Comparison of the phase compositions in Alloy 718 measured by atom probe tomography and predicted by thermodynamic calculations

https://doi.org/10.1016/S0921-5093(01)01881-0Get rights and content

Abstract

A comparison has been made of the phase compositions predicted by thermodynamic calculations and measured by atom probe tomography in a commercial Alloy 718 nickel based superalloy. The results indicate that caution should be taken in the application of the NiFe database/Thermocalc™ to superalloys because some of the predicted phases may not correspond to actual microstructure of commercial alloys. When microstructural information is taken into account and absent phases are suppressed in the calculations, reasonable agreement between the atom probe data and the thermodynamic predictions may be achieved.

Introduction

The design of new or modified nickel-based superalloys for high temperature applications is an extremely expensive and time consuming process. Therefore, thermodynamic calculations are being introduced in an attempt to reduce these costs. However, little experimental data are available on the effectiveness of these predictive tools in complex multiphase and multicomponent commercial alloys. The three-dimensional atom probe (3DAP) is able to determine the partitioning and concentrations of the alloying elements in the phases present in these complex systems [1].

In this study, thermodynamic predictions of the phase compositions [2] are compared to the compositions of the phases, as determined by atom probe tomography, at various annealing temperatures. The alloy selected for this study was a commercial Alloy 718 [3]. Niobium-containing Alloy 718 is strengthened by a combination of DO22-ordered Ni3(Nb,Ti) γ″ and Ni3(Al,Ti,Nb) γ′ phases in a face centered cubic γ matrix [3], [4], [5], [6], [7]. Previous atom probe tomography characterizations of Alloy 718 [8], [9], [10] revealed that the secondary or fine (<∼20 nm diameter) precipitates consisted of two distinct types of regions enriched in either niobium or in aluminum and titanium that are characteristic of the γ″ and γ′ phases, respectively. This study complements a previous atom probe tomography study on another Alloy 718 with a different multistep heat treatment [10].

Section snippets

Experimental

The nominal composition of the commercial Alloy 718 used in this study was Ni—21.5 at.% Fe, 19.6% Cr, 0.31% Co, 1.76% Mo, 3.24% Nb, 1.18% Ti, 1.27% Al and 0.19% C (Ni—20.7 wt.% Fe, 17.6% Cr, 0.32% Co, 2.92% Mo, 5.21% Nb, 1.05% Ti, 0.55% Al, and 0.04% C). The material was given a multistep heat treatment of 1 h at 1093 °C, 8 h at 718 °C, a slow cool at a rate of 55 K h−1 to 621 °C, 8 h at 621 °C and an air cool to room temperature. The microstructural characterizations were performed after the step 8

Thermodynamic predictions

The equilibrium compositions and amounts of phases present in Alloy 718 at different temperatures were calculated with the use of Thermocalc™ version M and a commercial 12 element NiFe database [2]. The phases considered in the calculation were face centered cubic γ, L12-ordered γ′, DO22-ordered γ″, σ, Laves, MC, M6C, M23C6, M7C3 and liquid. It was necessary to suppress the orthorhombic Ni3Nb δ phase from the calculations due to the interference with the γ″ phase.

The thermodynamic predictions

Microstructure

Transmission electron microscopy of this alloy has revealed a uniform distribution of precipitates in the γ matrix, as shown in Fig. 1. No significant difference in the microstructure was noted between the 8 h at 718 °C and subsequent stages of the heat treatment. The size of these precipitates was approximately 5–13 nm. No other intragranular phases, such as δ and Laves, were observed. All the grain boundaries were free of δ, Laves and NbC precipitates. The low chromium level of this particular

Conclusions

The results of this study indicate that caution should be taken in the application of the NiFe database/Thermocalc™ to superalloys because some of the predicted phases may not correspond to actual microstructure of commercial alloys. When microstructural information is taken into account and absent phases are suppressed in the calculations, reasonable agreement between the atom probe data and the thermodynamic predictions may be achieved. These results also indicate that microstructural

Acknowledgements

The authors would like to thank K.F. Russell and J.J. Haugh for their assistance. Research at the Oak Ridge National Laboratory SHaRE User Facility was sponsored by the Division of Materials Sciences and Engineering, U.S. Department of Energy, under contract DE-AC05-00OR22725 with UT-Battelle, LLC. The thermodynamic predictions were computed with the use of the NiFe database, a thermodynamic database for the calculation of phase equilibria in multicomponent alloys, N. Saunders, ThermoTech

References (10)

  • B. Sundman et al.

    Calphad

    (1985)
  • M.K. Miller et al.

    Mater. Sci. Eng.

    (1999)
  • M.K. Miller

    Atom Probe Tomography

    (2000)
  • E.A. Loria (Ed.), Proceedings of the Superalloy 718, The Minerals, Metals and Materials Society, Warrendale, PA, USA,...
  • M.G. Burke et al.

    J. Phys.

    (1989)
There are more references available in the full text version of this article.

Cited by (48)

  • Influence behavior and mechanism of γ′- or γ′′-precipitated phase types on damage resistance of heterogeneous interface in nickel-based superalloys

    2022, Journal of Alloys and Compounds
    Citation Excerpt :

    The anti-phase boundary energy of the γ′ or γ′′ precipitated phases hinders the movement of dislocations, resulting in the excellent properties of Ni-based superalloys [6,7]. Miller [8] reported that the precipitated phases in Ni-based superalloys are mainly γ′-Ni3(Al, Ti) (L12 crystal structure) and γ′′-Ni3(Nb, Al, Ti) (D022 crystal structure). Additionally, γ′-Ni3Al and γ′-Ni3Ti with L12 crystal structure are more stable than those with D022 crystal structure [9].

  • The effect of composition on microstructure and properties for additively manufactured superalloy 718

    2021, Journal of Alloys and Compounds
    Citation Excerpt :

    Another possible cause of this effect again may be due to the formation of the Ti and Nb-rich MC carbides. Both Ti and Nb are important alloying elements for the formation of the strengthening γ′ (Ni3Al,Ti) and γ′′ (Ni3Nb) phases, respectively [46–48]. Already, some amount of Ti was consumed to form detrimental nitrides in these alloys [34].

  • Microstructure-property relationships in directly aged Alloy 718 turbine disks

    2020, Materials Science and Engineering: A
    Citation Excerpt :

    These local variations in Nb may originate from coarsened γ"-precipitates not being captured in the APM datasets in Fig. 7. While local fluctuations in the Nb concentration are obvious due to the γ"-precipitate stoichiometry Ni3(Nb), partial solubility of Al in γ"-precipitates as found by Miller et al. [32] may has the same effect. The increase in Mo is an APM artefact due to the proximity of low-intensity Mo3+ peaks to stronger Ni2+ and Nb3+ peaks [33,34].

View all citing articles on Scopus
1

Present address: Bechtel Bettis Inc., West Mifflin, PA 15122, USA.

View full text