Ag-Ti soldering paste for ceramic copper-clad plate as well as preparation method and application method of Ag-Ti soldering paste

Abstract

Solid powder and an organic carrier are matched according to the mass ratio of (82-86): (14-18), silver powder and titanium dihydride powder which form the solid powder are matched according to the mass ratio of 96: 4, and a solvent, a thickening agent and a dispersing agent which form the organic carrier are matched according to the mass ratio of (91-95): (2-6): (1-5). According to the Ag-Ti soldering paste, silver powder and titanium dihydride powder serve as solid phases and are matched with a specific organic carrier, by optimizing the rheological property and the vacuum brazing technology, TiH2 is decomposed to generate active Ti atoms, the active Ti atoms react with AlN ceramic to generate a continuous TiN reaction layer, and high-strength and low-defect connection of an AlN ceramic base material and an oxygen-free copper metal copper layer is achieved; a traditional Cu element is abandoned in a soldering paste system, generation of a brittle phase is avoided, and the reliability of a connector under the extreme working condition is improved; when applied to preparation of the ceramic copper-clad plate, the film has the advantages of strong silk-screen printing adaptability, uniform film thickness and low voidage.

Description

TECHNICAL FIELD

[0001] The present application relates to the field of ceramic copper clad plate welding technology in the field of electronic packaging, in particular to the solder paste for welding the ceramic substrate and the metal copper layer in the ceramic copper clad plate and the preparation method and application method thereof. BACKGROUND

[0002] Insulated gate bipolar transistor (IGBT) is widely used in the fields of automobile, photovoltaic, energy storage, wind power, rail transit, power grid, industry, household appliances, etc. In the IGBT module, the ceramic copper clad plate is used as the core packaging carrier. Due to its excellent electrical insulation characteristics, excellent heat conduction efficiency, and material characteristics highly matched with the thermal expansion coefficient of the semiconductor chip and the heat dissipation base, the heat generated by the IGBT during work can be quickly conducted away, avoiding the performance degradation of the chip due to overheating, and improving the reliability and stability of the IGBT. In addition, the ceramic copper clad plate has high mechanical strength, which can provide stable physical support for the IGBT chip, ensuring its stable work in complex environments and helping the IGBT to play a greater role in various application fields. However, the connection reliability between the ceramic substrate and the metal copper layer in the ceramic copper clad plate is one of the main means to achieve effective connection, which is mainly achieved by active metal brazing technology.

[0003] Although the traditional Ag-Cu-Ti solder paste has many advantages, it is complex due to the presence of copper element, and the preparation of silver-copper-titanium ternary alloy powder is challenging. Most of them use mechanical mixing method to mix titanium powder with silver-copper binary alloy powder, and then use other technical means to prepare the required silver-copper-titanium solder paste. However, the silver-copper-titanium solder paste prepared by this method has not very ideal welding effect, and cannot meet some technical requirements. Under extreme working conditions such as high power density, continuous high temperature and thermal shock, moisture and corrosive medium, high voltage and electromigration, brittle phase failure is easy to occur, affecting the reliability of the joint. SUMMARY

[0004] The purpose of the present application is to provide a solder paste for welding the ceramic substrate and the metal copper layer in the ceramic copper clad plate, to solve the problem of complex traditional solder paste system and insufficient reliability, and to achieve high-strength and low-defect connection between ceramic and metal. The second purpose is to provide a preparation method for the solder paste. The third purpose is to provide an application method for the solder paste.

[0005] Technical scheme: An Ag-Ti solder paste for ceramic copper clad plate is prepared by using solid powder and organic carrier as raw materials. The solid powder and the organic carrier are mixed in a mass ratio of (82-86):(14-18), and the mass ratio of silver powder and titanium dihydride powder is 96:4.

[0006] The solid powder is silver powder and titanium dihydride powder, and the mass ratio of silver powder and titanium dihydride powder is 96:4.

[0007] The organic carrier is a solvent, a thickening agent and a dispersing agent, and the solvent, the thickening agent and the dispersing agent are in a mass ratio of (91-95):(2-6):(1-5) based on 100% of the mass of the organic carrier; the solvent is a mixture of terpineol and diethylene glycol butyl ether in a mass ratio of 3:2, the thickening agent is ethyl cellulose, and the dispersing agent is a mixture of di-n-octyl phthalate and 2,2,4-trimethyl-1,3-pentanediol monoisobutyrate in a mass ratio of 1:1.

[0008] Preferably, the solid powder and the organic carrier are in a mass ratio of 84:16, wherein:

[0009] The silver powder and the titanium dihydride powder are in a mass ratio of 96:4 based on 100% of the mass of the solid powder;

[0010] The solvent, the thickening agent and the dispersing agent are in a mass ratio of 93:4:3 based on 100% of the mass of the organic carrier.

[0011] Further, the silver powder has a particle size of 0.15-0.35 μm and a purity of not less than 99.95%, and the titanium dihydride powder has a mesh size of -325 mesh and a purity of not less than 99%.

[0012] Further, the silver powder has a bulk density of 0.6-1.0 g / cm 3 , a specific surface area of 1.6-3.8 m 2 / g, and a tap density of 1.0-1.8 g / cm 3 .

[0013] Preferably, the silver powder has a bulk density of 0.75 g / cm 3 , a specific surface area of 1.73 m 2 / g, and a tap density of 1.2 g / cm 3 .

[0014] Further, the ceramic substrate of the ceramic copper-clad plate is aluminum nitride, and the metal copper layer is oxygen-free copper.

[0015] A preparation method of the Ag-Ti solder paste for the ceramic copper-clad plate, comprising the following steps:

[0016] Step S10, preparing the organic carrier: adding ethyl cellulose into a mixture of terpineol and diethylene glycol butyl ether, heating and stirring in a 70°C water bath until completely dissolved, and then adding a mixture of di-n-octyl phthalate and 2,2,4-trimethyl-1,3-pentanediol monoisobutyrate after cooling and stirring uniformly;

[0017] Step S20, mixing the solder paste: adding silver powder and titanium dihydride powder into the organic carrier, and stirring to form a uniform paste.

[0018] Furthermore, in step S20, the stirring is done manually first and then mechanically. The manual stirring lasts for 3 minutes, and the mechanical stirring is done at 800 r / min for 30 minutes and then at 1600 r / min for 3.5 hours.

[0019] A method for applying Ag-Ti solder paste for the above-mentioned ceramic copper-clad laminate involves using a 200-mesh polyester screen printing stencil to print a pattern of solder paste onto an aluminum nitride ceramic substrate, with a film thickness of approximately 24 μm. Then, an oxygen-free copper metal layer is placed on the pattern, and the two are then placed together in a vacuum brazing furnace. The heating rate is 10 °C / min, the brazing temperature is 840 °C, and the holding time is 10 min. The solder paste forms a uniform solder paste layer, which connects the aluminum nitride ceramic substrate and the oxygen-free copper metal layer.

[0020] Furthermore, an aluminum nitride ceramic sheet of the same area as the test is applied to the copper layer, a pressure molybdenum block is applied to the aluminum nitride ceramic sheet, and then they are placed together in a vacuum brazing furnace.

[0021] Beneficial effects: Ag-Ti solder paste uses silver powder and titanium dihydrogen phosphate powder as solid phases, combined with a specific organic carrier. Through optimized rheological properties and vacuum brazing process, TiH2 decomposes to generate active Ti atoms, which react with AlN ceramics to form a continuous TiN reaction layer. This achieves a high-strength, low-defect connection between the AlN ceramic substrate and the oxygen-free copper metal layer, with a peel strength of up to 18.645 N / mm, far exceeding the industry standard of no less than 10 N / mm. The solder paste system eliminates the traditional Cu element, avoiding the formation of brittle phases and improving the reliability of the joint under extreme conditions. When applied to the preparation of ceramic copper-clad laminates, it has strong adaptability to screen printing, uniform film thickness, high controllability of vacuum brazing process, and low void ratio, making it suitable for large-scale production. It is of great significance for improving the stability of high-power electronic devices in high-frequency and high-temperature environments. Attached Figure Description

[0022] Figure 1 (a) and (b) are microscopic morphology images of silver powder and titanium dihydrogen phosphate powder, respectively.

[0023] Figure 2 This is a microscopic morphology image of the surface of an aluminum nitride ceramic substrate;

[0024] Figure 3 (a) Parameter settings details for the 3ITT thixotropic test. Figure 3 (b) shows the measurement results of the 3ITT thixotropic test;

[0025] Figure 4 Viscosity curves of solder pastes with different solid contents at different shear rates;

[0026] Figure 5 10s -1 Viscosity curves of solder paste with different solid contents at different shear rates;

[0027] Figure 6 3ITT thixotropic curves for different solid content solder paste;

[0028] Figure 7 3T recovery time histograms for different solid content solder paste 3ITT test 3T stage;

[0029] Figure 8 Edge pictures of screen printed and dried films for different solid content solder paste;

[0030] Figure 9 Viscosity curves for different thickener content solder paste at different shear rates;

[0031] Figure 10 10s -1 Viscosity curves for different thickener content solder paste at different shear rates;

[0032] Figure 11 3ITT thixotropic curves for different thickener content solder paste;

[0033] Figure 12 3T recovery time histograms for different thickener content solder paste 3ITT test 3T stage;

[0034] Figure 13 Edge pictures of screen printed and dried films for different thickener content solder paste;

[0035] Figure 14 Viscosity curves for different dispersant content solder paste at different shear rates;

[0036] Figure 15 10s -1 Viscosity curves for different dispersant content solder paste at different shear rates;

[0037] Figure 16 3ITT thixotropic curves for different dispersant content solder paste;

[0038] Figure 17 3T recovery time histograms for different dispersant content solder paste 3ITT test 3T stage;

[0039] Figure 18 Edge pictures of screen printed and dried films for different dispersant content solder paste;

[0040] Figure 19 Wetting topography of solder paste on oxygen free copper;

[0041] Figure 20 Wetting topography of solder paste on AlN ceramic;

[0042] Figure 21(a), (b) are acoustic scanning maps of ceramic copper clad plate under no-pressure block and pressure block conditions respectively;

[0043] Figure 22 (a), (b) are TGA-DTG test results of soldering paste and titanium dihydride powder respectively;

[0044] Figure 23 The brazing process curve is determined based on Figure 22

[0045] Figure 24 (a)-(d) are AlN / oxygen-free copper brazing interface microstructure diagrams of Ag-Ti soldering paste at 840℃ for 10min, which are solder joint as a whole, AlN ceramic side interface, solder joint area, and oxygen-free copper side interface respectively;

[0046] Figure 25 (a)-(f) are acoustic scanning maps of ceramic copper clad plate prepared at 800℃, 820℃, 840℃, 860℃, 880℃, and 900℃ for 10min respectively;

[0047] Figure 26 (a)-(f) are interface microstructures of ceramic copper clad plate cross section prepared at 800℃, 820℃, 840℃, 860℃, 880℃, and 900℃ for 10min respectively.

[0048] Figure 27 (a)-(e) are acoustic scanning maps of ceramic copper clad plate prepared at 840℃ for 0min, 5min, 10min, 20min, and 30min respectively;

[0049] Figure 28 (a)-(e) are interface microstructures of ceramic copper clad plate cross section prepared at 840℃ for 0min, 5min, 10min, 20min, and 30min respectively. DETAILED DESCRIPTION

[0050] The application will be further illustrated below in combination with the accompanying drawings and specific examples.

[0051] An Ag-Ti soldering paste for ceramic copper clad plate is prepared by taking solid powder and organic carrier as raw materials, and the solid powder and the organic carrier are matched in a mass ratio of (82-86):(14-18) based on 100% of the mass of the soldering paste, wherein:

[0052] ​The solid powder is silver powder and titanium dihydride powder, and the silver powder and the titanium dihydride powder are in a mass ratio of 96:4, based on 100% of the mass of the solid powder; the silver powder has a particle size of 0.15-0.35 μm, a purity of not less than 99.95%, and a loose bulk density of 0.6-1.0 g / cm 3 ; the specific surface area is 1.6-3.8 m 2 / g; and the tap density is 1.0-1.8 g / cm 3 ; the titanium dihydride powder has a mesh size of -325 mesh and a purity of not less than 99%;

[0053] The organic carrier is a solvent, a thickening agent and a dispersant, and the solvent, the thickening agent and the dispersant are in a mass ratio of (91-95):(2-6):(1-5), based on 100% of the mass of the organic carrier; the solvent is a mixture of terpineol and diethylene glycol butyl ether in a mass ratio of 3:2, the thickening agent is ethyl cellulose, and the dispersant is a mixture of di-n-octyl phthalate and 2,2,4-trimethyl-1,3-pentanediol monoisobutyrate in a mass ratio of 1:1.

[0054] The ceramic substrate of the ceramic copper-clad plate is aluminum nitride, and the metal copper layer is oxygen-free copper.

[0055] The preparation method of the Ag-Ti solder paste for the ceramic copper-clad plate includes the following steps:

[0056] In step S10, the organic carrier is prepared: ethyl cellulose is added to a mixture of terpineol and diethylene glycol butyl ether, and heated and stirred in a water bath at 70°C until completely dissolved; after cooling, a mixture of di-n-octyl phthalate and 2,2,4-trimethyl-1,3-pentanediol monoisobutyrate is added, and stirred uniformly; and then the mixture is placed for standby use;

[0057] In step S20, the solder paste is mixed: silver powder and titanium dihydride powder are added to the organic carrier and stirred; the stirring is first manual stirring and then mechanical stirring; the manual stirring is for 3 min, and the mechanical stirring is first stirring at 800 r / min for 30 min, and then stirring at 1600 r / min for 3.5 h; a uniform paste is formed, and the preparation of the solder paste is completed.

[0058] Example 1

[0059] The raw materials for preparing the Ag-Ti solder paste are as follows: 8.64 g of silver powder, 0.36 g of titanium dihydride powder, i.e. 9 g of solid powder in total, 1.5943 g of solvent (0.9566 g of terpineol and 0.6377 g of diethylene glycol butyl ether), 0.0686 g of ethyl cellulose, and 0.0514 g of dispersant (0.0257 g of di-n-octyl phthalate and 0.0257 g of 2,2,4-trimethyl-1,3-pentanediol monoisobutyrate), i.e. 1.7143 g of organic carrier in total.

[0060] The raw materials for preparing the solder paste used in the examples and experiments are: silver powder, titanium dihydride powder, terpineol, diethylene glycol butyl ether, ethyl cellulose, di-n-octyl phthalate, 2,2,4-trimethyl-1,3-pentanediol monoisobutyrate.

[0061] The properties of the silver powder and titanium dihydride powder are shown in Table 1, and the micro-morphology diagrams are Figure 1 (a), Figure 1 (b).

[0062] Table 1

[0063]

[0064]

[0065] The properties of terpineol, diethylene glycol butyl ether, ethyl cellulose, di-n-octyl phthalate, and 2,2,4-trimethyl-1,3-pentanediol monoisobutyrate are shown in Table 2.

[0066] Table 2

[0067]

[0068] In the examples and experiments, ceramic copper-clad plates were prepared using aluminum nitride (AlN) ceramic substrates as the base and oxygen-free copper with a purity of 99.99% as the metal copper layer material. To accurately grasp the micro features of the aluminum nitride ceramic substrate, a scanning electron microscope (SEM) was used for characterization, and the sample was gold sprayed before analysis to optimize the conductivity and imaging quality. The surface morphology is as shown in Figure 2 .

[0069] Experiment 1, Effect of Different Solid Contents on the Rheological Properties of Solder Paste and Screen Printing Quality

[0070] 1. Five groups of solder paste with different solid contents (82%, 83%, 84%, 85%, and 86%) were prepared by mixing solid powder and organic carrier. The effects of different solid contents on the rheological properties of the solder paste were investigated. The corresponding numbers were 1, 2, 3, 4, and 5, respectively. A total of 9g of solid powder was taken for each group, and the mass of each raw material is listed in Table 3.

[0071] Table 3

[0072]

[0073] 2. Preparation of Ceramic Substrate

[0074] (1) Sample cutting: Cu plates were cut into 8mm x 100mm, 5mm x 110mm strips and 13mm x 13mm, 20mm x 20mm square shapes; AlN ceramic substrates were cut into 8mm x 100mm strips and 15mm x 15mm, 20mm x 20mm and 40mm x 40mm square shapes.

[0075] (2) The samples were used after being removed of impurities, pickled, washed, and dried.

[0076] 3. Preparation of solder paste

[0077] 4. Screen printing

[0078] The solder paste was printed on the AlN ceramic substrate to form a pattern by screen printing.

[0079] 5. Rheological property testing

[0080] (1) Viscosity test, reflecting the degree of hindrance of the solder paste fluid to flow deformation.

[0081] (2) Thixotropy test, using a rheometer, and using the 3ITT test method to test the thixotropy of the solder paste. Figure 3 (a) Parameter setting details for 3ITT thixotropy test, Figure 3 (b) Measurement results for 3ITT thixotropy test.

[0082] The rheological property test conditions of the solder paste are as follows: the temperature is 25℃, and the smallest substrate is 25mm. To reduce the measurement error of the instrument, a waiting time is set before testing, and the solder paste is allowed to stand for 10s -1 The viscosity at the shear rate is taken as the measurement value (this shear rate is in the middle region of the experimental test range, can better reflect the viscosity characteristics of the solder paste under common working conditions, and can effectively avoid the data fluctuation interval caused by unstable factors in the initial stage of the experiment, can ensure that the experimental data is highly accurate, stable and comparable), the 3ITT test curve covers three key time stages and different shear rate conditions, and the three time stages are accurately set to 2min, 30s and 6min, corresponding to 1T, 2T and 3T in turn. The three shear rates are: 0.1s -1 in the 1T stage, simulating the situation of the solder paste in the static or slow flow state; the shear rate is switched to 300s -1 in the 2T stage, to simulate the stress state of the solder paste in the process of rapid application of external force such as screen printing; and the shear rate is returned to 0.1s -1 in the 3T stage, to observe the characteristics of the solder paste in returning to the initial state after experiencing high-speed shearing.

[0083] 6. Test results

[0084] (1) Viscosity test results

[0085] Figure 4 For solder pastes with different solid contents (numbered 1-5) at different shear rates (1-120 s) -1 The viscosity curves are shown below. As the shear rate gradually increases from a lower value to a higher value, the viscosity of the solder paste for each solid content shows a stable and continuous decreasing trend. Under the same shear rate conditions, as the solid content gradually increases from 82% to 86% in increments of 1%, the viscosity of the solder paste shows a steady increasing trend.

[0086] Table 4 shows 10s -1 Table of solder paste viscosity at different shear rates and solid contents, corresponding to Figure 5 10s -1 Viscosity curves of solder paste with different solid contents at different shear rates. As the solid content continues to increase, when the solid content reaches 84%, the viscosity of the solder paste further increases to 18.527 Pa·s. When the solid content is greater than 84%, the viscosity of the solder paste begins to increase sharply.

[0087] Table 4

[0088]

[0089]

[0090] (2) Thixotropic test results

[0091] Figure 6 The 3ITT thixotropic curves are for five solder pastes with different solid contents, numbered 1 to 5.

[0092] Figure 7 The bar chart shows the recovery time during the 3-time stage of the 3ITT test for five solder pastes with different solid contents (numbered 1-5). The 84% solid content solder paste exhibits extremely outstanding recovery characteristics, with the shortest recovery time to its initial viscosity value, approximately 20 seconds. The internal structure of the solder paste can quickly and stably recover to its initial state after being subjected to high-speed shear, resulting in a relatively ideal recovery effect.

[0093] (3) Results of screen printing dry film morphology test

[0094] Figure 8 Images (a)-(e) show the edges of the films after screen printing and drying of five different solid content solder pastes (numbered 1-5). The 84% solid content solder paste showed outstanding performance, smoothly filling the screen mesh during printing. After being transferred to the substrate, its edges were smooth and flat, maintaining a stable shape and effectively avoiding common defects such as over-expansion and missing prints.

[0095] Experiment 2: Effects of different thickener contents on the rheological properties and screen printing quality of solder paste

[0096] 1. Based on the optimal solid content of 84% determined in Experiment 1, and with the mass ratio of silver powder to titanium dihydrogen phosphate powder in the solder paste set at 96:4, the effects of different thickener (ethyl cellulose) contents on the rheological properties of the solder paste were investigated. The corresponding numbers are 6, 7, 8, 9, and 10. Assuming that 9g of solid powder is taken in each group, the mass of each raw material is listed in Table 5.

[0097] Table 5

[0098]

[0099] 2. Preparation of ceramic substrate, preparation of solder paste, and screen printing: refer to the procedures in Experiment 1.

[0100] 3. Test Results

[0101] (1) Viscosity test results

[0102] Figure 9 For solder pastes with different thickener contents (numbered 6-10) at different shear rates (1-120 s)... -1 The viscosity curves are shown below. As the shear rate increases, the viscosity values ​​of solder pastes with different ratios gradually decrease, exhibiting shear-thinning characteristics. Under the same shear rate conditions, the higher the ethyl cellulose content, the greater the solder paste viscosity.

[0103] Table 6 shows 10s -1 Table of solder paste viscosity with different shear rates and thickener contents, corresponding to... Figure 10 10s -1 Viscosity curves of solder paste with different shear rates and thickener contents. As the ethyl cellulose content continues to increase, when the ethyl cellulose content reaches 4%, the solder paste viscosity jumps to 18.527 Pa·s. When the ethyl cellulose content is greater than 4%, the increasing trend of solder paste viscosity becomes extremely obvious.

[0104] Table 6

[0105]

[0106] (2) Thixotropic test results

[0107] Figure 11 The 3ITT thixotropic curves are for five solder pastes with different thickener contents, numbered 6 to 10.

[0108] Figure 12 The bar chart shows the 3ITT recovery time of five solder pastes with different thickener contents (6-10) during the 3ITT test. The solder paste with 4% thickener content exhibits extremely outstanding recovery characteristics, recovering to its initial viscosity value in the shortest time, only 20 seconds, and maintaining high viscosity stability throughout the recovery process.

[0109] (3) Silk screen dry film appearance test results

[0110] Figure 13 (a)-(e) are the edge pictures of the five different thickener content solder paste film after silk screen printing and drying. The solder paste with 4% thickener content has suitable rheological properties, the edge after printing is smooth and regular, and the printing area is clearly defined without defects such as excessive extension and edge collapse.

[0111] Through the research and experiment of the rheological properties of the five different thickener content solder pastes numbered 6-10, combined with their actual performance in silk screen printing, it can be seen that the 4% ethyl cellulose solder paste has significant advantages in rheological properties and printing quality. Experiment three, the influence of different content of dispersant on the rheological properties of solder paste and silk screen printing quality

[0112] 1. On the basis of the best solid content of 84% determined in experiment one and the best thickener content of 4% determined in experiment two, and the silver powder and titanium dihydride powder in the solder paste are set to a mass ratio of 96:4, the influence of different dispersant contents on the rheological properties of the solder paste is investigated, and the corresponding numbers are 11, 12 and 13. If 9g of solid powder is taken in each group, the mass of each raw material is listed in Table 7.

[0113] Table 7

[0114]

[0115]

[0116] 2. The preparation of ceramic substrate, the preparation of solder paste and silk screen printing refer to the operation of experiment one.

[0117] 3. Test results

[0118] (1) Viscosity test results

[0119] Figure 14 The viscosity curve diagram of the three different dispersant content solder pastes numbered 11-13 under different shear rates (1-120s -1 ). Table 8 is the viscosity table of the solder paste with different dispersant contents at 10s -1 shear rate, and the corresponding Figure 15 is the viscosity curve diagram of the solder paste with different dispersant contents at 10s -1 shear rate.

[0120] Table 8

[0121]

[0122] Through the research and experiment of the rheological properties of the five different thickener content solder pastes numbered 6-10, combined with their actual performance in silk screen printing, it can be seen that the 4% ethyl cellulose solder paste has significant advantages in rheological properties and printing quality. Figure 14 , Table 8, Figure 15The systematic analysis of the experimental results shows that as the content of dispersant in the solder paste system gradually increases, the viscosity of the solder paste exhibits a highly regular trend.

[0123] When the dispersant content increased to 3%, the solder paste viscosity decreased significantly to 18.527 Pa·s. When the dispersant content further increased to 5%, the solder paste viscosity dropped to 15.294 Pa·s. When the dispersant content exceeded 1%, the decreasing trend of solder paste viscosity became very obvious. When the dispersant content exceeded 3%, although the solder paste viscosity continued to decrease, the rate of decrease slowed down, showing a relatively stable downward trend.

[0124] (2) Thixotropic test results

[0125] Figure 16 The 3ITT thixotropic curves are for solder pastes with different dispersant contents, numbered 11 to 13.

[0126] Through the Figure 16 Analysis of the test results shows that as the mass fraction of dispersant in the solder paste system gradually increases, the viscosity value of the solder paste shows a continuous decreasing trend in the 1T stage of the 3ITT test.

[0127] Further observation was conducted on the characteristics of the solder paste recovering to its initial viscosity value after undergoing a cycle of low-speed-high-speed-low-speed shear rate changes. Figure 17 The bar chart shows the 3ITT recovery time of solder pastes with different dispersant contents (numbers 11-13) during the 3T stage. When the mass fraction of dispersant is increased to 3%, the solder paste exhibits extremely outstanding recovery characteristics, with the shortest recovery time to the initial viscosity value, which is only 20 seconds. Furthermore, the viscosity value remains highly stable throughout the recovery process.

[0128] (3) Results of screen printing dry film morphology test

[0129] Figure 18 Images (a)-(c) show the edges of the solder paste after screen printing and drying for three different dispersant contents (numbers 11-13). It is clear that a dispersant content of 3% yields the best results. With a 3% dispersant content, the components in the solder paste are evenly dispersed, resulting in smooth and delicate edges with natural transitions. There are no obvious particle accumulations or gaps, and the printed area is clearly and neatly outlined, providing a good foundation for subsequent soldering.

[0130] Based on the research and experiments on the rheological properties of solder pastes with different dispersant contents (numbers 11-13), and their actual performance in screen printing, it can be concluded that a 3% dispersant addition is the most suitable.

[0131] The principle of the solder paste formula system of the present application: TiH2 is used to replace Ti powder, which decomposes into active Ti atoms and H2 at the soldering temperature, can avoid early oxidation, inhibit pore generation and be more stable and easy to store; the particle size of silver powder and titanium dihydride powder can balance the rheological property and welding performance of the solder paste, avoid the problems of too coarse or too fine, and improve the wettability; terpineol and diethylene glycol butyl ether as solvents can optimize the dissolution and dispersion, control the volatilization, balance the viscosity and flowability, and ensure the chemical stability; ethyl cellulose can form a network to give thixotropy, thicken to prevent sedimentation, and slow down the volatilization, and has chemical inertness and pH stability, and no residue after high temperature decomposition; di-n-octyl phthalate and 2,2,4-trimethyl-1,3-pentanediol monoisobutyrate as dispersants can enhance the dispersion of powders through adsorption and steric hindrance, optimize the rheological property and be compatible with each component.

[0132] The application method of the above-mentioned Ag-Ti solder paste for ceramic copper-clad plate uses a 200-mesh polyester screen printing screen to print the solder paste on the aluminum nitride ceramic substrate to form a pattern, uses a confocal laser scanning microscope (CLSM) detection equipment to accurately measure the thickness of the solder paste after printing, accurately measures that the printed film thickness of the solder paste is stabilized at about 24 μm, then places an oxygen-free copper metal copper layer on the pattern, and applies an aluminum nitride ceramic sheet of the same size as the oxygen-free copper metal copper layer on the metal copper layer, applies a pressure molybdenum block on the aluminum nitride ceramic sheet, and then places them together in a vacuum soldering furnace, with a heating rate of 10°C / min, a soldering temperature of 840°C, and a holding time of 10 min, and the solder paste forms a uniform solder paste layer to connect the aluminum nitride ceramic substrate and the oxygen-free copper metal copper layer.

[0133] Experiment four, wettability test of solder paste on oxygen-free copper and AlN ceramic surface

[0134] Figure 19 It is the wetting morphology of Ag-Ti solder paste on the surface of oxygen-free copper under the sintering condition of 840°C. The Ag-Ti solder paste has a large spreading area on the surface of the oxygen-free copper plate, the edge is in a natural expansion state and has no aggregation and contraction phenomenon, the contact angle is small, and it shows strong spreading driving force and wettability. At the same time, the solder paste and the copper plate interface are closely combined without defects such as pores and cracks, which indicates that a stable combination is formed.

[0135] Figure 20 It is the wetting morphology of Ag-Ti solder paste on the surface of AlN ceramic under the sintering condition of 960°C. The Ag-Ti solder paste presents a nearly circular uniform spreading morphology on the surface of AlN ceramic, the edge is regular and has no aggregation and splashing phenomenon.

[0136] In the actual production and preparation of ceramic copper-clad plate, the Cu element provided by the oxygen-free copper cooperates with the reaction system, the Cu, Ag and Ti form a ternary eutectic phase to reduce the liquidus temperature, promote the diffusion of Ti and interface reaction, improve the liquid metal wettability and inhibit the generation of brittle phase, so that high-efficiency sintering can be realized at a lower temperature.

[0137] Experiment five, void ratio detection, peel strength detection

[0138] The molybdenum block is applied with pressure, which helps the solder paste spread and flow better between the ceramic substrate and the copper-clad layer under pressure conditions, enhances the solder paste's filling capacity for the interface, promotes the easy discharge of gas, and effectively inhibits the formation of voids. The introduction of pressure significantly improves the quality of the ceramic copper-clad plate brazing, reduces internal void defects, and the void ratio is less than 5%, which is beneficial to improve the electrical performance and mechanical reliability of the product. Figure 21 (a) and (b) are acoustic scanning maps of ceramic copper-clad plates under pressure-free block and pressure block conditions.

[0139] Two sets of tests with pressure copper on the bottom and pressure copper on the top were set up and the peel strength was detected. The test results showed that the peel strength of pressure copper on the bottom was 15.232 N / mm, and the peel strength of pressure copper on the top was 18.645 N / mm. The test results showed that the peel strength of pressure copper on the top was higher than that of pressure copper on the bottom.

[0140] Experiment six, TGA-DTG test of solder paste

[0141] Figure 22 (a) and (b) are the TGA-DTG test results of solder paste and titanium dihydride powder. The solder paste and TiH2 respectively correspond to a significant endothermic peak at three temperature points of 165℃, 319℃ and 614℃. These three endothermic peaks are most likely to correspond to the decomposition temperatures of the organic carrier and TiH2. This result provides key temperature reference points for the accurate setting of subsequent sintering process curves, which helps to more effectively control the reaction process of the solder paste in actual production. In order to further protect the solder paste from oxidation during testing, argon gas is continuously introduced as a protective gas, and the argon gas flow is accurately controlled at 20 L / min. The entire test temperature range is set from room temperature 25℃ to 1000℃, and the heating rate is strictly controlled at 20℃ / min. Based on this, the brazing process curve is determined as shown in Figure 23 .

[0142] Experiment seven, effect of brazing process on the microstructure and properties of ceramic copper-clad plate

[0143] Figure 24 (a)-(d) are the AlN / oxygen-free copper brazing interface microstructure diagrams of Ag-Ti solder paste at 840℃ for 10min. A continuous and dense reaction layer is formed between the AlN ceramic base material and the solder joint area, indicating that effective element diffusion and chemical reaction have occurred at the interface. The diffusion behavior of Cu elements in the oxygen-free copper matrix to the solder joint area promotes the bonding between the base material and the solder. The whole joint presents a uniform microstructure, and no defects such as pores, inclusions or cracks are observed, which confirms the feasibility of the brazing process and the reliability of the joint performance.

[0144] From Figure 24 The AlN ceramic side interface microstructure of (b) shows that a dense reaction layer is formed between the weld zone and the AlN ceramic. The EDS detection results show that the atomic ratio of Ti and N elements at points 3 and 6 is approximately 1:1, indicating that Ti can react with AlN ceramic to form TiN compound. During brazing, the Ag-Ti brazing paste melts, Ti migrates to the surface of the AlN ceramic, reacts in situ with the released N, and generates a continuous and dense TiN reaction layer. The reaction layer forms a metallurgical bond with the original AlN ceramic (position 4, containing only Al and N), not only enhancing the interface strength through chemical bonding, but also relieving thermal expansion stress concentration through crystal structure matching, becoming a key barrier to inhibit interface cracks, and being the core structure of ceramic side connection reliability.

[0145] From Figure 24 The weld zone microstructure of (c) shows that a clear dendritic network structure is formed inside the weld, and the network shows significant morphological differences on both sides of the interface. The dendritic arm spacing near the oxygen-free copper interface region is significantly refined, while the dendrites near the AlN ceramic interface show thickening characteristics.

[0146] From Figure 24 The oxygen-free copper side interface microstructure of (d) shows that the structure on the oxygen-free copper side is dense. During brazing, Ag in the brazing paste diffuses into the Cu matrix, and the two form a solid solution to achieve metallurgical bonding.

[0147] Figure 25 (a)-(f) are acoustic scanning images of ceramic copper clad plates prepared at six different brazing temperatures of 800℃, 820℃, 840℃, 860℃, 880℃, and 900℃ for 10min, and the corresponding Figure 26 (a)-(f) are the interface microstructures of the cross-section of the ceramic copper clad plates prepared at six different brazing temperatures for 10min, and Table 9 is the thickness of the corresponding reaction layer. With the increase of brazing temperature, the thickness of the reaction layer shows a clear increasing trend.

[0148] Table 9

[0149] Temperature (°C) 800 820 840 860 880 900 Thickness (μm) 0.015 0.244 0.521 0.59 0.847 1.437

[0150] Figure 27 (a)-(e) are acoustic scanning images of ceramic copper clad plates prepared at 840℃ for 0min, 5min, 10min, 20min, and 30min respectively, and the corresponding Figure 28 (a)-(e) are the interface microstructures of the cross-section of the ceramic copper clad plates prepared at five different holding times, and Table 10 is the thickness of the corresponding reaction layer. With the extension of holding time, the reaction layer gradually becomes continuous and thickens.

[0151] Table 10

[0152] Time (min) 0 5 10 20 30 Thickness (μm) 0.206 0.368 0.521 0.635 0.894

[0153] It can be seen that increasing the brazing temperature or increasing the holding time will lead to the growth of the interface reaction layer thickness. On this basis, the ceramic copper clad plate is tested for peel strength. Table 11 is the detection result of the peel strength of the sample brazed at different brazing temperatures for 10 min. In the range of 800-840℃, the peel strength increases from 18.185N / mm to 18.645N / mm as the temperature increases. When the temperature exceeds 840℃ (860-900℃), the peel strength gradually decreases to below 15.591N / mm. Table 12 is the detection result of the peel strength of the sample brazed at 840℃ for different holding times. When the holding time is extended from 0 min to 10 min, the peel strength increases from 15.009N / mm to 18.645N / mm. When the holding time is further extended to 20 min and 30 min, the strength decreases to 17.134N / mm and 16.579N / mm, respectively, and appears to fluctuate and decline.

[0154] Table 11

[0155]

[0156] Table 12

[0157]

[0158] The principle of the solder paste application method of the present application is as follows:

[0159] (1) The Ag-Ti solder paste can realize stable combination of AlN ceramic and oxygen-free copper by vacuum brazing process, and the structure of the cross-sectional organization of the connection interface is AlN / TiN / [Ag solid solution + Cu solid solution] / Cu in turn.

[0160] (2) In the brazing process, the active element Ti in the liquid solder will diffuse towards the surface of the AlN ceramic and react with it to form a continuous TiN reaction layer. The solder mainly contains Ag-based solid solution and Cu-based solid solution; at the interface between the solder and the oxygen-free copper base material, mainly the Cu atoms in the base material diffuse into the liquid solder, thereby building a Cu-based solid solution.

Claims

1. An Ag-Ti solder paste for ceramic copper clad plate, characterized by: The soldering paste is prepared by using solid powder and organic carrier as raw materials, the mass ratio of the solid powder and the organic carrier is (82-86):(14-18) based on 100% of the mass of the soldering paste, wherein: The solid powder is silver powder and titanium dihydride powder, the mass ratio of the silver powder and the titanium dihydride powder is 96:4 based on 100% of the mass of the solid powder; The organic carrier is solvent, thickening agent and dispersant, the mass ratio of the solvent, the thickening agent and the dispersant is (91-95):(2-6):(1-5) based on 100% of the mass of the organic carrier; the solvent is terpineol and diethylene glycol butyl ether in a mass ratio of 3:2, the thickening agent is ethyl cellulose, and the dispersant is di-n-octyl phthalate and 2,2,4-trimethyl-1,3-pentanediol monoisobutyrate in a mass ratio of 1:

1.

2. The Ag-Ti solder paste for a ceramic copper-clad plate according to claim 1, characterized by: The mass ratio of the solid powder and the organic carrier is 84:

16. The mass ratio of the silver powder and the titanium dihydride powder is 96:4 based on 100% of the mass of the solid powder; The mass ratio of the solvent, the thickening agent and the dispersant is 93:4:3 based on 100% of the mass of the organic carrier.

3. The Ag-Ti solder paste for a ceramic copper-clad plate according to claim 1, characterized by: The particle size of the silver powder is 0.15-0.35 μm, and the purity is not less than 99.95%; the mesh number of the titanium dihydride powder is -325 mesh, and the purity is not less than 99%.

4. The Ag-Ti solder paste for the ceramic copper clad plate according to claim 1, characterized by: The bulk density of the silver powder is 0.6-1.0 g / cm 3 The specific surface area is 1.6-3.8 m 2 / g, and the tap density is 1.0-1.8 g / cm 3 .

5. The Ag-Ti solder paste for ceramic copper clad plate according to claim 4, characterized in that: The tap density of the silver powder is 0.75 g / cm 3 The specific surface area is 1.73 m 2 / g, and the tap density is 1.2 g / cm 3 .

6. The Ag-Ti solder paste for the ceramic copper clad plate according to claim 1, characterized in that: The ceramic substrate of the ceramic copper-clad plate is aluminum nitride, and the metal copper layer is oxygen-free copper.

7. A method for producing the Ag-Ti solder paste for the ceramic copper clad plate according to any one of claims 1 to 6, characterized by The method comprises the following steps: In step S10, the organic carrier is prepared: ethyl cellulose is added into a mixture of terpineol and diethylene glycol butyl ether, heated and stirred in a 70℃ water bath until completely dissolved, and then a mixture of di-n-octyl phthalate and 2,2,4-trimethyl-1,3-pentanediol monoisobutyrate is added after cooling and stirred uniformly; In step S20, the soldering paste is mixed: silver powder and titanium dihydride powder are added into the organic carrier, and stirred to form a uniform paste.

8. The method of claim 7, wherein: In step S20, the stirring is first manual stirring and then mechanical stirring, the manual stirring is 3 min, and the mechanical stirring is first stirring at 800 r / min for 30 min, and then stirring at 1600 r / min for 3.5 h.

9. A method for using the Ag-Ti solder paste for ceramic copper clad plate according to any one of claims 1 to 6, characterized in that: The soldering paste is printed on the aluminum nitride ceramic substrate to form a pattern by using a 200-mesh polyester screen printing screen, the film thickness is about 24 μm, then an oxygen-free copper metal copper layer is placed on the pattern, and then they are placed in a vacuum soldering furnace, the heating rate is 10℃ / min, the soldering temperature is 840℃, the holding time is 10 min, the soldering paste forms a uniform soldering paste layer, and the aluminum nitride ceramic substrate and the oxygen-free copper metal copper layer are connected.

10. The method of use of claim 9, wherein: An aluminum nitride ceramic sheet used for testing is applied on the metal copper layer, a pressure molybdenum block is applied on the aluminum nitride ceramic sheet, and then they are placed in a vacuum soldering furnace.

Images