Method for improving thermal warping of copper-clad ceramic carrier plate

Abstract

The invention discloses a method for improving thermal warping of a copper-clad ceramic carrier plate, and relates to the technical field of semiconductor devices. The residual copper rate is adjusted by changing the copper thickness of the pattern surface or the non-pattern surface, thermal expansion of the two surfaces is balanced, and therefore thermal warping is improved; the method specifically comprises the following steps: under the condition that the residual copper rate of a pattern surface is smaller than that of a non-pattern surface, reducing the copper thickness of the non-pattern surface; under the condition that the residual copper rate of the pattern surface is greater than the residual copper rate of the non-pattern surface, reducing the copper thickness of the pattern surface; and according to the optimal copper thickness, copper on the pattern surface or the non-pattern surface is thinned in a chemical thinning and mechanical grinding mode, then annealing is conducted, and a final product is obtained. According to the product prepared by the invention, a test chart obtained in a subsequent experiment shows that the thermal warping change is effectively improved.

Description

Technical Field

[0001] The invention relates to the technical field of semiconductor devices, in particular to a method for improving thermal warping of a copper-clad ceramic carrier. Background Art

[0002] With the rapid development of the automotive power module industry, copper-clad ceramic substrates (substrates for short) are increasingly used in power modules. When packaging power modules, a variety of components need to be soldered on the substrate, and the matching of the substrate surface and the warpage of the components is particularly important. Currently, customers not only have requirements for the warpage value before packaging, but also for the warpage after packaging (thermal warpage): This is because the substrate is a composite structure of ceramic and copper, and is affected by the different thermal expansion coefficients of the two materials, so the substrate itself has a certain warpage; during the welding process, the substrate is continuously heated, and its stress distribution changes accordingly, and the warpage will also change accordingly. The existence and changes of substrate warpage may cause the substrate surface to not match the components, resulting in defects such as ceramic cracking. Summary of the invention

[0003] The object of the present invention is to provide a method for improving the thermal warpage of a copper-clad ceramic carrier to solve the problems raised in the prior art.

[0004] To achieve the above purpose, the present invention balances the thermal expansion of the graphic surface and the non-graphic surface by adjusting the difference in residual copper rate on both sides, thereby improving thermal warping; wherein, the calculation formula of the residual copper rate is: residual copper rate = area ratio of copper on the product surface × copper layer thickness.

[0005] Since the copper area ratio is a fixed parameter on the drawing, although the copper surface size also has a certain tolerance, the copper surface needs to be mounted with chips, devices, wire bonding and other operations, so changing the copper surface size involves a large amount of work, while the impact of copper thickness is relatively small. Therefore, this solution only adjusts the residual copper rate by changing the copper thickness of the graphic surface or non-graphic surface. Specifically, when the residual copper rate of the graphic surface is less than that of the non-graphic surface, the thickness of the copper layer on the graphic surface is guaranteed to remain unchanged, and the non-graphic surface is used as a pre-thinning surface to reduce the thickness of the copper layer; when the residual copper rate of the graphic surface is greater than that of the non-graphic surface, the thickness of the copper layer on the non-graphic surface is guaranteed to remain unchanged, and the graphic surface is used as a pre-thinning surface to reduce the thickness of the copper layer. Generally speaking, the residual copper rate of the non-graphic surface is greater than that of the graphic surface, so normally only the copper thickness of the non-graphic surface is reduced.

[0006] (1) When the residual copper rate of the graphic surface is less than that of the non-graphic surface, the ideal copper layer thickness of the non-graphic surface is H1 = residual copper rate of the graphic surface / residual copper rate of the non-graphic surface × copper layer thickness of the graphic surface;

[0007] (2) When the residual copper rate of the graphic surface is greater than that of the non-graphic surface, the ideal copper layer thickness of the graphic surface H2 = residual copper rate of the non-graphic surface ÷ residual copper rate of the graphic surface × copper layer thickness of the non-graphic surface.

[0008] The specific steps are as follows:

[0009] Step 1: Compare the residual copper rate of the graphic surface with that of the non-graphic surface, select the pre-thinning surface, calculate the optimal copper thickness of the pre-thinning surface, and subtract the calculated optimal copper thickness from the pre-thinning surface copper thickness in the drawing to obtain the value X; if X is a negative value, select a thicker copper sheet as the pre-thinning surface for sintering; if X is a positive value, the product is sintered normally;

[0010] Step 2: Take the product sintered in step 1, cover the non-pre-thinning surface with a photosensitive film, perform UV exposure and curing, calculate the thinning amount and thin the pre-thinning surface, remove the photosensitive film by stripping, and then anneal the product.

[0011] Preferably, the thinning can be carried out by chemical thinning, mechanical grinding or a combination of both;

[0012] The calculation formula of the thinning amount is: Thinning amount Y = actual copper thickness of the pre-thinning surface (not copper thickness on the drawing) - optimal copper thickness

[0013] When the thinning amount is above 0.02mm, the thinning scheme adopts chemical thinning and mechanical grinding together; when the thinning amount is between 0-0.02mm, thinning is directly carried out through one or more mechanical grindings.

[0014] Preferably, the chemical solution used for chemical thinning includes hydrochloric acid, sodium chlorate and copper ions, wherein the equivalent concentration of hydrochloric acid is 1.7±0.2N, the conductivity of sodium chlorate is 15-60, and the specific gravity of copper ions is 1.28-1.30.

[0015] Preferably, mechanical grinding uses a ceramic brush roller; the purpose is to prevent uneven copper thickness that may be caused by chemical thinning; mechanical grinding reduces the copper thickness by 0-10 μm at a time.

[0016] Preferably, the annealing temperature in step 2 is set to 330±10°C and kept warm for 15 minutes; the purpose of annealing is to release stress in advance. After the above-mentioned thinning operation, the residual copper rate of the graphic surface and the non-graphic surface is ensured to be the same, and the warpage value of the product after stress release is greatly reduced; and after heating, the warpage changes caused by thermal expansion of the copper on the front and back sides of the product can offset each other, and the warpage change amount will become extremely low when heated again.

[0017] Compared with the prior art, the present invention has the following beneficial effects:

[0018] The present invention balances the thermal expansion of the graphic surface and the non-graphic surface by adjusting the difference in residual copper rate on both sides, thereby improving thermal warpage; the residual copper rate is modified by adjusting the copper thickness, which has little effect on the product; the residual copper rates on both sides are balanced, and then annealing is performed to release the residual stress, and the effect of improving thermal warpage is achieved under the dual effect; wherein, chemical thinning and mechanical grinding are used to reduce the copper thickness, thereby reducing the thickness error during thinning and improving the product precision; the following effects are achieved in application:

[0019] 1. The substrate has less warping at room temperature;

[0020] 2. There is no significant change in the warpage value during the packaging process;

[0021] 3. Does not affect other performance during packaging;

[0022] 4. Greatly reduce the risk of cracks during packaging;

[0023] 5. Compatible with different product specifications. BRIEF DESCRIPTION OF THE DRAWINGS

[0024] Figure 1 This is the thermal warpage data diagram of product a1;

[0025] Figure 2 This is the thermal warpage data diagram of product a2;

[0026] Figure 3 This is the thermal warpage data diagram of product b1;

[0027] Figure 4 This is the thermal warpage data diagram of product b2;

[0028] Figure 5 This is the thermal warpage data diagram of product c1;

[0029] Figure 6 This is the thermal warpage data chart of product c2. DETAILED DESCRIPTION

[0030] The following will be combined with the 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, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0031] Embodiment 1:

[0032] Select copper-ceramic combination A, in which the residual copper rate of the graphic surface is 78.15% and the copper thickness is 0.3mm; the residual copper rate of the non-graphic surface is 83.9% and the copper thickness is 0.3mm; the residual copper rate of the graphic surface is less than that of the non-graphic surface, so the copper thickness of the non-graphic surface is reduced;

[0033] According to the optimal copper thickness = 78.15% ÷ (83.9% ÷ 0.3mm), the optimal copper thickness is subtracted from the non-graphic copper thickness of 0.3mm to obtain a positive value. The copper is directly coated and sintered at 1080°C and an oxygen content of 100ppm. After sintering, the graphic surface is covered with a photosensitive film, and then ultraviolet light exposure and curing are performed. The non-graphic surface is thinned, and the thinning amount Y = 0.3mm-optimal copper thickness, between 0-0.02mm, and mechanical grinding is selected. A is thinned by 20μm with a ceramic brush roller, and then the above-mentioned photosensitive film is removed by film stripping. A is annealed and placed in a nitrogen oven. The annealing temperature is set to 330°C and kept warm for 15 minutes to obtain product a1.

[0034] Embodiment 2:

[0035] Select copper-ceramic combination B, in which the residual copper rate of the graphic surface is 79.33% and the copper thickness is 0.3mm; the residual copper rate of the non-graphic surface is 86.17% and the copper thickness is 0.25mm; the residual copper rate of the graphic surface is less than that of the non-graphic surface, so the copper thickness of the non-graphic surface is reduced;

[0036] According to the optimal copper thickness = 79.33% ÷ (86.17% ÷ 0.3mm), the optimal copper thickness is subtracted from the non-graphic surface copper thickness of 0.25mm, and X is obtained as a negative value. After replacing the 0.25mm copper on the non-graphic surface with 0.3mm copper, it is directly copper-coated and sintered at 1080°C and an oxygen amount of 100ppm; after sintering, the graphic surface is covered with a photosensitive film, and ultraviolet light exposure and curing are performed, and the non-graphic surface is thinned, and the thinning amount Y = 0.3mm-optimal copper thickness. If it is above 0.02mm, chemical thinning of 15μm is selected, and then thinning of 9μm with a ceramic brush roller is performed, and the above-mentioned photosensitive film is removed by film stripping, and then A is annealed, placed in a nitrogen oven, and the annealing temperature is set to 330°C, and kept warm for 15min to obtain product b1.

[0037] Among them, the chemical liquid used for chemical thinning includes hydrochloric acid, sodium chlorate and copper ions, among which the equivalent concentration of hydrochloric acid is 1.8N, the conductivity of sodium chlorate is 35, and the specific gravity of copper ions is 1.29.

[0038] Embodiment 3:

[0039] Select copper-ceramic combination C, in which the residual copper rate of the graphic surface is 88.06% and the copper thickness is 0.3mm; the residual copper rate of the non-graphic surface is 87.97% and the copper thickness is 0.3mm; the residual copper rate of the graphic surface is greater than that of the non-graphic surface, so the copper thickness of the graphic surface is reduced;

[0040] According to the optimal copper thickness = 87.97% ÷ (88.06% ÷ 0.3mm), the optimal copper thickness is subtracted from the copper thickness of the graphic surface 0.3mm to obtain a positive value. The copper is directly coated and sintered at 1080°C and an oxygen content of 100ppm. After sintering, the non-graphic surface is covered with a photosensitive film, and then ultraviolet light exposure and curing are performed. The graphic surface is thinned, and the thinning amount Y = 0.3mm-optimal copper thickness, between 0-0.02mm, and mechanical grinding is selected. After thinning by 5μm, the above-mentioned photosensitive film is removed by film stripping, and then A is annealed, placed in a nitrogen oven, and the annealing temperature is set to 330°C. The heat preservation is carried out for 15 minutes to obtain product c1.

[0041] Comparative Example 1: In the control experiment of Example 1, only the copper-ceramic combination A was sintered normally;

[0042] A copper-ceramic combination A was selected, in which the residual copper rate of the graphic surface of A was 78.15% and the copper thickness was 0.3 mm; the residual copper rate of the non-graphic surface was 83.9% and the copper thickness was 0.3 mm; direct copper coating and sintering was performed at 1080°C and an oxygen content of 100 ppm to obtain product a2.

[0043] Comparative Example 2: In the control experiment of Example 2, only the copper-ceramic combination B was sintered normally;

[0044] A copper-ceramic combination B was selected, in which the residual copper rate of the graphic surface of B was 79.33% and the copper thickness was 0.3 mm; the residual copper rate of the non-graphic surface was 86.17% and the copper thickness was 0.25 mm; direct copper coating and sintering was performed at 1080°C and an oxygen content of 100 ppm to obtain product b2.

[0045] Comparative Example 3: In the control experiment of Example 3, only the copper-ceramic combination C was sintered normally;

[0046] A copper-ceramic combination C was selected, in which the residual copper rate of the graphic surface of C was 88.06% and the copper thickness was 0.3 mm; the residual copper rate of the non-graphic surface was 87.97% and the copper thickness was 0.3 mm; direct copper coating and sintering was performed at 1080°C and an oxygen content of 100 ppm to obtain product c2.

[0047] The copper-ceramic combination specifications and treatment schemes of the above Examples 1-3 and Comparative Examples 1-3 are tabulated and organized as shown in Table 1:

[0048] Table 1

[0049]

[0050] Detection test

[0051] Thermal warpage test was conducted on products a1, a2, b1, b2, c1, and c2 to simulate reflow soldering conditions: RT-50℃-100℃-150℃-200℃-230℃-260℃-290℃-260℃-230℃-200℃-150℃-100℃-50℃, and the real-time warpage value corresponding to the real-time temperature in each temperature range was tested.

[0052] The methods for evaluating thermal warpage are as follows:

[0053] 1. Warping direction: When the temperature is above 200℃, the warping direction is positive, or the warping value is infinitely close to 0; if this requirement is not met, solder turbulence may occur during welding.

[0054] 2. The smaller the change in the warpage value during the thermal hysteresis process (especially the initial and final values), the better. The smaller the change in the warpage value, the larger the stress area of ​​the product when it is subjected to stress, and the less damage the product will suffer.

[0055] Conclusion: The test data of products a1, a2, b1, b2, c1, c2 prepared in Examples 1-3 and Comparative Examples 1-3 are shown in Figure 1-6 ;contrast Figure 1 and Figure 2 , Figure 3 and Figure 4 , Figure 5 and Figure 6 It can be seen from these three groups that the thinning and annealing method adopted in the present invention effectively improves the thermal warping phenomenon of copper porcelain.

[0056] It will be apparent to those skilled in the art that the invention is not limited to the details of the exemplary embodiments described above and that the invention can be implemented in other specific forms without departing from the spirit or essential features of the invention. Therefore, the embodiments should be considered exemplary and non-limiting in all respects, and the scope of the invention is defined by the appended claims rather than the foregoing description, and it is intended that all variations falling within the meaning and scope of the equivalent elements of the claims be included in the invention. Any reference numeral in a claim should not be considered as limiting the claim to which it relates.

Claims

1. A method for improving thermal warpage of a copper-clad ceramic substrate, characterized in that: The residual copper rate is adjusted by thinning the copper layer thickness of the graphic surface or the non-graphic surface to improve the thermal warpage; specifically, when the residual copper rate of the graphic surface is less than that of the non-graphic surface, the copper layer thickness of the graphic surface is kept unchanged, and the non-graphic surface is used as a pre-thinning surface to thin the copper layer thickness; when the residual copper rate of the graphic surface is greater than that of the non-graphic surface, the copper layer thickness of the non-graphic surface is kept unchanged, and the graphic surface is used as a pre-thinning surface to thin the copper layer thickness.

2. A method for improving thermal warpage of a copper-clad ceramic substrate according to claim 1, characterized in that: The calculation formula of the residual copper rate is: residual copper rate = area ratio of copper on the product surface × copper layer thickness.

3. The method for improving thermal warpage of a copper-clad ceramic substrate according to claim 1, characterized in that: (1) When the residual copper rate of the graphic surface is less than that of the non-graphic surface, the ideal copper layer thickness of the non-graphic surface is H1 = residual copper rate of the graphic surface / residual copper rate of the non-graphic surface × copper layer thickness of the graphic surface; (2) When the residual copper rate of the graphic surface is greater than that of the non-graphic surface, the ideal copper layer thickness of the graphic surface H2 = residual copper rate of the non-graphic surface ÷ residual copper rate of the graphic surface × copper layer thickness of the non-graphic surface.

4. The method for improving thermal warpage of a copper-clad ceramic substrate according to claim 1, characterized in that: The specific steps are as follows: Step 1: Compare the residual copper rate of the graphic surface and the residual copper rate of the non-graphic surface, select the pre-thinning surface, calculate the most ideal copper thickness of the pre-thinning surface, and sinter the product; Step 2: Take the product sintered in step 1, cover the non-pre-thinning surface with a photosensitive film, perform UV exposure and curing, calculate the thinning amount and thin the pre-thinning surface, remove the photosensitive film by stripping, and then anneal the product.

5. A method for improving thermal warpage of a copper-clad ceramic substrate according to claim 4, characterized in that: Thinning can be achieved by chemical thinning, mechanical grinding, or a combination of both.

6. A method for improving thermal warpage of a copper-clad ceramic substrate according to claim 4, characterized in that: The calculation formula for the thinning amount is: Thinning amount = actual copper thickness of the pre-thinning surface (not the copper thickness on the drawing) - ideal copper thickness.

7. A method for improving thermal warpage of a copper-clad ceramic substrate according to claim 6, characterized in that: When the thinning amount is above 0.02mm, chemical thinning and mechanical grinding are used together for thinning; when the thinning amount is between 0-0.02mm, thinning is directly performed through one or more mechanical grindings.

8. The method for improving thermal warpage of a copper-clad ceramic substrate according to claim 5, characterized in that: The chemical solution used in the chemical thinning includes hydrochloric acid, sodium chlorate and copper ions, wherein the equivalent concentration of hydrochloric acid is 1.7±0.2N, the conductivity of sodium chlorate is 15-60, and the specific gravity of copper ions is 1.28-1.

30.

9. The method for improving thermal warpage of a copper-clad ceramic substrate according to claim 5, characterized in that: Mechanical grinding uses a ceramic brush roller; mechanical grinding reduces the copper thickness by 0-10μm at a time.

10. The method for improving thermal warpage of a copper-clad ceramic substrate according to claim 4, characterized in that: In step 2, the annealing temperature is set to 330±10°C and kept at this temperature for 15 minutes.

Images