Pergamon Printed m Great Britam 0079-6425/97 $32.00
1.
2.
3.
4.
5.
6.
7.
8.
9.
10.
PII: SOO79-6425(97)00014-5
PROCESSING, PROPERTIES, AND APPLICATIONS OF NICKEL AND IRON
ALUMINIDES
S. C. Deevi*, V. K. Sikkat and C. T. Liul_
*Research Center, Philip Morris U.S.A., 4201 Commerce Road, Richmond, VA 23234, U.S.A. and TMetals and Ceramics Division, Oak Ridge National Laboratory, P.O. Box 2008, Oak Ridge, TN 37831, U.S.A.
CONTENTS
INTRODUCTION N&AI-BASED ALLOYS NiAI-BASED ALLOYS Fe,AI-BASED ALLOYS FeAI-BASED ALLOYS
PROCESSING OF INTERMETALLICS Exo-MeItTM PROCESS
CASTING OF INTERMETALLICS WELDING OF Ni,AI-BASED ALLOYS
CONCLUSIONS _
ACKNOWLEDGEMENTS REFERENCES
177 179 181 185 186 188 188 190 190 191 191 192
1. INTRODUCTION
Intermetallics have drawn enormous attention owing to their ability to provide significant advantages in manufacturing processes, technologies, and as commercial products. The ordered nature of intermetallic compounds exhibits attractive high-temperature properties due to the presence of long-range-ordered superlattices, which reduce dislocation mobility and diffusion processes at elevated temperatures. (‘. 2, Many intermetallic compounds exhibit brittle fracture through transgranular cleavage and intergranular separation, and the brittle fracture or low crack tolerance has been the primary barrier to the use of intermetallic compounds for load-bearing applications. (3) Both intrinsic and extrinsic factors contribute to the brittle behavior, and the intrinsic factors include: complex crystal structure, insufficient deformation modes, high work-hardening rate, constitutional defects, and planar slip. The extrinsic factors are segregants, interstitials, moisture in the environment, notches, poor surface finish, and hydrogen. In this paper, we focus our attention on polycrystalline materials even though significant progress has been made on single-crystal intermetallics.
Aluminide intermetallics are distinctly different from those of conventional (disordered) materials based on solid-solution alloys. For example, Ni,Al exhibits an increase in yield strength with an increase of temperature, whereas conventional alloys exhibit a general decrease in strength with temperature. c41 5, Also, nickel and iron aluminides possess sufficiently high concentrations of aluminum (i.e. in the range of 8 to 32 wt%) to form
177
Table 1. Attributes and upper-use temperature limit for selected intermetallic compounds Maximum use temperature
(“Cl
Intermetallic Property
Strength Corrosion
limit limit
N&AI NiAl FelAl TiAl T&AI Ni,Si
FeAl MoS&
Oxidation, carburization, and nitridation resistance; high- temperature strength
High melting point; high thermal conductivity; oxidation, carburization, and nitridation resistance
Oxidation and sulfidation resistance
Low density is the real advantage; good specific-strength properties and wear resistance
Low density; good specific strength
Oxidation and reducing environment resistance; good resist- ante in sea water and sulfuric acid; good resistance in ammonia reactor up to 900°C
Oxidation, sulfidation, molten salt, and carburization resist- ante
High melting point; excellent resistance to oxidation; exhibits metallic conductivity
1100 1150
1200 1400
700 1200
1000 800
760 650
800 1000
800 1200
1300 1600
a continuous and fully adherent alumina layer. In contrast, almost 98% of alloys and superalloys capable of operating above 700°C in oxygen environments contain less than 2 wt% Al, and invariably contain chromium to as high a concentration as 18 wt% for oxidation resistance.@) Alloys containing chromium form Cr,O, on exposure to air or oxygen, and the dissociation of Cr,O, to Cr03 limits their oxidation resistance to 950°C. On the other hand, nickel and iron aluminides provide excellent oxidation resistance in the range 1100 to 1400°C owing to their high aluminum contents and high melting points. Table 1 provides attributes of some selected intermetallic compounds and their maximum use temperatures. c7) Along with nickel and iron aluminides, also included are titanium aluminides, Ni,Si, and MoSi,. At the present time, only N&Al, NiAl, Fe,Al, and FeAl are being developed as high-temperature structural materials for a broad range of applications.
The major advantages that can be derived from the use of nickel aluminides (N&Al) include:
1.
2.
3.
4.
5.
6.
resistance to oxidation and carburization in both oxidizing and reducing carburizing atmospheres up to 1100°C;
good tensile and compressive yield strengths for temperatures up to 1100°C;
fatigue resistance superior to that of nickel-based superalloys, resulting from the elimination of second-phase particles such as carbides;
superior creep strength;
excellent wear resistance at high temperatures (2 600°C). In fact, wear resistance increases with temperature by a factor of 1000; and
surface pre-oxidation that provides good chemical compatibility for many environ- ments owing to the formation of an alumina layer.
The major advantages that can be derived from the use of iron aluminides include:
1. their density is lower than that of many stainless steels, and, therefore, they offer better strength-to-weight ratios;
2. their resistance to sulfidation in H,S and SO, gases is much better than that of any other iron- or nickel-based alloys;
3. they have excellent oxidation resistance at temperatures up to 1200°C;
4. they have a high electrical resistivity that increases with temperature; and 5. they have good corrosion resistance in many aqueous environments.
Although the research activity in intermetallics accelerated during the early 1980s W” commercial utilization of nickel and iron aluminides was hampered because of the lack of processing experience by industrial manufacturers and the low ductility of NiAl- and FeAl-based alloys. This paper provides recent advances in the alloy design, processing, properties, and applications of intermetallics. This discussion will be followed by critical areas of research for extensive industrial utilization of intermetallics.
2. NI,AL-BASED ALLOYS
The Ni,Al-based alloys exhibit { 11 l}(llO) slip and have sufficient slip systems for extensive plastic deformation. (I” Single crystals of N&Al are highly ductile, whereas polycrystals are brittle at ambient temperature. An extrinsic factor (i.e.
environmental embrittlement) has recently been shown ‘13’ to be the major cause of low ductility and brittle intergranular fracture in binary Ni,Al. Boron has been found to be the most effective addition in improving the tensile ductility of Ni,Al ( < 25 at.% Al) when tested in air at room temperature. (‘M’ Zirconium additions have also been used to ductilize Ni,Al.
Environment influences the ductility of polycrystalline Ni,Al alloys both at room and elevated temperatures. (I23 “’ The room-temperature embrittlement is associated with hydrogen released from the reaction of moisture in air with aluminum atoms in Ni,Al by eq. (1):
2Al+ 3H20 -+ Al,O, + 6H (1)
The atomic hydrogen penetrates along the grain boundaries at the crack tips, resulting in hydrogen-induced embrittlement. Embrittlement at elevated temperatures causes brittle intergranular fracture owing to the penetration of oxygen along the grain boundaries.““’ The oxygen-related environmental embrittlement in Ni,Al alloy (Ni-21.5 Al-O.5 Hf-O.01 B) is demonstrated in Ref. 19. Oxygen-induced embrittlement in Ni,Al can be reduced at elevated temperatures by alloy modification and control of surface condition and grain shape. The addition of 8 at.% Cr to Ni,Al has been found to be the most effective method(12’ of minimizing elevated-temperature embrittlement in oxidizing environments. The penetration of oxygen into grain boundaries is reduced because of a rapid formation of self-healing, protective chromium oxide films on the surface. The high ductility of chromium-containing alloys in vacuum as opposed to air suggests that environmental embrittlement cannot be completely eliminated even with the addition of chromium.
The Ni,Al alloy can dissolve a substantial amount of alloying elements, and its mechanical and metallurgical properties can be improved by controlling the solute
concentration and second-phase formation. The specific contribution of beneficial alloying elements in N&AI alloys is described below:
Cr:
Zr:
Reduces oxygen embrittlement at elevated temperatures.
Provides high-temperature strength through solid solution, reduces solidification shrinkage and macroporosity through the formation of low-melting-point (1172°C) eutectic, and improves oxide spallation resistance during thermal cycling.
Hf: Provides high-temperature strength through solid solution and prevents surface reaction of zirconium with the ceramic shell material during investment casting by forming a protective oxide film.
MO:
B:
Improves strength at low and high temperatures.
Reduces moisture-induced hydrogen embrittlement and enhances the grain boundary cohesive strength. The boron is most effective when N&Al is nickel-rich, with Al _< 24 at.%.
Based on an understanding of the room- and high-temperature environmental embrittlement mechanisms and influence of alloying elements on the strengths, several N&Al-based alloys have been developed for commercial applications for cast and wrought applications (see Table 2).
The mechanical properties of castable IC-221M and IC-396 N&Al-based alloys provide significant strength advantages”) over a commercially available FeNiCr alloy (designated as HU) that is used widely for high-temperature furnace applications.‘*‘) The yield strengths of IC-221M and IC-396 alloys are at least twice that of HU at room temperature and increase with rising temperature. On the other hand, the strength of HU decreases with rising temperature while the strength of cast IC-221M and IC-396 N&Al-based alloys increases with temperature and is over five times higher than that of HU. Ultimate tensile strengths of N&Al-based IC-221M and IC-396 alloys are also higher than the ultimate tensile strengths exhibited by the HU alloy. Tensile elongations of
Table 2. Compositions of cast N&Al-based alloys selected for commercial applications and compositions of some cast and wrought commercial alloys
Alloy (wt%)
Element K-221 M” IC-39@ HU IC-50d IC-218LZr’ IC-221W Haynes 214’ Alloy 800’
Al Cr MO
Zr : Fe Ti Ni Si Y
8.0 7.1 1.43
&8
- - 81.1
- -
1.98 1.72 3.02 0.85 0.005 -
- 80.42
- -
- 11.3
18.0 -
- -
- 0.6
- 0.02
0.55 - 42.45 -
- -
39.0 88.08
8.7 8.00
8.1 1.10
- 1.50
0.2 3.00
0.02 0.003
- -
- -
83.1 19.80
- - - -
4.5 16.0
- 76.35
0.1 0.02
0.4 21.0 - - 0.05 45.5
0.4 32.5 -
“Castable alloy for dynamic applications (minimum microporosity).
“Castable alloy for static applications (some microporosity).
‘Cast alloy.
dCold workable.
‘Hot and cold workable.
Wrought alloy.
N&Al-based IC-221M and IC-396 alloys indicate excellent room-temperature tensile ductility exceeding that of HU.
In wrought alloys, a commercially available nickel-based superalloy sold as Haynes 214(*‘) by Haynes International, Inc. (Kokomo, IN, U.S.A.) competes with the wrought N&Al-based alloys up to 1100°C (see Fig. 1). It is important to note that the yield strengths are similar even though the composition of IC-50 is much simpler than that of IC-218LZr and IC-221W alloys and is close to the composition of N&Al intermetallic.
On the other hand, powder-metallurgically processed IC-221 W exhibits a significantly higher strength than that of wrought Ni,Al-based alloys, even at 700°C [see Fig. l(a)].
Ultimate tensile strengths of N&Al-based alloys are significantly higher at room temperature, and the strength advantage of N&Al-based alloys over Haynes 214 is lost at or above 850°C [see Fig. l(b)]. High tensile elongations of wrought N&Al-based alloys suggest that room-temperature processing can be carried out similar to Haynes 214 [see Fig. l(c)].
High-cycle fatigue properties of IC-221M are also superior by an order of magnitude compared with those of commercially available IN-713C when tested in air at 650°C (see Fig. 2). Creep resistance of cast and wrought alloys is evaluated since creep is the primary deformation mechanism at high temperatures. Creep data, plotted as Larsen-Miller plots shown in Fig. 3, indicate the superior creep strength of IC-221M over HU by a factor of two to four even at high temperatures. A similar trend is observed when the creep strengths of wrought alloys are compared with the creep strength of Haynes 214.“)
The superior oxidation resistance of nickel aluminides was confirmed by oxidizing in air with 5% water vapor at 1100°C. No weight change was observed in the case of N&Al intermetallics even when oxidized with a 250 day exposure, whereas Alloy 800 lost over 150 mg cm-’ over a 40 day exposure. (‘) N&Al intermetallics offer excellent carburization resistance compared with many of the furnace fixture materials such as HU, HK, and Alloy 800 steels. The excellent oxidation and carburization resistances allow the use of Ni,Al-based alloys at high temperatures for furnace fixtures such as heat-treating trays, posts, and guide rolls.“)
3. NiAl-BASED ALLOYS
Nickel aluminide containing more than approximately 41 at.% Al starts to form a single-phase ordered B2 structure based on the body-centered cubic @cc) lattice. In terms of thermophysical properties, B2 NiAl offers more potential for high-temperature applications than Ll, N&Al. Even though NiAl has excellent oxidation resistance, higher thermal conductivity, higher melting temperature, and lower density, as compared with N&Al-based alloys, inadequate room-temperature ductility has been a key limiting factor for structural applications. Brittleness can be attributed to the fact that only three slip systems are independently operative in NiAl with a strongly ordered B2. Even when crystal structure favors five independent slip systems as required by the Von Mises criterion, the brittle mode may persist. The insufficient deformation modes, poor cleavage resistance, and brittle grain-boundary fracture(22~23) are all considered to be major causes of low tensile ductility in NiAl. Also, the structural use of NiAl suffers from: (1) poor fracture resistance at ambient temperature, and (2) low strength and creep resistance at elevated temperatures.
Single crystals of NiAl are quite ductile in compression, but both single-crystal and polycrystalline NiAl are brittle in tension at ambient temperature. The NiAl exhibits (100)
1500 3 5 1250 g1oaO E s 750
z 500
* t! 250
d 0
0 200 400 600 800 1000 1200 Temperature (“C)
lb) 3 fg15of) 5 1250
z --cIC218LZr
pOO v)
. 750 f
f 500 +
. 250 ti E
4 0
3 0 200 400 600 500 1000 1200 Temporrture (“C)
(Cl
0 200 400 600 a00 1000 1200 temperature (“C)
Fig. 1. A comparison of (a) yield strengths, (b) ultimate tensile strengths and (c) tensile elongations of wrought N&AI-based K-50, IC-218LZr, and IC-221W (PM) alloys with Haynes 214.
A 0
IN-71 3C c-221 M
650% AIR. R = 0.05
Fig. 2. Cyclic fatigue life of N&Al-based IC-221M alloy and a commercially available IN-713C superalloy.
slip rather than (111) slip as commonly observed for bee materials. NiAl shows a sharp increase in ductility above 400°C and becomes very ductile above 500°C at conventional strain rates.
There are at least three approaches possible for improving the ductility of NiAl, which include microalloying, macroalloying, and microstructural control. The influence of microalloying element on the tensile properties of NiAl’24’ indicated that microalloying elements seem to reduce the already limited tensile ductility of polycrystalline NiAl. A macroalloying approach was considered to modify the slip process. The general approach is based on the identification of ternary alloying elements for addition that should lower the ordering energy of NiAl, thereby making (111) slip easier. Chromium, manganese, and vanadium are reasonable choices for promoting (111) slip in NiAl, on the basis of calculations resulting from interatomic potential models. (*‘) Experimental studies do not confirm these calculations. Instead of attempting to alter the inherent properties, an alternative approach has been to improve the ductility and toughness of NiAl by
1 000
100
- . .- i___ ,_ .‘ .;. :
10 ..~.‘,.,,‘,,,,
;,
,h:, .,;
30 35 40 45 50 55 60
P=(f+460)*(20+logt,)*l O-3
Fig. 3. A comparison of creep resistance of cast N&Al-based IC-221M and IC-396 alloys with commercial HU alloy.
Table 3. Chemical composition of NiAl and NAL-109 alloy Alloy (%)
Element
NiAl NAL-109
Weight Atomic Weight Atomic
Ni 68.5 50.0 68.0 49.8
Al 31.5 50.0 31.2 49.1
MO - 0.45 0.20
Fe - - 0.26 0.20
Zr - 0.17 0.08
modification of the microstructure. The approach consists of microstructural variation from fine-grained materials to multiphase microstructure.
Recently, we alloyed NiAl with molybdenum, iron, and zirconium, as shown in Table 3, to drop the brittle-to-ductile transition temperature (BDTT) and are cornpositing with N&Al-based alloy to impart some ductility and improve the mechanical properties. Figure 4 shows the tensile elongations of binary NiAl, and an alloy of NiAl, designated as NAL-109, obtained by extrusion of the combustion- synthesized (otherwise known as reaction synthesis) powders at 1100°C. As can be noted from Fig. 4, the BDTT dropped from 525°C in binary NiAl to 350°C when alloyed with molybdenum, iron, and zirconium. A lower BDTT temperature can be obtained by extruding the powders at 930°C. The differences in the BDTT values due to extrusion temperatures can arise from two different factors: (1) a coarser grain size is obtained at higher extrusion temperature, and coarser grain size produces a material with poor ductility; and (2) material extruded at lower temperatures retains more mobile dislocations, and higher dislocation density has been known to increase the dislocation density in NiAl through dislocation multiplication. Alloying elements could slow down the grain growth and dislocation recovery process, and, therefore, for the same extrusion temperature of llOO”C, NiAl alloy possesses lower grain size and higher retained dislocation density than the binary NiAl. Grain-size measurements confirmed that the grain size of NiAl alloy is only about 11.7 pm while the grain size of
100 . . . . ..I... . . . ..I..
--a-- Cornbust ion- Synthesized NiAl
7c; -O- NiAI-109 Alloy
0:
0 200 400 600 600 1000 1200
Temperature (“C)
Fig. 4. A comparison of tensile elongation of binary NiAI and an alloy of NiAl (designated as
NAL-109) produced by hot extrusion of combustion-synthesized powders in a mild steel can at 1100°C and at a strain rate of 3.3 x lo-’ s-‘.
Temperature (“C)
Fig. 5. A comparison of ultimate tensile strengths of Ni,Al-based K-50 and NAL-109 alloys and a 50:50 mixture of IC-50 with NAL-109 produced by hot extrusion of powders in a mild steel can.
Water-atomized IC-50 was used for the experiments.
binary NiAl is 17.8 pm. Additional work needs to be carried out to determine the dislocation density.
Our alloying approaches showed that the strength of NiAl can be significantly improved, and the strengths are intermediate to those of N&Al-based IC-50 and NAL-109 alloys. This is shown in Fig. 5 with the ultimate tensile strengths of water-atomized N&Al-based IC-50 alloy, combustion-synthesized NiAl alloy NAL-109, and a 50:50 mixture by weight of N&Al-based K-50 alloy with NAL-109. While a composite of N&Al-based IC-50 alloy with NiAl has shown strength improvement, there has been no ductility improvement at all. Indeed, the room-temperature ductility of N&Al-based IC-50 alloy is lost when mixed with an alloy of NAL-109.
4. Fe,AL-BASED ALLOYS
The iron aluminides have been of interest since the 193Os, when their excellent oxidation resistance was first noted. Iron aluminide of a composition corresponding to Fe,Al exists in a DO3 structure, and DO, is stable in the 23 to 36 at.% Al range and from room temperature to 550°C. Above 55O”C, an ordered Fe,Al with DO, structure transforms to an imperfectly ordered B2 structure, which ultimately changes to a disordered solid solution. Iron aluminides offer relatively low material cost, conservation of strategic elements, and a lower density than stainless steels. These advantages, along with their excellent oxidation and sulfidation resistances, have led to the consideration of iron aluminides for many applications. However, limited ductility at ambient temperatures and a sharp drop in strength above 600°C have been major deterrents to their acceptance for structural applications.
Several methods have been adopted to improve the room-temperature ductility of Fe,Al-based alloys. For example, highly elongated microstructure with minimum transverse grain boundaries has been most resistant to hydrogen diffusion and resulted in highest ductility. On the other hand, increased recrystallization due to higher-temperature anneals increased the number of transverse grain boundaries, which in turn enhanced the
Table 4. Compositions of Fe,Al-based alloys developed for commercial applications and compositions of some commercial allovs
Allov (wt%)
Element FAS” FALb FA-129’ FAH Fecralloy MA-956 Type 310
Al Cr B Zr Nb C MO
Si Y Y*O9
Ti Ni Fe
15.9 15.9 15.9
2.20 5.5 5.5
0.01 0.01 -
- 0.15
- - 1.0
- - 0.05
- - -
- - -
- -
d d d
15.9 5.5 0.04 0.15 1.04 - 1 .oo - - - - d
4.5 - -
16.0 20.0 25.0
- 0.02 - 0.3 - - - - d
- - - - -
0.5 0.5 -ii
0.15 - 0.50
- -
20.0 d
‘Sulfidation-resistant alloy.
‘High room-temperature (RT) tensile ductility.
‘High-temperature strength with good RT ductility.
“Balance.
inward hydrogen diffusion. This resulted in lower ductility values. Only alloy changes resulted in more stable and reproducible improvement in ductility. It is important to note that even single-crystal FeAl also exhibited environmental embrittlement. The effect of precipitate-forming and solid-solution-alloying elements on the tensile properties of Fe,Al can be found in Ref. 26. Elements such as niobium, copper, tantalum, zirconium, boron, and carbon were considered for precipitation strengthening; chromium, titanium, manganese, silicon, molybdenum, vanadium, and nickel were added for solid-solution strengthening. In general, the addition of elements either for precipitation strengthening or solid-solution strengthening to improve high-temperature tensile strength and creep resistance resulted in low room-temperature tensile elongations. Chromium was found to be an effective addition to enhance the ductility of iron aluminides at higher aluminum contents, and a combination of alloying elements led to the optimization of properties. At least five alloy compositions have been identified for commercialization (see Table 4).
5. FeAl-BASED ALLOYS
The FeAl alloy is a B2 compound and exists over a wide range of compositions from 36 to approximately 50 at.% Al at room temperature. Aluminides based on FeAl exhibit better oxidation and corrosion resistances than FejAl alloys and are also lower in density by as much as 30 to 40%, compared with steels and other commercial iron-based alloys.
They also suffer from low room-temperature ductility. The major cause of poor ductility of FeAl-based alloys is their sensitivity to environmental embrittlement in the presence of water vapor.
The addition of boron has been tried to improve the ductility of FeAl-based alloys.‘*‘) The 36.5 at.% Al alloy showed a large increase in tensile ductility (2.2 to 17.6%) when the test environment was changed from air to dry oxygen. A similar result was not observed for the 40 at.% Al alloy. The 36.5 at.% Al alloy failed by transgranular cleavage in air, while the 40 at.% Al alloy showed brittle grain-boundary fracture. When tested in oxygen,
K”‘.. I’
- FA-365- FA-385 w/25 ppm Boron
-I?-- SA 316 Steel
- Fe-240 AIM
-Fe-240 VIM
- Fe-240 Hot Pressed
0 200 400 600 800 1000
Temperature (“C)
Fig. 6. (a) A comparison of yield strengths of binary FeAl with compositlon corresponding to Fe-24 wt % Al (Fe-240) obtained by air-induction melting, vacuum-induction melting and reaction svnthesis with an FeAl allov of Fe-21.1 Ala.42 Mo-0.1 Zr-O.03 C (in wt%). Yield
strengths of steel are also given
the fracture mode for the 36.5 at.% Al alloy changed from transgranular in air to grain-boundary failure in oxygen, while for the 40 at.% Al alloy, it remained unchanged.
These results suggest that the grain boundaries in 40 at.% Al alloy are relatively weak, and brittle intergranular fracture is a major cause for its low ductility at room tempera- ture.‘*‘) The boron addition of 300 ppm (by weight) increased the ductility of Fe40 at.%
Al alloy when tested in oxygen. This implies that boron strengthens the grain boundaries but is not effective in minimizing the environmental effect in Fe-40 at.% Al alloy.
The mechanical properties of binary FeAl are dependent on the aluminum content.
While the yield strengths increase with the increase of aluminum content up to 40 at.%
Al, room-temperature tensile elongations of aluminides decrease with the increase of aluminum content. Recent work showed that cast FeAl alloys, designated as FA-385, consisting of Fe-21.1 Al-O.42 MoO. 1 Zr-O.03 C (all in wt%), can possess reasonable room-temperature ductilities (of the order of 6 to 8%) and high-temperature strengths, particularly with microadditions of boron. Yield strengths and tensile elongations of FA-385, along with the properties of reaction-synthesized, air-induction-melted (AIM), and vacuum-induction-melted (VIM) binary FeAl with Fe-24 wt% Al composition (designated as Fe-240), are shown in Fig. 6. As can be noted from Fig. 6, the as-cast properties of Fe-24 wt% Al are similar to those of FA-385, and the casting method did not influence the observed properties. On the other hand, the fine-grained nature of hot-pressed Fe-24 wt% Al alloy from a powdered mixture of iron and aluminum exhibits much better yield strength than all other alloys. The properties of a conventional steel are also shown for comparison.
Low density, high specific strength, and high specific stiffness of FeAl alloys over steels and other nickel-based alloys are attractive for a variety of applications even though their creep resistance is low. Iron aluminides, unlike many commercial heating-element materials such as Ni-20 wt% Cr, iron alloys, nickel alloys, and superalloys based on nickel, cobalt, and iron, possess high electrical resistivities of the order of 150 uR cm or higher.
Thermal conductivities of iron aluminides are lower than the thermal conductivities of nickel-based alloys, and their thermal expansion coefficients are similar to those of stainless steels.“’
6. PROCESSING OF INTERMETALLICS
Melting remains the primary processing technique to obtain a variety of cast, wrought, and powder-metallurgical products, and the melting technique will determine whether an intermetallic can be obtained economically with good control of the composition, and with a minimal of defects and porosity in the cast structure.“’
Several factors to be considered in the melting of intermetallics include: (1) the melting points of aluminum and transition metals such as nickel, iron, titanium, and niobium; (2) the large amounts of aluminum present in intermetallics; and (3) the exothermic nature of formation of the intermetallic compound. Since melting of aluminum was only needed for the onset of aluminide formation in combustion synthesis of intermetallics, we considered exothermicity as an added advantage in melting and casting of intermetallics since it can save time and energy needed for melting. Of course, safety is a concern (as in the case of combustion synthesis) in melting of large quantities for structural parts.
Melting techniques, process parameters such as crucibles, melting atmosphere (air, inert gas, and vacuum), furnace loading of low- and high-melting-point metals, and, most importantly, the ability to use recycled material with close control of target composition must also be considered. The AIM technique is the most economical and can be used if a protective slag can be generated easily on top of the melt.
Our efforts to implement combustion-synthesis principles to the melting of nickel and iron aluminides resulted in modifying the conventional melting process in which aluminum melt stock is added to the molten nickel bath. We have shown that the addition of aluminum to the molten nickel melt stock initiated a violent exothermic reaction instantaneously.@’ A peak temperature of 2300°C was noted within 1 min, and the crucible was at a temperature of 2100°C for several minutes. The temperatures attained were above the melting temperatures of the commonly used crucibles such as alumina, zirconia, mullite, etc. Also, a high-temperature vapor cloud escaped from the crucible as a result of the oxidation of aluminum and zirconium.
7. Exo-MeltTM PROCESS
Recently, we addressed the above safety issues associated with the melting, casting, and processing of Ni,Al-, Fe,Al-, and FeAl-based alloys for the commercial utilization of intermetallics by the Exo-MeltTM process. The Exo-MeltTM process consists of dividing the melt stock into several parts in a furnace-loading sequence such that a very exothermic reaction with a high adiabatic combustion temperature is favored initially, leading to a molten product.
The furnace-loading sequence used in the Exo-MeltTM process is shown schematically in Fig. 7 for melting of nickel- and iron-aluminide alloys. It is important to note that the aluminum melt stock is placed vertically inside the wall of the zirconia crucible so that molten aluminum will react with part of the nickel melt stock in a controlled manner. All of the alloying elements of a particular melt stock are loaded between the top and bottom layers of nickel. The remaining nickel melt stock is loaded at the bottom. Temperature measurements and a video recording of the melting confirmed the occurrence of exothermic reaction in the top layer. Exothermic reaction continued until the reaction between the molten aluminum and the heated nickel melt stock was complete.‘28)
Thermocouple Leads
Zirconia Crucible
Thermoc 1.5 in.
spacing 1,2,&3 in.
Induction - Coils -Al
- Ni
- Alloying Elements - Cr, Zr, Mo,
B, etc.
- Ni
Fig. 7. A schematic representation of furnace loading employed in the Exo-MeltTM process for melting and casting nickel aluminides. A similar arrangement can be used for melting iron aluminides. Note that melting by the Exo-MeltTM process can be carried out in air under an argon
cover.
In the Exo-MeltTM process, the power required to melt a batch of IC-221M is 47% lower (2.92 kW h) than with the conventional process (5.5 kW h). This is due to the fact that exothermic reactions are initiated just after the melting of aluminum in the Exo-MeltTM process, whereas the melting of nickel melt stock is required in the conventional process. The time required to melt and pour the intermetallic by the Exo-MeltTM process is also reduced by 50% as opposed to a conventional process. All of these resulted in substantial savings in energy and a reduction in processing cost. While yield and ultimate tensile strengths did not differ significantly between the Exo-MeltTM process and the conventional process, the tensile elongations of conventionally melted alloy IC-221M are significantly lower than for the alloy processed by the Exo-MeltTM process.(“)
The target composition of an alloy was reached easily and safely with the Exo-MeltTM process as compared with the conventional process. The high levels of aluminum present in the nickel and iron aluminides promote rapid formation of a continuous A&O, film on top of the molten metal. Very low levels of oxygen and nitrogen elements in AIM
heats suggest that the oxide film is impervious for the penetration of oxygen and nitrogen. Typical oxygen and nitrogen values in nickel and iron aluminides are ~40 ppm by weight. These values are similar to oxygen levels of < 20 ppm by weight possible in aluminum-deoxidized steels.
8. CASTING OF INTERMETALLICS
Cast components of aluminides can be obtained by sand, investment, centrifugal, directional solidification, and innovative near-net-shaping techniques. The castability responses of the aluminides are related to the process as well as the product. The process-related castability concerns deal with the steps involved in the casting process, whereas the product-related issues deal with the porosity and microstructural features of the cast components.
Many components also require use of intermetallics in the fine-grained wrought condition. The commercial utilization of intermetallic alloys at a competitive cost requires their fabrication by conventional processes including: (1) hot extrusion; (2) hot forging;
(3) hot flat and bar rolling; (4) hot swaging; (5) cold flat and bar rolling; and (6) cold drawing into tube, rod, and wire. The general criteria that determine the processing response of an alloy include: (1) ductility (30% or greater) of the cast structure at desirable hot-processing temperature and at industrial strain rates approaching 10-l s-l, (2) flow stress of the cast structure at processing temperatures and strain rates one-fifth to one-tenth of the flow stress at room temperature, (3) broad temperature range with high ductility, (4) absence of the low-melting-point liquid formation at the processing temperature, and (5) absence of environmental effects and good intermediate-temperature ductility (2 10%) to prevent cracking during cooling from the hot-processing temperature.
9. WELDING OF N&Al-BASED ALLOYS
The full potential of nickel and iron aluminides can only be reached with further advances in welding and innovations in the alloy design and processing. We briefly discuss some of the welding and future work to be carried out in nickel and iron aluminides.
There are two issues in the welding of aluminides: the first deals with cosmetic repair of casting defects, and the second deals with the structural welds needed for the assembly of components. To make either one of the welds, the welding-wire composition and welding procedures need identification. In support of these needs, the welding-wire compositions have been identified. A composition identified as IC-221LA (in wt%, consisting of 16% Cr, 4.5% Al, 1.5% MO, 1.5% Zr, and balance Ni) is used for weld repairs and low-stress or less demanding structural welds. The composition identified as IC-221 W is used for normal structural welds. The currently used wire of both IC-221LA and IC-22 1 W compositions is prepared by incorporating a powder composition in a nickel sheath (powdered-core wire). The pre-alloyed powder composition is adjusted for the nickel content that comes from the sheathing. The powdered-core wire is acceptable for welding by gas tungsten arc (GTA) and machine inert gas (MIG) methods.
The GTA method has been used most commonly for welding nickel aluminides. Section thicknesses up to 25 mm (1 in) prepared by air melting using the Exo-MeltTM process have been successfully welded. For consistently good-quality welds, carbon, sulfur, silicon,
boron, and zirconium need to be closely controlled. These elements are easy to control but need to be recognized at the time of melting. The tensile and creep properties of the weldments meet or exceed those of the base metal.
The IC-221 LA and IC-221 W wires have also been successfully used in depositing Ni,Al-based alloy on carbon, chromium-molybdenum, and stainless steel.
10. CONCLUSIONS
The mechanical, oxidation, and carburization properties of intermetallic compounds based on Ni,AI-, NiAl-, Fe,Al-, and FeAl-based alloys are discussed and compared with some commercially available materials. Boron-doped Ni,Al-based alloys have significantly better tensile and creep properties than the commercially available HU alloy, and the properties are comparable to those of Haynes 214. The addition of boron is not effective in improving the tensile properties of NiAl. The tensile properties of Fe,Al-based alloys are comparable with the properties of yttria-dispersed MA-956 but exhibit poor creep resistance compared with MA-956.
We have shown that the exothermicity associated with the formation of aluminides can be harnessed to save time and energy needed for the melting and casting of intermetallics in air under an argon cover. Both processing and casting of intermetallics are discussed in detail, and it is shown that processing parameters influence the properties of the intermetallic. The Exo-MeltTM process is inherently safe and limits the temperature rise during the melting of intermetallics, as compared to a conventional melting process in which the addition of aluminum to molten nickel raises the temperature of the melt above the melting temperatures of the crucibles. The Exo-MeltTM process has been used to melt and cast a wide variety of components for structural use, and a total of over 15 240 kg (15 ton) of intermetallics was cast by the Exo-MeltTM process in 1995.
There is a renewed industrial interest in the use of nickel and iron aluminides for a wide range of structural and coating applications. Alloy design approaches, processing advances, and the development of the Exo-MeltTM process have enabled the realization of intermetallic properties, known for several decades, and paved the way for the commercial utilization of nickel and iron aluminides. The intermetallics discussed in this paper can be cast successfully, and the molten-metal technology will allow wrought and forged products only. The widespread use of nickel and iron aluminides requires the development of weldable alloys, welding methods, and technology for the production of sheet, wire, and rod forms.
ACKNOWLEDGEMENTS
S. C. Deevi is on a sabbatical at the Oak Ridge National Laboratory under the Philip Morris Fellowship Program. The authors thank Professor R. W. Cahn of the University of Cambridge, U.K. for encouragement; C. R. Howell for plotting the data;
R. W. Swindeman and E. K Ohriner for internal review; K. Spence for editing; and M.
L. Atchley for preparing the manuscript. Research for this work was sponsored by the U.S.
Department of Energy, Assistant Secretary for Energy Efficiency and Renewable Energy, Office of Industrial Technologies, Advanced Industrial Materials (AIM) Program;
the Office of Basic Energy Sciences, Division of Materials Sciences; the Office of Computational and Technology Research, Laboratory Technology Research Division;
and the Office of Fossil Energy, Advanced Research and Technology Development Materials Program [DOE/FE AA 15 10 10 0, Work Breakdown Structure Element ORNL-2(H)], under contract DE-AC05-960R22464 with Lockheed Martin Energy Research Corp.
REFERENCES I. N. S. Stoloff and R. Cl. Davies, Prog. Mater. Sci. 13, 1 (1966).
2. J. H. Westbrook (Ed.), Intermetallic~Compounds. John Wiley and Sons, Inc., New York (1967).
3. E. M. Schulson. in Brittle Fracture and Toughening of Intermetallic Comoounds (N. S. Stoloff and V. K. Sikka eds). Chapman’and Hall, New York (1966). y
4. S. M. Kopley and B. H. Kear, Trans. Metall. Sot. 239, 977 (1967).
5. P. H. Thornton, R. G. Davies and T. L. Johnston, Metall. Trans. 1, 207 (1970).
6. S. C. Deevi, unpubhshed work.
7. S. C. Deevi and V. K. Sikka, fnterme/allics (in press).
8. MRS Proceedings, Vols 81, 133, 288 and 213. Materials Research Society, Pittsburgh, PA (1991).
9. S. H. Whang, C. T. Liu, D. P. Pope and J. 0. Stiegler (eds), Hugh Temperature Aluminides and Intermetallics, Proc. TMS/ASM Symposium. TMS-AIME, Warrendale, PA (1990).
IO. R. Darolia, J. J. Lewandowski, C. T. Liu, P. L. Martin, D. B. Miracle and M. V. Nathal (Eds.), Structural Zntermetallics. The Minerals, Metals and Materials Society. Warrendale, PA (1993).
I I. R. W. Cahn, Metals, Materials and Processes 1, I (1989).
12. C. T. Liu, in Structural lntermetallics (R. Daroha. J. J. Lewandowski, C. T. Liu, P. L. Martin, D. B. Miracle and M. V. Nathal eds), p. 365. The Minerals, Metals and Materials Society, Warrendale, PA (1993).
13. C. T. Liu, Scripta Metall. Mater. 21, 25 (1992).
14. K. Aoki and 0. Ioumi, Nippon Kinzoku Gakkaishi 43, 1190 (1979).
15. C. T. Lm, C. L. White and J. A. Horton. Acta Metall. 33, 213 (1985).
16. A. I. Taub, S. C. Huang and K. M. Chang, Metall. Trans. A. 15A, 399 (1984).
17. M. Takeyama and C. T. Liu, Mater. Sci. Eng. A153, 538 (1992).
18. C. A. Hippsley and J. H. DeVan. Acta. Metall. 37, 1485 (1989).
19. S. C. Deevi and V. K. Sikka, Mechanical properties of N&AI intermetallics obtained by Exo-MeltTM process.
To be submitted for publication.
20. Cast Heat Resistant Alloys, Report No. 1196. The International Nickel Company, Inc., New York (1974).
21. Haynes Alloy No. 214. Haynes International, Kokomo, IN (1986).
22. E. P. George and C. T. Liu, J. Mater. Res. 5, 754 (1990).
23. R. Darolia, in Structural Intermetallics (R. Darolia, J. J. Lewandowski, C. T. LIU, P. L. Martin, D. B. Miracle and M. V. Nathal eds), p. 494. The Minerals, Metals and Materials Society, Warrendale, PA (1993).
24. R. D. Noebe, R. R. Bowman and M. V. Nathal. m Physical Metallurgy and Processing of Intermetallic Compounds (N. S. Stoloff and V. K. Sikka eds). Van Nostrand Reinhold, New York (1994).
25. T. Hong and A. J. Freeman, Phys. Rev. B 43, 6446 (1991).
26. C. G. McKamey, J. H. DeVan, P. F. Tortorelli and V. K Sikka, J. Mater. Res. 6, 8, 1779 (1991).
27. C. T. Liu, V. K. Sikka and C. G. McKamey, Alloyj Development of FeAl Aluminide Alloysfor Structural Use in Corrosive Encrronments. ORNL/TM-12199. Lockheed Martm Energy Systems, Oak Ridge Nat. Lab., Oak Ridge, TN (1993).
28. S. C. Deevi and V. K. Sikka, Intermetallics (m press).
