High Power Device Packaging Research Articles

Carbon-Based Thermal Interface Materials for High-Power Device Packaging

The packaging of high-power devices, like high-frequency GaN and GaAs devices, is a persistent challenge. These devices are often mounted with thermal interface materials (TIMs) that must provide both electrical and thermal conductivity to an underlying board. As these devices begin to face extreme thermal fluxes, the required thermal conductivity begins to exceed that of metals, and TIMs based on new materials must be developed for their reliable packaging. In most non-metals, the thermal conductivity is related to the atomic mass and the chemical bonds in the material. This spotlights carbon allotropes that have light atoms and some of the strongest chemical bonds found in nature. For example, three-dimensional (3D) diamond with sp3 hybridized carbon and two-dimensional (2D) graphene with sp2 hybridized carbon. As such, these materials exhibit record thermal conductivities. However, diamond is electrically insulating. Graphene has excellent electrical and thermal conductivity in-plane, but is highly anisotropic and has poor cross-plane conductivities (i.e., it is hard to get heat into graphene). Furthermore, graphene has finicky interactions with the surrounding environment. To lend perspective, simply placing a graphene on a substrate can suppress its thermal conductivity by an order of magnitude. While neither material may be a standalone solution, their integration together (like sp2-sp3 conversion) or within a composite (like an epoxy) may hold promise. To fully understand the electrical and thermal potential of such carbon-based TIMs, not only do the bulk properties of diamond and graphene matter, but also the interfacial properties. In fact, in nanostructured composites, the interfacial properties can severely limit the overall properties. As a result of this, diamond and graphene composites typically exhibit thermal conductivities below bulk metals. This work probes the fundamental limits of interfacial electrical and thermal conductivity between, diamond, graphene, and a host material. The aim of this work to understand and manipulate these interfacial properties in a manner that allows carbon-based TIMs to meet the packaging needs of increasingly higher-power devices.

Combined fluid flow simulation with electrochemical measurement for mechanism investigation of high-rate Cu pattern electroplating

BackgroundCu pattern electroplating plays an essential role in the interconnection and microstructure fabrication of high-power device packaging. However, improving efficiency while ensuring the preparation quality is an urgent breakthrough direction. MethodsA new electrolyte formula suitable for the high-rate Cu pattern electroplating was put forward, including 300 mg/L polyethylene glycol (PEG, molecular weight=8000) and 6 mg/L thiazolinyl polydipropyl sulfonate (SH110). The mechanism of high-rate Cu pattern electroplating was investigated by the combination of fluid flow simulation and the electrochemical measurements including galvanostatic measurements (GMs) and cyclic voltametric (CV) tests. The Cu pattern was electroplated in the Haring-Cell with air bubble as the forced convection method. Significant findingsThe simulation results indicate that the flow rate on the ring-shaped cathodic pattern surface gradually increases with the Cu growing up, which also increase from the pattern surface to the bulk electrolyte. The GMs results suggest that the cathodic polarization of the convection-dependent additives raises with the current density, which is strengthened by strong convection with current density lower than 6 ASD but have the opposite effect with current density larger than 8 ASD. The CV tests indicating that the aggregation of weak convection and strong polarization promote the antagonistic effect between thiolate of SH110 and PEG, the nucleation and the electroplating rate. Nevertheless, the synergistic effect of thiazoline of SH110 and PEG are enhanced with strong convection and weak polarization. The adsorption model during the Cu pattern electroplating process is established and verified by electroplating experiment with the optimum current density of 8 ASD. The Cu pattern is mirror bright with the high electroplating rate of over 90 μm/h.

Molecular dynamics investigation of the slip flow liquid–solid interfacial thermal conductance

With the integrated high-power device packaging structure rapidly developing, the embedded heat dissipation architectures are challenged by the local micro-/nanoscale massive heat flux. The slip flow molecular dynamics models were established to explore the liquid–solid interfacial thermal conductance. With stepwise declining shear forces (0.032 pN/200, 0.024 pN/200, and 0.016 pN/200 ps, respectively), the slip flow [the slip shear velocity is Si: (125.43 ± 0.92 m/s), graphite: (142.43 ± 1.92 m/s), and Cu: (180.93 ± 3.42 m/s), respectively] water–solid interfacial thermal conductance of different materials [Si: (8.11 ± 0.1) × 107 W/m2 K, graphite: (10.18 ± 0.1) × 107 W/m2 K, and Cu: (17.97 ± 0.1) × 107 W/m2 K] can be calculated. The rationality of the calculated values can be verified in the literature. The slip flow water–solid interfacial thermal conductance values are about 0.5 times higher than the static ones. It can be significantly affected by the slip shear velocity. The slip shear velocity increasing about five times can enhance the interfacial thermal conductance two times. From the water layer density distribution, it is found that the dependence of interfacial thermal conductance on velocity slip relies more on the dynamical properties than on the fluid structure. This molecular dynamics model provides an operative methodology to investigate the slip flow liquid–solid interfacial heat transfer for the various embedded cooling surfaces.

A low-temperature bonding method for high power device packaging based on In-infiltrated nanoporous Cu

With the rapid development of the third-generation semiconductor materials, an appropriate high-temperature-resistant die attach material has become one of the bottlenecks to fully exploit the excellent properties of the third-generation semiconductor power devices. At the same time, a low-bonding temperature is always the pursuit goal of packaging engineers to reduce the thermal residual stress in electronic devices. In this paper, a low-temperature bonding method was proposed to address the above-mentioned issue based on In infiltrating the nanoporous Cu. In, as a low melting point metal, can significantly reduce the bonding temperature, and the nanoporous Cu structure can provide a very large specific surface area, which greatly increases the consumption rate of In. Furthermore, the formed Cu-In IMCs with high-remelting temperature can withstand the high operating temperature. The microstructures of the bondlines before and after bonding were studied in detail. The results show that the bondline can completely consume the low melting point In, within 10 min at 165 °C under a pressure of 0.75 MPa. When the bonding temperature was further increased to 310 °C, the bondline was composed of η-Cu2In and δ-Cu7In3 phases, whose melting points were more than 600 °C. The average electrical resistivity was determined to be 5.53 ± 0.65 μΩ cm, and the thermal conductivities were 144.33 W m−1 K−1, 139.24 W m−1 K−1 and 129.79 W m−1 K−1 at 30 °C, 150 °C and 300 °C, respectively. The average shear strength were 19.09 ± 3.4 MPa, 20.47 ± 4.6 MPa and 30.73 ± 5.2 MPa at 30 °C, 250 °C, 310 °C, respectively. These results indicate that the nanoporous Cu infiltrated with In could meet the requirements of electrical, thermal conduction, and mechanical support as a die attachment for high-power devices.

Wetting behavior and vacuum soldering of novel Au-30Ga solder on Ni and Cu substrate

As a novel high-temperature solder with suitable eutectic temperature and high thermal conductivity, Au-30Ga solder is expected to be used in high-power device packaging. This research provides a timely and necessary study of its solderability. The wetting behavior and reactivity of Au-30Ga solder on Cu and Ni substrates was investigated through spreading tests in vacuum. The favorable wetting property of Au-30Ga solder on Ni and Cu substrate was observed. However, fundamental differences in the wetting behavior of liquid Au-30Ga alloy in contact with Ni and Cu substrates were immediately apparent. Their behavior changed from reactive wetting, characterized by formation of two thin Ni–Ga compound layers and a wide wetting ring, to dissolutive wetting, characterized by formation of a thick (Au, Cu)9Ga4 interface layer. Moreover, significant signs of solidification shrinkage of liquid solder are observed on the surface of wetting ring for the first time, which provides direct evidence for the formation mechanism of wetting ring. The wetting ring also reveals the microstructure of the Ni–Ga compounds during ripening. In addition, the soldering performance was evaluated by the shear strength of solder joints. The findings presented in this thesis add to understanding of solderability of the high-temperature Au-30Ga solder.

Ultrasonic-accelerated metallurgical reaction of Sn/Ni composite solder: Principle, kinetics, microstructure, and joint properties.

The high-melting-point joints by transient-liquid-phase are increasingly playing a crucial role in the die bonding for the high temperature electronic components. In this study, three kinds of Sn/Ni composite solder pastes composed of different sizes of Ni particles were synthesized to accelerate metallurgical reaction among Sn/Ni interfaces under the ultrasonic-assisted transient liquid phase (U-TLP) soldering. The temperature evolution, microstructure and mechanical property in joints composed by these composite solder pastes with or without ultrasonic energy were systemically investigated. The intermetallic joint consisted of high-melting-point sole Ni3Sn4 intermetallic compound with a little residual Ni was obtained under the conditions of no pressure and lower power (200W) in a high-temperature duration of only 10s, its shear strength was up to 45.3MPa. Ultrasonic effects significantly accelerated the reaction among the interfaces of liquid Sn and solid Ni, which attributed to the temperature rise caused by acoustic cavitation because of large number of liquid/solid interfaces during U-TLP, resulting in accelerated solid/liquid interfacial diffusion and growth of intermetallic compounds. This intermetallic joint formed by U-TLP soldering has a promising potential for applications in high-power device packaging.

Pressureless low-temperature sintering of plasma activated Ag nanoparticles for high-power device packaging

Abstract A controllable plasma-activated method to achieve pressureless sintering of Ag nanoparticles is proposed for low-temperature bonding. O2 plasma activation is employed to effectively decompose organics coated on the nanoparticles prior to sintering, avoiding the loose interfacial microstructures. Therefore, robust bonding Cu-Cu joints with high shear strength (>20 MPa) can be formed at a low temperature of 200 °C without requiring additional pressure. This rapid and controllable activation method significantly enhances the sinterability of the Ag particles, leading to a dense microstructure. The improved thermal conductivity is three times higher than the one prepared without plasma activation.

Study of Al−Cu compounds as soldering bond pad for high-power device packaging

Four equilibrium phases, Al2Cu (θ), AlCu (η2), Al3Cu4 (ζ2), and Al4Cu9 (γ2) were produced by annealing the solid-state Al/Cu diffusion couples. With a proper polishing process, the above four Al−Cu compounds were fabricated as soldering substrates. We found that the pure Sn solder ball can wet properly on the Al3Cu4, Al4Cu9 substrates, but, the pure Sn solder ball cannot wet on the Al2Cu and AlCu substrates. Interestingly, Sn can wet on the Al4Cu9 substrate as well as on the Cu substrate. In addition, we found that a dark Al-rich layer formed below the interfacial Cu6Sn5 compound, which is analyzed to be an Al-rich Al4Cu9 phase. We speculate that the Al-rich Al4Cu9 layer is caused by the Sn soldering reaction. During soldering reaction with Sn, the Cu in the Al4Cu9 layer would diffuse out to react with Sn and form the Cu6Sn5 compound. The Al in the Al4Cu9 layer does not react with Sn solder, so, the Al content would stay in the Al4Cu9 substrate. X-ray photoelectron spectroscopy (XPS) analysis has verified the metallic Al phase in the dark Al layer.

Fluxless bonding of Si chips to aluminum boards using electroplated Sn solder

The high thermal conductivity and light weight properties of aluminum (Al) make it a promising material in high power device packaging and automotive electronics. A primary challenge is its high coefficient of thermal expansion (CTE) of 23 ppm/°C. In this research, we investigated the possibility of surmounting this challenge by bonding large Si chips to Al substrates using fluxless tin (Sn) solder. The Si–Al pair probably has the largest CTE mismatch among all bonded structures in electronic packaging. In experiments, 0.1 μm Cr layer and 0.2 μm Cu layer were deposited on Al substrates, followed by an electroplated 25 μm copper (Cu) layer. The Sn solder layer was then electroplated over the Cu, followed immediately by thin (0.1 μm) silver (Ag) layer. The joint thickness was controlled either by bonding pressure or by incorporating Cu spacers. Microstructure and composition of the joints were studied with scanning electron microscopy and energy dispersive X-ray spectroscopy. Despite the large CTE mismatch, the bonded structures did not break. The minimum fracture force measured among six samples by shear test is higher than three times of what specified in MIL-STD-883H, method 2019.8. This preliminary result corroborates potential adaption of Al substrates and boards in electronic packaging where lightweight is essential.