Current production thermal-barrier coatings (TBCs) have been shown to be capable of reducing the average temperatures of metallic components by 50 to 80 °C and hot-spot temperature by up to 140 °C. This substantial temperature reduction has been used to extend the life of metallic components in aircraft turbines. However, for critical applications aimed at improving engine performance where significantly higher temperatures are involved, higher-durability TBCs are required. An improved bond coat incorporating metallic and cermet layers has been demonstrated to increase the thermal-fatigue life of a plasma-sprayed thermal-barrier coating (TBC) by a factor of two or more. These TBCs can be applied to components in gas turbines and in diesel engines.
A typical TBC comprises a single metallic bond-coat layer, 0.005 to 0.008 in. (about 0.13 to 0.020 mm) thick, coated with a single ceramic top-coat layer, 0.005 to 0.020 in. (about 0.13 to 0.50 mm) thick. The bond-coat layer is typically MCrAlX, where M signifies Ni, Co, or Fe and X signifies Y, Zr, Hf, Yb, or another reactive element. The ceramic top-coat layer is typically zirconia partially stabilized with 6 to 8 weight percent of yttria. The bond coat is typically processed by plasma spraying, while the top coat can be processed by either plasma spraying or electron-beam physical vapor deposition. For TBCs using a plasma-sprayed top coat, the bond coat is prepared with a rough surface to improve bonding.
In spite of the necessity of bond-coat roughness to enhance adhesion, the roughness also tends to intensify the stresses that occur at the interface between the ceramic and the bond coat. Recent work has shown that the high stresses are particularly significant in the vicinity of the peaks in the rough bond coat (see Figure 1). Detailed investigation has further shown that the stresses can be minimized by matching the thermal expansion of the peaks of the bond coat to the ceramic top coat.
Figure 2 illustrates a TBC design that addresses these problems through the use of a two- layer bond coat. The first layer of the bond coat is a typical MCrAlX, as described for a conventional TBC above. The second layer of the bond coat incorporates a fine dispersion of a particulate second phase in an MCrAlX matrix. The second phase is required to have a coefficient of thermal expansion as low as, or preferably lower than, the yttria stabilized zirconia ceramic layer, it must be stable up to the intended use temperature, chemically inert with respect to the MCrAlX matrix, and must be chemically compatible with the thermally grown alumina scale. Candidate second-phase materials include alumina, chromia, yttrium-aluminum garnet, nickel-aluminum spinel, yttria, mullite, and other oxides.
Since the goal is to achieve expansion matching of the second-layer peaks to the yttria stabilized zirconia, the particulate second phase must have dimensions less than that of the peaks, typically less than 5 μm, and must be well dispersed in the MCrAlX matrix. The volume fraction of the particulate must be high enough to achieve substantial matching of the peak expansion to that of the ceramic layer. For the case of alumina additions to MCrAlX, an alumina volume fraction of 0.71 is required to achieve a near-zero thermal expansion mismatch. In practice, the thermal expansion of the second layer must be balanced against the other requirements for the layer, such as ductility and oxidation resistance.
Coatings to date have been plasma sprayed using starting powders produced by mechanical alloying. The mechanical-alloying process that has been developed has produced plasma-spray starting powders with up to 20 volume percent of a fine dispersion of submicron alumina particles. The ceramic layer life was doubled for TBCs, using a bond coat of only 5 volume percent alumina additions. This technologically important, and repeatable, increase in life could be used to push the TBCs to higher operating temperatures.
Higher volume percentages of alumina, up to 20 volume percent, were expected to provide even longer lives due to better expansion matching with the ceramic. While some samples did exhibit longer lives, these compositions also exhibited widely varying oxidation responses. The net result of the erratic oxidation response was a reduction in the average life for these coatings. Alternative thermal-spray processes, such as high-velocity oxy-fuel spraying (HVOF), have proven to produce more homogeneous particle distributions and hold the promise of even higher gains in TBC life. The HVOF coatings are currently being tested.
This work was done by William J. Brindley and Robert A. Miller of Lewis Research Center and Beverly J. M. Aikin of Case Western Reserve University. For further information, access the Technical Support Package (TSP) free on-line at www.techbriefs.com under the Materials category.
Inquiries concerning rights for the commercial use of this invention should be addressed to
NASA Lewis Research Center
Commercial Technology Office
Attn: Tech Brief Patent Status
Mail Stop 7 - 3
21000 Brookpark Road
Refer to LEW-16390.