|
Items in cart [4] |
Questions? Call 888-624-8373 |
| Copyright © 2009. National Academy of Sciences. All rights reserved. Terms of Use and Privacy Statement |
Below are the first 10 and last 10 pages of uncorrected machine-read text (when available) of this chapter, followed by the top 30 algorithmically extracted key phrases from the chapter as a whole.
Intended to provide our own search engines and external engines with highly rich, chapter-representative searchable text on the opening pages of each chapter.
Because it is UNCORRECTED material, please consider the following text as a useful but insufficient proxy for the authoritative book pages.
Do not use for reproduction, copying, pasting, or reading; exclusively for search engines.
OCR for page 50
5
Materials Science Research Program
INTRODUCTION
Materials science provides the technical foundation for many of the nation's most vital industries:
Semiconductors, optoelectronics, ceramics, and steel making are a just a few examples. Because the
structure of a material determines its properties, the ability to produce a specific structure at will, and
hence tailor a material's properties for a particular application, has been a continuing goal for materials
scientists and engineers. However, understanding the fundamental relationship between process his-
tory, structure, and properties is complicated by the fact that material processing often involves the
liquid or vapor state, so that buoyancy-driven convection, surface tension effects, sedimentation, and
container reactions can mask the critical effects under study.
Materials scientists were among the first to recognize that the unique aspect of space experimenta-
tion the absence of buoyant fluid effects and sedimentation could provide insights that would be
difficult, or impossible, to gain by any other approach. Thus, the materials science program has been in
existence longer than any of the other microgravity disciplines. The objective of this space-based, or
microgravity, experimentation was to gain a deeper understanding of materials phenomena that could
then be used to improve Earth-based science and technology.
Early microgravity materials experiments were rudimentary and discovery-driven. In one of the
first experiments, tiny lead-tin solder ingots solidified in a teacup-size furnace during an Apollo mis-
sion, were examined for microstructural improvements expected from the low-gravity environment.
The Skylab and Apollo-Soyuz test programs of the 1970s successfully demonstrated the processing of
semiconductor, metal alloy, and halide crystals by solidification and vapor techniques. These platforms
used a specially designed multipurpose furnace to achieve a variety of experimental conditions that took
advantage of microgravity conditions to process different materials. Among the results of these early
studies was a demonstration that the low-convection conditions of space eliminated the compositional
variations due to oscillatory convection in semiconductor crystals these variations degraded the elec-
tronic properties of Earth-grown material. (Later, magnetic suppression of convection became standard
procedure in the silicon industry.)
50
OCR for page 51
MATERIALS SCIENCE RESEARCH PROGRAM
5
The space shuttle provided access to space for performing microgravity material science experi-
ments after Skylab. During the period of shuttle-based experimentation, NASA developed a cadre of
ground-based and flight investigators and improved flight hardware; it also developed new analytical
tools, computational modeling techniques, and advanced materials theories. Materials science experi-
ments in space evolved to become driven more by hypothesis and theory and less by discovery-oriented
goals. As is noted below, many of the successful experiments were directed at producing benchmark
data sets. Within materials science, the present microgravity research program, described below, has
strongly influenced scientific understanding in several technologically important areas, for example:
· Control of impurity segregation, interface stability, and dendrite formation in metals and semi-
conductors;
· Control of coarsening, liquid-phase sintering, and grain structure in industrial alloys; and
· Measurement of accurate thermophysical properties, such as surface tension, viscosity, and diffu-
sion, that are required for effective computational modeling.
In the early 1990s NASA broadened its program significantly beyond just those experiments that
were destined to be flown. The ground-based program now contains a number of experimental and
theoretical investigations. The purpose of the Round-based experimental program is to provide the
opportunity to establish conclusively the need for periormmg an experiment In mlcrogravlty and to
develop the protocols necessary to perform the experiment prior to the development of spaceflight
hardware. In addition, theoretical programs directed at modeling the processes under investigation are
funded through the ground-based program.
The current microgravity materials science program has a dual focus. Projects are conducted in
each major materials system, but they are organized by research themes that are common to all materials
systems. These research themes are fundamental in nature, but many also impact a range of practical
problems. The materials systems being investigated currently are electronic and photonic materials,
glasses and ceramics, metals and alloys, polymers, and more recently, biological materials. Common to
these materials systems are the phenomena that form the key microgravity research themes: (1) nucle-
ation and metastable states, (2) prediction and control of microstructures, (3) interracial and phase-
separation phenomena, (4) transport phenomena, and (5) crystal growth and defect control. Addition-
ally, work IS under way to improve the radiation resistance of shielding materials for use in the human
exploration of space.
.. . . .
~ . . _
IMPACT OF NASA'S MATERIALS RESEARCH
The materials science program currently funds research across a broad spectrum of areas, ranging
from crystal growth to materials for radiation shielding. Of the approximately 115 investigators who
have been in the program over the past 2 years, 9 are members of the National Academies of Engineer-
ing and Sciences, 5 are fellows of the Mining, Metals and Materials Society (TMS), a large materials
society with 100 living fellows and 10,000 members, 11 are fellows of ASM International, another
major materials society. and 17 are fellows of the American Physical Societv. These are large numbers
~ - , ~ ~ ~
. .. . - . ,% ,% .. ~ . ~ . . · . . . .. . ... ~ ..
given the comparatively narrow focus ot the NAbA materials program compared to the breadth ot the
interests represented by each of these materials societies. NASA has funded outstanding materials
science investigators during the formative period of their careers, two of whom have been subsequently
elected to the National Academy of Engineering. A major focus of the materials science program is
solidification and crystal growth. The Bruce Chalmers Award, sponsored by the TMS, is given to
OCR for page 52
52
ASSESSMENT OF DIRECTIONS IN MICROGRAVITY AND PHYSICAL SCIENCES RESEARCH AT NASA
individuals who have made outstanding contributions to the science and technology of solidification
processing. There have been only 12 winners of the award to date. Of the nine Americans who have
received that award, eight have been, or are, in the NASA materials program. It is clear that the program
has been successful in creating a community of high-quality investigators.
Discussed below is the impact of the materials science program in the five research-theme areas
mentioned previously. Not addressed are the new areas, since the research program in these areas is just
beginning. Research was judged to have had a significant impact on the field if the paper received 50 or
more citations or, for more recent publications, if it is being cited at a high rate. Other metrics were also
used to assess impact; these are mentioned under the particular program. All of these papers were
published as a result of NASA research funding.
Nucleation and Metastable States
Probing the thermophysical properties of liquids cooled significantly below their melting points, or
deeply undercooled, poses a great challenge for those systems in which nucleation of the stable crystal
occurs. By melting samples that are levitated it is possible in some cases to forestall nucleation of the
crystal as the liquid cools, thus allowing the properties of the undercooled liquid to be measured. A
recent example of this approach is provided by the study of alloys that form metallic glasses under slow
cooling rates. Usually metals crystallize upon cooling significantly below their melting points. In
contrast, in these "bulk metallic glasses" it is Possible to form a metallic class under very slow cooling
1 0 ~ O
rates. These materials have garnered much attention recently because they display unusual properties
such as strength (they are two to three times stronger than their crystalline counterparts) and the ability
to withstand very high strains and still remain linearly elastic, while enabling standard casting processes
to be employed to produce objects with large volumes. The thermodynamic properties of these novel
bulk metallic glasses were investigated using electrostatic levitation (Kim et al., 1994; Busch et al.,
1995). In these studies, the heat capacity of the glass, crystal as well as the levitated undercooled liquid,
was measured. By comparing these measurements with those of more standard metallic glass alloys it
was possible identify certain thermodynamic properties that favor bulk metallic glass formation, thus
pointing the way to the development of other bulk metallic glass alloys. Flemings and coworkers have
investigated phase selection within levitated liquid steel droplets for many years. It was found that
through deep undercooling of the liquid below the liquidus it is possible to nucleate phases that are not
normally found during near-equilibrium solidification, thus allowing a careful investigation of the
thermodynamics and kinetics of metastable phase formation. Similar phases are found during strip
casting of commercial steel alloys. This work thus has the potential to lead to better control of the
processing parameters during strip casting and to improvement in the quality of the steel (DiMicco,
2000).
Prediction and Control of Microstructure
As a liquid is cooled below the melting point the solid phase forms and grows into the liquid phase.
The growing solid displays a rich variety of morphologies, ranging from beelike objects called dendrites
to undulating solid-liquid interfaces, termed cells. This process is of intense interest from both scientific
and commercial viewpoints. Of scientific interest is the fact that these processes involve the formation
of patterns governed by highly nonlinear, but simple, equations. Understanding the factors that control
these patterns has been of great interest in the physics and materials communities. More practically,
OCR for page 53
MATERIALS SCIENCE RESEARCH PROGRAM
53
these patterns control the distribution of chemical elements in the solid and thus the resulting properties
of a casting. Since the solidification process occurs from a liquid phase, buoyancy-generated convec-
tion of the liquid can strongly perturb these processes, and thus the focus of the work funded by NASA
has been to understand the role of convection in these processes and to study the processes in the
absence of convection, which is usually not possible on Earth even with the application of magnetic
fields.
An analysis of the citations clearly indicates that there has been a long history of important theoreti-
cal work funded by NASA in this area. The effects of convection driven by composition and thermal
gradients on the morphological stability of a solid-liquid interface during solidification were first inves-
tigated by Coriell et al. (1980~. They showed that both morphological and convective instabilities can
occur and that the ability of the solid-liquid interface to deform leads to conditions for the onset of
instability that can be quite different from those in the absence of such a deformable boundary. Ungar
and Brown (1984, 1985) examined the formation of deep cells during directional solidification. They
did this through a direct numerical solution of the defining equations. They showed clearly the rich
bifurcation structure of these nonlinear systems and the conditions for cellular interfaces to form drops
of liquid at their roots like those observed experimentally. More recently, the numerical modeling of
solidification processes has been revolutionized by the introduction of the phase-field method. The
strength of this approach is that it is not necessary to track explicitly the location of the solid-liquid
interface during the solidification process while still accounting for the thermodynamics of the interface.
The phase field model thus enables numerical simulation of the myriad topologically complex solidifi-
cation morphologies that are sensitive functions of the interracial thermodynamics. NASA researchers
have been at the forefront of the development of this method by providing phase field models for
systems with realistic anisotropic interracial energies (McFadden et al., 1993), thermodynamic con-
straints on phase field models (Wang et al., 1993), and the first phase field model for the isothermal
solidification of binary alloys (Wheeler et al., 1992~.
Treelike structures known as dendrites are ubiquitous. They are found in processes ranging from
crystallization from the vapor in this context they are called snowflakes to solidification of a liquid
metal. An important advance (this paper received over 370 citations) in the study of dendritic solidifi-
cation was made by Huang and Glicksman in their work on dendritic growth in succinonitrile (Huang
and Glicksman, 1981~. They showed that dendrites grow into a liquid cooled below its melting point
with a velocity and a tip radius that are unique for a given liquid temperature. In addition, they
documented clearly the influence of the buoyancy-driven convection of the liquid that accompanies
dendritic growth on the dendrite morphology and growth rate. This work stimulated significant theo-
retical studies in the physics and mathematics communities in an attempt to understand why dendrites
select the unique growth velocity and tip radius that was measured experimentally. The results of the
experiments have been included in an undergraduate textbook on solidification (Kurz and Fisher, 1989)
and in more advanced monographs (Davis, 2001~. To remove buoyancy-driven convection and produce
a data set that can be compared directly to theory, the experiments were flown aboard USMP-2 in 1994
and two subsequent missions (Glicksman et al., 1994~. The experiments showed that the data on Earth
are compromised by convection, and that once these convective effects are removed there are significant
long-range thermal interactions between both neighboring dendrites and secondary dendrite arms that
form back from the dendrite tip. Analysis of the data is continuing not only by the PI but by many others
in the materials, physics, and mathematics communities, as evidenced by the numerous downloads of
the spaceflight data. The experiments have produced a set of benchmark data that will serve as a test of
existing and future theories of dendritic growth.
OCR for page 54
54
ASSESSMENT OF DIRECTIONS IN MICROGRAVITY AND PHYSICAL SCIENCES RESEARCH AT NASA
Interfacial Effects and Phase Separation
Liquid-phase sintering is a widely employed industrial process that can be used to produce an array
of advanced materials, from cutting tools to automotive turbochargers. The process begins with pow-
ders that are compacted into a desired shape and then heated to produce a two-phase, liquid-solid
mixture. The liquid provides a pathway through which vapor pores can escape. The goal is to reduce
the shape changes of the compacted powder that occur during sintering so that little machining is
required following processing. Experiments performed in space by German show, surprisingly, that the
shape distortion of samples processed in microgravity is considerably greater than that of terrestrially
processed samples. This finding, coupled with other insights gained from ongoing research at
Kennametal, Inc., enabled a better understanding of the underlying causes of the changes in shape of
powder compacts during liquid-phase sintering. Kennametal, Inc., is a market leader in the metal-
cutting tool industry, with annual sales revenues of $1.8 billion. Approximately 40 percent of the
production cost of its cutting tools is associated with the post-sinter machining necessary to bring the
dimensions of the tool into specification. Using this insight, it was possible to nearly eliminate the final,
expensive grinding step in the fabrication of parts (Latrobe et al., 1988~.
If a solid-liquid mixture is held isothermally, the solid particles will evolve in time to decrease the
total solid-liquid interracial area, and so the average size of the solid particles will increase in time. This
process, called Ostwald ripening, is observed in nearly all two-phase mixtures, from raindrops in clouds
near the equator to solids composed of mixtures of different crystals. As a result, the ripening process
influences the properties of many commercially important materials, among them a wide variety of
aluminum alloys and high-temperature materials used in jet turbines. NASA-funded work has provided
many new insights into this important phase transformation process. Akaiwa and Voorhees predicted
theoretically the interparticle spatial correlations and the kinetics of ripening in systems with the high-
volume fractions of ripening phase that are typically employed experimentally (Akaiwa and Voorhees,
1994~. In agreement with theory, terrestrial experiments showed that the ripening process in model
solid-liquid mixtures was a function of the volume fraction; however, the mixtures were found to ripen
faster than predicted by theory (Latrobe et al., 1988~. To avoid the sedimentation of the solid particles
that accompanies terrestrial experiments, ripening experiments were performed in space. As a result of
the ideal conditions afforded by the microgravity environment, it was possible to show that the classical
theory for ripening, first postulated in the 1960s, did not describe the experimental results (Alkemper et
al., 1999~. The experiments have produced a set of benchmark data that has been used by other
investigators in their research and teaching. The finding that systems undergoing Ostwald ripening may
not be described by the classical theory influenced the development of software that is used to simulate
the nucleation, growth, and ripening process in commercial alloys. In particular, the Precipicalc soft-
ware that has been developed by the materials design company Questek was designed to take advantage
of the spaceflight results. General Electric, and Pratt and Whitney, two jet engine manufacturers, are
currently using this new software.
Measurement of Thermophysical Data
With the advent of fast computers and powerful algorithms, it is now possible to simulate many
important processes that are used to produce materials. However, for such models to be truly predictive,
it is necessary to employ accurate values for the thermophysical properties for the material being
modeled. Many of these thermophysical parameters, such as diffusion coefficients of elements in high-
temperature semiconductor and metallic liquids, are extremely difficult to measure, and when measured
-0
OCR for page 55
MATERIALS SCIENCE RESEARCH PROGRAM
55
on Earth their values are frequently perturbed by buoyancy-driven convection. Thus, spaceflight experi-
ments are required to produce benchmark sets of thermophysical data. These experiments, however,
will involve only a limited number of systems due to the constraints imposed by space experimentation.
While these experiments have yet to be performed in space, the importance of performing this work is
well recognized by industry, as evidenced by letters of support for this research from the staff of the
Brimrose Corporation (Murphy, 2001) and the chairman and chief executive officer of the II-VI Corpo-
ration (Johnson, 2001~. Both companies are involved in the production of semiconductor materials for
sensor and electro-optic applications. In addition, certain II-VI materials have been and will be used in
NASA space-based telescopes, indicating that this work may have a direct impact on NASA if allowed
to proceed.
Crystal Growth and Defect Control
Nearly all materials used for structural applications are composed of more than one chemical
component. As a result, a mushy zone, a partially solidified region of significant spatial extent, forms
during solidification. Convection within the mushy zone can occur due to the variation of the liquid
density with temperature and composition. This leads to the formation of vertical plumes of liquid, or
freckles, within the mushy zone. The resulting strong microstructural inhomogeneity in the solidified
material has a considerable adverse impact on the properties of many cast materials. Major insights into
the physics underlying freckle formation were made by Hellawell and coworkers early in the NASA
materials science program (Sample and Hellawell, 1984; Sarazin and Hellawell, 1988~. They showed
by direct observation that convection was responsible for the formation of freckles and that rotating the
sample can damp the formation of freckles. Porier and coworkers were among the first to propose
physics-based models of convection and solute segregation within mushy zones (Felicelli et al.. 1991:
Ganesan and Poirier, 1990~.
", ",
Since crystals are grown from a melt or solution, buoyancy-driven convection of the fluid can play
an important role in the resulting distribution of chemical components and defects in the crystal. A
nonuniform distribution of chemical components can be quite deleterious to certain electronic and
mechanical properties. Before the infusion of NASA funding, understanding of the effects of convec-
tion on crystal growth was limited to a few simple models of the convection process, many of them ad
hoc. While providing some insight into the role of convection on the crystal growth process, these
theories did not address the coupling between fluid flow and mass transfer in a self-consistent manner.
A solution to the Navier-Stokes equations during crystal growth is required wherein the solid-liquid
interface is allowed to deform in concert with the diffusive and convective processes in the liquid.
Coriell et al. were the first to employ such an approach to examine the effects of solutal and thermal
convection on the stability of a solid-liquid interface during directional solidification (Coriell et al.,
1980~. Chang and Brown (1983) performed numerical simulations that allowed for large deformations
of the solid-liquid interface and both thermal and solutal convection in the liquid. They examined the
dopant distribution in the resulting crystal and correlated the distribution with the levels of convection,
the shape of the solid-liquid interface, and the design of the Bridgman furnace. They found that the
segregation can be as large as 60 percent of the mean concentration and that the design of the furnace
can change the degree of segregation. Further work by Adornato and Brown found that solute segrega-
tion was minimized for near-diffusion controlled growth or for situations where there is intense laminar
mixing (Adornato and Brown, 1987~. The implication of this result is that intermediate levels of
convection, such as those found at certain levels of reduced gravity or in weak magnetic fields, can
actually produce larger segregation and thus more undesirable material. In addition to these results, the
OCR for page 56
56
ASSESSMENT OF DIRECTIONS IN MICROGRAVITY AND PHYSICAL SCIENCES RESEARCH AT NASA
support provided by NASA had an impact on the theory and understanding of Czochralski growth
(Derby et al., 1985; Derby and Brown, 1986) and float zone growth (Coriell et al., 1977~. NASA-funded
research aided in the development of a community of scientists working on the mathematical modeling
of crystal growth and solidification.
FUTURE DIRECTIONS IN MATERIALS RESEARCH
Materials science has played a central role in many of the discoveries that have shaped our world,
from integrated circuits to low-loss optical fibers and high-performance composite materials. The
recommended research areas for NASA's materials science program, given below, will continue this
tradition of science-driven discoveries of great importance to the nation and to NASA. Many of these
promising areas build on research currently in the program, while others are new and unrelated to the
existing program. Furthermore, the committee expects that many other new and exciting areas will
emerge from the nation's research community.
Nucleation Within and Properties of Undercooled Liquids
The properties of many materials are strongly linked to the phases that nucleate during crystalliza-
tion. Hence, to tailor the properties of a material for a given application conditions must be chosen such
that only desirable phases nucleate during the processing of the material. Thus the nucleation process
plays a prominent role in setting materials properties. Unfortunately, the conditions governing the
nucleation of new phases are not well understood. Experiments in microgravity can play an important
role in providing a basic understanding of the nucleation process, since they eliminate a potent site for
nucleation the walls of a container while still permitting large volumes of liquid to be employed in
the experiment. In systems where nucleation on foreign particles in the bulk is not a problem, it is
possible to study carefully the nucleation process in bulk samples using a wide range of sample vol-
umes. An example is the formation of a glass from a liquid. Glasses typically form when a liquid is
cooled sufficiently fast to prevent the atoms from arranging themselves in a crystalline lattice. A new
class of metallic alloys has been discovered that can form glasses under very slow cooling rates, making
it easy to cool significant volumes of the liquid alloys far below their melting points. The ability of the
bulk metallic glass alloys to undercool allows the properties of the undercooled liquid state to be
examined. For example, recent studies appear to indicate that the classical theory for crystal nucleation
from a liquid is inapplicable in these materials and that the liquid alloys exhibit atomic transport and
theological characteristics that are very different from standard liquid metallic alloys. These materials
also exhibit novel properties: They are twice as hard as stainless steel with similar toughness. For
example, bulk metallic glass composites have recently been found to have significant ductility in
tension, opening a wide range of applications, from aircraft materials to medical implants. Fundamental
studies of the undercooled liquid state of these bulk metallic glasses can be performed in microgravity
since it is possible to undercool large volumes of liquid using containerless methods and thus avoid
heterogeneous nucleation due to the container. For example, if the properties of the melt that lead to
bulk metallic glass formation are better understood, this might allow aluminum-based bulk metallic
glass alloys to be formulated. Such bulk metallic glasses could have a huge impact on both industry and
NASA since they would possess high specific strength and tensile ductility and thus would be ideal for
a number of aerospace applications. Another example of nucleation from a liquid is quasicrystal
formation. Quasicrystals are crystals with an unusual fivefold crystallographic symmetry. Here the
issue is whether there are structural precursors to the quasicrystal in the undercooled liquid, a liquid that
OCR for page 57
MATERIALS SCIENCE RESEARCH PROGRAM
57
is cooled below its melting point. Establishing such a link between the structure of the liquid and
quasicrystal formation will shed new light on why certain alloys form quasicrystals and others do not.
Dynamics of Microstructural Development During Solidification
The microstructures of solidifying materials provide beautiful examples of spontaneous pattern
formation. The development of dendritic and cellular microstructures is governed by relatively simple
partial differential equations whose solutions are complex, possibly chaotic, and extremely sensitive to
small perturbations. Moreover, since the majority of metallic materials are cast, the solidification
process is of enormous industrial importance. The ability to directly link processing conditions to the
resulting materials properties is still not at hand as the mechanisms governing the development of
dendrite and cell morphology are not well understood. Outstanding questions include the effects of
interactions between individual dendrites or cells on their spatial distribution and morphology, the
evolution of dendrite morphology during transient heating or cooling, and the effects of noise and initial
conditions on the resulting patterns. The interactions between dendrites are particularly important in the
development of the mushy zone. A National Research Council study on the future of condensed matter
and materials physics identified the mushy zone as "perhaps the most important theoretical challenge"
in metallurgical pattern formation and also chose the study of the mushy zone as one of the research
priorities in nonequilibrium physics (NRC, 1999~. A major impediment to the study of these solidifica-
tion processes, however, is convection of the liquid phase, since convection makes it nearly impossible
to compare results with theoretical predications and greatly complicates the interpretation of experimen-
tal data. Performing experiments in a microgravity environment where convection is much reduced is
thus crucial to understanding these complex pattern-forming systems that are of great commercial
importance.
Morphological Evolution of Multiphase Systems
Materials used commercially are usually composed of more than one phase for example, the
strength of a jet turbine blade is linked to the size, shape, and spatial distribution of the precipitates that
are embedded in the matrix of the blade. These multiphase systems are created by a nucleation, growth,
and Ostwald ripening process through which a single phase decomposes into two or more phases.
Examples of such phase transformation processes abound. They occur in systems as diverse as poly-
mers, wherein a second phase of different composition can form by either spinodal decomposition or
nucleation with a liquid matrix, and in metallic alloys, wherein the morphology of dendrites in mushy
zones evolves in time by Ostwald ripening. In other cases, a two-phase mixture is created by physically
mixing two phases, such as the solid-liquid mixture found during liquid-phase sintering. During thermal
processing, the morphology of the mixture evolves and the volume fraction of vapor bubbles decreases.
Despite the clear commercial relevance and scientific importance, an understanding of the dynamics of
phase transformation processes is not at hand. Phase separation and the processing of materials where
one of the phases is liquid inevitably leads to sedimentation due to the density difference between the
component phases. Performing experiments in a microgravity environment greatly reduces the rate of
sedimentation and allows the dynamics of the transformation process to be investigated carefully. As a
result, microgravity experiments can provide new and important insights into the dynamics of the
evolution of multiphase materials. Research into Ostwald ripening and liquid-phase sintering has
affected industrial practice in the past, and the committee expects that there is a good probability that it
will in the future.
OCR for page 58
58
ASSESSMENT OF DIRECTIONS IN MICROGRAVITY AND PHYSICAL SCIENCES RESEARCH AT NASA
Computational Materials Science
Materials are typically designed using an extremely tedious and expensive trial-and-error approach.
However, with the advent of high-speed computers and modern algorithms, it is possible to simulate the
behavior of materials using a computer. Since the properties of materials are a function of processes that
occur over an enormous range of length scales, from the nanometer to the meter scale, a wide variety of
methods is required. On the smallest scales, quantum mechanical methods are used to simulate proper-
ties that depend on the electronic structure of a material. In the tens of nanometers to micrometer scales,
mesoscopic methods (such as the phase field method described earlier) are employed to describe the
evolution of the morphology of the constituent phases. Finally, the methods of computational mechan-
ics are used on the largest scales. As a result, numerical simulation is being viewed as an equal partner
with experiments in determining the properties of materials. A recent example of the power of this
approach is the work of Johannesson et al., wherein density functional theory was used to find the most
stable phases in alloys with four chemical components that can be constructed using 32 different metals
or 192,016 possible alloys (Johannesson et al., 2002~.
On the mesoscale, three-dimensional calculations of morphological development in alloys, while
still challenging, are becoming commonplace. The practical ramifications of such an approach are
profound; it is now possible to design a material using simulations to yield a desired set of properties.
No longer are parametric models required, since the calculations are predictive, efficient, and accurate.
This will yield a new paradigm for designing industrially relevant materials, since the materials will be
created with a minimum of costly, time-consuming experiments. This approach can have a significant
impact on NASA because it ensures that the materials properties of interest to NASA will be attained, in
much less time and at a lower cost. Such computational efforts are a natural outgrowth of the strong
effort in process modeling research currently supported by the program. Computational materials
science promises to yield a revolutionary new approach to designing new materials processes as well as
tailoring materials properties for a given application. Furthermore, the integration of process modeling
and diagnostics promises to one day be the key to rapid implementation of new materials-processing
technologies. These approaches are just beginning to be embraced by industry.
Thermophysical Data of the Liquid State in Microgravity
Computational modeling of materials processing requires accurate thermophysical data of the liquid
state. Obtaining such data on the ground that are not affected by convection is very difficult in low-
melting-point systems and nearly impossible for high-melting-point materials. For example, there are
very few accurate measurements of the solute diffusivities in liquid metallic or semiconductor alloys as
a function of temperature. Performing such experiments in microgravity will provide insights into the
physics of the liquid diffusion process as well as much needed thermophysical data for industry.
Accurate thermophysical data along with computational models will yield realistic predictions of quan-
tities such as the degree of microsegregation following solidification. The magnitude of the
microsegregation in turn can have a significant deleterious effect on a wide array of materials properties
Nanomaterials and Biomimetic Materials
There are many new avenues for materials research at the nanoscale and at the interface between the
biological and materials sciences. These new directions are discussed in Chapter 7.
OCR for page 59
MATERIALS SCIENCE RESEARCH PROGRAM
59
REFERENCES
Adornato, P.M., and Brown, R.A. 1987. Convection and segregation in directional solidification of dilute and non-dilute
binary-alloys: Effects of ampoule and furnace design. J. Cryst. Growth 80: 155-190.
Akaiwa, N., and Voorhees, P.W. 1994. Late-stage phase-separation: Dynamics, spatial correlations, and structure functions.
Phys. Rev. E 49: 3860-3880.
Alkemper, J., Snyder, V.A., Akaiwa, N., and Voorhees, P.W. 1999. The dynamics of late-stage phase separation: A test of
theory. Phys. Rev. Lett. 82: 2725.
Busch, R., Kim, Y.J., and Johnson, W. 1995. Thermodynamics and kinetics of the undercoated liquid and the glass-transition
of the Zr41.2 Til3.g CU12.5 Nilo.o Be22.s alloy. J. Appl. Phys. 77 4039-4043.
Chang, C.J., and Brown, R.A. 1983. Radial segregation induced by natural-convection and melt solid interface shape in
vertical Bridgman growth. J. Cryst. Growth 63: 343-364.
Coriell, S.R., Cordes, M.R., Boettinger, W.J., and Sekerka, R.F. 1980. Convective and interracial instabilities during unidirec-
tional solidification of a binary alloy. J. Cryst. Growth 49: 13-28.
Coriell, S.R., Hardy, S.C., and Cordes, M.R. 1977. Stability of liquid zones. J. ColloidInterface Sci. 60: 126-136.
Davis, S.H. 2001. Theory of Solidification. Cambridge Monographs on Mechanics. Cambridge University Press, Cambridge,
U.K.
Derby, J.J., and Brown, R.A. 1986. Thermal-capillary analysis of Czochralski and liquid encapsulated Czochralski crystal-
growth. 1. Simulation. J. Cryst. Growth 74: 605-624.
Derby, J.J., Brown, R.A., Geyling, F.T., Jordan, A.S., and Nikolakopoulou, G.A. 1985. Finite-element analysis of a thermal-
capillary model for liquid encapsulated Czochralski growth. J. Electrochem. Soc. 132: 470-482.
DiMicco, Daniel R., President and Chief Executive Officer, Nucor Corporation. 2000. Letter to M. Wargo, Enterprise Scien-
tist, NASA, dated October 30.
Felicelli, S.D., Heinrich, J.C., and Poirier, D.R. 1991. Simulation of freckles during vertical solidification of binary-alloys.
Metall. Trans. B 22~6~: 847-859.
Ganesan, S., and Poirier, D.R. 1990. Conservation of mass and momentum for the flow of interdendritic liquid during
solidification. Metall. Trans. 21: 173-181.
Glicksman, M.E., Koss, M.B., and Winsa, E.A. 1994. Dendritic growth velocities in microgravity. Phys. Rev. Lett. 73: 573-
576.
Huang, S.C., and Glicksman, M.E. 1981. Fundamentals of dendritic solidification: 1. Steady-state tip growth. Acta Metallurgica
29: 701-715.
Johannesson, G.H., Bligaard, T., Rub an, A.V., Skriver, H.L., Jacobsen, K.W., and N0rskov, J.K. 2002. Combined electronic
structure and evolutionary search approach to materials design. Phys. Rev. Lett. 88: 255506.
Johnson, Carl J., Chairman and Chief Executive Officer, II-VI Incorporated. 2001. Letter to the Honorable Barbara A.
Mikulski, dated August 29.
Kim, Y.J., Busch, R., Johnson, W., Rulison, A.J., and Rhim, W.K. 1994. Metallic-glass formation in highly undercooled
Zr41 2 Til3 ~ Cul2 5 Nilo 0 Be22 5 during containerless electrostatic levitation processing. Appl. Phys. Lett. 65: 2136-
2138.
Kurz, W., and Fisher, D.J. 1989. Fundamentals of Solidification. Trans Tech Publications, Switzerland.
Latrobe, P.A., Hardy, S.C., and Voorhees, P.W. 1988. Ostwald ripening in a system with a high volume fraction of coarsening
phase. Metall. Trans. A 19~11~: 2713-2721.
McFadden, G.B., Wheeler, A.A., Braun, R.J., Coriell, S.R., and Sekerka, R.F. 1993. Phase-field models for anisotropic
interfaces. Phys. Rev. E 48: 2016-2024.
Murphy, Diane C., Brimrose Corporation of America. 2001. Letter to the Honorable Barbara A. Mikulski, dated September 6.
National Research Council (NRC). 1999. Condensed-Matter and Materials Physics: Basic Research for Tomorrow's Technol-
ogy. National Academy Press, Washington, D.C.
Sample, A.K., and Hellawell, A. 1984. The mechanisms of formation and prevention of channel segregation during alloy
solidification. Metall. Trans. A 15: 2163-2173.
Sarazin, J.R., and Hellawell, A. 1988. Channel formation in Pb-Sn, Pb-Sb, and Pb-Sn-Sb alloy ingots and comparison with the
system NH4C1-H2O. Metall. Trans. A 19: 1861-1871.
Ungar, L.H., and Brown, R.A. 1984. Cellular interface morphologies in directional solidification: The one-sided model. Phys.
Rev. B 29~3~: 1367-1380.
OCR for page 60
60
ASSESSMENT OF DIRECTIONS IN MICROGRAVITY AND PHYSICAL SCIENCES RESEARCH AT NASA
Ungar, L.H., and Brown, R.A. 1985. Cellular interface morphologies in directional solidification: The formation of deep cells.
Phys. Rev. B 31: 5931-5940.
Wang, S.L., Sekerka, R.F., Wheeler, A.A., Murray, B.T., Coriell, S.R., Braun, R.J., and McFadden, G.B. 1993. Thermody-
namically-consistent phase-field models for solidification. Physica D 69: 189-200.
Wheeler, A.A., Boettinger, W.J., and McFadden, G.B. 1992. Phase-field model for isothermal phase-transitions in binary-
alloys. Phys. Rev. A 45: 7424-7439.
Representative terms from entire chapter:
crystal growth
