Refractory Alloys: Vanadium, Niobium, Molybdenum, Tungsten

CHAPTER

REFRACTORY ALLOYS: VANADIUM, NIOBIUM, MOLYBDENUM, TUNGSTEN

13

Lance L. Snead*, David T. Hoelzer†, Michael Rieth‡, Andre A.N. Nemith‡ Department of Materials Science and Chemical Engineering, State University of New York at Stony Brook, Stony Brook, NY, United States* Materials Science and Technology Division, Oak Ridge National Laboratory, Oak Ridge, TN, United States† Institute for Advanced Materials, Karlsruhe Institute of Technology, Karlsruhe, Germany‡

CHAPTER OUTLINE 13.1 Introduction ........................................................................................................................................585 13.2 Practical Routes for Refractory Alloy Production ...................................................................................587 13.2.1 Vanadium .................................................................................................................... 587 13.2.2 Niobium ...................................................................................................................... 590 13.2.3 Fabrication of Nuclear-Grade Molybdenum ..................................................................... 591 13.2.4 Practical Routes of Tungsten and Tungsten Alloy Production ............................................ 596 13.3 As-Fabricated Mechanical Properties ...................................................................................................600 13.3.1 Vanadium .................................................................................................................... 600 13.3.2 As-Fabricated Mechanical Properties of Niobium ............................................................. 606 13.3.3 As-Fabricated Mechanical Properties of Molybdenum ...................................................... 607 13.3.4 As-Fabricated Mechanical Properties of Tungsten ............................................................ 611 13.4 As-Irradiated Mechanical Properties ....................................................................................................615 13.4.1 As-Irradiated Mechanical Properties of Vanadium ............................................................ 615 13.4.2 As-Irradiated Mechanical Properties of Niobium .............................................................. 622 13.4.3 As-Irradiated Mechanical Properties of Molybdenum ........................................................ 625 13.4.4 As-Irradiated Mechanical Properties of Tungsten ............................................................. 629 13.4.5 Summary and Conclusions ............................................................................................ 634 References ..................................................................................................................................................635

13.1 INTRODUCTION Refractory metals as a class of materials are understood to share the common properties of very high melting temperature and mechanical properties and wear resistance. A narrowly defined class of refractory metals would include metals with melting points >2000°C: niobium, chromium molybdenum, tantalum, tungsten, and rhenium [1], while a wider class would also include those with melting points above 1850°C: vanadium, hafnium, titanium, Structural Alloys for Nuclear Energy Applications. https://doi.org/10.1016/B978-0-12-397046-6.00013-7 Copyright # 2019 Elsevier Inc. All rights reserved.

585

586

CHAPTER 13 REFRACTORY ALLOYS

zirconium, ruthenium, osmium, rhodium, and iridium. The current practical application of refractory metals is relatively widespread (though arguably for specialty application) with examples being casting molds, wire filaments, reactant vessels for corrosive materials, hard tooling, and a myriad of applications where high density is desired. This chapter focuses on a subset of refractory materials that are of particular interest for current and next-generation nuclear applications, specifically vanadium, niobium, molybdenum, and tungsten. It is noted that tantalum has historically been studied for space reactor applications, though is not discussed here, in part due to its similarity in behavior to niobium. A snapshot of how their physical properties compare is given in Tables 13.1 and 13.2. This chapter is supportive of content to Chapter 2 (Generation IV nuclear reactors), Chapter 3 (Fusion Reactors), and Chapter 5 (Radiation Effects and Thermomechanical Degradation Processes). Historically, refractory alloys have been considered in a wide array of nuclear applications including terrestrial, maritime, and space fission reactors as well as current and next-generation fusion reactors. Because refractory metals are a class of materials possessing extraordinary high-temperature properties, they are perennial contenders for high-temperature nuclear applications. However, their use to date has been limited, due in part to the difficulty in fabricating high-performance refractory parts, and their environmental degradation

Table 13.1 Summary of Physical and Chemical Properties of Selected Refractory Metals Vanadium

Niobium

Molybdenum

Tungsten

Atomic number Atomic volume in cm3/mol Average atomic mass g/mol Naturally occurring isotopes in %

23 8.78

41 10.87

42 9.4

74 9.53

50.9415

93.2

95.94

183.85

50

93 (
100%)

Pauling electronegativity Electronic shell Oxidation states

1.63

1.6

92(14.84%); 95(9.25%); 96(16.68%); 97(9.55%); 98 (24.13%); 100 (9.63%) 2.16

180(0.135%); 182(26.4%); 183(14.4%); 184(30.6%); 186(28.4%) 2.36

3d3, 4s2 5, 4, 3, 2, 0

[Kr] 5s1 4d5 6, 5, 4, 3, 2, 1, -1, -2 (strongly acidic oxide)

[Xe] 4f14 5d4 6s2 2, 1, 0, +2, +3, +4, +5, +6

0.452

4d4 5s1 5, 4, 3, 2, 1, 1, 3 (a mildly acidic oxide) 689.9 kJ/mol

598

824

21.5

30

37.48

35.4

490

265

24.06

0.13

19.7

0.00152

0.0534

5.5

0.307

53.7

138

1.74

1890 3407 5.8

2477 4744 8.57

2623 4639 10.28

3410 5660 19.3

Heat of vaporization in kJ/mol Heat of fusion in kJ/mol Specific heat in J/gK Electrical resistivity in μΩcm (@ 20°C) Thermal conductivity in W/cmK (@ 20°C) Melting point in °C Boiling point in °C Density in g/cm3 (@ 20°C)

V(0.25%), V(99.75%)

51

13.2 PRACTICAL ROUTES FOR REFRACTORY ALLOY PRODUCTION

587

Table 13.2 Typical Values for Single-Crystal Elastic Constants at Room Temperature

Young’s modulus—GPa Shear modulus—GPa Bulk modulus—GPa Poisson’s ratio

Vanadium

Niobium

Molybdenum

Tungsten

128 47 160 0.37

105 38 170 0.40

329 126 230 0.31

390–410 156–177 305–310 0.280–0.30

including irradiation effects. The following sections will discuss the current processing routes being taken to produce nuclear-grade refractory alloys, a general discussion of their properties, and the effects of irradiation on the materials.

13.2 PRACTICAL ROUTES FOR REFRACTORY ALLOY PRODUCTION 13.2.1 VANADIUM Vanadium-base alloys emerged as a candidate material for nuclear energy applications during the 1960s primarily due to the low neutron-induced activation characteristics (particularly for fast neutron spectra), reasonable hightemperature mechanical properties, and high thermal stress factors mainly due to high thermal conductivity [2–5]. The early development of vanadium-base alloys was for fuel cladding for space and liquid metal fast breeder reactors (LMFBR’s) during the 1960s in the United States [6] and related efforts in Europe [7]. The investigations conducted in this time frame led to the early awareness of physical metallurgy issues and the development of welldefined processing steps for producing small experimental heats (<10 kg) of vanadium-base alloys. Significant interest for the development of vanadium-base alloys grew in the US fusion power program (FPP) during the 1980s for first wall and blanket structures in fusion energy reactors, such as the international thermonuclear experimental reactor (ITER). Numerous laboratory-scale heats (<30 kg) of vanadium-base alloys were produced during the early phase of this program, mainly from the V, V-Ti, V-Cr, V-Cr-Ti, and V-Ti-Si systems with Ti and/or Cr concentrations within the range of
3–15 wt% and Si levels <1 wt%. The goal of the research on these heats was to determine the role that processing and alloy composition had on the physical and mechanical properties and the microstructure of the vanadium alloys [6, 8, 9]. A substantial amount of research was conducted on these alloys before and after irradiation and the results obtained from the investigations identified a V-4Cr-4Ti alloy as the US reference alloy for subsequent development in the United States [8, 9]. This alloy composition was subsequently selected in similar fusion energy R&D efforts in countries such as Japan, Russia, and China. A common goal shared with each of the international countries with active R&D programs on vanadium alloys was to obtain the processing experience and physical metallurgy knowledge for producing large industrial-scale heats of the V-4Cr-4Ti alloy with better chemical homogeneity and reduced impurity levels and to establish a comprehensive database of physical and mechanical properties before and after fast neutron irradiations and in lithium environments. During the 1990s, significant progress in the development of vanadium-base alloys occurred in the US fusion program led by a joint campaign between Argonne National Laboratory and Teledyne Wah Chang (Albany, Oregon). The fabrication steps are outlined in Fig. 13.1. This effort resulted in the first productions of industrial-scale heats of the V-4Cr-4Ti alloy; first, a 500 kg heat (#832665) that was used mainly for gaining valuable experience and knowledge in production, testing, and microstructural characterization studies and, second, a large 1200 kg heat (#832864) that was selected by General Atomics for the manufacture of the Radiative Divertor for the DOE DIII-D tokamak. Much of the experience for producing the large V-4Cr-4Ti heats was obtained from previous studies of fabrication, mechanical properties testing, and microstructural examination of laboratory scale heats (<30 kg) of V-(4-5)Cr-(3-5)Ti alloys [5, 6, 8, 9]. The specification for the chemical composition of the large heats of V-4Cr-4Ti called for careful control of impurities since numerous studies had shown that the mechanical properties, such as

588

CHAPTER 13 REFRACTORY ALLOYS Machined ingot clad in evacuated SS jacket Extruded to rectangular billet (63 mm x 200 mm) at 1130–1150°C Warm-rolled (~400°C) perpendicular to extrusion direction to 23 mm thick Annealed at ~1150°C, 2 h at 10–5 torr Warm-rolled parallel to extruded direction to 12.7 mm thick Annealed at ~1050°C, 2 h at 10–5 torr Warm-rolled parallel to extruded direction to 6.35 mm thick Annealed at ~1050°C, 2 h at 10–5 torr Warm-rolled parallel to extruded direction to 3.81 mm thick

FIG. 13.1 Illustration of the fabrication schedule for production of plates from V-4Cr-4Ti cast ingots developed by Teledyne Wah Chang. Developed by Allegheny Technologies Inc., Specialty Alloys and Components. Used with permission from Allegheny Technologies Inc.

tensile strength and ductility and ductile-to-brittle transition temperature (DBTT), were very sensitive to the impurity content and most notably to the interstitial O, C, and N concentrations [10]. Particular attention was given to optimization of the Si content of 400–1000 wppm to suppress swelling during neutron irradiation; minimization of Nb, Mo, and Ag to achieve low neutron activation; and limitation of other impurities (O, C, N, S, P, Ca, Na, etc.) to minimize grain boundary segregation and precipitation of second phase precipitates responsible for causing embrittlement of the V-based alloys. For production of the large heats of V-4Cr-4Ti, V ingots were first processed by electron beam melting that were then machined into large chips and consolidated with high-purity Cr and Ti chips by double vacuum arc melting. For the V stock, a purity level of 99.95 wt% was necessary and was typically obtained by an aluminothermic reduction method using high-purity V2O5 and additional electro-refining steps [2, 11]. Due to the high chemical reactivity of Ti, the recommendation was to use double- or triple-arc melting to achieve high-purity Ti stock. The fabrication practice developed for producing plates of V-4Cr-4Ti from the cast ingot is illustrated in Fig. 13.1. Once the cast ingot was produced, it was canned in stainless steel, heated for one or more hours at
1150°C, and then extruded to form rectangular-shaped billets. The billets were then decanned and subjected to a series of heat treatments and rolling operations to form plates and sheets. It was a common practice to perform a final rolling to <50% thickness reduction followed by annealing at 1050°C for 2 h in a high-vacuum furnace. The chemical analysis performed on both large heats of V-4Cr-4Ti after extrusion to the rectangular-shaped billet showed no additional pickup of impurities from the cast ingot indicating that the production and fabrication procedures were overall successful. Ensuing characterization studies of the large V-4Cr-4Ti heats produced in the United States led to many advances in the understanding of the physical metallurgy of V-based alloys. In many sets of specimens prepared from the 500 kg V-4Cr-4Ti for mechanical properties evaluations, inhomogeneous microstructures appearing as alternating bands of coarse- and fine-grain regions were reported [12, 13] as shown in Fig. 13.2. Detailed microstructural analysis of the banded grain-size regions showed that higher concentrations of 0.1–0.3 μm size precipitates of the Ti(OCN) phase were associated with bands of finer grains aligned along the rolling direction of plates [13]. It was determined that the Ti(CON) precipitates would form any time the temperature decreased below the solvus during fabrication of plates from the cast ingot of V-4Cr-4Ti. This temperature was suggested to be
1125°C [12, 14].

13.2 PRACTICAL ROUTES FOR REFRACTORY ALLOY PRODUCTION

(A)

589

(B)

FIG. 13.2 Optical metallographic micrographs showing the (A) alternating bands of coarse- and fine-grain regions and (B) bands of high concentration of Ti(OCN) precipitates associated with the fine-grain regions. The inset is a TEM micrograph of the Ti(OCN) precipitate.

Understanding the precipitation behavior of the Ti(OCN) phase was important since studies showed that removal of interstitial O, C, and N atoms from the bcc matrix lowered the DBTT and increased the lower shelf energy in V-4Cr-4Ti [12, 14, 15]. Crystallographic investigations determined the Ti(OCN) precipitates (inset of the TEM micrograph shown in 13.2b) were consistent with the fcc Ti-x phase, where x ¼ O, C, and N (
50% atomic percent total) and each possessing the Fm3m space group, which meant that each precipitate could accommodate different stoichiometries of the interstitial elements [16]. The globular-shaped Ti(OCN) precipitates nucleated preferentially in high dislocation density regions that formed during the fabrication steps. The inhomogeneous precipitation of Ti(OCN) complicated efforts to control formation of the banded microstructures of large V4Cr-4Ti heats assince it was not easy to redistribute them in the microstructure during thermal-mechanical treatments (TMT). In one study, it was shown that cross-rolling at room temperature produced a more uniform dispersion of Ti(OCN), but these results suggested that much more needed to be learned about the kinetics of recovery, recrystallization, and precipitation in V-4Cr-4Ti before optimal processing procedures could be developed. Following the pioneering efforts of producing large V-4Cr-4Ti heats in the United States, similar efforts occurred in Japan, Russia, and China on the same reference V-4Cr-4Ti composition. In the early 2000s, several 30–160 kg ingots of NIFS-HEAT-1 and NIFS-HEAT-2 were produced in the Japanese program [17–19]. The goal of the development efforts was to reduce the total interstitial impurity (O, C, N) content of V-4Cr-4Ti to below 300 wppm, while maintaining low concentrations of Al (<200 wppm) and of Mo and Nb (<10 wppm). Detailed characterization studies were conducted on each production step to determine the most effective methods for reducing impurity pick-up for producing pure vanadium ingots. Several small heats of pure vanadium (25 kg) were successfully prepared with total interstitial concentrations (O, N, and C) of
290 wppm. Alloying techniques were also reviewed and evaluated to identify promising methods for reducing impurity pick-up. In the Japanese program, the production method that was developed resulted in total interstitial concentrations of
340 wppm for the NIFS heats of V-4Cr-4Ti [9]. The significant difference between the NIFS heats and the two US heats was the lower O concentration levels of
150–180 wppm (NIFS) compared to
330–360 wppm (US), which allowed for investigations of the oxygen effects of various physical and mechanical properties to be conducted. In Russia, a similar objective was employed for development of high-purity, chemically homogeneous V-Cr-Ti alloy ingots, mainly of the reference V-4Cr-4Ti and V-5Cr-5Ti. Ingots of 30–110 kg, that is, RF-VVC-2 and VM-DPCH-9 heats, were produced by vacuum-arc melting from 2000 to 2013 [20–22]. The investigations of the Russian heats focused

590

CHAPTER 13 REFRACTORY ALLOYS

on mainly lowering the oxygen impurity level and optimizing the TMT for improving the chemical homogeneity of solute-alloying elements, mainly Ti, and Ti(OCN) precipitate distribution to enhance the properties of the V-Cr-Ti alloys. The maximum allowed concentration of O, N, and C was specified at 200, 100, and 150 wppm, respectively, which was achieved in the production of the V-4Cr-4Ti and V-5Cr-5Ti heats. Analysis of the V-4Cr-4Ti (42 kg) and V-5Cr-5Ti (45 kg) ingots showed good macroscopic homogeneity in the Cr and Ti alloying elements, but detailed analysis by a microscopy X-ray spectral analyzer showed Ti segregation fluctuations within 2%–7% compared to the average concentration of 4.38% Ti. Subsequent TMT studies indicated that the composition variations in Ti could be minimized to lower fluctuation values. Development of V-Cr-Ti alloys in China remained at the laboratory scale until a 30 kg V-4Cr-4Ti heat (SWIP-30) was produced around 2013 [23–25]. The significant outcome of this development effort was achieving the desired limiting sum of O, C, and N impurity solute concentrations to about 400 wppm and identifying several TMT routines that resulted in high number densities of Ti(OCN] precipitates and dislocations that enhanced the high-temperature strength of V-4Cr-4Ti. The achievements obtained in the United States and international programs have demonstrated that it is feasible to fabricate V-Cr-Ti alloys in largescale heats with the present industrial bases.

13.2.2 NIOBIUM Interest arose during the 1950s to develop niobium (Nb)-based alloys for proposed aerospace and terrestrial nuclear reactor applications due to the low thermal neutron absorption cross-section for Nb, good compatibility with alkali liquid metals, acceptable high-temperature mechanical properties, and good fabricability [26–29]. Around this period, significant Nb ore deposits were discovered in the Western hemisphere that contributed to this interest [30]. Although early research focused on development of a wide range of Nb alloys, the only alloy that emerged on a commercial production status by the end of the 1950s was Nb-1Zr [31]. For most of the Nb alloys, the research focused on improving the high-temperature strength and oxidation resistance without degrading the inherently good properties of ductility and fabricability. The strategies that led to improvements in creep and oxidation resistance of Nb alloys consisted of alloying with W, Mo, Ta, and V for solid-solution strengthening and with Ti, Zr, and Hf for precipitation strengthening by reacting with interstitial O, C, and N atoms. By the early 1960s, the research identified Zr additions as essential for achieving acceptable oxidation resistance and compatibility with liquid metals such as Li. Alloying elements that promoted carbide formation, such as Zr, also emerged as the most effective method for precipitation strengthening. However, for many of the Nb alloys it became evident that improving the high-temperature strength properties reduced their fabricability, which resulted in further acceptance of Nb-1Zr for many applications due to the overarching importance of easily fabricable products. One notable derivative of Nb-1Zr was Nb-1Zr-0.1C, which was known as PWC-11. This alloy possessed higher strengths with minimum degradation of room-temperature ductility and fabricability. By the late 1960s, program shifts away from space reactor concepts resulted in significant reductions in the availability of Nb alloys and commercial producers, from over 20 alloys and 7 vendors in the mid-1960s to 5 alloys and 4 vendors by the late 1970s. Table 13.3 summarizes the commercially available Nb alloys at the end of the 1970s [32]. Resurgence in Nb alloy development occurred in the mid-1980s during the SP-100 space powder program, but effectively stopped when the program was terminated in 1994. Nb-1Zr and C-103 (Nb-10Hf-1Ti-0.7Zr) are the only alloys currently produced by commercial vendors. The early production routes of Nb alloys involved casting and powder metallurgy. However, poor control of composition specifications and extensive thermo-mechanical processing that led to unacceptable interstitial O and C contamination levels in final product forms impeded the early alloy development programs. Better control of chemical composition with lower levels of impurities occurred with changes in casting techniques in the 1960s from arc-melting to electron-beam melting of the starting ingots. Further production improvements were achieved with consumable electrode arc melting methods during the 1970s and 1980s. This is illustrated in Fig. 13.3, which shows the effect of casting method on the impact toughness of unalloyed Nb [33, 34]. The higher purity levels achieved with electron beam casting showed the lowest DBTT. The typical process flow chart for the production

13.2 PRACTICAL ROUTES FOR REFRACTORY ALLOY PRODUCTION

591

Table 13.3 Summary of Nb Base Alloys Commercially Available by the Late 1970s [32] Solute Concentration (wt. %) Alloy

Vendor

W

C-103 C-129Y FS-85 Nb-1Zr SCb-291

Fansteel, Kawecki Berylco Industries (KBI), Wah Chang Wah Chang Fansteel Fansteel, KBI, Wah Chang, NRC Fansteel, KBI, Wah Chang

Mo

Ta

Ti

Zr

Hf

Y

0.5 10 10

0.5

1

0.7

10 10

0.1

10

10

28

0.8 1

300 1

Impact strength (J)

2 200 3 5

4 100 6

0 –200

–100

100 0 Test temperature (⬚C)

200

FIG. 13.3 Effect of temperature on impact properties of cast niobium [32, 33]. Curves 1–3 are from Charpy V-notch tests of (1) electron beam, and (2 and 3) of arc-melted samples. Curves 4–6 are from modified IZOD tests of (4) electron beam, (5) arc-melted, and (6) arc-melted and cold-worked materials.

of Nb alloys is shown in Fig. 13.4 [29]. Following the initial alloy melting by an electron beam, consumable electrode arc melting was used to introduce final alloying additions to the Nb alloy. Primary ingot breakdown to improve chemical homogeneity involved high-temperature extrusion or forging. Secondary final working was usually carried out at room temperature through rolling, drawing, or swaging. Thus, stringent process controls and materials specifications are necessary for the successful production of Nb alloys.

13.2.3 FABRICATION OF NUCLEAR-GRADE MOLYBDENUM Because molybdenum possesses one of the highest melting temperatures (2623°C) and volatilizes or produces scale at a relatively low temperature [35], traditional alloy fabrication methods such as smelting are not possible. Instead, molybdenum can be produced through a series of grinding and separation steps leading to, as example, one of three

592

CHAPTER 13 REFRACTORY ALLOYS

Electron beam melng

Purificaon and alloy addion

Double vacuum Consumable electrode ARC melng

Final alloy addion and composion control

Primary breakdown ingot 900–1300°C

Extrusion

Round sheet bar tubular

Forging

In process anneal (1100–1650°C)

Secondary breakdown (Ambient to 1300°C) Extrusion

Rolling

Swaging, rocking

In process anneal (1100–1650°C)

Plate, rob, tube redraw stock, pipe

Final working (Ambient)

Drawing or pilgering Wire tubing

Rolling Rod, bar, shape sheet Strip foil

Swaging

Rod wire

In process anneal -variable(1100–1650°C) Final annealing (1100–1650°C)

FIG. 13.4 Process flow diagram for the production of Nb base alloys [30].

types of product forms: sintered products, which are most of the commercial products available, arc cast products, which are also commercially available even though significantly more costly, and electron beam melted products, which are also quite costly and essentially laboratory-grade materials. Since polycrystalline molybdenum possesses inherently weak grain boundaries, grain boundary strengthening and refinement techniques have been pursued to mitigate embrittlement [36–40]. Significant improvement in ductility and fracture toughness has been achieved thus far mainly through impurity control, reduction of the grain size, and through addition of alloying elements such as Re, Ti, Zr, Al, B, and C, or oxide and carbide particles. In many cases, the control of impurities in both the matrix and grain boundary is crucial to obtain a product with good ductility. This is especially true for oxygen. Moreover, the Mo crystal shares the same room-temperature embrittlement issue as other Group VIA metals (tungsten, chromium), which has been long realized to be relieved through direct solid-solution alloying with Re. The underlying mechanisms of the so-called Re-effect [41], which is technologically important for both W and Mo, have been heavily debated in the literature with possible mechanisms including: (1) solid-solution softening attributed to improved ½h110i screw dislocations [42–47] and (2) a change in the predominant slip plane from 110 to 112. While the Mo-Re alloys are technologically important, with alloys such as Mo-5Re, Mo-41Re, and Mo-47.5 Re being readily available, they are not currently considered for nuclear application in part due to the extreme cost of Re, and in part due to the poor irradiation performance of the alloys (see Section 13.4.3).

13.2 PRACTICAL ROUTES FOR REFRACTORY ALLOY PRODUCTION

593

FIG. 13.5 Production route to common forms of molybdenum.

A schematic representation of the process for producing the three generic Mo products discussed above is given in Fig. 13.5. This process starts with a series of grinding and separation steps to isolate MoS2, which is then roasted in air to product MoO3 and SO2. The SO2 is then converted to sulfuric acid for commercial use, leaving a “technical oxide” containing approximately 57% Mo with
0.1% S. This material is then chemically purified to produce Mo metal products. Pure Mo products can then be formed by sintering, arc casting, or through electron beam melting the purified powder, or alloys can be produced through standard powder mixing or through ball milling. The bulk of

594

CHAPTER 13 REFRACTORY ALLOYS

Mo powder produced is sold in pressed and sintered mill products such as sheet, plate, rod, etc. These pure or alloyed powders are typically placed in an elastomeric mold, evacuated, and undergo a cold isostatic pressing (CIP). Typical pressures used are 200 MPa followed by a hydrogen sinter in the 1700–1800°C range. The hydrogen cover gas plays the crucial role of reducing the final oxygen content of the product to <20–30 wppm. A lesser amount of materials is produced following the press-sinter-melt (PSM) or arc-melt process originally developed by Climax Molybdenum. This semicontinuous process starts feeding the pure or alloyed product into a vacuum arc furnace heated by either consumable graphite electrodes (which suppress oxygen) or nonconsumable tungsten electrodes. Standard arc-cast molybdenum produces materials of very low oxygen content (<5 wppm) though at relatively high carbon content (
300 wppm.) A low-carbon arc-cast (LCAC) molybdenum is available with ASTM B386 specification of <100 wppm carbon and <15 wppm oxygen. Even higher purity levels or pure or alloyed material can be obtained by electron-beam melting in a copper crucible. For e-beam melted alloys uniformity can be an issue, typically addressed by repeated deformation and remelting. In both the arc-cast product and electron-beam product casts, grain sizes are quite large leading to low ductility and poor fracture toughness. This is mitigated by extruding the products at relatively high temperature to refine the grain size, which in many cases causes highly anisotropic grain structures and associated mechanical properties. Materials of specific interest to nuclear applications are unfortunately not the pressed and sintered materials most commonly available. As will be discussed in Section 13.4.3, this is attributed to the issue of the irradiation instability of grain-boundaries that are more difficult to control in the pressed and sintered materials. For this reason, the focus of attention over the past decade has been on the development and study of three types of Mo: high-purity LCAC Mo, the commercial alloy TZM produced through arc melting, and a lanthanum oxide particle containing oxide dispersion-strengthened “ODS-Mo.” Table 13.4 provides a comparison of the typical chemistry of a pressed and sintered bar product with the materials of most recent focus for nuclear structural applications [48]. The LCAC Mo and TZM products of Table 13.4 were produced through an arc-melt process, though it is noted that a sintered TZM product is also commercially available. The clear difference between the LCAC Mo and the sintered Mo product is the suppression of the oxygen, promoting strengthened grain boundaries. It is noted that TZM, which is the most widely available carbide-strengthened alloy, utilizes its alloying aids primarily to induce the formation of ZrC and TiC on the grain boundaries, which then inhibit grain-boundary grain-growth and grain-boundary failure. The presence of these carbides results in approximately twice the high-temperature (1095°C) strength as compared to pure Mo and a significant increase in material recrystallization temperature, potentially expanding the operating temperature window of this alloy. The ODS Mo alloy discussed in Table 13.4 is an extremely pure alloy with the exception of the intentional addition of La. Specifically (see Fig. 13.5), wet doping of the MoO2 powder with an La-nitrate solution takes place followed by pyrolysis, producing a fine dispersion of La-oxide “ODS” particles in the compact powder, which is then consolidated by hot extrusion and warm rolling. With the exception of the additional steps to produce the

Table 13.4 Chemical Composition of Nuclear-Grade Molybdenum Currently Under Study [48] Chemical Composition (wppm) Material a

Sintered bar LCAC Mo ODS Mo TZM a

C

O

N

Ti

Zr

Fe

Ni

Si

La

50 90 40 223

70 3 n/a 17

n/a 4 n/a 9

20 n/a <10 5000

n/a n/a n/a 1140

50 10 27 <10

20 <10 <10 <10

30 n/a <10 n/a

n/a n/a 1.08w/% n/a

Max Quantities, H.C. Starck. PD-7009, Issue 1–2009–08-10. Typical bar product typically contains 20–30 wppm.

13.2 PRACTICAL ROUTES FOR REFRACTORY ALLOY PRODUCTION

595

FIG. 13.6 Representative microstructure of ODS molybdenum microstructure [48]. From R. Baranwal and M. B. Burke. “Transmission Electron Microscopy of Oxide Dispersion Strengthened (ODS) Molybdenum: Effects of Irradiation on Material and Microstructure” Report B-T-3462. Bechtel Bettis 08/2007.

lanthanum oxide “nodules,” the fabrication is more typical of the press-and-sinter route to production. For this material, the presence of the lanthanum in the final form serves to suppress oxygen segregation at the grain boundaries. The relative size and location of the ODS particles in this form is shown in the transmission electron micrograph in Fig. 13.6. Molybdenum and its alloys can be welded through a number of methods including conventional gas tungsten arc welding (GTAW.) However, such welds are considered nonstructural as the heat-affected zone (HAZ) tends to have a much-reduced brittle-to-ductile transition (BDT) temperature and degraded strength that concentrates triaxial stresses in the weld region leading to (typically brittle) joint failure. Alternative methods to GTAW such as electron beam welding are often chosen in order to limit the absorbed heat and extent of the melt region. The underlying mechanism for weld embrittlement is oxygen contamination and segregation to grain boundaries and, for this reason, Mo materials are typically welded under controlled atmosphere or vacuum. Arc cast molybdenum, and in particular materials with higher carbon contents, have improved weldability as compared to typical powder metallurgical forms of molybdenum. Moreover, the reactive elements in TZM interact with oxygen during the welding process, mitigating its ability to embrittle grain boundaries, and making TZM a fairly robust alloy with regard to joining. In a similar vein, molybdenum alloys including Zr, B, C, and Al have been demonstrated to have the ability to preclude oxygen segregation to the grain boundaries (as determined though atom probe tomography studies [38, 39, 49]) during fabrication and welding, resulting in improved as-welded properties [40, 51].

596

CHAPTER 13 REFRACTORY ALLOYS

13.2.4 PRACTICAL ROUTES OF TUNGSTEN AND TUNGSTEN ALLOY PRODUCTION In this section, the practical fabrication route for tungsten and tungsten materials is described. The production route utilizes powder-metallurgy techniques due to the high melting point of tungsten (3422°C) and its susceptibility to produce volatile oxide products that makes traditional smelting processes impractical. Tungsten fabrication involves powder production, blending, compacting, sintering, and a hot- and cold-forming process. In addition, a near net-shape production route is introduced.

13.2.4.1 Powder production Tungsten is usually prepared by hydrogen reduction of high-purity tungsten oxides. Such tungsten oxides can either be tungsten trioxide (WO3) or tungsten blue oxide (WO3 x) and, in rare cases, tungstic acid (H2WO4). Reduction is performed in rotary or pusher furnaces. The operating temperature lies between 600°C and 1000°C. The hydrogen gas has the additional purpose of carrying away the water vapor that is formed during the reduction. Tungsten particles form as a result of vaporization and later deposition processes. Therefore, due to a variation of temperature and water vapor partial pressure processes summarized in Fig. 13.7, tungsten particle sizes between 0.1 and 100 μm can be easily achieved. The powder metallurgical production of tungsten materials usually uses particle sizes between 2 and 6 μm.

13.2.4.2 Blending For the production of tungsten alloys, tungsten powder is mixed with a powder of the alloying element. Common alloying elements include Re, Ta, or Mo. Most elements that form a solid solution form brittle phases at one point. Therefore, the alloying element has to be very evenly distributed. One approach is to mix rare oxides and co-reduce them to metal powder. In addition, often a binder is added.

13.2.4.3 Compacting The fabricated powder is consolidated into a compact form. Tungsten has a relatively high hardness and strength and therefore the deformation of the particles is not easy. This compacting process however is performed without a lubricant to exclude impurities that would affect the mechanical properties of the end product. Two main routes Tungsten powder average grain size

Temperature

Water vapor concentration

Tungsten oxide related

Hydrogen related

Tungsten mass flow

Flow rate

Powder layer height

Flow direction

Porosity of powder layer

Dew point

FIG. 13.7 Summary of tungsten oxide processing parameters that influence grain size.

13.2 PRACTICAL ROUTES FOR REFRACTORY ALLOY PRODUCTION

597

have been established to compact tungsten powder: die pressing and isostatic pressing. After compacting, the component is called a green compact.

13.2.4.4 Die pressing Die pressing of the power(s) is performed from the top or from the bottom and the top using rigid dies in mechanical or hydraulic presses. Mechanical presses can apply loads up to 1 MN, while hydraulic presses have a pressing force of up to 30 MN. Usually, pressures in the range of 200–400 MPa are needed in order to achieve a compact density of 55%–65% of the theoretical density. The pressure is also dependent on particle size and shape, as can be seen in Fig. 13.8.

13.2.4.5 Cold isostatic pressing Cold isostatic pressing (CIP) is more common than die pressing. The raw powder(s) are filled into flexible molds (rubber or elastomers) and compressed. Most common is the wetbag technique where powder is filled in flexible molds that are immersed in water (in rare cases other liquids) and pressure is applied isostatically. With this process, tungsten ingots of up to 1 ton can be pressed and more complex components can be compacted. Another subtype of CIP is drybag pressing. This form is used for simple shapes. The powder-filled mold is sealed and the compression occurs between the mold and the pressure vessel.

13.2.4.6 Presintering In some cases, the green compound has to be presintered to have a sufficient strength after the compacting. After compression, parts are then heat treated at 1100–1300°C under a hydrogen atmosphere.

Green density

40

62

60 30 58 Compressive strength 56

Compressive strength (MPa)

Green density (%)

64

20

54 1

2

3

4

5

6

7

8

9

Average particle size, FSSS (mm)

FIG. 13.8 Relationship between average tungsten grain size and green density as well as compressive strength for a constant compacting pressure (241 MPa) [52]. From Yih HW, Wang CT. Tungsten. Sources, Metallurgy, Properties, and Applications, 1979, reproduced with permission of SNCSC.

598

CHAPTER 13 REFRACTORY ALLOYS 100 400 MPa

Grain size (mm)

10

600

800 1000 1200 1400

800°C 10 min

1600

1

0.1

0.01 0.4

0.5

0.6

0.7

0.8

0.9

1.0

Fractional sintered density

FIG. 13.9 Example grain size and sintered density contours for a variety of tungsten particle sizes processed at 800°C for 10 min over a range of possible compaction pressures, showing that extremely high compaction pressures are needed to deliver a grain size below 100 nm and nearly dense material. The best strength combinations result from tungsten particles below 100 nm [53].

13.2.4.7 Sintering The sintering of tungsten is very challenging, as evidenced by the extraordinary pressures required to achieve a relatively modest as-sintered density at 800 C (see Fig. 13.9). For this reason, very high temperatures and pressures are generally used, with sintering temperatures between 2000°C and 3050°C commonplace. The high temperature leads to a volatilization of impurities in addition to particle growth. The temperature is either applied with selfresistance heating (several thousand amperes) or with the help of a resistance element heating system. The former option is called direct sintering, while the latter is called indirect sintering. Both methods are carried out under flowing hydrogen in order to remove the oxygen coating of the tungsten particle surface. The final density of the tungsten component is around 92%–98% of the theoretical density. The final grain size in as-sintered ingots is about 10–30 μm. In order to achieve high sintering densities in relatively short times, the temperature has to be above 1900°C. At 2400°C, the average time to obtain a density of 92% lies at about 1 or 2 h. At 3000°C, the average time is already <30 min. Particle size has a great influence on the sintering time as well. The smaller the particle size, the shorter is the average time to obtain a high density at a given temperature. Sintering can be accelerated if small amounts (0.5%–1%) of alloying elements like Ni or Pd that enhance the grain boundary diffusion in tungsten are present. This, however, has a negative influence on the mechanical properties.

13.2.4.8 Direct sintering The direct sintering process uses electric current to control the temperature and, therefore, the sintering process. At the beginning stages, the current is kept relatively low. Due to a limited contact between the tungsten particles/ grains, the electric resistance of the tungsten product is relatively high, particularly during the initial sintering period. The temperature is gradually raised up to 3000°C, with holding points in between. These hold times are essential to allow outgassing of undesired elements. The average holding time is approximately 30–60 min. Direct sintering has the advantage of having higher temperatures in the middle of the part. This favors diffusion processes and leads to an efficient evaporation of impurities.

13.2 PRACTICAL ROUTES FOR REFRACTORY ALLOY PRODUCTION

599

13.2.4.9 Indirect sintering Indirect sintering usually does not require presintering. The green compact is placed in a basket-shaped heating element of a furnace. Uniform shrinkage is achieved by placing the compacts on green tungsten shims [52]. The heating occurs at a slow rate leading to a hold at 2000–2700°C. Sintering times may range up to 24 h.

13.2.4.10 Forming process In order to form a preshape, to improve the properties, and to achieve a completely dense material, a complex, multistage, hot- and cold-forming process is applied. Working is usually performed underneath the recrystallization temperature and therefore all forming is, strictly speaking, a cold-working process. Some typical techniques are rolling (sheets) or swaging for rods, forging for larger parts, or drawing for wire and tubes. A detailed description of the most common forming processes can be found in the book “Tungsten,” by Yih and Wang [52]. The first part of the forming process is carried out at around 1500–1700°C. If tungsten is worked at lower temperatures, cracks are reported to develop. With increasing degree of deformation, the grain size and shape changes significantly and anisotropy develops. This change in microstructure goes along with a change in mechanical properties. Due to this change, the forming process during the latter stages can be carried out at lower temperatures between 1100°C and 1300°C. Table 13.5 summarizes a typical progressive hot-rolling treatment and intermediate stress relief annealing schedule for the fabrication of tungsten sheet.

13.2.4.11 Near net-shape and mass fabrication A time and cost-effective near-net-shape-forming process with the advantage of shape complexity, material utilization, and high final density is powder injection molding (PIM). This process was adapted and developed at KIT for tungsten and promising results have already been achieved. The key steps in PIM are kneading or extrusion

Table 13.5 Example of Rolling Schedules for Tungsten Sheet Thickness After Pass (mm)

Preheating Temperature (°C)

20.3 15.9 12.7 10.2

1450 1400 1400 1400

8.26 6.86 5.46 4.45

1350 1300 1300 1300

Starting thickness 24.4 mm 1 2 3 (Cross roll) 4 Anneal 5min at 1300–1350°C 5 6 7 (Cross roll) 8

Anneal 5min at 1250–1300°C; caustic pickling and condition 9 10 11 12 13

3.51 2.82 2.26 1.80 1.63

1250 1200 1200 1150 1150

Source: Data from G.C. Bodine Jr., Tungsten Sheet Rolling Program, Final Report, Fansteel Metallurgical Corp., North Chicago, Illinois, 1963.

600

CHAPTER 13 REFRACTORY ALLOYS

of a suitable feedstock (combination of powder and binder), where the heated powders (80°C) are mixed with a 50 vol% wax/thermoplastic binder system in a kneader at 120°C. The design and engineering of a tool (including filling simulation) is followed by injection molding of the so-called “green parts” at a feedstock temperature of 160°C and a mold temperature of 50–60°C. At the first step, the premold thimble is replicated; after that, the tool is moved around 180° and in the second step the tile is molded on top of the thimble. This process is very quick and effective. The de-binding (solvent and thermal) is carried out in n-hexane, followed by a thermal debinding step in dry hydrogen atmosphere. During the debinding step, not only are the binder and the impurities (oxygen and carbon) removed, but also the high residual stresses generated during injection molding are released. The last step of the process is a final heat-treatment process. The PIM process for tungsten is shown in Fig. 13.10. The process allows for an easy, reproducible, cost-effective mass production of parts of general shape. The process does not depend on the particular composition of the tungsten material. Further, the sintering process can be varied in different ways. For example, the HIP step can be replaced by a simple sintering step in a hydrogen furnace. Usually, one-component PIM (the produced parts consist of the same tungsten material) is sufficient for many applications.

13.3 AS-FABRICATED MECHANICAL PROPERTIES 13.3.1 VANADIUM Several studies have shown that the mechanical properties and deformation behavior of V-Cr-Ti alloys are sensitive to composition and thermomechanical processing variables [4, 5, 8, 12, 15, 19, 25, 54]. The alloying additions of Cr and Ti form solid solutions in the bcc V matrix at elevated temperatures and provide strengthening mainly by solid solution (Cr) and precipitation (Ti). The main purpose of the Ti addition is to serve as a gettering agent for the interstitial O, C, and N atoms that are present as impurities in the bcc V matrix, and to form Ti(OCN) precipitates that provide some alloy strengthening [12, 15, 55]. The thermomechanical treatments performed during processing of V-Cr-Ti alloys also contribute to the sensitivity factor since they influence the amount of pickup of interstitial impurities as well as their removal from the bcc V matrix by precipitation processes and the average size, size distribution, and uniformity of grains. The effect of solute Cr and Ti additions on the tensile properties of V-Cr-Ti alloys is shown in Fig. 13.11 [56]. Elevated levels of Cr and Ti were added to the V-4Cr-4Ti (#832665) heat and the processing conditions used resulted in similar O, C, and N concentrations of 368  14, 118  19, and 102  19 wppm, respectively. Compared to V-4Cr-4Ti, the increase to 10%Cr and to 10% and 15%Ti resulted in continuously increasing yield and ultimate tensile stresses with essentially no effect on the uniform elongation and only a slight decrease in total elongation. Since the interstitial O, C, and N atoms are removed from the bcc V matrix by the formation of Ti9(OCN) precipitates, ductility is not affected significantly with higher Ti additions to the V-Cr-Ti alloy at room temperature. The fracture toughness and Charpy impact properties of V-Cr-Ti alloys depend on composition and processing variables as well as numerous experimental factors, such as specimen geometry, notch characteristics, and test conditions, that is, strain rate and temperature [4, 15, 40, 41]. The results shown in Fig. 13.12 illustrate these factors. In Fig. 13.12A, data obtained from Charpy V-notched specimens show monotonic increases in the DBTT for a variety of Cr and Ti solute additions as the annealing temperature is increased above 1000°C, and monotonic increases in DBTT for each annealing temperature as the solute Cr and Ti concentrations are increased. The Charpy impact results shown in Fig. 13.12B for V-4Cr-4Ti (#832665) indicate that annealing at 1000°C resulted in no detectable DBTT down to 196°C (i.e., extremely good resistance to embrittlement down to very low test temperatures). Conversely, annealing at 1125°C resulted in a DBTT near 125°C with very low value of lower shelf energies. This poor Charpy impact behavior in the 1125°C annealed material was significantly improved by a subsequent anneal at 890°C. The increase in DBTT from annealing at 1125°C was attributed to a larger grain size and larger percentage of interstitial O, C, and N in solid solution, as opposed to finer grain size and Ti(OCN)

Torque [Nm]

20 18 16 14 12 10 8 6 4 2 0 0:0

W-feedstock Kneader 0:10

0:20

0:30

0:40

0:50

1:0

Kneading time [h:min]

Feedstock development

Filling simulation / design+engineering of a tool

Charpy impact test specimen Sleeve Tensile test specimen

Divertor part “tile”

Disc

Gear wheel

Green parts (dark) finished parts (bright)

HIP Presintering Debinding + heat-treatment process

FIG. 13.10 The PIM process for tungsten as developed at KIT.

Injection molding of green parts

13.3 AS-FABRICATED MECHANICAL PROPERTIES

W-powder + binder

601

602

CHAPTER 13 REFRACTORY ALLOYS 50

700

600 40

30

400

300

20

Elongation (%)

Stress (MPa)

500

200 10 100

YS

UE

UTS

TE

0

0 V-4Cr-4Ti V-10Cr-4Ti V-4Cr-10Ti V-4Cr-15Ti

FIG. 13.11 Comparison of tensile properties between the reference V-Cr-4Ti alloy with increases in the Cr content (V-10Cr-4Ti) and in the Ti content (V-4Cr-10 and V-4Cr-15Ti). The tensile tests were conducted at room temperature using a strain rate of 103/s. 20

100 1150⬚C

50

±20⬚C 15 1125⬚C/1h + 890⬚C/24h, T-L 1100⬚C 950⬚C

–50 1000⬚C

–100

Energy, J

DBTT (⬚C)

0

Annealing temp.

1000⬚C/1h, T-L

950⬚C 1000⬚C 1050⬚C 1100⬚C 1150⬚C

–150 –200

5 1125⬚C/1h, T-L

–250 5

(A)

10

10

15 [Cr + Ti] Wt %

20

25

0 –200

(B)

–150

–100 –50 Temperature (⬚C)

0

50

FIG. 13.12 Effect of annealing temperature on (A) the DBTT of V-Cr-Ti containing different solute additions [13, 57] and (B) the impact energy of the reference V-4Cr-4Ti (#832665) [14]. From R.J. Kurtz, M.L. Hamilton and H. Li, “Grain Boundary Chemistry and Heat Treatment Effects on the Ductile-To-Brittle Transition Behavior of Vanadium Alloys”, Journal of Nuclear Materials, Vol. 258–263, (1998), p. 1375.

precipitates that formed during annealing at 1000°C. The annealing at 890°C increases the volume fraction of Ti(OCN) precipitates that lowers the O, C, N interstitial solutes in solid solution. The temperature-dependent tensile properties of V-4Cr-4Ti (heat #832665) are shown in Fig. 13.13 [58]. The yield and ultimate tensile strengths (UTS) and total elongation decrease from room temperature to 300°C are followed by a nearly temperature-independent regime for strength and ductility up to 700°C. Above 700°C,

13.3 AS-FABRICATED MECHANICAL PROPERTIES

603

50

500

400

40

300

30

200

20

100

10 Yield stress

Uniform strain

Ultimate stress

Total elongation

0 0

100

200

300

400

500

600

700

800

Strain (%)

Stress (MPa)

Tested at 10–3/s

0 900

Temperature (°C)

FIG. 13.13 Temperature dependence of tensile properties for the V-4Cr-4Ti (heat 832,665) alloy.

the strength increases to a local maxima and ductility decreases to a minima near 750°C, followed by relatively pronounced decreases in strength and increases in ductility at higher temperatures due to the onset of thermal creep processes. The tensile curves obtained for V-4Cr-4Ti from 20°C to 850°C with a strain rate of 103/s are shown in Fig. 13.14 [59, 60]. The results illustrate the influence that interstitial O, C and N have on the mechanical properties [4, 5, 8, 12, 15, 19, 25, 54, 57, 59]. The tensile curves are offset from the stress and strain axes for clarity. For most of the curves, the upper yield stress, σy, is preceded by a small plastic pre-strain in the elastic regime followed by a L€uders extension ranging from 0.5% to 1.0%. Prior to the L€ uders extension, a prominent load drop occurs at 20°C, but is barely detectable at higher temperatures. Following the L€ uders extension, deformation occurs homogeneously throughout the specimen with decreasing strain-hardening rate until reaching the ultimate tensile stress, σu, which defines the point of plastic instability. For temperatures between 300°C and 750°C, and at sufficiently low strain rates, serrations occur in various regions of the flow curves. These serrations are related to the diffusion of solute atoms and the locking of dislocations during the test, that is, the phenomenon of dynamic strain aging (DSA) [59, 60]. As evident in Fig. 13.14, the DSA is most pronounced near 600°C and for plastic elongations of
10%–20% for the tensile tests performed at a 103/s strain rate. The DSA behavior of V-4Cr-4Ti was investigated using varying strain rates in tensile tests conducted over a temperature range from 20°C to 850°C [59, 60]. From the stress-strain curves, the strain rate sensitivity (SRS) parameter defined as [61] m¼

1 δσ σ δln_E E, T

(13.1)

was determined from the slope of the yield stress (σy), stress at 8% strain (σ8%), and ultimate tensile stress, σu plotted against the logarithm of strain rate. The temperature dependence of the ultimate stress, σu, as a function

604

CHAPTER 13 REFRACTORY ALLOYS 1000

Strain rate = 10–3/s

850°C 900

800°C

750°C

800

700°C

700

600°C 500°C

Stress (MPa)

600 500 400

400°C

300

300°C 200

200°C 100

20°C

100 0 0.00

0.05

0.10

0.15

0.20

0.05

0.25

0.30

0.35

0.40

0.45

0.50

Strain (%)

FIG. 13.14 Temperature-dependent tensile curves for V-4Cr-4Ti at a strain rate of 103/s illustrating the dynamic strain aging regime at 400–700°C.

of strain rate and the calculated values of m as a function of temperature for V-4Cr-4Ti are shown in Fig. 13.15. Several different flow regimes are evident, with the SRS m > 0 at low and high temperatures whereas m < 0 at intermediate temperatures. During normal deformation processes, the flow strength increases with strain rate and m is positive, which generally occurred below 300°C and above 700°C for V-4Cr-4Ti, depending on the strain rate and the flow stress parameters. However, within the DSA regime, the increase in strain rate causes a decrease in the flow strength and m is negative. This regime generally occurred between 300°C and 700°C for V-4Cr-4Ti. Since deformation occurs by thermally activated slip processes at elevated temperatures, the ability of solute atoms to diffuse and form atmospheres around mobile dislocation cores will influence the type of deformation behavior. At low temperatures, solute mobility is too low to form sufficient solute concentrations while at high temperatures, solute concentrations are too mobile to exert sufficient drag on mobile dislocations. For the DSA regime, the increase in strain rate decreases the time available for solute diffusion to dislocations, which decreases the drag effect, resulting in a lowering of the flow strength, that is, negative SRS behavior. For the V-4Cr-4Ti, the interaction between Ti and interstitial O, C, and N impurities to form Ti-OCN precipitates removes the interstitial impurities from the bcc matrix and reduces the DSA effect. Thus, the DSA behavior observed for V-Cr-Ti alloys is indicative of the concentration of interstitial impurities that are in solution. The high-temperature thermal creep properties of V-4Cr-4Ti are sensitive to changes in the interstitial concentrations due to the high affinity of V for interstitial solute atoms, especially O, and experimental factors such as long test times, high temperatures, and surrounding environment of specimens during the creep test [13, 19, 62–64]. The creep properties of V-4Cr-4Ti investigated in high vacuum using uniaxial tensile and biaxial pressurized tube

13.3 AS-FABRICATED MECHANICAL PROPERTIES

605

460 –0.028

440 420 –0.017

380

Ultimate stress (MPa)

0.017

–0.020

400

–0.014 –0.005

360 340 320 300

20°C

700°C

300°C

750°C

260

400°C

800°C

240

500°C

850°C

280

600°C

220 10–6

10–5

10–4

10–3

10–2

10–1

100

–1

(A)

Strain rate (s ) 0.08 Lower yield stress Stress for 8% strain Ultimate stress

0.06

Homogeneous deformation regime

Strain rate sensitivity

0.04

0.02

0

–0.02

Inhomogeneous deformation regime

–0.04 0

(B)

100

200

300

400

500

600

700

800

Temperature (°C)

FIG. 13.15 (A) Ultimate stress as a function of strain rate of V-4Cr-4Ti over the temperature range from 20°C to 850°C. (B) Strain rate sensitivity as a function of temperature for the three measured flow stress parameters.

606

CHAPTER 13 REFRACTORY ALLOYS

10–2 10–3

10–6 Uniaxial, 600°C [8] n = 9.9 725°C Biaxial, 800°C n = 2.7

Creep rate (s–1)

Effective creep rate (%/h)

10–1

10–4

800°C

10–7

700°C 650°C

10–8

10–5 –6

10

(A)

10

Biaxial, 700°C n = 3.7 100 Effective stress (MPa)

500

10–9

(B)

100

1000 Applied stress (MPa)

FIG. 13.16 Stress dependence of the minimum creep rate of V-4Cr-4Ti tested in vacuum at 600–800°C by (A) biaxial pressurized tube specimens [10] and (B) uniaxial tensile specimens [63].

specimens at temperatures >600°C are shown in Fig. 13.16 [13, 62, 64]. The results showed minimum creep rates that were consistent with the power-law creep expression:   Q ε_ ¼ Aσ n exp  RT

(13.2)

where σ is the applied stress, n is the stress exponent, Q is the activation energy for the controlling creep mechanism, R is the universal gas constant, and T is absolute temperature. The measured values of n were within the range of 2.7–4.0 for the two creep test methods. The measured activation energies for creep were
260–330 kJ/mol for the two sets of data, which was close to the self-diffusion of pure vanadium, or
270 kJ/mol. The predominant creep mechanism based on the data analysis was climb-assisted dislocation glide; however, thermally activated solute drag mechanisms could limit the rate of dislocation climb past obstacles in the microstructure at temperatures of 600–800°C and stresses of 50–400 MPa. Creep testing in liquid Li at 700°C and 800°C was conducted on V-4Cr-4Ti prepared from the United States (#832665) and Japan (NIFS-Heat-2) heats using biaxial pressured tube specimens [64]. The results showed that the creep strain rate was affected by the heat, tubing production method, and changes in the interstitial O and N levels that occurred during the creep test. The O levels decreased, while the N levels increased in the specimens during exposure in liquid Li, as expected from V-Li thermochemistry considerations (N and O being present in both the V alloy and molten Li as an impurity] [65]. The predominant creep behavior was an inverted primary transient creep period followed by steady state, or accelerating creep to rupture. Unlike the creep tests of V-4Cr-4Ti in vacuum, creep in liquid Li does not follow the classical three-stages of creep exemplified by primary, secondary, and tertiary creep.

13.3.2 AS-FABRICATED MECHANICAL PROPERTIES OF NIOBIUM While niobium is BCC and therefore subject to low-temperature embrittlement and highly susceptible to impurity embrittlement (e.g., oxygen, nitrogen, carbon, hydrogen), the BDT temperature in the nonirradiated state is quite low (see Fig. 13.3). For this reason, niobium has very good formability and therefore is a desirable alloy for a number of practical applications. This enhanced fabricability comes at the expense of strength, which is typically lower than the other refractory metals being discussed in this chapter. As example, the UTS of a number of niobium alloys

13.3 AS-FABRICATED MECHANICAL PROPERTIES

607

FIG. 13.17 Ultimate tensile strength of niobium alloys compared with vanadium and molybdenum.

studied for nuclear and aerospace applications (including the two alloys in current commercial production, Nb-1Zr and C-103) are provided in Fig. 13.17. Also provided in the figure for comparison purpose is the vanadium alloy V-4Cr-4Ti and the TZM molybdenum commercial alloy. As drawn from Fig. 13.17, the beneficial combination of ductility and fabricability comes at the cost of alloy strength, with the suite of niobium alloys all similar to or lower than that of vanadium, and substantially lower than the strain-relieved TZM alloy depicted or tungsten alloys (not depicted). In comparison with tantalum alloys, which are routinely considered alongside niobium for space and nuclear applications, the niobium alloys have similar strength at low-to-intermediate level. For example, Nb-1Zr is very similar to the tantalum alloy T-111 (Ta-8 W-2Hf) to approximately 800°C, though due to a coarsening of carbides in the niobium alloy, its tensile and fracture properties degrade at higher temperatures [66, 67]. As pointed out by Leonard [68], such microstructural instability in alloys such as Cb-752 and Cb-753 at elevated temperature leads to a degraded creep strength, as evidenced by a negatively sloping Larson-Miller plot and embrittlement (in the case of FS-85) of grain boundaries.

13.3.3 AS-FABRICATED MECHANICAL PROPERTIES OF MOLYBDENUM The transition metal molybdenum has the second highest melting temperature after tungsten, of the materials considered in this chapter. This high melting temperature, combined with the second commonality with tungsten-a BCC crystal structure, contributes to the limited low-temperature ductility of molybdenum. As discussed in the previous section, the primary Mo alloys of industrial interest are the pressed and sintered materials. The mechanical properties for each of these alloys, as with the higher purity alloys of more interest to nuclear applications, are strongly dependent on the amount of process deformation and annealing applied in order to gain some ductility. Fig. 13.17 provides a comparison of the temperature-dependent tensile yield stress for Plansee commercial pure sintered molybdenum and high-strength TZM alloy in the stress-relieved and recrystallized conditions. It can be seen that stress-relieved materials exhibit significantly higher strengths than recrystallized materials up to high

608

CHAPTER 13 REFRACTORY ALLOYS

temperatures, and that TZM alloy has much higher strength than pure Mo. In addition to increased yield strength, TZM possesses other beneficial engineering qualities such as approximately twice the stress rupture strength and a much lower steady-state creep rate. Inset within the figure are specific temperatures determined to provide 1 h recrystallization for 90% deformed molybdenum alloys (for lower deformation levels, longer times or higher temperature would be required). The 1-h recrystallization annealing times increase from
1100°C for pure sintered Mo up to 1500°C for the specialty alloys such as the Mo-Re alloys, TZM, and the hafnium carbide grain boundary stabilized Mo-MHC. Essentially independent of processing route, the work deformation required in the production of molybdenum produces material with very limited room-temperature ductility, mandating some level of heat treatment to allow use in structural applications. As seen in Fig. 13.18, the primary result of recrystallization for pure Mo and Mo alloys is a product with significantly lower strength. A more modest reduction in strength is obtained by applying a stress-relieving anneal, which also yields modest improvements in ductility. The relative stress relief, or full recrystallization of the material can have a large impact on the ductility, strongly dependent on the initial interstitial content of the material and any interstitial pick-up during the process. For materials of interest to nuclear applications, the interstitial content will be well controlled and very low precluding the sweeping of oxygen to grain boundaries and associated reduced fracture toughness and embrittlement upon recrystallization. As example, in Cockeram’s paper [51] on the development of weldable molybdenum alloys, both pure LCAC molybdenum and a number of “weldable” alloys improved from an as-worked total elongation of 2%–4% in the 100–1000°C range to 2%–18% in the stress-relieved condition to 14%–42% in the recrystallized condition. The tensile behavior for Mo materials of current developmental interest for nuclear structural application is given in Fig. 13.19. These materials include the low carbon arc-cast Mo, arc-cast TZM (contrasting with the sintered TZM of Fig. 13.18), a lanthiated (ODS) Mo, and a molybdenum alloy developed as a high-purity weldable alloy containing B-11, C, and Zr as alloying elements to scavenge oxygen [51]. The yield strength for all these

FIG. 13.18 Yield stress of commercial Mo alloys (Plansee technical data sheet).

13.3 AS-FABRICATED MECHANICAL PROPERTIES

609

FIG. 13.19 Comparison on non-irradiated tensile properties of various forms of interest to nuclear application in stress-relieved conditions along the rolling direction. Figure drawing upon data from [51,70,69] Inset, LCAC and TZM Mo fracture surface following ambient tensile test; Inset from Cockeram BV, Smith RW, Snead LL. The Influence of Fast Neutron Irradiation and Irradiation Temperature on the Tensile Properties of Wrought LCAC and TZM Molybdenum. Journal of Nuclear Materials. 2005;346:145–164.

tested Mo alloys decreases monotonically with increasing test temperature, with the weldable Mo alloy exhibiting the highest strength and the other three materials exhibiting strengths comparable to each other. In each case, the alloys possess moderate room-temperature ductility, which significantly decreases in the intermediate temperature range and then increases rapidly as test temperatures consistent with stress-relief annealing temperatures are achieved. Due to the relatively large amount of reduction molybdenum undergoes during the forming process, it develops a highly anisotropic lamellar or “pancake-like” microstructure. Depending on the size of the grains and strength of

610

CHAPTER 13 REFRACTORY ALLOYS

Table 13.6 Comparison of Tensile Properties of Stress Relieved Nuclear Grade Molybdenum in Longitudinal and Transverse Orientations [70]

LCAC—Longitudinal LCAC—Transverse TZM—Longitudinal TZM—Transverse ODS—Longitudinal ODS—Transverse

Yield Strength at RT (MPa)

Total Ductility at Room Temp.(%)

Yield Strength at 600°C (MPa)

Total Ductility @600°C (%)

751 809.3  9.7 803 895 735  32 823  24

17.6  1.5 13.1  2.8 19.6 11 14.2  2.6 9.6  2

475.6 33.4

3.8 0.5

546.5  26.9 637.4  9.3 445  20

4.7 0.2 2.5 0.4 4.3 0.4

the grain boundaries, this microstructural anisotropy can have a large effect on the mechanical properties. For the typical wrought alloys being discussed here, at temperatures near or above the DBTT, fracture initiates at grain boundaries (or perhaps at ODS particles) producing a separation of the adjoining lamella. An example of the typical microstructure in the as-deformed condition is provided as an inset to Fig. 13.19, showing the decohesion between lamellae in LCAC Mo and TZM following ambient temperature tensile testing. From studies on in-situ fractures, where fracture toughness was observed directly in a SEM, a crack-divider type of toughening mechanism for these materials has been inferred [69]. Given this failure microstructure, it is understandable that the longitudinal and transverse ductility of the materials would be significantly different. This is clearly seen in the data summarized in Table 13.6 for the longitudinal and transverse tensile ductility of LCAC Mo, TZM, and ODS Mo, which indicates a significant reduction in ductility in the transverse (across lamella) direction as compared to the longitudinal direction. The fracture toughness of nonirradiated molybdenum, while strongly affected by grain size, is also affected by orientation. Typically, the finest grain materials such as ODS Mo would possess the highest fracture toughness, while LCAC Mo with the largest grains exhibits the lowest fracture toughness and highest DBTT. Materials such as TZM, of intermediate grain size, fall in between. This can be seen by inspection of Fig. 13.20, along with the anisotropy in fracture toughness.

250 Fracture toughness (MPA/m–2)

LCAC Mo

TZM

Transverse

Longitudinal

200

ODS moly

Longitudinal

150

100 Transverse 50

Transverse Longitudinal

0 –200

0

200

400

600

Temperature (°C)

800

1000

0

200

400

600

Temperature (°C)

800

1000

0

200

400

600

Temperature (°C)

FIG. 13.20 Fracture toughness and anisotropy in fracture toughness of different types of molybdenum.

800

1000

13.3 AS-FABRICATED MECHANICAL PROPERTIES

611

13.3.4 AS-FABRICATED MECHANICAL PROPERTIES OF TUNGSTEN The outstanding mechanical properties of tungsten are the high tensile strength, including high thermal creep resistance at elevated temperatures. The biggest disadvantage, especially for the use as a structural material, is the intrinsic brittleness (measured by tensile, Charpy impact, or fracture mechanics tests) and the low fracture toughness, particularly near room temperature and below. In addition, the DBTT is especially high. Mechanical properties and therefore also fracture toughness of polycrystalline tungsten are strongly dependent on the underlying microstructure and the chemical composition. Table 13.7 compares the large difference in Vickers Hardness HV30 for polycrystalline tungsten at room temperature under recrystallized and work-hardened conditions [52]. Various tungsten materials and alloys have been developed in the past for structural applications, mainly in order to increase the ductility of tungsten. The listed values for tungsten alloy mechanical properties in this chapter should be considered as general tendencies due to the strong dependence of mechanical properties on microstructural features and the fact that in most cases the precise microstructure, type and concentration of impurity elements, and their distribution are not detailed in published articles.

13.3.4.1 Structured tungsten materials While tungsten is accepted as a very brittle material, deformation of high-purity single-crystal tungsten exhibits some ductility even at the low temperature of liquid helium. All tungsten materials show a distinct temperature dependence of the flow stress [71], typically explained by the thermally activated interaction between dislocations and interstitial impurities that are in solution. Above the critical temperature, a strong increase of the flow stress can be observed. For pure tungsten, this critical temperature is well above room temperature, at
330–530°C [72]. The stress-strain curves for high-purity tungsten tested as a single crystal and a polycrystalline material swaged and subsequently annealed for 1 h at 2000°C are shown in Fig. 13.21 [73] for tests performed at room temperature and 400°C. The polycrystalline material shows good ductility at 400°C but is completely brittle at room temperature, indicating that the BDT temperature for this tungsten material lies between 20°C and 400°C. The precise value of the BDT temperature is dependent on the material condition and the strain rate, along with other experimental variables. For tungsten, the reported tensile DBTT usually lies between 100°C and 200°C, but in cases of extreme high strain rates this value can rise up to 1000°C [74]. After fabrication, tungsten products can exhibit different microstructures; typical examples are shown in Fig. 13.22A and B. While tungsten rods (Fig. 13.21B), due to the uniform radial stress applied during extrusion, have resulting elongated tendril grains along the extrusion (rod) axis, (Fig. 13.22A) the in-plain stress applied during the rolling of a plate results in a lathe-like (staked pancake) structure for the plates. While a number of factors can contribute to the strength, elongation, and fracture properties of toughness, the processing-derived structure is a major factor. For example, Fig 13.22C provides the absorbed fracture energy as a function of temperature as dependent on the orientation of the sample with respect to plate rolling (note: the plate lathe structure corresponds to Fig 13.22A.) As seen from the figure, for the two directions in which a crack is forced to propagate normal to the pancake structures (LS and LT) a reasonable measure of energy is absorbed in the sample upon failure. As the notch and therefore the crack is oriented in the plane of the pancakes, little energy is absorbed in crack propagation.

Table 13.7 Macro-Vickers Hardness for Polycrystalline Tungsten is Given as (HV30) [52] Recrystallized Work hardened

300 >650

612

CHAPTER 13 REFRACTORY ALLOYS

Engineering stress (MPa)

1000 Single crystal 25°C

800

600 Polycrystal 25°C

400

Polycrystal 400°C Single crystal 400°C

200

0 0

20

60 40 Elongation (%)

80

FIG. 13.21 Stress-strain curves of polycrystalline and single-crystal tungsten performed at a strain rate of 2%/min [72]. From Allen BC, Maykuth DJ, Jaffee RI. The Recrystallization and Ductile-Brittle Transition Behaviour of Tungsten-Effect of Impurities on Polycrystals Prepared from Single Crystals. Journal of the Institute of Metals. 1961;90(120–128).

FIG. 13.22 (A) Pancake microstructure of plates, (B) fiber-like microstructure of rods. (C) Influence of the microstructure orientation on the Charpy properties of pure tungsten blanks.

The high sensitivity of fracture toughness regarding grain shape and orientation is due to the weak grain boundary cohesion in tungsten. Tungsten specimens are tougher when they are stressed parallel to the long side of the grains (fiber-like or “pancakes”). Stresses, acting perpendicular to the grain boundaries, cause easy crack initiation as well as easy crack propagation. Charpy specimens oriented parallel to the axis of tungsten rods do not show a direct transition from brittle to ductile fracture with raising test temperatures. Moreover, a broad range of the so-called delamination fracture can be observed. Specimens fabricated from plates or round blanks behave in a similar manner.

13.3 AS-FABRICATED MECHANICAL PROPERTIES

613

13.3.4.2 Dispersion strengthened tungsten

Creep stress (MPa)

Dispersion strengthening and precipitation hardening are the most effective ways to increase the high-temperature strength and creep resistance of tungsten. Such tungsten materials are often called ODS tungsten. Small amounts of insoluble elements like oxide or carbide powders are added to tungsten prior to the sintering process; the resulting material will consist of the typical tungsten microstructure but with an additional second phase at or around the grain boundaries. Either potassium solute additions in the ppm range (WVM) or oxides and carbides such as La2O3 (WL10), CeO2 (WC20), and ThO2 (WT20) are used. Other additions could be ZrO2, CeO2, ThO2, TiC, and HfC. Due to high activation, Th and Hf have to be excluded for most nuclear applications. Adding lanthanum oxide (WL10) results in a shift of around 100 K in the onset of brittle fracture to lower temperatures as shown in Fig. 13.23.

12 11 10 8

Ductile

7

tioi n

6

ina

5 4

Del am

Charpy energy (J)

9

Rod materials W, 6.9 mm W, 20 mm WL 10, 6.9 mm WL 10, 20 mm

3 2 1 0

tle

400

Brit

600

700

800

900

1000

1100

Test temperature (⬚⬚C)

FIG. 13.23 Top: Creep strength prediction showing tungsten and tungsten lanthanum oxide (WL10). Bottom: Charpy test results of pure tungsten and tungsten lanthanum oxide (WL10).

614

CHAPTER 13 REFRACTORY ALLOYS

At higher temperatures, oxides completely suppress ductile fracture, even at temperatures as high as 1100°C. It is very likely that the use of other oxides leads to similar results. In potassium-doped tungsten (WVM), the recrystallization temperature shifts by 200 K to higher temperatures. Since interstitially soluble elements like, for example, O, N, C, P, S, Al, or B, tend to segregate at grain boundaries, the addition of their small amounts would be another possibility for affecting the grain boundaries. While most of these elements will weaken the grain boundaries even more, only B is supposed to show a positive influence. Dispersion strengthening and precipitation hardening increase creep strength and recrystallization can be improved with only minor decrease in thermal conductivity by the use of dispersed oxides such as lanthanum that stabilize the grains. The intrinsic brittleness of tungsten (measured by Charpy tests), however, cannot be improved by oxide dispersion. On the contrary, intercrystalline fracture is enhanced even more.

13.3.4.3 Solution-strengthened tungsten Only rhenium is rigorously known to improve the ductility of tungsten by solid solution and is known to shift the BDT to lower temperatures. Unfortunately, its use for nuclear energy applications has been ruled out for reasons of cost and irradiation embrittlement. Iridium could also have a positive effect, though it is even more expensive than Re. Other elements forming solid solution with tungsten are: Ta, V, Nb, Ti, Nb, and Cr. Of these elements, only Ta, V, Nb, and Mo form solid solutes in the complete practical alloying range. Since Nb and Mo transmute to very longliving radioactive isotopes, they are considered unattractive for fusion application. This leaves only Ta and V as possible candidate solutes for fusion applications. However, both mechanical properties measurement of W-(x)Ta and W-(y)V and supporting theoretical density functional theory calculation show that these alloys are more brittle than high-quality pure tungsten. The apparent cause for the inferior performance of the Ta and V bearing is the formation of highly brittle intermetallic phases. Charpy tests in the temperature range of 400–1100°C have shown that in comparison to pure tungsten the BDT temperature for solid solutions in all ranges studied is increased (i.e., worse.) This is seen from the right-most image of Fig. 13.24, in particular from inspection of the significant increase in the W-5Ta DBTT. However, strength was improved for very small additions of Ta as an alloying element and only at temperatures higher than 1000°C. As an additional note, these alloying elements can suppress the tungsten thermal conductivity significantly. With the exception of W-Re, alloying tungsten in terms of establishing substitution solid solution does not improve the ductility. Grain boundary strengthening might improve the ductility. But so far it has not worked on the whole temperature range. B and TiC are candidates that still need to be tested. ODS tungsten materials like WL10 do not show fully ductile fracturing in Charpy tests at temperatures up to 1100°C.

5

Charpy energy (J)

4

Plate materials W, 3.6 mm WL10, 3.6 mm Round blanks W (RW), 30 mm WTa5, 30 mm

3 2 1 0 400

500

600 700 800 900 Test temperature (⬚⬚C)

1000

1100

FIG. 13.24 Left: Charpy test results of pure tungsten, W-Ta, and W-V alloys. Right: Charpy test results of different discussed tungsten materials.

13.4 AS-IRRADIATED MECHANICAL PROPERTIES

615

In ODS tungsten as well as potassium-doped tungsten (WVM), recrystallization is shifted to higher temperatures by at least 200 K. The creep strength of ODS tungsten is slightly increased. The thermal conductivity of ODS tungsten is comparable to pure tungsten parallel to the rolling direction. In plates, it is slightly reduced perpendicular to the rolling surface. As shown in the left graph of Fig. 13.24, pure tungsten with an optimal microstructure seems still to be best suited for structural applications. As demonstrated in the left-most image of Fig. 13.24, the W-5Ta alloy performs considerably worse than the W-1Ta alloy, presumably due to the higher concentration of intermetallic present.

13.4 AS-IRRADIATED MECHANICAL PROPERTIES 13.4.1 AS-IRRADIATED MECHANICAL PROPERTIES OF VANADIUM The effects of neutron irradiation on the mechanical properties of V-(4-5%)Cr-(4-5%)Ti alloys have been shown in several studies to be very detrimental at relatively low temperatures (<0.3TM, where TM is the melting temperature) even for low-dose (0.1–5 dpa) irradiation exposures [4, 6, 8, 12, 15, 19, 75–83]. In terms of tensile properties, the tensile curves shown in Fig. 13.25 following neutron irradiation in the High Flux Beam Reactor (HFBR) to 0.5 dpa at 110–420°C and to 0.1 dpa at 505°C reveal pronounced radiation hardening with concurrent severe reduction in strain-hardening capacity for irradiations at temperatures <400°C [78]. In this regime, significant increases in yield stress occur with yield peaks at uniform elongations <0.2%, followed by severe plastic instability (monotonically decreasing stress with increasing strain) for higher tensile strain levels. The slope of the tensile 800 110⬚C

270⬚C

Ttest ~ Tirr

700 325⬚C Engineering stress (MPa)

600 420⬚C 500

500⬚C

400 300 200 100

0.05

0 Normalized crosshead displacement

FIG. 13.25 Tensile stress vs strain curves for V-4Cr-4Ti following neutron irradiation in HFIR to 0.5 dpa at 110–420°C and to 0.1 dpa at 505°C [78]. The tensile curves are offset on the strain axis for clarity. From Snead LL, Zinkle SJ, Alexander DJ, Rowcliffe AF, Robertson JP, Eatherly WP. Summary of the investigation of low temperature, low dose radiation effects on the V-4Cr-4Ti alloy. Fusion Materials Semiannual Progress Report for Period Ending Dec 31, 1997. DOE/ER-0313/231997. p. 81–98. Original from L.L. Snead et al., Presented at 8th Int. Conf. on Fusion Reactor Materials, Sendai, (1997) to be publ. in Fusion Materials semiann. Prog Rep. for period ending Dec. 31 1997.

616

CHAPTER 13 REFRACTORY ALLOYS 800

25 Lower yield stress (irr) Lower yield stress (Unirr)

700

D F

H L 20

500

15

400 10

300

Uniform elongation (%)

Lower yield stress (MPa)

600

200 5

Uniform elongation (irr) Uniform elongation (Unirr) E G I

100

0

0 0

100

200

300

400

500

Temperature (°C) [temperature of irradiation and test]

FIG. 13.26 Temperature dependence of yield strength and uniform elongation for unirradiated V-4Cr-4Ti and after neutron irradiation in HFBR to 0.5 dpa at 110–420°C and to 0.1 dpa at 505°C [59, 78]. Tensile tests were performed at the irradiation temperatures.

stress-strain curves changes abruptly at elongations of
0.2%–0.4% for irradiations at temperatures <325°C. Conversely, for irradiations at temperatures >400°C, yielding occurs with no yield drop and the ensuing plastic deformation occurs with extensive strain hardening to the point of plastic instability at the ultimate tensile stress. Fig. 13.26 shows the temperature dependence of yield stress and uniform elongation for unirradiated V-4Cr4Ti before and after irradiation in HFBR to 0.5 dpa [59, 78]. The data indicate pronounced radiation hardening and loss of uniform elongation for temperatures up to 325°C, whereas progressive recovery of the irradiationhardening effects in V-4Cr-4Ti occurs for irradiations at temperatures >400°C. Above this irradiation temperature, the yield stress decreases and the uniform elongation increases with increasing irradiation temperatures until similar values occur for unirradiated and irradiated V-4Cr-4Ti at
500°C. Charpy impact test results of neutron-irradiated V-4Cr-4Ti also showed enhanced embrittlement behavior at relatively low temperatures and low-dose irradiation exposures. This is illustrated in Fig. 13.27 with absorbed energy data obtained from impact testing of machine-notched Charpy (MCVN) and pre-cracked Charpy (PCVN) specimens of V-4Cr-4Ti before and after neutron irradiation in HFBR to 0.5 dpa at 110–420°C [78]. In these figures, the tested specimens were miniaturized 1/3 CVN specimens with dimensions of
3.3  3.3  25 mm. In the unirradiated condition, the DBTT was near 200°C. However, neutron-irradiation exposures at 420°C resulted in severe embrittlement of V-4Cr-4Ti by shifting the DBTT to higher temperatures and causing brittle cleavage and intergranular fracture of MCVN specimens. The magnitude of the DBTT shift was 200–350°C (Fig. 13.27A). The

13.4 AS-IRRADIATED MECHANICAL PROPERTIES

617

20 ANNEALED 1000°C/2 h 1/3 CHARPY BLUNT NOTCH IRRADIATED IN HFBR TO 0.5 dpa

18 16

UNIRRADIATED IRRADIATED AT 108°C IRRADIATED AT 204°C IRRADIATED AT 235°C IRRADIATED AT 274°C IRRADIATED AT 315°C IRRADIATED AT 420°C

12

C

°C

8

8°

10

420

10

UNIRRADIATED

Energy (J)

14

23

4

5° C

31

5° C

6

2 0 –250 –200 –150 –100 –50

(A)

0

50

100 150 200 250 300 350

Temperature (°C)

Ductile to brittle transition temperature (°C)

250

200

150

100

50 MCVN 0

–50 100

(B)

PCVN

150

200

250

300

350

400

450

Irradiation temperature (°C)

FIG. 13.27 Charpy impact absorbed energy curves of V-4Cr-4Ti before and after low-dose (0.5 dpa) neutron irradiation at 110–420°C [78].

notch acuity effect observed for unirradiated and irradiated V-4Cr-4Ti led to even higher DBTT shifts of
50–120° C for PCVN specimens compared to MCVN specimens (Fig. 13.27B). One important issue that was resolved in the irradiation-embrittlement studies of V-4Cr-4Ti was that post-irradiation annealing at temperatures 650°C in vacuum restored the ductile failure behavior in specimens that had been embrittled by low-temperature neutron irradiation. Fig. 13.28 shows results obtained from post-irradiation annealing of V-4Cr-4Ti following neutron

618

CHAPTER 13 REFRACTORY ALLOYS

23°C Charpy impact energy (J)

15

12

9

V-4Cr-4Ti, 500-kg Heat #832665 double-annealed at »1050°C

Test at 23°C Test at –150°C

6

3

0 –200

EBR-II X-530 »4 dpa, »390°C 0

800 1000 200 400 600 Postirradiation annealing temperature (°C)

1200

FIG. 13.28 Impact energy measured on specimens of 500 kg heat of V-4Cr-4Ti after irradiation at
390°C to
4 dpa and postirradiation annealing at 23–1000°C [82].

irradiation at
390°C to 4 dpa in the Experimental Breeder Reactor-II (EBR-II) [63]. The neutron-irradiation exposure caused embrittlement of V-4Cr-4Ti by reducing the impact energy to <0.3 J in tests conducted at 150°C to +300°C (maximum tested temperature) in these miniaturized 1/3 CVN specimens. Following post-irradiation annealing at 650°C, the Charpy impact energy increased from <0.3 J to levels of
9–11 J for tests at 23°C, which was near the impact energy of
14 J for unirradiated V-4Cr-4Ti at 23°C (Fig. 13.27A). In microstructural investigations of irradiated V-(4-5)Cr-(4-5)Ti alloys, significant changes were observed at doses as low as 0.1 dpa over temperatures ranging from 110°C to 500°C [84–86]. Fig. 13.29 shows two examples of the radiation-induced microstructural changes that occurred in HFBR neutron-irradiated V-4Cr-4Ti specimens at 268°C and 390°C. Transmission electron microscopy (TEM) showed the formation of a high number density of
3nm diameter faulted dislocation loops (b ¼ a/2 h110i Burgers vectors) in specimens irradiated at <275°C (Fig. 13.29A). The faulted loops were found to have moderate dispersed barrier strengths (α < 0.5). TEM analysis of tensile tested specimens following irradiation at <275°C showed cleared dislocation channels
50 nm wide on [87] planes (Fig. 13.29A). Collectively, these microstructural results account for the pronounced radiation hardening and decrease in strain-hardening capacity to uniform elongations of <0.2% in tensile specimens at low irradiation temperatures (Fig. 13.25). For irradiation temperatures ranging from 275°C to 415°C, the dislocation loops changed from faulted loops to small perfect a/2 h111i loops with concomitant formation of large defect clusters lying on [001] habit planes at >315°C (Fig. 13.29B). TEM analysis of V-(4-5)Cr-(4-5)Ti alloys irradiated in EBR-II at 400°C also revealed the presence of a high number density of particles <6 nm diameter shown in Fig. 13.30 [88]. Diffraction from the particles showed radial streaks in the h200i direction near 3/4 h200i spots and tangential streaks near 2/3 h222i of the bcc matrix. Composition line profiles with electron energy loss spectroscopy showed Ti-enrichments of the particles. The diffraction characteristics of the particles were similar to those attributed to plate-shaped Ti-OCN (Fm3m space group) precipitates showing the Baker-Nutting Orientation Relationship, h001im //h011i p and [200]m // [200]p, between the matrix and precipitate [89]. For irradiation temperatures >415°C, the microstructural features were predominantly large [001] defect clusters that were enriched with Ti solute.

(A)

(B)

FIG. 13.29 TEM micrographs showing the microstructure of V-4Cr-4Ti irradiated to (A) 0.5 dpa at 268°C and tensile tested at 20°C and (B) 0.1 dpa at 390°C [85] In (B), the small arrows point to a/2 < 111> dislocation loops and the large arrows point to the [001] precipitates. The inset figure in (B) shows a magnified view of the 390°C microstructure.

Ti(30eV)/V(20eV)

0.15

0.1

0.05

0 0

10

20

30

40

50

Position (nm)

FIG. 13.30 Parallel electron energy loss spectroscopy line profile showing the correlation of Ti concentrations with the small particles in V-4Cr-4Ti irradiated at 400°C to 4 dpa [86]. From D.S. Gelles, P.M. Rice, S.J. Zinkle and H.M. Chung, “Microstructural Examination of Irradiated V-(4–5%)Cr-(4–5%)Ti”, Journal of Nuclear Materials, Vol. 258–263, (1998), p. 1380.

620

CHAPTER 13 REFRACTORY ALLOYS 1000 Ultimate tensile strength (MPa)

Ttest ~ Tirr 900

800

V-5Cr-5Ti 400⬚C, 4 dpa, EBR-II

V-4Cr-4Ti 330⬚C, 4-6 dpa, HFR CR+1050⬚C

TWCA/1050⬚C

700

600 V-4Cr-4Ti 200⬚C, 0.5 dpa, HFBR

500 10–6

10–5

0.0001 0.001 Strain rate (s–1)

0.01

0.1

FIG. 13.31 Ultimate tensile strength as a function of strain rate for neutron irradiated V-4Cr-4Ti and V-5Cr-5Ti at 200–400°C to 0.5–6 dpa [79].

The strain rate dependence of the tensile properties of neutron-irradiated V-(4-5)Cr-(4-5)Ti alloys were investigated and correlated with the DSA behavior of unirradiated V-4Cr-4Ti [59, 77, 79]. Fig. 13.31 shows a compilation of UTS data vs strain rate for V-(4-5)Cr-(4-5)Ti alloys irradiated to 0.5–6 dpa at 200–400°C. For V-4Cr4Ti specimens irradiated to 0.5 dpa at 200°C, the UTS increased slightly with increasing strain rate, consistent with a positive strain rate exponent that is commonly observed in irradiated materials. With irradiations to 4–6 dpa at 330°C, the UTS showed a weak, or slightly positive, strain rate exponent. Conversely, a slightly negative strain rate exponent for UTS was measured for V-5Cr-5Ti specimens irradiated to 4 dpa at the higher irradiation temperature of 400°C. The negative strain rate exponent (along with accompanying electrical resistivity measurements) suggests that a sizable concentration of interstitial O, C, and N solutes remained in the matrix of the irradiated V-Cr-Ti alloys and were not contained in Ti-OCN precipitates or defect clusters. The freely bound interstitial solutes were then able to interact with migrating dislocations during tensile testing causing the appearance of serrated flow in irradiated tensile curves, which is consistent with DSA [83]. Fig. 13.32 shows the SRS parameters obtained from tensile data for V-4Cr-4Ti before and after neutron irradiation in HFBR experiments to 0.1–4.0 dpa at 160°C to 500°C. The results showed that the irradiated strain rate exponent of the lower yield stress (LYS) was positive from 160°C to 500°C, but that the UTS changed from positive to negative values between
300°C and 400°C. The magnitude of the negative strain rate exponent of UTS was reduced after irradiation at >300°C compared to that of unirradiated V-4Cr-4Ti. The decrease in magnitude of the strain rate exponent for UTS indicated that a greater concentration of interstitial O, C, and N solutes were bound to small defect clusters or Ti-OCN precipitates in specimens irradiated at temperatures 400°C compared to unirradiated V-4Cr-4Ti (although a significant concentration of O, C, and N solutes was dissolved in the matrix of the irradiated samples at >300°C, as evident from the negative SRS parameter). A limited number of studies on the irradiation creep of V-4Cr-4Ti have been conducted and therefore understanding of effects of irradiation is still emerging [90–94]. Fig. 13.33 shows the creep strain rate (%/dpa units) plotted against applied stress for reactor irradiations performed in the ATR-A1 (to 5 dpa at 200°C and 300°C in lithium-filled capsules) [91–93], HFIR-12 J (to 6 dpa at 500°C in He filled capsules) [92], BR-10 (to 3 dpa at 445°C) [90], and HFIR-17 J experiments (to 3.7 dpa at 425°C and 600°C in lithium-filled capsules). Pressurized

13.4 AS-IRRADIATED MECHANICAL PROPERTIES

621

0.10 LYS unirradiated

0.08

FS unirradiate

Strain rate sensitivity

UTS unirradiated LYS (irradiated)

0.06

UTS (irradiated)

0.04 0.02 0.00 –0.02 –0.04 0

100

200

300

400

500

600

700

800

Temperature (°C)

FIG. 13.32 Comparison of the strain rate sensitivity parameter as a function of temperature for unirradiated V-4Cr-4Ti before and after fission neutron irradiation in HFBR to 0.5 dpa at 160–500°C [16, 78, 79]. Values for LYS and UTS were similar at the 160°C and 200°C irradiation temperatures.

0.8 HFIR 17J, 425°C/3.7 dpa HFIR 17J, 600°C/3.7 dpa ATR A1, 200°C/3 dpa ATR A1, 300°C/5 dpa HFIR 12J, 500°C/6 dpa Troyanov, 445°C/3 dpa

Effective strain (%/dpa)

0.7 0.6 0.5

V-4Cr-4Ti

0.4 0.3 0.2 0.1 0.0 0

50

100

150

Effective stress (MPa)

FIG. 13.33 Comparison of irradiation creep data for V-4Cr-4Ti [94].

200

622

CHAPTER 13 REFRACTORY ALLOYS

creep tube specimens were used in the ATR-A1, HFIR 12 J, and HFIR-17 J experiments and thin-walled tube specimens loaded in torsion were used in the BR-10 experiment. The creep data from the ATR-A1 experiment showed no linear stress dependence of creep strain due to the large scatter in results, which may have been the result of nonuniform irradiation temperatures and neutron flux gradients experienced by the specimens. Linear stress dependence of creep strain was observed for HFIR-12 J, HFIR-17 J, and BR-10, although bilinear creep behavior was also observed in the BR-10 experiment and a slightly nonlinear relationship was observed in the HFIR-17 J experiment at 600°C above
110 MPa. The nonlinear relationship may suggest that thermal creep mechanisms begin to dominate at >110 MPa and 600°C. The discrepancies with data showing linear vs bilinear creep behavior could not be explained. The creep coefficients measured from the data varied from 1.4  106 to 11  106/MPa dpa for irradiation doses of 3–6 dpa. However, the data showed no clear dependence on irradiation temperature or neutron dose for temperatures of
400–600°C and doses of
3–6 dpa. In general, the understanding of irradiation creep behavior of V-4Cr-4Ti remains unclear due to complexities in the production and fabrication of the alloys and neutron-irradiation experiments including the effects of dose, temperature, and environment on the specimens. Additional higher dose data over a wide range of temperatures are needed to improve understanding of irradiation creep in V alloys.

13.4.2 AS-IRRADIATED MECHANICAL PROPERTIES OF NIOBIUM The studies of neutron-irradiation effects on the mechanical properties of Nb alloys have been primarily based on temperature and neutron fluence ranges projected for space reactor operating conditions, along with limited studies for sodium-cooled fast reactor applications. The space reactor operating goals were for 2–7 years at temperatures of 1100°C to 1400°C (1373–1673 K) to fast fluences of 0.4–4  1022 n/cm2 in liquid (Li, K) or gas (He) environments [95, 96]. The main concerns for Nb alloys consist of radiation hardening at all temperatures up to
0.5 Tm, irradiation-assisted creep at
0.2–0.5 Tm, void or cavity swelling at <0.5 Tm, and He embrittlement at >0.5 Tm, where the melting temperature Tm ¼ 2741 K [95, 97, 98]. From the early development of Nb alloys in the mid-1950s to the present, there have been limited studies of irradiation effects and most have been on examinations of the microstructure and tensile properties of unalloyed Nb, Nb-1Zr, and Nb-1Zr-0.1C (PWC-11) [95]. Irradiation studies of Nb and Nb-1Zr showed rapid hardening occurs with complete loss of strain-hardening capacity due to rapid necking at temperatures <423 K [99–102]. Fig. 13.34 shows the reductions in uniform and total elongations that occurred for Nb-1Zr after irradiation to
0.005–12.3 dpa at
343 K [99]. Although the uniform elongation decreased to zero for irradiation doses >0.02 dpa, the total elongation remained at
10% up to 12.3 dpa and the tensile fracture modes were ductile with high reduction in area. In the radiation-hardening regime, the localized plastic deformation after yielding occurs by dislocation channeling of accumulated radiation-induced defect clusters and dislocation loops, which is shown in Fig. 13.35 [32, 99]. Fig. 13.36 shows tensile data obtained from several studies of Nb-1Zr and Nb-1Zr-0.1C exposed to neutron irradiation to 0.1–5 dpa and tensile tested at the irradiation temperatures ranging from
300 to 1573 K [32, 95, 103–105]. The irradiated UTS increases significantly below
800 K and then decreases monotonically with increasing irradiation temperature to
1000 K where it converges with the unirradiated UTS data. Since Nb-1Zr can become embrittled due to pickup of O, C, and N from the surrounding atmosphere at temperatures above
800 K, special care must be taken in irradiation experiments to minimize such contamination effects. The irradiated uniform elongation is nearly zero up to
723 K and begins to slowly increase at higher irradiation temperatures. The irradiated uniform elongation approaches acceptable “ductile” values of
5% for irradiation and test temperatures of 1070 K or higher [105]. Irradiation temperatures of
0.3–0.6 Tm can cause void swelling, He embrittlement, and irradiation creep in Nb alloys depending on the neutron fluences. For Nb and Nb-1Zr, swelling associated with void formation is <2% for fluences up to
2–5  1026 n/m2, E > 0.1 MeV [32, 99, 105]. Fig. 13.37 shows the temperature range for void

13.4 AS-IRRADIATED MECHANICAL PROPERTIES

623

50 Wiffen (1978)

Tensile elongation (%)

40 Total elongation

30

1 dpa = 2x10 21n cm–2 (E>0.1 MeV)

20

10

Uniform elongation

0 0 1 0.000

0.001

0.01

0.1

1

10

100

Displacement damage (DPA)

FIG. 13.34 Dose dependence of the total and uniform elongation of Nb-1Zr neutron irradiated at a temperature near 70°C and tested at room temperature [32, 98].

FIG. 13.35 Deformation microstructure of Nb-1Zr neutron irradiated at a temperature near 70°C and tensile tested at room temperature [32, 98].

624

CHAPTER 13 REFRACTORY ALLOYS

FIG. 13.36 Ultimate tensile strength (UTS) and uniform elongation (eu) of Nb-1Zr and Nb-1Zr-0.1C neutron irradiated to 0.1–5 dpa and tensile tested at the irradiation temperature [98]. Adapted from Zinkle SJ, Wiffen FW. Radiation effects in refractory alloys. In: El-Genk MS, editor. Space Technology and Applications International Forum-STAIF 2004 (AIP Conf Proc vol 699). Melville, NY: American Institute of Physics; 2004. p. 733–40. Figure 2.

Swelling (%)

3.0 POWELL et. al. JANG & MOTEFF SPRAGUE et. al. WIFFEN WIFFEN MICHEL & SMITH

2.0

5.0

2.5 1.0

0

17.6 13.2 17.2 16.7 14.0 15.8 15.3 17.8 5.4 5.0 1.9 5.0 1.1 400

1.9

500

Nb – 1 Zr Neutron Irradiation (Fluence shown in units 1022 N cm–2)

5.0 5.0

800 600 700 Irradiation temperature (⬚C)

5.0 900

1000

FIG. 13.37 Summary of void swelling behavior of neutron-irradiated Nb-1Zr [31, 99]. From Buckman RW, Jr. A review of tantalum and niobium alloy production. In: Cooper RH, Jr., Hoffman EE, editors. Proc Symp on Refractory Alloy Technology for Space Nuclear Power Applications, CONF-8308130: Oak Ridge National Lab; 1984. p. 86–97.101. Original from R. W. Powell, D. T. Peterson, M. K. Zimmerschiec, and J. F. Bates, Swelling of Several Commercial Alloys Following High Fluence Neutron Irradiation, J. NucL Mater., 103–104: 969–374 (1981).

swelling in Nb-1Zr is
800–1200 K for doses up to
35 dpa, with maximum swelling occurring near 1073 K [99]. Neutron irradiation [n, α) reactions can also cause embrittlement of Nb alloys by formation of He bubbles on grain boundaries. However, experimental studies have reported no severe embrittlement for He ions implanted to concentrations up to 100–500 appm, indicating Nb and Nb-1Zr have relatively good resistance to He embrittlement [87, 106, 107]. Experimental data on irradiation creep of Nb-1Zr are very limited, but the results suggest that irradiation creep rate would only produce minor deformations of <0.3% based on projected space reactor operating conditions [32, 104, 108].

13.4 AS-IRRADIATED MECHANICAL PROPERTIES

625

The effect of irradiation on the DBTT and fracture properties of Nb-1Zr and PWC-11 is nonexistent. However, estimates for Nb-1Zr fracture behavior based on irradiated tensile strengths at temperatures below
600–700 K suggests fracture toughness would be a concern for irradiation temperatures <800 K [32].

13.4.3 AS-IRRADIATED MECHANICAL PROPERTIES OF MOLYBDENUM As with other BCC metals, molybdenum is subject to severe base metal embrittlement for typical temperature of interest in nuclear applications. Molybdenum, and for example tungsten, also share similar methods of fabrication and grain boundary susceptibility to neutron-irradiation embrittlement leading to material failure. Also, in materials such as some powder metallurgically derived and ODS molybdenum developed for radiation resistance, significant reduction during process rolling leads to anisotropic nonirradiated mechanical properties and response to irradiation. Unlike tungsten, in the nuclear application temperature range the defects responsible for embrittlement appreciably evolve through a homogeneous nucleation and growth process following the initial neutron-induced cascade [109] and therefore embrittlement mitigation strategies such as alloying or second phase additions can be considered. However, while molybdenum has historically been a popular conceptual material for nuclear energy applications leading to extensive R&D, limited success toward development of ductile nuclear-grade structural molybdenum has been achieved. It is noted that improved irradiation performance has been achieved by developing higher purity materials (particularly minimizing O, C, and N impurities) and materials with cleaner and therefore more irradiation-resistant grain boundaries. It is noted that this is in contrast to the common strategy of adding grain boundary precipitates to enhance thermal creep resistance and irradiation stability for nuclear alloys. However, this underscores the particular susceptibility of refractory materials to grain boundary impurities. For much the same reason, the most commonly produced materials, specifically hot-pressed and sintered powders, are not discussed here because of the inherent impurity content and as-irradiated embrittlement of their grain boundaries. In fast and mixed spectrum reactors, 98-Mo (isotopically the most abundant) transmutes through a (n,γ) reaction first to 99-Mo and then to 99-Tc, a relatively long lived isotope that eventually produces 101-Ru. The majority of Mo transmutations simply create other Mo isotopes. Chain reactions with other isotopes produce similar, minor amounts of Zr and Nb, as discussed [110]. For doses <10 dpa, the transmutant levels are on the order of <1000 ppm and have negligible effect on hardening, although precipitates can be observed at higher irradiation temperatures [48]. Given the high melting temperature of molybdenum (2623°C), the homologous temperature range for its practical nuclear structural application temperatures (300–1000°C) is relatively low: 0.20–0.44Tm. For temperatures as low as 80°C, dislocation loops, rafts, and cavities (voids) are the major irradiation-induced defects observed in LCAC molybdenum microstructure [111]. At that temperature, loops were discernable by TEM at damage levels as low as 0.1 dpa with cavities detected by positron annihilation at levels <0.01 dpa, though higher temperature irradiation is required for cavities to become visible through TEM. Rafts of aligned dislocation loops emerge at 80°C irradiation at doses >0.01 dpa [111]. With increasing temperature (300°C), voids on the order of 1 nm become visible in TEM and are of similar size for LCAC and ODS molybdenum. Fig. 13.38 provides an overview schematic representation of the microstructural evolution in Mo as a function of dose and temperature as taken from the literature [48]. The void microstructure increases in number density with increasing dose and at first increases in size, then decreases within the limits of the current published work. At sufficiently high dose, the curious phenomenon of the ordered void lattice, a phenomenon first observed in molybdenum, can occur [112–115]. Increasing irradiation temperature results in larger voids of lower number density with a weak dependence on material microstructure: finer particle size materials such as ODS molybdenum result in somewhat finer scale void formation [48, 115]. Very high dose irradiation at high temperature leads to the formation of a low concentration of fine transmutation-induced precipitates [48]. As summarized in Fig. 13.38, the dominance of voids in the irradiated microstructure of molybdenum at intermediate to high temperatures raises the issue of irradiation-induced swelling. Void swelling under V+ ion irradiation has been studied to a damage level of 50 dpa (
900–1510°C), suggesting the potential for significant

626

CHAPTER 13 REFRACTORY ALLOYS

FIG. 13.38 Microstructural evolution map for irradiated molybdenum.

swelling [116]. Near the lower temperatures studied, a saturation level for void swelling (uncommon in most nuclear materials, and perhaps related to the propensity for void lattice formation in Mo alloys) approached 4%, though the saturation void swelling was closer to 10% at the upper end of the irradiation temperatures studied [116]. A compilation of the neutron-irradiation data has been assembled by Leonard [117] indicating <4% swelling for Mo irradiated to 8  1026 n/m2 (E > 0.1 MeV,
35 dpa) in the 400–900°C range [113]. Therefore, void swelling does not appear to produce unacceptably large-dimensional changes in neutron-irradiated Mo, at least up to moderately high doses (100 dpa) that have been studied to date. The highest dose data, such as that replotted in Fig. 13.39, suggests that a peak in swelling occurs in a temperature range from 550°C to 650°C, perhaps with a slight shift depending on the material type. For the lower end of the nuclear application temperature range, the irradiation-induced dislocation loops and tiny cavities result in significant matrix hardening. It is noted that at lower irradiation temperature and dose, and specifically as demonstrated in the systematic 80°C neutron-irradiation studies of Li [111], very low dose yield softening (doses <0.005 dpa) precedes the yield-hardening regime for Mo. However, for neutron irradiation in the range of 300–600°C on Mo materials selected and developed for irradiation resistance, substantial increased hardness occurs with increasing dose. Fig. 13.40 compares the change in room-temperature hardness for LCAC Mo, ODS Mo, and TZM irradiated in a mixed spectrum reactor at
300°C, 600°C, and 900°C [36]. Mo materials irradiated at nominally 300°C and 600°C are seen to rapidly harden with a very weak temperature dependence and exhibit an approach to saturation at a hardness value approximately twice that of the initial value. For higher temperature irradiation (900°C in Fig. 13.40), fewer loops and small cavities are produced and the void, which is less effective hardening agent, becomes the dominant microstructural feature. Moreover, irradiation hardening at temperatures approaching 0.4 of the homologous temperature competes with anneal-softening of cold work,

FIG. 13.39 Swelling of cold-worked and aged (CWA) and TZM molybdenum neutron irradiated to high dose as a function of temperature [116].

FIG. 13.40 Relative change in hardness of stress-relieved neutron-irradiated molybdenum and molybdenum alloys [36].

628

CHAPTER 13 REFRACTORY ALLOYS

FIG. 13.41 Relative change in hardness as a function of Re content for recrystallized and stress relieved Mo. Rhenium transmutation produced through pre-irradiation alloying (FFTF-MOTA irradiation) [118].

yielding a much-reduced relative effect of irradiation. At irradiation temperatures in the 900–1100°C range, recrystallization of stress-relieved materials would be expected. A similar trend can be seen for Mo-Re alloys irradiated in the 400–600°C range (significant hardening) and at 800°C (anneal softening), as shown in Fig. 13.41 [118–120]. In this figure, rather than having a ductilizing effect such as has been observed in unirradiated Mo alloys, the presence of Re tends to increase the irradiation hardening of the matrix in part through the promotion of the embrittling σ and χ phases. While such second phases are not in equilibrium for the Mo-Re alloy Re contents of less than
61%-eutectic temperature for the σ phase of
1130°C, and
77%-for the χ phase, under nonequilibrium irradiation conditions the undersized Re solute precipitates into the Re-rich phases. This phenomenon, known as Radiation Induced Segregation, is briefly summarized in Chapter 5 in this book. As expected, the significant matrix hardening results in increased tensile strength and reduced ductility of Mo. Many approaches have been explored to mitigate this issue including alloying, ultra-pure alloys, and incorporation of nano-oxides, with moderate success. However, irradiation embrittlement remains a central issue to molybdenum in the practical temperature ranges for nuclear application. In contrast with the limited data on the effect of irradiation on the deformation and fracture of tungsten and tungsten alloys, a fair body of literature exists for modern molybdenum alloys including both fundamental irradiation effects studies and more practical engineering data on alloys developed for radiation damage resistance. Fig. 13.42 provides a simple example of the progressive hardening and loss of tensile ductility for relatively low-temperature irradiated Mo (nominally 80°C) over a range of very modest irradiation doses up to 0.3 dpa. As seen, this low-carbon arc-cast molybdenum, which is currently a leading irradiation-tolerant form of molybdenum, possesses over 50% total elongation prior to irradiation. The data clearly show progressive hardening (increased yield/ultimate stress) and rapid loss of ductility with near-complete loss of strain tolerance by 0.072 dpa (2  1024 n/m2, E > 0.1 MeV.) As would be expected given the evolution of microstructure presented in Fig. 13.38 and the previous discussion on hardening, as the irradiation temperature is increased from
80°C molybdenum embrittlement becomes less severe. However, for LCAC molybdenum and high-purity ODS molybdenum the transition temperature from ductile to brittle behavior for un-notched (smooth) tensile specimens shifts from well below room temperature in the nonirradiated condition to hundreds of degrees Celsius in the as-irradiated condition [121]. This shift in temperature is somewhat mitigated as the

13.4 AS-IRRADIATED MECHANICAL PROPERTIES

629

FIG. 13.42 The effect of mixed-spectrum reactor irradiation at 80°C on the tensile behavior of low-carbon arc-cast molybdenum [111]. Mixed spectrum High Flux Isotope Reactor neutron irradiation. From Li M, Byun TS, Snead LL, Zinkle SJ. Defect Cluster Formation and Radiation Hardening in Molybdenum Neutron-Irradiated at 80°C. Journal of Nuclear Materials. 2007;367–370:817–22.

irradiation temperature is increased, though tensile embrittlement remains an issue for Mo for irradiation temperatures up to several 100°C. Fig 13.43 provides data comprising a broad transition point from brittle (lower left) to ductile (upper right) behavior for smooth tensile specimens as a function of irradiation temperature for a range of nominally radiation-resistant forms of molybdenum: TZM, LCAC Mo, and ODS (La oxide) Mo. As seen from the figure, the nonirradiated ductile-brittle transition for nonirradiated specimens is approximately 100°C, increasing to a broad band between nominally 400–800°C in the as-irradiated condition. For practical applications, it is important to note that this data describes tensile ductility, which should not be confused with the material’s ability to resist crack propagation, for example, fracture toughness. It has been demonstrated by Cockeram [122, 123], as typical of the fracture of most metallic systems, that tensile ductility of irradiated molybdenum is likely an optimistic measure of material performance (e.g., the brittle transition would be lower for a fracture toughness specimen as compared to a tensile specimen). An irradiation of the same materials depicted in Fig. 13.43 has been carried out on disc compact tension specimens and the fracture toughness was subsequently measured. In the nonirradiated condition, the fracture toughness of TZM, LCAC, and ODS forms of molybdenum in various orientations and recrystallized and strain-relieved conditions dipped below 30 MPa-m1/2 fracture toughness in the temperature range between
 200°C and 0°C. For materials neutron irradiated at 386–574°C, the point associated with very low fracture toughness (<30 MPa-m1/2) was in the range of 300–600°C [122].

13.4.4 AS-IRRADIATED MECHANICAL PROPERTIES OF TUNGSTEN As discussed in Section 13.2.4, the tungsten fabrication route has profound influence on the resulting ductility and mechanical performance of the finished product in the absence of irradiation. Similarly, the response of tungsten or tungsten alloys to neutron irradiation is also strongly influenced by the processing route for fabrication.

630

CHAPTER 13 REFRACTORY ALLOYS

FIG. 13.43 The effect of irradiation on the brittle transition temperature for smooth tensile specimens of Mo and TZM [70]. Mixed spectrum High Flux Isotope Reactor neutron irradiation in the 0.23–44  1025 n/m2 (E > 0.1 MeV) [121a–121f, 142].

Experimental studies reinforced with molecular dynamics simulations indicate that in metals with FCC crystal structure such as Cu, Ni, and Au, the formation of immobile “sessile” defect clusters primary occurs directly within the neutron-induced displacement cascade. This is in contrast with the behavior of BCC iron, for which defect clusters primarily form through a nucleation and growth process following the initial cascade event. In the latter case, such as in many steels, it is possible to tailor the alloy behavior under irradiation through mechanisms such as cold work, varying the type and level of interstitial solutes, and/or inclusion of secondary phases. Interestingly, as the atomic number of BCC metals increases, the relative impact of sessile defect formation and therefore the irradiation embrittlement increases. Molybdenum, which has mass between iron and tungsten, contains a mixture of sessile in-cascade and evolving defect clusters [69] while in-cascade sessile defect clusters are assumed to dominate the more massive tungsten. Hardening centers such as dislocation loops have been observed to form directly within tungsten displacement cascades [124]. As previously mentioned, rhenium has a ductilizing effect on unirradiated molybdenum, effectively reducing the BDT temperature to below room temperature. The current understanding of this effect is that, at low concentrations, rhenium softens the material through reduction of the Peierls stress while at higher concentrations it beneficially scavenges oxygen that would otherwise migrate to and embrittle grain boundaries. Upon neutron irradiation, and especially in mixed and thermal spectrum fission reactors, copious amounts of rhenium are built into tungsten through transmutation. This can be seen through inspection of Fig. 13.44, giving the atomic percent of transmuted rhenium as a function of dpa for a mixed spectrum fission reactor (in this case HFIR, a water-cooled thermal reactor used for materials irradiation studies), a fast reactor (a previous fast spectrum materials irradiation reactor), and a D-T fusion first wall spectrum (STARFIRE study). For the dose range of the plot, which can be

13.4 AS-IRRADIATED MECHANICAL PROPERTIES

631

FIG. 13.44 Neutron-induced transmutation concentration of rhenium from tungsten in various neutron spectra. Based on L.R. Greenwood, F.A. Garner, Transmutation of Mo, Re, W, Hf and V in various irradiation test facilities and STARFIRE, J. Nucl. Mater. 212–215 (1994) 635–639.

roughly interpreted as the yearly equivalent in a fission materials test reactor or fusion power system, several to ten atom percent of rhenium is produced and subsequently transmuted to osmium, which also accumulates within the tungsten. The effect of rhenium on irradiated microstructure has been historically studied [125, 126] and continues to be evaluated [127–129]. While the overall mechanism is somewhat in dispute, it is clear that the matrix embrittlement due to fast neutrons far outweighs any ductilizing effect of rhenium, especially at low rhenium levels (<1%) [125]. It is noted that Hasegawa [130] has reported that percent levels of rhenium in as-irradiated alloy do marginally reduce the Vickers hardening (ΔH) for low damage levels (<0.4 dpa). In a separate work, for higher levels of rhenium, the formation of the embrittling χ phase has been observed in W-26 Re irradiated in the temperature range of 373–800°C at doses of 2–9.5 dpa [131]. In any event, the use of rhenium as a route toward enhancing the ductility of as-irradiated tungsten seems ineffective. However, alloying or build-in of rhenium through transmutation does affect other thermophysical properties and the microstructural evolution. Rhenium is understood to suppress the growth of voids and therefore the amount of swelling [126]. In one of the few systematic studies on swelling, Matolich [132] compared the swelling in pure sintered tungsten to W-25Re irradiated in a fast spectrum (to 5.5  1025 n/m2 (E > 0.1 MeV)) in the temperature range of 700–1200 K. Over this irradiation temperature range, pure tungsten yields a bell-curve behavior in swelling with a maximum swelling of about 1.6% at
1100°C while the W-25% Re sample exhibits negligible swelling (
0.1%) at all temperatures. Finally, the presence of metallic rhenium will have a large negative impact on thermal conductivity, outweighing any impact of fast neutron damage on conductivity degradation. For the maximum concentrations schematically shown in Fig. 13.43 (
10%), tungsten would be expected degrade in thermal conductivity to approximately one-third of its original value [133]. Given the extremely high melting temperature of tungsten (3423°C), the homologous temperature range for its practical nuclear structural application temperatures (300–1000°C) is relatively low: 0.15–0.35 Tm. Given this, at the lower end of this range one would expect the irradiated tungsten microstructure to consist of finely dispersed defects of ill-defined character, which would coarsen as irradiation temperature and dose increased. Fig. 13.45 provides a map for the evolution in microstructure vs dose for pure tungsten spanning this temperature range, though at a relatively modest dose level (<2 dpa) compared to component lifetimes that would be expected in a nuclear power application. At the lowest temperature, the dense population of tiny defects is difficult to resolve,

632

CHAPTER 13 REFRACTORY ALLOYS

FIG. 13.45 Microstructural evolution of pure tungsten as a function of mixed-spectrum neutron irradiation. Based on A. Hasegawa, T. Tanno, S. Nogami, T. Tanaka, Property change mechanism in tungsten under neutron irradiation in various reactors, J. Nucl. Mater. 417 (2011) 491–494.

though with increasing temperature and dose both loops and small voids become evident. As dose and temperature are further increased, voids coarsen and void-lattice can occur. Tanno [134] has observed the formation of a void lattice by 1.54 dpa at 750°C, significantly lower than previously reported [135]. For very high dose, consistent with high build-in of rhenium, or for alloys initially containing rhenium, needle-like precipitates of χ-phase rhenium become evident [126, 131, 134, 136]. The impact of irradiation in this relatively low homologous-temperature regime is to produce significant hardening in the tungsten crystal as can be seen through inspection of Fig. 13.46 [137]. In this data set, the Vicker’s hardness of single-crystal W is provided for a mixed spectrum neutron irradiation spanning the temperature range of 90–750°C over a few orders of magnitude fluence. Within the significant statistical variation, there does not appear to be a strong function of irradiation temperature in this temperature/dose range. Moreover, the irradiation hardening of sintered, polycrystalline tungsten appears to be similar to that of the single-crystal W. As mentioned in the introduction section, the properties of tungsten are a strong function of both the grain orientation and grain boundaries, with the mechanical properties of grain boundaries profoundly influenced by the fabrication route and post-fabrication processing such as deformation and annealing to produce strain relief or recrystallization. Moreover, it is assumed (given the dearth of data) that sintered forms of tungsten with boundaries weakened by the presence of impurities such as oxygen will undergo exaggerated irradiation-induced degradation. For the single-crystal materials shown in Fig. 13.46, a significant increase in yield strength occurs accompanied by loss of ductility for both [100, 110] orientations. This impact on ductility occurs by
1  1024 n/m2 (E > 0.1 MeV.) Fig. 13.47 provides a snapshot of the limited available tensile data on sintered pure polycrystalline tungsten

FIG. 13.46 Effect of mixed spectrum neutron irradiation on the hardness of single-crystal and sintered tungsten [137].

FIG. 13.47 Tensile behavior of polycrystalline tungsten under neutron irradiation [138–140].

634

CHAPTER 13 REFRACTORY ALLOYS

FIG. 13.48 Brittle-to-ductile transition temperature for relatively low-temperature mixed spectrum neutron-irradiated sintered tungsten and tungsten-10% rhenium alloys in the range of 250–300°C [141].

irradiated in mixed and fast spectrum reactors at 90–425°C. As expected given the typical hardening of Fig. 13.46, yield strength also exhibits a rapid increase with dose. This hardening, and likely a weakening of the grain boundaries, also results in a rapid loss of ductility with a complete loss of uniform and total elongation by
1  1024 n/m2 (E > 0.1 MeV), similar to that noted in single-crystal material. The work of Krautwasser [141] summarized in Fig. 13.48 for non-precracked bars provides the irradiation-induced shift in DBT temperature of both pure sintered tungsten and W-10Re indicating the DBT shifts to well above room temperature even for the lowest dose studied, 0.52  1025 n/m2 (E > 0.1 MeV), and approaches 900°C for a dose an order of magnitude higher, with no apparent effect of rhenium.

13.4.5 SUMMARY AND CONCLUSIONS Refractory metals have historically been, and continue to be, perennial candidates for high-temperature nuclear systems. However, their application has been predominantly notional or limited to nonstructural applications due largely to the issues highlighted in this chapter: challenging manufacturing issues associated with sensitivity to O, C, N pickup, a structure with inherently low as-fabricated ductility in most cases (particularly for Mo and W alloys), and pronounced embrittlement due to neutron irradiation. While the irradiation-induced embrittlement at the crystal level is extremely challenging to mitigate for this class of materials, strides are being made in our ability to both fabricate complicated (e.g., graded composition layers) refractory structures and in our understanding of how to design structures with low-ductility materials. Given this coupled processing and design understanding, it is becoming more likely that refractory materials will find more widespread and important application within nuclear systems in the near future.

REFERENCES

635

REFERENCES [1] Fang ZZ. (Editor in Chief) International Journal of Refractory Metals and Hard Materials. [2] O.N. Carlson, F.A. Schmidt, E.N. Dalder, T. Grobstein, C.S. Olsen (Eds.), Vanadium Alloy Development in Evolution of Refractory Metals and Alloys, The Minerals, Metals, and Materials Society, 1994. [3] Ross RB. Metallic Materials Specification Handbook (Springer US). 1980. [4] H.M. Chung, B.A. Loomis, D.L. Smith, Development and testing of vanadium alloys for fusion applications, J. Nucl. Mater. 239 (1996) 139–156. [5] D.L. Smith, M.C. Billone, K. Natesan, Vanadium-base alloys for fusion first-wall application, Int. J. Refract. Met. Hard Mater. 18 (2000) 231. [6] H.R. Thresh, A.G. Hins, D.L. Smith, The fabrication of vanadium base alloys as canditate fusion reactor materials, J. Nucl. Mater. 155–157 (1988) 608. [7] H. B€ohm, W. Dienst, H. Hauck, H.J. Laue, Irradiation effects on the mechanical properties of vanadium-base alloys, in: W.C. Leslie (Ed.), Effects of Radiation on Structural Materials, American Society for Testing and Materials, Philadelphia, 1967, pp. 95–106. ASTM STP 426. [8] D.L. Smith, H.M. Chung, B.A. Loomis, H.-C. Tsai, Reference vanadium alloy V-4Cr-4Ti for fusion application, J. Nucl. Mater. 233–237 (1996) 356–363. [9] H.M. Chung, H.-C. Tsai, D.L. Smith, R. Peterson, C. Curtis, C. Wojcik, et al., Fabrication of 500 kg heat of V-4Cr-4Ti. Fusion Materials Semiannual Progress Report for Period ending Sept 30, 1994, DOE/ER-0313/17, Oak Ridge National Lab, 1994, pp. 178–182. [10] B.A. Loomis, O.N. Carlson, Investigation of the Brittle-Ductile transition in vanadium, in: W. Clough (Ed.), Reactive Metals, Interscience, London, 1959, pp. 227–243. [11] R.W. Buckman, Consolidation and fabrication of vanadium and vanadium-base alloys, Int. Metals Rev. 25 (1980) 158. [12] R.J. Kurtz, K. Abe, V.M. Chernov, V.A. Kazakov, G.E. Lucas, H. Matsui, et al., Critical issues and current status of vanadium alloys for fusion energy applications, J. Nucl. Mater. 283–287 (2000) 70–78. [13] A.F. Rowcliffe, D.T. Hoelzer, Microstructural inhomogeneities in V-4Cr-4Ti mechanical property specimens prepared for irradiation experiments in HFBR, BOR-60, and HFIR. Fusion Materials Semiannual Progress Report for Period ending December 31, 1998, DOE/ER-0313/25, Oak Ridge National Lab, 1998, pp. 42–58. [14] R.J. Kurtz, M.L. Hamilton, H. Li, Grain boundary chemistry and heat treatment effects on the ductile-to-brittle transition behavior of vanadium alloys, J. Nucl. Mater. 258–263 (1998) 1375–1379. [15] S.J. Zinkle, H. Matsui, D.L. Smith, A.F. Rowcliffe, E. van Osch, K. Abe, et al., Research and development on vanadium alloys for fusion applications, J. Nucl. Mater. 258–263 (1998) 205–214. [16] D.T. Hoelzer, Structural Analysis of Ti-Oxycarbonitrides in V-Cr-Ti Based Alloys. In Fusion Materials Semiannual Progress Report for Period ending December 31, 1998, DOE/ER-0313/25, p. 59. [17] T. Muroga, T. Nagasaka, A. Iiyoshi, A. Kawabata, S. Sakurai, NIFS program for large ingot production of a V-Cr-Ti alloy, J. Nucl. Mater. 283–287 (2000) 711–715. [18] T. Muroga, T. Nagasaka, in: Development of manufacturing technology for high purity low activation vanadium alloys, 18th IAEA Fusion Energy Conference, Sorento, Italy, 2000. [19] T. Muroga, T. Nagasaka, K. Abe, V.M. Chernov, H. Matsui, D.L. Smith, Vanadium alloys-overview and recent results, J. Nucl. Mater. 307–311 (2002) 547. [20] A.K. Ikov, V.M. Chernov, M.M. Potapenko, I.N. Gubkin, C.A. Drobyshev, V.S. Zurabov, Development of production process and study of low-activity V-(4-5)%Ti-(4-5)%Cr structural alloys for thermonuclear reactors, Metal Sci. Heat Treat. 46 (2004) 497. [21] A.N. Tyumentsev, I.A. Ditenberg, K.V. Grinyaev, V.M. Chernov, M.M. Potapenko, Multi-directional forge molding as a promising method of enhancement of mechanical properties of V-4Cr-4Ti alloys, J. Nucl. Mater. 329–333 (2011) 429. [22] V.M. Chernov, M.M. Potapenko, V.A. Drobyshev, D. Blokhin, N. Budylkin, E. Mironova, et al., in: Low activation alloy V-4Cr-4Ti for fusion and fission power reactors-the RF results, Presented O2B2:11th ISFNT, Barcelona, Spain, 2013. [23] H.Y. Fu, J.M. Chen, P.F. Zheng, T. Nagasaka, T. Muroga, Z. Li, et al., Fabrication using electron beam melting of a V-4Cr-4Ti alloy and its thermo-mechanical strengthening study, J. Nucl. Mater. 442 (Suppl. 1) (2013) S336. [24] J. Chen, P. Wang, H. Fu, Z. Xu, in: Research of low activation structural materials for fusion power reactor in SWIP, 24th IAEA Fusion Energy Conference, 2012. [25] J.M. Chen, C.M. Chernov, R.J. Kurtz, T. Muroga, Overview of the vanadium alloy researches for fusion reactors, J. Nucl. Mater. 417 (2011) 289.

636

CHAPTER 13 REFRACTORY ALLOYS

[26] E.M. Sherwood, Columbium, in: C.A. Hampel (Ed.), Rare Metals Handbook, Robert Krieger Publishing Co., Huntington, NY, 1971, pp. 149–177. [27] L.J. Pionke, J.W. Davis, Technical assessment of niobium alloys data base for fusion reactor applications. McDonnell Douglas, 1979 August, 1979. Report No.: C00-4247-2. [28] J.R. Stewart, W. Lieberman, G.H. Rowe, Recovery and recrystallization of columbium-1.0% zirconium alloy, in: D. L. Douglass, F.W. Kunz (Eds.), Columbium Metallurgy, Interscience Publishers, New York, 1961, pp. 407–434. [29] E.S. Bartlett, J.A. Houck, Physical and Mechanical Properties of Columbium and Columbium-Base Alloys, Defense Metals Information Center, Battelle Memorial Institute, Coloumbus, OH, 1960. Report No.: DMIC Report 125. [30] R.G. Frank, Recent advances in columbium alloys, in: I. Machlin, R.T. Begley, E.D. Weisert (Eds.), Refractory Metal Alloys Metallurgy and Technology, Plenum Press, New York, 1968, pp. 325–372. [31] R.W. Buckman Jr., in: R.H. Cooper Jr., E.E. Hoffman (Eds.), A review of tantalum and niobium alloy production, Proc Symp on Refractory Alloy Technology for Space Nuclear Power Applications, CONF-8308130, Oak Ridge National Lab, 1984, pp. 86–97. [32] S.J. Zinkle, F.W. Wiffen, J.R. DiStefano, Historical Basis for Selection of Nb-1Zr Cladding for Space Reactor Applications, Oak Ridge National Lab, Oak Ridge, TN, 2004 September 30, 2004. Report No.: ORNL/LTR/NR/04-05. [33] T.E. McGreevy, D.T. Hoelzer, S.J. Zinkle, C.E. Duty, K.J. Leonard, F.W. Wiffen, Outstanding Issues Concerning the Mechanical Behavior of Nb-1%Zr Alloys, Oak Ridge National Lab, Oak Ridge, TN, 2004. September 30, 2004. Report No.: ORNL/LTR/NR-JIMO/04-02. [34] C.A. English, in: H. Stuart (Ed.), The physical, mechanical and irradiation behavior of niobium and niobium-base alloys, Proc Symp on Niobium, The Metallurgical Society of AIME, Warrendale, PA, 1984, pp. 239–324. [35] E.A. Gulbransen, K.F. Andrew, F.A. Brassard, Oxidation of molybdenum 550°C to 1700°C, J. Electrochem. Soc. 110 (9) (1963) 952–959. [36] B.V. Cockeram, R.W. Smith, T.S. Byun, L.L. Snead, The change in the hardness of LCAC, TZM, and ODS molybdenum in the post-irradiated and annealed conditions, J. Nucl. Mater. 393 (2009) 12–21. [37] A.J. Bryhan, Welding Research Council Bulletin 312, United Engineering Center, 1986. [38] M.K. Miller, A.J. Bryhan, Effect of Zr, B and C additions on the ductility of molybdenum, Mater. Sci. Eng. A 327 (1) (2002) 80–83. [39] M.K. Miller, E.A. Kenik, M.S. Mousa, K.F. Russell, A.J. Bryhan, Improvement in the ductility of molybdenum alloys due to grain boundary segregation, Scr. Mater. 46 (2002) 299–303. [40] N.N. Morgunova, L.N. Demina, N.I. Kazakova, Met. Sci. Heat Treat. 28 (12) (1986) 895–899. [41] Y.N. Gornostyrev, M.I. Katsnelson, G.V. Peschanskikh, S.V. Trefilov, Phys. Staus Solid. 164 (1991) 185. [42] L. Romaner, C. Ambrosch-Draxl, R. Pippan, Effect of rhenium on the dislocation core structure in tungsten, Phys. Rev. Lett. 104 (2010). 195503(1-4). [43] M. Garfinkle, Room-Temperature Tensile Behavior of 100 Oriented Tungsten Single Crystals with Rhenium in Dilute Solid Solution, 1966, (1966)NASA TN D-3190 1966:17. [44] A. Gilbert, M.J. Klein, J.W. Edington, Investigation of Mechanical Properties of Chromium, Chromium-Rhenium, and Derived Alloys, NASA CR 81225, (1966). [45] P.L. Raffo, Yielding and Fracture in Tungsten and Tungsten-Rhenium Alloys, J Less Comm. Metals 17 (1969) 133. [46] D.R. Trinkle, C. Woodward, The chemistry of deformation how solutes soften pure metals, Science 310 (2005) 1665. [47] N.I. Medvedeva, Y.N. Gornostyrev, A.J. Freeman, Electronic origin of solid solution softening in BCC molybdenum alloys, Phys. Rev. Lett. 94 (2005) 136402. [48] B.V. Cockeram, R.W. Smith, N. Hashimoto, L.L. Snead, The swelling, microstructure, and hardening of wrought LCAC, TZM, and ODS molybdenum following neutron irradiation, J. Nucl. Mater. 418 (1–3) (2011) 121–136. [49] V.V. Bukhanovskii, N.G. Kartyshov, E.G. Polishchuk, V.K. Karchenko, M.I. Chikunov, Effect of heat treatment on the mechanical properties and nature of failure for sheet molybdenum alloys, Stren. Mater. 382 (1987) 229–241. [50] Deleted in review. [51] B.V. Cockeram, E.K. Ohriner, T.S. Byun, M.K. Miller, L.L. Snead, Weldable ductile molybdenum alloy development, J. Nucl. Mater. 382 (2008) 229–241. [52] H.W. Yih, C.T. Wang, Tungsten, Sources, Metallurgy, Properties, and Applications, Plenum Press, New York, 1979. [53] R.M. German, E. Olevesky, Int. J. Refract. Met. Hard Mater. 23 (2005) 294–300. [54] H.M. Chung, B.A. Loomis, D.L. Smith, Creep properties of vanadium-base alloys, J. Nucl. Mater. 212–215 (1994) 772–777.

REFERENCES

637

[55] T. Schober, D.N. Braski, The microstructure of selected annealed vanadium-base alloys, Metall. Trans. A. 20 (1989) 1927–1932. [56] D.T. Hoelzer, A.F. Rowcliffe, L.T. Gibson, Processing and fabrication of experimental vanadium alloys for HFIR RB-17J experiment. Fusion Materials Semiannual Progress Report, DOE/ER-0313/34, Oak Ridge National Lab, 2003, pp. 22–26. [57] D.T. Hoelzer, A.F. Rowcliffe, Investigating solute interactions in V-4Cr-4Ti based on tensile deformation behavior of vanadium, J. Nucl. Mater. 307–311 (2002) 596. [58] A.F. Rowcliffe, D.T. Hoelzer, S.J. Zinkle, High Temperature Tensile Properties and Deformation Behavior of V-4Cr-4Ti. Fusion Materials Semiannual Progress Report DOE/ER-0313/26, p. 25. [59] A.F. Rowcliffe, S.J. Zinkle, D.T. Hoelzer, Effect of strain rate on the tensile properties of unirradiated and irradiated V-4Cr-4Ti, J. Nucl. Mater. 283–287 (2000) 508–512. [60] A.F. Rowcliffe, et al., Investigating solute interactions in V-4Cr-4Ti, J. Nucl. Mater. 508 (2000) 283–287. [61] P. Haasen, Physical Metallurgy, Cambridge Univ. Press, New York, 1978, pp. 26–27. [62] R.J. Kurtz, Biaxial thermal creep of V-4Cr-4Ti at 700°C and 800°C, J. Nucl. Mater. 283–287 (2000) 628. [63] K. Natesan, W.K. Soppett, A. Purohit, Uniaxial creep behavior of V-4Cr-4T- alloy, J. Nucl. Mater. 307–311 (2002) 585–590. [64] M. Li, T. Nagasaka, D.T. Hoelzer, M.L. Grossbeck, S.J. Zinkle, T. Muroga, et al., Biaxial thermal creep of two heats of V4Cr4Ti at 700 and 800°C in a liquid lithium environment, J. Nucl. Mater. 367–370 (2007) 788. [65] D.L. Smith, K. Natesan, Influence of nonmetallic impurity elements on the compatibility of liquid lithium with potential CTR containment materials, Nucl. Technol. 22 (1974) 392–404. [66] R.L. Stephenson, The creep-rupture properties of some refractory metal alloys I. Effect of heat treatment on the properties of the niobium-base alloy D-43, J Less Comm. Metals 15 (3) (1996) 395–402. [67] D.M. Farkas, A.M. Mukherjee, Creep behavior and microstructural correlation of a particle-strengthened Nb-1Zr-0.1C alloy, J Mat Res. 11 (9) (1996) 2198–2205. [68] K.J. Leonard, J.T. Busby, D.T. Hoelzer, S.J. Zinkle, Nb-base FS-85 alloy as a candidate structural materials for space reactor applications: effects of thermal aging, Met Mater Trans A 40A (2009) 838–855. [69] B.V. Cockeram, R.W. Smith, K.J. Leonard, T.S. Byun, L.L. Snead, Irradiation hardening in unalloyed and ODS molybdenum and ODS molybdenum during low dose neutron irradiation at 300°C and 600°C, J. Nucl. Mater. 382 (2008) 1–23. [70] B.V. Cockeram, R.W. Smith, L.L. Snead, The influence of fast neutron irradiation and irradiation temperature on the tensile properties of wrought LCAC and TZM molybdenum, J. Nucl. Mater. 346 (2005) 145–164. [71] H. Schultz, Defect-sensitive properties of tungsten and other BCC transition-metals, Z. Metallkd. 78 (1987) 469–475. [72] G. Bar, D. Gras, V. Hasse, System-Nr 54, Tungsten, Suppl In Gmelin Handbook of Inorganic Chemistry, Springer-Verlag, Heidelberg, 1989. [73] B.C. Allen, D.J. Maykuth, R.I. Jaffee, The recrystallization and Ductile-Brittle transition behaviour of tungsten-effect of impurities on polycrystals prepared from single crystals, J. Inst. Metals 90 (1961) 120–128. [74] D.B. Snow, Dopant observations in thin foils of annealed tungsten wire, Metall. Trans. A. 2 (1972) 2553. [75] B.A. Loomis, H.M. Chung, L.J. Nowicki, D.L. Smith, Effects of neutron irradiation and hydrogen on ductile-brittle transition temperatures of V-Cr-Ti alloys, J. Nucl. Mater. 212–215 (1994) 799–803. [76] B.A. Loomis, D.L. Smith, Vanadium alloys for structural applications in fusion systems: a review of vanadium alloy mechanical and physical properties, J. Nucl. Mater. 191–194 (1992) 84–91. [77] S.J. Zinkle, D.J. Alexander, J.P. Robertson, L.L. Snead, A.F. Rowcliffe, L.T. Gibson, et al., Effect of Fast Neutron Irradiation to 4 dpa at 400°C on the properties of V-(4–5)Cr-(4–5)-Ti Alloys. Department of Energy Fusion Reactor Semiannual Period Ending December 31, 1996 DOE/ER-0313/21, Oak Rige National Lab, 1996, pp. 73–78. [78] L.L. Snead, S.J. Zinkle, D.J. Alexander, A.F. Rowcliffe, J.P. Robertson, W.P. Eatherly, Summary of the investigation of low temperature, low dose radiation effects on the V-4Cr-4Ti alloy, in: Fusion Materials Semiannual Progress Report for Period Ending Dec 31, 1997, pp. 81–98 DOE/ER-0313/231997. [79] S.J. Zinkle, L.L. Snead, J.P. Robertson, A.F. Rowcliffe, Strain rate dependence of the tensile properties of V-(4–5%)Cr(4–5%)Ti irradiated in EBR-II and HFBR, in: Fusion Materials Semiann Prog Rep for period ending Dec 31 1997, DOE/ ER-0313/23, Oak Ridge National Lab, 1997, pp. 77–80. [80] S.J. Zinkle, Thermophysical and mechanical properties of V-(4–5%)Cr-(4–5%)Ti alloys, in: Fusion Materials Semiann Prog Report for period ending Dec 31 1997, DOE/ER-0313/23, Oak Ridge National Lab, 1997, pp. 99–108. [81] E. Donahue, G.R. Odette, G.E. Lucas, J.W. Sheckherd, A.F. Rowcliffe, Effect of irradiation on toughness-temperature curves in V-4Cr-4Ti, in: Fusion Materials Semiann Prog Report for period ending Dec 31 1998, DOE/ER-0313/25, Oak Ridge National Lab, 1998, pp. 32–41.

638

CHAPTER 13 REFRACTORY ALLOYS

[82] H.M. Chung, D.L. Smith, Tensile and impact properties of vanadium-base alloys irradiated at <430°C, J. Nucl. Mater. 258–263 (1998) 1442–1450. [83] T. Muroga, Vanadium alloys for fusion blanket applications, Mater. Trans. 45 (2005) 405. [84] J. Gazda, M. Meshii, H.M. Chung, Microstructure of V-4Cr-4Ti after low-temperature irradiation by ions and neutrons, J. Nucl. Mater. 258–263 (1998) 1437–1441. [85] P.M. Rice, S.J. Zinkle, Temperature dependence of radiation damage microstructure in V-4Cr-4Ti neutron irradiated to low dose, J. Nucl. Mater. 258–263 (2) (1998) 1414–1419. [86] D.S. Gelles, P.M. Rice, S.J. Zinkle, H.M. Chung, Microstructural examination of irradiated V-(4-5%)Cr-(4-5%)Ti, J. Nucl. Mater. 258–263 (1998) 1380–1385. [87] F.W. Wiffen, Creep and tensile properties of helium injected Nb-1%Zr, in: J.S. Watson, F.W. Wiffen (Eds.), Radiation Effects and Tritium Technology for Fusion Reactors, CONF-750989. II, USERDA, Gatlinburg, TN, 1976, pp. 344–361. [88] H. Kayano, S. Yajima, Mechanical properties of neutron-irradiated niobium, J. Nucl. Sci. Technol. 15 (2) (1978) 135–140. [89] D.T. Hoelzer, S.J. Zinkle, Structural analysis of plate-shaped precipitates in neutron irradiated V-4Cr-4Ti, in: Fusion Materials Semiannual Progress Report for Period ending December 31 2000, DOE/ER-0313/29, Oak Ridge National Lab, 2000, pp. 19–25. [90] V.M. Troyanov, M.G. Bulkanov, A.S. Kruglov, E.A. Krjuchkov, M.P. Nikulin, J.M. Pevchykh, et al., Irradiation creep of V-Ti-Cr alloy in BR-10 reactor core instrumented experiments, J. Nucl. Mater. 233–237 (1996) 381–384. [91] H. Tsai, I.C. Gomes, D.L. Smith, A.J. Palmer, F.W. Ingram, F.W. Wiffen, Status of the irradiation test vehicle for testing fusion materials in the advanced test reactor, in: Fusion Materials Semiann Prog Report for period ending June 30, 1998, DOE/ER-0313/24, Oak Ridge National Lab, 1998, pp. 278–282. [92] H. Tsai, T.S. Bray, H. Matsui, M.L. Grossbeck, K. Fukumoto, J. Gazda, et al., Effects of low-temperature neutron irradiation on mechanical properties of vanadium-base alloys, J. Nucl. Mater. 283–287 (2000) 362. [93] D.S. Gelles, Microstructural Examination of Irradiated and Unirradiated V-4Cr-4Ti Pressurized Creep Tubes, J. Nucl. Mater. 386–388 (2002) 393. [94] M. Li, D.T. Hoelzer, M.L. Grossbeck, A.F. Rowcliffe, S.J. Zinkle, R.J. Kurtz, Irradiation creep of the US Heat 832665 of V-4Cr-4Ti, J. Nucl. Mater. 386–388 (2009) 618. [95] F.W. Wiffen, in: R.H. Cooper Jr., E.E. Hoffman (Eds.), Effects of irradiation on properties of refractory alloys with emphasis on space power reactor applications, Proc Symp on Refractory Alloy Technology for Space Nuclear Power Applications, CONF-8308130, Oak Ridge National Lab, 1984, pp. 252–277. [96] J.H. DeVan, J.R. DiStefano, E.E. Hoffman, in: R.H. Cooper Jr., E.E. Hoffman (Eds.), Compatibility of refractory alloys with space reactor system coolants and working fluids, Proc Symp on Refractory Alloy Technology for Space Nuclear Power Applications, CONF-8308130, Oak Ridge National Lab, 1984, pp. 34–85. [97] M. Kangilaski, Comparison of niobium (columbium) alloys for SP-100 space reactor applications (unpublished), General Electric Company, Advanced Nuclear Technology Operations, Sunnyvale, CA, 1984. November 1984. (Report No). [98] S.J. Zinkle, F.W. Wiffen, in: M.S. El-Genk (Ed.), Radiation Effects in Refractory Alloys, Space Technology and Applications International Forum-STAIF 2004 (AIP Conf Proc, vol. 699), American Institute of Physics, Melville, NY, 2004, pp. 733–740. [99] F.W. Wiffen, Plastic instability in neutron-irradiated niobium alloys, in: Alloy Development for Irradiation Performance Quarterly Progress Report, January–March, 1978, DOE/ET-0058/1, Oak Ridge National Lab, 1978, pp. 142–152. [100] T.T. Claudson, H.J. Pessl, Irradiation effects on high-temperature reactor structural metals, in: C.E. Weber (Ed.), Flow and Fracture of Metals and Alloys in Nuclear Environments, American Society for Testing and Materials, Philadelphia, 1965, pp. 156–170. ASTM STP 380. [101] R.P. Tucker, S.M. Ohr, M.S. Wechsler, Radiation hardening and transmission electron microscopy in niobium, in: Radiation Damage in Reactor Materials. I, IAEA, Vienna, 1969, pp. 215–233. [102] F.W. Wiffen, The tensile properties of fast reactor neutron irradiated BCC metals and alloys, in: R.J. Arsenault (Ed.), Proc Int Conf on Defects and Defect Clusters in BCC Metals and their Alloys, Nuclear Metallurgy, vol. 18, National Bureau of Standards, Gaithersburg, MD, 1973, pp. 176–196. [103] V. Kazakov, A.S. Pokrovsky, R. Melder, in: Mechanical properties of niobium alloys after neutron irradiation at 350–1300°C, Presented at 6th International Conference on Fusion Reactor Materials, Stresa, Italy, September 1993 (Unpublished Preprint), 1993. [104] J.A. Horak, M.L. Grossbeck, M.M. Paxton, Effects of irradiation on the tensile properties of Nb-1Zr, in: M.S. El-Genk, M. D. Hoover (Eds.), 11th Symp on Space Nuclear Power and Propulsion, AIP Conf Proc 301, American Institute of Physics, Woodbury, NY, 1994, pp. 413–420.

REFERENCES

639

[105] K.J. Leonard, J.T. Busby, S.J. Zinkle, Influence of thermal and radiation effects on microstructural and mechanical properties of Nb-1Zr, J. Nucl. Mater. 411 (2011) 286–302. [106] D.G. Atteridge, L.A. Charlot, J.A.B. Johnson, J.F. Remark, Effects of helium implanted by tritium decay on the high temperature mechanical properties of niobium, in: F.W. Wiffen, J.S. Watson (Eds.), Radiation Effects and Tritium Technology for Fusion Reactors, 1976 (Gatlinburg, TN). [107] A.A. Sagues, J. Auer, Mechanical behavior of Nb-1%Zr implanted with He at varioius temperatures, in: F.W. Wiffen, J. S. Watson (Eds.), IInd Radiation Effects and Tritium Technology for Fusion Reactors, 1976, pp. 331–343 (Gatlingburg, TN). [108] M.M. Paxton, C.R. Eiholzer, Biaxial creep of rhenium and Nb-1Zr Re-bonded pressurized type specimens at SP-100 service conditions. Westinghouse Hanford Co., 1994 September, 1994, (1994) Report No.: WHC-SP-1078. [109] M. Li, T.S. Byun, L.L. Snead, S.J. Zinkle, Defect cluster formation and radiation hardening in molybdenum neutronirradiated at 80°C, J. Nucl. Mater. 367–370 (2007) 817–822. [110] L.R. Greenwood, F.A. Garner, Transmutation of Mo, Re, W, Hf and V in various irradiation test facilities and STARFIRE, J. Nucl. Mater. 212–215 (1994) 635–639. [111] M. Li, T.S. Byun, L.L. Snead, S.J. Zinkle, Low-temperature thermally-activated deformation and irradiation softening in neutron-irradiated molybdenum, J. Nucl. Mater. 377 (2008) 409–414. [112] F.W. Wiffen, The effect of alloying and purity on the formation and ordering of voids in bcc metals, in: J.W. Corbett, L. C. Iannello (Eds.), Radiation Induced Voids in Metals, CONF-710601, National Technical Information Service, Springfield, VA, 1972, pp. 386–396. [113] J.H. Evans, Observations of a regular void array in high purity molybdenum and TZM irradiated at high temperatures with 2 MeV nitrogen ions, Radiat. Eff. 10 (1971) 55–60. [114] R.C. Rau, J. Moteff, R.L. Ladd, Comparison of radiation damage in molybdenum and some molybdenum-base alloys, J. Nucl. Mater. 40 (1971) 233–235. [115] V.K. Sikka, J. Moteff, Damage in neutron-irradiated molybdenum, J. Nucl. Mater. 54 (1974) 325–345. [116] F.A. Garner, J.F. Stubbins, Saturation of swelling in neutron-irradiated molybdenum and its dependence on irradiation temperature and starting microstructural state, J. Nucl. Mater. 212–215 (1994) 1298–1302. [117] K.J. Leonard, Radiation effects in Refractory Metals and Alloys, in: R.J.M. Konings (Ed.), Comprehensive Nuclear Materials, Elsevier Publishing, Amsterdam, 2012, pp. 181–213. [118] Y. Nemoto, A. Hasegawa, M. Satou, K. Abe, Y. Hiraoka, Microstructural development and radiation hardening of neutron irradiated Mo–Re alloys, J. Nucl. Mater. 324 (1) (2004) 62–70. [119] J.T. Busby, K.J. Leonard, S.J. Zinkle, Radiation-damage in molybdenum-rhenium alloys for space reactor applications, J. Nucl. Mater. 366 (2007) 388–406. [120] S.A. Fabritsiev, A. Pokrovsky, The effect of rhenium on the radiation damage resistivity of Mo-Re alloys, J. Nucl. Mater. 252 (1998) 216–227. [121] B.V. Cockeram, R.W. Smith, L.L. Snead, The influence of fast neutron irradiation and irradiation temperature on the tensile properties of wrought LCAC and TZM molybdenum, J. Nucl. Mater. 346 (2005) 145–164. [121a] V. Chakin, V. Kazakov, J. Nucl. Mater. 233–237 (1996) 570–572. [121b] A. Hasegawa, K. Abe, M. Satou, C. Ueda, C. Namba, J. Nucl. Mater. 233–237 (1996) 565–569. [121c] K. Abe, T. Takeuchi, M. Kikuchi, S. Morosumi, J. Nucl. Mater. 99 (1981) 23–37. [121d] K. Abe, M. Kikuchi, K. Tate, S. Morozumi, J. Nucl. Mater. 122–123 (1984) 671–675. [121e] T.H. Webster, B.L. Eyre, E.A. Terry, in: Proceedings of BNES conference on irradiation embrittlement and creep in fuel cladding and core components, British Nuclear Energy Society, London, 1972, pp. 61–80. [121f] M. Schibetta, R. Chaouadi, J.L. Puzzolante, J. Nucl. Mater. 283–287 (2000) 455–460. [122] B.V. Cockeram, T.S. Byun, K.J. Leonard, J.L. Hollenbeck, L.L. Snead, Post-Irrdiation fracture toughness of unalloyed molybdenum, ODS molybdenum, and TZM molybdenum following irradiation at 244°C to 507°C, J. Nucl. Mater. 440 (2013) 382–413. [123] K. Babinsky, S. Primig, W. Knabi, A. Lorich, R. Stickler, H. Clemens, Fracture behavior and delamination toughening of molybdenum in Charpy impact tests, J. Mater. 68 (11) (2016) 2854–2863. [124] D.R. Mason, X. Yi, M.A. Kirk, S.L. Dudarev, Elastic trapping of dislocation loops in cascades in ion-irradiated tungsten foils, J. Phys. Condens. Matter 26 (2014) 375701. [125] L.K. Keys, J. Moteff, J. Nucl. Mater. 34 (1970) 260. [126] R.K. Williams, F.W. Wiffen, J. Bentley, J.O. Stiegler, Irradiation induced precipitation in tungsten based, W-Re alloys, Mettal. Trans. A 14A (1983) 655–666.

640

CHAPTER 13 REFRACTORY ALLOYS

[127] M. Fukuda, K. Yabuuchi, S. Nogami, A. Hasegawa, T. Tanaka, Microstructural development of tungsten and tungstenrhenium alloys due to neutron irradiation in HFIR, J. Nucl. Mater. 455 (2014) 450–463. [128] P.D. Edmondson, A. London, A. Xu, D.E.J. Armstrong, S.G. Roberts, Small-scale characterisation of irradiated nuclear materials: part 1-microstructure, J. Nucl. Mater. 462 (2015) 369–373. [129] T. Koyanagi, N. Kumar, T. Hwang, L.M. Garrison, X. Hu, L.L. Snead, Y. Katoh, Microstructural evolution of pure tungsten neutron irradiated with a mixed energy spectrum, J. Nucl. Mater. 490 (2017) 66–74. [130] A. Hasegawa, T. Tanno, S. Nogami, T. Tanaka, Property change mechanism in tungsten under neutron irradiation in various reactors, J. Nucl. Mater. 417 (2011) 491–494. [131] Y. Nemoto, A. Hasegawa, M. Satou, K. Abe, Microstructural development of neutron irradiated W-Re alloys, J. Nucl. Mater. 283-287 (2000) 1144–1147. [132] J. Matolich, H. Nahm, J. Moteff, Swelling in neutron irradiated tungsten and tungsten-25 percent rhenium, Scr. Metall. 8 (7) (1974) 837–842. [133] M. Fujisuka, B. Tsuchiya, I. Mutoh, T. Tanabe, T. Shikama, Effect of neutron irradiation on thermal diffusivity of tungsten-rhenium alloys, J. Nucl. Mater. 283–287 (2000) 1148–1151. [134] T. Tanno, A. Hasegawa, J.-C. He, M. Fujiwara, S. Nogami, M. Satou, et al., Effects of transmutation elements on neutron irradiation hardening of tungsten, Mater. Trans. 48 (9) (2007) 2399–2402. [135] K. Krishan, Ordering of voids and gas bubbles in radiation environments, Radiat. Eff. 66 (1982) 121–155. [136] R. Herschitz, D.N. Seidman, Radiation-induced precipitation in fast neutron irradiated W-Re alloys: an atom probe field microscopy study, Nucl Instrum Meth Phys Res B 7/8 (1985) 137. [137] X. Hu, T. Koyanagi, M. Fukluda, N.A.P. Kiran Kumar, L.L. Snead, B.D. Wirth, Y. Katoh, Irradiation hardening of pure tungsten exposed to neutron irradiation, J. Nucl. Mater. 480 (2016) 235–243. [138] R.C. Rau, J. Moteff, R.L. Ladd, Comparison of microstructure with mechanical properties of irradiated tungsten, J. Nucl. Mater. 24 (1967) 164–173. [139] J.M. Steichen, Tensile properties of neutron irradiated TZM and tungsten, J. Nucl. Mater. 60 (1976) 13–19. [140] I.V. Gorynin, A.A. Ignatov, V.V. Rybin, S.A. Fabritsiev, V.A. Kazakov, V.P. Chakin, V.A. Tsykanov, V.R. Barabash, Y. G. Prokofyev, Effects of neutron irradiation on properties of refractory metals, J. Nucl. Mater. 191–194 (1992) 421–425. [141] P. Krautwasser, H. Derz, E. Kny, Influence of fast neutron fluence on the DBTT of tungsten, W-10Re and W-3.4Ni-1.6Fe, High Temp. High Pressures 22 (1) (1990) 25–32. [142] H.H. Smith, D.J. Michel, Journal of Nuclear Materials, 66 (1) (1977) 125–142.