Development of Ceramic Fibers for Reinforcement in Composite Materials
Refinements of the vertical arc fiberizing apparatus resulted in its ability to fiberize very different refractory glasses having wide ranges of properties. Although the apparatus, was originally designed as a laboratory research tool for the evaluation of many compositions daily, up to one quarter pound of fibers of a single composition could be produced in an 8-hour day. Fibers up to six and a half feet long were produced with the apparatus. Studies were conducted of two methods of fiberizing refractory glasses requiring rapid freezing from the melt. The first method consisted of fiberizing droplets of molten glass passing through an annular nozzle. The second method consisted of reconstructing the annular nozzle in. the shape of a horseshoe to achieve a shorter delay in blasting a molten droplet from the tip of a rod. Both methods were judged feasible for producing fibers of glasses requiring rapid freezing. The first method would be more amenable to volume fiber production. Studies of induction heating for fiber formation did not lead to its designation as a very efficient heating method. Problems. remain to be solved, in the design of a suitable susceptor for a higher heating rate, in protecting the susceptor from oxidation with an inert gas, in contamination of the melt from a refractory crucible, and in the protective radiation shielding of the induction concentrator coil. It is not considered practical to continue studies of this heating method. In the course of this program 151 refractory glass compositions were evaluated for fiber, forming characteristics. Of the various types of materials studied, the following showed promise in producing acceptable refractory fibers: sIlica- spinel (magnesium aluminate), silica- spinel-zirconia, silica-zirconia, silica-zinc spinel, aluminum phosphate glasses, and fluoride glasses. Compositions which did not produce acceptable fibers were high zirconia materials, barium spinels, and calcium aluminates. Improvements in the. testing apparatus for single fiber tensile strength increased the precision. of tests conducted on nine fibers. The highest mean tensile strength, a value of 295,000 pounds per square inch, was obtained with R-141 fibers. Treatment of R-74 fibers with anhydrous Linde A-1100 silane finish improved its mean fiber tensile strength by 25 percent. The lapse of time after fiber formation had no measurable effect on tensile strength. A static heating test conducted with various high melting fibers indicated that Fiberfrax and R-108 underwent no significant changes in bulk volume or resiliency on exposure to 2750 degrees Fahrenheit (1510 degrees Centigrade) in an oxidizing atmosphere. For fiber-resin composition fabrication, ten fiber materials were selected on the bases of high fiber yield, fusion temperature, and type of composition. Fiberfrax, a commercial ceramic fiber, was included for comparison. A new, more effective method of removing pellets from blown fibers was developed. The de-pelletized fibers were treated with a silane finish and felted into ten-inch diameter felts prior to resin impregnation. Composites containing 30 percent by weight of CTL 91-LD phenolic resin were molded under high pressure from the impregnated felts and post-cured to achieve optimum properties. Flexural strength, flexural modules of elasticity, and punch shear strength tests were conducted on the composite specimens. The highest average flexural strength obtained was 19,958 pounds per square inch with the R-74-fiber-resin composite. This compares very favorably with the military specification of 13,000 pounds per square inch flexural strength for randomly oriented fiber reinforced composites. The highest punch shear strength (11,509 pounds per square inch) was obtained with the R-89 fiber-resin composite. The effects of anhydrous fiber finishes on composite strength were not clearly indicated. Plasma arc tests at a heat flux of 550 British Thermal Units per square foot per second on eight composite materials indicated ablation rates generally equivalent to Fiberglas-Micarta No. 259-2. The composite reinforced with R-99 fibers had an average ablation rate of 0.008 inch per second and appears quite promising on the basis of these tests. Preliminary studies for processing fibers into yarn and fabric were conducted with R-99 fibers. The use of certain organic gums aided in 2 fabricating by hand several relatively strong yarns and a crude fabric swatch. This indicated the practicality of developing techniques for processing these fibers into yarn and fabric without significant damage to the fibers.