Deposition rate dependence of the critical optimization temperature for thin film properties: Evidence for the disordered area reevaporation mechanism

A wide variety of inorganic and organic vacuum‐deposited materials have previously been shown to exhibit sharply optimized properties when the condensation (substrate) temperatures T_s are held in a very narrow range near 0.33 of the respective normal boiling points. The optimized properties include many electrical, photoelectrical, optical, and structural parameters, and the immediate cause has been shown to be singularities in the crystallite size and orientation in films formed at the critical temperature T₀. It has been proposed that the underlying mechanism of this critical optimization effect is a new process involving reevaporation of small well‐defined but disordered (amorphous) areas of the film. This mechanism predicts an ∼7.5% change in T₀ for each decade change in deposition rate at typical rates. A definitive check of this prediction is in fact experimentally complicated, and none has previously been attempted. We report here a careful study of the deposition rate dependence of T₀ for a model material, metal‐free phthalocyanine (PC) (chosen because of its lack of stoichiometric complications). The experimental rate dependence of T₀ for PC (18±3 °C for a decade change in rate) is found to be in excellent agreement with the theoretical prediction (∼20 °C). This is good evidence for the disordered‐area reevaporation model. In addition, we discuss in detail the alternative possibility that only conventional surface diffusion factors are at work, and conclude that such mechanisms are not likely to be the explanation. Based on our confirmation of the rate dependence of T₀, we indicate how to calculate T₀ for any appropriate material and deposition rate. Finally, measurements on films whose initial layers were deposited at different values of T_s from the bulk suggest that the disorder which is present for T_s<T₀ is associated with difficulties in establishing the first fully crystalline layers of the film (probably because of the presence of the disordered areas); they also suggest that the shortcomings of films deposited at T_s>T₀ may arise during the later stages of deposition, perhaps because of excessive out‐of‐plane crystallite growth.