Investigation of the thermodynamics of iron phosphate melts and stability of iron phosphate glasses: effects of composition and iron redox ratio

Author

Melodie Linda Schmitt

Keywords and Phrases

Crystallization; Dissolution; Glass stability; Iron phosphate glasses; Iron redox ratio; Optical basicity

Abstract

“Iron phosphate glasses with Fe/P = 0.33 - 0.67, O/P = 3.0 - 3.5 and Fe²⁺/ΣFe = 0.16- 0.52 were prepared by altering batch compositions and melt conditions. Thermal analyses indicate polyphosphate glasses with intermediate chain lengths are most stable against devitrification. Heterogeneous crystallization occurs on the glass surface and is dependent on iron valence state, particle size, heating rate, temperature and oxygen availability in the atmosphere. Oxidation of ferrous iron on the glass surface occurs at temperatures as low as Tg, and thus crystallization behavior and stability can be altered via post-melt heat treatments. A novel preparation of thin glass “bubbles” was developed to allow examination of iron phosphate glasses using optical spectroscopy. Absorption near ~476 nm was determined to be predictive of iron valence state, and a deconvolution method is proposed to analyze iron coordination environments. Two models were developed using optical basicity and a statistical, bonding-level representation of the glass structure to investigate the effects of composition on iron redox equilibria and aqueous dissolution behavior. A new approach of calculating heats of formation from group basicities yields a linear correlation with thermochemical data for equivalent crystalline compounds, and the thermodynamic model of iron redox equilibria predicts trends in melt viscosity with bulk basicity, chain length and alkali mixing. An empirical fit of the model predicts iron redox ratios of glasses prepared here within experimental uncertainty. The bond hydration model yields similar spread in predicted versus experimental dissolution rates as previously established models based on free energies of hydration, yet is based solely on composition with no requisite for thermochemical data.”--Abstract, page iii.

Advisor(s)

Brow, Richard K.

Committee Member(s)

Fahrenholtz, William
Switzer, Jay A., 1950-Smith, Jeffrey D.
Kim, Cheol-Woon, 1966-