The chemical compositions of the metals and alloys used in this study are given in Table 1. Pure Ni, Cr, Fe were provided by MRC, Praxair, GoodFellow respectively. Wrought 718 is detailed in [10], Ti-6Al-4V in [11]. PM2000 was provided by Plansee. Some chromia-forming HEA alloys under development were tested, they are named F, G and H. AM1 [9], R77, DS200-Hf alloys were supplied by Safran Aircraft Engines. AMS20 is an alumina forming superalloy under development, AMS20-S was desulfirized by ONERA, CMSX-4 SLS, AMS19, GOL1, GOL2, AD730 and ST201 were provided by Safran Tech and alloys A, B, C, D and E are new alumina forming alloys under development.
Specimens as pure Ni, pure Fe, pure Cr, HEA alloys, AM1, AMS19, AMS20, GOL1, GOL2, AD730, ST201, DS200, R77 and alloys A-E with a diameter of 10 to 13 mm and a thickness of 1 or 2 mm were cut from a plain cylinder. CMSX-4 with a diameter of 20 mm and a thickness of 1–2 mm was cut as a semi cylinder. Specimens as wrought 718, Ti-6Al-4V and PM2000 with dimensions of 15 or 10 × 1 or 2 mm were cut from a plain ingot. All samples were ground to P600 grit surface finish. Each coupon was degreased in acetone and alcohol, then weighed before the oxidation test. The oxidation tests were conducted using SETARAM TAG24 S thermobalances. The samples were oxidized in a flow of synthetic air of 10 ml/min corresponding to a gas velocity of 0.53 mm/s at room temperature. All heating rates and initial cooling rates were set at 60°C/min. Data points were recorded every 10 seconds during the test.
Samples underwent stepwise multi-temperature thermogravimetric analysis (SMT-TGA). This testing procedure involves multiple steps and temperature dwells. The duration of each dwell is carefully selected to achieve optimal signal-to-noise ratio, aiming for approximately the same mass gain at each dwell. The mass gain rate should remain significantly higher than the drift of the TGA apparatus but not too large, allowing for several dwells without excessive oxidation of the sample. A specific temperature program was tailored for each alloy type, depending on the nature of the main oxide expected to form, such as NiO, FeO, TiO2, Cr2O3 or Al2O3 (see Fig. 1 and Supplementary information).
Table 1
Chemical composition (wt%) of the metals and alloys
wt % | Ni | Fe | Cr | Al | Co | Mo | Ta | Ti | W | Nb | Hf | Re | C ppm | S ppm | Others |
AMS20 | Bal. | | 4-5.5 | 5.4-6 | 5.5–7.5 | 0.1–0.7 | 7.5-9 | 0.1–0.25 | 4–5 | | 0.04–0.15 | 4.8–6.2 | | 8 | |
AMS20 -S | Bal. | | 4-5.5 | 5.4-6 | 5.5–7.5 | 0.1–0.7 | 7.5-9 | 0.1–0.25 | 4–5 | | 0.04–0.15 | 4.8–6.2 | | < 1 | |
AMS19 | Bal. | | 4–5 | 5.9–6.5 | 9.5–10.5 | 0.3–0.7 | 7.5–8.5 | 0.2–0.7 | 3.2-4 | | 0.1–0.2 | 3.7–4.5 | | | |
GOL1 | Bal. | | 4–8 | 6–8 | 12–15 | 0.5-3 | 4–6 | 0–3 | 0–2 | | 0-0.2 | 3.5–5.5 | | | |
GOL2 | Bal. | | 6–9 | 4–6 | 5–8 | 2–4 | 5–7 | | 2–5 | | 0,1 − 0,9 | 5–7 | | | |
ST201 | Bal. | | 15.7 | 5.8 | 5.7 | 0.7 | 2.7 | 1.25 | 1.8 | | 1.3E-3 | | | | |
CMSX-4 | Bal. | | 6.5 | 5.6 | 9 | 0.6 | 6.5 | 1 | 6 | | 9.3 E-4 | 3 | 40 | 0.32 | Mn 2.2ppm |
AM1 | Bal. | | 7.54 | 5.2 | 6.61 | 2.01 | 7.97 | 1.2 | 5.49 | | 4.9 E-4 | 8.4 E-4 | 47 | 0.1 | Mn 2ppm |
R77 | Bal. | | 14.6 | 4.3 | 15 | 4.2 | | 3.35 | | | | | | | |
DS200 | Bal. | | 9 | 5 | 10 | | | 2 | 12.5 | 1 | 2.1 | | | | |
IN 718 | Bal. | 18 | 19 | 0.5 | 0.5 | 3.1 | 0.1 | 1 | | 5.2 | | | 1000 | | Mn 0.2 Si 0.2 |
AD730 | Bal. | 7.6 | 17.3 | 2.09 | 8.75 | 3.03 | | 3.53 | | 1.09 | | | | | |
IN 738 | Bal. | | 17.4 | 7.1 | 8.2 | 1 | 0.6 | 4 | 0.8 | 0.6 | | | 800 | | B 0.1 Zr 0.1 |
PM2000 (Plansee) | | Bal. | 20 | 5.5 | | | | 0.5 | | | | | | | Y2O3 0.5 |
Cr (Praxair) | | | Bal. | | | | | | | | | | | | |
Fe (GoodFellow) | | Bal. | | | | | | | | | | | | | |
Ti6Al4V | | 0.16 | | 6.4 | | | | Bal. | | | | | | | V 3.5 |
Ni (MRC) | Bal. | | | | | | | | | | | | | | |
When a sample is oxidized at a given temperature after being oxidized at a succession of dwell at different temperatures, the microstructure of its oxide layer and the depletion profile of certain alloy elements in the metal are not the same as in simple isothermal oxidation. Therefore, the thermal history of the sample can influence its oxidation kinetics. To investigate potential effects of thermal history, such as the evolution of the parabolic constant over time or after high-temperature annealing of the oxide scale, the same temperature is used for two separate dwells in the SMT-TGA: one during temperature increase steps and the other during cooling steps. Effects of thermal history can then be identified by comparing kinetic parameters measured during the two dwells at the same temperature. If oxidation kinetics parameters remain similar across the two dwells, one may conclude the absence of history effects. Conversely, if oxidation kinetics parameters differ between the two dwells, a history effect is indicated.
Following each oxidation test, the oxide scale was examined using X-ray diffraction (not shown here) on a Bruker D8 GIXR instrument employing Cu-Kα radiation, with a step size of 0.02°, an angle of incidence of 10°, and a scan step time of 2 seconds within the 2θ range from 15° to 80°. Cross-sectional analyses were carried out using optical microscopy and a SEM FEI Quanta 450 equipped with a Bruker Quantax (SDD) EDS detector. EBSD analyses were employed to gain further insight into the microstructure of the oxide scale using a scanning electron microscope JEOL JSM-7100 equipped with a NordlyNano EBSD camera. To prevent any spallation of the oxide scale during cross-section polishing, the oxidized samples were coated with a thin layer of silver followed by an electroplated copper layer.