The high-temperature components of a jet turbine engine are made
from nickel-base superalloys. These components must be able to
withstand high stresses, fatigue, and corrosive reactions with
high-temperature gases. Such oxidation resistance is associated
with slow-growing Al2O3 scales that remain
adherent to superalloy components after many thermal cycles. Historically,
good oxidation resistance has been obtained by coating these components
with Ni-Cr-Al-Y coatings, where small additions of yttrium (Y)
were necessary for scale adhesion. Subsequently, it was found
that the Y aids scale adhesion by preventing sulfur from
segregating to the scale metal interface and thus preventing the
sulfur from weakening the oxide-metal bonds. Y is a difficult
element to incorporate in single-crystal superalloy castings,
but it was shown in early work at the NASA Lewis Research Center
that good adhesion could be obtained for low-sulfur, uncoated,
single-crystal superalloys, without Y additions (ref. 1).
Low sulfur contents for these uncoated superalloys were achieved
in the laboratory by a high-temperature hydrogen annealing process.
This process allows segregation and surface cleaning of sulfur
monolayers in a reducing environment. (Annealing in air or oxygen
simply traps the sulfur at the oxide-metal interface). Although
this process has been pursued by industry (ref. 2), another approach
is to remove sulfur from the alloy in the melting process. Both
processes involve extra effort and costs that must be balanced
against improved performance. The present study was designed to
establish a guideline for the minimum level of desulfurization
needed to achieve maximum performance (ref. 3).
Coupons of various thicknesses of the superalloy PWA 1480 were
hydrogen annealed at various times (8 to 100 hr) and temperatures
(1000 to 1300 °C), resulting in coupons with
sulfur contents ranging from about 0.05 to 5 ppm. This variation
occurs because sulfur removal is approximately controlled by diffusion
and the parameter (
Dt
/
L
2), where
D
is the diffusion coefficient of sulfur,
t
is diffusion
time, and
L
is the sample thickness.
Cyclic oxidation tests at 1100 °C were then used to assess adhesion
and spalling. The weight change of one set of 20-mil (0.5-mm)
samples, annealed for 20 hr at 1000, 1100, 1200, and 1300 °C, is
shown in the following figure. Clearly, the effect of the annealing
temperature is quite dramatic in that the higher temperatures
produced scales that spalled very little, whereas the lower temperatures
resulted in severe weight losses comparable to those for the as-received,
unannealed sample.
Similar effects were observed as a function of annealing time
or sample thickness, because these parameters determine the final
sulfur content from the hydrogen annealing process. Thus, spalling
behavior can be related to a single parameter: sulfur content.
This relationship is shown in the next figure, where the weight
change after 500 1-hr cycles for the 20-mil samples is plotted
against the sulfur content, as measured by glow discharge mass
spectrometry. (Individual points are identified by the annealing
temperature and time.) There is a steep reduction in the degree
of spallation as the sulfur content is reduced below the as-received
value of ~6 ppmw (parts per million by weight). Furthermore, there
is very little spallation for samples having less than 1 ppmw
sulfur, with no further benefit for samples having less than about
0.3 ppmw sulfur. These values translate to 1 to 2 monolayers of
total sulfur segregation possible for a given sample geometry
and sulfur level. A similar analysis of results, summarized from
many published studies, produced the same sulfur dependencies
for PWA 1484, René N5, René N6, René 142, and CMSX 4 superalloys
(ref. 3).
Thus, for typical airfoil dimensions in advanced aeroengines,
0.5 ppmw sulfur would be a commendable target level and should
result in excellent scale adhesion. Most advanced turbine blades
are also thermal barrier coated with an insulating layer of plasma-sprayed
ZrO2. The integrity of this coating will also depend
on good Al2O3 scale adhesion, since the
scale grows beneath the ZrO2. These low-sulfur alloys
will improve scale adhesion and, thus, should improve ZrO2
coating lives, even when NiCrAlY bond coats are employed.
Author: Dr. James L. Smialek
Headquarters program office: OA
Previous articleLast updated April 30, 1997
Responsible NASA Official:
Gynelle.C.Steele@nasa.gov
216-433-8258
Point of contact for NASA Glenn's Research & Technology reports:
Cynthia.L.Dreibelbis@nasa.gov
216-433-2912
SGT, Inc.
Web page curator:
Nancy.L.Obryan@nasa.gov
216-433-5793
Wyle Information Systems, LLC
