Wrought iron–nickel base superalloys play an important role in high-temperature applications. One of the most popular austenitic heat-resistant superalloy is Incoloy 901 which has widespread uses for parts such as turbine disks, shafts and rotors and requires high strength of up to 760°C and oxidation resistance of up to 870°C [1–4]. Therefore, research interest to improve the performance characteristics of Incoloy 901 is going on [5, 6]. The precipitation hardening process applied for these alloys gives rise to interesting mechanical properties. These alloys are strengthened by the presence of Ti, Al and C. The improvement of mechanical properties mainly relies on proper selection and design of the main alloy composition and/or additives and development of appropriate microstructure and phase(s) through appropriate heat-treatment conditions. It is shown that the appropriate heat treatment of this family of alloys produces an ultrafine precipitate in nanometric size of γ′ phase in a face-centered cubic (FCC) matrix .
Adding proper amount of carbon to superalloys can improve their strength. Carbon atoms occupy interstitial sites in FCC lattice and result in solution strengthening. Carbon also combines with reactive elements such as Ti to form carbide. Carbides play an important role in superalloy. The carbides act to impede deformation when they are in a grain. For high-temperature alloys, however, their most important role is to inhibit the movement of grain boundaries, which tend to slide when being stressed above about 0.5 of the absolute melting point. Carbides must be suitably dispersed along the boundaries and reasonably stable when a component is put into elevated temperature service [1–3].
Various works have reported the effects of carbide precipitating on the mechanical properties of superalloys [7–13]. By controlling and improving size, morphology and distribution of carbides in superalloys, their strength can be improved [10–12]. The addition of carbon to superalloys such as Nimonic 115 , IN 718 , CM-681 LC , IN 792  and Nimonic 80A  has been already reported to increase tensile strength, rupture life and toughness. In Nimonic 115 and Nimonic 80A alloys, increasing of carbon amounts results in considerable increase in creep and rupture life. Increasing of carbon amounts also increases ductility in CM-681 LC and IN 792 alloys. The improvement of mechanical properties of superalloys mainly relies on increasing carbide area fractions and/or mismatch between γ and γ′ phases [10–13]. However, for the IN 792, such a trend has also been followed by studying the effects of carbon on mechanical properties. Increase of carbon to more than 0.14 wt% has been reported to increase continuous carbides and large carbides and hence decrease mechanical properties.
Hale et al.  have reported the effect of carbon on the dynamic strain aging (DSA) of IN 718 superalloy. He suggests that carbon atoms diffuse to the strain fields around dislocation at lower temperatures and from the lattice diffusion of a substantial solute like Cr at higher temperatures. Also he has reported that yield point return in Inc. 718 is mainly related to the reorientation of carbon–vacancy pairs in the strain fields of dislocations.
The work presented here was therefore stimulated by considering the possible advantages of addition of carbon on microstructure and mechanical properties of Incoloy 901 and determining a proper amount of carbon to achieve good tensile properties at high temperatures.
Incoloy 901 alloys with various amounts of carbon were produced by VIM+ESR process. The dimensions of ESR ingots were 100×80×60 mm. The chemical compositions of the alloys are shown in Table 1. ESR ingots were homogenized at 1160°C for 3 h and then were rolled to plates with 10 mm thickness at 1,150°C. These plates were heat-treated according to a three-step heat treatment schedule (1095°C/2 h/WQ+790°C/2 h/WQ+740°C/24 h/AC) .
After heat treatment, subsized rectangular tension test specimens were prepared with the gauge length of 25 mm and thickness of 5 mm according to the ASTM-E8M . Tensile tests were carried out on a computer-controlled testing machine (Instron 8502) at 650°C at a constant strain rate of 0.001 s−1.
Microstructural observations were studied by optical microscope (OM) and scanning electron microscope (SEM) equipped with energy dispersive spectroscopy (EDS). All the samples were etched in glycergia solution (15 ml HCl+10 ml glycerin+5 ml HNO3). An image analysis method was used to measure of area fractions of carbide precipitates. For each alloy, a sample was sectioned from as-heat-treated plates and prepared for the standard electrolytic extraction of carbide and γ′. The electrolytes for the extraction of carbide and γ′ were a solution of 10% HCl in methanol and 20% H3PO4 in water, respectively. Carbide and γ′ powders were then dried and collected. Both powder and bulk material samples were examined by Philips Expert pro X-ray diffractometer using Cu-Kα radiation. For both carbides and γ′, XRD data were examined using X’Pert High Score software. The lattice parameters of γ and γ′ phases were calculated using Nelson–Riley extrapolation method .
Results and discussion
Figure 1 shows an SEM micrograph of as-heat-treated microstructure of the alloy containing 0.02 wt% carbon. The microstructures of alloys with 0.05 and 0.09 wt% carbon were also the same as those of the alloy with 0.02 wt% carbon. Microstructures consisted of the austenite and secondary phases. The secondary phases were nitride, carbonitride and carbide, which precipitated in the matrix and at grain boundaries. Carbides are seen in Figure 1. Annealing twinnings, which are one of the microstructural features of Incoloy 901, are also illustrated in Figure 1. γ′ precipitates have nanometric sizes in Incoloy 901 and they could not be detected at the magnifications of SEM.
XRD patterns of the extracted powders of carbides of the alloys are demonstrated in Figure 2. The observed phases that appeared in the XRD patterns of all the alloys could be indexed as MC carbides. The peaks of other types of carbides (M23C6 and M6C) were not observed in XRD patterns of the extracted carbides. Figure 3 demonstrates SEM micrograph of carbides in alloys with 0.05 and 0.09 wt% carbon. The EDS spectra from the observed carbides in all three alloys indicated that Mo was present in carbides while XRD results confirmed that carbides were MC. (Ti, Mo)C carbide was also reported in other superalloys such as U-500, M-252 and Rene 77  and it seems that the carbide in Incoloy 901 was MC and M elements were Ti and Mo. No research about the characteristics of carbides in Incoloy 901 superalloy has been observed but, in one research on precipitations in some austenitic alloys including Incoloy 901 , only TiC was identified as carbide phase in this superalloy.
As mentioned earlier, γ′ precipitates could not be detected in SEM micrographs and their characteristics were examined by extracting them from matrix by the process described in the previous section. The mismatches between γ and γ′ phases were calculated from : (1)where aγ and aγ′ are lattice parameters of the γ and γ′ phases, respectively.
The calculated lattice parameters based on XRD results for the γ and γ′ phases and mismatch for the samples with various amounts of carbon are listed in Table 3. It can be seen that values of aγ′ were bigger than those of aγ and, according to eq. (1), the mismatch values would be positive. Table 2 shows a decrease in both aγ and aγ′. The decrease of aγ is bigger than that of aγ′. So, the mismatch between γ and γ′ phases increased from 0.119 for alloy with 0.02 wt% carbon to 0.139 for alloy with 0.05 wt% carbon and 0.167 for alloy with 0.09 wt% carbon. The amounts of Ti for alloys are also presented in Table 1; it can be seen that the change of Ti contents was negligible and variations of lattice parameters and mismatch can be attributed to the variation of carbon amounts. The observed decrease of the lattice parameter with the increase of carbon was possibly due to the shorter radius of carbon atoms compared with that of the Ti atoms . Ti as a solute atom results in the expansion of FCC lattice of γ matrix. When carbon is added to alloy to form TiC carbides, titanium atoms exit from solubility; so, the lattice parameter of γ decreases. Also, for γ′ phase, the atomic radius of Ti is slightly larger than that of Al and, therefore, decreases of the amount of Ti in γ′ phase results in the decrease of aγ′.
Table 3 displays variations of the area fraction and average length of carbides with the increasing amount of carbon. As can be seen, both parameters increased with the increase of carbon. The area fraction of carbide increased from 0.73% to 1.9% when the amount of carbon changed from 0.02 to 0.09 wt%. The average carbide length also increased with the increase of carbon content from 3.4% to 4.3%.
The tensile properties of Incoloy 901 at 650°C were examined. Figure 4 shows the change of 650°C tensile properties of Incoloy 901 alloy with carbon content. The yield strength (YS) and ultimate tensile strength (UTS) of the alloy with 0.02 wt% carbon were 519 and 915 MPa, respectively. With the increase of carbon content to 0.05 wt%, the strength of the alloy increased and maximum YS and UTS were achieved as 568 and 978 MPa, respectively. With the increase of carbon content to 0.09 wt%, the strength of the alloy decreased. Also, ductility decreased slightly with the increase of carbon content. The values of elongation and reduction of area for the alloy with the maximum strength (alloy with 0.05 wt% carbon) were 25 and 26 pct, respectively.
Carbides have two strengthening effects in superalloys. First, when they are in grains, they act to impede dislocation movement and cause precipitation strengthening. Second, carbides precipitated at grain boundaries increase grain boundaries’ strength and inhibit their sliding particularly at high temperatures [11, 12]. Another important factor that determines the high temperature strength, especially rupture strength, of nickel and iron–nickel base superalloys is the degree of mismatch between γ and γ′ phases [1, 2]. Increase of mismatch between γ and γ′ phases increases coherent strain around the γ′ phase. This coherency strain imparts in strengthening and in particular affects the YS at high temperatures. For the alloy containing 0.02 wt% carbon, the area fraction of carbide was 0.73% and this amount of carbides was low in order to cause sufficient strengthening both in grains and at grain boundaries, as shown in Figure 5(a). As a result of increasing carbide area fraction (Table 3) and mismatch between γ and γ′ phases (Table 2), strength increased from the alloy containing 0.02 wt% carbon to the alloy containing 0.05 wt% carbon and then decreased for 0.09 wt% carbon containing alloy. Figure 5(b) shows the carbides in grains and discontinuous carbide precipitation at grain boundaries. This distribution is favorable for causing good strengths. The maximum YS and UTS (563 and 973 MPa, respectively) at 650°C were observed for the alloy containing 0.05 wt% carbon. Previous studies  have reported the decrease of grain boundary strength and formation of cavity at the interface between γ′+γ eutectic and γ′ phase and γ matrix in In792 alloy.
The sizes of carbides are another factor affecting strength of the alloy. When carbon content increased to 0.09 wt%, the average carbide length increased, as shown in Table 3. The elongated film-shaped carbides precipitated at grain boundaries and large blocky carbide dispersed in matrix decreases the strength of alloy because the interface of MC and matrix is incoherent and big blocky carbides are crack initiation sites. Furthermore, big blocky carbides may be cracked and led to the rapid formation and growth of crack . In alloy containing 0.09 wt% C, precipitating elongated carbides at grain boundaries, as shown in Figure 5(c), led to the reduction in strength; these carbides were easy crack expansion routes and caused early failure of alloy.
Microstructural examinations of Incoloy 901 superalloy in this study only revealed the presence of MC carbides. When carbon content was changed from 0.02 wt% to 0.05 and 0.09 wt%, the area fraction of these carbides varied from 0.73% to 0.90% and 1.3%, respectively.
Increase of carbon content gave rise to the higher amount of carbide sizes. The average carbide length increased from 3.4 µm for the alloy containing 0.02 wt% carbon to 3.8 µm for the alloy containing 0.05 wt% carbon and 4.3 µm for the alloy containing 0.09 wt% carbon.
Carbon content also affected the lattice parameters of γ and γ′ phases. With the increase of carbon content, the lattice parameters of γ and γ′ phases decreased and the mismatch between them slightly increased.
The maximum strength of Incoloy 901 superalloy was achieved at 0.05 wt% carbon. For the alloy with this level of carbon, YS and UTS were 570 and 980 MPa, respectively. Despite the increase of strength, ductility decreased slightly with the increase of carbon content.
Sims CT, Stoloff NS, Hagel WC. Superalloys II. London: Applied Science Publishers, 1981. Google Scholar
Donachie MJ, Donachie SJ. Superalloys: a technical guide. Materials Park, Ohio, USA: ASM, 2002. Google Scholar
Reed RC. The superalloys fundamentals and applications. New York, NY: Cambridge University Press, 2006. Google Scholar
Davis, JR, et al., editor. ASM specialty handbook: nickel, cobalt and their alloys. Materials Park, OH: ASM, 2000:68–91. Google Scholar
Momeni A, Abbasi SM, Morakabati M, Badri H, Wang X. Mater Sci Eng A 2014;615:51–60. Google Scholar
Tavakkoli MM, Abbasi SM. Mater Des 2013;46:573–8. Google Scholar
Furillo FT, Davidson JM, Tien JK. Mater Sci Eng 1979;39:267–73. Google Scholar
Xiao L, Chaturvedi MC. J Mater Eng Perform 2005;14:528–38. Google Scholar
Wei CN, Bor HY, Chang L. Mater Sci Eng A 2010;527:3741–74. Google Scholar
He LZ, Zheng Q, Sun XF, Hou GC, Guang HR, Hu ZQ. J Mater Sci 2005;40:2959–64. Google Scholar
Wei CN, Bor HY, Chang L. J Alloys Compd 2011;509:5708–14. Google Scholar
Yang J, Zheng Q, Jie M, Sun X, Hu Z. Mater Sci Eng A 2011;528:1534–9. Google Scholar
Xu Y, Jin Q, Xiao X, Cao X, Jia G, Zhu Y, et al. Mater Sci Eng A 2011;528:4600–7. Google Scholar
Hale CL, Rollings WS, Weaver ML. Mater Sci Eng A 2001;300:153–64. Google Scholar
ASTM E-8M. Annual book of ASTM standards, vol. 3.01. Philadelphia, PA: American Society for Testing and Materials, 2004:62.Google Scholar
Suryanarayana C. X-ray diffraction: a practical approach. New York, NY: Plenum Press, 1998:157–61. Google Scholar
Beattie Jr HJ, Hagel WC. Trans Metall Soc AIME 1961;221:28–35. Google Scholar
About the article
Published Online: 2015-02-06
Published in Print: 2015-12-01