Aqueous Processing of Alumina and Phase Behavior of Polymeric Additives

Abstract

The goal of this research was to produce ceramic alumina microstructures with controlled porosity attained through the exploitation of polymeric phase separation. In order to accomplish this goal, a fundamental understanding of the aqueous colloidal behavior and sintering of alumina were required. The ability to manipulate the polymeric phase separation process was necessary in order to tailor porosity. For this reason, background information regarding the phase separation process and the parameters associated with its thermodynamic and kinetic properties were essential. The dispersion of aqueous suspensions of two a-aluminas (APA-0.5 and A-16 S.G.), were investigated. APA-0.5 was a high purity powder, and A-16 S.G. had MgO added as a sintering aid and contained relatively high levels of other impurities, most notably Na2O. The APA-0.5 was used as a standard alumina, while the A-16 S.G. represented a common industrial powder. Acids (HCl and H2SO4) and bases (NaOH and NH4OH) were used to adjust suspension pH. HCl was the only electrostatic dispersant found to substantially alter suspension viscosity. The z -potential and critical coagulation concentration (CCC) of electrostatically dispersed with HCl suspensions were measured. A correlation was found between the CCC and the inability of H2SO4, NaOH, and NH4OH to disperse alumina. The inorganic anionic polyelectrolytes examined included sodium silicate, sodium hexa-metaphosphate, and sodium carbonate. Comparisons were made between plots of the different polyelectrolytes (viscosity as a function of dispersant level), as well as between the two aluminas. Rheological phenomena correlated with z -potential measurements, the dissociation behavior of the polyelectrolytes, and the powder surface chemistry. The organic anionic polyelectrolytes examined include the sodium and ammonium salts of poly(methacrylic acid) and poly(acrylic acid), and neutralized citric acid. Comparisons were made between dispersants (viscosity as a function of dispersant level), including effectiveness and general rheological trends. Rheological phenomena were correlated with z -potential measurements, the dissociation behavior of the polymers and the powder surface chemistry. HCl and NH4-poly(acrylic acid), PAA, were used to stabilize alumina suspensions. A dual minima was found in the PAA dispersed suspensions containing MgO. The dispersant level of the first minima was a function of the concentration of sodium, shifting to lower dispersant levels for higher sodium concentrations. The second minima corresponded to the PAA dispersion of alumina. The critical coagulation concentration was measured for suspensions stabilized to a minimum in viscosity via electrostatic (HCl), inorganic electrosteric (sodium silicate and sodium hexa-metaphosphate), or organic electrosteric (NH4- and Napolymethacrylates, NH4- and Na-polyacrylates, and citric acid) dispersants. The salts investigated were the chlorides and sulfates of sodium, magnesium, and calcium. The thickness of the electrical double layer (d) around an alumina particle was calculated as a function of the electrolyte concentration and valence of the counter-ion, using a capacitance model. All of the suspensions coagulated at the same critical d value (~0.96 nm). Estimations of critical coagulation concentration using the critical d values obtained from MgCl2 additions agreed with rheological observations. The dependence of microstructure development on the type of dispersant used in the processing of alumina was investigated. Bulk density, and percent linear shrinkage measurements were used to evaluate the “green” density of pellets slip cast then calcined to 1000°C. Firing temperatures of 1200°C, 1400°C, and 1600°C were used to evaluate the densification process using bulk density and shrinkage measurements. SEM micrographs of pellets fired to 1400°C, polished, then thermally etched, display variations in morphology and grain size. The addition of Na+ ions to the suspensions resulted in abnormal grain growth, the organic component inhibited grain growth, and the inorganic dispersants severely inhibited grain growth. Binary-polymer/solvent and ternary-polymer/solvent interactions were observed using optical microscopy and light scattering (turbidity). The polymers investigated were poly(methacrylic acid) (PMAA), poly(vinyl alcohol) (PVA), and poly(ethylene glycol) (PEG). A ternary mixture design was used to evaluate the dependence of turbidity on polymer concentration. An increase in the degree of polymeric interaction was correlated with an increase in turbidity. The effects of pH, background electrolyte, electrolyte concentration, and temperature on the maximum in turbidity was evaluated using a statistical experimental design. The polymer concentrations, and the ratio of one polymer to another, were found to have the most profound effect on increasing turbidity. The solution pH and electrolyte conditions played important roles in the complexation of PMAA with PVA and PEG. An increase in the solution temperature decreased the turbidity of PMAA-PEG solutions, and increased the turbidity of PVA-PEG and PMAAPVA solutions. Based upon the information gathered, the microstructures developed when dextran sulfate and PEG were added to an aqueous alumina suspension, were the result of polymeric phase separation; the pores of which were thermodynamically stable. Using the knowledge gained from the investigations, a suspension can be processed outside the phase separating region by manipulating the thermodynamic properties of the polymer system. Phase separation can then be induced by changes in the suspension pH, electrolyte level, and temperature. The size, shape and connectivity of the pores can be controlled through kinetics. Although only demonstrated for alumina, the approach could be extended to other systems, providing an easily fabricated sample with tailored porosity.