Materials And Experimental Methods

Nine different thermal spray powders used in this program were supplied by seven vendors. Additional powders were supplied at a later time by two of the original vendors, and a group of powders were supplied by an eighth vendor. Each material is nominally zirconia-8% yttria (by weight). The actual compositions were taken from the vendors' analysis sheets. As a rough test of the particle-to-particle consistency, five particles of each powder were examined by KEVEX™ X-ray florescence. In addition, a full field (120X) KEVEX™ chemistry was taken. A comparison of these analyses is shown in Table 1. The vendor-supplied analyses are shown in the wet chemistry column. The KEVEX™ column gives the full field chemistry and then lists (in parentheses) the particle-to-particle range. In both cases, the balance of the material is Zr.

Prior to spraying, physical data were obtained on each of the nine powders. The particle size distribution of each material was obtained with a Leeds and Northrup Microtrac™ (Model 7995-1, extended range). The samples were run in water using a trace of 100 surfactant when needed. The specific surface area of each powder was measured by the Branaur, Emitt, Teller (BET) method using a Quantachrome Model QS-8 Quantasorb. A 3-point BET was run for each powder using nitrogen as the adsorbant in a helium carrier-gas stream. The true powder par-' ticle density was measured in a Quantachrome PY-5 Null Pychometer employing pure helium gas. The average of three runs is reported. Powder flowability was measured using a Hall Flow Cup 191. Rather than reporting the time for 50 grams of material to flow, the data were inverted to provide flow rate (in grams/sec.). In addition, the powders were examined by Scanning Electron Microscopy (SEM) for overall shape and structure, and they were evaluated metallographically for microstructure.

The powders were air plasma sprayed using identical spray parameters onto Hastel-loy-X buttons (1" diameter x 0.0125" thick) with and 0.008 inch VPS Ni-22Cr-10A1-1Y bond coating. The spray parameters for both operations are shown in Table 2. The typical roughness of the bond coats was 600 p-in AA. The air plasma work was performed on an X-Y traverse to insure the reproducibility of the process throughout this test and future iterations.

The as-sprayed coatings were evaluated for erosion resistance by impinging 50 pm A1203 at 20" angle onto the surface. The depth of penetration was measured, and the results reduced to an erosion reading in seconds per mil. All of the values were normalized against a Lexan™ plastic standard to allow future comparison of the data. The coatings were also evaluated metallographically for structure and porosity.

were then evaluated metallographically to determine coating changes during thermal exposure.

In order to evaluate the effects of phase structure on TBCs, X-ray diffraction (XRD) analyses were performed on the as-received powders, the as-sprayed coatings, and the coatings after thermal cycling. A Philips-Norelco diffractometer, automated by a microcomputer'10' for data acquisition, storage and analysis was

0 0

Post a comment