Wire bonding is a popular joining technique in microelectronic interconnect. In this study, the effects of applied load, surface roughness, welding power and welding time on bonding strength were investigated using an ultrasonic bonding machine and a pull tester. In order to relate bonding strength to contact phenomena, the asperity model was used to compute real contact area and flash temperature between the wire and the pad. The experimental results show that a decrease in load or ultrasonic power produces a larger weldable range in which the combination of operation parameters allow the wire and pad to be welded. Regardless of roughness and applied loads, the bond strength increases to a maximum with increases in the welding time, and then decreases to fracture between wire and pad. The theoretical results and experimental observations indicate that bond strength curves can be divided into three periods. The contact temperature plays an important role in bonding strength in the initial period, and surface roughness is the dominant factor in the final period. The maximum bonding strength point occurs in the initial period for different loads and surface roughness values. Our results show that bond strength of ultrasonic wire bonding can be explained based on the input energy per real contact area.
