Because of suggestions that at least some alum crystals may become ferroelectric at low temperatures, and that a ferroelectric crystal would show anomalies in the absorption of sound, the attenuation of 10, 30 and 50 M.c. sound waves has been measured in alum single crystals over the temperature range of 300°K to 4.2°K. Results are available for both longitudinal and transverse waves propagated in the [111] direction. In the temperature region between 160°K and 70°K we have obtained some very large attenuation peaks. For 10 M.c. these peaks have attained values up to 14 DB/CM, with a normal background attenuation of approximately 1 DB/CM. For the case of 30 M.c. and 50 M.c. the peaks are much higher than for 10 M.c. However, due to the limitations on the electronic equipment our measured values were limited to 14 DB/CM. In AlNH4(SO4)2 · 12H 2O the attenuation peak is in the temperature region where both the real and imaginary parts of the dielectric constant become large, possibly indicating that the crystal is ferroelectric. A calculation, using a simple theory for the attenuation of sound waves in a ferroelectric material, gave results in rough agreement with our measured values. The single crystals used in this experiment were grown here at Rice. The procedure for growing alum single crystals is discussed in the appendix.

Bohmite, gamma-AlOOH, has been found to exist in nature and may be prepared in the laboratory in a variety of ways. From the close similarity of the powder x-radiograms of bohmite and the mineral lepidocrocite, gamma-FeOOH, it has been concluded that the structures are analogous. The existence of the closely related compound gamma-ScOOH has clearly been established by x-ray diffraction and dehydration isobaric studies, but quantitative x-ray studies have not been made. There are two crystalline modifications of the monohydrate of aluminum oxide with the formula AlOOH. The modification corresponding to diaspore, alpha-AlOOH, has been found to be isomorphous with goethite,alpha-FeOOH, of known structure. The second modification corresponding to bohmite, gamma-AlOOH, is isomorphous with lepidocrocite, gamma-FeOOH. The structure of lepidocrocite has been determined by Ewing employing oscillation and rotation photographs of a naturally occurring single crystal. Even though bohmite occurs in nature, it has not been possible to obtain individual crystals suitable for single crystal x-ray studies. Milligan and Weiser, from dehydration isobars of scandium trihydroxide and x-ray diffraction studies, reported the existence of scandium oxide monohydrate, gamma-ScOOH, which they believed to be isomorphous with bohmite. In the present investigation the unit structure, space group symmetry, and detailed atomic arrangement of gamma-AlOOH and gamma-ScOOH have been determined. Since numerous attempts to grow single crystals of both of these monohydrates failed, it was necessary to confine the work done here to powder techniques.

The goal of this research is to observe the variation of the zeta-potential with the concentration and valency of coagulating electrolyte. Using this data, the validity of Verwey and Overbeek's theory will be tested to determine whether or not the stability ratio W is a satisfactory criterion of colloid stability.

The addition of a solution of an alkali hydroxide to a solution of a heavy metal salt usually gives a voluminous, gelatinous precipitate containing an amount of water depending upon the conditions of formation and subsequent treatment of the sample. While many of the cherished hydrates of the early investigators have been shown to be non-existent, in many cases modern methods of investigation have shown the situation in may cases to be such more complicated than it was formerly supposed. Examples of this will be given in following sections of this paper. For the purposes of this paper a hydrous oxide shall be defined as an adsorptive oxide containing an amount of adsorbed water depending upon the conditions of formation and treatment of the sample. A hydrate shall be considered to be a definite, crystalline, chemical compound. In most definite "hydroxides" it has been shown by x-ray analysis that the crystal lattice consists of metal ions and hydroxyl ions, and that in certain "hydrates", H2O groups are arranged around some central group after the manner of Werner complexes. "Hydroxides" may also contain additional chemically bound water over and above the true hydroxide. However, since in so many cases the complete structure has not been determined by x-ray analysis no attempt shall be made, in general, to distinguish between "hydroxides" and "hydrates". It is not considered that anything would be gained by using the terms "water of crystallization" and "water of constitution".

X-ray investigations have been carried out on two organic compounds, ditropyl and xanthane hydride. Their unit cell dimensions and space groups are summarized in the table below. Compound \a1\ A \a2\ A \a3\ A alpha1 alpha2 alpha3 Space group Ditropyl 8.79 6.52 10.79 116°30' 93°10' 86°25' P1¯ Xanthane Hydride 4.08 10.65 12.85 90° -' 82°5' 90° -' P21/c It was shown that the tropyl ring is probably a seven-membered ring, rather than the proposed bicyclic alternate. Xanthane hydride was shown to be a planar molecule and a method for solving the structure is proposed.

A number of different methods for studying the fine structure of silica gel have been reported in the literature. Extensive sorption-desorption studies have been made and from these studies pore size and distribution, surface area, and average particle size have been determined. Sorption-desorption studies and porosity studies have been employed also to determine the effect of heat treatment on the structure of silica gel. Small-angle x-ray scattering data have been used to determine particle sizes in silica gel. However, the assumptions which were made in the method of analysis of the small-angle x-ray scattering data are not completely justified in the case of silica gel. No investigation of the effect of heat treatment on the small-angle x-ray scattering by silica gel has been reported in the literature. The electron microscopic studies of silica gel reported in the literature are at relatively low magnifications and show the morphology of the aggregates only. No systematic electron microscopic study of the fine structure of silica gel has been reported. If the optimum magnification and resolution of modern electron microscopes is utilized, it should be possible, in a well dispersed sample of silica gel, to resolve the very small particles and pores, which sorption-desorption studies and small-angle x-ray scattering studies indicate to be present. In this investigation an electron microscopic study has been made on a series of silica gels, which had been subjected to systematic heat treatments from 30°C to 930°C. There is also included in this investigation the results of electron diffraction and small-angle x-ray scattering studies on the same series of silica gels.

The purpose of this investigation was the calorimetric determination of the saturated heat capacity and the heats and temperatures of all phase transitions of propionitrile (ethyl cyanide) from about 15°K to room temperature. From this data, and gas imperfection data available in the literature, the entropy of this compound was calculated in the ideal gas state. This value was then compared with the entropy calculated from Raman and infrared spectroscopic data. Agreement between the two methods may be used as a check on the molecular model used and on the accuracy of the calorimetric data. No experimental value for the entropy of propionitrile has been reported. Janz and Duncan have calculated a value from spectroscopic data. There is some room to doubt the validity of their calculations, however, as will be shown in the section on the entropy. The Raman and infrared spectra have been measured by several authors. Lerner and Dailey and Laurie have reported the microwave spectrum with values for the principal moments of inertia and the torsional frequency. The sample used in this research was found to contain an appreciable amount of solid soluble impurity. Putnam investigated the effects of this type of impurity on sharpness of melting data. However, his results were not completely satisfactory, and this analysis was extended to include premelting heat capacity data.

For many years, the adsorption of gases and vapors by solids has been studied mainly by measuring the amount of adsorption as variant of pressure and temperature. Several theories of adsorption have been advanced, and equations derived to fit existing data with varying degrees of accuracy. Postulation of a mechanism to explain the processes of adsorption has been hampered by the lack of evidence regarding the physical properties of the adsorbed phase. Experiments to directly determine densities, indices of refraction, and similar properties of the adsorbed phase have not been devised. Recently, however, measurements of the dielectric constants and magnetic susceptibilities of sorbate-sorbent systems have been reported. Since both dielectric constants and magnetic susceptibilities are properties which may vary with physical and chemical changes, some insight into the problems of adsorption may be gained by appropriate measurement of these quantities. It is the purpose of this thesis to report the results of an investigation of the diamagnetic susceptibilities of several vapors adsorbed on silica gel.

We present several related studies which probe the energy flow dynamics of overtone initiated photochemistry using theoretical methods. The nature of the classical dynamics for two model isomerization systems has been investigated with an assumed initially localized excitation to a high overtone vibration state of a bond oscillator. For these systems, the classical equations of motion are propagated and analyzed using nonlinear resonance theory. We find energy flows along resonance enhanced pathways leading to nonstatistical chemical reactivity. There exist, however, profound perturbations for both models due to nonresonant energy flow. It is concluded that this perturbation leads to significant complications which render mode selective photochemistry unobservable. Since most experimental investigations have been unable to identify nonstatistical behavior, these theoretical results are in agreement with the majority of previous observations. We identify therefore, the primary difficulty which prevents the observation of mode selectivity to be nonresonant energy flow. We also investigate the nature of the excitation dynamics of the overtone excitation process itself. Completely classical descriptions for this process are highly inaccurate, showing no excitation of overtone levels. Quantum mechanically, the dynamics for excitation are allowed. The overtone excitation process is thereby identified as an example of dynamic tunneling, which is slow relative to intramolecular vibrational relaxation, resulting in completely delocalized excitations at all times. It is not accurate therefore to assume an initially localized excitation, but interpretations which make such an assumption may still be reasonable if there is no competing dynamic process such as dissociation. These delocalized excitations are very eigenstate specific, which leads to yet another hypothesis. Multiple laser excitation, each laser tuned to a different excited eigenstate, will lead to coherent superpositions of these states, which may lead to destructive interference in chemically significant portions of coordinate space. As examples, we calculate that such a coherent multi-color (CMC) excitation can lead to long time localized bond excitations of high overtones for a polyatomic model and show that phosphorescence can be eliminated as a radiative channel for diatomic carbon monosulfide. Finally, overall implications of these separate studies on the nature of overtone photochemistry are discussed.

Auger electron spectra of matrix-isolated metal atoms have been recorded for the first time. Differences in extra-atomic relaxation between bulk metal and argon matrix isolated atoms resulted in shifts in the energy of LMM Auger electrons, $-3.1$ eV for potassium and $-8.7$ eV for titanium. The sum of the shift and the known difference between the energies of LMM Auger electrons from gaseous and frozen argon is a good estimate of the total extra-atomic relaxation energy for the bulk metal after Auger processes. Polyimide Isoindolo Quinazolin-dione (PIQ) was studied in detail with Direct Recoil Spectroscopy (DRS), X-ray Photoelectron Spectroscopy (XPS) and Secondary Ion Mass Spectroscopy (SIMS) to determine the effects of argon and nitrogen ion implantation, oxygen recoil implantation, and vacuum annealing. A hydrocarbon contaminant that cannot be detected by XPS alone was confirmed to exist with DRS. Ion bombardment, even at relatively low energies, causes profound graphitization of the surface. DRS and XPS data show that appreciable amounts of nitrogen and oxygen can be implanted into the polymer, and that the implanted species are stable up to 300$\sp\circ$C. DRS evidence suggests that water dissociates on the 3 keV argon ion bombarded PIQ surface, adding hydrogen to the surface but not oxygen. DRS, XPS and SIMS have been used to examine chemisorption and annealing on polycrystalline titanium. Dynamic surface processes were studied with DRS on the time-scale of 30 seconds, revealing that a surface phase transition takes place on Ar$\sp{+}$ sputtered titanium at only 500$\sp\circ$K. DRS in conjunction with XPS showed conclusive evidence for formation of a subsurface nitride after a nitride titanium sample was exposed to oxygen. A 'trickle down' mechanism is proposed to account for the formation of the subsurface species. There are at least two surface sites for hydrogen, nitrogen and oxygen. At room temperature, the hydrogens' lower site is saturated, even after sputter-cleaning.