Purpose: The purpose of this paper is to investigate the diffusion behaviors of different atoms at the Cu/Cu3Sn interface and the vacancy formation energy, diffusion energy barrier and vacancy diffusion activation energy. Design/methodology/approach: The diffusion behaviors of different atoms at the Cu/Cu3Sn interface are analyzed, and the vacancy formation energy, diffusion energy barrier and vacancy diffusion activation energy are obtained using molecular dynamics simulation. The nudged elastic band method is used to evaluate diffusion energy barrier for Cu/Cu3Sn system. Findings: It is found that the vacancies in the Cu/Cu3Sn interface promote the interfacial diffusion, and the formation energy of Cu vacancy in the Cu crystal is larger than that in Cu3Sn crystal. In addition, the formation energies of Cu1 vacancy and Cu2 vacancy are close to each other in Cu3Sn crystal, and they are all less than the formation energy of Sn vacancy. Furthermore, the vacancy diffusion barrier and vacancy diffusion activation energy of the Cu/Cu3Sn interface are calculated, and the results show that the vacancy diffusion activation energy of Sn was higher than that of Cu. Finally, by comparison of diffusion activation energies of different diffusion mechanisms, Cu→Cu1vac is the most possible migration path at all temperatures. Originality/value: It is concluded that the vacancies in Cu/Cu3Sn interface promote interfacial diffusion, and the activation energy of vacancy diffusion in most diffusion mechanisms decreases with the increase of temperature.

Limit loads are widely studied and several limit load solutions are proposed to some typical geometry of weldments. However, there are no limit load solutions exist for the single edge crack weldments in tension (SEC(T)), which is also a typical geometry in fracture analysis. The mis-matching limit load for thick plate with SEC(T) are investigated and the special limit load solutions are proposed based on the available mis-matching limit load solutions and systematic finite element analyses. The real weld configurations are simplified as a strip, and different weld strength mis-matching ratio M, crack depth/width ratio a/W and weld width 2H are in consideration. As a result, it is found that there exists excellent agreement between the limit load solutions and the FE results for almost all the mis-matching ration M, a/W and ligament-to-weld width ratio (W-a)/H. Moreover, useful recommendations are given for evaluating the limit loads of the EBW structure with SEC(T). For the EBW joints with SEC(T), the mis-matching limit loads can be obtained assuming that the components are wholly made of base metal, when M changing from 1.6 to 0.6. When M decreasing to 0.4, the mis-matching limit loads can be obtained assuming that the components are wholly made of base metal only for large value of (W-a)/H. The recommendations may be useful for evaluating the limit loads of the EBW structures with SEC(T). The engineering simplifications are given for assessing the limit loads of electron beam welded structure with SEC(T).