Thesis (M.S.)--University of Michigan, 1986.

Thesis (M.S.)--University of Michigan, 1996. / Includes bibliographical references.

Thesis (Ph. D.)--University of Washington, 2006. / Vita. Includes bibliographical references (leaves 141-148).

This work was concerned with compositional analysis of archaeological ceramics from the Nile Valley and surrounding region. This is a vital step in the tracing patterns of ceramic production in Ancient Egypt, a topic which has not been investigated in any detail because of the apparent physical homogeneity of the raw materials. For this reason a large number of elements (22) was measured by neutron activation to give maximum differentiation between the samples. Multivariate clustering methods were used to identify sample groups on the basis of their composition. The main aim of the work was to examine the basis of the fabric classification procedure known as the Vienna System, in terms of the elemental composition of the ceramics. This was achieved, not only in terms of the major fabric groups ( Nile alluvium and Marl clays) but also in finer detail. The composition of these Egyptian pottery sherds is therefore a complex function of the fabric and of the geographical origin ( provenance) since clay compositions vary from place to place. In the case of the Marl fabrics, the compositional variations associated with the fabric are large and overshadow variations arising from provenance, whereas for the Nile alluvium pottery, fabric and provenance variations were more equal

No description available.

No description available.

No description available.

No description available.

No description available.

Although the oxides of the Groups VB and VIB elements (Cr, Mo, Nb, Ta, V, W) have frequently been used as nucleating agents in glass-ceramics, there has been little rigorous study of their effects. It has been suggested that they serve to reduce the interfacial tension between the subcritical nucleus and glass matrix, and thus encourage nucleus growth; this would cause an increase in nucleation rates in both homogeneously and heterogeneously-nucleating systems. There has been no attempt to demonstrate this, however. A test of the effects of these oxides on the steady-state nucleation rate and induction period in homogeneously-nucleating lithium disilicate glass was made by substituting one and two mole-percent of MoO₃, Nb₂O₅, Ta₂O₅, V₂O₅ or WO₃ for SiO₂, and heat-treating the glasses under identical conditions. The "two-step" method was used, consisting of nucleation for variable periods at 733 K and development at 843 K, followed by standard metallographic specimen preparation and inspection by optical reflection light microscopy. Crystal (i.e., nuclei) density was calculated using the stereological method of DeHoff and Rhines. It was shown that this method yielded results independent of crystal size for a given glass, and that the relationship between crystal size and shape supported the hypothesis that lithium disilicate nucleates as a rod-shaped particle, becoming more spherical as growth continues. It was found that MoO₃, Nb₂O₅, Ta₂O₅ and WO₃ substitutions reduced the nucleation rate in the experimental glasses and increased the induction period, the opposite of what was hypothesized. V₂O₅ substitutions yielded a slight nucleation-agent effect. The anti-nucleation agent behavior was associated with increased inhomogeneity in the heat-treated glass samples, resulting from varying cooling rates from the melt through the cross-section. Analysis of the equations linking nucleation rates and induction periods to glass properties suggests that the changes in both the nucleation parameters and the effect on the results of thermal inhomogeneity were primarily linked to higher viscosities in the substituted glasses.