Effect of density and Z-contrast on the visibility of noble gas precipitates and voids with insights from Monte-Carlo simulations

Highlights

• Noble gases implanted into glasses precipitate as supercritical fluid inclusions.

• The precipitates of Xe and Kr can be unambiguously differentiated from voids.

• The precipitates under ion irradiation transform into voids which shrink and eventually close.

Abstract

In this work, a detailed analysis of He, Ne, Ar, Kr and Xe precipitates in a complex borosilicate glass using transmission electron microscopy (TEM) with in-situ ion implantation is presented. With in-situ monitoring, the real-time dynamics of precipitate and void evolution under ion implantation was followed. Using appropriate equations of state and, Monte-Carlo simulations to supplement the TEM images, we then discuss in detail the possibility and ways of differentiating the precipitates of various noble gases from empty voids. It is shown that all the noble gases precipitate as inclusions of supercritical fluid. With the aid of the simulations, the crucial role played by the size and density of the precipitates and atomic number of the gas atoms in affecting the visibility of the precipitates is highlighted. The results show that the precipitates and voids can be unambiguously differentiated in the case of Xe and Kr whereas the precipitates of other lighter noble gases cannot be differentiated from the voids. However, the precipitate and void evolution under ion irradiation follow different dynamics, knowledge of which allows one to differentiate between the precipitates and voids even for lighter noble gases. Besides shedding light on the subject of noble gas precipitation and identification of the precipitates and voids, the study highlights the complexity in dissociating the behaviour of voids from the process of precipitate re-solution. This type of knowledge is pivotal in developing models describing the evolution of precipitates, voids and macroscopic porosity in a number of materials.

Introduction

There is a significant interest in the behaviour and precipitation of noble gases in condensed matter. Researchers in geological and planetary sciences are interested in the estimates of noble gas solubility in minerals and as such undertake routine experiments involving noble gas infusion at high temperature and pressure. Noble gas infusion is also of great importance in fundamental glass science where the gases are used as structural probes to access the so-called “glass free-volume” which is fundamental in deciphering the structure and connectivity of the glass network. Similarly, the nuclear industry is significantly interested in understanding the behaviour and precipitation of noble gases especially Xe, Kr and He in nuclear fuel (Xe and Kr) structural nuclear materials (He) and, nuclear wastepackages (He) to understand the evolution and degradation, if any, of such materials due to noble gas precipitation. Furthermore, the electronics industry routinely employs noble gas implantation to tailor the dielectric and thermoelectric properties of semiconductors and insulators (Oliviero et al., 2010; Zhao et al., 2017; Lorenzi et al., 2014). Due to low physical solubility and chemical reactivity (as a result of the closed shell structure), noble gases tend to precipitate at relatively low concentrations. This can result in the formation of dense solid/fluid precipitates, gaseous bubbles and voids depending on the nature of the material and temperature among some of the factors. For the sake of clarity, the terminology “bubbles” will be used to indicate the presence of a gaseous phase and “precipitate” to indicate the presence of a liquid/supercritical fluid and/or solid phase. From this standpoint, a cavity filled with noble gas atoms can be a bubble or a precipitate depending on the phase of its contents. An empty cavity, on the other hand, will be referred as a “void”. In general, we will use the term “inclusions” to refer to any of these features. Amongst a plethora of techniques, transmission electron microscopy (TEM) is an indispensable tool for the study of inclusions in a variety of materials as it offers the potential to characterize and identify such precipitates at the nanoscale. The density of inclusions not only depends on the concentration of the implanted ions/infused gas but also on material properties, temperature and size of the precipitates. Clear and unambiguous identification of these inclusions is therefore important to properly understand their behaviour and dynamics. This is also crucial to develop predictive models which depend on the behaviour of precipitation, growth, size distributions etc. for nuclear materials. TEM, when combined with in-situ ion implantation, becomes a versatile instrument not only to study the inclusions themselves but also reveal the real-time dynamics taking place during the evolution of the inclusions — something that cannot be captured with any other technique at the nano-scale.

A number of TEM studies have been performed on the formation of He (Vishnyakov et al., 2002; Stevens and Johnson, 1997; Donnelly et al., 1995a, b; Johnson and Mazey, 1995; Birtcher et al., 1994; Evron et al., 1994; Johnson et al., 1991; Olander and Wongsawaeng, 2006; Marochov and Goodhew, 1988), Ne (Marochov and Goodhew, 1988; Noordhuis and De Hosson, 1991; Luukkainen et al., 1985; Vom Felde et al., 1984), Ar (Olander and Wongsawaeng, 2006; Marochov and Goodhew, 1988), Kr (Oliviero et al., 2010; Lis and Birtcher, 1988; Pagano et al., 1997; Birtcher and Liu, 1989; Rest and Birtcher, 1989; Evans and Mazey, 1985; He et al., 2017) and Xe (Oliviero et al., 2010; Vom Felde et al., 1984; Lis and Birtcher, 1988; Rest and Birtcher, 1989; Ishikawa et al., 1997) precipitates in several materials. TEM diffraction studies on noble gas implanted metals (e.g. Al, Au, Cu, Ni) (Stevens and Johnson, 1997; Donnelly et al., 1995b; Birtcher et al., 1994; Vom Felde et al., 1984; Rossouw and Donnelly, 1985; Lis and Birtcher, 1988; Birtcher and Liu, 1989; Rest and Birtcher, 1989; Evans and Mazey, 1985; Ishikawa et al., 1997) have shown that dense solid precipitates of Ar, Kr and Xe of few nanometres in size can be formed at room temperature. For the same implantation conditions, the size of precipitates in a given material is usually inversely proportional to the elastic moduli of the material (Luukkainen et al., 1985). Despite a number of studies on the formation of noble gas precipitates, a clear differentiation of dense precipitates, low-density gas bubbles, empty voids and the dynamics of their inter-conversion, either under ion implantation or during annealing, still remain very challenging. To date, there are no studies on the precipitates/bubbles of all the noble gases in a single material type to decipher the way in which the atom type, temperature, precipitate size and their density affect their visibility in a TEM. In fact, it is not known at all whether a distinction between the precipitates of various densities and bubbles can be made at all. Therefore, delving into this subject to clarify such issues and lay down certain basic guidelines and criteria could be helpful to a broad community dealing with noble gases and TEM.

With all these issues in mind, our goal in the current article is to present studies of the precipitates of all the noble gases (except radon) in a single material. The studies were performed on a complex borosilicate glass referred as SON68 which is a French nuclear waste glass (see Table S3 in the supplementary information for its composition). The reason to select this material for this study was its complexity which reduces the diffusion coefficient and favours the precipitation of noble gases at room temperature. The work will address some of the issues raised above and highlight the complexities and possible solutions in ambiguous situations. Furthermore, we will also present results on the simulations of the TEM images of precipitates and voids using Monte-Carlo methods to address the effect of the atom type, size and density of the precipitates on their visibility.