Predicting Oxide Spallation from Sulphur-Contaminated Oxide/Metal Interfaces
- PDF / 465,973 Bytes
- 12 Pages / 439.37 x 666.142 pts Page_size
- 30 Downloads / 228 Views
Predicting Oxide Spallation from SulphurContaminated Oxide/Metal Interfaces H. E. Evans
Received: 25 June 2012 / Published online: 21 December 2012 Ó Springer Science+Business Media New York 2012
Abstract The deleterious effect of sulphur contamination on oxide spallation has been well demonstrated experimentally. The present paper attempts to account for this using finite-element modelling of crack growth along a sulphur-contaminated interface during cooling. Sulphur reduces the intrinsic work of adhesion, Wad, of this interface and, it is suggested, this is equivalent to reducing the characteristic fracture stress for that interface. It is shown that this approach does predict the typical trends in S-bearing alloys of spallation after shorter exposure times. A significant result is that the effective fracture energy for spallation, ceff, reduces with increasing sulphur contamination but is always 1–2 orders of magnitude larger than Wad. This high value for ceff arises because of creep relaxation within the alloy. Sulphur does not affect this process directly but reduces the extent of creep relaxation in a cooling transient by initiating spallation at a smaller value of temperature drop. Keywords Alumina spallation Sulphur effect Finite-element analysis Effective fracture energies
Introduction Alloy trace elements such as sulphur [1–7] and phosphorus [8, 9] and more substantial additions such as titanium [10, 11] are known to segregate to oxide/metal interfaces and can reduce the adherence of protective oxide layers such as alumina and chromia. In the case of titanium, the deleterious effect has been associated with the disruption to the layer caused by the local formation of titanium-rich oxides [10, 11]. With the trace elements, however, it seems likely that their segregation to H. E. Evans (&) School of Metallurgy and Materials, The University of Birmingham, Birmingham B15 2TT, UK e-mail: [email protected]
123
4
Oxid Met (2013) 79:3–14
Fig. 1 Cyclic oxidation kinetics (1-hour cycles at 1,100 °C) of the Ni-based superalloys AM1 and MCNG [14]
the oxide/metal interface directly reduces the work of adhesion of the oxide to the alloy. Particular attention, both experimental and theoretical, has been paid to the influence of sulphur and this will also be the focus of this paper. The deleterious effect of sulphur on oxide adherence is well established in experimental studies. It has been shown, for example, that oxide spallation increases with alloy sulphur content [12–14] for both alumina- and chromia-formers. An example for the alumina-forming AM1 alloy is shown in Fig. 1 [14] for alloy sulphur contents of 0.22, 0.41 and 3.2 ppm. Clearly, increasing alloy sulphur content initiates oxide spallation, as evidenced by a decline in mass gain, at shorter exposure times than alloys with lower sulphur content. Consistently, hydrogen annealing, to reduce alloy sulphur concentrations, has been shown to improve spallation resistance in a number of studies [15–18]. This body of work strongly supports the view th
Data Loading...
