This work outlines the development of a new natural gas/oxygen (NG/O2) fueled combustion rig located at the NASA Glenn Research Center for high-temperature environmental durability studies of advanced materials and components at atmospheric pressure. The NG/O2 burner rig can simulate the high-temperature, high-heat flux, and high-velocity thermal environments encountered in gas turbine engines. It also provides the capability to study environmental effects such as water vapor and other foreign contaminants relevant to these applications. The rig is anticipated to bridge the gap between other laboratory methods such as furnaces, jet-fueled burner rigs, high-heat flux lasers, and more expensive engine rig testing. The NG/O2 rig is expected to have maximum sample temperature capabilities over 3,000°F (1,649°C) and result in higher water vapor content compared to our Mach 0.3 to Mach 1.0 jet-A burner rigs which is important for characterizing current and next-generation environmental barrier coatings. This paper will provide an overview of the development of the NG/O2 burner rig, initial characterization, and current research and development efforts on environmental barrier coated ceramic matrix composites.

A FIB/STEM interfacial study was performed on a TBC/Ti2AlC MAX phase system, oxidized in an aggressive burner rig test (Mach 0.3 at 1300 °C for 500 h). The 7YSZ TBC, α-Al2O3 TGO, and MAXthal 211TM Ti2AlC base were variously characterized by TEM/STEM, EDS, SADP, and HRTEM. The YSZ was a mix of “clean” featureless and “faulted” high contrast grains. The latter exhibited ferro-elastic domains of high Y content tetragonal t″ variants. No martensite was observed. The TGO was essentially a duplex α-Al2O3 structure of inner columnar plus outer equiaxed grains. It maintained a perfectly intact, clean interface with the Ti2AlC substrate. The Ti2AlC substrate exhibited no interfacial Al-depletion zone but, rather, numerous faults along the basal plane of the hexagonal structure. These are believed to offer a means of depleting Al by forming crystallographic, low-Al planar defects, proposed as Ti2.5AlC1.5. These characterizations support and augment prior optical, SEM, and XRD findings that demonstrated remarkable durability for the YSZ/Ti2AlC MAX phase system in aggressive burner tests.

The process scale-up of fully oxide-based environmental barrier coatings (EBCs) on sintered SiC and chemical vapor infiltration (CVI) SiC/SiC ceramic matrix composite (CMC) sub-components was investigated using various slurry manufacturing processes (dip, spray, spin–dip). The performance of EBC-coated sub-components (SiC heating element, SiC/SiC ceramic matrix mini-composite, SiC airfoil, SiC/SiC CMC airfoil) was evaluated in steam oxidation and combustion rigs. Steam oxidation was conducted at 1427 °C in 90 vol.% H2O (g) + 10 vol.% O2 (g) with a 1 h hold at 1427 °C per cycle (1 h hot and 20 min cooling). For high-pressure combustion rig testing, the EBC surface temperature ranged between 1354 °C and 1538 °C with the temperature gradient through CMC + EBC ranging between 100 °C and 150 °C. Dip and spin–dip are non-line-of-sight processes, whereas spray is a line-of-sight process. The three processes, collectively, demonstrated the capability to manufacture slurry EBCs on sub-components with various shapes and sizes. There was no discernable disparity in the EBC steam oxidation performance between the coupons and sub-components in this study and coupons in a previous study. The dependence of steam oxidation rates on the substrate chemistry reported previously was confirmed. The steam oxidation rate of EBC-coated sintered SiC, compared with EBC-coated CVI CMC, was ~2 times and ~1.5 times higher after 100 h and 500 h, respectively, due to the boron sintering aid in sintered SiC. An EBC-coated CMC airfoil after 150 15-h-long cycles in a high-pressure combustion rig test showed only limited EBC spallation along the leading edge and more substantial spallation along the trailing edge, demonstrating the feasibility of an oxide-based bond coat to meet the extreme temperature requirements of next-generation EBCs.

A thermal barrier coating (TBC) system survived 500 hours in aggressive, 1300 degrees C burner rig testing. The yttria-stabilized zirconia (7YSZ) TBC was plasma sprayed on an oxidation-resistant Ti2AlC-type MAX phase and tested in a jet fuel burner at 100 m/s, using 5 hours cycles. No coating spallation or surface recession was observed; Al2O3-scale growth produced a slight 2.4 mg/cm(2) mass gain. The coating surface exhibited craze-cracked colonies of [111](flourite) textured columns, with no visible moisture attack. The 20 mu m alumina scale remained intact under the YSZ face, about twice that producing failure for TBC/superalloy systems. TiO2 nodules, initially formed on the uncoated backside, were removed, and Al2O3 was etched through volatile hydroxides formed in water vapor (similar to 10%). Overall, the test indicated exceptional stability of the YSZ/Al2O3/Ti2AlC system under turbine conditions due in large part to close thermal expansion matching.

A thermal barrier coating system survived burner rig testing at 1300 °C for 500 h. A 160 µm thick yttria stabilized zirconia (YSZ) coating was applied to a Ti2AlC MAX phase bar sample by plasma spray physical vapor deposition (PS-PVD) and tested face-on in an atmospheric Mach 0.3 jet fuel burner, using 5-h thermal cycles. No thermal barrier coating (TBC) spallation or recession was observed, only a 2.4 mg/cm2 mass gain. The modest weight gain precluded severe volatility losses under high velocity burner conditions. The coating surface exhibited colonies of (111)flourite fiber-textured columns separated by craze patterns, with no visible moisture attack. The metastable tetragonal t' YSZ phase was obtained initially, transitioning to equilibrium teq and cubic YSZ, but with little detrimental monoclinic. The thickness of the alumina TGO was ~21 to 23 μm under the heated YSZ face and ~13 to 15 μm on the uncoated, cooler backside. The backside exhibited removal of initial transient TiO2 nodules and partial etching of the underlying Al2O3 scale by volatile hydroxides formed in high temperature, high velocity water vapor. Aerodynamic forces produced some bending of the cantilevered sample via creep. The test indicated exceptional stability of YSZ coatings on Ti2AlC under turbine conditions, with thermal expansion matching playing a key role. The purpose of this study was to demonstrate long term durability of YSZ/MAX phase system in aggressive high temperature burner rig testing. MAX phases have been keenly studied because of their unique crystal structure and intriguing properties (Refs. 1 and 2). Having Mn+1(Al,Si)(C,N)n general composition, they are defined as ceramics, but possess unusual desirable attributes such as high conductivity, thermal shock resistance, easy machinability, and deformation tolerance. The mechanical properties derive from weak M-(Al,Si) bonding in the basal plane that leads to sliding and kinking in preference to catastrophic crack growth. Like most ceramics they are phase stable at high temperatures, generally up to 1500 °C. High temperature oxidation resistance is excellent for alumina-forming Ti3AlC2, Ti2AlC, and Cr2AlC, as reviewed by Tallman, et al. (Ref. 3). Compatibility with α-Al2O3 scales is further enhanced in cyclic exposures by a close matching of thermal expansion coefficients, (Ref. 4) i.e., (~9.3, 10.2, 11.3×10–6/K for Al2O3, Ti2AlC, and YSZ, to be discussed). Turbine environments generally contain 10 percent water vapor in the combustion gases, therefore moisture effects can be a concern for some materials (Ref. 5). Furnace tests of MAX phases in high temperature steam generally showed little effect on Al2O3 scale growth (Ref. 6). However, high velocity and high pressure gas can influence scale losses by the formation of volatile reaction products, such as TiO(OH)2 and Al(OH)3 (Refs. 7 to 10). This phenomenon had been discussed for 1100 to 1300 °C high pressure burner rig tests of Ti2AlC (Ref. 11). A single cubic growth rate parameter kcubic was measurably lower than comparable furnace TGA data, but it could be matched reasonably well if corrected for a slight volatility term. In general, a two-parameter cubic-linear growth-volatility law was believed to apply. Corresponding scale volatility loss rates, directly measured at 1300 °C on a pre-oxidized sample, were moderate (0.012 mg/cm2/h) and largely attributed to removal of the initial TiO2 transient scale. A related CH4 burner study of high purity Cr2AlC MAX phase demonstrated 1200 °C durability after 500 rapid (5 min. heat and 2 min. cool) thermal shock cycling (29 h hot time) (Ref. 12). Heating and cooling rates were ~1000 and 500 °C per minute, with a gas velocity of 5 m/s, producing a 75 °C/mm gradient. A 7 μm Al2O3 surface scale and a 13 μm Cr7C3 depletion zone formed with no signs of failure. No evidence of scale volatility was evident, although weight change was not provided, the velocity was moderate, and the total hot time was not extensive. The same high gradient BRT was used to produce 1400 °C surface temperatures for a YSZ/Cr2AlC/IN738 system in the first study of MAX phases used as bond coats for thermal barrier coatings (TBC) (Ref. 25). Here TBC failure was reported after 745 cycles, with only a 1.5 μm Al2O3 scale entrained within a porous, Cr7C3 bondcoat depletion phase. YSZ thermal barrier coatings have been considered to be a compatible complement to Al-MAX phases because of thermal expansion matching and extremely low volatility in water vapor. Initial studies showed superior oxidative stability up to 1300 °C, for long times (at least 500 h) for Ti2AlC substrates and less (268 h) for Cr2AlC, while withstanding large alumina TGO scale thickness (~35 to 40 μm) (Refs. 13 and 14). By comparison, typical superalloy systems can only survive 1150 °C maximum interface temperatures for extended periods, with a maximum sustained TGO below 10 μm (Ref. 15). High temperature SiC based systems are known to form slow-growing SiO2 scales. But these are subject to rate enhancement and volatile Si(OH)4 products in the presence of water vapor, as described comprehensively by Opila, et al. (Refs. 5, 16 to 19). Net weight losses are generally observed in high velocity, high pressure burner rig studies (e.g., 0.084 mg/cm2/h at 1300 °C) (Ref. 20). Furthermore, the loss rates have been shown from chemical physics to scale with v1/2 and pH2O 2 (Ref. 16). Low activity, moisture-resistant environmental barrier coatings (EBC), such as rare earth silicates, are needed to prevent substrate recession under turbine conditions (Refs. 21 to 23).

It has been noted by industry that in addition to dramatic variations of temperature over a given blade surface, blade-to-blade variations also exist despite identical design. These variations result from manufacturing variations, uneven wear and deposition over the life of the part as well as limitations in the uniformity of coolant distribution in the baseline cooling design. It is proposed to combine recent advances in optical sensing, actuation, and film cooling concepts to develop a workable active, closed-loop modulated turbine cooling system to improve by 10 to 20 the turbine thermal state over the flight mission, to improve engine life and to dramatically reduce turbine cooling air usage and aircraft fuel burn. A reduction in oxides of nitrogen (NOx) can also be achieved by using the excess coolant to improve mixing in the combustor especially for rotorcraft engines. Recent patents filed by industry and universities relate to modulating endwall cooling using valves. These schemes are complex, add weight and are limited to the endwalls. The novelty of the proposed approach is twofold 1) Fluidic diverters that have no moving parts are used to modulate cooling and can operate under a wide range of conditions and environments. 2) Real-time optical sensing to map the thermal state of the turbine has never been attempted in realistic engine conditions.

The overall goal of the Aeronautics Research Mission Directorate (ARMD) Seedling Phase II effort was to build on the promising temperature-sensing characteristics of the ultrabright thermographic phosphor Cr-doped gadolinium aluminum perovskite (Cr:GAP) demonstrated in Phase I by transitioning towards an engine environment implementation. The strategy adopted was to take advantage of the unprecedented retention of ultra-bright luminescence from Cr:GAP at temperatures over 1000 C to enable fast 2D temperature mapping of actual component surfaces as well as to utilize inexpensive low-power laser-diode excitation suitable for on-wing diagnostics. A special emphasis was placed on establishing Cr:GAP luminescence-based surface temperature mapping as a new tool for evaluating engine component surface cooling effectiveness.

This Technical Memorandum presents the current capabilities of the state-of-the-art Mach 0.3 Burner Rig Facility. It is used for materials research including oxidation, corrosion, erosion and impact. Consisting of seven computer controlled jet-fueled combustors in individual test cells, these relatively small rigs burn just 2 to 3 gal of jet fuel per hour. The rigs are used as an efficient means of subjecting potential aircraft engine/airframe advanced materials to the high temperatures, high velocities and thermal cycling closely approximating actual operating environments. Materials of various geometries and compositions can be evaluated at temperatures from 700 to 2400 F. Tests are conducted not only on bare superalloys and ceramics, but also to study the behavior and durability of protective coatings applied to those materials.

Physical vapor deposited thermal barrier coatings (TBCs) were thermally cycled to near-failure at It 50 degrees C. Normal failure occurred after 200-300 1 h cycles with only moderate weight gains (0.5 Mg cm(-2)). Delamination and buckling was often delayed until well after cooldown (desktop spallation), but could be instantly induced by the application of water drops, as shown in an accompanying video-recording. Moisture therefore plays a primary role in delayed desktop TBC failure. Hydrogen embrittlement is proposed as the underlying mechanism. Published by Elsevier Ltd. on behalf of Acta Materialia Inc.

A primarily silica-containing material with traces of organic compounds, as well as aluminum and calcium additions, was exposed to a Mach 0.3 burner rig at atmospheric pressure using jet fuel. The sample was exposed for 5 continuous hours at 1370 C. Post exposure x-ray diffraction analyses indicate formation of cristobalite, quartz, NiO and Spinel (Al(Ni)CR2O4). The rig hardware is composed of a nickel-based superalloy with traces of Fe. These elements are indicated in the energy dispersive spectroscopy (EDS) results. This material was studied as a candidate for high temperature applications under an engine technology program.

Silicon carbide (SiC) and silicon nitride (Si3N4) are proposed for applications in high temperature combustion environments containing water vapor. Both SiC and Si3N4 react with water vapor to form a silica (SiO2) scale. It is therefore important to understand the durability of SiC, Si3N4 and SiO2 in water vapor. Thermogravimetric analyses, furnace exposures and burner rig results were obtained for these materials in water vapor at temperatures between 1100 and 1450 C and water vapor partial pressures ranging from 0.1 to 3.1 atm. First, the oxidation of SiC and Si3N4 in water vapor is considered. The parabolic kinetic rate law, rate dependence on water vapor partial pressure, and oxidation mechanism are discussed. Second, the volatilization of silica to form Si(OH)4(g) is examined. Mass spectrometric results, the linear kinetic rate law and a volatilization model based on diffusion through a gas boundary layer are discussed. Finally, the combined oxidation and volatilization reactions, which occur when SiC or Si3N4 are exposed in a water vapor-containing environment, are presented. Both experimental evidence and a model for the paralinear kinetic rate law are shown for these simultaneous oxidation and volatilization reactions.

A commercially available, sintered silicon carbide was exposed to a temperature of 982°C for up to 50 h in a burner rig pressurized to 500 kPa. Synthetic sea salt added to the flame (5 ppm) resulted in the deposition of sodium sulfate and formation of a sodium magnesium silicate corrosion product. A 16% reduction in room‐temperature strength occurred after 5 h of exposure; this reduction was due to the formation of surface pits. Exposure for longer times resulted in continued strength reduction, up to 56% at 25 h. Samples exposed for 50 h were so degraded that mechanical tests could not be conducted. The strength after 25 h of exposure to a salt concentration of 2 ppm was similar to the as‐received strength, whereas exposures to 10 ppm of salt resulted in strengths similar to that observed with 5 ppm of salt.

Measuring the temperatures of advanced materials, such as ceramic matrix composites (CMC's), in a hostile environment has been a difficult task because of the poor adhesion of the measurement systems. Commonly used wire thermocouples (TC) cannot be attached to such ceramic-based materials via conventional spot-welding techniques, and commercially available ceramic cements fail to provide sufficient adhesion at high temperatures. Although advanced thin-film TC's provide minimally intrusive surface temperature measurement and adhere well on CMC's, their fabrication requires sophisticated, expensive facilities and is time consuming. In addition, the durability of lead wire attachments to both thin-film TC's and substrate materials requires further improvement.

A commercially available, hot isostatically pressed Si3N4 containing 4 wt% yttria was exposed to 982 degrees C for up to 75 h in a burner rig pressurized to 500 kPa. Synthetic sea salt added to the flame (5 ppm) resulted in formation of a sodium magnesium silicate corrosion product. A 33% reduction in room-temperature strength occurred after 5 h exposure. This is thought to be due to modification of the near-surface grain boundary phase and relief of surface compressive stresses. Exposures to longer times resulted in continued strength reduction, up to 46% at 75 h. Strength also decreased when salt concentration was increased, as shown by exposures using 2 and 10 ppm sea salt. In tests at 100 and 300 kPa with other variables held constant, post-corrosion strengths were similar to those after 500 kPa exposure.