Arnold L. Pundsack

Vacuum deposition onto softenable substrates: Formation of novel subsurface structures

Abstract The vacuum deposition of a wide range of evaporant materials onto soft (heated thermoplastic) substrates has been studied for the first time. It was previously known that, in some circumstances, selenium deposited onto heat-softened polymers forms a nearly close-packed monolayer of submicron, almost monodisperse, amorphous spheres embedded a few tens of nanometres under the surface; this novel structure has wide applications as a photographic film. It was not known whether this phenomenon was unique to selenium and its alloys (which are unusual in being amorphous and polymeric), and if not what the criteria were for formation of these or other subsurface structures. Under deposition conditions similar to those used for selenium (substrate temperature T s ∽120°C), tin, indium and silver have now also been observed by electron microscopy to form novel subsurface particulate structures. The tin and indium structures are similar to that of selenium, except that the spheres are nearly single crystal with greater poolydispersity and are arranged in a less pristine monolayer with a fair degree of multiple stacking. The silver structure consists of very small (about 25 nm) crystalline aggregates, very tightly packed to a depth of several particles below the polymer surface. For considerably higher T s , the embedded silver become discrete and well separated, resembling more the selenium, tin and indium structures. Under our typical deposition conditions, all other materials tried form only above-surface deposits, as do selenium, tin, indium and silver when either the substrate temperature is too low or the deposition rate too high. Subsurface particle formation is shown to be an activated process, the activation energies depending on both the evaporant and the polymer. The above-surface deposits generally grow in a Stranski-Krastanov mode, with a thin (1–3nm) continuous initial deposit and superposed island structures; this is extremely unusual for a non-single-crystal substrate. Our results, and a theoretical analysis to be published separately, suggest that the criteria for formation of subsurface structures are certain minimum levels of the evaporant surface diffusion and polymer fluidity at T s . Thus, for typical thermoplastics, this criterion usually requires a fairly low melting point evaporant; most other inorganic materials have been shown to have a thermodynamic tendency to form subsurface monolayers, and (like silver) they will probably overcome kinetic limitations if T s can be made high enough. Thermodynamic considerations suggest partially embedded structures for organics, and this was also observed.

Analysis of Thin‐Film Germanium Epitaxially Deposited onto Calcium Fluoride

An experimental investigation of the formation conditions of thin‐film Ge epitaxially deposited onto (111) surfaces of CaF2 and the resultant physical and electrical character is presented. The amorphous to crystalline transition region was found to be between 320° and 400°C, and the best single‐crystal orientation was found between 550° and 575°C. Average values of the mobility, conductivity, and the hole concentrations are also reported with an explanation of the problems encountered in their determination.

Subsurface particle growth kinetics in physical vapor deposition

In a well‐defined range of conditions, selenium vapor deposited onto a thermoplastic substrate forms a monolayer array of spherical particles located just beneath the surface. A model for the growth of these particles has been proposed. In it, the particles grow by two mechanisms: (i)capture of selenium molecules diffusing into the substrate, and (ii)coalescence with other particles when they grow into contact. Expressions for both rates are presented and analytical solutions for the time dependence of the numbers and sizes of the particles are devived for two limiting cases—complete condensation (in which all impinging materials become incorporated into the particles) and incomplete condensation (in which re‐evaporation of adsorbed species limits the kinetics). Experimental data confirm the model’s predictions.