Cu Ribbon Ultrasonic Bonding on Die-Top Nanotwinned Cu Films Evaporated on SiC Chips for Power Modules

Finite element modeling and analysis of ultrasonic bonding process of thick aluminum wires for power electronic packaging

Two sets of NiFe∕FeMn films with Ta and Ta∕Cu buffer layers were prepared by magnetron sputtering to study the influence of different buffers on the exchange coupling process. The results show that the exchange bias field (Hex) of NiFe∕FeMn films with a Ta∕Cu buffer is lower than that of films with a Ta buffer. There is no apparent difference in the texture and roughness of films both with Ta and Ta∕Cu buffers. However, the segregation of Cu atoms on the surface of NiFe in the trilayer of Ta∕Cu∕NiFe was found by using angle-resolved x-ray photoelectron spectroscopy (XPS). A decrease of Hex for NiFe∕FeMn films with a Ta∕Cu buffer layer is primarily caused by the segregation of Cu atoms to the interface of the NiFe and FeMn layers. In order to suppress the Cu segregation, we deposited Bi insetting layers at the interface of Cu and NiFe in Ta∕Cu∕NiFe∕FeMn∕Ta films and found that Hex of NiFe∕FeMn can be doubled when the proper quantity of Bi is deposited. XPS analysis shows that Bi insetting layers deposited at the interface of Cu and NiFe effectively suppress the Cu segregation on the NiFe layer. As a result, Hex is increased. However, if the insetting Bi is deposited excessively, it will partially migrate to the FeMn layer, damaging the antiferromagnetic property of the FeMn layer, thereby resulting in decreased Hex. When Ag and Pb were deposited at the interface of Cu and NiFe in Ta∕Cu∕NiFe∕FeMn∕Ta films, similar experimental results were obtained.

Welding characteristics of thick coated wire speci- mens using a 19 kHz ultrasonic complex vibration welding system are studied. Using the 19 kHz complex vibration, 1.0- mm-diameter or 1.0-mm- to 2.0-mm-wide flat copper wires with polyurethane or polyurethane and polyimide copolymer insulation coating layer are welded directly to copper plates or copper layer on substrates in short time. And also, required vibration amplitudes of a newly developed 150 kHz complex vibration welding equipments for polyurethane coated cop- per wires of 0.036 mm and 0.12 mm in diameter and copper plates are 1.4 µm and 1.6 µm (peak-to-zero value).

Bonding wire is one of the main interconnection techniques. Thick bonding wire is widely used in power modules and other high power applications. This study examined the case for extending the use of traditional thin wire reliability criteria, namely wire flexure and aspect ratio, to thick wires. Eleven aluminum (Al) and aluminum coated copper (CucorAl) wire samples with diameter 300 μm were tested experimentally. The wire response was measured using a novel non-contact method. High fidelity FEM models of the wire were developed and validated. We found that wire flexure is not correlated to its stress state or fatigue life. On the other hand, aspect ratio is a consistent criterion of thick wire fatigue life. Increasing the wire aspect ratio lowers its critical stress and increases its fatigue life. Moreover, we found that CucorAl wire has superior performance and longer fatigue life than Al wire.

Ultrasonic seam welding of metal plate specimens using a 28 kHz complex vibration disk welding tip vibrating in transverse and torsional vibration modes is studied. Conventional ultrasonic seam welding equipment using one-dimensional vibration locus can only join metal foil or very thin plate specimens and not join thick metal plate specimens. Using a complex vibration welding system with a welding tip vibrating in elliptical or circular locus, thick plate specimens can be welded with a more uniform and larger area weldment compared to a conventional ultrasonic welding system. An ultrasonic seam welding system consists of a complex vibration disk welding tip which is driven by a longitudinal vibration system through a longitudinal-torsional vibration converter. The disk welding tip vibrates in elliptical locus. Aluminum plate specimens 0.3 mm and 3.0 mm in thickness were successfully joined with such weld strengths that the specimens were broken at the plate parts adjacent to the welded area under tensile strength test. Aluminum plates 50 mm in width were successfully joined by the ultrasonic seam welding system.

It was established that the mode parameters determining efficiency of the plastics ultrasonic welding process include waveguide oscillation amplitude, welding static pressure, oscillation frequency, welding time and size of the fixed gap between the waveguide working end and the support. In each specific welding mode, an increase in the gap of more optimal value leads to a decrease in the breaking load; this is especially true in welding products of higher thickness. Welding could be performed on surfaces coated with various products. An expression was obtained to determine time and intensity coefficients of the ultrasonic welding mode. Fixture supports influence on quality and strength of the welded joints in the ultrasonic welding of polymeric materials was considered. When using an active support, strength and productivity of ultrasonic welding was much higher than in using the passive supports. Results of the experiments confirmed the relationship between the support oscillations amplitude and the heating, thickness and properties of the parts to be welded. A working cycle of the plastics ultrasonic welding is proposed, where the static welding pressure and the ultrasonic pulse time are ensuring maximum strength and tightness of the welded joint.

Hybrid integrated circuits used in automotive applications often use surface-mounted thick film circuits on ceramic substrate. The electrical interconnection of the respective contact areas of such circuit modules are typically made by wire bonding techniques, in which thin electrically conductive wires, about 25-300 μm in diameter, are bonded from the chip to bond pads on the substrate, and from the substrate to the leads of the metal case of the package. Typically 25-50 μm gold wire ball-wedge thermosonic bonding is used for interconnecting the chip to the substrate, while 100-300 μm thick aluminium wire wedge-wedge ultrasonic bonding from the substrate to the leads. The interconnection materials such as the wire, the bonding pads and surfaces, as well as, the bonding techniques and process parameters play important role in the quality and the reliability of the wire bonds, and finally of the circuit. Our research has dealt with the analyzes of the defects occurring at the aluminium wire bonding from thick film pads to the nickel-plated stainless steel - Ni leads of the component case.

The deformation behavior of thick Al wires and the expansion behavior of the bond area during ultrasonic wedge bonding to Al­Si, Si and SiO2 substrates were measured simultaneously in detail with a high-speed measuring system. The deformation of the wire by the application of the bonding force is completed immediately. The deformation restarts by the application of the ultrasonic vibration. The deformation induced by applying the bonding force consists of only elastic component, whereas that by ultrasonic vibration consists of only plastic component. The Al wire is not work-hardened by the plastic deformation during application of ultrasonic vibration. The adhered area expands to the direction perpendicular to the ultrasonic vibration. The evolution of the wire deformation behavior and the expansion of the adhered area show an intimate correlation with each other. [doi:10.2320/matertrans.MD201210]

Interface recombination is one of the factors limiting the performance of Cu(In,Ga)Se <sub xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">2</sub> (CIGS). Especially in the absence of band grading at the front and rear surface, interface passivation approaches become important to improve device performance. The integration of an oxide layer as passivation layer at the front surface of the CIGS requires meticulous considerations in order not to impact the further steps of the solar cell production. In this article, a novel approach is reported to try to tackle the problem of interface recombination at the front surface of CIGS without affecting further solar cell production steps. In this approach, an Al <sub xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">2</sub> O <sub xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">3</sub> /HfO <sub xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">2</sub> multi-stack layer with contact openings is applied. NaCl template patterning with preliminarily selected parameters was used to create a homogeneous pattern of contact opening on the CIGS surface and allow the current flow in the device. After the removal of the NaCl islands, the holes in the multi-stack (openings) were visualized by scanning electron microscopy. In addition, energy-dispersive X-ray spectroscopy (EDS) was performed before and after chemical bath deposition of the buffer layer. The EDS result confirmed that the undesired etching of the Al <sub xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">2</sub> O <sub xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">3</sub> layer during buffer layer deposition was prevented by using a thin HfO <sub xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">2</sub> layer. Solar cells were produced by using preliminarily selected parameters for the multi-stack design. As a result, without having a significant negative impact on the solar cell parameters, a device design was achieved which is almost comparable with the reference device. In addition, options for future improvement and development are discussed.

In this study, we evaluated a higher frequency ultrasonic welding system using a surface acoustic wave (SAW) device with inter-digital electrodes. In ultrasonic plastic welding, welding at higher frequencies has some merits. First, it is assumed that welding at high frequency makes the joining time shorter, because ultrasonic absorption by the polymer is proportional to the square of the frequency. Second, damage to joined parts can be avoided, because vibration displacement amplitude on joining tool is lower at high frequency. However, it is very difficult to maintain a wider joined area at a higher frequency using a conventional longitudinal-mode transducer system. Therefore, a joining system using a SAW device will be quite effective for high frequency joining. In this paper, we describe 1220.6 kHz SAW system with a 20×18 mm2 work area. Using the SAW system, we joined polyethylene films of 0.8 mm of thick and acrylic plates 2.0 mm of thickness. Furthermore, we compared the SAW system with a conventional 19 kHz longitudinal-mode welding system based on the results of joining.

Modelling was undertaken to investigate the role of bond wire size on reliability in power electronic converters. Experiments have shown that thin 125 μm Al wires used in place of 375 μm Al wires alleviate bond wire lift-off and further outlast other sources of failure such as solder degradation in a power module. To investigate the role of bond-wire size on wire lift-off, the effective plastic strain was estimated through thermo-mechanical simulation. Three-dimensional models were constructed for the thin and thick bond wires, respectively. For the critical deformation of the aluminium bond wires during thermal cycling, a temperature-dependent bi-linear plasticity model was used. The effect of a difference in yield strength for the thin wires was also investigated. Maximum as well as volumetrically averaged values of the effective plastic strain showed significant differences between the thick and thin wires and wires with different yield strengths. The modelling results show higher effective plastic strain for the thick wires - supporting the experimental findings.

Ultrasonic multi-spot continuous welding of metal plate specimens using a two-vibration-system welding equipment

Further understanding of the electronic structure at the ZnS/Cu(In,Ga)Se2 interface is necessary to enhance the electron injection across the interface in Cu(In,Ga)Se2 solar cells. The valence band structure and shallow core levels were investigated by ultraviolet photoelectron spectroscopy depth profile analysis with He II line excitation. ZnS film was grown by a chemical bath deposition on a Cu(In,Ga)Se2 absorber deposited by the co-evaporation of Cu, In, Ga, and Se elemental sources. The discontinuity of 2.0 eV in the valence band edge at the ZnS/Cu(In0.7Ga0.3)Se2 interface was directly determined. This type of valence band offset yields a spike conduction band alignment of 0.25 eV. The positions of the VBM and the Zn 3d core-level emission of the buffer underwent the substantial shifts of 0.36 eV and 0.64 eV to a lower binding energy levels during the etching process. The shifts are ascribed to the contribution of the band bending in the ZnS buffer layer and its graded chemical composition. This study is the first to determine the small conduction band offset at the interface formed by the chemical bath deposited ZnS layer and the Cu(In0.7Ga0.3)Se2 absorber. Our results also provide information toward the design optimization of the optoelectronic properties of the ZnS/Cu(In0.7Ga0.3)Se2 interface. To enhance the electron injection from Cu(In0.7Ga0.3)Se2 absorber to ZnS layer further lowering of the energy barrier is required. For this purpose, the bandgap of ZnS should be reduced by controlling the crystal structure and composition or its Fermi level should be upward shifted by appropriate doping.

A small number of fishers in Chiba Prefecture of eastern Japan use cotton gill nets to catch Japanese spiny lobster Panulirus japonicus. To examine the advantages of cotton gill nets, we analyzed changes in mesh breaking load of a new cotton gill net used in a fishing operation. A new cotton gill net was also soaked in a seawater tank to simulate ghost fishing conditions. The average mesh breaking load of new cotton mesh was 50.3 N. This value decreased to 19.0 N after 38 days (∼912 h), and after 82 days (∼1968 h) the mesh could be easily torn (breaking load 0.07 N). Under fishing conditions, the cumulative soak time was only 744.4 h over 19 months. The average breaking load at the end of this period was 43.1 N, a strength 86% that of the presoaked mesh. The mesh breaking load of a cotton gill net continuously soaked for 744.4 h was 26.1 N, as estimated from tank experiment data. Thus, a cotton gill net maintains reasonable strength under typical use conditions, but will degrade if lost at sea.

Ultrasonic fabrication of fiber reinforced plastics made from thermoplastic polymer films and carbon or glass fibers enables cycle times of a few seconds and requires investment costs of only some 10,000 €. Besides this, the raw materials can be stored at room temperature. A fiber content of 33 vol % and a tensile strength of approximately 1.2 GPa have been achieved by ultrasonic welding of nine layers of foils from polyamide, each 100 µm in thickness, and eight layers of carbon fibers, each 100 µm in thickness, in between. Besides unidirectional carbon fiber reinforced polymer composite (CFRP) samples, multi-directional CFRP plates, 116 mm, 64 mm and 1.2 mm in length, width and thickness respectively, were fabricated by processing three layers of carbon fiber canvas, each 300 µm in thickness, and eight layers of polyamide foils, each 100 µm in thickness. Furthermore, both the discontinuous and the continuous ultrasonic fabrication processes are described and the results are presented in this paper. Large-scale production still needs to be demonstrated.