Advertisement

Influence of Minor Grain Boundary Elements on the Solidification Behavior of a Re-containing Single-Crystal Superalloy

  • Jian ZhangEmail author
  • Jingxuan Zhao
  • Xiaotie Zhang
  • Yan Yang
  • Hao Chen
  • Hua Jiang
  • Li She
Conference paper
Part of the Springer Proceedings in Physics book series (SPPHY, volume 217)

Abstract

The re-containing single-crystal superalloys with different hafnium and boron additions were fabricated, and their microstructures and segregation behavior were investigated. The results indicated that the addition of hafnium and boron increase the volume fraction of γ/γ′ eutectic, and decrease the volume fraction of micropore. The solidification segregation ratio of Re, W, Mo, Al, and Ta obviously varied with hafnium and boron modified alloy. In addition, individual or simultaneous hafnium and boron additions obviously decrease the liquidus and solidus temperature as well as the incipient melting point.

Keywords

Hafnium Boron Re-containing superalloy Microstructure Segregation 

1 Introduction

Nickel-based single-crystal (SX) superalloys are widely used materials for turbine blades due to their balanced creep and fatigue properties at high temperatures (900–1100 °C). With the development of large complex single-crystal blades, some new problems have been discovered [1]. One problem was low angle grain boundaries (LABs) defects, which could lead to poor qualified rates. The solution to this trouble was to add carbon, boron, and hafnium to improve grain boundary strength [2].

In single-crystal superalloy, the elements carbon, boron, and hafnium have generally been excluded, since they lowered incipient melting temperatures. Some work reported that the as-cast microstructures and solidification segregation were also changed by these elements. However, previous work focused on the creep properties of the superalloy with minor addition of carbon, hafnium and boron [3, 4], or individual carbon addition [5, 6, 7]. However, hardly any research has been carried out on the individual and interactive influence of hafnium and boron additions on solidification characteristics and segregation behavior of single-crystal superalloys. Therefore, in the present paper, the single-crystal superalloy with and without hafnium and boron was fabricated. The as-cast microstructure, phase transformation temperature, and solidification segregation behavior were studied to clarify the influence of minor addition hafnium and boron additions.

2 Experiment Procedures

Four experimental Re-containing single-crystal superalloys were directionally solidified (HRS) using Bridgman technique, and the compositions were listed in Table 1. The compositions of four alloys are nominally identical with the only difference of the hafnium and boron content. The alloy without any intentional alloying additions was named alloy A, and the alloys B, C ,and D were the modified alloys, respectively.
Table 1

Nominal compositions of four experimental alloys (wt%)

Alloy

Ni

Ta

W

Re

Al

Cr

Co

Mo

C

Hf

B

A

Bal.

7.0

7.0

3.0

6.0

4.0

8.0

2.0

0.01

0

0

B

Bal.

7.0

7.0

3.0

6.0

4.0

8.0

2.0

0.01

0.40

0

C

Bal.

7.0

7.0

3.0

6.0

4.0

8.0

2.0

0.01

0

0.015

D

Bal.

7.0

7.0

3.0

6.0

4.0

8.0

2.0

0.01

0.40

0.015

Transverse samples used for microstructural observation were cut from the same section of SX superalloys. Samples were etched using a solution of 25% HNO3, 50% HCl, and 25% H2O in volume ratio. A scanning electron microscope (SEM) was used to observe microstructural change. The selective area diffraction (SAD) analyses were also conducted using transmission electron microscope (TEM). Differential scanning calorimetry (DSC) experiments were performed to analyze the incipient melting (TE), solidus (Ts) and liquidus (TL). Cylindrical samples (about 60–90 mg) were utilized with a rate of 10 °C min−1. Endothermic peak characterization of incipient melting (TE) was confirmed by metallographic methods. The γ/γ′ eutectic, micropore volume fraction was measured in optical images (OM) by the image analysis software. The volume fractions of secondary phase, such as carbides and borides were also determined using image analysis software using backscattered SEM images at a magnification of 2000×. A JEOL JXA-8100 electron probe microanalyser (EPMA) was carried out to analyze the dendritic segregation characteristics of the main elements. The more detailed procedures of this technique have been presented by Tin et al. [8], Scheil [9].

3 Results and Discussion

3.1 Microstructures of As-Cast Alloys

Due to little difference of the composition of the experimental single-crystal alloys, they almost exhibit the same microstructure. Figure 1 exhibit the OM images of the four experimental superalloys, respectively. In these four alloys, PDAS (Primary dendritic arm spacing) was similar, in the range of 350–400 μm. The volume fraction of the coarse γ/γ′ eutectic in alloy A is 5%, but the addition of hafnium and boron increases the volume fraction of the eutectic to 7% in alloy B (0.4 wt% hafnium), 8% in alloy C (0.015 wt% boron) and 13% in alloy D (both 0.4 wt% hafnium and 0.015 wt% boron). Clearly, the individual addition of hafnium or boron almost has the similar effect on the volume fraction of the eutectic.
Fig. 1

Typical optical images of as-cast microstructures of alloys A (a), B (b), C (c) and D (d)

Figure 2 illustrates the SEM backscattered (BSE) images of the four alloys, clearly showing the morphology and distribution of carbides, borides, and micropore. Near the eutectic, carbide phase in white contrast and micropore were both distributed in alloys A and B, as shown in Fig. 2a, b. Carbide phase in white contrast and boride phases in gray contrast were both distributed near the eutectic in alloys C and D, as shown in Fig. 2c, d. TEM image and SAD pattern of as-cast borides in alloy D are illustrated in Fig. 2e, f, respectively. The analysis on volume fraction of carbide exhibited that it was increased from 0.08 to 0.15% with 0.4 wt% hafnium addition, but not changed in alloys C and D with boron addition. EDS analysis indicates that MC type carbide was rich in Ta and Nb element in original alloy A, and the MC carbide was rich in Hf and Ta with addition of 0.4 wt% hafnium in alloy B and alloy D. Meanwhile, about 0.33% and 0.32% of the boride was found in alloys C and D with 0.015 wt% boron. According to the EPMA analysis of boride phases (Fig. 2f) illustrated in Table 2, the Re, W, and Mo content in this phase was very high. During the solidification of superalloys, the formation of γ/γ′ eutectic in interdendritic regions would result in an remarkable increase of the contents of Re, Mo, Cr, Co, and W in the last residual liquid, because these elements have relatively low solubility in γ′ phase, resulting in low concentrations in γ/γ′ eutectic as γ′ phase is the main phase inside γ/γ′ eutectic. Meanwhile, the boride phase is also formed during the growth of the γ/γ′ eutectic. Therefore, the newly formed boride is rich of elements Re, Cr, Mo, Co, W, and B. The formation of these large size phases may result in premature creep rupture in high-temperature and high-stress turbine applications, due to the depletion of strengthening elements from matrix and their intrinsic brittleness. In addition, no Ni5Hf phase was found in Hf modified alloys B and D [10].
Fig. 2

Microstructures of the as-cast alloy A (a), B (b), C (c), D (d), and TEM image (e) and SAD pattern (f) of borides of the M3B2 boride in alloy D

Table 2

The main compositions of borides in alloy D in weight percent using EPMA

Elements

Al

Ta

Cr

Mo

W

Re

Co

Ni

Weight percent (wt%)

0.3

5.1

7.6

14.6

22.5

23.7

9.1

17.1

In single-crystal superalloys, micropore will promote formation of the creep and fatigue cavity during high-temperature (900–1100 °C) application service and obviously decreases mechanical properties [11, 12]. The micropores were mainly distributed in interdendritic area and near the eutectic, as shown in Fig. 2a. A high level of volume fraction micropores of 0.22 and 0.18% was observed in alloys A and B with the size of about 5–10 μm, respectively. The micropores volume fraction decreased obviously to 0.05 and 0.04% with the size of about 2–5 μm in alloys C and D, due to boron addition. The result is in accordance with the result of Antony [13] and Zhao [3]. When the interdendritic liquid channels close up, the liquid flow is blocked, therefore the final eutectic liquid pools leave fine micropores at the end of solidification shrinkage. M3B2 borides are formed in the last stage of the solidification [14]. Compared with γ/γ′ eutectic, the size of M3B2 is smaller and the volume is larger. Thus, the formation of M3B2 borides in the last eutectic pools will fill more space than the same amount of γ/γ′ eutectic and decrease the shrinkage in the interdendritic area. The microstructures analysis of four as-cast alloys were listed in Table 3.
Table 3

Microstructures analysis of four as-cast alloys

Alloy

PDAS/μm

As-cast γ/γ′ eutectic vol./%

Carbides vol./%

Borides vol./%

Micropores vol./%

A

359.1 ± 12.1

5 ± 1

0.08 ± 0.05

0

0.22 ± 0.05

B

358.2 ± 11.3

7 ± 2

0.15 ± 0.07

0

0.18 ± 0.03

C

368.2 ± 9.3

8 ± 2

0.07 ± 0.04

0.33 ± 0.06

0.10 ± 0.04

D

371.2 ± 7.3

13 ± 2

0.16 ± 0.04

0.32 ± 0.07

0.11 ± 0.05

As illustrated in Fig. 3, the incipient melting points (TE), solidus (TS) and liquidus (TL) temperatures of investigated alloys were achieved from DSC heating curves between 1330 and 1410 °C. The phase transformation temperatures of the four experimental alloys were listed in Table 4. The DSC results clearly indicated that individual or simultaneous additions of 0.4 wt% hafnium and 0.015 wt% boron could decrease the liquidus temperature and solidus temperature. In these four alloys, the incipient melting points were also tested by DSC to study the influence of hafnium and boron additions. In Fig. 3b, the first peak TE is the melting point of γ/γ′ eutectic. Compared with the 1355 °C in alloy A, hafnium addition resulted in little decrease of the incipient melting point to 1353 °C, however boron addition was found to exert more serious effect and decrease the incipient melting point to 1338 °C. Further, both hafnium and boron additions decreased incipient melting point to 1334 °C. The decrease of solution heat treatment temperature to avert incipient melting gives rise to incomplete solution of carbides and borides as well as γ/γ′ eutectic, which has deleterious effects on the mechanical properties.
Fig. 3

The DSC heating curves of liquidus, solidus and incipient melting temperatures of four as-cast alloys. a Overall curves of four as-cast alloys, b partial enlarged curves of alloy D

Table 4

The incipient melting points (TE), solidus (TS) and liquidus (TL) temperatures of investigated alloys

Alloy

TE/°C

TS/°C

TL/°C

(TL − TS)/°C

Base

1355

1369

1407

38

B

1353

1362

1402

40

C

1338

1359

1395

36

D

1334

1358

1392

34

3.2 As-Cast Segregation

During the directionally solidified (HRS) of ten or more components Ni-based single-crystal superalloys, γ′ phase forming elements (Al, Ta, Ti and Hf) and γ phase forming elements (W, Re and Co) show a preference of segregation into interdendritic regions and dendrite cores, respectively. In this study, measured difference in the segregation characteristics for W, Re, Mo, Al, and Ta due to the minor hafnium and boron additions were quantified. It can be seen that Re, W, Al, and Ta segregate more serious with the addition of 0.4% hafnium and then the segregation content also increases when the content of boron was 0.015%. In alloy D with simultaneous hafnium and boron additions, compared to the alloy A, segregation coefficient of Re and Ta increased, and the segregation coefficient of W and Al changed not obviously. The segregation coefficient of Mo exhibited no obvious tendency.

The strong tendency of refractory alloying elements to partition to dendrite cores, providing a driving force for the TCP phases (μ and P phases) after long time exposure at elevated temperature [15, 16]. The existence of the γ/γ′ eutectic within the as-cast microstructure demonstrates that serious dendritic segregation occurred during solidification. In this study, the decrease in the volume fraction of the γ/γ′ eutectic revealed that hafnium addition enhanced the overall degree of solidification segregation. The results are well consistent with the segregation ratios tested for Hf modified alloys. The results in Fig. 4 show that hafnium addition in alloys B and D increased the segregation tendency of Re and W during solidification. However, boron addition increased the volume fraction of γ/γ′ eutectic but decreased the solidification segregation ratio of Re and W. These results are consistent with previous reports [17].
Fig. 4

Comparsion of dendritic segregation for the four experimental alloys

0.4 wt% Hf addition increases the segregation, the volume fraction of γ/γ′ eutectic and MC carbides, but affects slightly on the incipient melting point, allowing the complete solution of γ/γ′ eutectic and will not decrease the high-temperature creep properties because of its effect on strengthening the low angle grain boundary (LAB) defects in Re-contained single-crystal superalloy.

0.015 wt% boron addition decreases the percentage of microporsity, which is good for the improvement of high-temperature creep properties. However, it decreases the incipient melting point in some content, restraining the complete solution of as-cast γ/γ′ eutectic and will decrease the high-temperature mechanical properties. The M3B2 boride also consumes some of Re, W, Mo, and Cr. Nonetheless, Re, W, Mo and Cr are all important strengthening element in Ni-based superalloy, so it may obviously alleviate the strengthening effect in superalloy. In the meantime, the large size M3B2 boride will become the origination site of creep fracture in high-temperature tests.

4 Conclusions

  1. (1)

    The volume fraction of γ/γ′ eutectic and MC carbides in as-cast Ni-based superalloys increase with the addition of 0.4 wt% hafnium. The segregation extent of refractory alloying elements W and Re during solidification increase with the hafnium addition.

     
  2. (2)

    Introducing 0.015 wt% boron enhanced the volume fraction of as-cast eutectic, and resulted in the M3B2 boride formation, thus lowing the micropore amount. The extent of Re and W segregation during solidification also obviously varies with boron addition.

     
  3. (3)

    Individual or simultaneous hafnium and boron additions obviously decrease the liquidus and solidus temperature and the incipient melting point.

     
  4. (4)

    M3B2 borides were found to rich in strengthening element Re, W, Mo, and Cr in this Re-containing superalloy, which is detrimental to the strengthening effect.

     

References

  1. 1.
    B.B. Seth, Superalloys 2000 (Minerals, Metals and Materials Society/AIME, USA, 2000), pp. 3–16Google Scholar
  2. 2.
    D.M. Shah, A. Cetel, Superalloys 2000, Champion, PA (2000), pp. 295–300Google Scholar
  3. 3.
    Y.S. Zhao, C.G. Liu, Y.Y. Guo, Y.F. Guo, Prog. Nat. Sci. Mater. Int. 28, 483–488 (2018)CrossRefGoogle Scholar
  4. 4.
    Q.Z. Chen, N. Jones, D.M. Knowles, Acta Mater. 50, 1095–1112 (2002)CrossRefGoogle Scholar
  5. 5.
    K. Al-Jarba, G. Fuchs, JOM 56, 50–55 (2004)CrossRefGoogle Scholar
  6. 6.
    L.R. Liu, T. Jin, N. Zhao, Z. Wang, X. Sun, H. Guan, Z. Hu, Mater. Lett. 58, 2290–2294 (2004)CrossRefGoogle Scholar
  7. 7.
    L.R. Liu, T. Jin, N.R. Zhao, Z.H. Wang, X.F. Sun, H.R. Guan, Z.Q. Hu, Mater. Sci. Eng., A 385, 105–112 (2004)CrossRefGoogle Scholar
  8. 8.
    S. Tin, T. Pollock, W. Murphy, Metall. Mater. Trans. A. 32, 1743–1753 (2001)CrossRefGoogle Scholar
  9. 9.
    E. Scheil, Metallforschun 2, 7 (1947)Google Scholar
  10. 10.
    L.Y. Sheng, F. Yang, J.T. Guo, X.I. Ting-Fei, Trans. Nonferr. Metals Soc. China 24, 673–681 (2014)CrossRefGoogle Scholar
  11. 11.
    L.G. Fritzemeier, Superalloys 1988, 265–274 (1988)Google Scholar
  12. 12.
    A.A. Hopgood, J.W. Martin, Mater. Sci. Eng. 82, 27–36 (1986)CrossRefGoogle Scholar
  13. 13.
    K. Antony, J. Radavich, in Proceedings of the Third International Symposium (Claitor’s Publishing, 1976Google Scholar
  14. 14.
    P. Zhou, J. Yu, X. Sun, H. Guan, Z. Hu, Mater. Sci. Eng. A 491, 159–163 (2008)CrossRefGoogle Scholar
  15. 15.
    S. Tin, T.M. Pollock, Mater. Sci. Eng. A 348, 111–121 (2003)CrossRefGoogle Scholar
  16. 16.
    W.S. Walston, J.C. Schaeffer, W.H. Murphy, in Superalloys 1996, Minerals, ed. by R.D. Kissinger, D.J. Deye, D.L. Anton, A.D. Cetel, M.V. Nathal, T.M. Pollock, D.A. Woodford (Metals Mater. Soc., Warrendale, 1996), pp. 9–18Google Scholar
  17. 17.
    R.M. Kearsey, J.C. Beddoes, K.M. Jaansalu, W.T. Thompson, P. Au, in K.A. Green, T.M.Google Scholar

Copyright information

© Springer Nature Singapore Pte Ltd. 2019

Authors and Affiliations

  • Jian Zhang
    • 1
    Email author
  • Jingxuan Zhao
    • 1
  • Xiaotie Zhang
    • 1
  • Yan Yang
    • 1
  • Hao Chen
    • 1
  • Hua Jiang
    • 1
  • Li She
    • 1
  1. 1.Science and Technology on Advanced High Temperature Structural Materials LaboratoryBeijing Institute of Aeronautical MaterialsBeijingChina

Personalised recommendations