JWJ Journal of Welding and Joining J Korean Weld Join Soc 2466-2232 2466-2100 The Korean Welding and Joining Society Korea jwj-40-3-248 10.5781/JWJ.2022.40.3.6 Research Paper Deposition of a Porous Cu Layer by Electroplating and Feasibility Evaluation of Cu-Cu Die Attachment by Sinter Bonding Using Thermo-Compression 전해도금에 의한 다공성 Cu 층의 증착과 열압착 소결접합에 의한 Cu-Cu 다이 접합 공정성 평가 http://orcid.org/0000-0002-6982-8311 Namgoong Doyeop * http://orcid.org/0000-0002-3792-0404 Lee Jong-Hyun * * Dept. of Materials Science and Engineering, Seoul National University of Science and Technology, Seoul, 01811, Korea †Corresponding author: pljh@snut.ac.kr 30 06 2022 14 06 2022 40 3 248 255 01052022 18052022 02062022 Copyright: © The Korean Welding and Joining Society 2022 This is an Open Access article distributed under the terms of the Creative Commons Attribution Non-Commercial License (http://creativecommons.org/licenses/by-nc/4.0/) which permits unrestricted non-commercial use, distribution, and reproduction in any medium, provided the original work is properly cited.

A porous Cu layer was deposited on a dummy Cu die and a dummy substrate, respectively, by electroplating to form a bondline that can endure high temperatures, such as 300 °C. The plating at a current density of 40 mA/cm2 in a plating solution containing a crystal controlling agent resulted in different porous layers from fine dendritic grains to coarse a typical grains and an increased layer thickness with increasing plating time. Die-attachment sinter bonding at 5 MPa and 300 °C in the air by thermo-compression using a 2-min-plated die and substrate after drying a reducing agent absorbed in the layer on a substrate. The boding for 1 min resulted in sufficient bondline shear strength of 26.6 MPa with an excellent microstructure of near full density. Identical bonding with exchanging nitrogen blowing increased the shear strength to 30.3 MPa with significantly reduced oxidation area in the inner edge of the die. The 1- and 3-min samples exhibited lower strength values owing to the local formation of large voids between non-uniform thickness of the deposited layers and the relatively less dense bondline structure, respectively.

Die attachment Thermo-compression Sinter bonding Cu-Cu bonding Porous layer Shear strength
1. Introduction

Regulations on the mandatory use of green energy are intensifying throughout the world, and the automobile industry is actively transitioning from internal combustion engines to electric vehicles1,2). A power module is an automobile part that changes direct current from a battery to alternating current and increases or decreases the voltage. Large numbers of power modules are installed on electric vehicles, and they are used in several areas for the same purpose during the process of storing, moving, and using electricity that is created by various methods. As such, their usage is expected to steadily increase3,4).

Current Si-based power devices are being converted into SiC or GaN wide band-gap compound semiconductors for the need for high-voltage power modules as well as the purpose of improving electric conversion efficiency and ensuring fast switching speed in the aforementioned operational processes4-8). The operating temperature of power devices that are based on these compounds is expected to reach a maximum of 300 °C9-11), and in such cases, the use of current solder bondlines will no longer be possible. Specifically, normal lead-free solder (Sn-Ag-Cu alloys), which has a melting point of 217 °C, will undergo remelting, and even solder with high Pb content, which has a melting point above 300 °C, will inevitably suffer from reduced reliability due to a significant decrease in strength at such an operating temperature, considering that this operating temperature is a high homologous temperature in comparison to the solder’s melting point. Therefore, the sinter bonding process is actively being studied as a next-generation bonding method that prevents the formed bond from melting or experiencing a rapid decrease in mechanical strength at temperatures near 300 °C11-13).

The most actively studied material for implementing sinter bonding is Ag particles, which are applied by making a paste or film that contains them12-19). Film materials are a prominent part of current mass production processes17), and they have found more acceptance as materials that make direct contact with the chip in processes such as die bonding because they do not flow like pastes. Despite this, the high raw material cost of Ag has led to research on lowering the cost of sinter bonding material, and research is now being actively conducted on the development and use of sinter bonding materials that use Cu particles20,21). Unlike Ag, Cu continuously experiences surface oxidation in air, which can greatly reduce its sintering properties, and the degree of oxidation becomes more severe as the temperature increases22). However, unlike paste, it is very difficult to include a reducing agent in a film-type bonding material; therefore, research on sinter bonding materials that use Cu particles has been unable to move beyond current paste-type material forms23-27).

Therefore, rather than develop a film-type Cu sinter bonding material, this study deposited a Cu layer on the bonding surface of a die and a substrate, added a liquid reducing agent before sintering, and then performed die bonding using thermocompression. Deposition of the Cu layer was performed using an electroplating process so that a sufficient thickness could be achieved within a short amount of time, and a porous structure with an enhanced surface area was formed in order to ensure effective sinter bonding and rapid bonding speed.

2. Experiment Method 2.1 Porous Cu Layer Formation

To prepare the plating solution for forming the porous Cu layer, 0.2 M of copper sulfate pentahydrate (CuSO4· 5H2O, 99%, Duksan Pure Chemicals) and 0.06 M of sulfuric acid (H2SO4, 98%, Daejung Chemicals & Metals), and a crystal control agent with which the sizes of precipitated crystals can be reduced were added as traces to 300 mL of distilled water and dissolved. The prepared plating solution was poured into a cuboid container, and a pair of Cu electrodes, which were used as the anode and the cathode, were inserted in the plating solution. They were spaced 1 mm apart, and then the current density was set while electroplating was performed for 1-3 min. The Cu plate to be used as the dummy die and substrate was placed in contact with the cathode, and a porous Cu layer formed on the surface of a plate. The die material was cut into a 3 x 3 mm size, and the substrate material was cut into a 10 x 10 mm size. In the substrate, masking was applied to the material except for a 3 x 3 mm area in the middle. The specimen that had been taken out after plating for a certain amount of time was immersed in distilled water to wash it, and then it was dried in a low-vacuum chamber.

X-ray diffraction analysis (K-Alpha+, Thermo Fisher Scientific) was used to examine the composition of the deposited porous Cu layer. The layer’s thickness was measured using a scanning electron microscope (SEM, SU8010, Hitachi) after mounting the deposited layer that formed over time and polishing its cross section. The scanning electron microscope was also used to observe the deposited layer’s surface microstructure.

2.2 Compression Sinter Bonding

A reducing agent that was developed by this study was added to the porous Cu layer that formed on the dummy Cu substrate, and it was absorbed. Then, the dummy die surface with another porous Cu layer was turned downward. The die was aligned with the porous layer on the substrate and placed to form a sandwich structure. Next, the sandwich-type specimen was placed on a 150 °C hotplate for 1 min to partially dry the reducing agent. Then, it was transferred to a thermo-compression bonder, and die bonding was performed. To accomplish the sinter bonding for the die bonding, pressure (5 MPa) was applied to the die while the temperature was raised to 300 °C at a rate of around 28 °C/s for a total bonding time of 1-3 min. The heating time was included in the total bonding time.

The microstructures of the formed bondlines were observed with a scanning electron microscope after mounting the sandwich-structure samples and polishing their cross sections. The strength of the formed bondlines was measured through shear tests using a micro shear tester (Hawk 8200S, Kovis Technology). Specifically, a shear tip was placed at a height of 200 ㎛ above the substrate surface, and then the tip was moved at a rate of 200 ㎛/s. The shear strength value was determined by measuring the maximum stress value before the bondline fractured. After the shear tests, the microstructure of the fracture surface was observed with a scanning electron microscope and an optical microscope.

3. Experimental Study 3.1 Microstructure of the Deposited Porous Cu Layer

Fig. 1 shows the X-ray diffraction analysis results for the Cu plate that was used as the Cu die and substrate, as well as the Cu layer that was deposited by the electroplating method that was performed in this study. In the Cu plate that was used as the dummy die and substrate, there was only a native oxide layer owing to surface polishing with a 2,000 mesh sandpaper. The oxide phase was not detected in the X-ray analysis result, and a peak corresponding to the Cu (200) plane was the main one. In the case of the specimen that was plated for 30 min at a low current density of 10 mA/cm2, the observed results were similar to the initial Cu plate’s XRD result with the exception of the slight increase in the peaks at the Cu (111) and (220) plane due to the thin Cu plating layer. On the other hand, in the case of the specimen that was plated for 10 min at a current density of 100 mA/cm2, the Cu(111) plane peak was the main one, and the creation of a small amount of Cu2O phase was also clearly observed. These results show that Cu(111) plane growth mainly occurred by the electroplating, and the amount of Cu oxidation increased according to the increase in the thickness of the Cu plating layer. The oxidation of the plated Cu layer was found to occur because a deposition layer forms as the crystals with large surface area, forming a porous structure.

X-ray diffraction results of an initial Cu plate and porous Cu layers on the plate deposited by electroplating at different current densities

Fig. 2 consists of surface images of the Cu dies that were used in the sinter bonding, and these images show the changes in the microstructure of the deposited porous Cu layer according to the electroplating time at 40 mA/cm2. In the specimen that was plated for 1 min (Fig. 2(a)), very fine dendrite-shaped particles were formed, and a porous Cu layer consisting of mutually aggregated structures was observed. The cross-section structure of the deposition layer was very wispy, and its thickness reached around 7.2 ㎛. When the plating time was increased to 2 min (Fig. 2(b)), the dendrite shapes disappeared, and particles with fairly flat shapes aggregated to create fractal-shaped aggregate particles. At the same time, a porous Cu layer with more structure than the 1-min deposition layer was observed due to the additional aggregation between the particles. However, the cross-section structure of this specimen was still wispy, and its thickness was around 16.8 ㎛. Finally, when the plating time was increased to 3 min (Fig. 2(c)), irregular particles that were coarser than the basic particle size were formed, and a porous Cu layer was observed in parts due to the aggregation between these coarse particles. The cross-section structure of this specimen was dense in parts, and its thickness reached approximately 24.1 ㎛. In conclusion, even when the electroplating was performed at the same current density, the average size of the Cu basic particles that made up the formed porous Cu layer clearly tended to increase as the plating time increased, and as a result, changes in the size of the pores and the porosity inside the layer were observed.

Surface SEM images of porous Cu layers on a Cu dummy die electroplated at 40 mA with different electro-plating times: (a) 1 min, (b) 2 min, and (c) 3 min

3.2 Compression Sinter Bonding Properties in Air

Fig. 3 shows the microstructural changes in the final formed bondlines according to the electroplating time when the porous Cu layers were sinter-bonded for 1 min in air at 300 °C with 5 MPa compression. When the Cu die and substrate that were plated for 1 min were used (Fig. 3(a)), a very fine sinter bondline structure was observed, but the formed bond’s final thickness was very thin at around 2.6 ㎛. On the other hand, when the Cu die and substrate that were plated for 2 min were used (Fig. 3(b)), a fairly fine sinter bondline was formed in which only the particle contours were observed, and its thickness increased to around 4.9 ㎛. The formed bondline contained almost no coarse pores, which indirectly confirms that the sinter bonding was performed relatively uniformly through the entire area. Specifically, when considering the microstructure of the porous deposition layer shown in Fig. 2(b), the rearrangement of micro-particles in the compression sintered bondline and the rapid sintering between the contacting particles caused the relatively fine bond microstructure that was observed in Fig. 3(b) in a short time of 1 min, and the rapid sintering between the contacting particles was caused by the initial bonding layer structure which had a large surface area due to the formation of a fine Cu particle-based porous layer28). Finally, the bonding that used the Cu die and substrate that were plated for 3 min (Fig. 3(c)) formed the most coarse sinter bondline, but when considering the initial pore structure before sinter bonding as shown in Fig. 2(c), it was found that the actual sinter bonding progressed quite far. In addition, the specimen that was bonded under these conditions showed a thick bondline formation of up to approximately 15.9 ㎛ due to an initial deposition thickness that was markedly thicker than the other specimens and a relatively dense initial microstructure. Based on the results above, it was observed that when the Cu particles, which constituted the initial bonding layer that was formed by electroplating, were smaller (i.e. had a larger surface area), it resulted in a finer microstructure in the bondline that was formed after performing the heated sinter bonding under the identical pressure of 5 MPa. However, as a result of the thin initial bonding layer by the short plating time, the thickness of the final bondline was very thin.

When a dummy Cu die containing the porous Cu layer was sinter-bonded on a dummy Cu substrate for 1 min under 5 MPa compression at 300 °C in air, cross-sectional SEM images of bondlines formed with different electroplating times in a die: (a) 1 min, (b) 2 min, and (c) 3 min

Fig. 4 shows the results of the shear strength according to the initial plating time for formation of a bonding layer in the formed bondlines of the specimens that were compression sintered for 1 min in air at 300 °C with 5 MPa of pressure applied. The results show that when Cu dies and substrates that were plated for 1 min, 2 min, and 3 min were used, the shear strength values were 20.1 MPa, 26.6 MPa, and 20.7 MPa, respectively, indicating that all specimens obtained excellent bonding results that exceeded the strength values required for industrial use29). The maximum shear strength value was obtained by the specimen on which the initial bonding layer was formed for 2 min. In the 3-min specimens, it is estimated that the reduction in the sintering degree between particles was a major factor for the slightly low strength. In the 1-min specimens, it is estimated that the formation of localized unbonded regions within the bondline due to the thin initial thickness of the bonding layer was a major factor.

After the sinter bonding for 1 min under 5 MPa compression at 300 °C in air, average shear strength of bondlines formed in relation to electroplating time in a die

Fig. 5 shows a comparison of the fracture microstructures that formed owing to the shear tests according to the initial bonding layer plating time. In all bonding specimens, the fractures formed inside the bondline, not on the upper or lower interface, indicating a cohesive failure mode. Looking at the fracture that formed in the specimen that was sinter-bonded after 1 min of plating (Fig. 5(a)), there was a localized ductile fracture structure with the result of a cup and cone fracture surface, which is observed in very tough metal bonds, as shown in the magnified image below. However, as indicated by the arrows, large pores were also observed in places. On the other hand, in the specimen that was sinter-bonded after 2 min of plating, there were no large pores remaining, and there were many fracture structures that were plastically deformed through elongation in the shear direction, which again confirms that a fairly fine sinter bondline was formed. Finally, the specimen that was bonded after 3 min of plating also had no large pores, but both slightly elongated fracture structures and fracture structures containing slightly sintered particles were observed in places, and it was found in the microstructure that the sintering degree had been the least overall in this specimen. Looking at the results above, it was found that some locally non-bonded microstructures and resultant coarse pores remained in the bondline of the specimen that was plated for 1 min owing to the thin thickness in the initial bonding layer. In the bond of the specimen that was plated for 3 min, a porous Cu layer was formed with coarse Cu particles, and the sintering between particles did not progress sufficiently. As a result, it was found that the 1-min specimen and the 3-min specimen each showed lower average shear strength values than the specimen that was bonded after 2 min of plating. Specifically, in the specimen that was bonded after 2 min of plating, the coarse pores were removed first due to the sufficient rearrangement and plastic deformation of the Cu particles caused by the compression, and then the sintering between the particles progressed to a certain extent so that ultimately the specimen exhibited the highest shear strength value.

 Schematics of the fracture site and different magnification SEM images of the fracture microstructure in bondlines formed with respect to electroplating time in a die: (a) 1 min, (b) 2 min, and (c) 3 min

3.3 Compression Sinter Bonding Properties When Using a Nitrogen Atmosphere

Fig. 6 shows the shear strength of the formed bondlines according to the initial plating time when the compression sinter bonding was identically performed under nitrogen blowing. When the sintering was performed while blowing nitrogen, each specimens showed improved shear strength values. In particular, a specimen with the bonding layer that was plated for 2 min obtained an excellent shear strength value of 30.3 MPa, which was a 14% increase in strength compared to when bonding was performed in air. The reason for this is expected to be due to the reduction in oxidation during the process of sinter bonding for bondline formation. Next, optical microscope images were used to analyze qualitative changes in the extent and distribution of oxidation on the fracture surfaces according to whether or not nitrogen blowing was used.

After the sinter bonding for 1 min under 5 MPa compression at 300 °C during nitrogen blowing, average shear strength of bondlines formed with different electroplating times in a die

Fig. 7 shows fracture surface images that were captured by an optical microscope after the shear tests according to the initial plating time and the atmosphere that was used when sinter bonding for 1 min at 5 MPa and 300 °C. When bonding was performed in air atmosphere, severe oxidation occurred along the edges of the Cu die in all specimens regardless of the initial plating time. On the other hand, when sinter bonding was performed while blowing nitrogen, the oxidation on the edges of the die was markedly reduced, and it was observed that oxidation was significantly reduced in the 2-min plating specimen in particular, which corresponds well with the results in Fig. 6 showing that the 2-min specimen had the largest increase in shear strength owing to nitrogen blowing. As a result of the coarse pores that remained in the 1-min plating specimen and the slow sintering between particles in the 3-min plating specimen, the partial oxidation at the die edges that was caused by the permeating oxygen could not be prevented as much as the degree in the 2-min plating specimen even when nitrogen was blown.

After the sinter bonding for 1 min under 5 MPa compression at 300 °C, optical microscopy images of the fracture microstructure in bondlines formed with respect to atmosphere type during sinter bonding and initial electroplating time in a die

The formation process of porous Cu bonding layer by electroplating, for performing the sinter bonding in this study, can be used instead of the conventional metallization coating or plating process, which forms an additional Ag layer on top of the Cu layer on the die and substrate. With the merit that excellent thermo-compression sinter bonding properties that were obtained in air or nitrogen environment can be realized without the use of pastes or film material, this process can be used as an alternative die bonding process of power devices in the future.

4. Conclusions

As an electroplating method that can be used instead of existing metallization coatings or plating process, this study deposited a porous Cu layer as the bonding parts of both a substrate and die, and then applied a reducing agent to the substrate’s plating layer and dried it. Next, the temperature was rapidly raised to 300 °C under 5 MPa of pressure to perform thermo-compression bonding for 1 min. The following results were obtained by this research.

1) When electroplating was performed for 1-3 min, it was observed that the thickness of the Cu plating layer increased greatly as the length of the plating time increased. Concurrently, the size of the formed particles increased markedly with increasing the plating time, and their microstructure changed greatly. As such, when plating was performed for 1 min, a thin porous Cu layer consisting of small dendrite-shaped particles was formed, but when plating was performed for 2 min, the dendrite shapes collapsed, and a porous Cu layer based on fractal-shaped aggregated particles was formed. Also, when plating was performed for 3 min, irregular coarse particles aggregated, and a very thick porous Cu layer formed.

2) When thermo-compression sinter bonding was performed in the air for 1 min, it was possible to obtain a fine bondline structure even after a short bonding time of only 1 min. The highest bondline shear strength value of 26.6 MPa was obtained by the specimen that was plated for 2 min. Although the densest bondline microstructure was obtained in the specimen that was plated for 1 min, the strength did not surpass that of the 2-min sample to the irregularity of its bondline structure. The specimen that was plated for 3 min also had a lower strength value than that of the 2-min specimen because its bondline microstructure was not the densest.

3) When nitrogen blowing was performed during the thermo-compression sinter bonding of the specimen that was plated for 2 min, the oxidation areas in the bondline at the edges of the die were greatly reduced, and the shear strength value was improved by around 14%, increasing to 30.3 MPa.

Acknowledgements

This work was supported by the National Research Foundation of Korea (NRF) grant funded by the Korea government (MSIT) (No. 2020M3H4A3081764). In addition, this paper was supported by Korea Institute for Advancement of Technology (KIAT) grant funded by the Korea Government (MOTIE) (P0008458, HRD Program for Industrial Innovation).

Riesz J Sotiriadis C Ambach D Donovan S Quantifying the Cost of a Rapid Transition to Electric Vehicles Appl. Energy 180 2016 287 300 https://doi.org/10.1016/j.apenergy.2016.07.131 Hertzke P Muller N Schenk S Wu T The Global Electric-Vehicle Market is Amped Up and on the Rise Mckinsey Cent. Futur. Mobil 2018 1 8 Bayerer R Higher junction Temperature in Power Modules - a Demand from Hybrid Cars, a Potential for the Next Step Increase in Power Density for Various Variable Speed Drives PCIM Europe 2008 27 29 Yadlapalli R. T Korapati A Kandipati R Balusu S. R Koritala C. S Advancement in Energy Efficient GaN Power Devices and Power Modules for Electric Vehicle Applications:A Review Int. J. Energy Res 45 9 2021 12638 12664 https://doi.org/10.1002/er.6683 Matallana A Ibarra E Kopez I Andreu J Garate J.I Jorda X Rebollo J Power Module Electronics in HEV/EV Applications:New Trends in Wide-Bandgap Semiconductor Technologies and Design Aspects Renew. Sust. Energ. Rev 113 2019 109264 https://doi.org/10.1016/j.rser.2019.109264 Roussel P Azemar J Technology, Industry and market Trends in WBG Power Module Packaging International Conference on Integrated Power Electronics Systems (CIPS 2014), Nuremberg, Germany 2014 1 3 Lee H Smet V Tummala R A Review of SiC Power Module Packaging Technologies:Challenges, Advances, and Emerging Issues IEEE J. Emerg. Sel. Top. Power Electron 8 1 2019 239 255 https://doi.org/10.1109/JESTPE.2019.2951801 Chin H. S Cheong K. Y Ismail A. B A Review on Die Attach Materials for SiC-Based High-Temperature Power Devices Metall. Mater Trans. B 41 4 2010 824 832 https://doi.org/10.1109/jestpe.2019.2951801 Drevin-Bazin A Lacroix F Barbor J. F SiC Die Attach for High-Temperature Application J. Electron. Mater 43 2014 695 701 https://doi.org/10.1007/s11664-013-2718-5 Chang T. C Lee C. C Hsieh C. P Hung S. C Cheng R. S Electrical Charateristics and Reliability Performance of IGBT Power Device Packaging by Chip Embedding Technology Micronelectron. Reliab 55 2015 2582 2588 https://doi.org/10.1016/j.microrel.2015.10.004 Li L. L Liu Z Tseng M Qi F. D Prediction of IGBT Power Module Remaining Lifetime Using the Aging State Approach Micronelectron. Reliab 102 2019 113476 https://doi.org/10.1016/j.microrel.2019.113476 Suganuma K Sakamoto S Kagami N Wakuda D Kim K. S Nogi M Low-Temperature Low-Pressure Die Attach with Hybrid Silver Particle Paste Microelectron. Reliab 52 2012 375 380 https://doi.org/10.1016/j.microrel.2011.07.088 Yu F Johnson R. W Hamilton M. C Pressurless Sintering of Microscale Silver Paste for 300 ℃ Applications IEEE Trans. Compon. Packag. Manuf. Technol 5 2015 1258 1264 https://doi.org/10.1109/TCPMT.2015.2455811 Sochi S Suganuma K Low-Temperature and Low- Pressure Die Bonding Using Thin Ag-Flake and Ag- Particle Paste for Power Devices IEEE Trnas. Compon. Packag. Technol 3 6 2013 923 929 https://doi/org/10.1109/TCPMT.2013.2256463 Zhang H Wang W Bai H Zou G Liu L Peng P Guo W Microstructural and Mechanical Evolution of Silver Sintering Die Attach for SiC Power Dvices During High Temperature Applications J. Alloys Compd 774 2019 487 494 https://doi.org/10.1016/j.jallcom.2018.10.067 Bai J. G Yin J Zhang Z Lu G. Q van Wyk J. D High-Temperature Operation of SiC Power Devices by Low-Temperature Sintered Silver Die- Attachment IEEE Trans. Adv. Packag 30 3 2007 506 510 https://doi.org/10.1109/TADVP.2007.898628 Zhao Y Mumby-Croft P Jones S Dai A Dou Z Wang Y Qin F Silver Sintering Die Attach Process for IGBT Power Module Production IEEE Applied Power Electronics Conference and Exposition (APEC 2017) Florida, USA 2017 3091 3094 https://doi.org/10.1109/APEC.2017.7931138 Kim M. S Oh C. M Hong W. S Pressureless Silver Sintering Property of SiC Device and ZTA AMB Substrate for Power Module J. Weld. Join 37 2 2019 15 20 https://doi.org/10.5781/JWJ.2019.37.2.3 Kim Y. C Yu D. Y Son J Kim J. S Kim M. S Lee C. W Byeon J. W Bang J Terminal Sinter- Bonding Using Silver Paste for Aluminum Nitride Heater J. Weld. Join 39 4 2021 384 391 https://doi.org/10.5781/JWJ.2021.39.4.6 Choi E. B Lee J. H Sub-1 min Sinter-Bonding Technique in Air Using Modified Cu Dendritic Particles for Formation of a High-Temperature Sustainable Bondline Met. Mater. Int 27 12 2021 5278 5284 https://doi.org/10.1007/s12540-020-00806-6 Hwang B. U Jung K. H Min K. D Lee C. J Jung S. B ressureless Cu-Cu Bonidng Using Hbrid Cu-Epoxy Paste and Its Reliability J. Mater. Sci. Mater. Electron 32 3 2021 3054 3065 https://doi.org/10.1007/s10854-020-05055-2 Tanaka S Yabuki A Oxidation Behavior of Copper Nanopaticles at Low Temperature Mater. Res. Bull 16 2 2011 2323 2327 https://doi.org/10.1016/j.materresbull.2011.08.043 Kamikoriyama Y Imamura H Muramatsu A Kanie K Ambient Aqueous-Phase Synsthesis of Copper Nanoparticles and Nanopastes with Low-Temperature Sintering and Ultra-High Bonding Abilities Sci. Rep 9 1 2019 1 10 https://doi.org/10.1038/s41598-018-38422-5 Lee B. S Back J. H Yoon J. W Effect of Sintering Conditions on Microstructure and Mechanical Strength of Cu Micro-Particle Sintered Joints for High-Power Semiconductor Module Applications J. Weld. Join 37 2 2019 26 34 https://doi.org/10.5781/JWJ.2019.37.2.5 Huang H. J Wu X Zhou M. B Zhang X. P A highly Reliable Die Bonding Approach for High Power Devices by Low Temperature Pressureless Sintering Using a Novel Cu Nanoparticle Paste Electronic Components and Technology Conference (ECTC 2020) Florida, USA 2020 1697 1702 https://doi.org/10.1109/ECTC32862.2020.00266 Peng Y Mou Y Liu J Chen M Fabrication of High-Stregnth Cu-Cu Joint by Low-Temperature Sintering Micron-Nano Cu Composite Paste J. Mater. Science:Mater. Electron 31 11 2020 8456 8463 https://doi.org/10.1007/s10854-020-03380-0 Zhang B Damian A Zijl J van Zeijl H Zhang Y Fan J Zhang G In-Air Sintering of Copper Nano- particle Paste with Pressure-Assistance for Die Attachment in High Power Electronics J. Mater. Sci.:Mater. Electron 32 4 2021 4544 4555 https://doi.org/10.1007/s10854-020-05196-4 Mohan K Shahane N Raj P. M Antoniou A Smet V Tumamala R Low-Temperatrue, Oragnics-Free Sintering of Nanoporous Copper for Reliable, High- Temperature and High-Power Die-Attach Interconnections IEEE Applied Power Electronics Conference and Exposition(APEC 2017), Florida, USA 2017 3083 3090 https://doi.org/10.1109/APEC.2017.7931137 Suganuma K Kim S. J Kim K. S High-Temerature Lead-Free Solders:Procperties and Possibilities JOM 61 1 2009 https://doi.org/10.1007/s11837-009-0013-y