Sputter-induced isotopic fractionation at solid surfaces

Abstract

Elemental and isotopic mass fractionation in both binary and multicomponent media are investigated within the framework of the familiar collision-cascade model for sputtering. Some of the most salient features of the phenomenon are explicable on this basis. It is found that the partitioning of beam-deposited energy among the various target components can account for differentiations in the secondary recoil fluxes only on the order of one part per thousand, indicating the importance of the surface potentials when large enrichment effects occur. A mechanism governing the translation of internal recoil fluxes into external sputtered fluxes is proposed in order to account for isotopic fractionation, for which the surface binding effects are assumed to be negligible. The predicted initial fractionations are δf(40Ca : 44Ca) =33 parts per thousand in the calcium-containing mineral plagioclase and δf(40Ca : 44Ca) =24 parts per thousand in CaF2, in reasonable agreement with recent data on isotopic fractionation.

Possible isotopic fractionation effects in material sputtered from minerals

“…The 'surface flux' model was adopted because the nonstoichiometric emission of isotope species expected from the bulk recoil flux alone was found to be much smaller than indicated by experiment. Thus Russell et al [1980] found 8(n•Ca: 4øCa) for material sputtered from a plagioclase target to be on the order of • -20% o, while predictions based on nonstoichiometries in the bulk recoil flux indicated values no less than -1%o [Watson and Haft, 1980]. On the other hand, the surface flux model is in adequate agreement with the data 2, respectively, and furthermore we define AM/M, --e. To il-of Russell et al [1980] for Ca fractionation in plagioclase and lustrate the fractionation behavior for small mass differences in fluorite.…”

“…The expression given by (2) and hence the result (6) arise from a detailed solution of the transport equation describing energy sharing among recoiling atoms [ Watson, 1980;Watson and Haft, 1980]. A source of particles is created by primary collisions between the incident ion and a target atom.…”

“…The actual calculations are thus perturbation theory calculations, with the target surface introduced only at the final step as the boundary across which ponent medium. According to Watson and Haft [1980], the bulk flux of recoiling atoms is very nearly stoichiometric (except for extreme differences in target atom masses or for the very highest energy particles, which, however, constitute only a small portion of the total flux). This flux has an energy dependence of the form An•/E •, for species i, where A is a constant.…”

“…pression for the isotopic fractionation expected in the material sputtered by the collisional recoil process from a surface containing isotopes of species k and l (This expression for 8k,• is the one given in Watson [1980]. It differs slightly from that found in Watson and Haft [1980]…”

“…Sequential collision events such as those shown in Figure 2 are included in the calculation but cannot physically occur in the true target since the collision point is outside the target surface. Watson [1980] and Watson and Haft [1980] made an attempt to include effects introduced by the presence of a surface in a less rigorous way. They postulated that the atoms comprising the extreme outer surface layer of the target cannot participate fully in the recoil cascade, primarily because the geometry makes it difficult for them to transfer energy to subsequent atoms after they themselves have been struck.…”