melt locally, and then polishing it mechanically to improve material removal efficiency. It effectively removes material, and the laser heating softens the material, reducing the difficulty and time of polishing. However, it requires advanced equipment technology, and precise control of laser power parameters is crucial, as it may cause thermal damage to the material.

[0010] In summary, chemical mechanical polishing is currently the main development direction for silicon carbide polishing, especially in high-precision surface treatment. As a key fine polishing step after rough polishing, its processing method and polishing process parameters have a crucial impact on the surface of the wafer and the performance of the device.

[0011] While chemical mechanical polishing (CMP) can achieve highly smooth silicon carbide surfaces, its equipment process parameters are extremely complex (a common chemical polishing process can be found in CN102049734A), and the preparation of polishing slurries requires extensive experimentation to determine suitable formulations. Traditional polishing equipment process parameters are mostly set based on engineers' personal experience or parameters from other equipment manufacturers. Although relatively stable process parameters can be found within a certain range, problems such as uneven polished surfaces, low polishing efficiency, and large parameter errors still exist. Later, experimental design methods were adopted to ensure process stability, but the experimental processes and steps are often manually calculated, previous experimental experience is not effectively utilized, and there is no complete process parameter database, making it impossible to identify trends in various silicon carbide indicators. Furthermore, the performance and selectable parameter ranges of CMP machines from different manufacturers vary, so there are currently no universally applicable polishing parameters for different CMP machines. Summary of the Invention

[0012] In view of the shortcomings of the prior art, one objective of the present invention is to provide a method for optimizing the process parameters of chemical mechanical polishing for wafers; another objective of the present invention is to provide a chemical mechanical polishing method for wafers.

[0013] To solve the above-mentioned technical problems, the technical solution of the present invention is as follows:

[0014] A method for optimizing chemical mechanical polishing process parameters for wafers includes the following steps:

[0015] S1. Using a polishing process, the wafer is polished with a chemical mechanical polishing machine, and the polishing result is measured.

[0016] The polishing process includes at least two parameters;

[0017] S2. Change at least one of the parameters of the polishing process, polish the wafer using a chemical mechanical polishing machine, and measure the polishing result 1';

[0018] Compare polishing result 1 and polishing result 1';

[0019] S3. If polishing result 1' is not better than polishing result 1, then continue to change the parameters changed in S2 and repeat S2-S3 until polishing result 1' is better than polishing result 1. At this time, the polishing process corresponding to polishing result 1' is the polishing process after the relevant parameters have been optimized.

[0020] If polishing result 1' is better than polishing result 1, then the polishing process corresponding to polishing result 1' at this time is used as the polishing process after the relevant parameters are optimized, and S4 is performed directly;

[0021] S4. Based on the polishing process after the relevant parameters determined in S3 have been optimized, change at least one of the other parameters in the polishing process, repeat S2-S3, and obtain a polishing process after multiple parameters have been optimized.

[0022] S5. Repeat S4 until the polishing process with all target parameters optimized is obtained.

[0023] In each step, the chemical mechanical polishing machine used is the same or the same model, and the wafer used is the same type of wafer.

[0024] Furthermore, the parameters of the polishing process include several of the following: large disc rotation speed W, pressure head rotation speed w, pressure P, polishing fluid flow rate F, whether the polishing fluid is circulated, whether the polishing head is oscillating, and ambient temperature.

[0025] Further, S2, change the pressure head rotation speed w and pressure P in the polishing process, use a chemical mechanical polishing machine to polish the wafer, and measure the polishing result 1';

[0026] Compare polishing result 1 and polishing result 1';

[0027] S3. If polishing result 1' is not better than polishing result 1, then continue to change the pressure head rotation speed w and pressure P described in S2, and repeat S2-S3 until polishing result 1' is better than polishing result 1. At this time, the polishing process corresponding to polishing result 1' is the polishing process after the pressure head rotation speed w and pressure P are optimized.

[0028] If polishing result 1' is better than polishing result 1, then the polishing process corresponding to polishing result 1' at this time is taken as the polishing process after the pressure head speed w and pressure P are optimized, and S4 is directly performed;

[0029] S4. Based on the polishing process after the pressure head speed w and pressure P determined in S3 are optimized, change the large disc speed W in the polishing process, repeat S2-S3, and obtain the polishing process after the pressure head speed w, pressure P, and large disc speed W are optimized.

[0030] S5. Repeat S4 until the polishing process is optimized, with the pressure head rotation speed w, pressure P, large disc rotation speed W, and whether the polishing head oscillates.

[0031] Furthermore, the polishing results include one or more of the following indicators: the thickness difference of the wafer before and after polishing, the polishing rate (i.e., the rate of change of wafer thickness during polishing, typically measured in μm / h), the surface defect state of the polished surface, and the lifespan of the polishing slurry. Optionally, the surface defect state of the polished surface can be visually inspected under strong light. Optionally, the power of the strong light is ≥20W.

[0032] Furthermore, the wafer is a silicon carbide wafer, preferably a silicon carbide piezoelectric wafer. Optionally, the silicon carbide wafer is 6 inches in size.

[0033] A chemical mechanical polishing method for wafers, wherein the wafer is polished using a polishing process with optimized target parameters obtained by the optimization method described above.

[0034] The optimization method of the present invention can be used to systematically optimize the parameters of the polishing process used by a chemical mechanical polishing machine for polishing semiconductor wafers, thereby improving the scientificity and rationality of the polishing process parameter settings.

[0035] The optimization method of this invention can optimize the polishing process parameters of a chemical mechanical polishing (CMP) machine, helping to improve the precision surface quality of wafers and enhance the stability of the polishing process. The polishing process optimized using this method significantly improves polishing efficiency, reduces material loss, enhances the smoothness and uniformity of the polished surface, and significantly reduces surface defects. This optimization method is applicable to the optimization of polishing processes for different CMP machines and semiconductor materials of different specifications, demonstrating good versatility. Furthermore, the optimization method of this invention has a complete, simple, and logically sound process.

[0036] Compared to traditional mechanical polishing, chemical mechanical polishing (CMP) can achieve high-quality nanoscale surfaces on silicon carbide wafers through the synergistic effect of chemical polishing solutions and mechanical thinning. This invention optimizes polishing process parameters in a step-by-step manner, resulting in high efficiency and facilitating a more accurate assessment of the impact of relevant polishing process parameters on wafer surface quality. Attached Figure Description

[0037] Figure 1 This is a flowchart of an optimization method of the present invention.

[0038] Figure 2 The results are the flatness characterization test results of the polished silicon carbide wafer in S5 of Example 1.

[0039] Figure 3 This is a schematic diagram showing the dimensional measurement results of the polished silicon carbide wafer in Example 1.

[0040] Figure 4 This is a schematic diagram of the wafer in Example 1 during polishing on a chemical mechanical polishing machine.

[0041] In the diagram, 1-large disk, 2-polishing pad, 3-silicon carbide wafer, 4-carrier disk, 5-pressure head, 6-infusion tube, 7-polishing fluid. Detailed Implementation

[0042] The present invention will be described in detail below with reference to embodiments. It should be noted that, unless otherwise specified, the embodiments and features in the embodiments of the present invention can be combined with each other.

[0043] Example 1

[0044] See Figure 1 The method for optimizing chemical mechanical polishing process parameters for wafers in this embodiment includes the following steps:

[0045] S1. The wafer (silicon carbide wafer) was polished using a chemical mechanical polishing machine, and the polishing results were measured. See Table 2.

[0046] The assembly state of the wafer with the chemical mechanical polishing machine is as follows: Figure 4 As shown; the parameters of the polishing process include: the large disk rotation speed is set to 51 rpm, the pressure head rotation speed range is 42 rpm, the pressure setting range is 63 kg, the flow rate setting range is 1.2 L / min, the polishing slurry is circulated (acidic polishing slurry, self-circulating), the polishing head is oscillating, the wafer and the carrier disk are bonded by a coating machine, and the experimental temperature is set to 30℃.

[0047] The polishing solution contains 5 wt% sulfuric acid and 5 wt% potassium permanganate.

[0048] S2. Change the pressure head rotation speed w (to 46 rpm) and pressure P (to 66 kg) in the polishing process, and use a chemical mechanical polishing machine to polish the wafer. Measure the polishing results 1', see Table 2;

[0049] Compare polishing result 1 and polishing result 1';

[0050] The settable range of the relevant parameters is shown in Table 1.

[0051] Table 1

[0052]

[0053] S3. Polishing result 1' is better than polishing result 1. The polishing process corresponding to polishing result 1' at this time is the polishing process after the pressure head speed w and pressure P are optimized, and S4 is directly performed.

[0054] S4. Based on the polishing process after the pressure head speed w and pressure P are optimized as determined in S3, change the large disc speed W in the polishing process (change it to 57 rpm), repeat S2-S3, and obtain the polishing process after the pressure head speed w, pressure P, and large disc speed W are optimized.

[0055] S5. Based on the polishing process after the pressure head speed w, pressure P, and large disc speed W are optimized as determined in S4, change whether the polishing head oscillates (change it to not oscillate), repeat S2-S3, and obtain the polishing process after the pressure head speed w, pressure P, and large disc speed W are optimized; until the polishing process after the pressure head speed w, pressure P, large disc speed W, and whether the polishing head oscillates are all optimized is obtained.

[0056] Throughout all steps, the same chemical mechanical polishing (CMP) machine, a precision single-sided CMP machine from Shanghai Zhiling Semiconductor Co., Ltd., was used. The wafers used were 6-inch silicon carbide wafers; the polishing time for each step was 4 hours.

[0057] The polishing results for each step are shown in Table 2.

[0058] Table 2

[0059]

[0060] Here, PAD refers to the polishing pad. TV5 refers to the difference in wafer thickness before and after polishing. During polishing, scratches appearing on the wafer are considered the end of the polishing slurry's lifespan.

[0061] Table 3 shows the surface roughness characterization test results of the silicon carbide wafer after polishing in S5 (obtained by a third-party testing agency). It can be seen that the surface roughness of the silicon carbide wafer obtained by the final optimized process is 0.16±0.05nm, which is extremely high. This shows that the optimized polishing process of the present invention has practical reliability and effectiveness.

[0062] Table 3

[0063] Sample Name Test Project Test Results 6-inch silicon carbide wafer Surface roughness Ra 0.16±0.05nm

[0064] Figure 2 The figure shows the flatness characterization test results of the silicon carbide wafer after polishing in S5 (obtained by a third-party testing agency). As can be seen from the figure, the flatness of the wafer after polishing obtained by the final optimized process is good, with no high points or defects on the surface. This indicates that the surface quality of the silicon carbide wafer after polishing by this process is very good.

[0065] Figure 3 This is a schematic diagram showing the dimensional measurement results of a silicon carbide wafer. As can be seen from the diagram, the size of the silicon carbide wafer remains 6 inches after polishing.

[0066] The above embodiments should be understood as being used only to illustrate the present invention more clearly, and not to limit the scope of the present invention. After reading the present invention, any modifications of the present invention in various equivalent forms by those skilled in the art fall within the scope defined by the appended claims.

Claims

1. A method for optimizing chemical mechanical polishing process parameters for wafers, characterized in that, Includes the following steps: S1. Using a polishing process, the wafer is polished with a chemical mechanical polishing machine, and the polishing result is measured. The polishing process includes at least two parameters; S2. Change at least one of the parameters of the polishing process, polish the wafer using a chemical mechanical polishing machine, and measure the polishing result 1'; Compare polishing result 1 and polishing result 1'; S3. If polishing result 1' is not better than polishing result 1, then continue to change the parameters changed in S2 and repeat S2-S3 until polishing result 1' is better than polishing result 1. At this time, the polishing process corresponding to polishing result 1' is the polishing process after the relevant parameters have been optimized. If polishing result 1' is better than polishing result 1, then the polishing process corresponding to polishing result 1' at this time is used as the polishing process after the relevant parameters are optimized, and S4 is performed directly; S4. Based on the polishing process after the relevant parameters determined in S3 have been optimized, change at least one of the other parameters in the polishing process, repeat S2-S3, and obtain a polishing process after multiple parameters have been optimized. S5. Repeat S4 until the polishing process with all target parameters optimized is obtained. In each step, the chemical mechanical polishing machine used is the same or the same model, and the wafer used is the same type of wafer.

2. The optimization method according to claim 1, characterized in that, The parameters of the polishing process include several of the following: large disc rotation speed W, pressure head rotation speed w, pressure P, polishing fluid flow rate F, whether the polishing fluid is circulated, whether the polishing head is oscillating, and ambient temperature.

3. The optimization method according to claim 2, characterized in that, S2. Change the pressure head rotation speed w and pressure P in the polishing process, use a chemical mechanical polishing machine to polish the wafer, and measure the polishing result 1'; Compare polishing result 1 and polishing result 1'; S3. If polishing result 1' is not better than polishing result 1, then continue to change the pressure head rotation speed w and pressure P described in S2, and repeat S2-S3 until polishing result 1' is better than polishing result 1. At this time, the polishing process corresponding to polishing result 1' is the polishing process after the pressure head rotation speed w and pressure P are optimized. If polishing result 1' is better than polishing result 1, then the polishing process corresponding to polishing result 1' at this time is taken as the polishing process after the pressure head speed w and pressure P are optimized, and S4 is directly performed; S4. Based on the polishing process after the pressure head speed w and pressure P determined in S3 are optimized, change the large disc speed W in the polishing process, repeat S2-S3, and obtain the polishing process after the pressure head speed w, pressure P, and large disc speed W are optimized. S5. Repeat S4 until the polishing process is optimized, with the pressure head rotation speed w, pressure P, large disc rotation speed W, and whether the polishing head oscillates.

4. The optimization method according to any one of claims 1-3, characterized in that, Polishing results include one or more of the following indicators: thickness difference of wafer before and after polishing, polishing rate, surface defect state of polished surface, and lifetime of polishing slurry.

5. The optimization method according to any one of claims 1-3, characterized in that, The wafer is a silicon carbide wafer.

6. The optimization method according to any one of claims 1-3, characterized in that, The wafer is a silicon carbide wafer.

7. A chemical mechanical polishing method for wafers, characterized in that, The wafer is polished using a polishing process in which the target parameters are optimized by the optimization method described in any one of claims 1-6.

Images