Next Article in Journal
Anomaly Monitoring Model of Industrial Processes Based on Graph Similarity and Applications
Previous Article in Journal
Reaction Behavior of Sm and Valence State Evolution of Sm3+ During the Reduction of SmF3
 
 
Font Type:
Arial Georgia Verdana
Font Size:
Aa Aa Aa
Line Spacing:
Column Width:
Background:
Article

High Cu-Cu Bonding Strength Achievement Using Micron Copper Particles Under Formic Acid Atmosphere

1
Beijing Smartchip Microelectronics Technology Company Limited, Haidian District, Beijing 100192, China
2
School of Mechanical and Electrical Engineering, Guilin University of Electronic Technology, Guilin 541004, China
*
Authors to whom correspondence should be addressed.
Processes 2025, 13(4), 1042; https://doi.org/10.3390/pr13041042
Submission received: 16 January 2025 / Revised: 19 March 2025 / Accepted: 28 March 2025 / Published: 31 March 2025
(This article belongs to the Section Materials Processes)

Abstract

:
This study demonstrates the achievement of robust Cu-Cu bonding strength through thermocompression bonding (TCB) under a formic acid (FA) atmosphere. When subjected to sintering at 300 °C for 1 min under FA, sintering joints exhibit an average shear strength of 50.9 MPa. This strength further increases to an average of 131 MPa when the sintering duration is extended to 20 min at the same temperature under FA. Molecular dynamics simulations are employed to model the sintering behavior of copper particles of various sizes and thus understand the diffusion mechanism. The analysis of mean square displacement (MSD) and radial distribution function from these simulations suggests that the presence of small particles aids in the sintering of large ones. A copper paste, formulated by mixing micron-sized copper particles with organic solvents, is utilized in a series of experiments to explore different sintering methodologies aimed at enhancing the mechanical integrity of the sintering joints while simultaneously addressing issues associated with copper particle oxidation. Innovative strategies, including redox processes, are applied to improve the shear strength of the sintering joints and to minimize the detrimental effects of oxidation on the copper particles. Results indicate that preoxidation, which was used to form a nano surface structure, and using an FA atmosphere, remarkably enhance the shear strength of the Cu-Cu joints created via TCB. The findings of this research are pivotal for the advancement of rapid Cu-Cu bonding techniques using micron-scale copper pastes and can have profound implications for the development of future electronic packaging and interconnection technologies.

1. Introduction

In pursuit of advanced interconnected solutions for third-generation semiconductor power devices, such as SiC and GaN, characterized by their high-power density, superior thermal conductivity, and wide bandgaps, the industry seeks interconnected materials that can withstand high-temperature operation. Power devices operate under high-temperature and high-frequency conditions, where the working temperature often exceeds the melting point of tin-based solders, thereby limiting the application of tin-based solders in the field of power devices. The sintering of Ag and Cu particles emerged as a promising technique in power device packaging due to its high thermal flux and reliability [1,2,3,4]. Ag is the metal with the highest electrical conductivity, while Cu ranks second. Both exhibit excellent thermal conductivity and can withstand high current densities. The sintered interconnection structures formed by Ag paste and Cu paste can meet the requirements for high-temperature service. Ag paste demonstrates superior performance, and its electrical conductivity, thermal conductivity, and mechanical properties have been proven in wide-bandgap semiconductor packaging. However, the high cost of Ag paste, along with issues such as electromigration-induced short-circuit failures and thermal coefficient mismatch, limits its industrial application [5,6,7]. By contrast, Cu-based particle pastes are a promising low-cost alternative to Ag-based particle pastes. They exhibit similar resistivity and thermal conductivity, but Cu offers greater durability and lower ion migration tendency [8,9,10,11,12,13]. Additionally, Cu paste possesses the notable advantage of strong resistance to electromigration, making it a subject of extensive research.
However, Cu is a readily oxidizable material and Cu oxides are not readily decomposable. Moreover, Cu oxide hinders the sintering of Cu paste and reduces the mechanical and electrical properties of Cu sintered joints. Currently, the oxidation tendency of Cu remains an important obstacle to the formation of high-quality Cu-Cu bonding [14].
To address these challenges, many works explored the enhancement of Cu’s antioxidation properties through the application of organic coatings derived from organic solvent. A specific copper paste formulation, incorporating terpineol with copper particles, demonstrated solder joint shear strengths of 23 MPa post a 300 °C, 30 min sintering process [15].
Further refinement in the nanoparticle size distribution within the paste led to the development of a mixed copper paste, which, under a sintering condition of 250 °C and 4 MPa pressure, can achieve joint shear strengths of approximately 15 MPa [16]. To improve the shear strength of Cu-Cu bonded joints, experiments were conducted on nano-copper slurries using different sintering processes. The sintered joint heated at 100 °C in a low-temperature vacuum can achieve an average strength of 40 MPa [17]. Not only have researchers studied improving the shear strength of sintered joints, but they also employed quasi in situ methods to investigate the grain growth and twinning formation mechanisms of sintered copper nanoparticles [18]. In addition, the sintered joint obtained by sintering micro and nano mixed copper paste under a nitrogen environment, 0.4 MPa auxiliary pressure, and 350 °C reached 40 MPa [19].
While micro-sized copper particles are favored for their ease of production and inherent antioxidative qualities, their large size and low surface energy present sintering challenges. A novel approach involving a mixture of micro- and nano-scale particles has been shown to increase the density of the sintered material [20]. Additionally, in situ surface modification techniques, specifically oxidation reduction bonding (ORB), have been employed to enhance the sintering strength of micro-sized copper particles. This process includes a preoxidation step at 300 °C for 30 min, followed by reduction bonding in a formic acid (FA) atmosphere at the same temperature for 60 min, culminating in solder joint shear strengths of up to 31 MPa [21].
Despite these advancements, the shear strengths of Cu-Cu bonded joints remain inferior to traditional tin–lead solder joints [22]. Due to the problems of tin whisker growth, bridging failure, tin solder overflow short circuit, electromigration, and thermal cycling in traditional tin–lead solder joints [23]. Copper paste has high electrical resistivity and thermal conductivity, as well as excellent resistance to electromigration, making it widely studied [8,9,10,11,12].
However, the application of thermocompression bonding (TCB) in conjunction with an in situ FA vapor treatment has been shown to increase the mean shear strength of Cu-Cu bonds to over 150 MPa [24]. Most of these previous studies were aimed at improving the shear performance of sintered joints by using different copper paste formulations and sintering processes to enhance the shear strength of sintered joints. This study builds upon these findings, leveraging micrometer-sized copper particles in copper paste formulations and combining ORB and TCB methods to refine the process parameters that govern the mechanical properties of Cu-Cu bonded joints.
The purpose of this study is to investigate different sintering processes after preparing copper paste and ultimately form high-performance Cu-Cu bonding joints. Firstly, simulate sintering experiments using copper particles of different sizes. Secondly, analyze the movement of copper particles. Lastly, select micron-sized copper particles with a size range distribution for sintering experiments. The combination of ORB and TCB processes is used to improve the sintering performance of micro copper paste, aiming to achieve optimal mechanical strength. In addition, an FA atmosphere is used to address the challenge of easy oxidation of copper particles. Results reveal that the ORB process and FA atmosphere remarkably enhance the TCB process. This study contributes a valuable reference for expediting the development of Cu-Cu bonding in electronic packaging.

2. Materials and Methods

2.1. Simulation Methodology

Currently, the molecular dynamics (MD) simulation method is in the lead in academic and industrial circles. To study and observe the atomic motion during the sintering process of materials, many researchers [25] used the MD simulation method to simulate the sintering process. In this study, the diffusion motion of copper particles during sintering is explored via MD simulation. Many researchers investigated the formation mechanism of nanocrystal coalescence during sintering by simplifying the model. The results reveal that once the two nanoparticles are brought into contact, they often go through drastic structural changes with the inter-particle grain boundary quickly eliminated [26].
This study adopted the simplified model approach to study the sintering process of nanosized copper particles growing on the surface of micron-sized copper particles. The NVT ensemble (i.e., constant number of atoms, constant volume, and constant temperature) with an initial temperature of 300 K, a running time of 3500 ps, and a time step of 0.5 fs was chosen, and then energy was minimized to study the effect of temperature on the system. The nano-copper particles simulate the sintering process. Copper particles with a diameter of 10 and 100 Å are created, and the model is shown in Figure 1. The 10 Å copper particles pair the sintering process between nano-copper granules that simulate Cu microparticles, 10 and 100 Å copper particles on the surface of the Cu microparticles, and the sintering process between 10 and 100 Å copper particles.
The 10 Å particles have a total of 89 copper atoms, the 10 Å and 100 Å particles have a total of 44,571 copper atoms, and the 100 Å copper particles have a total of 89,8043 copper atoms. First, simulations are conducted on particles of different sizes. Subsequently, regular compression and non-periodic boundary conditions are selected, and the velocity form of the Verlet algorithm is chosen for the equations of motion. Finally, the results of different copper particles during the sintering process are analyzed. Visualization software OVITO [27] was used to analyze the change in the copper granular sintering process.
LAMMPS [28] is a software developed by the key laboratory of the US National SANDIA National Laboratory. It supports the embedded atom method [29], which divides the general trend of the entire simulation system into the interaction between the nucleus and the electron cloud in the lattice. The potential energy is
E i = F j j i N ( ρ j r i j ) + 1 2 j i N ( φ i j r i j ) .
In the formula, Ei is the embedded energy of i atom; rij is the distance between the atom i and j; Fj is the embedded function of the electronic density; ρj is the electronic density of the j atom at the atom i; and φij is a short-range potential function.
The radial distribution function (RDF) [30] can be employed to characterize the distribution state of atoms within a system. The RDF reflects the degree of atomic order during sintering. The higher the atomic density, the lower the randomness of the spatial arrangement of atoms. A high density corresponds to a low randomness in the spatial arrangement of atoms. The peak position, intensity, and broadening of the RDF curve disclose the local and global structural characteristics of the material, encompassing atomic spacing, order, and disorder. Researchers [31] utilized the RDF to investigate the influence of temperature on materials. In this study, the RDF is applied to analyze the dynamics during sintering.
A relationship exists between the diffusion of the first peak in the RDF curve and the degree of disorder of the peak. The broadening of the peak reflects the distribution range of atomic distances. The greater the broadening, the more uneven the distribution of atomic distances and the more disordered the local structure. The first peak reflects the distance and coordination number of the nearest atoms. The higher the peak intensity, the more atomic pairs present at this distance, and the more ordered the structure. The formula is
G ( r ) = d N ρ 4 π r 2 d r .
N is the total number of atoms in the formula; it is the ratio of the local density to the average density of the atomic; the G(r) is the particle distribution function of the atomic distance r.
The balance of the root-based root displacement MSD [32] reflects the thermal vibration and diffusion form of the atom and analyzes the change in copper particles during sintering. The formula is
R M S D = 1 N i = 1 N [ r i ( t ) r i ( 0 ) 2 ] .
In the formula, N is the total number of atoms in the entire system; ri(t) is the position of the i atom under time t.

2.2. Experimental Procedure

The experiment was conducted to investigate the sintering process of micron-sized copper particles, employing commercially available large spherical copper particles with a size range of 1–100 μm (Jincai Shengte Metal Materials Co., Ltd., Hebei, China). Copper particles were mixed with terpene alcohol at a mass ratio of 8:2 to prepare a slurry, with stirring performed to mix the copper paste thoroughly and remove any air bubbles. Given the high cost of power chips, copper blocks were used to simulate power chips in the upper position and substrates in the lower position to achieve Cu-Cu bonding, thereby reducing expenses. Subsequently, a spring clamp was utilized to fix the interconnected structure and provide pressure, controlling the pressure values through the compression distance and measuring the pressure values of the spring clamp under identical compression distances using a pressure sensor. Pressure values were repeatedly tested to minimize the pressure required for the formation of the interconnected structure, with pressures above 226 MPa deemed sufficient to form the interconnected structure. This sintering-assisted pressure was adopted for the experiments. Once the Cu-Cu structures were secured by the clamps, they were subjected to sintering within a sintering furnace. The chamber of the sintering furnace could control various temperature and atmosphere conditions. When FA gas was introduced, the FA atmosphere facilitated the reduction in copper particles, as specified by the chemical Equation (4). Upon completion of the sintering process, the Cu-Cu structures underwent shear strength testing. The schematic illustrating the experimental procedure is presented in Figure 2.
C u 2 O + H C O O H ( g ) 2 C u + H 2 O ( g ) + C O ( g )

2.3. Sintering Process

Given that pressureless or low-pressure sintering of micron copper paste fails to achieve interconnected structures with excellent mechanical properties, this study utilized the TCB process to perform the relevant sintering experiments. Sintering was performed using a sintering furnace (RS220, Zhongke Comrade Technology Co., Ltd., Beijing, China), as illustrated in Figure 3a,b, which depicts the sintering processes involved in the ORB and TCB techniques, focusing on the investigation of the effects of preoxidation times, reduction times, and different atmosphere conditions on the mechanical properties of the joints. Figure 3b illustrates the sintering of the TCB technique, examining the influence of varying sintering parameters on the mechanical properties of the Cu-Cu sintered joints.
Table 1 and Table 2 show experimental parameters. Three samples were tested under each sintering condition, and their average shear strength was determined. The sintering joint was performed, and the shear strength was evaluated using the shear test machine (MFM1200). The shear speed was 200 μm/s. SEM was used to analyze the microstructure obtained by the interconnected structure.

3. Results and Discussion

3.1. Effect of Oxidatively Grown Small Copper Particles on the Sintering Process

Aiming at the sintering process in which it is difficult to observe preoxidation to generate small copper particles, it built different sizes of copper granules and analyzed the effects of small copper particles that oxidize growth on large copper particles during sintering. The MD simulation method was employed to establish copper particle models of different sizes for simulating the sintering process and to investigate the influence of small copper particles formed through oxidation growth on large copper particles during sintering. The slope of the mean square displacement (MSD) curve represents the diffusion coefficient D, which reflects the diffusion rate of particles. Currently, the MSD [33] is utilized to study particle diffusion.
Figure 4a,b shows the MSD curves and RDF curves of copper particles of different sizes during sintering. In the MSD curves of Figure 4a, an increase in temperature enhances atomic vibrations [34]. The MSD curves indicate that during sintering, heating intensifies atomic vibrations, increases the contact area between particles, reduces the surface energy, and drives the system toward a more stable state. In Figure 4a, two 10 Å copper particles reach a stable vibration state first, followed by the 10 and 100 Å particle models, and finally the two 100 Å particle models. Before reaching the stable state, the slope of the MSD curve represents the diffusion coefficient D. The larger the diffusion coefficient D, the more intense the atomic vibrations. The MSD curves suggest that smaller copper particles are easier to sinter, and they facilitate the sintering of larger particles. Larger particles are more difficult to sinter but have better oxidation resistance and higher stability. Sintering larger copper particles requires higher temperatures, which makes it more challenging to meet the requirements of “low temperature sintering and high temperature service” for power device sintering. Simulation results indicate that mixing large and small copper particles can promote the sintering process. Considering oxidation resistance and cost, small copper particles formed by oxidation can enhance the sintering of larger copper particles.
Figure 4b shows the RDF curves. There is a certain relationship between the broadening of the first peak of the RDF curve and the degree of peak disorder [31]. By comparing the broadening of the first peak of the RDF curves, the two 10 Å copper particles exhibit the largest broadening, which indicates that the distribution of atomic distances is more uneven, the local structure is more disordered, and the degree of disorder is higher. The 10 and 100 Å copper particles follow next in terms of broadening, while the two 100 Å copper particles show the smallest broadening, corresponding to the lowest degree of disorder. Comparing the intensity of the first peak of the RDF curves, the two 10 Å copper particles have a high peak intensity, suggesting that the number of nearest-neighbor atomic pairs is relatively large, and a certain degree of local structural order exists at short-range scales. The number of nearest-neighbor atomic pairs in the 10 and 100 Å copper particles is greater than that in the two 100 Å copper particles, indicating that adding small-sized copper particles during sintering can enhance the local structural order at short-range scales and increase the disorder of the local structure at larger scales.
Figure 5 shows the changes in the potential energy of the sintering process of 100 and 10 Å copper particles. The general potential curve of the copper particles suddenly increases within the two-temperature range, breaking the linear crease, indicating that the phase changes occurred. The first stage is 300–325 K. At this stage, large particles and small particles are fused. The small particles are active close to large particles and enter into large particles. The second stage is 1240–1355 K. The particles are melted to reach the melting point of the large particles, and the result shows that the melting point is 1357.77 K. The melting point of the material can be obtained from the potential energy results of the simulation using the LAMMPS software. The melting point is the temperature of the atomic energy of the entire particles to return to the temperature when the stable slope [25]. Figure 5 displays the internal structure of the copper granules. The FCC crystal type becomes a white unskilled crystal type, and the internal is still a green FCC crystal type. As the temperature reaches the 1240 K phase change point, the crystal shape changes. When the temperature rises to 1355 K, the copper particles are all white unskilled crystals inside and outside. Results show that the generated small particles promote large particle sintering. Common sintering can increase phase change points, and increasing the general trend can help the sintering process.

3.2. Effect of Preoxidation Time on TCB

The ORB process enables the enhancement of the shear strength of interconnected joints through surface modification of copper particles [21]. During the preoxidation stage, Cu2O oxide forms on the particle surfaces, with Cu2O oxidation products nucleating and growing into Cu nanoparticles. These nanoparticles on the copper particle surfaces act as sintering promoters [35]. An oxide film grows on the particle surfaces, allowing Cu ions to diffuse outward. When the oxide film reaches a certain thickness, Cu is prevented from further contact with oxygen, and instead, oxygen diffuses inward through cracks in the film while copper ions continue to diffuse outward. The formation of the surface oxide film impedes the sintering process of copper particles, which can be mitigated by reacting FA gas with the oxide layer on the surface of Cu nanoparticles, forming an amorphous cupric formate layer that subsequently decomposes to form copper atoms, thereby reducing the Cu2O film back to Cu [36]. Investigating the effect of preoxidation time on the TCB process warrants further study. Analysis of how to improve the mechanical properties of sintered joints under varying ORB processes led to the design of experiments for the preparation of copper pastes and their subsequent sintering and the analysis of the mechanical properties and microstructural characteristics of the sintered joints.
Figure 6 illustrates the shear strength of Cu-Cu sintered structures obtained under varying ORB processes. Figure 6a exhibits the results of a preoxidation process conducted at 250 °C under an N2 atmosphere, followed by reduction sintering at 300 °C for 20 min under an FA atmosphere. Figure 6b demonstrates the sintering performed at 300 °C for 20 min without a preoxidation step under different atmosphere conditions. The results reveal that the average shear strength of the sintered joints subjected to a 20 min preoxidation process is 144.97 MPa. The shear strength decreases as the preoxidation time is reduced, with the lowest average shear strength of 128 MPa observed in the absence of preoxidation, indicating that a preoxidation duration of 20 min is most suitable. Throughout the preoxidation process, longer preoxidation times do not necessarily lead to enhanced mechanical properties; the average shear strength after 30 min of preoxidation is lower than that following 20 min of preoxidation, suggesting that an appropriate preoxidation duration can facilitate the TCB process and enhance the mechanical properties of the sintered joints. When the TCB process is performed at 300 °C for 20 min without preoxidation, the average shear strength is compared under FA and N2 atmospheres. As shown in Figure 6b, the average shear strength under an FA atmosphere is 128 MPa, while under an N2 atmosphere, it is 119.53 MPa, indicating that an FA atmosphere promotes the sintering process to some extent, thereby enhancing the mechanical properties of the interconnected layer. After the thermal oxidation process and subsequent TCB sintering, mechanical polishing of the interconnect layer is conducted. As depicted in Figure 6c, measurements taken using the software Nano Measurer indicate that 69.23% of the solder layer is distributed between 0.5 and 0.6 μm, and 15.38% is distributed between 0.4–0.5 and 0.6–0.7 μm, with an average thickness of 0.54 μm. The solder layer is extremely thin and exhibits clear bulk copper metallization, indicating that the solder layer formed is denser than sintered joints produced by other methods [37]. This result suggests that the TCB process achieves an extremely thin interconnected layer with excellent mechanical properties.
Figure 7 presents SEM images of the cross-sectional morphologies of Cu-Cu sintered structures obtained under different sintering processes. Figure 7a–c depicts the cross-sectional morphologies of Cu-Cu joints subjected to preoxidation for 20 min at 250 °C followed by FA reduction sintering for 20 min at 300 °C. Figure 7d–i illustrates the cross-sectional morphologies of Cu-Cu joints sintered for 20 min at 300 °C under various atmosphere conditions. Figure 7a–c demonstrates that the majority of particle shapes are truncated, with the fracture surfaces presenting numerous small hill-like tensile morphologies, indicating that plastic tensile deformation occurs at the particle fracture surfaces. The coalescence Cu particles show remarkable stretch under shear forces, indicating the ductile fractures’ process during shear testing [38,39,40]. Additionally, large areas of bulk copper exhibit tensile morphologies with a flowing appearance, and the fracture surfaces of particles display many micro-nanoparticle aggregates. Nanoparticles growing on the particle surfaces show signs of fractured sinter necks, suggesting effective diffusion between particles and a high degree of particle-to-particle fusion. Figure 7d–f reveals fracture surfaces with numerous small hill-like dimple tensile morphologies, smooth surfaces of bulk copper with minor quantities of micro-nanoparticles, and a profusion of sharp-ended elongated ductile fracture traces observable on the fractured surfaces, indicating the formation of a dense connection and a robust interconnected structure. Figure 7g–i shows that the surfaces of bulk copper contain many protruding micro-copper particles, with fewer hill-like morphologies present on the fracture surfaces of copper particles. Micro-copper particles cover the fracture surfaces of copper particles and bulk copper, indicating the formation of a micro-interconnected structure arising from the growth of micro-copper particles. Consequently, the experimental results suggest that during preoxidation and sintering, micro-copper particles form on the surfaces of copper particles. These micro-copper particles endow the copper particles with increased surface energy and reduced fusion heat, facilitating the formation of bulk copper microstructures with plastic deformation under FA or N2 atmospheres [41]. To compare the shear strength results of the Cu-Cu sintered structures, it is indicated that the preoxidation process does not considerably affect the TCB process. However, FA atmosphere promotes the reduction in micro-oxide particles on the surfaces of copper particles, enhancing the sintering process of oxidized copper particles. Oxide layers on the surfaces of copper atoms impede the sintering process, resulting in the mechanical properties of Cu-Cu sintered structures obtained under the N2 atmosphere being inferior to those obtained under an FA atmosphere.

3.3. Effect of Different Sintering Environment on TCB

An analysis of the shear strength and microstructure of sintered joints was conducted to address the challenge associated with the large size of micrometer copper particles that hinders their sintering, employing methods to enhance the mechanical properties of the sintered joints through TCB processes. The sintering temperature, time, and atmosphere were altered, and shear mechanical property tests were performed on the sintered samples to investigate the impact of varying TCB processes on the mechanical properties of the sintered joints. As illustrated in Figure 8, Figure 8a,b shows the performance of Cu-Cu interconnected structures obtained after sintering for 20 min at different temperatures under various atmosphere conditions. Figure 8c,d illustrates the mechanical performance of Cu-Cu interconnected structures obtained under different atmosphere conditions and sintering times at a sintering temperature of 300 °C. High-performance interconnected structures could not be achieved at 200 °C for 20 min. To attain enhanced interconnectivity, the duration of sintering could be increased or the temperature raised. The shear strengths were found to be low at 200 °C for 20 min and 250 °C for 20 min. The shear strength of Cu-Cu sintered structures under an FA atmosphere was superior to that under an N2 atmosphere. At low temperatures or short sintering times, FA molecules did not react sufficiently with copper oxides to promote the sintering of copper particles [36]. An increase in temperature, an extension of sintering time, and the utilization of FA atmosphere were all conducive to improving the mechanical properties of the sintered joints.
When the sintering temperature is set at 300 °C, the mechanical properties are compared for sintering durations of 5, 10, and 20 min, revealing a notable increase in the average shear strength. This increase indicates that extending the sintering time yields sintered joints with enhanced mechanical properties. As the temperature rises, FA molecules become more active, decomposing the oxides on the surface of copper particles, thereby enhancing the mechanical properties of the sintered body. At a sintering temperature of 300 °C, an average shear strength of 51.39 MPa is achieved under N2 after a sintering time of 1 min, while an average shear strength of 66.53 MPa is observed under an FA atmosphere. For TCB, when the temperature reaches 300 °C, active FA molecules enter the sintering interconnected layer region, promoting the reduction in copper particles. This process involves the reduction in copper (II) oxide or copper (I) oxide particles on the particle surfaces back into copper particles and the growth of ultrafine nano-copper particles with higher activation energy [42], thus fostering the sintering fusion and the formation of bulk copper. This result indicates that under low sintering temperatures, the FA atmosphere has minimal effect on the formation of interconnected joints via TCB. However, under high temperatures, the FA atmosphere remarkably enhances the mechanical properties of the interconnected joints.
Figure 9 illustrates cross-sectional SEM images obtained after sintering at 300 °C for 1 min under various atmospheric conditions. Figure 9a–c depicts the sintered cross-sections under the FA atmosphere, showing abundant ductile fracture traces, sheared-off tensile-shaped sinter necks, and filamentary structures of Cu oxidation on the joint’s cross-section, which reduce particle porosity [15,43]. The cross-section exhibits a fracture of bulk copper, with a few particle agglomerates visible. After particle growth and fusion, the volume increases, and the particle surfaces contain a few micro-nanoparticles, with sintered neck structures present between the particle surfaces. Figure 9d–f shows the cross-sections formed under an N2 atmosphere, presenting a relatively flat surface within the particles. Interparticle connectivity is evident, and the cross-section reveals slip bands with a morphology resembling small hills indicative of tensile deformation. Distinct sintered neck structures form among the micro-nano-copper particles distributed on the particle surfaces, with a greater abundance of micro-nano-copper particles on the particle surfaces compared with those treated under the FA atmosphere [44]. Under an FA environment, the process of sintered neck formation and the reduction in surface micro-nano-copper particles occur simultaneously. After reduction, the micro-nano-copper particles possess high surface energy, which facilitates the fusion of large copper particles into a large-area bulk copper structure. When sintered in N2, numerous micro-nano-copper particles form on the particle surfaces, and the growth of these micro-nano-copper particles aids sintering. The formation of sintered necks between particles, along with tensile and extensive particle fracturing within the cross-section, indicates the occurrence of plastic deformation.

4. Conclusions

In this study, the influence of preoxidation time on TCB and that of sintering atmosphere, time, temperature, and other parameters on the shear strength of sintering joints are investigated to realize the effectiveness of fast and high-quality sintering interconnected structures.
(1)
As the sintering temperature increases, the motion mode changes from atomic vibration to diffusion. The smaller the copper particle size, the easier the sintering, and the smaller copper particle size facilitates the sintering process of the larger copper particle size. The introduction of small copper particle sintering can improve the local structure orderliness on a short-range scale and increase the local structure disorderliness on a large scale. Small copper particles can increase the phase change point, increase the total potential energy, and promote the sintering of copper particles. The oxidized small particles on the surface of micron copper particles can promote the sintering process of micron copper particles.
(2)
Preoxidation has minimal effect on the improvement of the mechanical properties of TCB. TCB provides extremely thin and excellent mechanical properties. The nail acid atmosphere helps to reduce micro-oxide particles on the particle surface and promote sintering of oxide particles. Oxides on the surface of copper atoms impede sintering, making the mechanical properties lower than those of the environment.
(3)
Different redox processes have a limited effect on the mechanical properties of low-temperature and high-pressure sintering joints, and the sintering joints are ductile fractures. TCB can realize short-term sintering and obtain sintering joints with excellent mechanical properties, thereby providing theoretical support for the rapid realization of Cu-Cu bond exploration in the future.

Author Contributions

Conceptualization, D.L. (Dengjian Li); methodology, B.L., D.L. (Dengjian Li) and Y.L.; software, Y.L. and S.H. (Siliang He); validation, M.C. and D.L. (Dengjian Li); formal analysis, B.L. and Y.L.; investigation, S.H. (Shunfeng Han); resources, S.H. (Shunfeng Han) and D.L. (Dameng Li); data curation, D.L. (Dameng Li) and B.Y.; writing—original draft preparation, B.L. and B.G.; writing—review and editing, S.H. (Shunfeng Han), S.H. (Siliang He), B.Y. and B.G.; visualization, B.L., B.G. and D.L. (Dameng Li); supervision, B.L., S.H. (Shunfeng Han), M.C. and D.L. (Dengjian Li); project administration, B.L., D.L. (Dameng Li), B.Y. and D.L. (Dengjian Li); funding acquisition, B.G. and D.L. (Dengjian Li). All authors have read and agreed to the published version of the manuscript.

Funding

This research was supported by the Laboratory Specialized Scientific Research Projects of Beijing Smart-chip Microelectronics Technology Co., Ltd. (No. SGSC0000SJQT2400942).

Data Availability Statement

The original contributions presented in this study are included in the article. Further inquiries can be directed to the corresponding author.

Conflicts of Interest

Authors Bofu Li, Dejian Li, Dameng Li, Baobin Yang, Baoliang Gong, Shunfeng Han were employed by the company Beijing Smartchip Microelectronics Technology Company Limited. The remaining authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest. The company had no role in the design of the study; in the collection, analyses, or interpretation of data; in the writing of the manuscript, or in the decision to publish the results.

Abbreviations

The following abbreviations are used in this manuscript:
TCBThermocompression bonding
FAFormic acid
MSDMean square displacement
ORBOxidation reduction bonding
RDFRadial distribution function
SEMScanning electronic microscope

References

  1. Loulijat, H.; Zerradi, H.; Mizani, S.; Achhal, E.M.; Dezairi, A.; Ouaskit, S. The behavior of the thermal conductivity near the melting temperature of copper nanoparticle. J. Mol. Liq. 2015, 211, 695–704. [Google Scholar] [CrossRef]
  2. Feng, D.; Feng, Y.; Yuan, S.; Zhang, X.; Wang, G. Melting behavior of Ag nanoparticles and their clusters. Appl. Therm. Eng. 2017, 111, 1457–1463. [Google Scholar] [CrossRef]
  3. Zhang, H.; Wang, Y.; Li, X.; Chen, M. Improved thermal conductivity and reliability through graphene reinforced nanopaste for power devices in new energy vehicles. IEEE Trans. Compon. Packag. Manuf. Technol. 2024, 14, 52–60. [Google Scholar] [CrossRef]
  4. Qu, G.; Guo, W.; Zhang, C.; Xue, J.; Peng, Z.; Yin, C.; He, S.; Zou, G.; Jia, Q.; Zhang, H. Improving thermal stability and reliability of power chips by sintering foam structure layer. Appl. Mater. Today 2024, 40, 102397. [Google Scholar]
  5. Yamagiwa, D.; Matsuda, T.; Furusawa, H.; Sato, K.; Tatsumi, H.; Sano, T.; Kashiba, Y.; Hirose, A. Pressureless sinter joining of bare Cu substrates under forming gas atmosphere by surface-oxidized submicron Cu particles. J. Mater. Sci. Mater. Electron. 2021, 32, 19031–19041. [Google Scholar]
  6. Zuo, Y.; Carter-Searjeant, S.; Green, M.; Mills, L.; Mannan, S.H. High bond strength Cu joints fabricated by rapid and pressureless in situ reduction-sintering of Cu nanoparticles. Mater. Lett. 2020, 276, 128260. [Google Scholar]
  7. Mou, Y.; Peng, Y.; Zhang, Y.; Cheng, H.; Chen, M. Cu-Cu bonding enhancement at low temperature by using carboxylic acid surface-modified Cu nanoparticles. Mater. Lett. 2018, 227, 179–183. [Google Scholar] [CrossRef]
  8. Yan, J.; Zou, G.; Hu, A.; Zhou, Y.N. Preparation of PVP coated Cu NPs and the application for low-temperature bonding. J. Mater. Chem. 2011, 21, 15981–15986. [Google Scholar]
  9. Jeong, S.; Woo, K.; Kim, D.; Lim, S.; Kim, J.S.; Shin, H.; Xia, Y.; Moon, J. Controlling the thickness of the surface oxide layer on Cu nanoparticles for the fabrication of conductive structures by inkjet printing. Adv. Funct. Mater. 2008, 18, 679–686. [Google Scholar] [CrossRef]
  10. Nishikawa, H.; Hirano, T.; Takemoto, T.; Terada, N. Effects of joining conditions on joint strength of Cu/Cu joint using Cu nanoparticle paste. Open Surf. Sci. J. 2011, 3, 60–64. [Google Scholar]
  11. Cheon, J.; Lee, J.; Kim, J. Inkjet printing using copper nanoparticles synthesized by electrolysis. Thin Solid Film. 2012, 520, 2639–2643. [Google Scholar] [CrossRef]
  12. Ishizaki, T.; Watanabe, R. A new one-pot method for the synthesis of Cu nanoparticles for low temperature bonding. J. Mater. Chem. 2012, 22, 25198–25206. [Google Scholar] [CrossRef]
  13. Zuo, Y.; Shen, J.; Xie, J.; Xiang, L. Influence of Cu micro/nano-particles mixture and surface roughness on the shear strength of Cu-Cu joints. J. Mater. Process. Technol. 2018, 257, 250–256. [Google Scholar] [CrossRef]
  14. Zuo, Y.; Shen, J.; Hu, Y.; Gao, R. Improvement of oxidation resistance and bonding strength of Cu nanoparticles solder joints of Cu–Cu bonding by phosphating the nanoparticle. J. Mater. Process. Technol. 2018, 253, 27–33. [Google Scholar] [CrossRef]
  15. Gao, R.; He, S.; Li, J.; Shen, Y.-A.; Nishikawa, H. Interfacial transformation of preoxidized Cu microparticles in a formic-acid atmosphere for pressureless Cu–Cu bonding. J. Mater. Sci. Mater. Electron. 2020, 31, 14635–14644. [Google Scholar]
  16. Zuo, Y.; Shen, J.; Xu, H.; Gao, R. Effect of different sizes of Cu nanoparticles on the shear strength of Cu-Cu joints. Mater. Lett. 2017, 199, 13–16. [Google Scholar] [CrossRef]
  17. Yamakawa, T.; Takemoto, T.; Shimoda, M.; Nishikawa, H.; Shiokawa, K.; Terada, N. Influence of joining conditions on bonding strength of joints: Efficacy of low-temperature bonding using Cu nanoparticle paste. J. Electron. Mater. 2013, 42, 1260–1267. [Google Scholar]
  18. Zuo, Y.; Zhao, C.; Robador, A.; Wickham, M.; Mannan, S.H. Quasi-in-situ observation of the grain growth and grain boundary movement in sintered Cu nanoparticle interconnects. Acta Mater. 2022, 236, 118135. [Google Scholar]
  19. Gao, Y.; Zhang, H.; Li, W.; Jiu, J.; Nagao, S.; Sugahara, T.; Suganuma, K. Die bonding performance using bimodal Cu particle paste under different sintering atmospheres. J. Electron. Mater. 2017, 46, 4575–4581. [Google Scholar] [CrossRef]
  20. Liang, P.; Pan, Z.; Tang, L.; Zhang, G.; Yang, D.; He, S.; Yan, H. Molecular dynamics simulation of sintering densification of multi-scale silver layer. Materials 2022, 15, 2232. [Google Scholar] [CrossRef]
  21. Liu, X.; Nishikawa, H. Low-pressure Cu-Cu bonding using in-situ surface-modified microscale Cu particles for power device packaging. Scr. Mater. 2016, 120, 80–84. [Google Scholar] [CrossRef]
  22. Namgoong, D.; Siow, K.S.; Lee, J.H. Improvement of bondability by addition of carboxylic acid to the sinter-bonding paste containing bimodal-sized Cu particles and rapid bonding in air. Met. Mater. Int. 2023, 29, 457–466. [Google Scholar] [CrossRef]
  23. Kim, D.; Chen, C.; Lee, S.-J.; Nagao, S.; Suganuma, K. Strengthening of DBA substrate with Ni/Ti/Ag metallization for thermal fatigue-resistant Ag sinter joining in GaN power modules. J. Mater. Sci. Mater. Electron. 2020, 31, 3715–3726. [Google Scholar]
  24. Jangam, S.C.; Bajwa, A.A.; Mogera, U. Fine-Pitch (≤10 m) Direct Cu-Cu Interconnects Using In-Situ Formic Acid Vapor Treatment. In Proceedings of the 2019 IEEE 69th Electronic Components and Technology Conference (ECTC), Las Vegas, NV, USA, 28–31 May 2019; pp. 620–627. [Google Scholar]
  25. Song, Z.; Luo, W.; Fan, X.; Zhu, Y. Atomic fast dynamic motion on the Cu Nanoparticle’s surface before melting: A molecular dynamics study. Appl. Surf. Sci. 2022, 606, 154901. [Google Scholar]
  26. Bingqing, C.; Alfonso, H.; Ngan, W. The crystal structures of sintered copper nanoparticles: A molecular dynamics study. Int. J. Plast. 2013, 47, 65–79. [Google Scholar]
  27. Stukowski, A. Visualization and analysis of atomistic simulation data with OVITO—The open visualization tool. Model. Simul. Mater. Sci. Eng. 2010, 18, 015012. [Google Scholar]
  28. Plimpton, S. Fast parallel algorithms for short-range molecular dynamics. J. Comput. Phys. 1995, 117, 1–19. [Google Scholar]
  29. Mishin, Y.; Mehl, M.J.; Papaconstantopoulos, D.A.; Voter, A.F.; Kress, J.D. Structural stability and lattice defects in copper: Ab initio, tight-binding, and embedded-atom calculations. Phys. Rev. B Condens. Matter Mater. Phys. 2001, 63, 224106. [Google Scholar]
  30. Li, Q.; Liu, C.; Chen, X. Molecular characteristics of dissociated water with memory effect from methane hydrates. Int. J. Mod. Phys. B 2014, 28, 1450062. [Google Scholar] [CrossRef]
  31. Jiang, J.; Chen, P.; Sun, W. Monitoring micro-structural evolution during aluminum sintering and understanding the sintering mechanism of aluminum nanoparticles: A molecular dynamics study. J. Mater. Sci. Technol. 2020, 57, 92–100. [Google Scholar] [CrossRef]
  32. Jiang, S.; Zhang, Y.; Gan, Y.; Chen, Z.; Peng, H. Molecular dynamics study of neck growth in laser sintering of hollow silver nanoparticles with different heating rates. J. Phys. D Appl. Phys. 2013, 46, 335302. [Google Scholar]
  33. Marquardt, R. Mean square displacement of a free quantum particle in a thermal state. Mol. Phys. 2021, 119, 17–18. [Google Scholar]
  34. Malti, A.; Kardani, A.; Montazeri, A. An insight into the temperature-dependent sintering mechanisms of metal nanoparticles through MD-based microstructural analysis. Powder Technol. 2021, 386, 30–39. [Google Scholar]
  35. 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. 2021, 27, 5278–5284. [Google Scholar]
  36. Li, J.; Xu, Y.; Zhao, X.; Meng, Y.; Yin, Z.; Wang, Y.; Suga, T. Enhancement and mechanism of copper nanoparticle sintering in activated formic acid atmosphere at low temperature. ECS J. Solid State Sci. Technol. 2021, 10, 054004. [Google Scholar]
  37. Li, J.; Yu, X.; Shi, T.; Cheng, C.; Fan, J.; Cheng, S.; Liao, G.; Tang, Z. Low-temperature and low-pressure Cu–Cu bonding by highly sinterable Cu nanoparticle paste. Nanoscale Res. Lett. 2017, 12, 255. [Google Scholar] [PubMed]
  38. Gao, Y.; Jiu, J.; Chen, C.; Suganuma, K.; Sun, R.; Liu, Z.-Q. Oxidation-enhanced bonding strength of Cu sinter joints during thermal storage test. J. Mater. Sci. Technol. 2022, 115, 251–255. [Google Scholar]
  39. Aliouat, A.; Antou, G.; Rat, V.; Pradeilles, N.; Geffroy, P.-M.; Maître, A. Investigation of electrical transitions in the first steps of spark plasma sintering: Effects of pre-oxidation and mechanical loading within copper granular media. Materials 2022, 15, 4096. [Google Scholar] [CrossRef]
  40. Yang, W.; Wan, J.; Wu, C.; Xie, C.; Huang, Z. Sintering of mixed Cu Ag nanoparticles pretreated by formic acid vapor for Cu-Cu low temperature bonding. Microelectron. Reliab. 2023, 141, 114890. [Google Scholar] [CrossRef]
  41. Liu, J.; Mou, Y.; Liu, J.; Peng, Y.; Chen, M. Low-temperature Cu-Cu bonding by using Cu2O nanoparticle coated hierarchical structure. IEEE Trans. Compon. Packag. Manuf. Technol. 2022, 12, 878–882. [Google Scholar]
  42. Yuan, Y.; Wu, H.; Li, J.; Zhu, P.; Sun, R. Cu-Cu joint formation by low-temperature sintering of self-reducible Cu nanoparticle paste under ambient condition. Appl. Surf. Sci. 2021, 570, 151220. [Google Scholar]
  43. Xie, J.; Shen, J.; Deng, J.; Chen, X. Influence of aging atmosphere on the thermal stability of low-temperature rapidly sintered Cu nanoparticle paste joint. J. Electron. Mater. 2020, 49, 2669–2676. [Google Scholar]
  44. Dai, D.; Li, J.; Qian, J.; Wang, Z.; Zheng, K.; Yu, J.; Chen, X. The formation of Cu-Cu joints by low temperature sintering Cu NPs with copper formate layer and its oxidation enhancement. Mater. Lett. 2023, 339, 134087. [Google Scholar] [CrossRef]
Figure 1. Established copper particle pair model: (a) diameter of 10 Å, (b) diameters of 10 Å and 100 Å, and (c) diameter of 100 Å.
Figure 1. Established copper particle pair model: (a) diameter of 10 Å, (b) diameters of 10 Å and 100 Å, and (c) diameter of 100 Å.
Processes 13 01042 g001
Figure 2. Schematic graph of the experimental procedure in this study: (a) preparation of Cu paste; (b) Cu paste; (c) bonding process under different atmosphere; and (d) shear strength test.
Figure 2. Schematic graph of the experimental procedure in this study: (a) preparation of Cu paste; (b) Cu paste; (c) bonding process under different atmosphere; and (d) shear strength test.
Processes 13 01042 g002
Figure 3. Schematic of the sintering process: (a) with preoxidation; (b) without preoxidation.
Figure 3. Schematic of the sintering process: (a) with preoxidation; (b) without preoxidation.
Processes 13 01042 g003
Figure 4. Changes in the sintering process of copper particles with different sizes: (a) change in MSD over time; (b) changes in RDF during sintering.
Figure 4. Changes in the sintering process of copper particles with different sizes: (a) change in MSD over time; (b) changes in RDF during sintering.
Processes 13 01042 g004
Figure 5. Effect of particles with diameter of 100 and 10 Å on the change in potential energy with temperature during sintering.
Figure 5. Effect of particles with diameter of 100 and 10 Å on the change in potential energy with temperature during sintering.
Processes 13 01042 g005
Figure 6. Results of experiment: (a) shear strength for different preoxidation sintering times, (b) shear strength without preoxidation, and (c) cross-sectional SEM image.
Figure 6. Results of experiment: (a) shear strength for different preoxidation sintering times, (b) shear strength without preoxidation, and (c) cross-sectional SEM image.
Processes 13 01042 g006
Figure 7. SEM images of the fracture surfaces of bonding joints fabricated under different processes, (ac) preoxidation process (FA), (df) FA, and (gi) N2.
Figure 7. SEM images of the fracture surfaces of bonding joints fabricated under different processes, (ac) preoxidation process (FA), (df) FA, and (gi) N2.
Processes 13 01042 g007
Figure 8. Shear strength of the joint under different sintering conditions: (a) FA (20 min); (b) N2 (20 min); (c) FA (300 °C); and (d) N2 (300 °C).
Figure 8. Shear strength of the joint under different sintering conditions: (a) FA (20 min); (b) N2 (20 min); (c) FA (300 °C); and (d) N2 (300 °C).
Processes 13 01042 g008
Figure 9. SEM images of the fracture surface of the joint manufactured at 300 °C for 1 min (ac) FA, and (df) N2.
Figure 9. SEM images of the fracture surface of the joint manufactured at 300 °C for 1 min (ac) FA, and (df) N2.
Processes 13 01042 g009
Table 1. Experimental design of the preoxidation process.
Table 1. Experimental design of the preoxidation process.
Experiment 1Oxidation BondingReduction BondingAtmosphere
Temperature (°C)Time (min)Temperature (°C)Time (min)
12504030020FA
23020FA
32020FA
41020FA
5520FA
6020FA
7020N2
Table 2. Experimental design of the sintering process.
Table 2. Experimental design of the sintering process.
Experiment 2Temperature (°C)Time (min)Atmosphere
13001FA/N2
25
310
420
12501
25
310
420
12001
25
310
420
Disclaimer/Publisher’s Note: The statements, opinions and data contained in all publications are solely those of the individual author(s) and contributor(s) and not of MDPI and/or the editor(s). MDPI and/or the editor(s) disclaim responsibility for any injury to people or property resulting from any ideas, methods, instructions or products referred to in the content.

Share and Cite

MDPI and ACS Style

Li, B.; Luo, Y.; Li, D.; Li, D.; Yang, B.; Gong, B.; Han, S.; He, S.; Cai, M. High Cu-Cu Bonding Strength Achievement Using Micron Copper Particles Under Formic Acid Atmosphere. Processes 2025, 13, 1042. https://doi.org/10.3390/pr13041042

AMA Style

Li B, Luo Y, Li D, Li D, Yang B, Gong B, Han S, He S, Cai M. High Cu-Cu Bonding Strength Achievement Using Micron Copper Particles Under Formic Acid Atmosphere. Processes. 2025; 13(4):1042. https://doi.org/10.3390/pr13041042

Chicago/Turabian Style

Li, Bofu, Yinyin Luo, Dejian Li, Dameng Li, Baobin Yang, Baoliang Gong, Shunfeng Han, Siliang He, and Miao Cai. 2025. "High Cu-Cu Bonding Strength Achievement Using Micron Copper Particles Under Formic Acid Atmosphere" Processes 13, no. 4: 1042. https://doi.org/10.3390/pr13041042

APA Style

Li, B., Luo, Y., Li, D., Li, D., Yang, B., Gong, B., Han, S., He, S., & Cai, M. (2025). High Cu-Cu Bonding Strength Achievement Using Micron Copper Particles Under Formic Acid Atmosphere. Processes, 13(4), 1042. https://doi.org/10.3390/pr13041042

Note that from the first issue of 2016, this journal uses article numbers instead of page numbers. See further details here.

Article Metrics

Back to TopTop