Determination of biaxial creep strength of T-111 tantalum alloy
Biaxial creep strength of T-111 tantalum alloy tubing in high temperature, high vacuum environment
SEARCH · Search NASA
Search indexed NASA NTRS and DOE OSTI research on propulsion, heat transfer, battery materials and energy systems. Follow report and document links to the original sources.
Quote a phrase for an exact phrase match. Source license links do not imply unrestricted reuse.
Biaxial creep strength of T-111 tantalum alloy tubing in high temperature, high vacuum environment
High temperature, low cycle fatigue tests of tantalum in inert argon atmosphere
Unalloyed tantalum as containment material in mercury Rankine cycle SNAP 8 system boiler for 5 year service life
An investigation was conducted to determine the thermomechanical and thermochemical behavior of a high temperature, oxidation resistant, hafnium-20 percent tantalum alloy. The elastic and shear moduli of this alloy were determined in air up to 1000 C and in vacuum up to 2000 C using a mechanical resonance technique. The internal friction of the alloy was measured up to temperatures greater than 1400 C. Room temperature stress-strain behavior of the oxidized and unoxidized alloy was established. The effect of annealing on the elastic and shear moduli of the extruded rod material was investigated. The martensitic-type phase transformation occurring in the alloy was studied using hot stage metallography and electron microscopy. Static oxidation tests were conducted on the alloy at temperatures from 1000 C to 1700 C with weight gain measurements made as a function of time and temperatures. Surface morphology studies were conducted on the oxide coatings formed at the different temperatures using scanning electron microscopy and X-ray diffraction techniques.
Computer-processed data from low-speed (10 rpm) slipring experiments with two similar (but of opposite polarity) gallium-lubricated tantalum slipring assemblies (hemisphere against disk) carrying 50 amperes dc in vacuum (10 to the minus 9th power torr) showed that the slipring assembly with the anodic hemisphere had significantly lower peak-to-peak values and standard deviations of coefficient-of-friction samples (a measure of smoothness of operation) than the slipring assembly with the cathodic hemisphere. Similar data from an experiment with the same slipring assemblies running currentless showed more random differences in the frictional behavior between the two assemblies.
The post-aging embrittlement of T-111 (tantalum - 8-percent tungsten - 2-percent hafnium) following exposure for up to about 10,000 hours at 1040 C in either vacuum or liquid lithium was investigated for sheet and tubing samples. This thermal aging was shown to greatly increase the sensitivity of T-111 to hydrogen embrittlement during subsequent room temperature specimen processing or testing. The hydrogen embrittlement problem can be avoided by preventing exposure to the T-111 to moisture during post-aging processing or testing. Aging at 1040 C also resulted in formation of HfO2 particles at grain boundaries, which may contribute to the observed embrittlement.
Low cycle fatigue tests have been performed on the tantalum base alloys T-111 and ASTAR 811C with synchronized, independently programmed temperature and strain cycling. The thermal-mechanical cycles applied fell into three basic categories: these were isothermal cycling, in-phase thermal cycling, and out-of-phase thermal cycling. In-phase cycling was defined as tensile deformation associated with high temperature and compressive deformation with low temperature, while out-of-phase thermal cycling was defined as the reverse case. The in-phase thermal cycling had a pronounced detrimental influence on the fatigue life of both alloys, with the life reduction being greater in the solid solution strengthened T-111 alloy than in the carbide strengthened ASTAR 811C alloy. The out-of-phase tests also showed pronounced effects on the fatigue life of both alloys, although not as dramatic.
Tubular transition joints providing a metallurgically bonded connection between tantalum and 316 stainless steel pipe sections were comparatively evaluated for durability under thermal cycling conditions approximating the operation of a SNAP-8 mercury boiler. Both coextruded and vacuum brazed transition joints of 50mm (2 inch) diameter were tested by thermal cycling 100 times between 730 C and 120 C(1350 F and 250 F) in a high vacuum environment. The twelve evaluated transition joints survived the full test sequence without developing leaks, although liquid penetrant bond line indications eventually developed in all specimens. The brazed transition joints exhibited the best dimensional stability and bond line durability.
Five experimental tantalum alloy compositions containing 13-18% W+Re+Hf solid solution solute additions with dispersed phase strengthening achieved by carbon or nitrogen additions were prepared as 1.4 inch diameter ingot processed to 3/8 inch diameter rod and evaluated. Elevated temperature tensile and creep strength increased monotonically with increasing solute content. Room temperature elongation decreased for 20% to less than 2% as the solute content was increased above 16%. Phase identification indicated that the precipitating phase in the carbide containing alloys was Ta2C.
The method of pseudo-resonance cross sections is used to analyze published temperature-dependent neutron transmission and self-indication measurements on tantalum in the unresolved region. In the energy region analyzed, 1825.0 to 2017.0 eV, a direct application of the pseudo-resonance approach using a customary average strength function will not provide effective cross sections which fit the measured cross section behavior. Rather a local value of the strength function is required, and a set of resonances which model the measured behavior of the effective cross sections is derived. This derived set of resonance parameters adequately represents the observed resonance hehavior in this local energy region. Similar analyses for the measurements in other unresolved energy regions are necessary to obtain local resonance parameters for improved reactor calculations. This study suggests that Doppler coefficients calculated by sampling from grand average statistical distributions over the entire unresolved resonance region can be in error, since significant local variations in the statistical distributions are not taken into consideration.
The liquid ammonia and vacuum distillation techniques were found to be satisfactory for removing lithium from 1040 C aged T-111 (tantalum - 8-percent tungsten- 2-percent hafnium). Results of ductility tests and chemical analysis show that these two methods are adequate for removing lithium without embrittlement or contamination of the T-111. Moist air exposure of T-111 with traces of lithium on the surface produced mixed results. Some specimens were ductile; others were brittle. Brittle T-111 had an increased hydrogen content. Water removal of lithium from T-111 caused brittleness and an increased hydrogen concentration.
A series of average transmission and average self-indication ratio measurements were performed in order to investigate the temperature dependence of the resonance self-shielding effect in the unresolved resonance region of depleted uranium and tantalum. The measurements were carried out at 77 K, 295 K and approximately 1000 K with sample thicknesses varying from approximately 0.1 to 1.0 mean free path. The average resonance parameters as well as the temperature dependence were determined by using an analytical model which directly integrates over the resonance parameter distribution functions.
A differential-thermal expansion method was developed to line T-111 (tantalum - 8 percent tungsten - 2 percent hafnium) tubing with a tungsten diffusion barrier as part of a fuel element fabrication study for a space power nuclear reactor concept. This method uses a steel mandrel, which has a larger thermal expansion than T-111, to force the tungsten against the inside of the T-111 tube. Variables investigated include lining temperature, initial assembly gas size, and tube length. Linear integrity increased with increasing lining temperature and decreasing gap size. The method should have more general applicability where cylinders must be lined with a thin layer of a second material.
The effects were studied of low-pressure air on contamination and corrosion in the tantalum alloy T-111/lithium system at 980 and 1260 C. Capsules of T-111 containing lithium were exposed to six vacuum levels between 1 x 10 to the 8th power and 0.0003 torr by controlled air leakage into a vacuum system. Capsules exposed at 980 C and 0.0002 torr failed from intragranular oxidation. The remainder of the capsules completed the 96-hour tests. The depth of oxygen contamination was greater at 980 C than at 1260 C. Tests made at 0.0001 and 0.00001 torr levels caused large increases in the oxygen content of the T-111. Tests at 0.000001 torr or less produced no significant contamination. No lithium corrosion of the T-111 was observed under any of the conditions.
High vacuum creep tests on the tantalum T-222 alloy at 0.42 to 0.52 T sub m show that the major portion of the creep curves, up to at least 1 percent strain, can be best described by an increasing creep rate, with strain varying linearly with time. Correlation and extrapolation of the creep curves on the basis of increasing creep rates results in more accurate engineering design data than would use of approximated linear rates. Based on increasing creep rates, the stress for 1 percent strain in 10,000 hours for T-222 is about four times greater than for the Ta-10W alloy. Increasing the grain size results in increased creep strength. Thermal aging prior to testing caused precipitation of the hexagonal close packed (Hf,Ta) sub 2 C, which initially increased creep strength. However, this dimetal carbide was converted during creep testing to face-centered cubic (Hf,Ta)C.
Results are reported of a research program to develop a reliable high-performance, fused slurry silicide protective coating for a tantalum-10 tungsten alloy for use at 1427 to 1538 C at 0.1 to 10 torr air pressure under cyclic temperature conditions. A review of silicide coating performance under these conditions indicated that the primary wear-out mode is associated with widening of hairline fissures in the coating. Consideration has been given to modifying the oxidation products that form on the coating surface to provide a seal for these fissures and to minimize their widening. On the basis of an analysis of the phase relationships between silica and various other oxides, a coating having the slurry composition 2.5Mn-33Ti-64.5Si was developed that is effective in the pressure range from 1 to 10 torr.
Two tantalum base alloys, Ta-12W-1.0Re-0.7Hf-0.025C(ASTAR-1211C) and Ta-15W-1.0Re-0.7Hf-0.025C(ASTAR-1511C), were cast as 12.5 cm (5 inch) diameter ingots and processed to swaged rod, sheet, forged plate, and tubing. Swaged rod was evaluated with respect to low temperature ductility, elevated temperature tensile properties, and elevated temperature creep behavior. A standard swaging process and final annealing schedule were determined. Elevated temperature tensile properties, low temperature impact properties, low temperature DBTT behavior, and extended elevated temperature creep properties were determined. A process for producing ASTAR-1211C and ASTAR-1511C sheet were developed. The DBTT properties of GTA and EB weld sheet given post-weld anneal and thermal aging treatments were determined using bend and tensile specimens. High and low temperature mechanical properties of forging ASTAR-1211C and ASTAR-1511C plate were determined as well as elevated temperature creep properties. Attempts to produce ASTAR-1211C tubing were partially successful while attempts to make ASTAR-1511C tubing were completely unsuccessful.
The causes of aging embrittlement in T-111 (Ta-8W-2Hf) and the effect of alloy modification were investigated. Results show that T-111 possesses a critical combination of tungsten and hafnium that leads to loss in ductility at -196 C after aging near 1040 C. It was found that this occurs because tungsten enhances hafnium segregation to grain boundaries, which also leads to increased susceptibility to hydrogen embrittlement. Aging embrittlement was not observed in tantalum alloys with reduced tungsten or hafnium contents; most of the alloys studied have lower strengths than T-111 and exhibit susceptibility to hydrogen embrittlement.