Read Microsoft Word - New High Temp Oxide Fibers 7-11-01.doc text version

New High Temperature Oxide Fibers

D. M. Wilson

3M Co., St. Paul, MN


During the last decade, oxide fiber-reinforced oxide composites have been the subject of increasing interest. This renaissance is based in part on the availability of two new oxide fibers, NEXTELTM Ceramic Oxide Fiber 720 and NEXTELTM Ceramic Oxide Fiber 650, which have good strength, microstructural stability and long-term creep resistance at temperatures as high as 1200°C. Superior high temperature structural properties result from the presence of crystalline, creep-resistant phases such as Y3+-doped Al2O3 and mullite. The physical properties, high temperature strength, strength retention and creep resistance of these composite Nextel fibers will be reviewed Further improvements in high temperature properties of polycrystalline oxide fibers are possible. High temperature BSR (bending stress rupture) results on experimental YAG fibers have demonstrated creep performance superior to Nextel 720 fiber. However, the successful synthesis of high strength, fine-grained YAG fibers has been elusive. In this work, two -phase YAG-alumina fibers were synthesized with uniform, fine grain size and high strength. The tensile creep of YAG-alumina fibers was measured and compared with commercial oxide fibers and single-phase YAG fibers.


In comparison to SiC fibers, oxide fibers have been traditionally considered to have insufficient strength and creep resistance for the reinforcement of high temperature ceramic composites. However, in the mid 90's, two new fibers, Nextel 650 fiber and Nextel 720 fiber, were developed that have uniquely high tensile strength and creep resistance. The creep resistance of Nextel 720 fiber1,2,3 allows the fabrication of oxidation-stable ceramic composites with useful load-bearing capability above 1100Û84,5,6. A new rare-earth doped alumina fiber, Nextel 650 fiber, has recently been developed that has 100 times lower creep rate than NEXTELTM Ceramic Oxide Fiber 610 7. One unique and critical difference between these fibers and other commercial oxide fibers is that they are fully crystalline. Most commercial fibers contain silica or other non-crystalline phases, including the older NextelTM fibers such as NEXTELTM Ceramic Oxide Fibers 312, 440 and 550. Amorphous phases become viscous at high temperatures, so are detrimental to creep performance. Equally critical for fibers designed for composite reinforcement is the reactivity of amorphous phases compared with fully crystalline fibers. Crystalline fibers containing high amounts of -Al2O3 that are free of glassy phases are very chemically stable. High chemical stability leads to good environmental stability in corrosive atmospheres, and less interaction with a variety of ceramic matrices. Several successful high temperature composites utilize porous oxide matrices4,6. Porous oxide matrices promote crack deflection at fiber-matrix interfaces, leading to fiber debonding and pullout, thereby providing high composite toughness. This approach to composite design requires highly stable and non-reactive fibers to prevent strong matrix-fiber

interactions. This type of composite, which does not require fiber coatings, is finding increasing application in part because of the significant cost advantages relative to other systems. Nextel 720 fiber was developed for load-bearing applications at temperatures in excess of 1000Û8ÃÃUurÃrvÃuvtuÃrrhrÃprrÃrshprÃsÃIrryÃ&!ÃsvirÃryÃsÃhà high content of mullite, which has much better creep resistance than alumina. Additionally, Nextel 720 fiber consists of 0.5 µm globular grains of mullite; this is five times larger than grains in Nextel 610 fiber. The microstructure of Nextel 720 fiber has been described in more detail elsewhere7,8. In fine-grained oxides, creep rate is inversely proportional to grain size, so the larger grain size of Nextel 720 fiber reduces creep. Lastly, the presence of acicular and globular grains of mullite and alumina reduces grain boundary sliding. However, these microstructural factors that improve creep represent a tradeoff with respect to strength. High strength fibers should preferably have a fine and uniform grain size and high contents of alumina, which has higher fracture toughness than mullite. Nextel 650 fiber was recently developed for high temperature composite applications in which the presence of mullite and other silicon-containing phases is not desirable and where higher composite strength is required. Additional information on the microstructure and properties of Nextel 650 fibers was recently published9. Nextel 650 fibers were developed with the goal of producing a fiber with improved strength and chemical stability relative to Nextel 720 fibers with superior creep performance relative to Nextel 610 fibers. High alumina fibers are more corrosion resistant in many environments, such as those containing alkalis, than fibers containing silica. Also, Nextel 720 fiber has a strength of only 2.1 GPa, significantly less than Nextel 610 fiber, as a result of its larger grain size. French10 and others11 have reported that doping alumina with rare earth oxides such as Y2O3 reduces creep in alumina by 1 to 2 orders of magnitude. The use of Y2O3 to reduce diffusivity in Al2O3 oxidation scales formed on metal alloys is also well known12. The reduction in creep is attributed to reduced grain boundary diffusivity caused by segregation of Y3+ to grain boundaries13. Since creep in fine-grained alumina is controlled by diffusion-related phenomena on grain boundaries, rare earth doping reduces the creep rate in alumina. Yttrium aluminum garnet (YAG) has been identified as a promising candidate fiber for high temperature composites. Small diameter, polycrystalline YAG fibers have been developed in several laboratories, including at 3M, that have creep performance 50-100°C superior to Nextel 720 fibers, as measured using the BSR technique14,15,16,17,18. These results have fueled the desire for commercial YAG fibers in reinforcement applications in the 1100°C-1300°C range. However, the strength of all of these fibers was below 1 GPa; this was probably caused not only by processing defects but the large grain size of these fibers (1-3 µm). The reason for the superior creep resistance of YAG fibers compared with Nextel 720 fiber is not necessarily obvious. The creep resistance of single crystal YAG is the highest of any metal oxide19. However, the creep rate of fine-grained, polycrystalline YAG is only one order of magnitude superior to Al2O310,20. Since the creep rate of Y3+-doped Al2O3 is 1-2 orders of magnitude lower than pure Al2O3, French10 concluded that the creep rate of YAG is actually inferior to Y3+-doped Al2O3.

Experimental Procedures

Single filament strength testing was performed by directly gripping fibers with rubberfaced clamp grips with 25 x 25 mm grip faces. A 25 mm gauge length and a fiber loading rate

of 0.02/min was used. Ten filaments were broken at each condition. Filament diameters were measured using a MEASURE-RITETM image analysis system (Model M25-6002 Dolan-Jenner Industries, Lawrence, MA), attached to a light microscope at 1000X magnification. In this system, fibers were measured end-on relative to a round template. This method has a resolution of 0.3 µm and has been determined to be accurate to within 0.3 µm relative to SEM measurements. For single filament strength testing, filament diameter was not measured on all individual filaments; rather an average value was used in the calculations. This has been determined to be accurate for Nextel 610 fibers, primarily because the coefficient of variability in diameter within a sample is less than 5%18. For high temperature strength testing, diameters were calculated from denier (weight per length of the tow) and number of fibers, assumed to be 410 for 1500 denier and 760 for 3000 denier. Calculated diameters were 11.52, 11.22 and 12.34 µm for 1500 denier Nextel 610, 650 and 720 fibers, respectively. For testing after high temperature heat treatment, 42-54 filament diameters were tested for each type of fiber after several of the heat treatments. Measured average fiber diameters were 11.52 (±0.22) µm, 11.46 (±0.35) µm and 12.22 (±0.23) µm for Nextel 610, 650 and 720 fibers, respectively. This is within 0.2 µm of calculated values (3000 denier Nextel 650 fibers used for the heat treatment study had a calculated diameter of 11.67 µm). High temperature fiber strength testing was performed vertically using a slotted furnace with a hot zone 20 mm long. Individual fibers were loaded into the test frame, the hot furnace was slid around the fiber and equilibrated for 90 seconds before beginning the test. The overall gauge length was 280 mm and the strain rate was 0.68 mm/minute. 100 hour heat treatments were carried out in a box furnace with MoSi2 heating elements. The fibers were placed uncovered on Nextel 610 fabric, which functioned as a getter for potential atmospheric contaminants. Temperature was confirmed with a separate thermocouple placed next to the fibers. Creep testing was performed using a single filament, dead load testing system to provide an accurate and consistent load1. Sample elongation was measured with a Zygo® laser extensometer (Model 1101, Zygo Corp, Middlefield, CT). The creep system used a resistance-heated, three zone alumina tube furnace with Nichrome heating elements. Hot zone temperature was accurate to within 3°C of the setpoint over the entire 106 mm gauge length.


Table 1 summarizes the properties of Nextel 610 fiber, Nextel 650 fiber and Nextel 720 fiber. Nextel 610 fiber is >99% -Al2O3. Nextel 610 has properties expected of an alumina fiber; the elastic modulus and density are slightly below theoretical (400 GPa, 3.98 g/cm 3), reflecting a small amount of porosity. Thermal expansion is very close to measured values for monolithic alumina. Nextel 650 fiber has the composition Al2O3 + 10% ZrO2 + 1% Y2O3. The microstructure of Nextel 650 fiber consists primarily of 0.1 µm -Al2O3 grains; ZrO2 is present as 5-30 nm grains on both grain boundaries and within alumina grains. The key additive in Nextel 650 fibers is Y2O3, which is added to reduce creep rate. ZrO2 additions are used to reduce grain growth, which is accelerated by Y2O3 doping. Zirconia is stabilized by Y2O3 in the cubic phase. The addition of ZrO2 had only a minor effect on creep rate for either rare earth doped or undoped alumina fibers. Density and thermal expansion are higher than Nextel 610 fiber due to the presence of 10% ZrO2. Elastic modulus was measured to be slightly lower than Nextel 610 fibers, as expected since ZrO2 has lower modulus than Al2O3.

Nextel 720 fiber contains 15% SiO2 in the form of mullite; volume calculations indicate that mullite comprises 55-60% of the fiber volume. The high content of mullite lowers density and thermal expansion by 13% and 30%, respectively. These can be significant advantages for aerospace and thermally loaded applications.

Table 1. Property Chemical Composition Properties of NextelTM Ceramic Oxide Fibers 610, 650, and 720 Units wt. % Nextel 610 >99 Al2O3 Nextel 650 89 Al2O3 10 ZrO2 1 Y2O3 -Al2O3 + cubic ZrO2 2.8 358 4.1 8.0 1080 Û& Nextel 720 85 Al2O3 15 SiO2 -Al2O3 + mullite 2.1 260 3.4 6.0 1150Û&

Crystal Phases Tensile Strength Tensile Modulus Density Thermal Expansion (25-1000°C) Maximum Use Temperature (1% strain/69 MPa/1000hr) GPa GPa g/cc ppm/ °C °C


3.3 373 3.9 7.9 1000Û&

High Temperature Fiber Properties

Figure 1 compares high temperature single filament strength of Nextel 610, Nextel 650, and Nextel 720 fibers. To highlight the differences between fibers, the data is presented as a percentage of room temperature strength. The room temperature strength of these samples of Nextel 610, 650 and 720 fibers at 25mm gauge length was 3.63, 2.75 and 2.49 GPa, respectively. Nextel 610 fibers retained 70% of room temperature strength up to 1000°C, Nextel 650 fibers up to 1200°C and Nextel 720 fibers up to 1300°C. The single filament data suggests an improvement in temperature capability for Nextel 650 fibers relative to Nextel 610 fibers of 100°C; Nextel 720 fiber has a temperature capability a further 100°C higher than Nextel 650 fiber. The relative high temperature strengths of these fibers correlated well with relative creep rates (Figure 5). Figure 2 compares the room temperature tensile strength of each type of fiber after heat treatment for 100 hours at temperatures between 1000°C and 1400°C. Fiber strength was near 2 GPa or above for all three fibers after heat treatment at 1300°C. The strength of Nextel 720 fibers declined the least, retaining 85% of its strength after 1300°C/100 hour heat treatment. Nextel 610 and 650 fiber both retained 80% of their initial strength after 1200°C hour heat treatment. At 1300°C and especially 1400°C, all fibers were welded together and more difficult to separate; the Nextel 650 sample at 1400°C was severely welded and could not be


% Strength Retention



40 Nextel 610 20 Nextel 650 Nextel 720 strain rate = 0.68 mm/min 0 0 200 400 600 800 1000 1200 1400

Test Temperature (°C)

Figure 1. Relative tensile strength retention, normalized to room temperature, of single filaments of Nextel 610, 650 and 720 fibers at elevated temperature.

3.5 3.0 2.5 2.0 1.5 1.0 0.5 0.0 0 200 400 600 800 Nextel 610 Nextel 650 Nextel 720

100 hr. heat treatment

Tensile Strength, GPa




Heat Treatment Temperature, C

Comparison of relative strength retention of single filaments of Nextel 610, 650 and Figure 2. 720 fibers after heat treatment at elevated temperature. Nextel 610 and 650 fibers retained 80% of strength up to 1200°C and Nextel 720 fibers up to 1300°C.

tested. Welding has been shown to be a primary fracture origin in Nextel fibers21. growth has not been measured precisely but became significant at 1300°C.


YAG-Alumina Fiber Synthesis

Both YAG and two-phase Al2O3-YAG fibers with up to 50% YAG (yttrium aluminum garnet, Y3Al5O12) content have been synthesized in our laboratory. The synthesis of relatively strong, crystalline 100% YAG fibers with BSR creep 50°C superior to Nextel 720 fibers has been previously reported19. However, because of difficulties encountered both in grain growth during crystallization, and in eliminating porosity in pure YAG fibers, recent effort has focused on producing two-phase fibers containing both YAG and alumina. The synthesis of these fibers, with lower YAG content, was more straightforward. Two compositions were investigated: 1) fiber containing 50% Al2O3 and 50% YAG by weight, designated 50A:50YAG, and 2) fiber containing 75% Al2O3 and 25% YAG by weight, designated 75A:25YAG. Both types of fibers were relatively strong and easily handled. Only limited strength testing was performed. The single filament strength of 50A:50YAG fibers was measured to be 700 ± 201 MPa at a fiber diameter of 16.2 µm. While the strength is much less than commercial Nextel fibers, the flexibility and uniformity of these fibers suggests that strengths >2GPa should be possible. Early versions of both Nextel 610 and 650 fibers had strengths below 1 GPa. The fracture surface of both YAG-Al2O3 fibers showed faceted grains with no trace of amorphous material or visible porosity (Figures 3a and 3b). Backscattered electron SEM indicated that the YAG and Al2O3 phases were very uniformly distributed; each grain of YAG is nearly surrounded by Al2O3 and vice versa. The 50A:50YAG fibers had a grain size of 0.2 µm. The 75A:25YAG fibers had a larger and bimodal grain size distribution, with many grains 0.2 µm surrounding larger 0.4 µm grains. The improved densification relative to pure YAG fibers may be related to the delay in forming YAG; the transient YAlO3 phase likely has superior sinterability compared to YAG. Also, the conversion of YAlO3 (an intermediate in the crystallization of both YAG-alumina fibers) to YAG + -Al2O3 involves an 11% volume expansion22. This expansion would assist in the elimination of porosity. Both the 75A:25YAG and 50A:50YAG fibers were very stable with respect to grain growth. Figure 4 shows a high resolution SEM micrograph of a 50A:50YAG fiber heated at 1300°C for 100 hours. The grain size of this fiber is 0.2 µm, unchanged from the as-prepared fiber. The thermal stability of the microstructure is ~100°C superior to either Nextel 610 or 720 fibers, and suggests potential utility in long-term applications as high as 1300°C. After heat treatment for 100 hours at 1200°C, the strength of 50A:50YAG fibers did not decline; single filament strength was 680 ± 280 MPa.

High Temperature Creep Performance of YAG-Alumina Fibers

The stress dependence of the steady state creep rate of polycrystalline oxides such as alumina and YAG can be described by the following equation: d/dt = A (1/d)p n (1) where d/dt is the steady-state creep rate, A is a constant, d is the grain size, p is the grain size exponent, is the applied stress, and n is the stress exponent. The deformation of



Figure 3. SEM micrograph of a) 75A:25YAG and b) 50A:50YAG composition fibers, heattreated at 1400°C. Microstructure was uniform and fine-grained for both fibers.

Figure 4. SEM micrograph of 50A:50YAG composition fibers, heat-treated at 1300°C/100 hr. Grain size was uniform and near 0.2 µm.

polycrystalline oxides with a grain size less than a few micrometers is controlled by grain boundary mechanisms due to the short diffusion distances and large grain boundary areas. Creep mechanisms for submicrometer oxides include grain boundary (Coble) creep, interfacecontrolled creep, i.e., defect creation or annihilation rates at grain boundaries, grain boundary sliding, grain switching, and grain rotation23,24. Creep models and empirical results on alumina and zirconia25,26 indicate that both the stress exponent n and grain size exponent, p, are usually 2-3. As shown in Figure 5, Nextel fibers have a stress exponent of 2-3, consistent with this model. The grain size exponent of reinforcement fibers is less well understood. In one study, Nextel 610 was reported to have a grain size exponent of greater than three27. The creep rate of two-phase microstructures, i.e., materials with two mutually insoluble phases such as YAG and alumina, is intermediate between the two single components. The creep of two-phase microstructures has been analyzed by French10 to be most similar to an "isostrain" mixing model, where each of the two phases experiences equal strain (rather than stress). The overall creep rate was less than would be expected from linear mixing; instead, the creep rate of two-phase mixtures was more similar to the creep rate of the more creepresistant phase.

High Temperature Creep Results

Figure 5 compares the tensile creep rates of commercial Nextel fibers with experimental 75A:25YAG and 50A:50YAG fibers at 1100°C. The creep rate of the commercial fibers decreased in the order Nextel 610 fiber > Nextel 650 fiber > Nextel 720 fiber. The difference in creep rate was a factor of 10-100 times between each type of fiber. At a stress of 100 MPa, the strain rate of Nextel 610 fiber was 1 x 10-6/sec, Nextel 650 fiber, 1 x 10-8/sec, and Nextel 720 fiber, 1 x 10-10/sec. This magnitude of difference increases the temperature capability of the fiber by ~100Û8à à Vvtà hà pvrvà sà %(à HQhà rà hqà hà Èà hvà hvà và à hours, Nextel 610 fibers have a temperature capability of ~1000Û8ÃÃVvtÃurÃhrÃpvrvà the temperature capability for Nextel 650 and Nextel 720 fiber is 1080Û8à hqà 150Û8à respectively. The stress exponent for both Nextel 610 and 720 fibers was very close to 3. The stress dependence of creep for Nextel 650 fiber was 1.8, less than the other fibers. The reason for this is not known; French10 reported no change in stress dependence with Y3+ doping. At 100 MPa, 75A-25YAG fibers had a creep rate of 5 x 10-9/sec, slightly superior to Nextel 650 fibers, but significantly inferior to Nextel 720 fibers. Despite its higher YAG content, the creep of 50A:50YAG fibers was 4 x 10-8/sec, ~10X more rapid than 75A-25YAG fibers. Multiple batches of Al2O3-YAG fibers were used for tensile creep testing, so these results were reproducible. The stress exponent was similar to other Nextel fibers, suggesting that creep mechanisms were similar.

0.00001 Test Temperature: 1100C 0.000001 Nextel 610

Strain Rate, 1/sec

50% Al2O3-50% YAG

1E-07 Nextel 650 1E-08

75% Al2O3-25% YAG


Nextel 720

1E-10 10 100

Load, MPa


Figure 5. Comparative creep rate of Nextel fibers at 1100Û8à hà hà spvà sà rà à Uurà prrà rate of 75A:25YAG and 50A:50YAG fibers was similar to Nextel 650 fiber. The creep rate of Nextel 720 fiber was 1-2 orders of magnitude lower.

Why was the creep rate of two-phase Al2O3-YAG fibers inferior to previously reported data on YAG fibers? Consider the similarities between Nextel 650 fibers and 75A:25YAG fibers. Both fibers consist primarily of submicrometer -Al2O3 saturated with Y3+ (the

solubility limit of Y3+ is <.3%), along with minor second phases. The second phase in Nextel 650 fibers is ~7 vol% yttria-stabilized ZrO2, in 75A:25YAG, ~ 23 vol%YAG. If, as discussed above, Y3+-doped Al2O3 is more creep-resistant that either YAG or ZrO2, then the effect on creep of the second phases would be relatively minor. This explanation is consistent with the measured creep performance. The better creep rate of the 75A-25YAG fibers relative to Nextel 650 fibers probably results from its larger grain size. The grain size of 75A-25YAG fibers was 0.2-0.4 µm compared with 0.1µm for Nextel 650 fibers. As indicated by Equation 1, larger grain size will reduce creep rate by (1/d) p. If the grain size exponent, p, is 2-3, then doubling the grain size would reduce creep rate by a factor of 4-8, again consistent with the creep results. The higher creep rate of the 50A:50YAG fibers compared with the 75A:25YAG fibers would then follow from the smaller grain size (0.2 µm) of these fibers and their higher YAG component compared with Nextel 650 fibers. This strong dependence on creep rate with grain size would also be consistent with the reported creep performance of single phase YAG fibers. Although tensile creep rates for YAG fibers was not measured (probably due in part to low fiber strength); 50°-100°C better creep rate than Nextel 720 fibers would imply that tensile creep rate would be less than 1x 1011 /sec at 100 MPa. With grain sizes near 1 µm, the single phase YAG fibers would have a creep rate a factor of 103, or 1000, less than Nextel 650 fibers, which is in the range of the measured improvement in creep performance. This data suggests that fiber grain size must be strongly considered in designing oxide fibers with improved creep resistance. Fiber microstructure must be carefully designed so that high strength is maintained despite the large grain size.


Two commercial oxide fibers, Nextel 650 and 720 fibers, have been developed for high temperature composite reinforcement. These composite fibers all have the chemical stability and strength provided by -Al2O3, and contain different modifying additives to improve high temperature properties. Nextel 720 contains a majority of large, 0.5 µm, creep-resistant mullite grains; as a result, it is suitable for load bearing applications at temperatures as high as 1200Û8ÃÃIrryÃ%$ÃsvirÃuhÃrvÃrtuÃphrqÃvuÃIrryÃ&!ÃsvirÃhqÃuhÃtqà load-bearing capability up to 1100Û8à à Bqà prrà rvance is provided by Y2O3 doping; good strength and chemical stability is provided by a fine, uniform, high alumina structure. Nextel 650 and 720 fibers maintained 80% of their room temperature strength after long-term heat treatment at 1200°C and 1300°C, respectively. Experimental two-phase YAG-Al2O3 fibers with good strength and handleability were synthesized. Tensile creep measurements found that creep rate was similar to Nextel 650 fibers. The relatively poor creep resistance compared with single phase YAG fibers was attributed to their fine-grained microstructure. Fibers with 25% YAG content had lower creep rate than fibers with 50% YAG content, consistent with literature reports that Y3+-doped alumina has superior creep performance compared with YAG.


The assistance of Larry Visser, Harrison Malinoff, Joe Schneider, Sandy Deppe and Ruth Ann Williams in single filament testing is greatly appreciated. The efforts of Jim Reimer and

Harrison Malinoff for creep measurements and Steve Pittman for SEM microscopy are also acknowledged. Garnet fiber work was supported by Rich Goettler and Rich Wagner of McDermott Technology and DOE under contract DE-FC02-92CE40945. Such support does not constitute an endorsement by DOE of the views expressed in the article.


1. Wilson, D. M., Lieder, S. L. & Lueneburg, D. C., Cer. Eng. Sci. Proc.1995, 16, 10051014. 2. Wilson, D. M., Lieder, S. L. & Lueneburg, D. C., Materials Research Society Symposium Proceedings, Vol. 350, Intermetallic Matrix Composites III, (Mat. Res. Soc., Pittsburg 1994) p. 89-98. 3. Goering, J. & Schneider, H., Cer. Eng. Sci. Proc. 1997, 18, 95-102. 4. Kramb, V. A., John, R. & Zawada, L. P., J. Am. Ceram. Soc. 1999, 82, 3087-96. 5. Levi, C. G., Yang, J. Y. Dalgleish, B. J. Zok, F. W. & Evans, A. G., J. Am. Ceram. Soc., 1998, 81, 2077-86. 6. Heathcote, J. A., Gong, X-Y. Ramamurty, U. & Zok, F. W., J. Am. Ceram. Soc. 1999, 82, 2721-30. 7. Deliglise, F., Berger, M. H., Jeulin, D., Bunsell, A. R., J. Europ. Ceram. Soc. 2001, 21, 569-580. 8. Hay, R. S., Boakye, E. E., Petry, M. D., Berta, Y. Von Lehnden, K., & Welch, J., Cer. Eng. Sci. Proc. 1999, 20, 153-163. 9. Wilson, D. M. & Visser, L. R., Cer. Eng. Sci. Proc. 2000, 21, 363-73. 10. French, J. Zhao, M. Harmer, H. Chan & G. Miller, J. Am. Ceram. Soc. 1994, 77, 2857-65. 11. Sato, E. & Carry, C., J. Am. Ceram. Soc. 1996, 79, 2156-60. 12. Pint, B. A., Garrat-Reed, A. J. & Hobbs, L. W., J. Am. Ceram. Soc. 1998, 81, 305-14. 13. Bruley, J. , Cho, J. , Chan, H. M. , Harmer, M. P. & Rickman, J. M., J. Am. Ceram. Soc. 1999, 82, 2865-70. 14. Eur. Patent Appl. 0600588A1, Sept. 30, 1993. 15. King, B. and Halloran, J., J. Am. Ceram. Soc. 1995, 8, 2141-48. 16. Popovich, D. and Lombardi, J., Cer. Eng. Sci. Proc. 1997, 3, 65-72. 17. Pullar, R. , Taylor, M., and Bhattacharya, A. J. Europ. Ceram. Soc. 1999, 19 1747-1758. 18. Wilson, D. M. and Visser, L. R., Composites: Part A 2001, 32, 1143-1153. 19. Corman, G., Ceram. Eng. Sci. Proc. 1991, 12, 1745-66. 20. Parthasarathy, T., Mah, T. and Keller, K., Ceram. Eng. Sci. Proc 1991, 12, 1767-73. 21. Wilson, D. M., J. Mat. Sci. 1997, 32, 2535-42. 22. Hay, R.. S., J. Am. Ceram. Soc. 1994, 77, 1473-85. 23. Sherby, O. D. and Wadsworth, J., Mat. Res. Soc. Symp. Proc. Vol. 196 1990 Materials Research Society, p. 3-14. 24. Gordon, R., Adv. in Ceramics, vol. 10, American Ceramic Society, 1984, p.418-437. 25. Wakai, F., Br. Ceram. Trans. J. 1989, 88, 205-08. 26. Kellett, B., Carry, C., and Mocellin, A., J. Am. Ceram. Soc.1990, 73, 1922-27. 27. Hammond, V., Wawner, F., and Elszey, D., Proc. 12th Int. Conf. On Comp. Materials (ICCM-12) 1999, Paper #606.


Microsoft Word - New High Temp Oxide Fibers 7-11-01.doc

10 pages

Find more like this

Report File (DMCA)

Our content is added by our users. We aim to remove reported files within 1 working day. Please use this link to notify us:

Report this file as copyright or inappropriate


You might also be interested in

Microsoft Word - New High Temp Oxide Fibers 7-11-01.doc