Theories and experiments on the mechanism of growth of crystals from aqueous solution phase and the related phenomena are reviewed critically. The results of experiments which have been carried out systematically mainly for NaClO_3 are summarized on the growth rate vs. supersaturation relations, the in-situ observations of growth and dissolution processes, and on the dislocation configurations revealed by X-ray topographic studies. The experimental results are confronted with the theories. Based on these results, growth mechanisms of crystals bounded by flat or hopper faces from aqueous solutions are discussed, and the general scheme is presented.
Except in particular cases a negative view of the whisker growth due to axial screw dislocation prevails on the basis that its verifications are few. Judging from observations of the Te-whisker growth process and the growth conditions and configuration of whiskers in which axial dislocation is recognized, however, the posibility of a dislocation slipping out of whiskers by thermal agitation can be speculated. Meanwhile theoretical studies on whisker due to VLS mechanism, though several theories about its rate-determining process have been reported, seem to be beginning. As compared with the initial Si-whisker due to VLS mechanism, the recent studies are expanding into a considerably anomalous area through diversification of kinds and methods. Stimulated by this fact, reports of research of whisker growth under the effect of an unknown impurity are steadily increasing. Thus the recent trend indicates that a new growth mechanism under the effect of impurity is gradually establised.
Growth rates of quartz crystals are discussed on the basis of the in situ determination of the solution temperature in an autoclave, in order to obtain exact values of kinetics in hydrothermal growth. The solution temperature under hydrothermal condition was directly determined using a specially designed autoclave, in which thermometers were inserted. The temperature of the growth solution in the autoclave was generally indicated between 10 and 30℃ lower than that measured from the outside of the autoclave body. It was found that the temperatures of the growth zone show periodical oscillation at ± 2 or ± 3℃, whereas that of dissolution zone of nutrient was nearly held at a constant temperature. Hydrothermal growth kinetics of quartz crystals was reconsidered on the basis of the solution temperature in the autoclave. The activation energies in the normal direction of the basal face (0001) gradually change from 16 kcal/mol to 22 kcal/mol, as the temperature difference ΔT between growth and dissolution zones decreases from 25 to 10℃. The activation energies in the direction of the a-axis <1120> were independent of the ΔT's and showed 13 kcal/mol.
Experimental methods are presented which allow crystal growth and thermochemical studies up to 2500℃, in evacuated and electron-beam-sealed W-crucibles. Crystal growth from the vapour phase by chemical transport and sublimation; from the melt by solidification (Bridgman's). Phase diagram studies with DTA (Metter Thermoanalyser, SHT Furnace) up to 2400℃ in specially erroded W-capsules ( φ 4mm, length 10mm), sealed by electron beam. In this way no losses due to evaporation or changes of stoichiometry can take place. Vapour pressure measurements are made in the same apparatus or in a specially designed UHV-furnace capable of 2200℃ (Vacuum after baking 10^<-9> torr, when at temperature 10^<-6> torr). For equilibrium evaporation measurements, Knudsen cells made out of W are used (hole φ 0.1mm). For Langmuir free evaporation open cells and plates are used. Measurements of heats of formation or heats of reaction are made with our LKB Solution Calorimeter--Scanning Electron Microscopy with resolution of approximately 200A. Some examples of the use of these methods to investigate high temperature crystal growth phenomena are given concerning Surface roughning near the m.p. of solid-vapour interfaces Segregation of impurities at the interface during evaporation. Diffusion over the steps seems inhibited. Segregation of one component at the interface due to incongruent evaporation. Thermodynamics of high temperature evaporation in EuTe and YbTe between 900 and 2200℃. Phase diagram studies up to 2400℃.
Valence instabilities are found in solids with narrow bandgap, and cations which may achieve two different valence states. The change in valence results here not by transfer of electrons from the cation to the anion, but by promotion of the electrons of the cation from a localized (4f or 5f) state (characteristic of the divalent cation) to a delocalized 5d-Band (characteristic of the trivalent cation). Application of pressure increases the crystal field splitting of the 5d-Band and makes the energy gap of e.g. semiconducting, black, divalent SmS smaller. At 6.7 Kbr a semiconductor-metal transition takes place to the trivalent, golden, metallic SmS. XPS measurements show that both Sm^<2+> and Sm^<3+> are present in the collapsed state. It is not, however, clear if a new mixed valent state appears. Unique for this phase transition however is an approximately 10% decrease in volume (due to the Me^<2+> → Me^<3+> transformation) without change of the NaCl-structure. TmSe contains already at normal pressure mixed valence. Up to now, samples of TmSe showed irreproducible average valence and therefore large scattering of their lattice constant (Bucher et al.). In this lecture, it is shown that a) The lattice constant of TmSe samples (and therefore the valence of Tm) becomes reproducible when the nonstoichiometry is controlled. b) The valence of Tm can be varied between Tm^<3+> and Tm^<2.71+> by proper adjustment of stoichiometry. Variations almost in the whole range between Tm^<3+> and Tm^<2+> can be achieved in the systems TmSe-TmTe and TmSe-EuSe. c) Measurements of the density show that these variations of valence are followed by the formation of high concentration of vacancies. d) Investigations of the Tm-Se phase diagram up to the m.p. of TmSe (2030℃) show several phase transition which may be the result of valence segregation (separation of Tm^<3+>Se and Tm^<2+>Se). e) The phase diagram shows that the m.p. of the stoichiometric compound is lower. This is an indication of instability for a composition at which the valence fluctuation is expected to appear.