# Low-Pressure, High-Solids Forming

September 28, 2000
Large ceramic hearths for appliances, such as this ceramic oven plate, are being injection molded at about 60 volume percent solids in less than 30 seconds.
Over the years, certain methods of forming, pressing and extruding have become established in spite of their identified, inherent weaknesses. Inertia is difficult to overcome, despite the obvious advantages of other processes. Dry powder pressing, for example, yields bodies with large density gradients that cause distortion on sintering. It is limited to uniaxial pressing of simple shapes with aspect ratios of 4 to 1 or less. In addition, most methods are restricted to the production of green parts of less than 60% density. Shrinkage during subsequent operations is too great to yield net shape, or even near net shape parts, as fired.

Low-pressure forming offers an alternative to dry powder pressing and other standard methods. The low-pressure forming process, previously referred to as low-pressure injection molding (LPIM), is capable of forming as-molded green parts having solids loadings in excess of 85 volume percent (v/o). These loadings depend on the relative density of the fluid medium, the binder and the ceramic powder.

Volume fractions express solids loadings independent of powder densities. Solids loadings are then, in effect, the green densities of the ceramic bodies. The volume fraction of solids is calculated using the simple formula below:

(v/o)s = {(wt/o)s / Ps} / {[(wt/o)s / Ps] + [wt/o)l / Pl]}

where v/o is the volume fraction, wt/o is the weight fraction, P is the density, s represents the solid phase and l represents the liquid phase.

Table 1 lists the maximum solids loadings of compositions that have been molded and compares weight percentages and volume percentages of these materials. Densities of the materials are included to complete the comparison.

Several items produced from rebounded fused silica.

## Formulating High-Solids Mixes

The development of high solids loadings has been previously described in detail in papers such as that by McGeary.[1] The difficulty comes in blending such high solids mixes with wetting agents and other surface active materials, collectively referred to as "surfactants," in such a way that the particles will readily flow past one another at low pressures.

Formulating these high-solids, low-viscosity mixes requires the application of two technologies. First, the powders must be blended so that the particle size distribution covers an extended range. The greater the range, the higher the solids loading that can be achieved. Second, the particle surfaces must be treated in such a way as to assure that they are fully wetted, and that agglomerates are minimized or, ideally, totally eliminated.

The blending of size fractions of ceramic powders follows a general rule based on simple geometrical relationships-each successively smaller size fraction should fit into the interstices formed among the larger ones. In the case of perfect spheres, diameter ratios of one-seventh that of the next larger sphere and one-tenth its volume have been derived. These rules, of course, are intended for spherical-shaped, monosized particles. In practice, a long particle size distribution with a gradual decrease in concentration with decreasing particle size is recommended. Shape has to be accepted for those raw materials available for the composition being formulated. Acicular and lamellar shapes complicate matters, generally leading to greatly reduced solids loadings. These irregular shapes should be avoided whenever possible, unless such geometries provide a desired property to the finished body.

A large (20+ kg) titanium diboride injection molded part.
Once the powders have been fully blended, surfactants are added to wet and deagglomerate the particles. The use of surfactants can, as yet, hardly be called a science. Some generalities can be made, for example, that an alkaline surfactant will be more effective on ceramic powders of an acidic nature. However, variances exist even within a single composition, such as carbon. For instance, it has been shown that the surfactants that work well with amorphous carbon are of little use with graphite, and vice versa.

Since practically all ceramic powders have a surface that is modified by sorbed water (physically or chemically attached), the best results can generally be achieved with surfactants that contain amine, hydroxyl or carboxylic acid groups. These surfactants contain active hydrogens that react and bond to the particles, splitting off water in the process, and they yield the most stable dispersions.

The long-term stability of these paraffin-based dispersions has been clearly demonstrated. In one instance, the excess portion of a production mix was solidified and stored on a shelf in plastic bags for several years. With the customer's request for additional parts, the mix was remelted and stirred, and parts were produced under the same conditions that were used to produce the original parts. They were identical in all respects to those in the original lot.

Paraffin is the material of choice for these high-solids, low-viscosity mixes for a number of reasons. First, the paraffins are inexpensive and non-toxic. Furthermore, unlike water, they do not evaporate under normal conditions, and thus the concentrations and fluidity of mixes do not vary over time. During the debinding operation, these paraffins can be easily and completely removed without evidence of decomposition or the formation of carbon residues.

Once a laboratory curiosity, low-pressure forming of high-solids ceramic greenware has become a routine production method. Using two cavity molds, thousands of parts are being produced in a single shift. By extending this operation to 12 cavity molds and reducing fill cycles to less than 15 seconds, production rates of 25,000 small parts per shift can be obtained. Continuous production of six parts per cycle and 15 seconds per cycle has been demonstrated for a period of eight hours. Additionally, large, complex parts approximately 2 ft square by less than a 1à2 in. thick are being molded in less than 20 seconds.

Using high-solids loadings offers numerous advantages. During the debinding stage (the removal of the organic phase), a highly loaded composition will undergo significantly less than 1% linear shrinkage, minimizing the chance of surface crazing, or worse, the forming of internal defects. Other advantages include greatly reduced shrinkage and dimensional control during firing to high density; far less severe firing conditions required to achieve full density; minimized grain growth and reduced furnace degradation, as well as power and maintenance expenses; the use of inexpensive forming equipment; and greatly increased green strength, leading to reduced handling losses during production.

Additionally, high-solids mixes are not restricted in their use to injection molding or forming by filling a mold. Preliminary experiments have shown that they can be adapted to forming by tape extrusion, as well as by simple uniaxial pressing in a die. Essentially, the LPIM technology has been shown to be applicable to a variety of forming techniques besides molding-making it a promising forming method for numerous ceramic manufacturing applications.

## For Further Reading

Corbett, W.J. and Shaffer, P.T.B., "Injection Molding of Advanced Ceramics and Composites," 3rd International SAMPE Conference, San Diego, CA, March 31, 1988.

Corbett, W.J., Lunde, M.C. and Shaffer, P.T.B., "Shaped Bodies Containing Short Inorganic Fibers or Whiskers, and Methods of Making Same," U.S. Patent 5, 108, 964, April 28, 1992.

Corbett, W.J. and Shaffer, P.T.B., "Complex Shaped Preforms for Forming Metal Matrix Composites by Injection Molding," J. Metals, 41 [8], pp. 52-53 (1989).