Strains in Thermally Growing Alumina Films Measured In-situ Using Synchrotron X-rays

Strains in thermally grown oxides have been measured in-situ, as the oxides develop and evolve. Extensive data have been acquired from oxides grown in air at elevated temperatures on different model alloys that form Al2O3. Using synchrotron x-rays at the Advanced Photon Source (Beamline 12BM, Argonne National Laboratory), Debye-Scherrer diffraction patterns from the oxidizing specimen were recorded every 5 minutes during oxidation and subsequent cooling. The diffraction patterns were analyzed to determine strains in the oxides, as well as phase changes and the degree of texture. To study a specimen's response to stress perturbation, the oxidizing temperature was quickly cooled from 1100 to 950C to impose a compressive thermal stress in the scale. This paper describes this new experimental approach and gives examples from oxidized β-NiAl, Fe-20Cr-10Al, Fe-28Al-5Cr and H2annealed Fe-28Al-5Cr (all at. %) alloys to illustrate some current understanding of the development and relaxation of growth stresses in Al2O3. Introduction. The existence of a growth stress associated with oxidation has long been envisioned [1], since most oxides have higher volumes than their corresponding metals. Later, it was proposed that growth stress can arise from oxide formation within the scale as the diffusing cations and anions meet and react, particularly along oxide grain boundaries [2]. Several other mechanisms have also been proposed to explain the origin of growth stresses [3]; however, actual in-situ measurements of these stresses are scarce. Values are usually deduced from the difference between measured residual stresses at room temperature and the calculated thermal stress based on thermal expansion coefficients (CTE) of the oxide and the alloy. However, this method depends on CTE values, which are often unreliable, cannot properly account for stress relaxations and fails to follow the development of growth stresses with time. Among the limited experimental efforts to determine the magnitude of oxidation stresses insitu are deflection [4-6] and x-ray diffraction [7-9] techniques. The former makes use of an asymmetrical specimen that is long and thin with a protective layer on one side to prevent one face of the specimen from oxidizing. Any stress generated from oxidation of the bare surface, therefore, will cause the specimen to deflect. The direction and degree of deflection give an indication of the sign and magnitude of the growth stress respectively. The technique, however, relies on a protective layer that cannot fully stop oxidation at elevated temperatures. Furthermore, analysis of the stress level is complicated by the visco-elastic response of the alloy at high temperatures [10]. X-ray diffraction, especially the sinψ method, is a well known and proven method for stress measurements. The challenge is to perform the experiment at elevated temperatures with sufficiently high accuracy and precision and sufficiently fast response time. In this paper, a novel x-ray technique (a variant of the sinψ method) is described, where the elliptical distortion of Debye-Scherrer diffraction rings is measured for all accessible ψ angles to obtain the in-plane strain. In-situ measurements were performed with synchrotron radiation at beamline 12BM at the Advanced Photon Source at Argonne National Laboratory. Some results are presented to illustrate a few important issues concerning the development and relaxation of stresses during oxide growth. Experimental. Fig. 1(a) shows a simple schematic of the experimental setup. The test sample is positioned on a thin alumina shelf, cemented into an open ended horizontal tube furnace. A Type S Pt-Pt(Rh) thermocouple (TC), whose temperature was calibrated by melting a gold foil in the sample position, was placed under the shelf, immediately beneath the sample. The sample surface is usually polished to a 1 μm finish using diamond slurry. During the test, a 21.6 keV xray beam, with a spot size of about 0.1 mm (FWHM) x 1 mm, strikes the sample surface at an incidence angle of 2 to 5°. Half circles of Debye-Sherrer diffraction rings from the sample are recorded using an image plate detector; an example of these half rings is shown in Fig. 1(b). For precise determination of the x-ray scattering center, and any detector misorientation, a porous, sintered alumina sheet is used as a reference sample. The reference is placed into the x-ray path, using a pneumatic actuator, after the test specimen is lowered. The x-rays penetrate through this thin reference and yield full-circle diffraction spectra (Fig. 1(c)) that can be analyzed to obtain the location of the beam centroid to within ± 2 μm. The reference and sample motions are computer driven; typically, a reference spectrum is recorded after five successive sample measurements, each of 5 minutes duration. Figure 1: (a) Schematic of the experimental setup. (b) Half rings of α-Al2O3 reflections from a NiAl sample. (c) Full α-Al2O3 Debye-Scherrer rings from the alumina reference. ψ incident xray beam 4° q surface normal diffracted ray