Effect of composition and stress on nickel-hafnium amorphization by interdiffusion

The rate of solid state amorphization reaction, SSAR, in the Ni/Hf system has been monitored by Rutherford backscattering spectrometry, RBS, thickness and composition measurements. Amorphous alloys, a-Ni_CHf_1−C , in contact with one of the terminal phases are used to measure the interdiffusion coefficient, D˜, averaged over limited composition ranges. It is found that both the SSAR rate and D˜ are composition-dependent. Numerical simulations provide semi-quantitative estimates of D˜(C) and corresponding composition profiles. Direct measurements of common-tangent compositions have been made based on absence of reaction; these differ from previously published determinations in a manner consistent with the calculated composition profiles. It is concluded that the average composition of the amorphous phase formed in elemental diffusion couples is determined by both thermodynamics and kinetics. This work is characterized by the first direct determination of the common-tangent composition and estimates of the composition dependence of D˜. Even under metastable equilibrium, caution is urged in estimating common-tangent compositions from observed composition profiles.; X-ray diffraction, XRD, and curvature measurements have been used to monitor stresses during SSAR. In the present work, little decrease in D˜ is observed at long reaction times. During SSAR in elemental bilayer diffusion couples, the amorphous layer forms under large tensile stresses which relax during further reaction. This causes the overall sample stress to increase to a maximum, then decline to an apparent steady-state level. Upon low-temperature ion irradiation of a partially reacted sample, the overall tensile stress increases; further solid state amorphization results in additional stress increase, followed by decrease. Neither stress changes accompanying SSAR nor those induced by irradiation result in any change in the reaction rate law—D˜ is unchanged. Restraint is urged in interpretating RBS and XRD measurements: RBS cannot indicate (a) relative atomic mobilities unless inert markers are used, nor (b) the presence/absence of voids. XRD can only yield stresses when multiple scattering vector orientations are sampled. Agreement between XRD and curvature-based stress measurements should not be assumed. This work is characterized by monitoring stress evolution by multiple techniques, irradiation-induced stress changes, and an unprecedented thorough use of control samples necessary to draw the conclusions presented.