intermetallic alloys: an assessment

(1)

ELSEVIER

Ordered

Intermetallics 5 (1997) 579-596

PII: SO966-9795(97)00045-9

OVERVIEW

intermetallic alloys: an assessment

C. T. Liu,” J. Stringer,b J. N. Mundy,c L. L. Horton” & P. Angelini”

“Metals and Ceramics Division, Oak Ridge National Laboratory, Oak Ridge, TN 37831, USA bEPRI, Palo Alto, CA 94303, USA

‘Division of Materials Sciences, Office of Basic Energy Sciences, Department of Energy, Germantown, MY 20874-1290, USA

(Received 19 June 1997; accepted 19 June 1997)

The paper summarizes our present understanding, as established at a recent workshop, of two classes of intermetallic alloys: nickel and iron aluminides, which are currently used by industries; and advanced intermetallic alloys including silicides and Laves-phase alloys, which have a great potential to be developed as new high-temperature structural materials for future industrial use. The workshop emphasized close interaction and co-operation between basic research, applied research, and industrial development, and stressed discussion of critical scientific and technological issues. The current status of these intermetallic alloys was assessed, and the directions for future research and development, as well as emerging opportunities, were identified. The information presented in the text is summarized from the presentations at the workshop, and so no references are given to the published literature. However, an extensive bibliography is appen- ded, in which further details may be found. 0 1997 Elsevier Science Limited Keywords: A. iron aluminides (based on FesAI and FeAl), Laves phases, nickel aluminides (based on N&Al), silicides, B. allov design. environmental embrittle-

G. load-bearing applications, miscel- ment, C. joining, E. ab-initio calculations.

laneous.

1 INTRODUCTION

Ordered inter-metallic alloys based on aluminides and silicides constitute a new class of metallic materials that have unique properties for structural applications at elevated temperatures in hostile environments. Their attractive properties include good high-temperature strength, resistance to oxidation and corrosion, relatively low material density and high melting point. Most intermetallic alloys, however, exhibit brittle fracture and low tensile ductility at ambient temperatures, and their use as engineering materials is restricted in many cases by their poor fracture resistance and limited fabricability. In addition, a number of them are sensitive to moisture in the environment at lower temperatures. For the past two decades, considerable effort has been devoted to the research and development of inter-metallic alloys. Basic research has been focused on understanding the mechanisms of deformation and brittle fracture; while applied research has concentrated on improving their mechanical and metallurgical properties by alloying additions, microstructural control, and the

579

optimization of materials processing variables.

As a result of these efforts, a number of new intermetallic alloys based on nickel, iron, and titanium, which possess attractive mechanical properties for many industrial applications, have been developed.

A good example is N&Al, which has been rendered ductile by alloying additions. Polycrystalline N&Al showed brittle grain-boundary fracture without appreciable tensile ductility at room temperature.

Microalloying with boron in ppm (parts per million) ranges resulted in a dramatic improvement in the tensile ductility (from 2 to 50%) of N&Al alloys with hypostoichiometric compositions (Nis +,A1 1 -,). Industrial interest in these intermetallic alloys is high, and several new aluminide alloys developed recently are on the market for industrial use.

In order to assess the current status and explore new and emerging opportunities, a workshop was organized to review the recent progress and to identify new directions in the research and development of ordered’intermetallic alloys. The Work- shop on Intermetallic Alloys held in Atlanta, Georgia, 3-4 June 1996, was jointly sponsored by

(2)

580 C. T. Liu et al.

three offices of the US Department of Energy (DO): (1) the Division of Materials Sciences of the Office of Basic Energy Sciences; (2) the Advanced Industrial Program of the Office of Industrial Technologies; and (3) the Advanced Research and Technology Development Materials Program of the Office of Fossil Energy. The Workshop sought close interaction and co-operation between basic research, applied research, and industrial development, and stressed discussion of critical scientific and technological issues with the aim of helping to define the future research and development work needed. Under the co-chairpersons C. T. Liu and J. Stringer, the\41 materials scientists and engineers listed in the Appendix participated in the workshop.

The workshop focused on two groups of intermetallic alloys: nickel and iron aluminides, which are currently used by industries; and advanced intermetallic alloys, including silicides and other intermetallics (such as Laves phase alloys), which have a great potential to be developed as high- temperature materials for future industrial use.

Titanium aluminides based on TiAl and Ti3A1 constitute a promising class of inter-metallic alloys, which have been broadly studied recently. Since many conferences and workshops have been focused on these inter-metallic alloys, they were not included in this workshop. The original suggestion and general guidance for conducting this workshop were provided by DOE Program Managers, John Mundy, Jim Carr, and Charlie Sorrell.

2 NICKEL AND IRON ALUMINIDES

The critical scientific and technological issues for the nickel and iron aluminide alloys are summarized in the following six topic areas: alloy design and structural use of NisAl Alloys; ductility and fracture of N&Al; alloy design and structural use of iron aluminides; deformation and fracture of iron aluminides; welding of nickel and iron aluminides;

and corrosion of nickel and iron aluminides.

2.1 Alloy design and structural use of Nidl alloys Development in alloy design, processing, and fabrication technology of NisAl has led to ductile, high-strength, castable and weldable N&Al-based alloys. Some of the attributes of NisAl-based alloys include: resistance to oxidation and carburizing atmospheres at temperatures up to 1100°C; high yield strength at 650 to 1100°C; fatigue resistance superior to many superalloys; high creep strength;

excellent wear resistance at high temperatures ( > 600°C); and superior corrosion resistance due to the formation of protective oxide films.

In the alloy design efforts, various elements are important to produce alloys with superior properties. Microalloying with boron is found to dramatically improve tensile ductility and suppress brittle grain-boundary fracture at ambient temperatures.

High ductility is achieved by controlling the boron levels in conjunction with controlling the aluminum concentration to substoichiometric values (below 24 at%). Properly, these alloys should be written NQ +X * All_,, but in this text all alloys will be labeled NisAl for simplicity. High strength and oxidation resistance are also of great importance for structural materials, and significant develop- mental efforts have been devoted to this area.

Testing of the oxidation behavior of some of the initial alloys at high temperatures has revealed a ductility minimum existing at intermediate temperatures (from 700 to 900°C). This behavior has been found to be due to a dynamic oxygen embrittlement effect. The ductility can be improved, however, by replacing part of the aluminum content with chromium in order to produce a protective oxide film that can form fast enough at these temperatures. The use of molybdenum in the alloys is also an important factor in improving strength at room and elevated temperatures. Sev- eral N&Al-based alloys have been developed which have improved strength and oxidation resistance over conventional alloys at temperatures up to 1100°C. Alloy development has been utilized in solving many issues; however, the need for increased understanding of many of the basic mechanisms and process/structure relationships remains.

The alloy design and processing technologies development have focused on castable and weldable N&Al alloys. Limited effort has been placed on wrought alloys thus far. The focus on castable alloys is a result of industrial input. The casting process is a near or near-net-shape technology that can provide components directly applicable in industrial settings. Specific alloy development to ensure castability is necessary, and various castable alloys have been developed.

It is critical also that alloys should be weldable.

Although casting allows production of near-net- shape components, these must often be assembled to form larger structures. The joining process may be mechanical (bolting or riveting, for example);

but welding is preferred in many cases. In general, one may distinguish two different situations in

(3)

Ordered intermetallic alloys: an assessment 581 assembly by welding: the first is the welding of a

material to itself: the second (and more demand- ing) is the welding of a material to another material-the ‘dissimilar weld’. It is also generally required that materials should be repairable, and welding is important for this objective, too. The repair may involve cutting out a damaged section, and welding in a replacement; alternatively the welder may deposit new material on a damaged component, building up a replacement section. All of these approaches present a challenge to materials. Most importantly, localized heating and melting may impose severe stresses on both the base and weld metal. These can sometimes be relieved by post-weld heat treatment, but this is not always easy. In addition, it is not always easy to preserve the composition and microstructure of the component materials in the vicinity of a weld, and where these are critical to the performance of a material (as is often the case with the advanced intermetallics) this may be a critical issue in their applicability.

The mechanical response of a weld must be understood with respect to the alloy microstructures and processing conditions. Properties of both the solid and liquid phases must be known. Mod- eling of solidification, welding, etc., requires that the thermophysical properties and other properties of alloys be known. The composition of Ni3Al alloys has a very strong influence on the casting and welding response with the zirconium content being very critical. Systematic studies have been performed of the effects of alloying and residual elements on the welding and casting response of N&Al. These studies show that important residual elements that need to be controlled include silicon, sulfur, carbon, boron, and iron. Joining efforts have also focused on weld wire development.

Melting technology is also a critical process especially when large size industrial heats of materials are produced. The use of the exothermic energy released during the reaction process has been demonstrated to be safely controlled in the melting of large heats. The process known as Exo- melt@ also results in a more energy efficient melting process. The technology has been transferred to industry (United Defense LP/Steel Products Divi- sion) and over 60000 lbs (27 200 kg) of N&Al alloys have been successfully melted in a year.

Castings have been poured in green sand, air-set, and Replicas@ molds. ORNL has worked with industry to develop secondary standards for the chemistries and to characterize thermal and physical properties of these alloys.

Industrial applications include heat-treating trays, thermal processing equipment, and forging dies. The payoffs for these applications include:

reduced downtime due to worn or broken parts;

total reduced costs for replacement parts, reduc- tion in unscheduled down time, and associated costs; and increased process efficiencies through higher operating temperatures.

2.2 Ductility and fracture of Ni&l

Although N&Al has the high-symmetry Lls crystal structure, it has long been known to be brittle in the polycrystalline form. Until recently, the cause for brittleness was believed to be the strongly ordered nature of this compound, which was thought to result in intrinsically weak grain boundaries. Over the course of the last few years, however, an extrinsic source of grain-boundary brittleness has been discovered in N&Al, namely, the water vapor (HzO) present in ambient air. The embrittlement mechanism is similar to that found in many ordered intermetallic alloys containing reactive elements (e.g. Al, Ti, Si) and has recently been confirmed by temperature-programmed thermal desorption and X-ray photoelectron spectroscopy experiments. The dissociative adsorption of HZ0 on N&Al has been found to produce 0 atoms (or OH radicals) attached to the Al atoms and H atoms attached to the Ni atoms.

The atomic H thus produced is then thought to embrittle the crack-tip regions. Consistent with this HzO-induced embrittlement mechanism, the ductility of Ni3A1 is found to increase with increasing strain rate, decreasing temperature, and decreasing amounts of Hz0 in the test environment.

Low-pressure ( < 10 torr) hydrogen gas has been found to be relatively benign compared to HZ0 at similar pressures, presumably because H2 does not dissociate readily into atomic H on the surfaces of N&Al. However, Hz does dissociate into H in the vicinity of the hot tungsten filaments of ion gauges (devices commonly used to measure gas pressures).

As a result, the ductility of specimens measured with an ionization gauge in the on position is found to be reduced from 50 to 25% of that measured with the gauge in the off position. However, when the availability of atomic H is carefully limited (by testing in an ultrahigh vacuum environment with the ion gauge off), polycrystalline N&Al is found to be intrinsically ductile, with tensile elongations exceeding 40% and predominantly ( > 70%) transgranular fracture.

(4)

582 C. T. Liu et al.

Doping N&Al with 100 wppm (parts per million by weight) B has long been known to suppress intergranular fracture and dramatically improve the room-temperature ductility of N&Al. The conventional view has been that B improves ductility by either enhancing grain-boundary cohesion or facilitating slip transfer across grain boundaries.

Recent results, however, show that the main role of B is to counteract the embrittlement caused by moisture in ambient air (although it also appears to have a role in improving grain-boundary cohesion). There is some evidence to indicate that B decreases H diffusion along NisAl grain boundaries, and that it slows down the dissociative adsorption of water vapor on freshly fractured surfaces of N&Al. However, additional research is needed to clarify the detailed mechanism. Interest- ingly, B also appears to embrittle N&Al in low- pressure HZ: in IO-torr Hz the ductility of N&Al doped with 50 to 1OOwppm B is actually lower than that of B-free N&Al. A possible explanation of this unexpected result is that B facilitates the dissociation of Hz into atomic H on the surfaces of N&Al. In other words, while B suppresses the formation and/or uptake of HzO-produced atomic H, it appears to enhance the generation of HZ-produced atomic H. These results underscore the importance of understanding the fundamental atomic-level mechanisms contributing to ductility and fracture of N&Al and its alloys. Other recent results, not discussed in great detail at the Work- shop, suggest that B does indeed increase grain boundary cohesive strength, and that perhaps these stronger boundaries require more hydrogen to embrittle them. The scientific aspects of these issues are discussed in more detail in Section 3.

2.3 Ahoy design and structural use of iron aluminides

There are a number of advantages associated with the development and use of iron aluminides (for example, high-temperature corrosion resistance, density, electrical resistivity, and lack of toxic elements), but several drawbacks remain. The principal ones include poor room-temperature ductility and toughness, and lack of strength at temperatures (> 600°C) where iron aluminides are still corrosion resistant and weldable.

Much of the alloy development efforts have focused on addressing the issue of room-temperature ductility and the associated moisture- (hydrogen-) embrittlement phenomenon. It was found that several metallurgical or processing modifica-

tions could improve room-temperature ductility (to about lO-20% elongation): alloying with chromium; addition of specific amounts of B; stabiliza- tion of the B2 structure (rather than the more ordered DO,); refinement of grain structure and control of grain shape; complete disorder (16 at%

Al); and processing with surface protection layers.

Less work has been done with respect to fracture toughness. Limited work on FeAl alloys indicates that their impact toughness can be substantially improved by control of microstructure in powder- metallurgy alloys and in hot-extruded ingot-metallurgy alloys. Recent efforts devoted to improving the creep resistance of ingot-processed FesAl alloys have shown that proper heat-treatment of an Fe-28Al-5Cr (at%) composition yields a lifetime at 650°C almost as long as that shown by Type 316 stainless steel. Fatigue resistance is affected by the ordered structure (DO3 is worse than B2) and alloying.

The applications of iron aluminides are, for the most part, based on their excellent high-temperature corrosion resistance in environments that destroy Fe-Cr-Ni steels and other alloys. Two of the most promising applications appear to be: hot- gas filters for coal gasification and combustion systems, and radiant burner tubes. FeAl alloys have recently demonstrated outstanding resistance to carburization, various molten salts, and molten aluminum; and relevant new applications are being pursued. Thin sections of iron aluminides may have application in catalytic converters as catalyst support structures. Because of elevated-temperature strength limitations, the corrosion resistance of iron aluminides may be best utilized as clads or coatings.

2.4 Deformation and fracture of iron aluminides Much of the fundamental work on the deformation and fracture of iron aluminides has been done with the FeAl system rather than the Fe3Al system.

While the fundamental mechanisms underlying environmental embrittlement are not yet fully elu- cidated, the phenomenon is clearly important in this class of alloys. As expected, these ambient- temperature environmental effects are strain rate dependent. At high strain rates, no environmental influence is observed: at lower ones the effects of alloying additions can be dominated by the HzO/

H2 embrittlement reactions. There is evidence that hydrogen- dislocation and hydrogen-impurity interactions play key roles in embrittlement and alloy design has been effective in mitigating this environment effect.

(5)

Ordered intermetallic alloys: an assessment 583

Vacancies play a particularly important role in determining mechanical behavior in the FeAl system. It is relatively easy to form vacancies, but their mobility is restricted by a high activation energy. Yield strength increases with increasing aluminum concentration (CA,) corresponding to a concomitant increase in the vacancy concentration.

However, correlations of yield strength with vacancy concentration (C,) are complicated by changes in operative slip systems with temperature.

Dramatic changes in C, with CA, are found near the 50 at% Al composition. There is a yield strength anomaly with temperature-a local maxi- mum is found around SOO”C-that can be associated with vacancy formation and dislocation pinning. This yield-strength-anomaly model may be applicable to other B2 alloys.

2.5 Welding of nickel and iron aluminides

Ten years ago it was shown that the welding behavior of N&Al alloys showed that they were susceptible to heat-affected zone (HAZ) cracking, and that cracking was minimized at boron levels of about 200wppm. Additional basic studies showed that N&Al alloys containing 10 wt% Fe were more resistant to cracking than binary alloys. An incon- sistent correlation of hot ductility to cracking response led to computer modeling of thermal stresses during welding. Modeling showed that high thermal conductivity could counteract low hot ductility and result in crack resistant alloys. It was also found that fine domain structures and chromium additions could improve weldability and ductility at elevated temperatures.

More recently, programmatic emphasis on technology transfer activities has led to a focus on issues related to weld repair of cast NissAl alloys (mainly, E-22 1 M: Ni-&OAl-7.7Cr- 1*4Mo-

1*7Zr-O.O08B, wt%). These alloys are susceptible to solidification cracking during welding. Microstruc- tural analysis indicated that the cracking was related to the zirconium concentration. Control of the zirconium level produced alloys that were resistant to solidification cracking, and could produce crack-free weld deposits on the cast Ni3A1 alloys. Further alloy development efforts produced additional crack-resistant compositions for weld filler metals. Important issues being addressed in current activities include: difficulties associated with producing welding wire; the influence of impurity elements on cracking, welding procedure development; and weldment mechanical properties.

Less effort has been devoted to welding studies of iron aluminides, but the basic effects of boron and major alloying elements on weldability have been examined. Both solidification (hot) and cold- cracking have been observed. Increased carbon and decreased boron concentrations improved resistance to solidification cracking. While hot-cracking can normally be avoided by alloying and preheat, cold-cracking is more problematic owing to the susceptibility of iron aluminides to hydrogen embrittlement. Refinements of fusion zone microstructures have improved resistance to hydrogen- induced cracking.

Significant effort has recently been focused on the development of procedures and materials for depositing weld overlays of iron aluminides (both Fe3Al and FeAl) on ferrous alloy substrates for use as high-temperature, corrosion-resistant coatings.

Hydrogen cracking of these overlays is a problem, but it often can be avoided by appropriate processing, composition control, and/or postweld heat treatment. Filler wires for both gas tungsten arc and gas metal arc deposition of these coatings have been produced, but much work remains to be done with respect to process development and control (including automation) and the tailoring of weld microstructures and compositions for specific environments.

2.6 Corrosion of nickel and iron aluminides

The corrosion resistance of nickel or iron aluminides can only be uniquely specified when the tar- geted application area (for example, a type of industrial process or a particular energy production method) is known as there is a wide range of temperatures and environments (oxidizing, sulfi- dizing, carburizing, etc.) of interest. However, in general, both nickel and iron aluminides have good-to-excellent high-temperature oxidation resistance when an A1203 surface scale forms and remains adherent. The large aluminum contents of the nickel and iron aluminides facilitate the growth of a continuous, slow growing, protective A1203 scale particularly at relatively low partial pressures of oxygen. Alloying additions that improve (or degrade) alumina scale adhesion have been identified for both nickel and iron aluminides. Typically, small additions (about 0.1 at%) of zirconium, yttrium, hafnium, or certain other reactive elements are effective in improving resistance to scale spallation.

Alumina scales can also provide high-temperature corrosion resistance in mixed-gas or salt

(6)

584 C. T. Liu et al.

environments if the oxygen partial pressure (or, more properly, oxygen availability) is sufficiently high to allow the rapid establishment of a continuous external A1203 layer. This is the case for Fe3Al and FeAl alloys exposed to SO2 (combustion) and H$H2-H20 environments. Even at high sulfur (1 Om6 atm) and low oxygen (1 O-22 atm) partial pressures. certain iron aluminides show excellent corrosion resistance at high temperatures because surface alumina can form. Consequently, information from more fundamental studies of the formation, growth and adhesion of protective alumina scales on iron aluminides has relevance to corrosion in mixed gases. The effects of variations in the concentrations of aluminum, chromium andminor alloying elements in Fe-Al alloys on sulfidation in low-PO,, environments have been characterized. In contrast to the iron aluminides, the sulfidation resistance of nickel aluminides is not good, even if preoxidation is used to form an A1203 layer prior to exposure to a sulfur-bearing gas.

There is a limited amount of data on the high- temperature corrosion of nickel and iron aluminides in other mixed gases and salts. Alloys based on Ni3Al have good carburization resistance, and FeAl alloys showed relatively low corrosion rates in highly oxidizing molten nitrate salts. The aluminides are not very corrosion resistant in sulfate salts (but chromium additions may help). While early laboratory results on the corrosion of iron aluminides in a boiler-ash deposit were not encouraging, recent data from exposures of FesAl alloys in an operating coal-fired power plant (in the presence of ash) showed rates as good, or better, than stainless steel.

In certain high-temperature environments, the corrosion resistance of nickel and iron aluminides is maintained at temperatures where these alloys are not strong. Consequently, the application of these materials as corrosion-resistant coatings is attractive (particularly for iron aluminides in sulfur-bearing gases). Corrosion studies with iron- aluminide coatings produced by weld-overlay techniques and electrospark deposition have shown good behavior in a variety of mixed gases, except when HCl was present in a combustion environment. Recently, efforts to improve the high-temperature strength of iron aluminides by the use of oxide dispersions have begun. Associated corrosion studies have shown that the good oxidation and sulfidation resistance of iron aluminides can be maintained with the proper types and amounts of oxide additions.

3 CRITICAL SCIENTIFIC AND TECHNICAL ISSUES IN Ni AND Fe ALUMINIDE

At the workshop many issues were raised with regard to the areas outlined in Section 2. This section discusses the more critical issues as defined by the participants. The issues are divided into scientific and technical categories, although with considerable overlap.

3.1 Critical scientific issues

The discussions are summarized separately for N&Al and for the iron aluminides, but it must be remembered that many issues are essentially the same.

3.1 .I Ni3Al

3.1 .I .l Environmental embrittlement. As indicated earlier, this is a pervasive problem, not only for N&Al, but also for FeAl, Fe3A1, NisSi, and other inter-metallic alloys; the general feeling of the meeting was that this is the most important issue for the successful practical development of these materials. The principal form of environmental embrittlement is due to hydrogen, in most cases introduced as a result of the dissociation of H20, although the nature of the dissociation step is far from fully understood.

The two most common suggestions for the mechanism of the embrittlement are: (a) decohe- sion or (b) H-enhanced dislocation mobility. In-situ straining experiments in the transmission electron microscope indicate that both phenomena occur;

however, when intergranular failure takes place, it is generally along the boundary plane, suggesting decohesion. The presence of atomic hydrogen, generated for example by an ionization gauge within the experimental system, enhances the embrittlement; the cause for this is not fully understood. It appears that the grain-boundary morphology in polycrystalline materials is a factor, with material with elongated grains being more ductile, but it is difficult to separate the effects and be sure which is associated with hydrogen. There has been little study of slip in polycrystalline alloys or of the role of the detailed grain-boundary morphology. So far as slip modes are concerned, the major studies have been on the changes of slip systems with temperature in N&Al single crystals; but in the case of Fe3A1, the DO3 structure is more sensitive to embrittlement than the B2 structure, implying that slip mechanisms may be involved.

(7)

Ordered intermetallic alloys: an assessment 585 At elevated temperatures, oxygen can also

induce embrittlement in N&Al alloys but not in iron aluminide alloys. The detailed mechanism of oxygen embrittlement in N&Al alloys is not well understood at the present time.

In the case of Ni-24 at% Al, it appears necessary to consider the interacting effects of small amounts of sulfur, boron, and hydrogen. Sulfur segregates to grain boundaries in the absence of boron.

Hydrogen cosegregates with sulfur. Boron greatly reduces the segregation of sulfur to the grain boundaries and reduces (but does not eliminate) H segregation to the grain boundaries. However, while boron reduces the embrittlement of N&Al in moist ambient air, it appears to embrittle it in low- pressure hydrogen.

3.1.1.2 Effect of grain-boundary structure and orientation on mechanical properties. Most of the available experimental data on N&Al polycrystals are for the special cases of grain-boundary charac- ter and the boundary habit plane, which are controlled by alloy composition and processing history. In order to obtain intrinsic mechanical properties of grain boundaries and understand the extrinsic sources of grain-boundary embrittlement, quantitative characterization of the distribution of grain-boundary characters and orientations is a prerequisite for assessing mechanical behavior of N&Al polycrystals.

3.1 .I.3 Elevated-temperature creep-fatigue-envir- onmen tal interactions and thermal mechanical fatigue (TMF) . For any high-temperature application, the TMF behavior is of critical importance. The creep-fatigue behavior is also of importance, and in both of these cases, the effect of the environment may be significant. However, these effects are scar- cely mentioned in the literature. The effect of minor solute additions may have both positive and adverse effects on the same alloy. For example, Zr enhances the fatigue crack growth resistance of Fe3A1 but degrades weldability. Optimum solute contents for beneficial effects are unknown for most solutes.

3.1.2 Iron aluminides

3.1.2.1 Environmental embrittlement, including solute effects, mechanisms, and kinetics of hydrogen

ingress (FejAl and FeAI). The issue of most importance here appears to be the development of coatings or other surface modification techniques to inhibit environmental embrittlement. Refinement of grain structure and control of grain shape and degree of recrystallization are all important in reducing environmental embrittlement in iron aluminides at ambient temperatures.

3.1.2.2 Vacancy-solute-property interactions (FeAZ). The interaction between vacancies and their clusters with the slip modes need to be investigated. To better understand the observed yield strength anomaly and to develop solid-solution strengthening at elevated temperatures, interactions among vacancies and their clusters, glissile dislocations of the active slip system, and substitutional and/or interstitial solutes need to be investigated. The role of solute atoms in vacancy clustering and the dynamic interaction of superdi- slocations with vacancies are believed to be important aspects of the overall problem.

3.1.2.3 Improved impact toughness while retaining high-temperature strength (FejAl and FeAI). At high-strain-rate deformation (such as impact tests), the embrittlement due to the environmental effect is expected to be suppressed. However, both Fe3Al and FeAl alloys generally show poor impact resistance, which is independent of tensile ductility at room temperature. The mechanism governing the impact properties of iron aluminides is not clearly understood at the present time, but it will be important to remember that any remedies must be consistent with retaining the high-temperature strength.

3.2 Critical technological issues 3.2.1 Impact toughness

Much of the work on mechanical properties of intermetallics has been concerned with strength and ductility, but for service, toughness is an issue of major importance. This issue is a major concern for the iron aluminides but not for N&Al. Creep was not considered to be a critical issue for many applications.

3.2.2 Fabricability

Many issues were discussed, including welding and joining, machinability, and castability; some issues noted in the next sub-section. Castability includes the issue of directional solidification (also single crystal products). Coating, both in the sense of coating the intermetallic structural components to protect them from environmental degradation and in the sense of using the intermetallics as coatings themselves for structural components of other materials, was a topic which was considered to be important.

3.2.3 Scale-up to industrial production

An obvious issue is increase of the melt size and the size of the components to be manufactured.

The variability in properties associated with scale-

(8)

586 C. T. Liu et al.

up was discussed as well as the associated need for quality assurance. This implies a recognition of the properties of most importance and the availability of techniques for assessing the process variability in regard to these key properties. The importance of relating benefits to cost, and also the need to define specific applications before any of the above optimizations or the potential benefits can be assessed was emphasized. The examples of diesel- engine turbochargers and pistons, heat-treatment furnace furniture, radiant burner tubes, hot-gas filters for coal gasification and combustion systems, and catalyst supports in catalytic converters were instanced; it is immediately apparent that the optimum property set is significantly different for each application, and the fabricability requirements are also different. Thought must be given to developing a database and establishing primary standards to enable increasing industrial participation.

3.2.4 Other issues

The effect of minor elements excited significant interest. The effects of sulfur, boron, and carbon have been discussed. The welding behavior of the intermetallics can be very sensitive to the presence of minor constituents. Some minor substitutional species-notably zirconium, yttrium, and cerium- may be beneficial in one way or another. The important role played by minor elements will have an impact on the use of scrap in the overall production process, and it also means that mold/metal reactions must be considered, either in the ingot- production step or in the final product if this is cast.

The fatigue life of nickel aluminide turbochargers was inferior to other products, and it appears that this is due to the interaction of Zr in the N&Al alloys with shell materials and the formation of Zr oxides during investment casting. It is not easy to machine N&Al-base alloys. A development of eco- nomic ways for surface machining is important for commercialization of nickel aluminide alloys. The machinability may also be dependent on the detailed alloy chemistry. The chemistry of the alloys at a commercial scale will need to be controlled carefully for a number of minor constituents, and chemical analysis standards will have to be developed.

4 ADVANCED INTERMETALLIC ALLOYS 4.1 MoSi, and composites

MoS&based materials have significant potential for elevated-temperature applications in oxidizing

and combustion environments. MoSiz possesses a number of attractive high-temperature properties and attributes, including a melting point of 203O”C, oxidation resistance up to 1900°C; ther- modynamic and chemical stability with other ceramic reinforcements (Sic, SisN4, ZrOz, AlzOs, mullite (3A1203.2Si02), TiB2, and TIC); and the potential for alloying with other high melting-point silicides (W& NbSi2, MO&, Ti&). Some of the critical issues which have hindered the development of MoSiz for high-temperature structural applications, however, include its intermediate- temperature oxidation pest behavior, low fracture toughness (3 MPa.m’/*) below lOOO”C, and poor high-temperature creep resistance above 1000°C.

Investigations to minimize and eliminate the oxidation pesting behavior of MoSi2 at intermediate temperatures (~SOO’C) have focused on a number of approaches; however, none has eliminated the problem. Some of the investigations have included alloying MoSi2 with other materials (Al) to alter the characteristics of the oxide and the oxidation mechanisms; minimizing porosity and microcrack- ing, which have been shown to promote pesting in MoSi2 at intermediate temperatures; the use of secondary reinforcements, such as Si3N4, to promote the formation of stable oxides (Si20N2) at the intermediate temperatures where pesting occurs;

and using metallic coatings for protection at intermediate temperatures.

Composite approaches have been used to enhance the fracture toughness of MoSi2-based materials. Composite approaches have included the use of continuous, discontinuous, and laminate reinforcements of the MoSi2 matrix. Secondary reinforcements which have been investigated include Sic, Si3N4, ZrO2, TiB2, Nb, and Ta.

To improve the high-temperature creep resistance of MoSi,, investigators have focused on minimizing the glassy phase along the MoSi2 grain boundaries to prevent grain-boundary sliding. The use of sub- micron Sic and Si3N4 as secondary reinforcements inthe MoSi2 matrix has resulted in a substantial increase in the creep resistance of MoSi2 at temperatures up to 1600°C. Development of larger matrix grain sizes during processing has also been investigated.

MoSi2-based structural components and coatings are currently being investigated for a number of industrial applications. These include structural components and coatings for glass melting opera- tions, industrial gas burners and lances, furnace heating elements and igniters, and aerospace and industrial gas turbine components.

(9)

Ordered inter-metallic alloys: an assessment 587

Various methods have been used to fabricate components including powder processing and melt- based techniques. Melt processing techniques have been used to produce coatings and structural components. Current melt processing techniques which are being investigated include plasma spraying and spray-forming technologies and non-conventional melt processing using laser sintering of MoSi;!

powders.

4.2 A& silicides

Recent work has shown that A&-type materials (where A is usually a transition-metal element) doped with B, C, or 0 have superior creep and oxidation resistance which may make them suitable for many applications above 1300°C in hostile environments.

Ternary light element additions can significantly affect properties of binary A& materials. This is thought to be caused by the formation of interstitial bonding in the otherwise unchanged structure (Ti&- based materials) or by promoting formation of a multiphase system (Mo$&based materials).

These types of materials can be densified by the use of pressureless sintering. Compacts can be consolidated by sintering under argon at temperatures near 1800°C. For Mo&-based materials, the densified material contains about 2% porosity.

The compound Ti5Si3 is particularly attractive because of its high melting point of 213O”C, ade- quate oxidation resistance at moderate temperatures, and low density of 4.32 gcmM3. The increased oxidation resistance of Ti$i3 doped with either B, 0, or C seems to be correlated to the effect of ternary atoms on the bonding of titanium atoms within the Mn& structure. Single-crystal X-ray diffraction data indicate a contraction of the Ti-Ti bond lengths in response to the presence of an interstitial atom. Additionally, increases are observed in Ti-Si bond lengths. These effects may lead to the more rapid formation of a continuous silica layer on the material surface. Further research, including experimental and modeling efforts, is required to better understand these effects.

Mo& is also the most refractory compound in the Mo-Si binary diagram with a melting point of 2180°C. At 800°C undoped MO.& exhibits pest oxidation and is converted to loose powder after only 35 h. Above 9OO”C, the material does not exhibit pest behavior, but oxidation proceeds rapidly and MoSSi3 is completely converted to silicon dioxide in less than 1 h at 1200°C. The iso- thermal oxidation resistance of MogSi3 at 800°C

has been improved with the addition of less than

2wt% boron. The oxidation rate slows markedly after a mass loss due to volatilization of the initially formed molybdenum oxide, and the material does not exhibit pest behavior at 800°C. A thick, glassy, borosilicate scale forms on the boron-doped material compared to pure silica glass which forms on MO&.

However, it is not known whether the pesting behavior has been eliminated. Cyclic oxidation testing from ambient temperature to 800°C or static low-temperature oxidation tests have not been made.

Compressive creep tests have shown boron- doped MogSi3 maintains the good creep resistance of undoped MO&. The material tested consisted of a MogSi3 matrix material with Mo3Si and Mog(Si,B)3 as second phases. At 1240°C and at 140 to 180 MPa, the creep rate of MoSSi is only slightly lower than that of boron-doped MO&. The slightly faster creep rate of boron-doped MO&

may be due to its three-phase microstructure. The other phases may be able to deform plastically at high temperatures due to dislocation motion.

Additional effort is required to understand the various mechanisms affecting creep behavior.

4.3 Brittle fracture and alloy design of Ni$3i alloys Of all silicides (MO&, MO&, Ti&, Ni$i, FeSi, and FeS&), Ni3Si appears to be the only one that can be made ductile and fabricable. Binary Ni3Si shows brittle grain-boundary fracture and low tensile ductility at ambient and intermediate temperatures, and possesses a liquidus temperature of 1143°C.

Like NisAl, Ni3Si shows a yield anomaly at elevated temperatures. The material is resistant to acid corrosion, particularly in H2S04 solutions, and possesses excellent strength at temperatures up to 650°C.

Alloying additions and microstructural control substantially improve the ambient-and elevated- temperature ductility and strength. The brittle grain-boundary fracture at ambient temperatures appears to be caused by both poor grain-boundary cohesion and moisture-induced hydrogen embrittlement. Microalloying Ni3Si with boron has been found to be effective in alleviating environmental embrittlement when tested in air at ambient temperature, and Auger electron spectroscopy has shown that the boron segregates to the grain boundaries. Titanium is effective in the enhancement of grain-boundary cohesion in Ni3Si, and grain-boundary fracture can be completely eliminated by doping Ni3(Si,Ti) alloys with B.

Both alloy stoichiometry and alloying additions affect the grain-boundary chemistry and the mechanical properties of Ni3Si. Alloying with elements

(10)

588 C. T. Liu et al.

including Cr, MO, Fe, V, results in superior mechanical properties including yield strengths of greater than 1OOOMPa and ultimate tensile strength of greater than 1200MPa up to 600°C. Superplastic behavior has also been observed in many NisSi- based alloys. Strain to failure from 150% to nearly 600% has been observed. Alloying additions of Nb and Ti substantially improve the acid corrosion resistance of NisSi alloys.

4.4 Silicide coatings

Refractory metals provide attractive mechanical properties for use in very high-temperature applications including rocket thrusters used on satellites and the Space Shuttle. In addition, current development of the Cr-Cr2Nb alloy has led to renewed interest in Cr-base alloys for high-temperature applications. With melting temperatures between 2468 and 3410°C alloys of niobium, molybdenum, tantalum, and tungsten exhibit good strength well beyond the temperature limitations of Ni- and Co- base superalloys and can be used at temperatures typically reserved for ceramics.

to dissolve considerable amounts of ternary alloying additions provides over 900 combined binary and ternary Laves phases. Despite their abundance these alloys have found little industrial application because their precipitation in grain boundaries of steels and superalloys resulted in their embrittlement. Later work showed Laves phase precipitates (TaFeJ distributed in the matrix of ferritic steels yield remarkable wear-resistant properties and good elevated-temperature strengths without sacri- ficing the low-temperature ductility. Indeed, Laves phases have unique properties which make them attractive for high-temperature structural applications. At half their homologous melting temperature, they retain > 0.85% of their ambient yield strength, which is higher than all other intermetal- lies. Many of the Laves phases also have high melting temperatures, excellent creep properties, reasonably low densities, and, for alloys containing Cr, Al, Si or Be, good oxidation resistance. Despite these useful properties, the tendency for low-temperature brittleness has limited the potential application of this large class of alloys.

However, the use of these alloys has not been fully realized due to the poor oxidation resistance they exhibit at elevated temperatures. For example, niobium rapidly oxidizes at temperatures above 76O”C, molybdenum oxidizes to form gaseous Moos above 1 lOO”C, and tantalum oxidizes rapidly to form Ta205. Therefore, it is crucial to protect these alloys from oxidizing atmospheres at high temperatures.

Silicide coatings are by far the most common method used to protect the refractory alloys from environmental degradation. The main technological issue for silicide coatings in use today is the minimization of coating defects and their deleter- ious effects on the coating properties. Coating fail- ures seem to be nearly always related to defects that are inherently present due to the thermal expansion mismatch between the coating and the substrate. The full potential of refractory alloys does not seem probable unless the effects of coating defects are minimized and/or adherent coatings with no defects can be produced.

Recent results suggest that Laves phases have intrinsic features that might favor deformability at ambient temperatures. For example, monolithic Laves phases such as HfV2 and NbCrz have high Poisson ratios (O-39 and 0.34, respectively) and low shear modulus values (30 and 80GPa, respectively). The high Poisson ratios are an indication of a lack of strong directional bonding as compared to other intermetallics with Poisson ratios near 0.2 (MoSiz, NiAl, and TiAl). The low shear modulus values suggest a potential for resistance to brittle failure. For example, the Rice-Thomson criterion suggests that when Gb/y < 10, (where G = shear modulus, b = Burgess vector, and y= surface energy), there is a tendency for resistance to brittle failure. For both HfV2 and NbCr*, their Rice- Thomson value is estimated to be 5.0 (HfV2 x 3.7 and NbCr2 z 6.0).

4.5 Laves-phase intermetallic alloys

Although experience indicates that monolithic Laves phases are brittle, dual-phase alloys of a body-centered cubic (bee) phase and Cl5 phase demonstrate encouraging deformability. Multilayers of HfV, + Nb have been cold-rolled up to 30%

without fracture. In addition, dual-phase NbCrz/

Cr(Hf) indicates promising ambient fracture toughness, with values ranging from 7 to 15 MPam112.

The Laves phases represent an abundant class of To fully develop the potential of Laves phases, intermetallic alloys with possible high-temperature three main issues need to be addressed: (1) alloy structural application. Laves phases form at or design, (2) structural stability, and (3) property near the AB2 composition, and over 360 binary evaluations. Alloy design is required to fully Laves phases are known. The ability of these alloys recognize the enormous base of alloying potential

(11)

Ordered intermetallic alloys: an assessment 589 in Laves phases, and the role of defects, stacking

faults, and bond strengths needs to be better understood. Structural stability is governed by the electronic structure of the Laves phases, and a structural instability may give rise to the tendency for twinning at ambient temperatures. Different sample quality may have resulted in the spread of the literature values for the physical and mechanical properties of Laves phases, and there is a need for full characterization. Moreover, the deformation modes and mechanisms require better defini- tion to couple with the mechanical behavior of the material. In dual-phase bcc/Laves alloys, the deformability of Laves phases has been observed, and the role of the bee phase in the Laves phase deformation needs more attention.

4.6 Superplasticity and superplastic forming of intermetallics

Superplastic forming is a viable near-net-shape forming technology for hard-to-form or hard-to- shape materials. It is especially attractive for intermetallics because intermetallic alloys are difficult to machine, resulting from their high hardness and low toughness. Superplasticity has now been demonstrated in a number of intermetallics, including the L12 structure (e.g. Ni3Al and Ni$i), iron aluminide, titanium aluminide (TiAl), and tri- titanium aluminides (T&Al). Two approaches were normally used to produce fine-grained structures in intermetallics-to add fine particles or solutes to retard grain boundary migration and to produce a two-phase microstructure via appropriate thermomechanical routes. The dominant superplastic deformation mechanism in these fine-grained intermetallics is believed to be grain-boundary sliding accommodated by a slip or diffusional process.

The key to obtaining superplasticity is, in fact, to develop a thermomechanical route to produce a fine-grained structure. However, the thermomechanical process is often extensive and elaborate.

Thus, from a scientific point of view, basic understandings of the phase diagram for a multialloy system, microstructure-deformation relationship, thermomechanical routes to produce a fine-grained structure, microstructural evolution during superplastic deformation (e.g. dynamic recrystallization and grain growth), and the post-formed properties are all needed in order to optimize the superplastic forming process. From a technological point of view, on the other hand, there also exist many challenges. For example, large bulk and sheet materials with a uniform microstructure, a rela-

tively high forming temperature (-lOOO’C), a protective atmosphere to prevent excessive oxidation, thickness control in a formed part, and the development of a computer model to describe the actual forming process are all needed to warrant a successful forming operation.

5 CRITICAL ISSUES IN ADVANCED INTERMETALLIC ALLOYS

Attractive properties at elevated temperatures has been achieved in a few intermetallics; however, the ability to make optimal use of these materials requires a good deal more understanding of their basic properties and behavior. The scientific principles derived from these understandings will be used as guidelines for property improvement through alloy design efforts.

5.1 Material processing

Brittle intermetallics, such as molybdenum sili- tides, with high melting points (T, > 2000°C) are difficult to prepare and fabricate. Mechanical properties of these intermetallics are generally very sensitive to trace impurities, and in some cases, these impurities completely mask the intrinsic properties of the intermetallics. One of the promi- nent examples is the poor strength and creep resistance of MoSi;! at temperatures above 1200°C. The poor mechanical properties at these temperatures were considered to be intrinsic to MoS& until they were proved to be a result of the formation of silicate glass phases due to oxygen trapped in the silicide during material processing. In order to understand the intrinsic mechanical properties, it is vital to prepare high-temperature intermetallics with high purity. Both high-purity charge materials and clean processing environments are required to synthesize these intermetallics for property evaluation.

Fabrication of sound materials is another major concern for engineering use of these high-temperature intermetallics. The understanding of heir deformation behavior together with knowledge of their microstructure and phase relationships are the keys to successfully fabricating these materials at elevated temperatures. Many ordered interme- tallies generally show super-plastic behavior even with a relatively coarse grain structure (as compared with metals and ceramics). It appears that superplastic forming is a feasible way to fabricate high-temperature intermetallics into net shapes.

(12)

590 C. T. Liu et al.

Technical issues for superplastic forming include (a) developing microstructure/superplastic forming correlation; (b) identifying conditions (high strain rate, lower temperature, and atmosphere) for superplastic forming at low cost; (c) demonstrating superplastic forming of large pieces (thickness, microstructural uniformity, etc.); and (d) computer-modeling of superplastic behavior.

5.2 Alloy design

The present methods for alloy design are rather

‘Edisonian’ in approach. They are largely based on empirical rules founded on careful experimental observations and ‘good guesses’. Future progress will be greatly assisted by a close coupling between the theoretical understanding of alloy formation and properties and the experimental development and testing of alloys.

Many of the complex new alloys require greater attention to the characterization of crystal structures, site occupation in the lattice, and grain- boundary chemistry and structure, as well as the correlation of these parameters with the mechanical properties. As the intermetallic ‘phase space’

opens to investigation, additional phase diagram work needs to be carried out, particularly on multicomponent systems. Before this becomes feasible, some preselection of systems in relation to potential applications and the required properties of an eventual product needs to be made.

As the alloys used become more complex, the factors which control fracture toughness in complex systems must be better understood. The role of multiple deformation mechanisms, including slip, twinning, and localized creep, must be understood.

High-temperature phase stability must be deter- mined in many of the systems of interest. This requires knowledge and measurement of alloy diffusivities and the effects of stoichiometry on defect structures and diffusivities. Advances in understanding will be aided by the availability of well- characterized single crystals and polycrystalline materials.

5.2.1 Molybdenum silicides

Oxidation problems remain, although the formation of protective glassy phases appears to be promising. Protective coatings ‘beyond glassy phases’

need to be explored. In particular, if coatings are to be depended upon for systems used for their mechanical properties, the self-healing properties of these coatings need to be understood and controlled.

The factors which control the fracture toughness of the silicides are not at all well understood.

Additional understanding is required to guide attention toward grain-boundary structure and chemistry, dislocation properties, or macro-design (e.g. layered structures) in trying to improve the fracture toughness.

The high-temperature creep behavior and mechanisms of molybdenum silicides need to be better understood in order for alloy designers to address the improvement of these properties. At present, it is not known whether the most promising route is second-phase dispersion, grain-boundary control, or design for single crystals, The effects of environmental attack at high temperatures in hydrogen-, sulfur-, and water-vapor-containing atmospheres are not understood. The potential impact of

‘down-time corrosion’ on the high-temperature performance must be recognized.

The potential for alloying exists in the case of the Mo& compound, and alloying behavior with different elements needs to be explored both from fundamental calculations and alloying experiments. It is important to determine the effective- ness of alloying elements in MO& and their effect on the thermophysical and mechanical properties with temperature. The alloying approach can potentially be used to optimize high-temperature creep resistance, room-temperature elongation, and oxidative stability of the silicides.

Very little information exists on the wetting behavior of various silicides with a variety of metals, compounds, and interlayers. It is important to determine the basic and fundamental information such as wetting angles, depth of penetration, and the chemistry of intermediate layers, if any.

These data can serve as building blocks for determining how silicides can be joined with other intermetallic compounds such as nickel aluminides, Laves phases, superalloys and other ceramic materials. The high temperature stability of the silicides may require developing new alloys for interlayers to successfully make use of the full potential of the silicides. Interfacial strength, interfacial products, mechanical strength, and thermal stability of the joints should be clearly understood before they can find any applications.

Silicides can be used as coatings for a variety of applications, and coatings may be obtained by pack cementation or thermal spray techniques. In the case of pack cementation, fundamental information on the kinetics of the coating processes is not available. In the case of thermal spray techniques, the stable bond coats for the materials need

(13)

Ordered intermetallic alloys: an assessment 591 to be developed along with the coating process

parameters. Mechanical and thermal stability of the coatings needs to be addressed from a fundamental viewpoint, and the data must be compared with the monolithic properties.

5.2.2 Laves phases

The Laves phases hold great potential for development of new, high-temperature materials, due to the very large phase space in which they can exist.

Most of this phase space has not been studied. A theoretical understanding of what controls the properties of these phases is required to guide the researchers through this ‘wilderness’. Both the difficulty and the potential of these alloys are very great. Little is known about, and attention must be paid to, the effects of alloying on the nature of the atomic bonding and the properties of multiphase alloys containing Laves phases.

For the Laves intermetallics, as for the other advanced high-temperature intermetallic systems such as the molybdenum silicides, it is important to identify ‘show-stoppers’ in critical areas at early stages of alloy development. Poor fracture toughness (typically of 2 to 4 MPa&ii) appears to be the major concern for these brittle intermetallics, with environmentally induced degradation a close second. The suggestions for metallurgical ways to improve the fracture resistance at ambient and elevated temperatures include: enhancement of metallic bonding; increases of the activity of mobile dislocations; promotion of deformation twinning through the control of crystal structure; addition of ductile second-phase particles to bridge cracks;

refinement of grain structure and control of grain- boundary chemistry (i.e. elimination of glass phases in silicides); design of interfacial properties to blunt microcracks; the design of multiphase microstructures; and control of defects and impurities.

Many Laves phases may have interesting properties other than their mechanical properties, such as anomalous elastic properties, soft phonon modes, and hydrogen-storage capabilities, (among others); and these should be explored.

5.3 Other practical issues

There is a such large amount of phase space that the potential use of these materials should be the guide for fundamental studies. Some factors to be considered are: does the intermetallic offer the possibility of use at temperatures of the order of 15OO“C?; how widespread is the interest in using these alloys?; what are the concerns of potential

users and producers, and how can these be addressed?; are the economics of the alloy systems favorable?

6 THEORY

6.1 First-principles calculations and atomistic modeling

Over the past seven years or so, first-principles quantum mechanical calculations based on the local-density-functional (LDF) method were used to determine fundamental bulk and defect properties of many transition-metal aluminides. The cal- culated results provide not only relevant physical quantities related to micromechanisms for the mechanical behavior of known or as yet untested alloys, but also information on electronic structure from which a clear understanding of observed properties can be developed. Particularly important contributions have been made in the physical quantities that are not available or are impossible to measure, notably in the predictions of point defect structures, planar fault energies, and site preference of ternary alloying additions. Significant progress has also been made in understanding the effect of interstitial solutes (hydrogen, boron, and carbon) in the bonding characteristics and resulting mechanical behavior.

As these ab-initio first-principles calculations are applied to relatively more complex intermetallic systems, such as transition-metal silicides with the low symmetry crystal structures or Laves phase with a large number of atoms per unit cell, the degree of computational difficulty will increase.

This could be overcome by utilizing not only ever increasing computer speed and memory, but also improved computational methodologies.

Semi-empirical methods such as the embedded atom method (EAM) have been used extensively for the past 10 years to calculate detailed properties of intermetallic alloys. Most of these calculations have centered on the Ni-Al system. It has been recently recognized that angular forces may be an important factor in some intermetallic materials, (e.g. those containing silicon or iron). The EAM does not well represent materials that have strong angular forces. It is crucial to determine if EAM potentials are sufficiently accurate in materials with directional bonding to describe properties such as elastic constants, fault energies, dislocation core structure, etc. If not, alternative methods, e.g. the modified EAM (MEAM) and tight-binding (TB)