Materials Science and Engineering A 462 (2007) 436–440 Analysis of creep curves of a nickel base superalloy in a wide stress/temperature range M. Maldini ∗ , G. Angella, V. Lupinc CNR-IENI, Via Cozzi 53, 20125 Milano, Italy Received 31 August 2005; received in revised form 11 November 2005; accepted 14 November 2005 Abstract The creep behaviour of the nickel base superalloy Nimonic 263 has been studied at constant load and temperature in the 750–30 MPa/600–950 ◦ C range. The experimental results have shown a very strong dependence of the creep curve shape with the applied stress/temperature. At high stresses, when an instantaneous plastic strain occurs during the initial loading, the creep curves are charac terised by an important decelerating/primary creep while, at lower applied stresses, the primary stage becomes very short and small and other strain stages dominate the creep curves. The experimental results can be rationalised supposing that the mobile dislocation density decreases/increases during creep, from an initially high/low value in the tests run above/below the alloy yield stress. With this hypothesis, the creep results have been modelled using coupled differential equations of the Kachanov form. © 2006 Elsevier B.V. All rights reserved. Keywords: Constitutive equations; Creep; Nickel alloy 1. Introduction When the experimental creep tests of engineering alloys cover only a limited stress/temperature range, the shapes of the creep curves often exhibit common features independent of the duration of the tests. For instance, in most experimental studies on γ ′ reinforced nickel base superalloys the creep curves show, after a small and short primary creep, a dominant accelerating creep stage where the strain rate, ˙ ε, increases linearly with the strain, ε [1–4]: ˙ ε = ˙ ε 0 (1 + Cε) (1) where ˙ ε 0 is the back-extrapolated strain rate at the beginning of the test, and C a parameter controlling the shape of the creep curve. Usually no steady state creep rate has been observed, but only a point of inflection in the strain versus time creep curve, corresponding to the minimum creep rate between the succes- sive decelerating primary and accelerating stages. Although Eq. (1) has demonstrated good interpolation properties of experi- mental creep curves [1–4], one must be cautious to extrapolate the creep properties too far from the experimentally explored ∗ Corresponding author. Tel.: +39 0266173378; fax: +39 0266173321. E-mail address: maldini@ieni.cnr.it (M. Maldini). stress/temperature range, since new qualitative and quantitative characteristics may appear in the creep behaviour of the mate- rial. The aim of the present paper is to modify and extend Eq. (1) in order to correctly model the creep behaviour of the nickel base superalloy Nimonic 263 in the very wide stress/temperature field the alloy can experience during service. 2. Material and experimental procedure Nimonic 263 is a polycrystalline Ni-base superalloy used in combustion chambers of gas turbines. The nominal chemical composition of the alloy is given in Table 1. The heat treatment sequence is: 2 h/1150 ◦ C water quenched +8 h/800 ◦ C air cool- ing. The alloy is reinforced by γ ′ particles with spherical shape and size around 20 nm. The dependence of the γ ′ volume fraction with temperature has been studied in [4]: 10% and almost con- stant between room temperature up to 800 ◦ C, at 900 ◦ C the γ ′ volume fraction is reduced approximately to 3% and it becomes nil above the γ ′ solvus temperature, at about 920 ◦ C, when the alloy becomes mainly a single phase solid solution. The alloy grain size is about 0.1 mm. Constant load creep tests were run on cylindrical specimens in the stress/temperature range 750–36 MPa/600–950 ◦ C produc- ing rupture times in the 10–1250 h range. The creep specimens with cylindrical symmetry 5.6 mm gauge diameter and 28 mm 0921-5093/$ – see front matter © 2006 Elsevier B.V. All rights reserved. doi:10.1016/j.msea.2005.11.084