Assessing thermal spike model of swift heavy ion-matter interaction via Pd$_{1-x}$Ni$_x$/Si interface mixing

TL;DR

This study evaluates the thermal spike model's explanation of swift heavy ion-induced interface mixing by experimentally measuring Pd-Ni/Si mixing and comparing it with theoretical predictions, revealing discrepancies that suggest additional mechanisms.

Contribution

It provides an experimental assessment of the thermal spike model's validity in explaining ion-induced mixing, highlighting limitations of current theoretical approaches.

Findings

Experimental mixing variation with composition shows a minimum not predicted by theory.

Density functional theory calculations of e-p coupling do not match the experimental trend.

Disparity suggests other mechanisms beyond the thermal spike model influence ion mixing.

Abstract

Thermal spike model (TSM) is presently a widely accepted mechanism of swift heavy ion (SHI) - matter interaction. It provides explanation to various SHI induced effects including mixing across interfaces. The model involves electron-phonon (e-p) coupling to predict the evolution of lattice temperature with time. SHI mixing is considered to be a result of diffusion in transient molten state thus achieved. In this work, we assess this conception primarily via tuning the e-p coupling strength by taking a series Pd$_{1-x}$Ni$_x$ of a completely solid soluble binary, and then observing 100 MeV Au ion induced mixing across Pd$_{1-x}$Ni$_x$/Si interfaces. The extent of mixing has been parametrised by the irradiation induced change $\Delta \sigma^2$ in variances of Pd and Ni depth profiles derived from X-ray photoelectron spectroscopy. The $x$-dependence of $\Delta \sigma^2$ follows a curve that is concave upward with a prominent minimum. Theoretically, e-p coupling strength determined using density functional theory has been used to solve the equations appropriate to TSM, and then an equivalent quantity L$^2$ proportional to $\Delta \sigma^2$ has been calculated. L$^2$, however, increases monotonically with $x$ without any minimum, bringing out a convincing disparity between experiment and theory. Perhaps some mechanisms more than the TSM plus the transient molten state diffusion are operative, which can not be foreseen at this point of time.