Engineering topics
Publications and source records attributed to Uhlmann, D. R..
Mathematical model calculates rate of small oxygen-bubble contraction in soda/lime/silicate melts in presence of foreign nondiffusing gas. Process of interest in optical glass fabrication.
When analyzing crystal nucleation in glasses, it is difficult to distinguish between homogeneous and heterogeneous nucleation. A method is proposed to investigate the nature of crystal nucleation in Li2O-2SiO2, and calculations are performed to determine under which conditions saturation effects help ascertain the nature of the nucleation. The capability of detecting impurity particles in unheated glass via standard micrographic techniques (SEM or REM) and small angle X-ray scattering (SAXS) is explored. Results indicate that if the maximum impurity particle densities do not exceed 2 x 10 to the 7th per cu cm then heterogeneous nucleation may be excluded and most impurity particles with densities in excess of 10 to the 10th per cu cm are detectable by SAXS. It is concluded that crystal nucleation in this system most probably occurs homogeneously.
Analytical expression used to calculate time it takes for stationary bubbles of oxygen and carbon dioxide to dissolve from glass melt. Technique based on analytical expression for bubble radius as function time, with consequences of surface tension included.
The process of glass formation is discussed from several perspectives. Particular attention is directed to kinetic treatments of glass formation and to the question of how fast a given liquid must be cooled in order to form a glass. Specific consideration is paid to the calculation of critical cooling rates for glass formation, to the effects of nucleating heterogeneities and transients in nucleation on the critical cooling rates, to crystallization on reheating a glass, to the experimental determination of nucleation rates and barriers to crystal nucleation, and to the characteristics of materials which are most conducive to glass formation.
Cooling rates required to form glassy or partly-crystalline bodies of 14 lunar compositions have been estimated using a previously introduced, simplified model. The calculated cooling rates are found to be in good agreement with cooling rates measured for the same compositions. Measurements are also reported of the liquidus temperature and glass transition temperature for each composition. Inferred cooling rates are combined with heat flow analyses to obtain insight into the thermal histories of samples 15422, 14162, 15025, 74220, 74241, 10084, 15425, and 15427. The critical cooling rates required to form glasses of 24 lunar compositions, including the 14 compositions of the present study, are suggested to increase systematically with increasing ratio of total network modifiers/total network formers in the compositions. This reflects the importance of melt viscosity in affecting glass formation.
Explore the source record for details and available documents.
The simplified model of glass formation and the development of partial crystallinity in cooled bodies has been applied to lunar compositions 10060, 15028, 15086, 15101, 15286, 15301, 15498, 15499, 60255, 65016, 77017, Apollo 15 green glass and LUNA 24 highland basalt. The critical cooling rates for glass formation predicted by the simplified model are found to be in good agreement (to within an order of magnitude) with those predicted by the exact treatment of crystallization statistics. These predicted critical cooling rates are in even better agreement (a factor of 2) with measured values of the rates required to form glasses of the materials.
The basic techniques used in the processing of glasses and crystalline ceramics under terrestrial conditions are briefly reviewed, and the features of the space environment relevant to the processing of glasses are examined. These include reduced gravitational forces, a vacuum of essentially unlimited pumping capacity, unique radiation conditions, and the unlimited dimensions of space. Of these factors, particular attention is given to reduced gravitational forces, and the advantages of containerless processing are discussed. Finally, current programs concerned with glass processing in space are reviewed along with additional areas which merit investigation.
Finite difference solutions of the mass transport equations governing the dissolution (growth) of a rising gas bubble, containing a single gas, in a glassmelt were obtained. These solutions were compared with those obtained from an approximate procedure for a range of the controlling parameters. Applications were made to describe various aspects of O2 and CO2 gas-bubble behavior in a soda-lime-silicate melt.
A model of solute partitioning in a finite body under conditions of continuous cooling is developed for the determination of cooling rates from concentration profile data, and applied to the partitioning of zirconium between ilmenite and ulvospinel in the Apollo 15 Elbow Crater rocks. Partitioning in a layered composite solid is described numerically in terms of concentration profiles and diffusion coefficients which are functions of time and temperature, respectively; a program based on the model can be used to calculate concentration profiles for various assumed cooling rates given the diffusion coefficients in the two phases and the equilibrium partitioning ratio over a range of temperatures. In the case of the Elbow Rock gabbros, the cooling rates are calculated from measured concentration ratios 10 microns from the interphase boundaries under the assumptions of uniform and equilibrium initial conditions at various starting temperatures. It is shown that the specimens could not have had uniform concentrations profiles at the previously suggested initial temperature of 1350 K. It is concluded that even under conditions where the initial temperature, grain sizes and solute diffusion coefficients are not well characterized, the model can be used to estimate the cooling rate of a grain assemblage to within an order of magnitude.
Explore the source record for details and available documents.
The variation of viscosity with temperature was determined in the 200-2000 K range for a Muong Nong tektite material. The viscosity at the liquidus temperature of 1320 C is 20,000 P; treatments between 900 and 1300 C do not result in significant crystallization in the natural sample except when the sample is heated in contact with a synthetic tektite composition. Two synthetic microtektite with lower SiO2 contents than the Muong Nong material were also examined; heat flow calculations were performed for 2.5 to 10 cm spheres of tektite when cooling by radiation.
The effect of a foreign nondiffusing gas on the rate of shrinkage of an oxygen bubble in a soda-lime-silica melt was studied. The rate of change of bubble radius with time was computed using the quasi-stationary approximation. The effects of melt undersaturation and initial fraction of foreign gas in the bubble are considered and compared with those calculated using previously derived expressions.
Explore the source record for details and available documents.
The paper describes an analytical method for calculating in detail the size distributions of small crystallites and nuclei in supercooled liquids as they are being quenched to form a glass and subsequently reheated above the glass transition to produce crystallization. This method is applied to experiments performed using differential thermal analysis (DTA) to estimate the barriers to crystal nucleation and the cooling rates required to form glasses or bodies with various degrees of crystallinity. DTA data and derived nucleation barriers are reported for anorthite and for the following lunar compositions: 15498, 15418, matrix and intrusion compositions of breccia 15286, Apollo 15 green glass, Luna 24 highland basalt, and 65016.
The growth rates and interface morphologies of crystals of synthetic compositions in the anorthite (CaAl2Si2O8)-albite (NaAlSi3O8) plagioclase feldspar system are measured in an investigation of the crystallization of igneous rocks. Mixed plagioclase glasses with compositions of 75% and 50% anorthite were observed using the microscope heating technique as they crystallized at temperatures near the liquidus, and 75%, 50% and 20% anorthite crystals were treated by resistance heating and observed at greater degrees of undercooling. Growth rates were found to be independent of time and to decrease with increasing albite content, ranging from 0.5 to 2 x 10 to the -5th cm/min. The crystal morphologies for all compositions are faceted near the liquidus and become progressively skeletal, dendritic and fibrillar with increasing undercooling.
Contents: a kinetic treatment of glass formation; effects of nucleating heterogeneities on glass formation; glass formation under continuous cooling conditions; crystallization statistics; kinetics of crystal nucleation; diffusion controlled crystal growth; crystallization of lunar compositions; crystallization between solidus and liquidus; crystallization on reheating a glass; temperature distributions during crystallization; crystallization of anorthite and anorthite-albite compositions; effect of oxidation state on viscosity; diffusive creep and viscous flow; high temperature flow behavior of glass-forming liquids, a free volume interpretation; viscous flow behavior of lunar compositions; thermal history of orange soil material; breccias formation by viscous sintering; viscous sintering; thermal histories of breccias; solute partitioning and thermal history of lunar rocks; heat flow in impact melts; and thermal histories of olivines.
A simplified model of glass formation based on the formal theory of transformation kinetics is presented, which describes the critical cooling rates implied by the occurrence of glassy or partly crystalline bodies. In addition, an approach based on the nose of the time-temperature-transformation (TTT) curve as an extremum in temperature and time has provided a relatively simple relation between the activation energy for viscous flow in the undercooled region and the temperature of the nose of the TTT curve. Using this relation together with the simplified model, it now seems possible to predict cooling rates using only the liquidus temperature, glass transition temperature, and heat of fusion.