Effect of sulfur removal on scale adhesion to PWA 1480

  • PDF / 1,947,934 Bytes
  • 9 Pages / 613 x 788.28 pts Page_size
  • 44 Downloads / 214 Views

DOWNLOAD

REPORT


I.

INTRODUCTION

IT is well documented and accepted that the addition of reactive elements to oxidation-resistant, hightemperature MA1, MCr, and MCrA1 alloys leads to a dramatic improvement in A1203 or Cr203 scale adhesion. One of the proposed mechanisms (Smeggil et al. tl]) is that the reactive elements getter sulfur impurities in the alloy. This prevents sulfur segregation at the oxide-metal interface and weakening of the scale-metal bond. tl,2,31 A number of studies have examined the effect of reduced sulfur content on scale adhesion for alloys free of any reactive elements. These studies have been remarkably successful for undoped NiCrA1, I4-71 FeCrA1,18,91 NiCr, tml and Cr. [3] The purpose of this study was to determine whether adherence could be improved for a complex superalloy, PWA 1480. This alloy contains no Y, Hf, or Zr reactive elements, precluding adhesion effects due to these dopants. Prior studies of this alloy tj~l have shown it to be relatively oxidation resistant for 200 hours at 1050 ~ (0.71 and 0.14 m g / c m 2 weight gain for isothermal and cyclic exposures, respectively? but rapidly degraded at 1200 ~ (4.56 and 263 m g / c m for isothermal and cyclic exposures, respectively, where 1 m g / c m 2 = 0.01 kg/m2). The scale phases were primarily A1203 and C r T a O 4 in isothermal tests, with additional Cr203, NiA1204, and NiO phases in cyclic tests. Therefore, 1100 ~ cyclic exposures were chosen for this study in order to achieve moderate scale growth rates and spalling. Interracial segregation layers were removed by repeated oxidation and scale removal via polishing, t5,6,7~ The effectiveness of sulfur purging by annealing in hydrogen, as originally demonstrated for Cr by Lees 131was also investigated. Residual sulfur content was determined by glow-discharge mass spectroscopy, which is sensitive to parts per billion (ppb) even though accuracy is limited to +25 pct without low-level standards.* The *Charles Evans and Associates, Redwood City, CA.

weight change, scanning electron microscopy (SEM), X-ray diffraction (XRD) and the area of interfacial spalling. More detailed results can be found in a preliminary publication. 1121Recent similar works also document the effectiveness of hydrogen annealing for desulfurization and improving scale adhesion for PWA 1480, PWA 1484, and Rene'N5. t~3,141 II.

EXPERIMENTAL PROCEDURE

A. Cyclic Oxidation~Polishing Series PWA 1480 is a y / y ' superalloy: Ni-10Cr-5AI-1.3Ti12Ta-4W-5Co and less than 15 ppmw S. There are no intentional additions of reactive elements to this alloy. Oxidation coupons were machined to 0.8 x 13.8-mm-diameter discs from single-crystal rods and to 0.8 x 13.0 x 25.5-mm rectangles from polycrystalline melt stock. The single-crystal and polycrystalline samples were finished to 600-grit emery, ultrasonically cleaned, weighed, and oxidized in air at 1100 ~ for 1-hour cycles. Weight changes were recorded after each cycle. All oxidized samples were lightly brushed prior to weighing to remove minor amounts of loose oxide. Samples were repolished