The kinetics for the crystallization of amorphous Ni(P) films and the formation of intermetallic compounds in Sn/Ni(P) films during isothermal aging treatment were studied with in situ intrinsic stress measurements. The intrinsic stress changes from crystallization were about 200 and 150 MPa for Ni(14P) and Ni(11.7P) films, respectively, and according to Johnson–Mehl–Avrami analysis, the Avrami exponents were about 3.6 ± 0.46 and 4.2 ± 0.39, and the activation energies were 242 and240 kJ/mol, respectively, for the crystallization of Ni(14P) and Ni(11.7P) films. The stress due to the formation of intermetallic compounds such as Ni3Sn4 and Ni3P in Sn/Ni(11.7P) films was about 320 MPa. Application of in situ stress measurementsto the empirical growth model during isothermal phase transformation of Sn/Ni(P) showed that the intermetallic compounds growth was interface reaction-controlled (n = 0.91 ± 0.08) in the early stage and then became diffusion-controlled (n =0.38 ± 0.01), and the activation energy was about 35.9 kJ/mol.

As-deposited thin films grown by vapor deposition often exhibit large intrinsic stresses that can lead to film failure. While this is an “old” materials problem, our understanding has only recently begun to evolve in a more sophisticated fashion. Sensitive real-time measurements of stress evolution during thin-film deposition reveal a generic compressive–tensile–compressive behavior that correlates with island nucleation and growth, island coalescence, and postcoalescence film growth. In this article, we review the fundamental mechanisms that can generate stresses during the growth of Volmer–Weber thin films. Compressive stresses in both discontinuous and continuous films are generated by surface-stress effects. Tensile stresses are created during island coalescence and grain growth. Compressive stresses can also result from the flux-driven incorporation of excess atoms within grain boundaries. While significant progress has been made in this field recently, further modeling and experimentation are needed to quantitatively sort out the importance of the different mechanisms to the overall behavior.