on the cover · this report describes the major technical activities and accomplishments in the...

On the Cover:Fig. 1 – (on left, on back cover) Ferromagnetic resonance measurements of aPermalloy film on a specially roughened copper substrate as a function of applied fielddirection. These measurements are part of a study to determine the effects ofroughness on magnetization dynamics.

Fig. 2 – (b&w image) An Al91Fe7Si2 alloy after electron-beam surface melting withscan velocity of 50 cm/sec (transmission electron micrograph). This observation of aglassy phase which has characteristics of a first-order transformation (as do normalcrystalline phases) holds promise for producing stable metallic glass phases.

Fig. 3 – FEM model of a rigid 100 nm diameter sphere indenting an Al sample to adepth of 10 nm. The FEM mesh is fine enough to use the predicted elasticdisplacement fields to generate boundary conditions and initial atom positions for anatomistic simulation using classical potentials.

Fig. 4 – (near edge of front cover) MgO cubes with surfaces decorated by small goldparticles. These particle ensembles are used in model studies of 3D chemical imagingat the nanoscale (electron tomography).

i

Materials Science andEngineering Laboratory

FY 2004 Programs andAccomplishments

MetallurgyDivisionCarol A. Handwerker, Chief

Frank W. Gayle, Deputy

NISTIR 7127

September 2004

National Institute ofStandards and TechnologyArden L. Bement, Jr.Director

TechnologyAdministrationPhillip J. BondUndersecretary ofCommerce for Technology

U.S. Departmentof CommerceDonald L. EvansSecretary

ii

Certain commercial entities, equipment, or materials may be identified in this document in order todescribe an experimental procedure or concept adequately. Such identification is not intended to implyrecommendation or endorsement by the National Institute of Standards and Technology, nor is it intendedto imply that the entities, materials, or equipment are necessarily the best available for the purpose.

iii

Table of Contents

Table of Contents

Executive Summary ...................................................................................................................1

Technical Highlights ...................................................................................................................3

Formation of Glass by a First Order Transition ..................................................................4

Interactions with Industry: Delivering Data andStandard Test Methods for Sheet Metal Forming ...............................................................6

Breakthroughs in Magnetic Refrigeration ...........................................................................8

Grand Challenges in Nanomagnetics: High Coercivity FePtAlloys for Future Perpendicular Magnetic Data Storage ................................................. 10

On-Chip Interconnects:Extending Performance of Sub-100 nm Lines .................................................................. 12

The Structural Steel of the World Trade Center Towers .................................................. 14

Advanced Manufacturing Methods ......................................................................................... 16

Sheet Metal Forming for Automotive Applications:Anelasticity and Springback Prediction ............................................................................ 17

Microstructural Origins of Surface Rougheningand Strain Localizations .................................................................................................... 18

Plasticity, Fabrication Processes, and Performance ......................................................... 19

Standard Tests and Data for Sheet Metal Formability ..................................................... 20

The NIST Center for Theoretical and ComputationalMaterials Science .............................................................................................................. 21

Underlying Processes of Plastic Deformation in Metal Alloys ........................................ 22

High Speed Machining ...................................................................................................... 23

Mechanisms for Delivery of Thermodynamic and Kinetic Data ..................................... 24

Phase Field Modeling of Materials Systems:Simulation of Polycrystalline Growth ................................................................................ 25

Hardness Standardization: Rockwell, Vickers, Knoop ..................................................... 26

Nanometrology ........................................................................................................................ 27

Nanoscale Characterization by Electron Microscopy....................................................... 28

Nanostructure Fabrication Processes:Patterned Electrodeposition by Surfactant-Mediated Growth .......................................... 29

Nanostructure Fabrication Processes:Thin Film Stress Measurements ........................................................................................ 30

Nanomechanics: Coupling Modeling with Experiments ................................................... 31

iv

Table of Contents

Biomaterials ............................................................................................................................. 32

USAXS and USAXS Imaging of Biomaterials ................................................................. 33

Materials for Electronics ......................................................................................................... 34

Combinatorial/Phase Diagram Approach forMetallization to Wide-Band-Gap Semiconductors ............................................................ 35

Lead-Free Surface Finishes: Sn Whisker Growth ........................................................... 36

Lead-Free Solders and Solderability ................................................................................. 37

Electrical Properties of On-Chip Interconnections ........................................................... 38

Nanomagnetodynamics ..................................................................................................... 39

Electrodeposited Pt1–x (Fe,Co,Ni)x Alloys ......................................................................... 40

Novel Magnetic Materials for Sensors andUltra-High Density Data Storage ..................................................................................... 41

Discovery of Spin Density Waves in a Ferromagnet: Fe-Al ........................................... 42

Safety and Reliability ............................................................................................................... 43

Analysis of Structural Steel in the World Trade Center .................................................... 44

Standard Test Methods for Fire-Resistant Steel ............................................................... 45

Pipeline Safety: Corrosion, Fracture and Fatigue ............................................................ 46

Frangible Bullets and Soft Body Armor ............................................................................ 47

Marine Forensics and Preservation of Historic Shipwrecks ............................................ 48

Kinetic Penetrators and Amorphous Metals ..................................................................... 49

Metallurgy Division FY04 Annual Report Publication List ..................................................... 51

Metallurgy Division .................................................................................................................. 59

Research Staff ......................................................................................................................... 60

Organizational Charts .............................................................................................................. 61

1

Executive Summary

Executive Summary

This report describes the major technical activitiesand accomplishments in the Metallurgy Division ofNIST Materials Science and Engineering Laboratoryin FY 2004 (October 2003 through September 2004).In this report, we have tried to provide insight into howthe capabilities of the NIST Metallurgy Division arebeing used to solve problems important to the nationaleconomy and the materials metrology infrastructure,and how we establish new programs in response tochanges in national priorities. We welcome feedbackand suggestions on how we can better serve the Nationand encourage increased collaboration to this end.

MissionOur mission is to provide critical leadership in the

development of measurement methods, standards, andfundamental understanding of materials behavior neededby U.S. materials users and producers to become orremain competitive in the changing global marketplace.As a fundamental part of this mission, we areresponsible not only for developing new measurementmethods with broad applicability across materialsclasses and industries, but also for working with otherfederal agencies and industrial sectors to develop andintegrate measurements, standards, software tools, andevaluated data for specific, important applications.

Establishing PrioritiesWe examine a wide range of research opportunities

and set priorities based on the following criteria:the match to the NIST mission, the magnitude andimmediacy of industrial need, the determination that theNIST contribution is critical for success, the anticipatedimpact relative to our investment, the ability to respondin a timely fashion with high-quality output, and theopportunity to advance mission science. We makesuch decisions using a variety of methods, includingroadmapping activities, workshops, technical meetings,standards committee participation, and consultationwith individuals representing U.S. industry and otherfederal agencies.

Technology trends strongly influence the technicaldirections addressed by NIST. We prefer to workin rapidly evolving technologies, where advancesin measurement science are needed to understandthe limitations on system behavior, and, thus, ourcontributions are likely to have a substantial impact onthe course of technology. For NIST as a whole andthe Metallurgy Division in particular, we are committedto having an impact on nanotechnology, homelandsecurity, health care, and information technology.

Over the last two years, we have shifted substantialresources into the areas of nanomagnetics,nanomechanics, and the application of thermodynamicsand kinetics to nanostructure fabrication, and intoa new MSEL Program in Safety and Reliability ofInfrastructure Materials with a focus on homelandsecurity and critical infrastructure protection.The Division highlights and the individual projectdescriptions demonstrate our rapid progress inthese and other longer-standing areas.

Research PortfolioOur 2004 research portfolio focuses on fulfilling

current and future measurement needs of the magneticdata storage, microelectronics packaging, automotive,optoelectronics, and energy distribution industries,on establishing national traceable hardness standardsneeded for international trade, and on developing newmeasurements that could underpin nanotechnologydevelopment. Our output consists of a variety offorms, from a fundamental understanding of materialsbehavior to measurement techniques conveyed throughthe scientific literature and oral presentations, standardreference materials, evaluated data and online databases,software tools, and sensors for on-line process control.

Division Structure and ExpertiseThe Division is composed of 34 scientists,

supported by 5 technicians, 6 administrative staffmembers, and more than 80 guest scientists, andit is organized into five groups that represent theDivision’s core expertise in Metallurgical Processing,Electrochemical Processing, Magnetic Materials,Materials Structure and Characterization, andMaterials Performance. However, by virtue of theinterdisciplinary nature of materials problems in theindustrial and metrology sectors that we serve,Program teams are assembled across group, divisionand laboratory boundaries and with external partnersto best meet project goals. We are committed toassembling the expertise and resources to fulfill ourtechnical goals with the speed and quality necessaryto have the desired impact.

Recognition for Division StaffWe are proud of the accomplishments of the

Metallurgy Division staff in delivering the missionscience tools — measurements, standards, data,modeling — needed by our customers. In FY2004,Division members were recognized for the impact andquality of their work by a wide range of organizations,

2

Executive Summary

including the award of two Gold Medals, the highesthonor conferred by the Department of Commerce, andone Silver Medal from the Department of Commerceand NIST.

Bill Egelhoff was awarded a 2003 Department ofCommerce Gold Medal for his world leadership inthe science and engineering of magnetic thin films.His contributions to the understanding of magnetism,electron transport and thin film deposition have beenmade possible by his Magnetic Engineering ResearchFacility, which has enabled him to measure andcontrol the deposition processing conditions betterthan anyone else in the world. Bill has earned aninternational reputation as a leader in this field andhas produced significant economic benefits to theU.S. magnetic data storage industry.

Carol Handwerker, Ursula Kattner, Bill Boettinger,Maureen Williams, Frank Gayle, Frank Biancaniello,and Tom Siewert were awarded the Departmentof Commerce Gold Medal for their scientific andtechnical leadership in providing the microelectronicsindustry with the critical measurements, modeling,and data necessary for the successful world-wideconversion to lead-free solder in microelectronicsmanufacturing.

Richard Fields was awarded the Departmentof Commerce Silver Medal for his importantcontributions to the science of materials propertymeasurements and modeling. His work has resultedin the recognition of NIST as an important sourceof expertise and leadership by other governmentagencies and the industrial sector when materialsperformance questions need solutions.

Ursula Kattner was named a Fellow of ASMInternational for her outstanding achievement inthermodynamics and the application of phasediagrams to important industrial metallurgicalprocesses.

Bob McMichael was named an IEEE MagneticsSociety Distinguished Lecturer for 2004.

Carol A. HandwerkerChief, Metallurgy Division

3

Technical Highlights


The following Technical Highlights section includesexpanded descriptions of research projects that havebroad applicability and impact. These projects generallycontinue for several years. The results are the productof the efforts of several individuals. The TechnicalHighlights include: Formation of Glass by a First Order Transition Interactions With Industry: Delivering Data and

Standard Test Methods for Sheet Metal Forming Breakthroughs in Magnetic Refrigeration Grand Challenges in Nanomagnetics: High Coercivity

FePt Alloys for Future Perpendicular MagneticData Storage

On-Chip Interconnects: Extending Performanceof Sub-100 nm Lines

The Structural Steel of the World TradeCenter Towers

4


Formation of Glass by a First Order Transition

Recently the aerospace, military and electronicsindustries have devoted much effort to thedevelopment of practical materials that are100 % bulk metallic glass but do not requireextremely rapid quenching. Most glassy solidsare thought to be kinetically frozen liquids,thus the usual approach is to cool fast enough toprevent formation of crystals. We have presentedevidence that an isotropic non-crystalline metallicphase (dubbed “q-glass”) in the Al-Fe-Si systemforms during rapid cooling of the melt by afirst order transition, like a crystalline phase,suggesting the possibility of low-energy,non-periodic structures and the formationof glasses by relatively slow growth.

Leonid A. Bendersky and John W. Cahn

Much DoD research has focused recently onthe development of bulk metallic glasses

(e.g., DARPA SAM program) for use as both armormaterials and kinetic penetrators. Production of suchglasses requires moderate rates of melt quenchingin contrast to the rapid quenching required forconventional metallic glasses. These studies haveall assumed that the glassy phases form only as afrozen liquid. In order to gain an understanding ofthe processing routes available to control these andrelated materials, it is essential to determine howthese structures actually form. Such distinctions inthe formation of various glassy phases may play animportant role in their relative ductility, corrosionresistance and stability. The compositional andstructural nature of the glassy phases is, thus,a question of more general interest.

An Unusual MicrostructureFound in Rapidly Solidifed Al-Fe-Si

Recently we presented evidence that an isotropicnon-crystalline metallic phase (dubbed “q-glass”) couldbe formed by growth from a melt. This phase is thefirst phase, or “primary phase,” to form in the Al-Fe-Sisystem during rapid cooling of the melt.[1] There isstrong evidence that this phase forms from the meltby a first order transition. According to transmissionelectron microscope (TEM) observations, Figure 1a,the q-glass nucleates from the melt as isolatedparticles, which initially grow as spheres (nodules).The diffraction pattern taken from a nodule shows itto be an isotropic glassy phase, Figure 1c. Dark field,

Figure 1b and high-resolution TEM imaging, Figure 2,also support the isotropic, but highly speckled, natureof the phase. There is a nucleation barrier, whichimplies an interface between the glass and the melt.Further evidence for the first order transition is foundin the solute partitioning at this interface; the q-glassrejects Al into the melt, which ultimately solidifies as aeutectic composed mainly of crystalline fcc aluminumand more q-glass. Presumably glass and melt aredistinct phases with an interface having a positive,surface-free energy. Such solidification microstructuresare familiar and well understood, except that here theglass is the first phase growing from the melt andeventually becomes surrounded by eutectic, insteadof being the last phase (sometimes the only phase)formed when atomic motion ceases in the melt.

Figure 1: a) Microstructure of an Al91Fe7Si2 alloy afterelectron-beam surface melting with scan velocity of 50 cm/sec.b) Dark field image of a nodule and the glassy phase in themonotectic, obtained by positioning a SAD aperture on thestrongest diffuse ring in c).

5


Figure 2: High-resolution TEM image showing an interfacebetween a glassy nodule (q-glass) and Al. Note the isotropicphase contrast of the q-glass.

Q-Glass PropertiesStability. A lower entropy implies a higher degree

of order, which in turn is consistent with a processof growth that selects some components and rejectsothers. This seems to be the case for the q-glass whereAl is rejected into the melt. The lower entropy impliesthat at low temperatures the q-glass may have greaterstability than a conventional glass (frozen liquid) of thesame composition, and that there would be a latent heatassociated with its formation which, as is true for allincongruent freezing, is spread over a freezing range.

The q-glass in Al-Fe-Si is metastable, as were theoriginal AlMn quasicrystals. Based on the quasicrystalexperience, it is expected that, in time, fully stableq-glasses will be found in systems with morecomponents.

Structure and Order. The implied order andnarrower composition range are closely related.We expect q-glasses to have a chemical order witha smaller set of local environments. Recently,there have been many advances in creating orderlyarrangements which are not crystalline.[2] This worksuggests that it might be possible that a glassy structureis composed of unit cells which are not parallelepipedsand cannot be packed in a periodic or quasiperiodicway. Unlike the pinwheel tilings in two dimensionswhich take on every orientation but need matching rulesto prevent them from packing periodically, there are3-D Schmitt–Conway–Danzig tiles which can fill spaceonly in an aperiodic way.[2] If atoms pack into such a

unit cell, q-glasses could be stoichiometric and havelow entropy, as do crystalline compounds.

There is some confirmation for such order inhydrogen-storage alloy glasses. For these glasses, thecharging and discharging pressure have a long plateau,just as for crystals. This unusual behavior for glassesimplies a highly ordered glass structure in which thereis only one type of site for the stored hydrogen.

On-Going WorkThe concept of q-glass was introduced at the MRS

meeting last year at “Symposium MM: Amorphous andNanocrystalline Metals.”[1] Since then, our work hasreceived support from the Air Force Office of ScientificResearch. Current work is focused on three areas:

1. Metastable Al-Fe-Si phase diagram. The ternaryAl-Fe-Si phase diagram has been extensively studied.Thermodynamic principles of metastable phaseequilibria are well developed, and much is alreadyknown about metastable crystalline phases in thisternary system, but not about the q-glass. We areinvestigating conditions for q-glass formation andits subsequent conversion to other phases to mapout the bounds of q-glass on the metastable phasediagram. In addition, calorimetric studies arebeing conducted to evaluate the thermodynamicsof q-glass formation.

2. Other glass forming Al-based systems. Judgingfrom the current literature, it is possible that theq-glass forms in other Al-based systems.[3] We usemorphological criterion to distinguish the formationof q-glass from conventional glass, with a special focuson systems known for forming metallic glasses.

3. Other methods of growing q-glass. Feasibilitystudies to fabricate and study the formation andstability of metallic glasses will be conducted usingcombinatorial methods. Libraries of amorphousmetal thin films will be fabricated using the newMetallurgy Division e-beam evaporation system bydepositing elemental wedge-shaped layers controlledby moving shutters. This combinatorial approachwill be applied to measuring the stability and controlof glasses.

For More Information on this Topic1. J.W. Cahn and L.A. Bendersky, “An isotropic glass

phase in Al-Fe-Si formed by a first order transition,”MRS 2003 Proceedings.

2. M. Senechal, Quasicrystals and Geometry, NewYork: Cambridge University Press, Chap. 7, (1995).

3. A. Inoue, M. Watanabe, H.M. Kimura, A. Nagata,and T. Masumoto, Matls. Trans., 8, 723 (1992).

6

The automotive industry’s push to incorporatelighter sheet metal parts into new car designshas been hampered by lack of detailed data onmechanical behavior for forming. The MetallurgyDivision has developed new metrology and testmethods to address specific problems identifiedby industry as inhibiting broad adaptation ofhigh-strength steel and aluminum sheet inautomotive applications.

Tim Foecke, Mark Stoudt, and Mark Iadicola

The Sheet Metal Forming program within theMetallurgy Division has been designed from the

outset to be relevant to the American manufacturingsector. Surveys conducted by NIST staff through theIndustrial Liaison Office (ILO) sought to determinewhich aspects of sheet metal forming were most inneed of better data and metrology. The results revealedthat industry had a clear understanding of how NISTexpertise could be applied and predicted a high potentialimpact if the research was successfully implemented.

The U.S. automotive industry estimates that thecosts associated with the inability of the finite elementmodels to adequately predict friction, multiaxialhardening, or springback are on the order of hundredsof millions of dollars annually. These shortcomingsare the direct result of insufficient property data andthe physical understanding that such data can reveal.The projects described here are specifically designedto address the quality of the sheet metal property dataand models used to develop predictive finite elementmodels of forming operations.

The Microstructural Origin ofRoughening and Strain Localization

Surface roughness is a major contributor to frictionduring sheet metal forming, and thus must be wellcontrolled in order to obtain a consistent product.Furthermore, the surface qualities of the finished productdetermine ultimate aesthetic characteristics of the vehicle.Ability to accurately characterize surface roughnessis, therefore, of critical importance to the industry.

There are many sources of error in surfaceroughness measurements. Recent studies of thediscrepancies between measured and predicted surfaceroughness show that the greatest error is introducedby the numerical tools used routinely by industry tointerpret the roughness data. Parameters such as

the rms roughness (Rq) compress complex surfacedetails into singular expressions that do not adequatelyrepresent the surface features involved (Figure 1).Changes in the surface feature arrangement, whichare not captured by the rms roughness, have a greaterinfluence on friction. Hence, any prediction of frictionthat is based solely on a tool such as the rms roughnessis likely to yield erroneous results.

Interactions with Industry: Delivering Data andStandard Test Methods for Sheet Metal Forming


Figure 1: Two surface conditions exhibiting the same roughnessmathematically, but very different appearances and frictionalbehaviors.

These results were presented to the automotiveindustry at several major research conferences in thepast year. Invited presentations included talks at the“2004 Society of Automotive Engineers World Congress”in Detroit, MI, the “2004 North American DeepDrawing Research Group” spring meeting at NIST,the “Springback Compensation Program” and the “AgileFlexible Binder Control” quarterly meetings held at theU.S. Council for Automotive Research (USCAR) inSouthfield, MI, as well as the “Automotive Sheet SteelsSymposium” in Chicago, IL. The results were alsopublished in both archival journal and in conferenceproceeding formats.

Overall, the industrial response has been verypositive, and there is a high level of interest in theoutcome of new studies. The high-resolutionsurface topography data obtained with the MetallurgyDivision’s scanning laser confocal microscope has beenparticularly well received and has stimulated severalnew opportunities identified by our industrial contactsfor direct collaboration with the automotive industry.

Multiaxial Flow Surfaces andStress-Based Forming Limit Curves

An X-ray stress measuring system was recentlyinstalled on the sheet metal formability station. Thisunique instrumentation allows for the direct, in situ

7

measurement of the stress in a given direction while thesample is under multiaxial load. This year, the multiaxialstress–strain surfaces have been mapped for threealuminum sheet alloys of particular interest to theautomotive industry: AA5182, AA6111 and AA6022.With this experimental system, shown in Figure 2,we have made the first-ever recorded measurementsof the biaxial hardening exponent. These data willprovide a substantial advantage to industry, allowing thereplacement of current practices that estimate biaxialflow behavior from a series of extrapolated uniaxialtensile tests. This new approach is expected to enablemore accurate designs and greater cost savings.

Press releases and technical articles have broughtthese new capabilities to the attention of industry. Visitsand inquiries from companies such as ISG-InternationalSteel Group, Volvo, and Avesta–Polarit have stimulatedthe exploration of mutually beneficial research that canproduce enabling data and new insights into formability.

Accumulated plastic strain has been used fordecades to describe the deformation limits of sheetmetal. However, this formalism does not account forpath dependence (i.e., how deformation responsedepends on how it is deformed). Stress, on the otherhand, is an intrinsic property of metals, and the abilityto describe a process in terms of critical stresses ratherthan strains would remove the dependence on path.Recent work at GM suggests that this approach wouldgreatly improve finite element predictions by betterdefining the safe working envelope of the materials.Still needed are direct measurements of multiaxial flowsurfaces, measurements that only NIST can provide.

A collaboration with GM has been developed usingthese new data to remove the pure isotropic andkinematic hardening models currently in use.Improvements to the modeling and limit formalismswill be directly transferred to the auto industry.

NUMISHEET 2005A triennial international conference focusing on

simulation techniques for sheet metal forming die designwill be held in Detroit in 2005. An integral part of thisconference is a round-robin simulation exercise wherevarious modeling groups try to predict the final shapeof a prototypical forming operation. NIST has beenasked to take the key role of measuring materialproperties and then forming the prototype part.This year, the new NIST capability to measure stressesin situ has allowed an increase in the complexity andsophistication of the round-robin simulation. Predictionof both through-thickness residual stresses in a sheetthat has undergone multiple forming operations andthe stresses in orthogonal directions in a sheet underload in a die have become vital components of thebenchmark for the conference, representing a“next-level” challenge for modelers to predict.

Springback Prediction“Springback,” during and after forming operations,

limits the accuracy and consistency of the final part.The cup test for measuring springback is a fairly recentdevelopment and has not been completely characterized.NIST development and testing of the springback cuptest has now been largely completed. A collaborationwith U.S. Steel has shown that computer models thathave been recently modified to account for springbackcan now accurately predict the final geometry of thesample. More significantly, the models come close topredicting the residual stress profiles that NIST hasmeasured with neutron and X-ray diffraction.

Working with ASTM and industry, NIST hassubmitted the flat-bottomed springback cup test forconsideration as a standard measurement protocolfor the capacity for elastic springback. A committee,led by Ford and including both material suppliersand end-users, will consider this new test. Datagenerated by NIST on the robustness of the testand the sensitivity to experimental variations willform the basis of round-robin analyses.

For More Information on this Topic

L.E. Levine (Metallurgy Division, NIST);T. Gnäupel–Herold, H. Prask (NIST Center forNeutron Research); M. Shi (USS); E. Chu (ALCOA);C. Xia (Ford); T. Stoughton, M. Wenner, C.T. Wang(GM); A. Lin (Univ. of Illinois)


Figure 2: Newly installed X-ray stress measuring system as ittakes data on sheet metal being deformed in biaxial tension.

8


Breakthroughs in Magnetic Refrigeration

In the past 10 to 15 years, there has been asubstantial effort to find alternative coolingtechnologies to vapor compression systems.This has been driven largely by the need toreduce the high energy costs of present systemsalong with their use of harmful CFC refrigerants.Magnetic refrigeration technology is one suchalternative. In principal, because it is a reversibleprocess, magnetic cooling should be more efficientthan the competing irreversible vapor compressionsystems. However, magnetic cooling technologyhas historically been relegated to cryogenic,high-magnetic field applications since efficiencywas so poor at higher temperatures. In 2004, wemade progress in identifying materials with greatlyimproved performance close to room temperature.For U.S. industry to take advantage of these newmaterials, their behavior needs to be understoodand methods of proper measurement determined.We are providing the metrology to do that.

Robert Shull, Virgil Provenzano, andAlexander Shapiro

In 1997 a large (“giant”) magnetocaloric effect wasreported in the Gd5Ge2Si2 compound between 270 K

and 300 K, thereby generating a great deal of interest asa potential near-room temperature magnetic refrigerant.Unfortunately, large hysteretic losses occur in the sametemperature range where this compound exhibits itslarge magnetocaloric effect, making the magneticrefrigeration with the material inefficient.

In Nature [Vol. 429, p. 853 (2004)], we reportedon the discovery of an effective and straightforwardmethod for greatly reducing (by more than 90 percent)the large hysteresis losses in Gd5Ge2Si2. This isaccomplished by alloying the compound with a smallamount of iron, which has the additional benefit ofshifting the magnetic entropy change (∆Sm) peakto higher temperature, from 275 K to 305 K, andbroadening its width. Although the addition of Fe yieldscomparable refrigerant capacity (RC) values, a greaterresultant value is obtained for the Fe-containing alloywhen the hysteresis losses are subtracted. TheFe-containing alloy is clearly better for near-roomtemperature magnetic refrigeration applications.

Figure 1 shows the magnetization (M) vs. appliedmagnetic field (H) data measured in the importanttemperature range of peak magnetocaloric effects

for the Gd5Ge2Si2 compound. In this material, theenhanced values of the magnetocaloric effect havebeen ascribed to the existence of a field-inducedcrystallographic phase change in the material froma low-field paramagnetic structure to a high-fieldferromagnetic crystal structure. As a consequence ofthis field-induced phase change, there is hysteresis inits transformation, shown by the shaded areas enclosedwithin the field cycles, corresponding to the cost inenergy to make one cycle. The magnitude must beassessed when determining whether a particularrefrigerant is useful in a given application.

Figure 2: Magnetic hysteresis measurements showing nofield-induced ferromagnetic transition and nearly zero hysteresisin the Gd5Ge1.9Si2Fe0.1 alloy.

Figure 1: Magnetic hysteresis measurements showing afield-induced ferromagnetic transition and large hysteresis inthe “giant magnetocaloric effect” Gd5Ge2Si2 compound.

We have found that by adding a small amount ofFe to the Gd5Ge2Si2 compound, the hysteresis maybe significantly reduced. This effect is shown in theM vs. H data for this alloy in the important temperaturerange of potential use in Figure 2, where the shadedhysteresis regions are nearly zero.

9


A magnetic field is alternately applied and removedfrom a refrigerant during a single cooling cycle in amagnetic refrigerator, thus any magnetic hysteresiswill reduce the usefulness of a material. Figure 3shows exactly how much of a difference in hysteresisthere is between the Gd5Ge2Si2 and Gd5Ge1.9Si2Fe0.1alloys.

The reduced hysteresis in the quaternary alloy is dueto a lack of field-induced paramagnetic-to-ferromagneticcrystallographic phase change. This change in behavioris caused by the Fe combining with primarily the Si inthe ternary compound to form an Fe-rich secondaryphase, thereby reducing the Si content in the majorityGd-Ge-Si phase. As a consequence, the thermodynamicsof the Gd-Ge-Si alloy were changed from that of theprecursor Gd5Ge2Si2 compound.

The magnetic entropy change (∆Sm) for a 5 Tapplied field as a function of temperature for bothalloys is shown in Figure 4. The computed refrigerantcapacities (RC) are indicated on the figures and shownby the shaded areas. Compared to Gd5Ge2Si2, theiron-containing compound exhibits a smaller, butbroader ∆Sm peak, and a ∆Sm peak at a highertemperature. Purely on this basis, Gd5Ge2Si2 is thesuperior material, but the Fe-containing alloy might beattractive for higher temperature applications due toits higher temperature peak.

One method to account for the hysteresis losses ofeach alloy is to simply subtract it from the correspondingrefrigerant capacity value. Subtraction of the averagehysteresis in the temperature range of the ∆Sm peakfor each alloy (~ 65 J/kg and ~ 4 J/kg, respectively,for the ternary and quaternary alloys) from thecorresponding refrigerant capacity values yieldedapproximate values of 240 J/kg for the Fe-free alloyand 355 J/kg for the Fe-containing alloy. Consequently,when the hysteresis losses are taken into account,it is clear that the Gd5Ge1.9Si2Fe0.1 alloy is a much

better magnetic refrigerant than the Gd5Ge2Si2compound despite its lower ∆Sm peak value.Most prior evaluations of magnetic refrigerants havedisregarded the material hysteresis losses, and insome materials these losses can be quite large; thisexample shows the weakness in that approach.

“The results are pretty amazing,” states MichaelDiPirro [Science News 165, 405 (2004)], a cryogenicengineer at the NASA Goddard Space Flight Centerin Greenbelt, Maryland.


Virgil Provenzano, Robert D. Shull, andAlexander J. Shapiro, “Reduction of hysteresislosses in the magnetic refrigerant Gd5Ge2Si2 bythe addition of Iron,” Nature 429, 853 (2004).

Figure 3: Hysteresis energy loss as a function of temperaturefor the Gd5Ge2Si2 and Gd5Ge1.9Si2Fe0.1 alloys.

Figure 4: Computed ∆Sm [for ∆H = 3980 kA/m (5 T)],normalized with respect to sample mass, of the Gd5Ge2Si2compound (top) and of the Gd5Ge1.9Si2Fe0.1 alloy (bottom),plotted as a function of temperature showing the presence ofpeaks centered near 270 K and 305 K respectively for Gd5Ge2Si2and Gd5Ge1.9Si2Fe0.1.

10


The magnetic data storage industry is seekingto sustain the 30-year trend of exponentiallyincreasing storage density. This must be achievedat the same time as maintaining the stability of therecorded data. The Metallurgy Division at NISTis collaborating with Seagate Technology todevelop the processes required to producepatterned media of high-magnetic coercivitythat will meet this challenge.

Jonathan J. Mallett, Thomas P. Moffat, andWilliam F. Egelhoff, Jr.

Conventional magnetic data storage media are basedon granular films in which many fine grains are

used to define each magnetic bit. The minimumnumber of grains per bit is limited by the irregularityof the shape and spacing of the grains, which producesirregular boundaries between bits. This irregularityresults in noise in the read signal and can only beconstrained at an acceptable level by maintaining afixed minimum number of grains per bit. The naturalapproach to increasing data storage density has beento reduce the grain size, while maintaining the samenumber of grains per bit. The limit to this approachoccurs when the magnetic energy barrier to the reversalof magnetization of each grain approaches the energy ofrandom thermal fluctuations. At this point, the mediumis no longer stable against spontaneous magnetizationreversal and consequential data loss.

The grain size at which this limit occurs dependson the magnetocrystalline anisotropy energy of the

Grand Challenges in Nanomagnetics: High Coercivity FePtAlloys for Future Perpendicular Magnetic Data Storage

magnetic material. FePt in its L10 phase has asufficiently large anisotropy energy to allowmagnetically stable grains of 5 nm diameter (Figure 1).Furthermore, FePt deposited by a variety of meanstypically forms grains approaching this size.Unfortunately, the annealing treatment required totransform the as-deposited A1-structured FePt tothe desirable L10 phase is widely found to resultin an increase in grain size to 100 nm. Currentefforts at NIST focus on developing a method ofelectrochemically depositing FePt into a regularpatterned template, circumventing the problem of graingrowth. Simultaneously, the imposed regularity of themagnetic cells opens the possibility of addressing singlecells (i.e., of using one “grain” per bit). The estimatedachievable density from such an approach is 7 Tbitsper square inch, which is 100 times higher than currentstorage densities. Electrodeposition is an obviouschoice for deposition into high aspect ratio templates,since it does not suffer from the shadowing effectthat characterizes vacuum deposition techniques.

Figure 2: The double cell, designed to minimize the concentrationof Fe3+ and dissolved O2 in the FePt plating solution.

Figure 1: The dependence of storage density on magnetocrystallineanisotropy energy. The most promising materials are FePtand CoPt.

Electrodeposition of FePt from aqueous solutionspresents many challenges. The double cell shown inFigure 2 has been developed to address the problem ofthe instability of Fe2+ solutions, which readily oxidizeto produce insoluble Fe(OH)3. The plating solution,containing FeCl2 and PtCl4 (left) is separated from theanode by a cation selective membrane, preventing theoxidation of Fe2+ to Fe3+ that would otherwise occurat the anode. Fe3+ formation by oxidation with air isalso minimized by blanketing the cell in nitrogen, andresidual Fe3+ is reduced to Fe2+ by an auxiliaryplatinum grid electrode.

11


The alloy composition of 300 nm thick filmsdeposited from this solution was found to dependmainly on the applied potential and to be insensitiveto the concentration of the solution components.A thermodynamic regular solution model was usedto describe the dependence of film composition onapplied potential. The comparison of theory toexperimental data can be seen in Figure 3.

The deposited films were found to be remarkablysmooth, with RMS roughness values less than 5 nmfor micron-thick deposits. This was surprising giventhe large platinum overpotential (supersaturation),which usually results in growth instabilities andconsequent roughening.

The control of crystal orientation is an essentialconsideration for high-density recording applications.A transition in the recording industry is currently inprogress to perpendicular media, in which the bits aremagnetized perpendicular to the plane of the medium.Maximum advantage can be derived from L10 FePt as aperpendicular medium when it is correctly aligned withits magnetic easy-axis perpendicular to the substrateplane. A careful choice of substrate and annealingparameters is required to recrystallize the as-depositedrandom fcc alloy to appropriately oriented L10.

Figure 3: The theoretical and experimental dependence ofcomposition on the applied potential.

FePt electrodeposited onto a Cu (001) substratehas recently shown great promise. X-ray diffractionrevealed the transition to L10 upon annealing, and X-raypole figures indicated favorable orientations. Figure 4shows FePt L10 (001) and (110) pole figures. The(001) figure indicates perpendicular orientation of themagnetic easy-axis (the c-axis), while the (110) figureindicates an in-plane texture. The in-plane texture is anadded advantage, as it results in a narrower switchingfield distribution.

Copper additions to the alloy have been found tolower the A1/L10 phase transition temperature by upto 90 oC. It may be speculated that interdiffusion ofcopper from the substrate during the anneal allowed therecrystallization to proceed from the interface with thesubstrate. It is likely that this would allow the film toreplicate the orientation of the substrate.

Figure 4: X-ray pole figures showing favorably oriented FePt L10 .

Figure 5: Magnetic hysteresis measurements showing a 10 kOecoercivity.

Magnetic hysteresis measurements performedusing a Kerr magnetometer are shown in Figure 5.The magnetic coercivity of 10 kOe is comparable tofigures quoted in the literature for vacuum-depositedFePt. It is believed to be the highest value reportedfor an electrodeposited film.

Current efforts focus on controlling theinterdiffusion between substrate and film to minimizeloss of magnetization in the alloy, and on movingfrom planar films to through-mask deposited arraysof FePt pillars.


W.F. Egelhoff, Jr. (Metallurgy Division, NIST);E.B. Svedberg (Seagate Research, Pittsburgh)

12


Conductors in on-chip metallizations are nowreaching dimensions so small that defectiveseed layers are impacting manufacturing yields,and electron scattering on surfaces and grainboundaries is reducing electrical transport inthe buried wires. Our goal is to provide toolsto overcome these barriers. Recent efforts havequantified sources of the increased resistivity inwires made of silver, the most conductive of allmetals, demonstrated seedless processing routesfor copper wires, and improved understandingand modeling of the superfill fabrication process.

Daniel Josell and Thomas P. Moffat

The steady reduction of transistor dimensions inintegrated circuits has been accompanied by similar

size reductions of the on-chip interconnects that carryelectrical signals, pushing the industrial technology forcopper seed deposition close to its limit for defect-freesidewall coverage. Defects in seed layers, whicharise from limitations in existing sputter technology,lead to voiding during electrodeposition of thecopper metallization.

Additionally, with the dimensions of the copperwires in these metallizations now approaching theintrinsic mean-free-path length of the conductionelectrons, scattering on the wire surfaces has begunto significantly reduce electron transport and, thus,the associated electrical conductance of the wires.With grains in these conductors similarly sized, areduction is also to be expected from grain boundaryscattering. While the penalty for both effects isincreased power dissipation and reduced clock speed,the appropriate approach for mitigation requiresquantitative determination of the relative sizes ofthe contributions.

On-Chip Interconnects:Extending Performance of Sub-100 nm Lines

Technical DetailsA tri-layer titanium, palladium and silver seed

(Figure 1) was shown to yield smooth, conductivesurfaces for the electrodeposition of silver wires forthe electrical properties study. The poor seed coveragethat is visible toward the bottom of the smallesttrench (Figure 1c) is a technical challenge noted in theInternational Technology Roadmap for Semiconductors.Such seed defects motivated the “seedless” rutheniumbarrier-based process detailed in last year’s report andcontinued by this year’s demonstration of seedlesscopper superfill in trenches with ruthenium or iridiumbarriers (Figure 2) deposited by perfectly conformalatomic layer deposition (ALD).

Figure 1: Ti/Pd/Ag seed layer in sub-100 nm deep trenches.

Figure 2: Trenches containing copper that was electrodepositeddirectly on an ALD iridium barrier (thin bright layer).

Figure 3: Cross-section views of silver wires fabricated byelectrodeposition in the seeded trenches followed by removal of themetal from the field (transmission electron microscope).

13


Figure 4: Electrical resistivities of 300 nm tall silver wirescompared to predictions that account for surface scattering withvarying amounts of grain boundary scattering. Resistivities forwires less than 100 nm wide are impacted by defects.

For the electrical studies, silver wires (Figure 3) werefabricated by silver electrodeposition on the tri-layer seeds(Figure 1), using a superfill process developed in theMetallurgy Division, followed by removal of the metalin the field adjacent to the wires through chemical–mechanical planarization and ion polishing.

The wires were studied in a standard four-point probegeometry that permitted measurement of wire resistances.Resistivities, obtained from the resistances using measuredwire dimensions, increased significantly with decreasingwire width (Figure 4). To assess the origin of the resistivityincrease, the Fuchs–Sondheimer analysis for diffusescattering of electrons on surfaces was extended topermit analysis in the presence of both specular anddiffuse scattering. The resulting equations, along witha previously published equation for grain boundaryscattering, permitted quantitative evaluation of theexperimental data. Significantly, modeling of the datashowed that the increase of resistivity with decreaseof wire size arises as much from scattering on grainboundaries as from scattering on the wire surfaces.This indicates that efforts to mitigate size effects mustincrease grain size as well as surface specularity ifthey expect to be more than modestly successful.

The continuing industrial need for predictivesimulation of feature filling spurred experiment andmodeling of the superfill process itself. Consumptionof adsorbed accelerator, a detrimental deviation fromthe surface segregation behavior that is responsible forthe superfill process, was measured. Inclusion of suchconsumption in our Curvature Enhanced AcceleratorCoverage (CEAC) model and computer code have madethe filling predictions even more accurate.

This research has continued to impact industry,indicated by requests for our superfill code from Intel,

Applied Materials, ST Microelectronics, and ATMI;an invited article on superfill in the IBM Journal ofResearch and Development; and invited presentationsgiven at Cookson–Enthone, in addition to publicationsand presentations at conferences.

Selected Project Publicationsfor FY2004

T.P. Moffat, D. Wheeler, M. Edelstein and D. Josell,“Superconformal Film Growth: Mechanism andQuantification,” IBM J. Res. and Dev., in press.

D. Josell, C. Burkhard, Y. Li, Y.-W. Cheng, R.R.Keller, C.A. Witt, D. Kelley, J.E. Bonevich, B.C. Baker,T.P. Moffat, “Electrical Properties of SuperfilledSub-Micrometer Silver Metallizations,” J. Appl. Phys.96 (1), 759–768, (2004).

D. Wheeler, T.P. Moffat, G.B. McFadden, S. Corielland D. Josell, “Influence of Catalytic Surfactanton Roughness Evolution During Film Growth,”J. Electrochem. Soc. 151 (8), C538–C544, (2004).

T.P. Moffat, D. Wheeler, and D. Josell,“Electrodeposition of Copper in the SPS-PEG-Cl AdditiveSystem: I. Kinetic Measurements: Influence of SPS,”J. Electrochem. Soc. 151 (4), C262–C271, (2004).

W.J. Evans, D.G. Giarikos, D. Josell, and J.W. Ziller,“Synthesis and Structure of Polymeric Networks ofSilver Hexafluoroacetylacetonate Complexes of THF,Toluene, and Vinyltrimethylsilane,” Inorg. Chem. 42,8255–8261, (2003).

For More Information on this TopicD. Josell, T.P. Moffat (Metallurgy Division, NIST);

G. McFadden (Mathematical and Computational SciencesDivision, NIST); R.R. Keller, Y.-W. Cheng (MaterialsReliability Division, NIST); C. Witt (Cookson–Enthone);C. Burkhard, Y. Li (Clarkson University); D. Wheeler(University of Maryland); T.K. Aaltonen, M. Ritala,M. Leskelä (University of Helsinki, Finland)

Figure 5: Simulations of superconformal feature filling nowaccount for consumption (incorporation) of adsorbed catalyst.

14


The Structural Steel of the World Trade Center Towers

Figure 1: Enhanced image of the impact zone has alloweddetermination of failure modes in the perimeter columns.An outline of the plane is superimposed. Steel panels recoveredfor the investigation are highlighted in color.

Task 2 — Categorize Failure MechanismsBased on Visual Evidence

The damage to the steel components of the buildingsrepresents an important check on the accuracy of the fireand impact modeling, with the caveat that the recoveredsteel could have been damaged before the collapse,during the collapse, or during the salvage efforts.

Image enhancement techniques were used tostudy photographs of the impact zones of the towersto establish the failure mechanisms of the perimetercolumns. By making montages of many differentimages, it was possible to produce a nearly smoke-free,composite view of the columns that the aircraft struck.The columns failed by several different mechanisms.One important result was the demonstration that severalkey impact-zone columns in the NIST inventory weredamaged primarily in the aircraft impact, and not in thesubsequent collapse or recovery efforts. Figure 1 is amap of the impact zone of WTC 1 that illustrates thenature of the damage to the individual columns.

Tasks 3 & 4 — Determine SteelProperties to Support Airplane Impactand Structure Performance Studies;Correlate Properties with ThoseSpecified in the Construction

One important part of this task was to assess thequality of steel and determine whether it met the originalspecifications. In support of this goal, the group conductedmore than 200 room-temperature tensile tests on samplesfrom 39 different building components, which represented

In August of 2002, NIST took responsibility for theinvestigation of the World Trade Center disaster. Theinvestigation objectives include determining: whythe buildings collapsed; the procedures and practicesused in the design, construction, operation andmaintenance of the buildings; and areas in codesand practices that warrant revision. The eightinterdependent projects include Project 3 — Analysisof Structural Steel, led by the Materials Science andEngineering Laboratory. The objective of the projectis to determine and analyze the properties andquality of the steel, weldments, and connectionsfrom steel recovered from the World Trade Center.

Stephen W. Banovic, Richard J. Fields,Timothy J. Foecke, William E. Luecke,J. David McColskey, Christopher N. McCowan,Thomas A. Siewert, and Frank W. Gayle

In terms of the steel used, the World Trade Centerwas very complex. Plans called for steels of fourteen

different yield strengths between 36 ksi and 100 ksi.Many of these steels were proprietary and were notsupplied to any existing ASTM specification. To furthercomplicate matters, four different fabricators providedsteel for the upper stories, and each used steel frommore than one supplier. Most of the records thatdocumented the steel actually used were either notpreserved or destroyed in the collapse.

The MSEL Metallurgy and Materials ReliabilityDivisions have determined properties of the steelrecovered from the WTC site and characterized failuremodes associated with pre-collapse damage. Thisinformation has been provided for use in models of thebuilding response to airplane impact and fire. In 2004,all major tasks were completed.

Task 1 — Collect and CatalogPhysical Evidence

The Structural Engineers Association of New York(SEAoNY), later supplemented by NIST personnel, spentcountless hours selecting steel in the New Jersey recyclingyards for later forensic analysis. After shipment to NIST,the Metallurgy Division cataloged the 236 recovered steelcomponents. Since many pieces were stamped and paintedwith unique serial numbers correlated with the buildingplans, it was often possible to identify the exact originallocation in the building.

As a second part of this task, we also researchedcontemporaneous construction documents and steelspecifications. This task helped identify steel suppliersand fabricators, as well as material substitutions.

15


Figure 2: Yield strength — measured vs. specified minimum.The few anomalous data can be attributed to damage incurredin the collapse that removed the yield point behavior, or to thenatural and accepted strength variability of structural steel.

all the steels relevant to the investigation. Tests indicatethat the steels generally met or exceeded the specifiedminimum strengths (Figure 2).

Steel specifications also regulate the chemicalcompositions. More than 350 different specimens,representing nearly every relevant steel component inthe building, were analyzed for chemical compositionand all of the sections relevant to the investigation. Thechemical analysis helped confirm archival informationthat an American steel mill supplied steel for parts ofthe perimeter columns. The analyses also support theconclusion that the steels in the buildings met thespecifications called for in the building plans.

The strength of steel increases as the deformationrate increases. Accurately modeling these rate effectsis essential for estimating the energy and momentum ofthe aircraft remnants which damaged the core columns.To establish the strain rate sensitivity of the relevantsteels, we have conducted more than 100 tests onspecimens from both the perimeter and core columnsat rates up to 2000 s–1 (200,000 % deformationper second).

Ultimately, the fires in the towers weakened the steelstructure leading to collapse. We conducted more than100 high-temperature tensile and creep tests to providemodels of the deformation of structural steel at elevatedtemperatures for use in models of the building responseto the fires.

Task 5 — Analyze Steel toEstimate Temperature

Just as the nature of the deformation and failureof the recovered steel reveals information about the

impact and collapse, the steel can also contain evidenceof its exposure to the elevated temperatures in the fire.The situation is made more complex because muchof the steel was also exposed to extended fires inthe rubble before recovery. Several different methods,both conventional and novel, were examined forestimating high-temperature excursions seen by thesteel. Only one method proved to be robust and easyto implement: paint on steels that reached temperaturesover 250 °C cracked from the difference in thermalexpansion between the paint and the steel.

Figure 3: Illustration of the method to establish the temperatureexcursions of recovered steel. The segment of the columnprotected by the floor slab remained at lower temperature.

From photographs showing fire in windows, allrecovered columns that may have been exposed topre-collapse fire were identified and characterizedusing the paint crack technique. Figure 3 illustratesone of the columns that showed evidence of havingexceeded 250 °C.

OutputsOutputs for the project include parts of the

June 2004 Progress Report on the World Trade CenterDisaster (NIST SP 1000-5) and numerous memos forcontractors that explain how to employ Project 3 dataand deformation models. The final NIST investigationreport will be released in the spring of 2005.


T.A. Siewert (Materials Reliability Division, NIST);F.W. Gayle (Metallurgy Division, NIST)

16

Program Overview

The competitiveness of U.S. manufacturersdepends on their ability to create new product conceptsand to speed the translation from concept to marketwhile decreasing product cost. This is equally truefor well-established “commodity” industries, suchas automotive and aerospace, and rapidly growingor emerging industries, such as biotechnology andnanotechnology. For existing products, manufacturingis a critical step in reducing product cycle time. Rapid,low-cost development of manufacturing processes isneeded to incorporate new materials into complexproduct shapes with higher performance at equivalentor lower cost as the competing, established materialsand methods. For innovative product concepts, newmaterials with increasing functionality are needed totranslate these concepts to reality.

To realize such improvements in materials andmanufacturing, MSEL is developing robust measurementmethods, models, standards, and materials and processdata needed for design, monitoring, and control ofmanufacturing processes. A growing challenge is beingable to design, monitor, and control such materials andmanufacturing processes at size scales from nanometersto meters. The Advanced Manufacturing ProcessesProgram focuses on the following high-impact areas: Combinatorial, high-throughput methods for materials

ranging from thin films and nanocomposites tomicro- and macroscale material structures;

Industry-targeted R&D centered on uniquemeasurement facilities in forming of lightweightmetals for automotive applications, polymerprocessing, and high-speed machining;

Innovative testbeds for emerging materials,including carbon nanotubes and fuel cells;

National traceable standards having a major impacton trade, such as hardness standards for metalsand process standards for polymers; and

Innovative, physics-based process modeling tools.

Our research is often conducted in closecollaboration with industrial consortia and standardsorganizations. These collaborations not only ensurethe relevance of our research, but also promote rapidtransfer of our research to industry for implementation.Three projects focused on Advanced ManufacturingProcesses are highlighted below.

NIST Combinatorial Methods Center (NCMC)The NCMC develops novel high-throughput

measurement techniques and combinatorial experimentalstrategies specifically geared towards materials research.

Advanced Manufacturing Processes

These tools enable rapid acquisition and analysis ofphysical and chemical data, thereby accelerating thepace of materials discovery and knowledge generation.By providing measurement infrastructure, standards, andprotocols, and expanding existing capabilities relevantto combinatorial approaches, the NCMC lowers barriersto the widespread industrial implementation of this newR&D paradigm. MSEL uses a two-pronged strategy foraccelerating the development and implementation of theseapproaches: an active intramural R&D program thatdemonstrates the ability of combinatorial methods toproduce cutting-edge scientific research and an ambitiousoutreach activity; key to this effort is the validation ofthese approaches with respect to traditional “one at a time”experimental strategies.

Forming of Lightweight MetalsAutomobile manufacturing is a materials intensive

industry that involves about 10 % of the U.S. workforce.In spite of the use of the most advanced, cost-effectivetechnologies, this globally competitive industry has majorproductivity issues related to measurement science anddata. Chief among these is the difficulty of designingstamping dies for sheet metal forming. An ATP-sponsoredworkshop (“The Road Ahead,” June 20–22, 2000) identifiedproduction of working die sets as the main obstacle toreducing the time between accepting a new design andactual production of parts. This is also the largest singlecost (besides labor) in car production. To benefit fromweight savings enabled by new high-strength steelsand aluminum alloys, a whole new level of formabilitymeasurement methods, models, and data is needed,together with a better understanding of the physicsbehind metal deformation. MSEL is working withU.S. automakers and their suppliers to fill this need.

Polymer ProcessingPolymers have become ubiquitous in the modern

economy because of their processability, high functionality,and low cost. However, these materials can exhibitcomplex and sometimes catastrophic responses to theforces imposed during manufacturing, thereby limitingprocessing rates and the ability to predict ultimateproperties. The focus of our polymer research is onmicrofluidics and microscale processing, modeling ofprocessing instabilities, and on-line process monitoringof polymers. Our unique extrusion visualization facilitycombines in-line microscopy and light scattering for thestudy of polymer blends, extrusion instabilities, and theaction of additives. These measurements are carried outin close collaboration with interested industrial partners.

Contact: Carol A. Handwerker (Metallurgy Division)

17


Increased computing power coupled with finiteelement modeling methods (FEM) has broughtspringback-compensated die design withinreach of the automotive industry. In developingthis technology, industry has determined thatbetter measurement methods and an improvedunderstanding of the influence of large plasticstrains on the properties and behavior of materialsduring springback is required. This project seeksto develop the understanding, measurements,and data to meet this need.

Richard E. Ricker

AU.S. Council for Automotive Research (USCAR) Consortium on Springback Prediction has

developed finite element springback prediction modelsto reduce die design and tryout costs. However,these models consistently underpredict springback.Three alloy properties could vary with plastic strainand contribute to prediction errors: (1) elastic modulus,(2) anelastic modulus, and (3) strain-dependentanelastic behavior (reverse creep or plastic hysteresis).Currently, FEM modelers compensate for theseerrors with arbitrary constants estimated from trialsor prior experience. The objectives of this study are todetermine the origin of these errors, evaluate relevantmeasurement methods, and provide materials datathat improves FEM springback prediction codes.

In FY 2003, we demonstrated that the large biaxialstrains typical of forming can significantly reduce theelastic modulus of Al alloys (18 %) and that texturechanges cannot explain this behavior. In FY 2004,finite element models were used to analyze the resultsof NIST 3-point bend springback measurements.Combining these results with modulus data obtained inFY 2003, it was determined that the strain and elasticmodulus variations are too small to measure in theseexperiments. During FY 2004 these results were

presented at industry workshops (USCAR Consortium,Deep Drawing Research Group) and at metallurgy andphysics conferences (ASM, TMS, APS).

Also in FY 2004, Mg and Mg alloys were studiedto determine how widespread this anomalous modulusbehavior is and to gain additional clues as to its origin.Mg is an extremely lightweight HCP metal that wouldbe a prime candidate for automotive applications ifnot for its poor formability and high chemical reactivity.Biaxial strain experiments did not produce a significantchange in the elastic modulus, but the limited ductilityprevented measurements at the high strains required tounambiguously reduce the modulus in Al alloys. Dynamicmodulus analysis (DMA) found that plastic strain produceda broad peak in the imaginary component of the complexmodulus. Previous researchers attributed this peak tograin boundary sliding, but plastic strain and twinning werefound to be responsible for this peak which analysisindicated contains 4 subpeaks (Figure 1).

Sheet Metal Forming for Automotive Applications:Anelasticity and Springback Prediction

In a separate study, niobium is a BCC metal thatDOE’s Jefferson Laboratory (JL) uses to fabricatesuperconducting structures by stamping and formingoperations similar to those used by the automotiveindustry to form BCC steel. The Jefferson Laboratoryhas funded NIST to investigate the catastrophic effectof impurities on the forming behavior of high purity Nb.DMA experiments detected variations in the interstitialimpurity content of this high purity metal (Figure 2),which will now be correlated with forming behavior.

Contributors and Collaborators

S.W. Banovic, R.J. Fields, L. Ma, D.J. Pitchure(Metallurgy Division, NIST); D. Dayan, A. Munitz(NRC-Negev); V. Luzin (NCNR); G. Myneni (JL);E. Chu (Alcoa); USCAR Springback Consortium(Alcoa, DaimlerChrysler, Ford, GM, LSTC, U.S. Steel)Figure 1: DMA peak in deformed Mg alloy AZ31.

Figure 2: DMA peaks from two lots of Nb compared to thecalculated locations for O, C, and N.

18

Other topics explored during this fiscal year include:1) statistical analyses of surface roughness data aimedat determining what roughness measures would be mostuseful for industry; 2) EBSD/SLCM studies of AlMgbinary alloys as a function of strain; and 3) exploratorywork on new methods for measuring friction underconditions designed to emulate actual forming operations.


J. Liu, T. Foecke, R.E. Ricker, L.E. Levine (MetallurgyDivision, NIST); M.D. Vaudin (Ceramics Division, NIST);T. Gnäupel–Herold, (NIST Center for Neutron Research);J.B. Hubbard (CSTL, NIST); R. Reno, E. Moore (UMBC)


The existing data, measurement methods, andbasic understanding of metallurgical factors thatinfluence friction, tearing, and surface finishduring sheet metal fabrication are insufficient tomeet the predictive modeling requirements of theautomotive industry. This project addresses theseneeds by exploring the microstructural originsof the distribution of slip, surface roughening,and strain localization during plastic straining.The primary focus of these investigations isthe relationships between the initial materialcharacteristics and the deformation behaviorof Al and Fe base sheet materials.

Mark R. Stoudt and Stephen W. Banovic

Replacement of conventional steel sheet by aluminumalloys and high-strength low-alloy steels would

significantly reduce automobile weight and increase fuelefficiency. However, wide spread application of thesematerials by the automotive industry is limited primarilydue to formability issues and deficiencies in materialdatabases/constitutive laws affecting finite elementmodeling. This project examines the underlying structureproperty relationships associated with formability in anattempt to improve the numerical simulations that predictdeformation behavior of materials.

Microstructural Origins of Surface Rougheningand Strain Localizations

In FY 2004, several experimental techniques wereused together to explore the relationships between strain,local crystallographic texture and surface roughening ofaluminum alloys. Figure 1a is an electron backscattereddiffraction (EBSD) map showing the variations in surfacegrain orientation on a 6xxx series aluminum sample afterapplying a uniaxial strain of 0.08. As reflected in the figureinset, the surfaces of grains 1, 2 and 3 have <001>, <101>and <111> orientations, respectively. A 3D topographof the same region obtained with a scanning laser confocalmicroscope (SLCM) is shown directly below Figure 1ain Figure 1b. The labels indicate the same grains inboth figures. There are clear differences in the slip bandstructures of the grains. The roughness profiles shownin Figure 2 were acquired from the SLCM data along thewhite lines in Figure 1b. The different crystallographicorientations in the three grains produced distinctly differentroughness characters for the same level of macroscopicuniaxial strain. New statistical analysis techniques arecurrently being developed to investigate correlationsin these data.

Figure 1: Surface Characterization of 6xxx series aluminumdeformed to 0.08 uniaxial strain. a) EBSD generated grainorientation map; b) SLCM surface topograph of same region.

Figure 2: Roughness profiles obtained from the three individualgrain orientations shown in Figure 1.

19


Predicting the final shape of a part producedusing a complex fabrication process can beextremely challenging. This is particularly truewhen microstructure and residual stresses playsignificant roles. Understanding the interplay ofthese factors in industrially important test cases isthe goal of this project. Work on this project wascompleted this fiscal year.

Lyle E. Levine and Hank Prask (856)

Diffraction provides a powerful means of veryaccurately measuring both microstructure and

mechanical behavior in a way that provides insightsimultaneously into both, and offers particularopportunities for understanding plasticity. This projectcollaborated with the developers of the NIST ObjectOriented Finite Element (OOF) program to implementa plasticity model into OOF and devise criticalexperimental and computational benchmarks to validateOOF using the NCNR neutron diffractometer tomeasure residual stress, texture and elastic properties.Conventional X-ray sources were also used wherestress states measured at or near surfaces wereimportant for understanding the behavior of materialsor to compare with predictions. Synchrotron X-rayswere used to deal with subsurface residual stressesin the thin, highly textured samples that result fromsimulated forming operations. Specifically, thefollowing tasks were part of this project:

Experimentally validate calculation of the importantrole played by residual stresses on high-precisionmachining and forming operations (includescollaborative contributions from MEL, ALCOA andBoeing Corporation);

Provide technical assistance and validation testingto CTCMS staff engaged in adding 3D plasticityto OOF;

Develop the ability to characterize near-surfaceresidual stresses; and

Further develop neutron and synchrotron radiationsources to provide high-resolution data needed torelate microstructure, residual stresses, and theireffect on springback.

Overall Project Accomplishments Validation of Distortion Prediction Method: In

2002, we validated residual stress measurement and

prediction methods used by industry. In 2003 and2004, we used the NCNR to measure residual stressdistributions within 7000 series aluminum bars andused high-speed machining to produce test parts.The shapes of the final parts were measured using acontact measuring machine and ALCOA will comparethese shapes to their model predictions.

Test of OOF: In FY 2003, textured polycrystallinesamples were tested for elastic modulus dependenceon texture. These results were compared withpredictions using Electron Backscatter Diffractionand OOF.

Measurement of residual stresses in standardspringback cups in FY 2002: Measurements wereprovided for the project on Standard Tests and Datafor Sheet Metal Formability. In addition, analyseswere done to understand the origin of these stresses.

Incorporation of plasticity: Project personnelworked with CTCMS staff to provide algorithmsand techniques for incorporating plasticity into OOF.

Although this project is ending this fiscal year,outgrowths of the work on residual stressmeasurements of deformed sheet metal and plasticitymodeling for OOF continue in other projects.


R. Fields, R. deWit (Metallurgy Division, NIST);T. Gnäupel–Herold, V. Luzin (NIST Center for NeutronResearch); R. Ivester, R. Polvani (ManufacturingMetrology Division, NIST); D. Bowden (Boeing);E. Chu (ALCOA); R. Reno (University of Maryland)

Figure 1: Aluminum parts machined to test shape changescaused by relaxation of residual stresses.

Plasticity, Fabrication Processes, and Performance

20

To meet the PNGV goals for fuel efficiency, theU.S. automotive industry is moving to lighter,high-strength materials for auto bodies. NISThas surveyed industry and found that providingdesigners with accurate material properties, andmethods to incorporate them into finite elementmodels of sheet metal forming dies, is a criticalneed for the US auto industry. This project seeksto develop new standard tests and metrology toaccurately determine sheet metal mechanicalresponse under forming conditions.

Tim Foecke and Mark Iadicola

For the U.S. automotive industry to transition tonew materials for formed sheet metal parts, they

must be able to mechanically characterize the startingmaterials under realistic forming conditions, and inputthis information into die design models. The MetallurgyDivision is conducting research to develop two sheetmetal formability tests, along with associated metrology,that can be standardized and used by industry.

Springback is the elastic shape change to a partassociated with the residual stresses that develop duringthe stamping process. This shape change complicatesassembly and accurate fit-up; thus, the automotiveindustry has a strong desire to be able to either avoid it,or at least be able to predict its magnitude and designdies to account for it. The proposed test for springbackconsists of splitting open a ring cut from the sidewallof a deep drawn cup.

This year, the springback cup test was introducedto ASTM at its national meeting in Florida, leading toindustry’s proposal to form a new ASTM committeeto explore standardization of this test. A crucial stepin this process has been taken by identifying severalindustrial partners that are willing to participate.Additional participants from both the producer andend-user sides of the sheet metal industry are nowbeing sought for this new committee. The initialrobustness data set generated at NIST last yearforms a solid basis for the design of a round-robintesting matrix to be implemented as part of thestandardization process.

Last year, an x-ray stress measuring system wasinstalled on the sheet metal formability station. Thisunique instrumentation allows for the direct, in situmeasurement of the stress in a given direction whilethe sample is under multiaxial load. This year we madesignificant progress in mapping the multiaxial stress–strainsurface for three aluminum sheet alloys of interest to theautomotive industry: 5182, 6111 and 6022. Typical datais shown in the Figure 1, where for the first time thebiaxial hardening exponent has been measured. In addition,both the flow stress and the hardening exponent, n, werefound to differ in the transverse and rolling directions ofthe sheet. These data will be used by industry to eliminateseveral assumptions in place that extrapolate estimatedbiaxial flow behavior based on a series of uniaxial tensiletests. This should lead to more accurate designs andcost savings.

This year, NIST was asked to generate materialsproperty data for simulations of prototype parts as part ofthe NUMISHEET 2005 conference. In addition, takingadvantage of our unique capabilities to measure stress ina forming part, another layer of complexity of the prototypepart simulation will require reproducing the measuredstresses at points of the part under the forming load.

Plans for FY 2005 involve completing measurementof the multiaxial stress–strain surfaces for the threeabove-mentioned aluminum alloys, as well as for aseries of steel alloys. This new project, in collaborationwith modelers at GM, will involve studying how theflow surface evolves with different types and amountsof multiaxial prestrain.


T. Gnäupel–Herold (NIST Center for NeutronResearch); M. Shi (USS); E. Chu (ALCOA); C. Xia(Ford); T. Stoughton, M. Wenner, C.T. Wang (GM);A. Andersson (Volvo); E. Schedin (Avestapolarit)

Standard Tests and Data for Sheet Metal Formability

Figure 1: Balanced biaxial flow stress for 5182 Al alloy intransverse (TD) and rolling (LD) directions.


21

The NIST Center for Theoretical andComputational Materials Science was foundedin 1994 in order to fulfill its three-fold mission:to investigate important problems in materialstheory and modeling with novel computationalapproaches; to create opportunities forcollaboration where CTCMS can make a positivedifference by virtue of its structure, focus, andpeople; and to develop powerful new tools formaterials theory and modeling and acceleratetheir integration into industrial research. TheCTCMS supports numerous materials theory andmodeling projects both within and externally to theMaterials Science and Engineering Laboratory.

James A. Warren and Benjamin P. Burton

The NIST Center for Theoretical and ComputationalMaterials Science (CTCMS) is a research program

addressing industry needs for theory and modeling toolsfor materials design and processing. The CTCMS is acenter of expertise in computational materials researchthat develops tools and techniques and fosterscollaborations.

CTCMS integrates ongoing research at variousinstitutions by forming temporary multidisciplinary andmulti-institutional research teams as required to attackkey materials issues of national importance. TheCTCMS has three principal activities, all operatinginteractively: planning, research, and technologytransfer. Workshops are held as the first step indefining technical research areas with significanttechnological impact, identifying team members, andbuilding the infrastructure for collaborative research.The CTCMS provides infrastructure and support forits members, including an interactive World Wide Webserver (www.ctcms.nist.gov) and modern computingand workshop facilities.

Current research areas include theory and simulationof a wide range of materials behavior, including phasetransformation kinetics and morphology, micromagnetics,composite materials, foams, microstructure anddynamics of disordered and partially ordered materials,complex fluids, materials reliability, reactive wetting,pattern formation, crystal growth, sintering, phasetransitions in biological systems, and solidification.Current CTCMS working groups include the following:

High-Throughput Analysis of MulticomponentMultiphase Diffusion Data.

The NIST Center for Theoretical and ComputationalMaterials Science


Phase Field Modeling Tools: The phase fieldmethod has become one of the most flexible andpowerful methods for predicting the evolution ofmaterials microstructure. This effort focuseson the development of both new applications forthis method and tools enabling the solution of thecomplex equations which emerge from these models.

Effective Hamiltonian Methods: Use of fundamental,quantum-mechanical descriptions to derive theproperties of matter is the ultimate goal of materialsmodeling. This effort uses electronic structurecalculations to derive the phase stability of alloys.

WWW Tools for Scientific Collaboration: CTCMSis working with information science specialists todevelop web-based tools for scientific collaboration.

Tools for Neutron Scattering Measurements:While neutron scattering has become a critical toolfor the probing of material structure and properties,interpretation of the results of experiments presentsa host of challenges for the scientist. This effortattempts to develop a better theoretical frameworkfor the interpretation of such experiments.

Object-oriented finite element modeling of compositematerials: This team of researchers is developing aset of object-oriented finite element modeling tools toimprove the characterization and property predictionof composite materials. Public domain software toolsare available at www.ctcms.nist.gov.

The CTCMS also hosts web pages with resources andtools in the following areas: an interactive, electroniclibrary of Green’s function and boundary element solution;accurate, standardized micromagnetics modeling tools;software tools to improve electronic packaging processes;and a tool to compute equilibrium crystal shapes.More detail can be found at www.ctcms.nist.gov.

Mechanisms for Collaboration with CTCMS

The CTCMS facilitates numerous interactions betweenindustry, academia, NIST, and other government andnational labs to apply materials theory and modeling tosolve U.S. industrial problems in materials design andprocessing. Researchers interested in joining existingefforts or starting new ones are encouraged to contactthe CTCMS. The CTCMS participates in the NationalResearch Council postdoctoral fellowship program andhosts short-term and long-term visitors.

22

A substantial increase in the use of aluminumalloys and high-strength steels in automobileswould greatly increase fuel efficiency. The primaryreason why this has not yet occurred is a lack ofaccurate deformation models for use in designingthe stamping dies. This project is developing aphysically based model of plastic deformationusing a combination of statistical physicsapproaches, atomistic modeling and advancedmeasurement techniques.

Lyle E. Levine

Plastic deformation of metals (as in cold rolling,stamping, drawing, and metal fatigue) is of great

importance to industries worldwide, and improvementsin the basic technology would have a significanteffect on the U.S. economy. Unfortunately, existingconstitutive equations cannot accurately predict thematerial behavior, and many tryout and redesign stepsare required. Another related difficulty is in the designof new alloys with improved formability characteristics.Currently, alloy design is done empirically with littleunderstanding of how the various constituents affectthe mechanical properties.

Addressing both of these issues, this project isfocused on developing constitutive laws based uponthe underlying physical processes that produce theobserved mechanical behaviors in metal alloys. Suchconstitutive laws are inherently multi-scale since theymust describe phenomena on length scales rangingfrom the atomistic all the way to the macroscopicstress–strain behavior of bulk material.

Since the underlying processes are extremely complex,most of our theory, modeling and experimental effortsover the past several years were concentrated onsingle-crystal pure Al as a model system. This workculminated in the development of the segment lengthdistribution (SLD) model, that describes how macroscopicdeformation arises from the statistical behavior of largenumbers of dislocations. The model correctly predictsthe flow stress (Figure 1), the linear behavior of stage IIhardening, the non-linear hardening in stage III and thedevelopment of slip lines and slip bands. The statisticalvariables, <nL> and N0, represent the density of primarydislocations and pinning points, respectively, in the sample.Alloying the Al changes the mechanical properties byaffecting these statistical variables. Thus, this criticalsurface is a universal surface for all Al alloys. The nextstep is to connect the evolution of the statistical variablesto the underlying physics for Al alloys using atomisticsimulations that accurately incorporate the effects ofalloy chemistry.

Quantum-mechanics based (ab initio) modeling isrequired to adequately describe the effects of chemistryin atomistic simulations. Unfortunately, such simulationsare extremely CPU intensive and thereby limited to afew hundred atoms. Atomistic modeling using classicalpotentials can handle much larger systems but failsfor large bond distortions and chemistry effects.A ground-breaking approach for directly coupling thetwo methods has been developed that allows the use oflarge-scale atomistic simulations with simple classicalpotentials while limiting the more accurate ab initioapproaches to critical regions. Preliminary calculationsof vacancy energies near dislocations are now underway.

Washington State University (WSU) is conductingexperimental validation tests of the SLD model usingchemi- and photo-emission techniques to study thetime dependence of new surface production duringdeformation. NIST researchers designed the UHVtensile stage and single-crystal growth facilities, andWSU paid for the fabrication with funding from DOE.

Finally, NIST researchers are supporting thedevelopment of supporting science by serving on theexecutive committee of the international conference“Dislocations 2004” and coorganizing symposia at theFall 2004 MRS meeting and at “Plasticity 2005.”


D. Pitchure, F. Tavazza (Metallurgy Division, NIST);A. Chaka (Physical & Chemical Properties Division,NIST); T. Dickinson, S. Langford, M. Cai (WSU)

Underlying Processes of Plastic Deformation in Metal Alloys

Figure 1: Universal flow surface (critical surface) for Al alloysat ≈ 15 % strain.


23

U.S. industry annually spends $200 billion onthe machining of metal parts, and potentialcost savings are a driver towards high-speedmachining. Traditional knowledge-basedapproaches have not been effective in improvingthe efficiency of this rapidly developing field:One recent study showed that industry chosethe correct machining parameters less than halfof the time. Through this project, NIST willprovide measurement capabilities, materialsdata, and assessment of constitutive laws anddeformation models, which will provide users witha capability of producing a first part correct.

Carelyn E. Campbell

To improve the efficiency of high-speed machiningand tooling, industry has employed finite element

modeling (FEM) to predict correct tooling andmachining parameters. However, FEM has had limitedsuccess as much of the needed material property datafor high speed machining processes, which involvelarge strains (>>1), high-strain rates (up to 106 s–1),and high-heating rates (greater than 105 ºC/s), isinsufficient. This project emphasizes two complementaryareas: (1) the development of fundamental data ondynamic material behavior at high strain and heatingrates and the assessment of the constitutive lawsused to model this behavior; (2) the modeling andcharacterization of the microstructural changes thatoccur during the rapid deformation and rapid heatingin real machining operations. Residual stresses,responsible for part distortion after machining, are beingmeasured at the NIST Center for Neutron Research.

The first research effort is focused on obtainingthe materials data and assessing whether currentlyused constitutive laws are appropriate. A dynamicmaterial testing facility using a pulse-heated Kolskybar has been developed at NIST and is being used tomeasure dynamic stress–strain data at high-strain rates(approximately 500 s–1 to 105 s–1) with heating ratesranging from 103 Ks–1 to 1 Ks–1. Measurementson 1045 steel showed significant differences in thestress–strain behavior depending on the heating andstrain rates (Figure 1). The two microstructuresshown in Figure 1 demonstrate the observedmicrostructural differences resulting from differentheating rates. Test 504 was rapidly heated andquenched, leaving no time for the pearlite to dissolveand reform. In contrast, test 509 was heated, held attemperature for 1.4 s, then quenched, leaving enough

time for the pearlite to dissolve and then re-crystallizeduring quenching, producing a finer grain size.

These results demonstrated that the non-time dependentconstitutive relationships used by industry to describethe stress–strain data for machining simulations are notcorrect. A time-dependence must be included in theconstitutive relationship to correctly model the materialbehavior during high-speed machining. Using the dynamicmaterial property data collected using the pulse-heatedKolsky bar, time-dependent constitutive laws will bedeveloped and then tested in machining simulations.

High Speed Machining

Figure 1: The effect of various heating and strain rates on boththe stress–strain behavior and the microstructure of 1045 steel.

The second area of research focuses oncharacterizing and modeling microstructural changes inthe deformation/hot zone of aluminum alloys duringmachining. Transmission electron microscopy (TEM)correlated with optical microscopy of the chip hasrevealed bands of re-crystallized grains. The worksurface is being characterized using microhardnessmeasurements, TEM, optical microscopy and residualstress measurements. A model based on thesemicrostructural characterizations and temperature vs.time data from thermal images will be used as input to amicrostructure model. The microstructure model,along with the time-dependent constitutive relationship,will be implemented into a finite element code formachining simulation.


D. Basak, L. Levine, R. Fields, R. deWit, W. Boettinger,L. Bendersky, T. Gnäupel–Herold, H. Prask (NIST/MSEL);R. Ivester, R. Rhorer, K. Jurrens, J. Soons, R. Polvani,E. Whitenton, M. Kennedy, B. Dutterer, G. Blessing(Manufacturing Engineering Laboratory, NIST); T. Burns(Information Technology Laboratory, NIST); H. Yoon(PL); M. Davies (U. North Carolina–Charlotte)


24

The availability of reliable materials data is thekey to successful design of new materials andmanufacturing processes. Many commercialprocesses are controlled by the thermodynamicand diffusion properties of the material.Thermodynamic and diffusion mobility databasesprovide an efficient method of storing the wealthof these data, and software tools allow the userto efficiently retrieve the needed information.

Ursula R. Kattner and Carelyn E. Campbell

The complexity of traditional thermodynamics anddiffusion kinetics prevented their direct application

to the design of complex materials and processes inthe past. By the same token, graphical representationof multicomponent systems is too complex, and storageof every single datum is inefficient due to the enormousamount of data. However, mathematical functions thatrepresent thermodynamic and diffusion properties ofthe phases permit efficient storage. Since these functionsare based on physical models, they further provide thepower of extrapolation of binary and ternary systems tohigher order systems. Software then allows calculationof the desired quantities. This approach is called theCALPHAD methodology. Unfortunately, previoussoftware was tailored to be efficient for expert users andwas difficult for the occasional, less-experienced user.

The databases and software developed in thisproject are designed to provide users with simple toolsto retrieve and disseminate information needed forefficient materials and process design.

Mechanisms for Delivery of Thermodynamic and Kinetic Data

Figure 1: Graphical user interface for solidification calculations.

A modern, user-friendly front end for user input hasbeen developed for solidification calculations (Figure 1).The results of the calculation are delivered to the userin graphical as well as in tabular form. This programis bundled with the NIST superalloy and solderthermodynamic databases, but has the capability towork with user-supplied databases as well. Results canbe used with a Mathematica script for DTA analysissimulation. The software, databases and scripts areavailable on the Metallurgy Division website(www.metallurgy.nist.gov).

Figure 2: Schematic of an optimization scheme to allow directinput of diffusion couple composition profiles.

Multicomponent, multiphase diffusion couples canbe simulated using NIST thermodynamic and diffusionmobility databases, and experimental and calculatedcomposition profiles can be compared. However,the complexity of such couples prevents experimentaldiffusion coefficients from being easily extracted.Current work focuses on methods to allow experimentalcomposition profiles to be directly inputted into a dataassessment process, so the composition profiles canbe directly related to the diffusion mobility parametersstored in the database (Figure 2).

Interactions developed through the “High ThroughputAnalysis of Multicomponent Multiphase Diffusion”workshop series have led to several software codes aimedto improve the analysis of multicomponent diffusioncouples based on combinatorial libraries. These codesand other resources are listed on the group’s website(www.ctcms.nist.gov).

Work continues to convert the NIST Diffusion DataCenter into an electronic searchable form. A preliminaryversion should be available by December 2004.


W.J. Boettinger (Metallurgy Division, NIST);A.R. Roosen (Center for Theoretical and ComputationalMaterials Science, NIST); J.-C. Zhao (GeneralElectric); L. Höglund (Thermo-Calc AB)


25

Many properties of structural and functionalmaterials depend on the distribution of composition,phases, grain orientations, internal surface stresses,and microstructure. These structures span lengthscales from nanometers to meters. Predictivemodels are needed to reduce the enormous costsand development times involved in designing andinserting new materials into products. Phase fieldmodels of solidification, coupled to stress, grainorientation, and solute and vacancy dynamics arebeing developed to address these needs.

James A. Warren

Modeling of microstructures produced by solidifi-cation, grain evolution, and stress, as well as other

processes, involves mathematical solution of equationsfor heat flow, fluid flow, current flow and/or solutediffusion. Boundary conditions on external surfaces reflectthe macroscopic processing conditions, while boundaryconditions at internal interfaces correspond to the liquidcrystal (grain) or grain–grain interfaces. These internalinterfaces are moving boundaries and require boundaryconditions with thermodynamic and kinetic character.

To deal with the complex interfacial shapes that developduring solidification, grain growth, and the application ofexternal stress or pressures, the phase field method hasbecome the technique of choice for computational materialsscientists. This approach often requires numericaltechniques to solve the model equations but readily dealswith complex interface shapes and topological changes.The research, conducted in collaboration with the PolymersDivision, is also supported by the NIST Center forTheoretical and Computational Materials Science.

This year, significant advances were made in theunderstanding of the formation of multi-grained(polycrystalline) materials. Polycrystals often arise byeither the impingement and growth of grains nucleated ina liquid (equiaxed grains), or via the nucleation of columnargrains on a surface. In collaboration with researchersin Japan and Hungary, we developed a phase field modelto study such growth. In a surprising development, thissame model has now elucidated a third mechanism ofpolycrystalline growth: growth front nucleation. Thismode of growth is manifested when new orientationsnucleate on the front of a growing crystal, yielding adensely branched morphology in which the structuralsymmetries that arise due to effects of surface energyanisotropy are disrupted by the nucleation process, yielding

an isotropic pattern. This model demonstrated that thistype of growth can be initiated by either static or dynamicheterogeneities in the solidifying system, solving a riddlethat has been of substantial interest to materials scientists.This work, published in Nature Materials, has attractedextensive academic and industrial interest.

Phase Field Modeling of Materials Systems:Simulation of Polycrystalline Growth

Figure 1: Two simulations of seaweed-like growth of an alloy.The left column shows the solute profile while the right columnshows grain orientation. The upper row is a single crystal whilethe lower is polycrystalline. Using such techniques, the potentialfor polycrystalline growth to disrupt the influence of surfaceenergy anisotropy (which gives dendritic growth) is demonstrated.

Although a remarkable amount of progress has beenmade, there are still physical phenomena that need to beproperly included in phase field models if engineeringimpact is desired. With this in mind, much effort hasbeen devoted to account for the effects of stress andvacancies on both solidification and interdiffusion. It isalso believed that our model for polycrystalline growthmay explain spherulitic growth as well as provide insightinto the observations of Bendersky and NIST coworkersof a stable glass phase in highly supercooled alloys.


W. Boettinger, J. Guyer, D. Wheeler (MetallurgyDivision, NIST); J. Douglas (Polymers Division,NIST); L. Granasy, T. Pusztai (RISSPO, Hungary)


26

Hardness is the primary test measurementused to determine and specify the mechanicalproperties of metal products. The MetallurgyDivision is engaged in all levels of standardsactivities to assist U.S. industry in makinghardness measurements compatible with othercountries around the world. These activitiesinclude the standardization of the nationalhardness scales, development of primary referencetransfer standards, leadership in national andinternational standards writing organizations,and interactions and comparisons with U.S.laboratories and the National MetrologyInstitutes of other countries.

Carlos Beauchamp and Sam Low

At the international level, we are leading the Working Group on Hardness (WGH) under the

International Committee for Weights and Measures(CIPM). The primary goal is to standardize hardnessmeasurements worldwide. As secretary of the WGH,we have led an effort this year to better define theRockwell hardness test procedure used by NationalMetrology Institutes (NMIs). Other activities includechairing the ASTM-International committee onIndentation Hardness Testing and heading the U.S.delegation to the ISO committee on hardness testingof metals, which oversees the development of therespective hardness test method standards publishedby those organizations; and as the Secretariat forthe International Organization of Legal Metrology(OIML) committee on hardness, we have revisedthe international requirements for regulatingRockwell hardness machines.

The primary task at the national level is tostandardize the U.S. national hardness scales and toprovide a means of transferring these scale valuesto industry. Currently, we are producing test blockStandard Reference Materials® (SRMs) for theRockwell, Vickers, and Knoop hardness scales, aswell as developing new reference standards. Twelvedifferent microhardness SRMs for Vickers and Knoophardness are now available. Fifteen units of a newsteel SRM are in the process of being certified at anominal hardness of 760 kgf/mm2 for each of theVickers and Knoop hardness scales. These SRMsare being certified at loads of 2.943 N, 4.905 N, and9.81 N. In addition, renewal of low inventory of twoof the Knoop Scale SRMs is underway. Ten units ofeach of the copper and nickel SRMs are being certifiedwith nominal hardness levels of 125 kgf/mm2 and

600 kgf/mm2, respectively, at loads of 0.245 N,0.490 N, and 0.981 N.

We also introduced two new SRMs for theRockwell B scale (HRB) to complement the threeSRM Rockwell C scale blocks currently available.The HRB scale is used for testing softer metals, such asaluminum, copper and brass. Work is also continuingto produce a new Rockwell hardness diamond indenterSRM, which has become feasible due to the successfulcompletion of a Small Business Innovative Research(SBIR) Phase-2 project with Gilmore Diamond Toolsto develop an improved method for manufacturinggeometrically correct Rockwell diamond indenters.

Other activities at the national level occurringthis year included: the assessment of commercialsecondary hardness calibration laboratories for theNIST National Voluntary Laboratory AccreditationProgram (NVLAP) providing direct linkage to the useof the NIST SRMs; the development of a NVLAPproficiency testing program for hardness calibrationlaboratories; and the development of indentation FiniteElement Analysis models that have been used to analyzethe effect of using different indenter materials forRockwell hardness tests in support of proposedrevisions to international test standards.


C. Johnson, J. Fink, D. Kelley, L. Ma, H. Gates(Metallurgy Division, NIST); J. Song (ManufacturingEngineering Laboratory, NIST); W. Liggett, Jr.,N. Zhang (Information Technology Laboratory, NIST);S. Doty, B. Belzer, D. Faison (National VoluntaryLaboratory Accreditation Program, NIST); W. Stiefel(Technology Services, NIST); M. Mihalec (GilmoreDiamond Tools, Providence, RI)

Hardness Standardization: Rockwell, Vickers, Knoop

Figure 1: Prototype of new steel microhardness SRM.


27

Nanometrology

in determining mechanical properties of thin filmsand nanostructures, focuses on developing traceablecalibration methodologies and standard test methods.We also use atomic force acoustic microscopy,surface acoustic wave spectroscopy, and Brillouin lightscattering to measure the mechanical properties ofthin films. In addition, we are developing micro- andnano-scale structures and test methods to measurestrength and fracture behavior of interfaces andmaterials having very small volumes.

The chemical and structural characterization andimaging utilize neutron and x-ray beam lines at threefacilities: the NCNR; the National Synchrotron LightSource at Brookhaven National Laboratory; andthe Advanced Photon Source at Argonne NationalLaboratory. Innovative scattering and spectroscopymethods are advancing our ability to obtain a widerange of chemical and structural information at thenanoscale, including chemical bond identification andorientation, polyelectrolyte dynamics, and equilibriumstructures. In collaboration with three other NISTlaboratories, we are developing electron microscopyand spectroscopy instrumentation for quantitative,3D chemical imaging at the nanoscale. Othercharacterization projects include work on gradientreference specimens for the calibration of advancedscanning probe microscopy, and the application ofcarbon nanotubes as physical probes of cell membranes.

Efforts in the fabrication and monitoring ofnanoscale processes and events include the study ofelectrochemical and microfluidic methods for fabricatingnanostructures, novel approaches to nanocalorimetryfor the study of interfacial reactions, in situ observationsof nanoparticle and nanotube dispersion and alignment,and advanced instrumentation for nanotribologyexperiments.

Finally, we have extensive efforts in the theory,modeling, and prediction of material properties andbehavior extending from nanoscale to macroscaledimensions. Modeling efforts include large-scale finiteelement methods, multiscale Green’s function methods,classical atomistic simulations, first principles, andquantum mechanical calculations using densityfunctional theory. Often, several modeling methodsmust be combined into one study to accurately describethe material behavior; thus, we pay great attention tothe correct interfacing between models operating atdifferent length scales, to ensure that our modelsproperly capture the physics of both componentsand total systems.

Contact: Gery R. Stafford or John E. Bonevich(Metallurgy Division)

The burgeoning field of nanomaterials extendsacross the full range of traditional material classes,including all forms of metals, polymers, and ceramics.No previous materials technology has shown soprodigiously a potential for concurrent advances inresearch and industry as does the field of nanomaterialsin mechanical devices, electronic, magnetic, and opticalcomponents, quantum computing, tissue engineeringand other biotechnologies, and as-yet unanticipatedexploitations of as-yet undiscovered novel propertiesof nanoscale assemblies of particles. Already, thereis growing excitement surrounding the ability ofsome molecules or particles to self-assemble atthe nanoscale to form new materials with unusualproperties. Nanometrology, i.e., the ability to conductmeasurements at these dimensions, to characterize thematerials, and to elucidate the structure and nature ofthese new and novel assemblies, is a requisite andfundamental cornerstone that must be establishedsecurely if this technology is to flourish.

NIST is uniquely positioned to lead the developmentof the measurement methods, instrumentation, standards,and reference materials that, together, will form themetrological infrastructure essential to the successof nanotechnology.

The MSEL Nanometrology Program incorporatesbasic measurement metrologies to determine materialproperties, process monitoring at the nanoscale,nanomanufacturing and fabrication techniques, andstructural characterization and analysis techniquessuch as advanced imaging and multiscale modeling.The Program comprises 22 projects in the Ceramics,Materials Reliability, Metallurgy, and PolymersDivisions, and includes structural characterizationusing neutron scattering at the NIST Center forNeutron Research (NCNR). The projects cover awide range of measurement and characterizationmethods grouped into the areas of mechanicalproperty measurement, chemical and structuralcharacterization and imaging, fabrication and monitoringof nanoprocesses and events, and modeling of nanoscaleproperties. In each area, we work to advance basicmeasurement capabilities and lead the intercomparison,standardization, and calibration of test methods. Thenewly completed Advanced Measurement Laboratoryat the NIST Gaithersburg site provides an incomparableenvironment for accurate nanoscale metrology.

In the area of mechanical property measurement,we are developing and standardizing techniques fordetermining nanoscale elastic properties (elastic moduli,Poison’s ratio, and internal stress), plastic deformation,density, adhesion, friction, stiction, and tribologicalbehavior. Work in nanoindentation, used extensively

Program Overview

28

Nanometrology

Electron microscopy is used to characterize thestructure and composition of materials at thenanometer scale to better understand and improvetheir properties. New measurement techniques inelectron microscopy are being developed andapplied to materials science research. The MSELElectron Microscopy Facility primarily serves theMetallurgy, Ceramics, and Polymers Divisions aswell as other NIST staff and outside collaborativeresearch efforts.

John E. Bonevich

Atomic-scale structure and compositional characterization of materials can lend crucial

insights to the control of their properties. For instance,direct observation of local structures by transmissionelectron microscopy (TEM) provides importantfeedback to the optimization of crystal growth andprocessing techniques. Various characteristics may beobserved such as crystal structure and orientation, grainsize and morphology, defects such as stacking faults,twins, grain boundaries and second phase particles.Structure, composition and internal defect structures,as well as the atomic structure of surfaces andinterfaces, can provide powerful knowledge forengineering and understanding materials.

The MSEL Electron Microscopy Facility consists oftwo transmission electron microscopes, three scanningelectron microscopes, a specimen preparation laboratory,and an image analysis/computational laboratory. TheJEM3010 TEM can resolve atomic structures and employsan energy selecting imaging filter (IF) and X-ray detector(EDS) for analytical characterization of thin foil specimens.The S-4700-II FE-SEM employs electron backscattereddiffraction/phase identification (EBSD) and EDS systemsto characterize the crystallographic texture andcomposition of materials.

Nanoscale Characterization by Electron Microscopy

Figure 1: MgO cubes with surfaces decorated by small goldparticles. These particle ensembles are used in model studies of3D chemical imaging at the nanoscale (electron tomography).

Figure 2: Confined Si single electron transistor device.

Highlights from the EM Facility for FY2004 include: A new, high-sensitivity EBSD CCD camera

which acquires in excess of 70 patterns/second,was installed on the FE-SEM.

3D Chemical Imaging at the Nanoscale, a collaborativeproject with CSTL and PL on tomographiccharacterization of materials, was initiated (Figure 1).

Research collaboration with the SemiconductorElectronics and Electricity Divisions (EEEL) hascharacterized quantum effects in confined Si devices(Figure 2).

Sub-100 nm Ag interconnects formed bysuperconformal electrochemical deposition wascharacterized.


D. Josell, T. Moffat, L. Bendersky (MetallurgyDivision, NIST); I. Levin, B. Hockey (Ceramics Division,NIST); J.H.J. Scott (Surfaces & Microanalysis ScienceDivision, NIST); Z. Levine (Optoelectronics Division,NIST); E. Vogel (Semiconductor Electronics Division,NIST)

29

Nanometrology

Novel structures and devices can be formedthrough template electrodeposition. The ultimatepattern resolution is determined by the size andpacking of molecules that comprise the templateand their ability to inhibit or catalyze variouselectrochemical reactions. In the past year, theeffect of molecular functionality on the metaldeposition process has been explored.

Thomas P. Moffat and Michael J. Fasolka

Over the last decade, electrochemical processing has been undergoing a renaissance with the

fabrication of new materials and novel microstructures.The development of new measurement techniques andmetrological tools for studying controlled growth atfine length scales is the primary objective of this effort.

2. The role of molecular functionality of the surfactantand robust design rules; and

3. The anisotropy induced in the electrocrystallizationreaction rate for a given surfactant.

As an example, an alkanethiol monolayer film caneither block or accelerate metal deposition dependingon the molecular functionality of its terminal group.A dynamic range spanning several orders of magnitudeis possible depending on the system in question (Figures1 and 2). Based on these measurements, selective metaldeposition may be obtained by patterning a substratewith the appropriate molecule.

Nanostructure Fabrication Processes:Patterned Electrodeposition by Surfactant-Mediated Growth

Figure 1: Influence of small changes in molecular structure,i.e., SO3

– vs. CH3, on the rate of copper deposition.

Figure 3: Spatially patterned monolayer films used to controlnucleation and anisotropic growth of copper crystals.

Through contact printing, higher resolution patterningis possible, along with some interesting opportunities forhigh-throughput combinatorial research. By varying thesurface chemistry, growth of isotropic or faceted crystalshas been shown to be possible (Figure 3).


D. Josell, J. Mallett, W.F. Egelhoff (MetallurgyDivision, NIST); M. Walker, L. Richter (CSTL, NIST)

Figure 2: Copper deposition on an alknethiol derivatized goldsubstrate only occurs in regions bearing –SO3

– terminatedmolecules.

The three key metrology issues addressed are:1. The rate differentiation accessible via surfactant

mediated metal film growth;

30

Nanometrology

The microelectronics community useselectrodeposition to produce solderable surfacefinishes, magnetic recording media, and copperinterconnections in printed circuit boards andintegrated circuits. These films tend to developsizable mechanical stresses that can lead to loss ofadhesion and the generation of bulk and surfacedefects. This project focuses on the measurementof these stresses which should enable thedevelopment of effective mitigation strategies.

Gery R. Stafford and Ole Kongstein

Electrodeposited films tend to develop sizablemechanical stresses during deposition due to the

nucleation and growth process or solution additives andalloying elements. We have established an optical benchdedicated to the in situ measurement of growth andresidual stress during electrodeposition using the wafercurvature method. In one approach, a substrate ofborosilicate glass is evaporated with 250 nm of gold.The force exerted on the cantilever by the electrodepositcauses curvature of the cantilever, which is monitoredduring electrodeposition. Forces on the order of0.03 N/m can be resolved during film deposition.

negative deposition potentials. As the deposit thickens,the average film stress decreases and becomescompressive in deposits formed at small depositionoverpotentials. In the physical vapor deposition literature,this compressive stress has been attributed to thenon-equilibrium concentration of mobile ad-atoms onthe surface that are driven into the grain boundaries.

Nanostructure Fabrication Processes:Thin Film Stress Measurements

Figure 1: Average film stress during copper deposition.

Figure 1 shows the average in-plane stress associatedwith the deposition of copper from an additive-free sulfateelectrolyte. The rapid rise in tensile stress (within thefirst 20 nm) and its dependence on deposition potentialare consistent with nuclei coalescence and grain boundaryformation. The highest tensile stresses are associatedwith high nucleation densities which are obtained at more

Figure 2: Force/ width exerted on the cantilever during theelectrochemical processing of Sn on Cu.

We have also examined the force exerted onto a coppercantilever electrode during the deposition of matte tin (Sn),Figure 2. This is a particularly important system sincethe growth of tin whiskers, known to produce electricalshort circuits and device failure, has been attributed toboth the residual stress in the electrodeposit as well asthat generated from intermetallic formation at the Sn-Cuinterface. The force curves show the following features:a tensile to compressive transition during deposition, asignificant tensile relaxation when plating is discontinued,the development of tensile stress while the deposit is atopen circuit, and a residual tensile stress after the depositis electrochemically dissolved. The two latter featuresare attributed to the formation of the Cu6Sn5 intermetallic(IMC) at the Sn-Cu interface. A tensile stress is generatedsince the IMC has approximately 6 % smaller volume thana rule of mixture combination of the Sn and Cu reactants.Calculating the IMC thickness from the difference in Sndeposited and stripped, a nominal stress in the intermetallicwas estimated to be 1.2 GPa after only one hour. Futurework will focus on determining to what degree thisincreases the compressive stress in the Sn electrodeposit.


W. Boettinger, U. Bertocci (Metallurgy Division, NIST)

31

Nanometrology

Knowledge of mechanical behavior is criticalwhen designing for device performance andreliability, even for “non-mechanical” systems.However, nanoscale mechanical behavior(including failure) is inherently difficult to measureaccurately, and existing modeling tools are onlyqualitative at best. We are developing modelingtechniques that provide quantitative predictionsand are validating these results experimentally.

Lyle E. Levine and Anne M. Chaka (838)

Mechanics at the nanoscale is inherently difficult tomodel accurately. Finite element modeling (FEM)

can effectively capture the elastic behavior of macroscopicstructures but includes no accurate failure criteriasince this depends upon atomic-scale behavior.Classical atomistic simulations can handle enough atoms(millions to billions) to model such events, but thesepotentials become inaccurate for large strains and theycannot effectively handle chemistry. Quantum-mechanics-based simulations using density functional theory (DFT)are extremely accurate and handle the chemistry exactly,but such simulations are so CPU intensive that theycan handle only a few hundred atoms. A combinationof all three modeling techniques is required toaccurately model device behavior at the nanoscale.

the predicted elastic displacement fields to generateboundary conditions and initial atom positions foran atomistic simulation using classical potentials.

The use of classical potentials in a large simulation cellallows the correct propagation of the long-range stressesto the critical regions where bond distortions are large orwhere chemistry effects need to be explored. In thesecritical regions, the techniques embed a DFT simulation.Iteratively, the critical region is relaxed using DFT, andthe classical cell is relaxed using a Monte-Carlo algorithm.The first application of this hybrid technique was todetermine the vacancy formation energy in aluminum asa function of distance (at a fixed angle) away from anedge dislocation. The simulation geometry is shown inFigure 2. The boundary conditions and initial atomicpositions were calculated from the known elasticdisplacement field of an edge dislocation.

Nanomechanics: Coupling Modeling with Experiments

Figure 1: FEM model of a rigid 100 nm diameter sphereindenting an Al sample to a depth of 10 nm.

Over the past year, we have developed techniquesto handle such multiscale modeling for quasistaticapplications. At the macroscale, FEM is used to simulatethe elastic behavior of a nanomechanical system. Figure 1shows an example in which an Al sample is being indentedby a rigid 100 nm diameter sphere. The indenter andthree of the sample quadrants have been removed tohighlight the resulting Von Mises stress distribution afterindenting 10 nm. The FEM mesh is fine enough to use

Figure 2: Diagram of a hybrid simulation for obtaining thevacancy formation energy at different positions relative to anedge dislocation. The large box represents the EAM cell inwhich the DFT region containing the vacancy was embedded.

After an initial dislocation nucleation event,nanoindentation progresses through the complexevolution of dislocation structures. For example, theraised lip around the indent is produced by largenumbers of dislocations exiting the surface. We areworking on modeling the early stages of this processusing 3D dislocation dynamics and assuming a randomdistribution of dislocation sources in the sample.

Finally, connection to experimental measurementsrequires careful force calibration of the indenter andcalibrated atomic force microscopy measurements (AFM)of the indenter tip. These calibrations are mostly complete,and the AFM data will be used to generate a FEM meshfor the simulations. Bulk single-crystal copper samplesare being cut and polished using non-contact chemicalmethods to minimize the dislocation density.


S.M. Khan, G. Levi, L. Ma, F. Tavazza (MetallurgyDivision, NIST); B. Hockey, R. Machado, D. Smith,R. Wagner, D. Xiang (Ceramics Division, NIST)

32

Program Overview

and enamel. This approach will ultimately lead tomaterials with improved durability, toughness, andadhesion to contiguous tooth structure. We alsocollaborate with the ADAF to develop metrologyfor the biocompatibility of synthetic bone grafts.

In this era of interdisciplinary research,we provide an added dimension. By taking aphysical/mechanical approach to how cells function,respond, and remodel in interaction with syntheticmaterials, we provide skill sets typically absent in thebiomedical community. Mechanical properties issuesalso arise when considering synthetic bone graftsand tissue engineering scaffolds. Complementingtraditional bulk mechanical property measurements,combinatorial approaches are being developed toidentify compositions and surface features thataffect properties such as biocompatibility andmechanical durability.

Our mechanical property metrology extends furtherto biological systems that span the range from individualneurons and muscle cells to complete pulmonaryarteries. This necessitates the development of uniquemechanical testing platforms and application of amaterials science approach to understanding integratedproperties. Recently, we have developed a bioreactorcapable of applying biaxial stresses and allowingmonitoring of the stress and strain of a two-dimensionalscaffold sheet during tissue growth.

Fundamental to the Biomaterials program isrecognition of the need for an integrated systemsapproach. Collaborations among and between projectteams are critical to progress against the ambitiousgoals of this program.

Contact: Eric J. Amis (Polymers Division),Lyle E. Levine (Metallurgy Division)

New materials and devices are radically changingthe treatment of injury and disease, yet it is clear thatwithin this rapidly evolving segment of the materialsindustry, a basic measurement infrastructure doesnot exist. The Biomaterials Program developsmeasurement methods, standards, and fundamentalscientific understanding at the interface between thematerials and biological sciences. For the health careindustry, we focus on dental and medical sectors thatapply synthetic materials for replacement, restoration,and regeneration of damaged or diseased tissue.Three primary foci exist within this program:biocompatibility, biomaterials characterization, andmaterials measurements applied to biological systems.

Whether the medical issue involves implantinga hip- or knee-joint prosthesis, a synthetic bone graft,or a tissue engineering scaffold into the human body,one primary issue is biocompatibility. Using ourexpertise in materials science, we have developedsuitable Reference Materials (RM) for investigatingbiocompatibility and implant suitability. Research hasfocused on measuring cellular response to powdersand bulk materials that are candidates for implants;recently, we produced a realistic wear particle StandardReference Material (SRM® 2880) for bioactivity testing.

Work on quantitative methods of biomaterialscharacterization includes assays for adhesion,viability, proliferation, and differentiation of bonecells, 3-dimensional structural/functional imagingof tissue in-growth, and biochemical assays toquantify inflammatory responses to synthetic materials.The focus of this effort is bridging the gap betweenfundamental knowledge and the product developmentneeds in industry. For example, in collaboration withthe Chemical Science and Technology Laboratory,we are developing measurement methodologies andreference materials to assess interactions in complexsystems of living cells with synthetic materials.The expected outcome of this work includes referencesubstrates that induce specific cellular responses,and engineered DNA vectors to act as fluorescentreporters of cellular responses.

Another example of our effort to bridge this gapis our collaboration with the dental industry, whichis primarily composed of small manufacturers withlimited R&D capability. Collaborations with theAmerican Dental Association Foundation (ADAF)develop improved materials and materials measurementstechniques, patent and license these inventions, and,most importantly, provide a technical foundation.Research focuses on improved understanding of thesynergistic interaction of the phases of polymer-basedcomposites and the mechanisms of adhesion to dentin

Biomaterials

33

Biomaterials

Modeling the behavior of complex materialsrequires detailed knowledge of the underlyingthree-dimensional microstructure. Ultra-small-angleX-ray scattering (USAXS) imaging is a new classof synchrotron X-ray imaging techniques that usesUSAXS as the contrast mechanism. The techniqueprovides imaging and statistical data that cannotbe obtained using any other experimental methods.It has now been applied to characterize artificialtissue scaffolds and human cartilage in 3D.

Lyle E. Levine

USAXS imaging was developed by NIST researchersand was first demonstrated in May 2000. Advantages

over existing X-ray imaging techniques are its inherentlyhigher contrast and its USAXS-derived ability to providequantitative data on object shapes and size distributions.USAXS imaging is applicable to a wide range of materialsystems including metals, ceramics, polymers, compositesand biological materials. Over the past year, work hasconcentrated on two main areas. First, the final hardwarecomponents were installed, greatly improving the USAXSand USAXS imaging capabilities, and, second, a wide rangeof material systems, including biological ones, havebeen explored to determine where USAXS imagingcan have the greatest impact.

The USAXS imaging experiments are conducted onthe UNICAT sector 33 insertion-device beamline at theArgonne National Laboratory Advanced Photon Source(APS). A high-intensity, nearly-parallel, monochromaticX-ray beam passes through entrance slits that definethe size, shape and position of the beam. The angulardivergence in the X-ray beam is reduced by using multipleBragg reflections on <111> Si crystals referred to asthe collimator. Within the sample, local density variationsfrom the microstructure produce X-ray scattering atsmall angles. The X-rays leaving the sample are thenangle-filtered using a pair of <111> Si crystals referredto as the analyzer. The only X-rays from the sample thatcan pass through the analyzer are those scattered by themicrostructure at a specific angle, where this angle canbe selected by rotating the analyzer. Images are thenformed by either exposing nuclear emulsion plates orusing the NIST high-resolution X-ray camera systemthat was completed in FY 2003.

During FY 2004, the USAXS imaging instrumentwas declared fully operational after installation andtesting of the last major hardware component, a newcollimator system in which the surfaces of the two Sicrystals were chemo-mechanically polished to minimize

image aberrations. The new collimator was designed,built, and paid for by the APS and meets all of ourspecifications. Only minor problems were encounteredduring testing, and the required modifications arenearly completed.

USAXS and USAXS Imaging of Biomaterials

Figure 1: Identical sample volumes of an artificial tissue scaffoldimaged at scattering vectors of a) q = 0 Å–1 and b) q = 0.0005 Å–1.

Samples studied this year include artificial tissuescaffolds composed of porous poly-caprolactone (PCL),human cartilage, several metal alloys and a composite ofcarbon black in poly-methylmethacrylate. Figure 1 showsimages of a PCL sample with 50 % porosity that wascultured with osteoblasts for 28 days. Figure 1a wastaken with a 0º scattering angle (radiographic) and showsthe overall structure of the scaffold. Figure 1b, fromthe same sample volume, was acquired at a scatteringvector of q = 0.0005 Å–1 and shows the distribution ofosteoblasts on the scaffold surfaces. Stereopairs wereused to examine the cell positions in 3D. A full rotationsequence was acquired at q = 0 Å–1, and a tomographicreconstruction produced a complete 3D description ofthe scaffold structure.

A detailed paper describing the USAXS imaginginstrument along with the underlying theory wascompleted and accepted for publication.


J. Dunkers (Polymers Division, NIST); J. Ilavsky,G. Long (Argonne National Laboratory); P. Jemian(UNICAT); C. Muehleman (Rush Medical College)

´ ´

´

´

34

Program Overview

Materials for Electronics

The U.S. electronics industry faces stronginternational competition in the manufacture of smaller,faster, more functional, and more reliable products.Many critical challenges facing the industry requirethe continual development of advanced materialsand processes. The NIST Materials Science andEngineering Laboratory (MSEL) works closely withU.S. industry covering a broad spectrum of sectorsincluding semiconductor manufacturing, devicecomponents, packaging, data storage, and assembly,as well as complementary and emerging areas such asoptoelectronics and organic electronics. MSEL has amultidivisional approach, committed to addressing themost critical materials measurement and standardsissues for electronic materials. Our vision is to bethe key resource within the Federal Government formaterials metrology development and will be realizedthrough the following objectives: Develop and deliver standard measurements and data; Develop advanced measurement methods needed by

industry to address new problems that arise with thedevelopment of new materials;

Develop and apply in situ as well as real-time, factoryfloor measurements, for materials and devices havingmicrometer- to nanometer-scale dimensions;

Develop combinatorial material methodologies forthe rapid optimization of industrially importantelectronic materials;

Provide the fundamental understanding of thedivergence of thin film and nanoscale materialproperties from their bulk values;

Provide the fundamental understanding of materialsneeded for future nanoelectronic devices, includingfirst principles modeling of such materials.

The NIST/MSEL program consists of projects ledby the Metallurgy, Polymers, Materials Reliability, andCeramics Divisions. These projects are conducted incollaboration with partners from industrial consortia(e.g., International SEMATECH), individual companies,academia, and other government agencies. The programis strongly coupled with other microelectronicsprograms within the government such as the NationalSemiconductor Metrology Program (NSMP). Materialsmetrology needs are also identified through theInternational Technology Roadmap for Semiconductors(ITRS), the IPC Lead-free Solder Roadmap, theNational Electronics Manufacturing Initiative (NEMI)Roadmap, the Optoelectronics Industry DevelopmentAssociation (OIDA) Roadmap, IPC (the InternationalPackaging Consortium), and the National [MagneticData] Storage Industry Consortium (NSIC) Roadmap.

In each of these areas, MSEL researchers have madesubstantial contributions to the most pressing technicalchallenges facing industry, from new fabrication methodsand advanced materials in the semiconductor industry, toadvanced packaging materials, to magnetic data storage.Below are just a few examples of MSEL contributionsover the past year.

Advanced Gate DielectricsTo enable further device scaling, the capacitive

equivalent thickness (CET) of the gate stack thicknessmust be 0.5 nm to 1.0 nm. This is not achievable withexisting SiO2/polcrystalline Si gate stacks. Given thelarge number of possible choices for these new layers,the only feasible approach to understanding the complexmaterials interactions that result at the gate dielectric/substrate and gate dielectric /metal gate electrodeinterfaces is through the application of combinatorialmethodologies. This same methodology and apparatusare applicable to a wide variety of problems in theelectronic materials field.

Sub-100 nm NanofabricationThe continual decrease in feature size has been the

driving force for advances in the semiconductor industry.Current structures have 90 nm dimensions with plannednodes at 65 nm and 35 nm structures. Advancedmeasurements of the patterning materials (photoresists),are needed to enable future large scale manufacturing ofsmaller devices. MSEL utilizes advanced x-ray andneutron tools to provide insight into the feasibilityand optimization of these important processes.

Advanced MetallizationElectrodeposited copper is rapidly replacing aluminum

for on-chip “wiring” because of its lower electricalresistivity, superior electromigration behavior, and theability to fill fine features without the formation of seamsor voids. As feature dimensions go below 100 nm,difficulties in maintaining performance are anticipated.These issues are addressed through a combination ofmodeling and experimental efforts.

Test Methods for Embedded Passive DevicesSignificant advantages arise if passive devices are

integrated directly into the circuit board as embeddedpassive devices rather than discretely attached withautomated assembly. New metrology methods weredeveloped to address the needs of the electronic industry.Two test methods were completed and have receivedwide acceptance by industry as new methods to acceleratethe development of embedded passive device technology.

Contact: Martin L. Green (Ceramics Division),Eric K. Lin (Polymers Division),Daniel Josell (Metallurgy Division)

35


The development of wide-band-gap semiconductoroptoelectronic and electronic devices is hinderedby poor electrical contact performance. Reliabilityissues for contacts include the requirements tobe low-resistance, morphologically smooth andthermally stable. This project develops a strategy toimprove electrical and morphological characteristicsof Ohmic contacts to GaN thin films by optimizingmetallization schemes using an integrated phasediagram/combinatorial approach.

Albert V. Davydov and William J. Boettinger

Ti/Al/Ti/Au metallizations are the most commonlyused electrical contact to n-type GaN-based

device structures. However, the overall composition(i.e., thickness ratios) and the thermal processingof metal layers are not yet optimized. A rationale fordesigning the optimum metallization scheme also hasnot been developed. This work demonstrates that thephase diagram approach, along with a combinatorialexperimentation method, is a useful tool in the designand optimization of electrical contacts to GaN andother wide-band-gap semiconductors.

A combinatorial library of Ti/Al/Ti/Au metallizationwas prepared and characterized electrically andmicrostructurally in order to establish optimumcomposition and processing parameters for the realizationof high-quality Ohmic contacts. An array of metallicelements with varying Ti/Al/Ti/Au thicknesses wasdeposited by combinatorial ion-beam sputtering (CIBS)on n-GaN/c-sapphire substrate followed by rapid-thermalannealing (RTA) in the 600 oC to 900 oC temperaturerange. The library design of metal compositions and RTAtemperatures was guided using the Al-Au-Ti phase diagramso that various Ti/Al/Ti/Au contact compositionsrepresented different regions on the phase diagram.

Four-probe transmission-line-method and Hall electricalmeasurements were used to assess contact resistivityand GaN sheet resistance. X-ray diffraction, white-lightinterferometric microscopy, atomic-force microscopy,field-emission scanning-electron-microscopy, andtransmission-electron-microscopy were used to evaluatethe microstructure and morphology of contacts andthe metal/GaN interfaces. The most Au-rich/Ti-poormetallization (E in Figure 1a), Ti(20 nm)/Al(85 nm)/Ti(15 nm)/Au(150 nm) with an RTA of 800 oC/30 sin argon, produced the lowest (ρ = 1.3 × 10–5 Ω × cm2)contact resistivity with reasonable (rms < 150 nm) surfacemorphology. However, the most Al-rich metallization(D in Figure 1a), Ti(20 nm)/Al(170 nm)/Ti(5 nm)/Au(50 nm) with RTA of 750 oC/30 s in argon, producedsuperior surface morphology (rms = 20 nm) andcomparable resistivity of ρ = 2.2 × 10–5 Ω × cm2. It issuggested that scheme (E) with the Au-rich contact canbe used for operating in oxidizing environments, wherethick Au top layer will serve as a protective cap layer,while scheme (D) can be suitable for applications wherecontact surface roughness is critical. According to XRDresults, the superior surface morphology of contactsin the D-series correlates with the absence of phasetransformations in the 600 oC to 900 oC temperatureinterval after the initial formation of Al2Au and Al3Tiphases in metal films.

The advanced metallizations developed in the projectwill be validated in industrial high-electron mobilitytransistors (HEMTs) at Northrop–Grumman Corporation.

Contributors and CollaboratorsL.A. Bendersky, D. Josell, U.R. Kattner, A.J. Shapiro

(Metallurgy Division, NIST); J.E. Blendell, R.S. Gates,P.K. Schenck, M.D. Vaudin (Ceramics Division, NIST);T. Zheleva (Army Research Laboratory); H.G. Henry(Northrop–Grumman); Q.Z. Xue (Intematix Corp.);A. Motayed (Howard University); I. Takeuchi (Univ. ofMaryland)

Combinatorial / Phase Diagram Approach forMetallization to Wide-Band-Gap Semiconductors

Figure 1: a) 36-element combinatorial library of annealedTi/Al/Ti/Au contacts on n-GaN: metal compositions changealong the x-axis from A to F, annealing temperature changesfrom 600 oC to 900 oC along the y-axis; “X”s mark the libraryelements with the lowest resistivity and smoothest morphology;b) map of surface roughness of the combinatorial array:yellow-shaded area represents the contacts with the smoothest(rms < 60 nm) morphology; c) map of electrical resistivity of thecombinatorial array: red-shaded area represents the contactswith the lowest (ρ < 3 × 10–5 Ω × cm2) resistivity.

36


As the microelectronics industry moves towardsPb-free assemblies, reliability has become a majorconcern. Pb-free coatings of nearly pure tin,used as a protective layer to maintain solderabilityon Cu leadframes and connectors, tend to grow“whiskers” which can cause shorts acrosscomponent leads. Copper additions to the Snprotective layer have been considered by industrysince the Sn-Cu-Ag is likely to be the Pb-freebulk solder of choice for industrial applications.However, measurements made at NIST indicatethat the addition of Cu tends to promote whiskergrowth. Our goal is to better understand howcopper and other solutes impact the growthof whiskers on electrodeposited Sn.

Gery R. Stafford, William J. Boettinger,and Kil-Won Moon

Whiskers are generally believed to grow torelieve residual stress in tin and other coatings.

However, the origin of this stress has not beendefinitively determined. The addition of certain solutes,such as Bi and Pb, are known to retard whisker growthin electrodeposited tin coatings, while solutes, such asZn are known to promote growth. Previous workat NIST has shown that low levels of Pb reduce theresidual compressive stress in Sn-Pb electrodeposits,while the addition of Cu increases the residualcompressive stress. Current research focuses on theinfluence of solute type and concentration on residualfilm stress, stress relaxation, and whisker formation.

Metallographic analysis of Sn-Cu electrodepositsindicates that the Cu is present as the Cu6Sn5 intermetalliccompound (IMC) which is observed both in the grainsand along the grain boundaries. It is unlikely that significantamounts of this complex IMC are formed directly duringdeposition. Rather, we believe that the as-depositedalloy is a supersaturated solid solution of Cu in Sn thattransforms to an equilibrium mixture of Sn and Cu6Sn5either during or soon after deposition.

In solid solution, Cu is presumed to occupy interstitialsites within relatively large square channels that exist inthe body-centered tetragonal (bct) Sn structure. Thisgives rise to the anomalously fast diffusion of Cu in thec-direction of the Sn lattice. In this configuration, it isassumed that the Cu does not alter the lattice volume ofthe Sn. It is further presumed that as Cu6Sn5 precipitateswithin the deposit, the Cu exits the interstitial sites in theSn phase and occupies volume more typical of its atomicsize; thus, the volume of the entire deposit increases.

We estimate the volumetric strain for Cu6Sn5 precipitationfrom a 1 % Cu solid solution to be about +0.004. Theresultant stress generated in the electrodeposit as a resultof the IMC precipitation is then on the order of –95 MPa(compressive). This, of course, assumes that Cu6Sn5forms from the solid solution after the deposit hasbeen formed.

Lead-Free Surface Finishes: Sn Whisker Growth

Figure 1 shows a series of x-ray diffraction patternsthat were taken at the Advanced Photon Source, usingsynchrotron radiation, during and after the depositionof a Sn-Cu alloy onto a tungsten substrate. The insetpattern was taken during the 2-minute deposition whilethe remaining patterns were taken 3 minutes apart afterdeposition was complete. The data show that althougha small amount of Cu6Sn5 is present when depositionis stopped (inset), the bulk of the IMC forms in the45 minutes following deposition. Although the Snreflections decrease as the IMC forms, the peaks donot change position indicating that the lattice volume ofthe Sn is constant. This supports the assumption thatinterstitial Cu does not alter the lattice volume of bct Sn.These results indicate that IMC formation occurs afterdeposition is complete and is a likely cause of the highercompressive stress observed in Sn-Cu alloy deposits.


J.E. Guyer, C.E. Johnson, M.E. Williams(Metallurgy Division, NIST); M.D. Vaudin (CeramicsDivision, NIST); D.R. Robinson, D. Wermeille(Advanced Photon Source, ANL)

Figure 1: In situ x-ray diffraction (λ = 0.127 Å) showing growthof the Cu6Sn5 110 reflection in the 45 minutes following depositionof bright Sn-Cu alloy from commercial methanesulfonate electrolyte.

37

Lead-Free Solders and Solderability


Solders and solderability are increasingly tenuouslinks in the assembly of microelectronics as aconsequence of ever-shrinking chip and packagedimensions and the international movement towardenvironmentally friendly lead-free solders. Incollaboration with the NEMI Pb-Free AssemblyProject, we are providing the microelectronicsindustry with measurement tools, data, andanalyses that address national needs in theimplementation of lead-free solders.

Ursula Kattner and Carol Handwerker

Since 1999, NIST has served a major role in NationalElectronics Manufacturing Initiative (NEMI)

projects to assist the microelectronics industry implementPb-free solders. NEMI is an industry-led consortiumof approximately 65 electronics manufacturers,suppliers and related organizations brought together tofacilitate leadership of the North American electronicsmanufacturing supply chain. This move toward Pb-freesolders is a direct result of the European Union ban, startingin 2006, on Pb-containing solders in electronic products.

NEMI and NIST have worked together to respondto the identified needs by:

1) Identifying and providing the most importantlead-free solder data for the microelectronicscommunity (including the definitive databaseon thermodynamic properties of Pb-free alloys),providing an analysis of the mechanical behavior ofSn-Ag-Cu alloys, and disseminating such data on theNIST website (see “Delivery of Thermodynamicsand Kinetic Data” in this volume).

2) Developing and widely disseminating a RecommendedPractice Guide on Test Procedures for DevelopingSolder Data.

3) Providing a list of literature references on alloys,processing, reliability, environmental issues, andcomponents for the implementation of lead-free solders.

4) Completing the microstructure-based failure analysison all thermally cycled assemblies as part of theNEMI project’s full-scale reliability trials.

5) Providing chapters on the materials science ofPb-free solder alloys for inclusion in four separatebooks on Pb-free solder implementation.

Although the NEMI Pb-Free Assembly Project endedin 2003, in 2004 we continued to serve as an informationresource to the microelectronics industry on issues relatedto the effects of alloy composition, reflow temperature

and furnace profiles for best melting and solidificationbehavior of solders when circuit boards are assembled,and to the effects of Pb contamination on alloy meltingand assembly reliability. This activity culminated in aCircuits Assembly article by Alan Rae, VP of Technologyfor Cookson Electronics, and Carol Handwerker of NISTon the range of acceptable alloy compositions in theSn-Ag-Cu system. The following excerpt gives theNEMI position, and effectively the U.S. position, onthe acceptable composition ranges for the new internationalstandard Pb-free alloy and allows both the U.S. and theJapanese preferred alloy compositions based on analysesof NIST thermodynamic data:

Although the composition of some of the alloysbeing commercialized varies slightly from the NEMIcomposition, the NEMI alloy is representative ofthe acceptable range of lead-free solders. Tin–silver–copper formulations with silver content between3.0 [mass] % and 4.1% and copper between 0.5%and 1.0% are virtually indistinguishable in terms ofmelting point and process features. The NEMI alloyprovides a model system for industry that is wellcharacterized, and several NEMI members currentlyare using the alloy in production. The focus ona single lead-free alloy has helped to accelerateindustry convergence on standard solderformulations, manufacturing processes and,ultimately, the timely and cost-effectiveconversion to lead-free assembly.

“The NIST group performed a significant service toindustry by being the focal point for development of areliable technology base to support the choice of a newlead-free alloy. They also led the way for further workon issues such as tin whiskers, which emerged fromthe initial lead-free work. NIST provided not only astrong technical basis but, by verifying reliability andcomparing alternatives, enabled industry to choosean alloy based on extensive investigation.”

Dr. Robert C. Pfahl, Jr., DirectorInternational and Environmental R&D,Motorola, (retired), and Vice PresidentOperations, NEMI, (current)


W.J. Boettinger, K.W. Moon, T. Siewert,D. McCowan, L.C. Smith, S.W. Claggett, M.E. Williams(NIST); J.P. Clech (EPSI, Inc.); J. Bath (Solectron);R. Gedney (NEMI); E. Bradley (Motorola); J.E. Sohn(NEMI, formerly Lucent); A. Rau (Cookson Electronics);E. Benedetto (HP); R. Charbonneau (StorageTek)

38


Copper wiring in on-chip electrical conductorshas reached dimensions so small that electricalresistivity is no longer constant. The higherresistivity induces greater power dissipation,aggravating chip cooling problems. We havequantified the microstructural factors behind thisincrease for silver wires, where silver has thehighest electrical conductivity of any metal.

Daniel Josell and Thomas P. Moffat

As dimensions of transistors in integrated circuits shrink, the dimensions of the metal wires

connecting them shrink as well. Dimensions of thewires, and internal grains, are now at approximately100 nm, similar to the intrinsic mean-free-path lengthsof the conduction electrons. Scattering from wiresurfaces and grain boundaries is significantly reducingelectrical conductivity in such small conductors.

Upcoming dimensional requirements, as specifiedin the International Technology Roadmap forSemiconductors, require determination of the originand magnitude of the resistivity increase in order todirect research that might mitigate these size effects.

To address these issues, we have measuredsize-dependent resistivity of silver wires 100 nm to300 nm tall for widths ranging from approximately50 nm to 840 nm, produced by electrodeposition onpatterned wafers followed by a combined chemicalmechanical planarization and oblique ion-polishingprocess. Silver was selected because of its highelectrical conductivity and our recent developmentof a bottom-up “superfill” process for creatingvoid-free and seam-free wires (Figure 1).

Electrical Properties of On-Chip Interconnections

Figure 1: Transmission electron microscope images of silverwires cross-sectioned after electrical measurement. The higheraspect ratio (height/width) wires contain multiple grains whilethe lower aspect ratio wires contain only twin boundaries.

To interpret the resistivity data, the Fuchs–Sondheimerformalism (diffuse scattering of electrons on the wiresurfaces) was extended to permit a nonzero specularcomponent for surface scattering. The analysis used anexisting formalism to account for grain boundary scattering.The resulting equations permit quantitative comparisonto experiment (Figure 2).

Significantly, the analysis shows that surfacescattering alone is not capable of explaining theobserved behavior. Grain boundary scattering is foundto contribute a similar, or larger, amount to the overallincrease. As a result, the potential payoff of increasingthe grain size to decrease grain boundary scattering isrecognized to be at least as large as that which might beobtained by smoothing surfaces to increase specularity.


G. McFadden (Mathematical and ComputationalSciences Division, NIST); R.R. Keller, Y.-W. Cheng(Materials Reliability Division, NIST); C. Witt (Cookson–Enthone); C. Burkhard, Y. Li (Clarkson University)

Figure 2: Electrical properties of 300 nm tall silver wires as afunction of the wire width. Curves are the result of calculationsincluding intrinsic, surface and grain boundary scattering ofelectrons. The sharp increase at ≈ 100 nm is associated with defects.

39


In order to pursue the rapid development pathset out for hard drives and magnetic memory,industry needs the ability to measure and controlmagnetization on nanometer length scales andnanosecond time scales. This project focuseson the metrology of dynamics and damping inmagnetic thin films, especially on the effectsof nano-scale dimensions and defects inferromagnetic resonance measurements.

Robert D. McMichael

We are developing ferromagnetic resonance (FMR)techniques to measure the static and dynamic

properties of thin films and patterned arrays oftechnologically important ferromagnetic metals and theirinterfaces with normal metals. The primary resultsinclude measurements of surface anisotropy, whichbecomes increasingly important for nanoscale damping,which governs behavior on the sub-nanosecond timescale, and assessment of the magnetic homogeneityof thin films or patterned arrays of magnetic bits.These results are communicated to the magnetic datastorage industry through conference presentations,journal articles and site visits.

For nanostructured materials, surface effectsand interfacial effects are important. For example,micromagnetic calculations show that switchingbehavior of magnetic memory cells will be affected bysurface anisotropy on the edges of the lithographicallypatterned elements, an effect that has not beenappreciated until recently. Measuring the anisotropyof films of varying thickness with ferromagneticresonance enables the determination of surfaceanisotropy values for Permalloy that will be usedin the micromagnetic design of MRAM cells.

A continuing thrust of this project has been thedevelopment of models that describe broadening of theFMR linewidth by defects. These models, which arespecifically designed for thin films, account for bothinhomogeneity and magnetic interactions. Our earliermodels were restricted to “local” defects such asanisotropy. One important application of these modelsis as a method for measurement of anisotropy variationsand axis alignment in perpendicular recording media.This year, we have capped off the model developmentwith a model for the non-local effects of film roughness.

Nanomagnetodynamics

Figure 1: Surface anisotropy measurements for Permalloy filmswith different normal metal over/underlayers. The slopes of the fitlines correspond to surface anisotropy.

Testing of the roughness-linewidth model involvedmaking variable roughness substrates, characterizingthe roughness by atomic force microscopy (AFM),and measuring nearly conformal Permalloy overlayersin FMR. Using only the AFM microstructural dataand the known properties of Permalloy, and assumingconformal roughness, the model predictions agreedextremely well with the measurements. Allowingone free parameter describing thickness variations,the agreement with the experiment is nearly perfect.Validation of this model is the first example oflinewidth modeling based entirely on the measuredmicrostructure.


J.O. Rantschler, B.B. Maranville, J. Mallett,T. Moffat, W.F. Egelhoff, Jr., A.P. Chen (MetallurgyDivision, NIST); J.A. Borchers, S.K. Satija (NISTCenter for Neutron Research); W. Bailey (Columbia U.);A. Arrott (Virginia State University)

Figure 2: FMR line broadening measured in a Permalloy filmand calculated almost entirely from the microstructure of theunderlayer shown in the inset.

–0.60

–0.65

–0.70

–0.75

–0.80

–0.85

–0.90

Effe

ctiv

e an

isot

ropy

(T)

0.60.50.40.30.20.10Inverse permalloy thickness (nm–1)

Ag / Ni80Fe20 / AgCu / Ni80Fe20 / CuTa / Ni80Fe20 / Ta

40


Platinum-iron group alloys have at least twoimportant potential applications: as a medium formagnetic recording and as CO tolerant catalystsin fuel cells. Alloy formation by electrodepositionpresents a convenient and inexpensive alternativeto conventional vacuum methods. A simplethermodynamic model has been used to accountfor the observed dependence of alloy compositionon the deposition potential.

Jonathan J. Mallett, William F. Egelhoff, Jr.,and Thomas P. Moffat

Perhaps the most exciting application of iron-groupPt alloys involves their high magnetic coercivity and

anisotropy in the L10 phase. This makes them idealcandidates for future magnetic storage media, whichwill require high coercivity material to sustain thereductions in bit size needed to increase storage density.Controlled alloy composition and microstructure areessential for achieving the L10 phase, which requiresclose to a 50:50 composition.

Iron-group Pt alloys may also find applicationas poison-resistant hydrogen fuel cell catalysts.A primary concern is the lifetime and efficiency ofthe catalyst used to oxidize hydrogen gas in the cell.Recent research suggests that replacing the traditionalPt catalyst with a Pt alloy containing up to 15 %of an Fe-group element can dramatically improvethe resistance of the catalyst to CO poisoning.An inexpensive, reliable process for producing suchalloys is, therefore, likely to be of considerable utilityto fuel cell development.

An electrodeposition method has been developedthat allows smooth (< 5 nm RMS roughness) alloyfilms to be deposited to a thickness of between 20 nm

and 3000 nm. Deposition takes place from a bathcontaining 3 mM PtCl4 and 0.1 M of the reactivemetal ions (Fe2+, Ni2+, or Co2+). The depositionpotentials were chosen such that deposition of thereactive metal in its bulk form would have beenenergetically impossible. Platinum on the other handreadily deposits at all of the chosen potentials. Theincorporation of the reactive metal occurs due to thehighly negative enthalpy of mixing of the alloy. Thisreduces the activity of the reactive element in the alloyand allows its deposition to occur at more positivepotentials than would be possible for the pure element.The reactive ions are in much higher concentration thanthe noble complexed platinum ions, which allows theformer to be reduced in a kinetically facile way thatenables equilibrium alloy compositions to be achieved.Under these conditions, the alloy composition andmorphology was found to remain remarkably invariantto significant perturbations in constituent concentrations.The strong dependence of composition on potential isshown in Figure 1 for three alloys.

Electrodeposited Pt1–x (Fe,Co,Ni)x Alloys

Figure 1: The theoretical (curves) and experimental (data points)dependence of composition on potential.

Figure 2: The increase in coercivity on annealing.

Figure 2 shows the effect of annealing an FePt filmgrown on a Cu(001) substrate. The transformationto the L10 phase results in a dramatic increase inmagnetic coercivity. The 10 kOe coercivity producedby the 650 ºC anneal is suitable for next-generationmagnetic media.

Current activities are focused on extendingthe investigation to ternary alloys and expandingthe available deposited structures to includenanoparticles (useful as catalysts) and template-formednanostructures (for future magnetic recording media).


E.B. Svedberg (Seagate Technology)

41


Magnetic sensors play a central role in manyimportant technologies ranging from health care tohomeland security. A common need among thesetechnologies is greater sensitivity and smaller size.In ultra-high density data storage, one of the mostpressing needs is for nano-structured media that storedata at ever-increasing densities. Improved methodsfor the magnetic isolation of grains in ultra-thin filmsare a key need; we have initiated research programsin both areas.

William F. Egelhoff, Jr.

NIST’s Magnetic Engineering Research Facility(MERF) is one of the most versatile facilities in the

world for the fabrication and analysis of novel magneticthin films. Two new areas of research, magneticsensors and magnetic media, illustrate this versatility.The common link is that both require novel magneticthin films.

the correct composition for complex ultra-thin films.MERF will soon become the only deposition system inthe world equipped for ultra-thin film deposition by IBSD,AC and DC magnetron sputtering, and molecular beamepitaxy. Such versatility will allow MERF to continueto play a leading role in magnetic thin-film research.

In the area of novel magnetic media we have beencollaborating closely with Seagate. Seagate is theworld’s leading manufacturer of hard-disk drives, andfor the past two years, they have been sending a Ph.D.physicist from their research labs to work with us,first on ballistic magnetoresistance, and now on novelmagnetic media.

Novel Magnetic Materials for Sensors andUltra-High Density Data Storage

Figure 1: The Magnetic Engineering Research Facility (MERF).

It appears that ultra-sensitive magnetic sensors willrequire the use of thin films of ultra-soft magnetic alloys.These alloys often have a very complex composition suchas Fe73.5Si13.5B9Nb3Cu1 and require a very specificannealing process. These materials are well-known inbulk form but have never been produced in thin-film form.The project goal is to carry out the metrology neededto produce these materials as ultra-thin films.

We are presently modifying the MERF to add thecapability of ion-beam sputter deposition (IBSD). IBSD,which is based on sputtering a target of the desiredcomposition, is the only reliable method for maintaining

Recent work determined a novel method formagnetically decoupling grains in CoPd media. CoPdmultilayers are one of the leading candidates for thenext generation of magnetic media. We have found thatif a Au film is deposited on top of the CoPd and annealin air, two unexpected phenomena occur as a result ofrapid diffusion of atoms along grain boundaries. One isthat Co atoms diffuse to the surface, react with oxygen,and remain at the surface. The other is that Au atomsdiffuse into the grain boundaries replacing Co. The neteffect is that the grain boundaries are demagnetized, andeach grain can magnetically switch independently of itsneighbors. This method provides the best magneticisolation of grains yet found.


E.B. Svedberg, D. Weller (Seagate); R.D. McMichael,T.P. Moffat, J. Mallett (Metallurgy Division, NIST);D.P. Pappas (Electromagnetics Division, NIST)

Figure 2: Close collaborations with Seagate have contributed tothe success of this project.


42

Magnetism is endemic to our society. Ferromagnetsare the basis for information storage in computersand recording devices; they are found in all motors,transformers, and generators; and they are used ina host of other products like credit cards, cellulartelephones, radios, and televisions. However, themagnetic characteristics of the ferromagnet usedin each of these applications varies with theproduct. Consequently, industry needs to knowthe interconnection between material characteristicsand magnetic properties, and how to control them.A critical element in such an understanding isknowledge of the fundamentals of spin interaction,and how to measure it. NIST provides the metrologyneeded by industry to do this.

Robert D. Shull and Jeffrey W. Lynn (NCNR)

In 1962, a new type of interaction between magnetic spins was predicted — a band structure effect caused

by an instability in the conduction electron gas near theFermi surface. The instability creates a new low-energystate in which the magnetic spin orientations form astanding wave, called a spin density wave (SDW).Similarly, a wave of charge, called a charge density

wave (CDW), was also predicted. Charge density waveshave since been found experimentally, and the SDWconcept has been successfully used to explain the unusualincommensurate antiferromagnetism in Cr. However,evidence for SDWs in a ferromagnet has proven elusive.

Using neutron diffraction measurements at NIST,we recently discovered a SDW in a single crystal ofFe60Al40, a material with ferromagnetically interactingspins. This was shown (Figure 1) by the presence ofscattering intensity at locations where there was noatomic reciprocal lattice vector (i.e., at places wherethere would normally be no coherent atomic scattering).In fact, the data showed there are several SDWs inthe material with wave vectors near (110) directions(shown in Figure 2). No such waves were foundalong either (100) or (111) directions.

Discovery of Spin Density Waves in a Ferromagnet: Fe-Al

Figure 1: Neutron scattering intensity for a) polycrystallineFe60Al40 at 120 K and 1.7 K with their difference in the inset andb) single crystal Fe60Al40 at 12 K shown as contours about the(1/7, 1/7, 1/7) peak.

Figure 2: The reciprocal lattice spots in Fe60Al40, showing theSDW wave vectors (black arrows) (Physical Review Letters 91,217201-1 (2004).

The presence of SDWs shows the exchangeinteractions between magnetic spins in Fe-Al is a bandstructure (i.e., non-local) effect and explains the largemagnetism in these materials containing a large contentof “non-magnetic” atoms. This discovery should be asimportant to the understanding of magnetism in metals asthe SDWs in Cr. In Cr, the SDW wave vector, close to a(100) direction, makes moments of nearest neighbors inthe bcc structure almost oppositely aligned.


A.S. Arrott (Virginia State University); D.R. Noakes(Virginia State University); M.G. Belk (Virginia StateUniversity); S.C. Deevi (Phillip Morris Co.); D. Wu(Dartmouth College)

Program Overview

Safety and Reliability

While various structural failures have capturednational attention over the years, the events ofSeptember 11, 2001 generated a greatly increasedawareness of vulnerabilities in our nationalinfrastructure. The extent of these vulnerabilitiesdepends to a large degree on the performance ofmaterials in situations outside of the original designconsiderations. It is now recognized that a criticaland urgent national need exists to establish thebehavior of materials under such extreme loadings,and to disseminate guidance and tools to assess andreduce future vulnerabilities.

The goal of providing a technical basis for cost-effectivechanges to national practices and standards, coupled witha need for an integrated effort drawing on capabilities andexpertise of a broad collaborative team, has lead to thedevelopment of the Safety and Reliability Program withinMSEL. This program draws on the expertise of severaldivisions in MSEL and across NIST.

Project selection is guided by an identification andassessment of the particular vulnerabilities within ourmaterials-based infrastructure, and focusing on thoseissues which would benefit strongly by improvedmeasurements, standards, and materials data. Ultimately,we intend to moderate the effects of acts of terrorism,natural disasters, or other emergencies, all throughimproved use of materials.

Our vision is to be the key resource within the FederalGovernment for materials metrology development asrealized through the following objectives: Identify and address vulnerabilities and needed

improvements in U.S. infrastructure; Develop and deliver standard measurements and data; Develop advanced measurement methods needed by

industry to address new problems that arise with thedevelopment of new materials;

Support other agency needs for materials expertise.

This program responds both to customer requests(primarily other government agencies) and to theDepartment of Commerce 2005 Strategic Goal of“providing the information and framework to enablethe economy to operate efficiently and equitably.”For example, engineering design can produce safe andreliable structures only when the property data for thematerials is available and accurate. Equally important,manufacturers and their suppliers need to agree onhow material properties should be measured.

The Safety and Reliability Program works towardsolutions to measurement problems on scales ranging

from the macro to the micro, in three of the Laboratory’sDivisions (Materials Reliability, Metallurgy, and Polymers).The scope of activities includes the development andinnovative use of state-of-the-art measurement systems;leadership in the development of standardized testprocedures and traceability protocols; development ofan understanding of materials in novel conditions; anddevelopment and certification of Standard ReferenceMaterials (SRMs). Many of the tests involve extremeconditions, such as high rates of loading, high temperatures,or unusual environments (e.g., deep underwater). Theseextreme conditions often produce physical and mechanicalproperties that differ significantly from the handbookvalues for their bulk properties under traditional conditions.These objectives will be realized through innovativematerials, property measurement and modeling.

The MSEL Safety and Reliability Program isalso contributing to the development of test methodstandards through committee leadership roles instandards development organizations such as theASTM International and the International StandardsOrganization (ISO). In many cases, industry alsodepends on measurements that can be traced to NISTStandard Reference Materials (SRM®).

In addition to the activities above, all three divisionsprovide assistance to various government agencies onhomeland security and infrastructural issues. Projectsinclude assessing the performance of structural steelsas part of the NIST World Trade Center Investigation,advising the Bureau of Reclamation on metallurgical issuesinvolving pipelines and dams, advising the Department ofthe Interior on the structural integrity of the U.S.S. ArizonaMemorial, and collaborating with both the Departmentof Transportation and the Department of Energy onpipeline safety issues.

Contact: Frank W. Gayle (Metallurgy Division)

43

44


Analysis of Structural Steel in the World Trade Center

In 2002 NIST became the lead agency in theinvestigation of the World Trade Center collapse.The investigation addresses many aspects of thecatastrophe, from occupant egress to factorsaffecting how long the Twin Towers stood after beinghit by the airplanes, with a goal of gaining valuableinformation for the future. A critical aspect ofthe investigation is the metallurgical analysis ofstructural steels from the Twin Towers. The analysisincludes characterization of properties, failuremodes, and temperature excursions seen by the steel.

Stephen W. Banovic, Richard J. Fields,Timothy J. Foecke, William E. Luecke,J. David McColskey, Christopher N. McCowan,Thomas A. Siewert, and Frank W. Gayle

The collapse of the World Trade Center (WTC)Towers on September 11, 2001, was the worst building

disaster in human history. Engineers, emergencyresponders, and the nation were largely unprepared forsuch a catastrophe. NIST is investigating the disaster(see http://wtc.nist.gov/), and a primary objective is todetermine why and how the towers collapsed after theinitial impact of the aircraft. As part of this investigation,the Metallurgy and Materials Reliability Divisions inMSEL are studying recovered structural steel fromthe WTC site. Progress in this study is outlined here.

Task 1 — Collect and catalog physical evidence:236 structural pieces of the WTC towers havebeen collected, brought to NIST and studied. Reportshave been issued documenting the steel, the as-builtlocations, the structure of the towers based on designdocuments, and the standards at the time of construction.

Figure 1: Enhanced image of the impact zone of WTC 2 hasallowed determination of failure modes in the perimeter columns.

Figure 2: Normalized yield strength and ultimate tensile strengthmeasured as function of temperature. Model curve is based onhistorical data for WTC era steels.

Task 2 — Categorize failure mechanismsbased on visual evidence: Available photographicevidence and recovered steel have been examined anddocumented as to failure mechanisms (Figure 1).

Task 3 — Determine steel properties tosupport structure performance and airplaneimpact modeling studies: 29 different steelshave been characterized for room temperaturemechanical properties. High-temperature andhigh-strain-rate tests are now complete. Creep, ortime–temperature-dependent behavior, has beendetermined for floor truss and column steel.

Task 4 — Correlate determined steel propertieswith specified properties: Most steel was found toexceed the specified minimum values by 5 to 10 percent.

Task 5 — Analyze steel to estimate temperatureextremes: We have developed a technique ofpaint characterization to provide a quick mapping oftemperature excursions seen by the steel. Challengesincluded deciphering pre- and post-collapse exposure(see related Highlight in this report).

Data on material properties have been provided to aidin the modeling of plane impact and structural responseto the fires. A final draft of the investigation report willbe released for public comment in December 2004.


R. Santoyo, L. Rodine (Materials Reliability Division,MSEL, NIST); M. Williams, S. Claggett, M. Iadicola,R. Jiggetts (Metallurgy Division, MSEL, NIST)

45


Standard Test Methods for Fire-Resistant Steel

The fires in the World Trade Center and theensuing collapse focused attention on thevulnerability of structural steel to fire. Recently,steel mills in Europe and Japan have begun tomarket steels designated as “fire-resistant.”This project is developing a standard test methodfor quantitatively evaluating and comparingthe resistance to high-temperature deformationand failure of structural steels.

William E. Luecke, J. David McColskey,Richard J. Fields, and Frank W. Gayle

At room temperature, the stress–strain behavior of structural steel is considered to be independent of

time. Above about 300 °C, structural steel begins to losestrength. At still higher temperatures, about 500 °C,creep, or time-dependent deformation, further reducesthe load-carrying capability. By 600 °C, most structuralsteels have lost more than half their strength. For thisreason, structural steels in buildings are insulated tominimize the temperature rise and resulting strength loss.

Fire resistant (FR) steels are intended to be drop-inreplacements for structural steel. They typically meetthe same specifications with similar welding propertiesand cost only marginally more. Yet these steels retainsuperior high-temperature strength, offering potentialfor extra time for building occupants to escape a fire.

In Japan and Europe, FR steels are qualifiedbased on yield strength retention at high temperature.This definition may not be the most appropriate sinceretained yield strength is a short-time property, yetfire-resistance is often measured in hours. Also, withincreasing temperature, strength becomes increasinglysensitive to the strain rate. Current U.S. standards forload-bearing components in fire are based on time toreach a given temperature, so effectively all steels areconsidered to have the same high-temperature strengthreduction, regardless of actual properties.

We are currently studying a hybrid of creep andconventional tension tests. The test specimen is heldunder constant load as the test temperature rampsupward linearly. Over a narrow temperature range,which can be approximated as a critical temperature,the deformation rate increases dramatically, and thespecimen fails. This critical temperature can be usedas a measure of fire resistance.

One goal of this project is to refine this testtechnique and offer it as a draft standard. Publishedresearch has established that the temperature ramp test

Figure 1: Comparison of “run-away creep” deformation ofFR and ordinary construction steel.

can rank steels, but definitive studies comparingordinary and FR steels are lacking. Until a standardtest method is developed, it is difficult to confidentlycompare results between laboratories. This researchwill focus on understanding the limitations, repeatability,and reproducibility of the method by characterizingseveral different classes of construction steels.Interlaboratory studies will be used to probe themethod’s limitations, precision and bias.

A second goal is to generate constitutive modelsfor modeling high-temperature deformation of steels.If test results are not predictable from underlyingdeformation behavior, a test cannot possibly indicatereal-world performance. A second benefit is thatmuch-needed data will be supplied for finite elementmodeling of modern structural steel deformation in fire,a prerequisite for performance-based fire resistant design.

Figure 1 illustrates the concept of criticaltemperature for “run-away” creep for two FR steelsand one conventional construction steel. The resultsdemonstrate the potential performance advantage ofFR steel. Much more work is necessary, however, tounderstand the variability in performance of ordinaryand FR steels.

During 2004, we also spearheaded the developmentof an ASTM task group under committee A01 to evaluatetest methods for fire resistance of structural steel.


M. Bykowski (SURF)

46


Pipeline Safety: Corrosion, Fracture, and Fatigue

A critical element of the nation’s infrastructure isthe more than 2 million miles of natural gas andhazardous liquid pipelines that provide over halfof the nation’s energy. While pipelines are thesafest means of transporting these materials, theyare not immune to failures. In 2002, Congresspassed the Pipeline Safety Improvement Act toreduce pipeline failure rates, increase publicsafety, and ensure a continuous and uninterrupteddelivery of the nation’s energy supply. NIST isworking with the pipeline industry, DoT, and DoEto provide the measurement methods, standards,and data needed to accomplish this objective.

Richard E. Ricker

In December 2002, Congress passed the Pipeline Safety Improvement Act (PSIA), through which

the DoT Office of Pipeline Safety (DoT/OPS), DoE,and NIST will coordinate pipeline R&D. In FY 2004,DoT/OPS, DoE, and NIST signed a memorandumof understanding (MOU) and a Five-Year Joint R&DPlan detailing agency responsibilities in a programof research, development, demonstration, andstandardization for pipeline integrity. The Departmentof the Interior/MMS has also been included in theplan because of its responsibility for off-shore oilpipelines. Under these agreements, NIST is responsiblefor materials research, fire safety, and standardsdevelopment that pertain to pipeline safety, reliability,and damage prevention.

A number of factors have combined to increase theneed for improvements in the safe operation of pipelines.Among these are increases in national energy needs,

population densities near pipelines, environmentalconcerns, and average pipeline age. While recent accidentshave brought increased attention to this issue, pipelineincident rates have not changed significantly since 1986(Figure 1). While failure mechanisms are not alwaysthoroughly documented, the DoT/OPS incident data canbe used to estimate the distribution of failure mechanisms(Figure 2). Clearly, fracture, fatigue and corrosion accountfor a significant proportion of the failures. Understandingand developing better standards for preventing thesematerials-related failure mechanisms is the objectiveof NIST R&D in 2004–2005 project.

Figure 1: Pipeline incidents and damage since 1986 (DoT/OPS).

Figure 2: Estimated pipeline failure mechanism distributionbased on DoT/OPS data over ≈ 10 years.

NIST has a history of contributions to pipeline safety.For example, NIST conducted welding and fractureresearch for the Alaskan pipeline, and the NIST 14-yearstudy of underground corrosion is the basis for severalcurrent pipeline standards. The current program willreview existing standards, the current corrosion, fatigue,and fracture issues, and evaluate where improvementsmay be possible. This work will be done in collaborationwith industry, particularly the Pipeline Research CouncilInternational (PRCI) Corrosion and Inspection TechnicalCommittee, and with private sector consensus standardssetting organizations. NIST and its collaboratorshave discussed and reviewed NIST R&D plans andcoordination. A Government/Industry PipelineR&D Forum is being organized for April 2005 topresent the Five-Year Joint R&D Plan.


R.J. Fields, T.J. Foecke, F.W. Gayle,C.A. Handwerker (Metallurgy Division, NIST);T. Siewert, C. McCowan (Materials Reliability Division,NIST); C. Sames (PRCI); J. Merritt, S. Gerard,R. Smith (DoT/OPS); C. Freitas, R. Anderson,D. Driscoll (DoE); M. Else (DoI)

47

Frangible Bullets and Soft Body Armor


Police officers’ lives often depend upon theeffectiveness of their soft body armor. Continuedintroduction of new types of body armor andbullets (including lead-free frangible models) areoverwhelming the enforcement community’s abilityto test everything adequately. At the request ofthe Office for Law Enforcement Standards (OLES),we are conducting a thorough study to identifythe determining factors for effective protectionagainst the various types of bullets on the market.

Lyle E. Levine and Richard Rhorer (822)

The impact of a bullet on soft body armor is aremarkably complex process involving complex

geometries (including the density and weave pattern ofthe body armor fibers) and high-strain-rate propertiesof the bullet and armor materials. Complicating theissue further, recent lead-free bullet designs are beingproduced with a wide range of materials (copper, tin,iron, tungsten, zinc and polymers) and processingconditions (sintering, cold-pressing and injectionmolding). The mechanical behavior of these new bulletdesigns can be dramatically different from those of theconventional lead bullets for which soft body armor hasbeen designed. Degradation of body armor materialswith age, mechanical wear and exposure conditions(moisture and light) is also important. A long-term,comprehensive project to address all of these issueswas started this fiscal year.

Planned tasks for bullets include:

High-strain-rate testing of bullets (longitudinal anddiametral) and bullet materials using the NIST Kolskybar facility. Approximately 600 tests will be required.

Corresponding low-strain-rate testing using aconventional mechanical testing machine.

Metallographic and chemical analyses of bulletmaterials.

Planned tasks for soft body armor materials include:

Design and construction of a miniature Kolsky barfor high-strain-rate testing of fibers.

Design and construction of tensile cages forminiature and existing Kolsky bars.

Exposure of fiber test samples to degradingconditions (light, moisture and abrasion)

High-strain rate testing of as-received and degradedindividual fibers, multiple non-braided fibers, andbraided fiber bundles. Approximately 2000 testswill be required.

Corresponding low-strain-rate testing using aconventional mechanical testing machine.

These tasks will be complemented by finite elementmodeling of: High-strain-rate bullet tests using measured

high-strain-rate behavior of bullet materials. Fiber and fiber bundle tests using measured

high-strain-rate behavior of fiber materials. Bullet impacts with woven body armor fabric.

Expected ImpactsThe test data will be used by the U.S. Department of

Justice to determine what actions are required to ensurethat soft body armor provides adequate protection topolice officers in the field. The experimentally validatedfinite element models will be used to screen combinationsof bullets and body armor designs to identify potentialthreats at an early stage.


S. Banovic, R. Fields, L. Ma, S. Mates(Metallurgy Division, NIST); M. Kennedy, E. Whitenton(Manufacturing Metrology Division, NIST); T. Burns(Mathematical and Computational Sciences Division,NIST)

Figure 1: Frangible 9 mm bullet and remains after impact.

48


Marine Forensics and Preservation of Historic Shipwrecks

Historic shipwrecks are valuable culturalresources continuously under attack by theirenvironment. Preservation techniques aimed atslowing corrosion processes and mechanicallystabilizing shipwreck structures are being modeledand designed in coordination with the NationalPark Service, NOAA, and the U.S. Navy. Thesetechniques are being applied to such wreck sitesas the USS Arizona, the Ellis Island Ferry, RMSTitanic, CSS Hunley, and the USS Monitor.

Tim Foecke

Since the attack on Pearl Harbor on December 7th,1941, the wreck of the USS Arizona has been

slowly corroding and collapsing. The monument thatstraddles the wreck sees over 700,000 visitors per year,and there is immense interest in preserving the wreckfor as long as possible. Aside from the historicalinterest, there are substantial environmental risksassociated with any large-scale collapse of the wreck,as there is still an estimated 500,000 gallons of fuel oiltrapped below water.

The Submerged Cultural Resources Unit (SCRU) ofthe National Park Service (NPS) has responsibility formaintaining and preserving the wreck. Based on ourpast experience in marine forensics, SCRU approachedthe Metallurgy Division for assistance in several areas.

Figure 1: Finite element model of the center section of theUSS Arizona, where fuel oil is believed to be concentrated.

Advice is being given on methods to measure theremaining thickness of the hull through the existingbioencrustation. A finite element model (Figure 1) isunder construction, and techniques are being developedto map the corrosion current data gathered by the NPSover the last 20 years onto the model in the form ofdecreasing plate thicknesses.

The tools developed in this project will eventuallybe transferred to companies and organizations thatare responsible for monitoring submerged wreckscontaining hazardous materials, allowing them tobetter simulate critical conditions.

Figure 2: Cross-section of a rivet extracted from the submarineHL Hunley, showing slag flow lines on the left. The submarine, thefirst to sink an enemy ship, is currently on shore in Charleston, SC).

In addition to the USS Arizona, we are providingassistance and technical expertise to NPS, NOAA andthe Navy regarding preservation of the Ellis IslandFerry, the CSS Hunley (Figure 2), the USS Monitor(partially salvaged), and the RMS Titanic. NISTparticipated in a multi-agency remote survey of theTitanic wreck in June 2004 at the University of RhodeIsland, satellite linked to the NOAA Ship Ron Brown.Ongoing analysis of the photos and videos taken duringthis expedition will attempt to map the degradation ofthe wreck and chart the impact of any human visits.


L. Ma, S. Banovic (Metallurgy Division, NIST);L. Murphy, M. Russell, D. Conlin (Submerged CulturalResources Unit, National Park Service); Lt. J. Weirich(NOAA); Cdr. J. Morganthou (CINCPAC, U.S. Navy)

49


Kinetic Penetrators and Amorphous Metals

Environmental and political debate on thecontinued use of depleted uranium (DU) alloyshas led to increased efforts at developing low-cost,reduced-toxicity replacement penetrator materialswith similar ballistic performance. The MetallurgyDivision has been working with the Army ResearchLaboratory (ARL) on a novel tungsten-amorphousmetal composite material as one of these newDU alloy replacements. The project relies onthe Metallurgy Division’s expertise in rapidsolidification processing and instrumentedpowder consolidation to provide these compositematerials to the ARL for ballistic testing.

Stephen D. Ridder

Bulk amorphous metal (BAM) alloys are a relativelynew class of alloys that solidify amorphously in

castings with cross-sections as large as 10 mmacross, a significant improvement over the sub-1 mmcastings of conventional amorphous metal alloys.The Army’s interest in high-density BAM alloys is aresult of their need for a new armor-piercing projectilematerial to replace the current Depleted Uraniumtechnology.

Modern, tank-fired armor-piercing munitions arehigh-velocity projectiles machined from DU alloys orfrom two-phase, tungsten-based heavy alloy (W–)composites. High-density materials are chosen tomaximize the kinetic energy available for armorpenetration. The projectile material is partiallyconsumed by erosion or back extrusion as it entersarmor plate. Current DU alloys are superior to availableW-composites due to “continuous tip sharpening”compared to the tip blunting and “mushrooming”seen in the W-composites. This tip sharpening isgenerally attributed to adiabatic shear localization(ASL) and failure, a specific failure mode exhibitedby DU alloys when rapidly loaded. Heat generatedwithin the penetrator during high-strain-rate loadingproduces sufficient thermal softening of the DU alloystructure to overcome work-hardening in ASL bandsconcentrated near the tip of the penetrator thatpromotes rapid removal of deformed material.The continuously sharpened penetrator concentratesthe remaining kinetic energy at the end of a relativelynarrow tunnel, thus reducing energy loss andimproving penetrator efficiency.

Recent work by the ARL has shown thatBAM alloys have promise in providing the necessary

The Metallurgy Division designed andbuilt a small-scale gas atomizer to study thecomposition-dependent solidification behavior ofBAM alloys using small batch (250 g) processingto reduce material and processing costs. Figure 1shows spherical amorphous Hf alloy powderparticles recently produced. The material producedby this device is preferred to the more readilyavailable comminuted melt-spun amorphous ribbonmaterial because these spherical particles can beblended with tungsten powder and consolidated toproduce a more homogeneous W-composite billet.

A hot isostatic press (HIP) has been instrumentedwith an eddy current sensor developed at NIST tofollow the composite consolidation and provide thenecessary quantitative, real-time monitoring of theprogress of densification. This technology willallow full densification of the W-composite withoutinducing large-scale crystallization of the Hf basedBAM powder.


F.S. Biancaniello, C.E. Campbell, R.J. Fields,R.D. Jiggetts, U.R. Kattner, S.P. Mates (NIST),L. Kecskes (U.S. Army)

Figure 1: Amorphous Hf alloy powder particles produced inNIST small-scale gas atomization facility.

ASL-like behavior that could lead to a W-compositepenetrator material with performance similar toDU alloys.

51

Metallurgy Division FY04 Annual Report Publication List


Anelasticity and Springback PredictionMa, L., T.F. Zahrah, and R.J. Fields, “Numerical

three dimensional simulation of cold compaction andspringback for particulate prealloyed reinforcedcomposites,” Powder Metallurgy Journal 47 (1),31–36, 2004.

Munitz, A., D. Dayan, D.J. Pitchure, andR.E. Ricker, “Dynamic mechanical analysis of pureMg and Mg AZ31 alloy,” Magnesium Technology 2004,edited by A.A. Lu, TMS, 2004.

Paul, R.L., H. Chen–Mayer, G.R. Myneni,W.A. Lanford, and R.E. Ricker, “Hydrogen uptakeby high purity niobium studied by nuclear analyticalmethods,” Matériaux and Techniques 91 (7-8-9),23–27, 2003.

Microstructural Origins ofSurface Roughening and Strain Localization

Liu, J., S.W. Banovic, F.S. Biancaniello, andR.D. Jiggetts, “Through-thickness textureinhomogeneity in a recrystallized Al-Mg alloy,”Met. Trans. A Communications, (submitted), 2004.

Liu, J., and J.G. Morris, “Recrystalliationmicrostructures and textures in AA 5052 continuouscast and direct chill cast aluminum alloy,” MaterialsScience Eng A, (submitted), 2004.

Stoudt, M.R., S.W. Banovic, and T. Quarrick,“Evolution of deformation-induced surfacemorphologies developed on Fe-based sheet metal,”45th MWSP Conference Proceedings, edited byM.A. Baker, Warrendale, PA: Iron and Steel Society,pp. 507–517, 2003.

Stoudt, M.R., J.B. Hubbard, and S.W. Banovic,“Evolution of deformation induced surfacemorphologies generated in Fe-based sheet,”Test and Measurement World, (submitted), 2004.

Stoudt, M.R., J.B. Hubbard, and S.W. Banovic,“Evolution of deformation induced surfacemorphologies generate in Fe-based sheet,” SpecialPublication # 1837; Innovations in Modeling andTesting of Steel Structures for Automotive Applications,edited by R. Mohan, et al., Warrendale, PA: SAE, 2004.

Stoudt, M.R., J.B. Hubbard, and S.W. Banovic,“Evolution of deformation induced surfacemorphologies generated in Fe-based sheet,”Transactions of the SAE, (submitted), 2004.

Stoudt, M.R., J.B. Hubbard, and S.W. Banovic,“Evolution of deformation induced surfacemorphologies generated in Fe-based sheet,”SAE Transactions: Journal of Materials andManufacturing, (in press), 2004.

Stoudt, M.R., A. Munitz, S.W. Banovic, andR.J. Fields, “Effect of uniaxial strain on the surfaceroughness of pure Mg and Mg AZ31 alloy,”Magnesium 2004, (submitted), 2004.

Stoudt, M.R., A. Munitz, S.W. Banovic, andR.J. Fields, “Effect of uniaxial strain on the surfaceroughness of pure Mg and Mg AZ31 alloy,” MagnesiumTechnology 2004, edited by A.A. Luo, Warrendale, PA:TMS, pp. 269–274, 2004.

Underlying Processes ofPlastic Deformation in Metal Alloys

Cai, M., S.C. Langford, L.E. Levine, andJ.T. Dickinson, “Determination of strain localizationin aluminum alloys using laser-induced photoelectronemission,” Journal of Applied Physics, (submitted),2004.

Delos–Reyes, M.A., M.E. Kassner, and L.E. Levine,“X-ray diffraction and the existence of long rangeinternal stresses,” Dislocations, Plasticity and MetalForming, edited by A.S. Khan, R. Kazmi, and J. Zhou,Fulton, MD: Neat Press, p. 262, 2003.

Khan, S.M, H.M. Zbib, and D.A. Hughes,“Multi-scale dislocation dynamics plasticity: applicationfor Frank’s dislocation boundaries,” (submitted), 2004.

Kramer, D.E., M.F. Savage, and L.E. Levine,“AFM observations of slip band development in Alsingle crystals,” Acta Materialia, (submitted), 2004.

Levine, L.E., “Multiscale phenomena in materials —experiments and modeling related to mechanicalbehavior,” Symposium Proceedings Vol. 779, edited byH.M. Zbib, D.H. Lassila, L.E. Levine, and K.J. Hemker,Warrendale, PA: MRS, 2003.

Levine, L.E., and R.M. Thomson, “A statisticalconnection between dislocations and mechanicalproperties,” Dislocations, Plasticity and MetalForming, edited by A.S. Khan, R. Kazmi, and J. Zhou,Fulton, MD: Neat Press, p. 286, 2003.

Tavazza, F., L.E. Levine, and A. Chaka, “Atomisticinsights into dislocation dynamics in metal forming,”Proceedings of 3rd International ComputationalModeling and Simulation of Materials, Acireale, Italy,(in press), 2004.

52


High Speed MachiningBasak, D., H.W. Yoon, R. Rhorer, T.J. Burns, and

T. Matsumoto, “Temperature control of pulse heatedspecimens in a Kolsky bar apparatus using microsecondtime-resolved pyrometry,” International Journal ofThermophysics, 25 (2), 561–574, 2004.

Basak, D., H.W. Yoon, R. Rhorer, and T.J. Burns,“Microsecond time-resolved pyrometry duringrapid resistive heating of samples in a Kolsky barapparatus,” American Institute of Physics, 684, Part 2,pp. 753–757, 2003.

Burns, T., M. Davies, R. Rhorer, D. Basak,H. Yoon, R.J. Fields, L.E. Levine, E. Whitenton,M. Kennedy, and R. Ivester, “Influence of heating rateof flow stress in modeling of machining processes,”Proceedings of the CIRP International Workshop onModeling and Machining, Cluny, France, (in press),May 2004.

Mechanisms for Delivery ofThermodynamic and Kinetic Data

Campbell, C.E., W.J. Boettinger, T. Hansen,P. Merewether, and B.A. Mueller, “Examination ofmulticomponent diffusion between two Ni-basesuperalloys,” Properties of Complex Inorganic Solids 3,edited by P. Turchi, et al., New York: KluwerAcademics/Plenum Publishers, (in press), 2003.

Campbell, C.E., J.C. Zhao, and M.F. Henry,“Comparison of experimental and simulatedmulticomponent Ni-base superalloy diffusion couples,”Journal of Phase Equilibria and Diffusion, 25 (1),6–15, 2004.

Kattner, U.R., J.E. Morral, and W.J. Boettinger,“Databases for computational thermodynamics anddiffusion modeling,” Journal of Phase Equilibria andDiffusion, 24 (3), 416–421, 2003.

Phase Field Modeling Tools forMicrostructure and Stress Evolution

Alikakos, N.D., P.W. Bates, J.W. Cahn, P.C. Fife,G. Fusco, and G.B. Tangoglu, “Extreme Anisotropy ofDiffusion Interfaces in Crystals,” SIAM Journal ofApplied Mathematics, (in press), 2003.

Cahn, J.W., and J. Taylor, “A unified approachto motion of grain boundaries, relative tangentialtranslation along grain boundaries, and grain rotation,”Acta Materialia, (in press), 2003.

Granasy, L., T. Pusztai, T. Borzsonyi, J.A. Warren,and J.F. Douglas, “A general mechanism ofpolycrystalline growth,” Nature Materials, (in press),2004.

Granasy, L., T. Pusztai, G. Tegze, J.A. Warren,and J.F. Douglas, “On the growth and form ofspherulites,” NATURE, (submitted), 2004.

Granasy, L., T. Pusztai, and J.A. Warren,“Modelling polycrystalline solidification using phasefield theory,” Journal of Physics: Condensed Matter,(in press), 2004.

Guyer, J.E., W.J. Boettinger, J.A. Warren,and G.B. McFadden, “Phase Field Modeling ofElectrochemistry I: Equilibrium,” Physical Review E,69 (2) Art. No. 021603, Part 1, February 2004.

Guyer, J.E., W.J. Boettinger, J.A. Warren,and G.B. McFadden, “Phase Field Modeling ofElectrochemistry II: Kinetics,” Physical Review E,62 (2), Art. No. 02164, 2004.

Guyer, J.E., D. Wheeler, and J.A. Warren,“A finite volume PDE solver using Python (FlPy),”http://www.python.org/dc2004/papers/28,WEB Publication, (submitted), 2004.

Lewis, D.J., T. Puztai, L. Granasy, J.A. Warren,and W.J. Boettinger, “Phase field models for eutecticsolidification,” Journal of Materials, 56 (4), 34–39,April 2004.

Sekerka, R.F., and J.W. Cahn, “Solid–liquidequilibrium for non-hydrostatic stress,” ActaMaterialia, (submitted), 2003.

Tegze, G., J.A. Warren, L. Granasy, T. Puztai,T. Borzsonyi, and J.F. Douglas, “The influence offoreign particles in the formation of polycrystallinesolidification patterns,” Solidification Processes andMicrostructures: A Symposium in Honor of Prof.W. Kurz, M. Rappaz, C. Beckermann, and R. Trivedi,TMS, pp. 379–385, 2004.

Upmanyu, M., D.J. Srolovitz, A.E. Lobkovsky,J.A. Warren, and W.C. Carter, “Simultaneous grainboundary migration and grain rotation,” ActaMaterialia, (submitted), 2003.

Warren, J.A., R. Kobayashi, A.E. Lobkovsky,and W.C. Carter, “Extending phase field models ofpolycrystalline materials,” Acta Materialia, 51,6035–6058, 2003.

Warren, J.A., L. Loginova, L. Granasy,T. Borzsonyi, and T. Pusztai, “Phase field modelingof alloy polycrystals,” Proceedings for ModelingCasting and Advanced Solidification Processes X,edited by D. Stefanescu, J.A. Warren, M. Krane,and M. Jolly, p. 45–52, 2003.

53


Hardness Standardization — Rockwell,Vickers, Knoop

Ma, L., S. Low, and J. Fink, “Effect of steel andtungsten carbid ball indenters on Rockwell hardnesstests,” Proceedings of HARDMEKO, (submitted), 2004.

Stibler, A., S.R. Low, and R. Ellis, “Comparativemeasurements of Slovenian and USA Rockwell Chardness scales,” Hardmeko 2004, (submitted), 2004.

3D Electron MicroscopyFord, A.X., J.E. Bonevich, R.D. McMichael,

M.D. Vaudin, and T.P. Moffat, “Structure and magneticproperties of electrodeposited Co on n-GaAs(001),”Journal of Electrochemical Society, Vol. 150, Issue 11,pp. C753–759, September 2004.

Lehman, S.Y., A. Roshko, R.P. Mirin, andJ.E. Bonevich, “Investigation of the shape of InGaAs/GaAs quantum dots,” Proceedings of MaterialsResearch Society, Vol. 737, 2003, (in press).

Nanostructure Fabrication Processes:Electrochemical Processing of Nanowires

Kongstein, O., U. Bertocci, and G.R. Stafford,“In situ stress measurements during copperelectrodeposition on (111)-textured Au,” (submitted),2004.

Wheeler, D., T.P. Moffat, G.B. McFadden,S.R. Coriell, and D. Josell, “Influence of a catalyticsurfactant on roughness evolution during film growth,”Journal of Electrochemical Society, 151 (8),C538–544, July 2004.

USAXS Imaging of BiomaterialsLevine, L.E., and G.G. Long, “X-ray imaging

with ultra-small-angle x-ray scattering as a contrastmechanism,” Journal of Applied Crystallography,(in press), 2004.

Measurements for CompoundSemiconductors

Davydov, A.V., L.A. Bendersky, W.J. Boettinger,D. Josell, M.D. Vaudin, C.S. Chang, and I. Takeuchi,“Combinatorial investigation of structural quality ofAu/Ni contacts on GaN,” Applied Surface ScienceJournal 223 (1–3), 24–29, 2004.

Lu, C.J., L.A. Bendersky, H. Lu, and W.J. Schaff,“Suppression of threading dislocations in InN thin filmsgrown on (0001) sapphire with a GaN buffer layer,”Applied Physics Letters 83, 2817, 2003.

Lu, C.J., A.V. Davydov, D. Josell, andL.A. Bendersky, “Interfacial reactions of Ti/n-GaNcontacts at elevated temperature,” Journal of AppliedPhysic, Vol. 94, No. 1, pp. 245–253, 2003.

Schenck, P., D.L. Kaiser, and A.V. Davydov,“High-throughput characterization of the opticalproperties of compositionally graded combinatorialfilms,” Applied Surface Science Journal, 223 (1-3),200, 2004.

Pb-free Surface Finishes: Sn Whisker GrowthBogustavsky, I., P. Bush, E. Kam–Lum, M. Kwoka,

J. McCullen, K. Spalding, K. Vo, and M.E. Williams,“NEMI tin whisker test method standards,” SMTAIProceedings, (submitted), 2003.

Pb-free Solder: Manufacturability andReliability

Allen, S.L., M.R. Notis, R.R. Chromik, R.P. Vinci,D.J. Lewis, and R.J. Schaefer, “Microstructuralevolution in lead-free solder alloy Part II: directionallysolidified eutectic tin–silver–copper, tin–copper andtin–silver alloys,” Journal of Materials Research,(submitted), 2004.

Bertocci, U., “An EQMB examination of Cu surfaceoxides in borate buffer,” Electrochemica Acta 49,1831–1841, 2004.

Handwerker, Carol, Jasbir Bath, ElizabethBennedetto, Edwin Bradley, Ron Gedney, Tom Siewert,Polina Snugovsky, and John Sohn, NEMI Lead-FreeAssembly Project: Comparison Between PbSn andSnAgCu Reliability and Microstructures, Benedetto,Proceedings of Surface Mount TechnologyInternational, Chicago, 2003.

Handwerker, C.A., U.R. Kattner, and K.-W. Moon,“Materials Science Concepts in Lead-Free Soldering,”Pb-Free Solders, edited by K. Suganuma, MarcelDekker, 2004.

Handwerker, C.A., E.E. deKluizenaar, K. Suganuma,and F.G. Gayle, “Major International Lead (Pb)-FreeSolder Studies,” Handbook of Lead-Free SolderTechnology for Microelectronic Assemblies, editedby K. Puttlitz and K. Stalter, McGraw Hill/IEEE,pp. 665–728, 2004.

Moon, K., and W.J. Boettinger, “Difficulties indetermining eutectic compositions: the Sn-Ag-Cuternary eutectic,” JOM 56 (4), 22–27, 2004.

54


Rae, A., and C.A. Handwerker, “NEMI’sCharacterization of Lead-Free Alloy Applicable toToday’s Commercially Available Alloys,” CircuitsAssembly, April 2004.

Schaefer, R.J., and D.J. Lewis, “Directionalsolidification in a AgCuSn eutectic alloy,” Journal ofElectronic Materials, (in press), 2003.

Electrical Properties of On-ChipInterconnections

Gayle, F., “FY2003 Programs and Accomplishments– MSEL Materials for Micro- and Optoelectronics,NISTIR 7019, 2004.

Josell, D., C. Burkhard, D. Kelley, Y.-W. Cheng,R.R. Keller, T.P. Moffat, Y. Li, B.C. Baker, C.A. Witt,and J.E. Bonevich, “Electrical properities of superfilledSub-100 nm silver metallizations,” Journal of AppliedPhysics, 96 (1), 759–768, 2004.

Josell, D., S. Kim, D. Wheeler, T.P. Moffat, andS.G. Pyo, “Quantifying superconformal filling of thesubmicrometer features through surfactant catalyzedchemical vapor deposition,” Proceedings of 2003 SpringMeeting of Electrochemical Society in Paris, France,PV 2003-8, pp. 98–103, 2003.

Josell, D., T.P. Moffat, and D. Wheeler, “An exact,algebraic solution for the incubation period of superfill,”Journal of the Electrochemical Society 151 (1),C19–C24, 2004.

Moffat, T.P., and D. Josell, “Electrodepositionof copper in the SPS-PEG-C1 additive system:I. kinetic measurements: influence of SPS,” Journal ofElectrochemical Society 151 (4), C262–C271, 2004.

Turchi, P.E.A., R.M. Waterstrat, R. Kuentzler,V. Drchal, and J. Kudrnovsky, “Electronic and phasestability properties of V – X (X = Pd, Rh, Ru) alloys,”Journal of the Physics of Condensed Matter, (in press),2004.

NanomagnetodynamicsLewis, A.C., D. Josell, and T.P. Weihs, “Stability

in thin film multilayers and microlaminates: the roleof free energy, structure, and orientation at interfacesand grain boundaries,” Scripta Materialia 48 (8),1079–1085, 2003.

McMichael, R.D., “A classical model of extrinsicferromagnetic resonance linewidth in ultra-thin films,”IEEE Transactions on Magnetics 40 (1), June 2004.

McMichael, R.D., “Ferromagnetic resonancelinewidth models for perpendicular media,” Journalof Applied Physics 95 (1), 1 June 2003.

Magnetic Materials for Sensors andUltra-high Density Data Storage

Buchanan, J.D., T.P. Hase, B.K. Tanner,C.J. Powell, and W.F. Egelhoff, Jr., “Interfaceintermixing and in-plane grain size in aluminum-transition metal bilayers,” Physical Review B,(submitted), 2004.

Chopra, H.D., Yang, D.X., Chen, P.J., andW.F. Egelhoff, Jr., “Domain structure andmagnetoeleastic properttes of Fe/Pd ferromagneticthin films and multilayers,” Journal of AppliedPhysics, (submitted), 2004.

Egelhoff, Jr., W.F., “Coercivities above 10 kOein CoPd superlattices,” Journal of Applied Physics,(submitted), 2004.

Egelhoff, Jr., W.F., L. Gan, H. Ettedgui, Y. Kadmon,C.J. Powell, P.J. Chen, A.J. Shapiro, R.D. McMichael,J. Mallet, T.P. Moffat, M.D. Stiles, and E.B. Svedberg,“Artifacts that mimic ballistic magnetoresistance,”Journal of Magnetism and Magnetic Materials,(submitted), 2004.

Egelhoff, Jr., W.F., L. Gan, H. Ettedgui, Y. Kadmon,C.J. Powell, P.J. Chen, A.J. Shapiro, R.D. McMichael,J. Mallet, M.D. Stiles, and E.B. Svedberg, “Artifacts inballistic magnetoresistance measurements,” Journal ofApplied Physics, 95, 7554, 2004.

Egelhoff, Jr., W.F., L. Gan, E.B. Svedberg,C.J. Powell, A.J. Shapiro, R.D. McMichael, J. Mallet,M.D. Stiles, and T.P. Moffat, “Artifacts that couldbe misinterpreted as ballistic magnetoresistance,”Digest of PMRC 2004, p. 248, 2004.

Hong, J.I., S. Sankar, A.E. Berkowitz, andW.F. Egelhoff Jr., “Perpendicular anisotropy in Co/Pdmultilayers,” Journal of Magnetism and MagneticMaterials, (submitted), 2004.

Huang, H., K. Seu, W.F. Egelhoff, Jr., and A. Reilly,“Ultrafast pump-probe laser spectroscopy of thehalf-metal CrO2,” Journal of Applied Physics,(in press), 2004.

Lee, C.G., J.-G. Jung, V.S. Gornakov,R.D. McMichael, A. Chen, and W.F. Egelhoff, Jr.,“Effects of annealing on the GMR and domain structurestabilization in a Py/Cu/Py/MnIr spin valve,” Journalof Magnetism and Magnetic Materials 272, 1887,2004.

Mallett, J.J., E.B. Svedberg, H. Ettedgui,T.P. Moffat, and W.F. Egelhoff, Jr., “A search forballistic magnetoresistance (BMR) in electrodepositedNi nanocontacts: feedback control used to selectthe contact resistance,” Applied Physical Letters,(in press), 2004.

55


Mallet, J., E.B. Svedberg, S. Sayan, A.J. Shapiro,P.J. Chen, W.F. Egelhoff, Jr., and T.P. Moffat,“Compositional control in electrodeposited CoPt films,”Electrochemical Solid State Letters, (in press), 2004.

Misra, R.D.K., S. Gubbala, A. Kale, andW.F. Egelhoff, Jr., “A comparison of the magneticcharacteristics of nanocrystalline Ni, Zn, and Mnferrites synthesized by reverse micelle technique,”Materials Science and Engineering, (in press), 2004.

Osofsky, M.S., R.J. Soulen, Jr., G. Woods,and W.F. Egelhoff, Jr., “Enhanced Tc near themetal/insulator transition: new insight into novelsuperconducting materials,” Physical Review Letters,(submitted), 2003.

Sanz, M., A. Papageorgopoulos, W.F. Egelhoff, Jr.,M. Nieto–Vesperianas, and N. Garcia, “Wedge-shapedabsorbing samples look left handed: the problem ofinterpreting negative refraction, and Its Solution,”Physical Review E 67, Art. No. 067601, 2003.

Svedberg, E.B., J.J. Mallett, H. Ettedgui, L. Gan,P.J. Chen, A.J. Shapiro, N. García, T.P. Moffat, andW.F. Egelhoff, Jr., “Possible ballistic magnetoresistance-like artifacts in electrodeposited nanocontacts,” Journalof Applied Physics, (in press), 2004.

Svedberg, E.B., J. Mallett, S. Sayan, A.J. Shapiro,W.F. Egelhoff, Jr., and T.P. Moffat, “Recrystallizationtexture and magnetic properties of electrodepositedFePt on Cu (001),” (submitted), 2004.

Woods, G.T., R.J. Soulen, Jr., I.I. Mazin,B. Nadgorny, M.S. Osofsky, J. Sanders, H. Srikanth,W.F. Egelhoff, Jr., and R. Datla, “Analysis of point-contact Andreev reflection spectra in spin polarizationmeasurements,” Physical Review B, (in press), 2004.

Yang, D.X., H.D. Chopra, P.J. Chen, andW.F. Egelhoff, Jr., “Mixed-surfactants as a novelapproach to atomic engineering of spin values,”IEEE Transactions of Magnetics, (submitted), 2004.

Yang, D.X., E.J. Repetski, H.D. Chopra, B.J. Spencer,D.C. Parks, P.J. Chen, and W.F. Egelhoff, Jr., “A pinholecoupling strength in giant magnetoresistance spinvalves: a statistical approach,” Physical Review B,(in press), 2004.

Electrodeposited Pt1–x (Fe,Co,Ni)x AlloysMallet, J., E.B. Svedberg, S. Sayan, A.J. Shapiro,

W.F. Egelhoff, Jr., and T.P. Moffat, “Compositionalcontrol in electrodeposition of FePt films,”Electrochemical Solid State Letters, (submitted), 2004.

Zheng, L.A., E.V. Barrera, and R.D. Shull,“Magnetic properties of the Co-C(d60) and Fe-C(d60)nanocrystalline magnetic thin films,” Journal ofApplied Physics, (submitted), 2003.

Magnetic Properties andStandard Reference Materials

Provenzano, V., J. Li, T.T. King, E.R. Canavan,P. Shirron, M.J. DiPorrp, and R.D. Shull, “Enhancedmagnetocaloric effects in R3(Ga1–xFe x)5O12(R = Gd, Dy, Ho; (O < x < 1) nanocomposites,”Journal of Magnetism and Magnetic Materials266, 185–193, 2003.

Provenzano, V., A.J. Shapiro, T.T. King,E.R. Canavan, P. Shirron, M. DiPirro, and R.D. Shull,“Peak magentocaloric effects in Al-Gd-Fe- alloys,”Journal of Applied Physics, (submitted), 2004.

Provenzano, V., A.J. Shapiro, and R.D. Shull,“Near-elimination of large hysteresis losses in theGd5Ge2S12 magnetic refrigerant compound by asmall alloying addition of iron,” Nature Journal,(submitted), 2004.

Shir, F., L.H. Bennett, E. Della Torre, C. Mavripis,L. Yanik, and R.D. Shull, “Details of the sequentialsteps in magnetocaloric regeneration,” IEEETransactions on Magnetics, (submitted), 2004.

Shir, F., E. Della Torre, L.H. Bennett, C. Mavripis,and R.D. Shull, “Modeling of magnetization anddemagnetization in magnetic regenerative refrigeration,”IEEE Transactions on Magnetics, (submitted), 2004.

Shir, F., E. Della Torre, L.H. Bennett, C. Mavripis,and R.D. Shull, “Modeling of magentization anddemagnetization in magnetic regenerative refrigeration,”IEEE Transactions of Magnetics, (submitted), 2004.

Shull, R.D., E. Quandt, and A.J. Shapiro, “MOIFobservations of domain motion in magnetostrictivematerials under stress,” Journal of Applied Physics,(submitted), 2004.

Taketomi, S., R.V. Drew, and R.D. Shull,Anomalous magnetic aftereffect of a frozen magneticfluid,” Journal of Applied Physics, (submitted), 2004.

Taketomi, S., R.V. Drew, and R.D. Shull,“Magnetic fluids’ phase transition between micelleand mono-dispersed phases,” Physical Review E,(submitted), 2004.

Taketomi, S., and R.D. Shull, “Experimentalverification of interactions between randomlydistributed fine magnetic particles,” Journal ofMagnetism and Magnetic Materials, (in press), 2003.

56


Turchinskaya, M.J., L.A. Bendersky, A.J. Shaprio,K.S. Chang, I. Takeuchi, and A.L. Roytburd, “Rapidconstructing magnetic phase diagrams by magneto-optical imaging of composition spread films,” Journalof Materials Research, (submitted), 2004.

Analysis of Structural Steel in theWorld Trade Center

Gayle, F.W., S.W. Banovic, T. Foecke, R.F. Fields,W.E. Luecke, J.D. McColskey, C. McCowan, andT.A. Siewert, “The Structural Steel of the World TradeCenter Towers,” Advanced Materials and Processes,(in press), October 2004.

Gayle, F.W., R.J. Fields, T.A. Siewert, andW.E. Luecke, “Interim Report on ContemporaneousStructural Steel Specifications,” June 2004 ProgressReport on the Federal Building and Fire SafetyInvestigation of the World Trade Center Disaster, NISTSpecial Publication 1000-5, pp. E–1 to E–61, 2004.

Banovic, S.W., “Interim Report on Inventory andIdentification of Steels Recovered from the WTCBuildings,” June 2004 Progress Report on the FederalBuilding and Fire Safety Investigation of the WorldTrade Center Disaster, NIST Special Publication1000-5, pp. F–1 to F–64, 2004.

Pipeline SafetyDatta, P.K., H.L. Du, J.S. Burnell–Gray, and

R.E. Ricker, “Corrosion of intermetallics,” ASM MetalsHandbook, Vol. 13, Corrosion, (in press), 2004.

Kolsky Bar (OLES Project)Ma, L., J. Song, E. Whitenton, and T. Vorburger,

“A bullet signatgure measurement system for NIST RM(Reference Material) 8240 standard bullets,” Journal ofForensic Sciences 49 (4), 649–659, 2004.

Rhorer, R., D. Basak, G. Blessing, T. Burns,M. Davies, B. Dutterer, R.J. Fields, M. Kennedy,L.E. Levine, E. Whitenton, and H. Yoon, “Kolsky barwith electrical pulse heating of the sample,” Proceedingsfor Society of Experimental Mechanics AnnualMeeting, (submitted), 2003.

Rhorer, R.L., L.E. Levine, T.J. Burns, R.J. Fields,H.W. Yoon, M.A. Davies, D. Basak, E.P. Whitenton,G.V. Blessing, B.S. Dutterer, and M.D. Kennedy,“Constitutive model data for machining simulation usingthe NIST pulse-heated kolsky bar,” Dislocations,Plasticity and Metal Forming, edited by A.S. Khan,R. Kazmi, and J. Zhou, Fulton, MD: NEAT Press,p. 103, 2003.

Yoon, H.W., D. Basak, R. Rhorer, E. Whitenton,T. Burns, R.J. Fields, and L.E. Levine, “Thermalimaging of metals in a kolsky-bar apparatus,”Thermosense XXV, edited by E. Cramer andX.P. Maldague, Proceedings of SPIE, 5073,284–294, 2003.

Marine forensicsHooper, J.J., T.J. Foecke, L. Graham, and

T.P. Weihs, “Metallurgical forensic analysis of wroughtiron from the RMS titanic,” Marine Technology,SNAME N40(2), pp. 73–81, April 2002.

Kinetic Penetrators and Amorphous MetalsBiancaniello, F.S., T.F. Zahrah, R.D. Jiggetts,

L. Kecskes, L.J. Rowland, S.P. Mates, andS.D. Ridder, “Structure and Properties of ConsolidatedAmorphous Metal Powder,” Powder Materials:Current Research and Industrial Practices III, editedby F.D.S. Marquis, Warrendale, PA: TMS, 2003.

OtherBasak, D., R.A. Overfelt, and D. Wang,

“Measurement of specific heat capacity and electricalresistivity of industrial alloys using pulse heatingtechniques,” International Journal of Thermophysics,24 (6), 1721–1733, 2003.

Benderksy, L.A., I.D. Fawcett, and M. Greenblatt,“TEM study of two-dimensional incommensuratemodulation in layered La2–2x Ca1+2x Mn2O7(0.6 < x < 0.8),” Chemical Materials, (submitted), 2004.

Bendersky, L.A., and I. Takeuchi, “Use oftransmission electron microscopy in combinatorialstudies of functional oxides,” Macromolecular RapidComm., 25 (6), 695, 2004.

Bobordis, K., A. Seifter, A.W. Obst, and D. Basak,“Radiance temperatures and normal spectral emittances(in the wavelength range of 1500 nm to 5000 nm) oftin, zinc, aluminum, and silver at their melting points bya pulse-heating technique,” International Journal ofThermophysics, (in press), 2003.

Bobordis, K., A. Seifter, A.W. Obst, and D. Basak,“Validation of a new high-speed fiber-coupledfour-wavelength infrared pyrometer in the range of505 k to 1234 k by a pulse-heating technique, usingthe melting points of tin, zinc, aluminum, and silveras reference points,” International Journal ofThermophysics, (in press), 2003.

57


Boettinger, W.J., C.E. Campbell, T. Hansen,P. Merewether, and B.A. Mueller, “Properties ofcomplex inorganic solids,” edited by P.E.A. Turchi,A. Gonis, A. Meike, and K. Ragan, New York: KluwerAcademics/Plenum Publishers, (in press), 2003.

Boettinger, W.J., D. Josell, S.R. Coriell, andD. Basak, “Rapid melting of Nb – 47 mass % Ti:effect of heating rate and grain size,” SolidificationProcesses and Microstructures, edited by Rappaz,Beckermann and Trivedi, Warrendale, PA: TMS,pp. 87–98, 2004.

Chang, K.-S., M. Aronova, C.-L. Lin,M. Murakami, M.H. Yu, J. Hattrick–Simpers,O.O Famodu, S.Y. Lee, R. Ramesh, M. Wuttig,I. Takeuchi, C. Gao, and L.A. Bendersky, “Explorationof artificial multiferroic thin-film heterostructures usingcompositon spreads,” Applied Physical Letters, 84(16), 3091, 2004.

Famodu, O.O., J. Hattrick–Simpers, M. Aronova,K.-S. Chang, M. Murakami, M. Wuttig, T. Okazaki,Y. Furuya, L. Knauss, L.A. Bendersky, F.S. Biancaniello,and I. Takeuchi, “Combinatorial investigation offerromagnetic shape-memory alloys in the Ni-Mn-Alternary system suing a composition spread technique,”Materials Transactions, 45, 173, 2004.

Kramer, D.E., M.F. Savage, A. Lin, andT.J. Foecke, “Novel method for TEM characterizationof deformation under nanoindents in nanolayeredmaterials,” Scripta Materialia, 50 (6), 745–749, 2004.

Lu, C.J., L.A. Bendersky, K. Chang, andI. Takeuchi, “HRTEM study on the extended defectstructure of epitaxial Ba0.3Sr0.7TiO3 thin films grownon (001) LaAl03,” Proceedings of MRS Symposium,Vol. 751, 23.3, 2003.

Mates, S.P., F.S. Biancaniello, and S.D. Ridder,“NIST experiments in gas atomization 1986–1999,”Proceedings of the 2004 International Conference onPowder Metallurgy and Particulate Materials, MPIF,Princeton, NJ, (submitted), 2004.

Mates, S.P., F.S. Biancaniello, and S.D. Ridder,“The effect of swirl on gas velocity decay in a genericannular close-coupled nozzle,” Powder Materials:Current Research and Industries Practices III, editedby F.D.S. Marquis, Warrendale, PA: TMS, 2003.

Mates, S.P., and G.S. Settles, “A study of liquidmetal atomization using close-coupled nozzles, part 1:gas dynamics,” Atomization and Sprays, (in press),2003.

Mates, S.P., and G.S. Settles, “A study of liquidmetal atomization using close-coupled nozzles, part 2:atomization behavior,” Atomization and Sprays,(in press), 2003.

Ricker, R.E., “Modeling the influence of crackpath deviations on the propagation of stress corrosioncracks,” Hydrogen Effects on Materials Behaviorand Corrosion Deformation Interactions, edited byN.R. Moody, A.W. Thompson, R.E. Ricker, G.S. Was,and R.H. Jones, Warrendale, PA: TMS, pp. 629–638,2003.

Ricker, R.E., and A.K. Vasudevan, “The influenceof grain boundary precipitation on the stress corrosioncracking of Al-Li and Al-Li-Cu alloys,” HydrogenEffects on Materials Behavior and CorrosionDeformation Interactions, edited by N.R. Moody,A.W. Thompson, R.E. Ricker, G.S. Was, andR.H. Jones, Warrendale, PA: TMS, pp. 927–935, 2003.

Stoudt, M.R., and R.E. Ricker, “Electrochemicalevaluation of a corrosion fatigue failure mechanism ina duplex stainless steel,” Metallurgical and MaterialsTransactions A 35 (A) (8), 2427–2437, 2004.

Takeuchi, I., W. Yang, K.-S. Chang, M. Aronova,T. Venkatasan, R.D. Vespute, and L.A. Bendersky,“Monolithic multi-channel UV detector arrays andcontinuous phase evolution in MgxZn1–xO compositionspreads,” Journal of Applied Physics 94, 7736, 2003.

Tsuda, T., C.L. Hussey, and G.R. Stafford,Electrodeposition of Al-Mo Alloys from the lewis acidicaluminum chloride-1 ethyl-3-methylimidazolium chloridemolten salt,” Journal of Electrochemical Society,Vol. 151, Issue 8, pp. 379, 2004.

Tsuda, T., C.L. Hussey, G.R. Stafford, andO. Kongstein, “Electrodeposition of Al-Zr alloys fromthe lewis acidic aluminum chloride-1ethyl-3-methylimidazolium chloride melt,” Journal ofElectrochemical Society, 151, C445, 2004.

58


59

Metallurgy Division

Metallurgy Division

ChiefCarol A. HandwerkerPhone: 301–975–6158E-mail: [email protected]

Deputy ChiefFrank W. GaylePhone: 301–975–6161E-mail: [email protected]

NIST FellowsWilliam J. BoettingerPhone: 301–975–6160E-mail: [email protected]

John W. CahnPhone: 301–975–5664E-mail: [email protected]

Group LeadersElectrochemical Processing GroupGery R. StaffordPhone: 301–975–6412E-mail: [email protected]

Magnetic Materials GroupRobert D. ShullPhone: 301–975–6035E-mail: [email protected]

Materials Performance GroupStephen D. RidderPhone: 301–975–6175E-mail: [email protected]

Materials Structure and Characterization GroupFrank W. GaylePhone: 301–975–6161E-mail: [email protected]

Metallurgical Processing GroupStephen D. RidderPhone: 301–975–6175E-mail: [email protected]

60

Research Staff

Research Staff

Drew, Rosetta [email protected] maintenance

Egelhoff Jr., William [email protected] thin filmsUltrasoft magnetsMagnetic sensorsUltrahigh density data storage media

Fields, Richard [email protected] propertiesMechanical testingPowder consolidationMetal forming

Fink, James [email protected] standardsHardness testingMechanical properties measurementMetal powder consolidationEnvironmental corrosionMetallography

Foecke, Timothy [email protected] forming standardsNanostructured materialsExperimental fracture physicsSEM, TEM, SPMMicro- and nano-mechanics of materialsDislocation-based deformation mechanismsHistorical metallurgy and failure analysis

Gates, Hilary [email protected] reference materials

Gayle, Frank [email protected]/property relationshipsTransmission electron microscopyIntrastructural materials

Guyer, Jonathan [email protected]–V semiconductorsMolecular beam epitaxyAlloy film depositionStress-induced mass transport

Banovic, Stephen [email protected] formingMechanical propertiesTexture analysis

Beauchamp, Carlos [email protected] reference materials

Bendersky, Leonid [email protected] ceramicsAnalytical and high-resolution TEMMetal glassesCombinatorial methods

Biancaniello, Frank [email protected] deposition measurements and diagnosticsInert gas atomization: metal powder

measurements and consolidationNitrogenated steels, standard reference materialsSpecial alloys, bulk metallic glassesBiomaterials

Boettinger, William [email protected] of alloy microstructures to processing

conditions; solidificationThermal analysisSolder

Bonevich, John [email protected] holographyInterfacial structure and chemistryHigh resolution/analytical electron microscopyMagnetic materials

Campbell, Carelyn [email protected] liquid phase bondingMulticomponent diffusions simulationsAlloy design methodology

Claggett, Sandra [email protected] preparation for electron microscopyDigital imagingSpecimen preparation for scanning electron

microscopyImage analysisMetallography of materials

61

Research Staff

Handwerker, Carol [email protected] thermodynamics and kineticsSolder scienceMicrostructure evolution

Iadicola, [email protected] propertiesMechanical testingStress measurementMetal formingShape memory alloys

Jiggetts, Rodney [email protected] analysisGrit blastingRoughness measurementsMetallography of coatingsImage analysisMicro analysis/characterizationHot isostatic processing

Johnson, Christian [email protected] coatingsElectroless deposition processesMetallic glass alloy depositionMicrohardness SRM researchChromium depositionPulse alloy depositionLead-free solder

Johnson, [email protected] magnetic materialsNanostructured metallic alloysMagnetocaloric materialsMagnetic properties measurementsMetallographic analysis

Josell, [email protected] and structural properties

of multilayer materialsElectrical properties of submicrometer

interconnectsFilling of sub-100 nm features by

electrodeposition

Kattner, Ursula [email protected] thermodynamicsAlloy phase equilibria evaluationsSolder alloy systemsSuperalloy systemsMetal hydrogen systems

Kelley, David [email protected] SRM developmentDye penetration SRM developmentPrecious metal electrodepositionPlating on aluminumElectroplating

Levine, Lyle [email protected] theoryDislocation-based deformationSynchrotron x-ray techniquesPercolation theoryTEM, SEM, SPMKolsky bar techniquesNanomechanicsMultiscale modeling

Low II, Samuel [email protected] standardsHardness testingMechanical properties of materialsMechanical testing of material

Luecke, William [email protected] mechanical testingFire-resistant steelCreep of ceramicsSilicon nitride

Maranville, Brian [email protected] resonancePerpendicular magnetic mediaAtomic force microscopy

Mates, Steven [email protected] fluid flowMetal spraysFluid flow visualization

McMichael, Robert [email protected] magnetoresistanceFerromagnetic resonanceNanocomposites

Moffat, Thomas [email protected] probe microscopyNanostructuresSurfactant mediated growth (ADD)

62

Research Staff

Shull, Robert [email protected] susceptibilityMossbauer effectX-ray and neutron diffractionMagneto-caloric and magneto-optical effectsMagnetic domain imaging

Stafford, Gery [email protected] transientsElectrodepositionMolten salt electrochemistryStress in thin films (ADD)

Stoudt, Mark [email protected] surface roughnessMechanical propertiesMetal fatigueThin filmsEnvironmentally assisted cracking

Warren, James [email protected] simulations of solidificationDendrite pattern formationModeling of solder/substrate wetting processesModeling grain boundaries

Williams, Maureen [email protected] thermal analysisPowder x-ray diffractionSolder WettabilitySEM

Pierce Jr., Thomas [email protected] measurementsMetallography

Pitchure, David [email protected] testingComputer-aided designMechanical engineering

Rantschler, James [email protected] resonanceMagnetic characterizationDamping formalisms

Ricker, Richard [email protected] mechanismsDeformation mechanismsStress corrosion cracking (SCC)Hydrogen embrittlementCorrosion fatigueDynamic mechanical analysis

Ridder, Stephen [email protected] powder production and characterizationMetal spray processingProcess metallurgy, modeling, and control

Shapiro, Alexander [email protected] domain imagingMossbauer effectScanning electron microscopy (SEM) and

X-ray microanalysisImage analysis

63

Organizational Charts

Organizational Charts

Materials Science and Engineering Laboratory

PolymersE.J. Amis, Chief

C.R. Snyder, Deputy

MetallurgyC.A. Handwerker, Chief

F.W. Gayle, Deputy

CeramicsD.L. Kaiser, Chief

R.G. Munro, Deputy

Materials ReliabilityA.F. Clark, Chief

T.A. Siewert, Deputy

NIST Center forNeutron Research

P.D. Gallagher, Director

DirectorDeputy Director

National Institute of Standards and Technology

DirectorBoulder Laboratories

Baldrige National QualityProgram

Director forAdministration

& Chief FinancialOfficer

TechnologyServices

AdvancedTechnologyProgram

ManufacturingExtension

Partnership

ManufacturingEngineeringLaboratory

Chemical Scienceand Technology

Laboratory

Materials Scienceand Engineering

Laboratory

Electronics andElectrical Engineering

Laboratory

InformationTechnologyLaboratory

PhysicsLaboratory

Building andFire ResearchLaboratory

L.E. Smith, DirectorS.W. Freiman, Deputy Director

on the cover · this report describes the major technical activities and accomplishments in the...

Documents