the Exploratory Studies Facility iscompleted at Yucca Mountain, we willproceed with more extensive integratedtests to validate our models andmethods.

Key Words: engineered barrier system(EBS); high-level radioactive waste; spentfuel; Yucca Mountain Project.

References1. R. B. Stout and H. R. Leider,

Preliminary Waste Form CharacteristicsReport, Lawrence Livermore NationalLaboratory, Livermore, CA, UCRL-ID-108314 (Rev. 1) (1994).

2. Initial Summary Report forRepository/Waste Package AdvancedConceptual Design, TRWEnvironmental Safety Systems,Document No. 00000000-01717-5705-00015, Rev. 1 (August 29, 1994).

3. R. A. Van Konynenburg et al.,Engineered Materials CharacterizationReport for the Yucca Mountain SiteCharacterization Project, LawrenceLivermore National Laboratory,Livermore, CA, UCRL-ID-119564(Vols. 1, 2, and 3) (1995).

4. D. G. Wilder, Preliminary Near-FieldEnvironmental Report, Volume I:Technical Bases for EBS Design;Volume II: Scientific Overview of Near-Field Environment and Phenomena,Lawrence Livermore NationalLaboratory, Livermore, CA, UCRL-LR-107476 (Vols. 1 and 2) (1993).

16

Science & Technology Review March 1996

Nuclear Waste Repository

For further information contact Bill Clarke (510) 423-4571(clarke1@llnl.gov).

WILLIS L. CLARKE received his B.S. in metallurgicalengineering from the University of Nevada, Reno, in 1960. Hejoined the Laboratory’s Chemistry and Materials ScienceDepartment in 1989 after holding several research andmanagerial posts, including lengthy service as principalengineer at the Vallecitos Nuclear Center at Pleasanton,California. Since 1991, he has been project leader at LawrenceLivermore for the Yucca Mountain Site CharacterizationProject, managing a staff of from 40 to 80 researchers

focusing on the design of an engineered barrier system for a permanent nuclearwaste repository. He has published more than 80 articles on materials performance,including the effects of radiation, oxidation, and corrosion on metals and alloys.

About the Scientist

Integrated Corrosion Facility

To project effects on candidate metals over a hundred centuries, we need to dolaboratory corrosion testing for as long as possible, several years at least. Our newcorrosion-testing laboratory at Livermore (Building 435) allows us to investigate modesof degradation in candidate materials for the required times.

This facility contains several dozen large tanks approximately 1 meter square and 2 meters high in which we can simulate conditions that are possible at a repository. Testsolutions are varied and controlled for temperature, pH (acidity), solution chemistry, andmany other variables. Metal samples are immersed in the aqueous solutions or subjectedto the vapor phase to study generalized, localized, and stress-assisted corrosion.

Some samples will be exposed for five years or more, still just a fraction of the timethe material must last in the repository. To measure changes in corrosion rates, we willremove samples of candidate materials at six-month intervals for kinetic andmechanistic analysis. Some of our exposure conditions, such as electrochemicalpolarization, intentionally accelerate the corrosion process. For different exposureconditions, we use computer models to project corrosion effects to much longer times.Thus, the effects we assess can correspond to the vastly longer exposure times in anunderground repository.

17

In the absence of nuclear testing, the Laboratory’s diamond anvil cell ishelping to assure the safety and reliability of our nation’s nuclearstockpile. Because it uses very small samples, the diamond anvil

cell is a cost effective way to collect accurate, reliable data about the physical and chemical behavior

of weapons materials under the ultrahigh pressures encountered in an imploding

nuclear weapon without the possibility of radioactive

contamination.

The Diamond Anvil Cell:Probing the Behavior of Metals

under Ultrahigh Pressures

OW materials behave under extremeconditions is of more than scientific interest

to Livermore researchers. Issues related to nationalsecurity are a major motivation. During theimplosion of a nuclear weapon, the materials aredriven inward, reaching enormously high pressuresand temperatures, until they achieve thesupercritical state that is necessary for nuclearfission. During the process, the ultrahighcompressions subject the weapon’s materials to

continual change in physical properties such asvolume, structural state, and density. These changesstrongly affect the course of the implosion andtherefore the final explosion. Weapon designersneed to know exactly what those material propertiesare and how they change during the implosionprocess if they are to calculate and reliably predictthe performance of a weapon. However, the greatviolence and brevity of a nuclear event combine toinhibit the collection of precise data.

H

physicists with the experimental datathat allow them to improve thecalculations upon which weaponscodes are based without doing actualnuclear tests.

The Compressing MechanismThe diamond anvil cell is a small

mechanical press that forces the small,flat faces (the culets) of two flawless,brilliant-cut diamonds together on amicrogram-size sample to create veryhigh pressures in the sample (seeFigure 2).1 It uses diamonds because,as the hardest known solid, they do notbreak or deform under the intensepressures of the DAC and aretransparent to light and x rays. Themechanism for applying the pressure isa stout lever with a mechanicaladvantage of 10:1. It is actuated by aheavy screw and Belville springs at thelong end. (Belville springs are cuppedwashers stacked back to back aroundthe screw to apply a balancedpressure.) The diamonds, which rangefrom one-eighth to one-third carateach, are in an opposed anvilconfiguration and mounted overzirconium pads on a pair oftungsten–carbide rockers. Theserockers (hemicylinders with their axesat right angles) can be tilted to alignthe culet faces perfectly parallel.Apertures in the rockers permit x raysand other kinds of radiation to enterand exit through the diamond anvils,thus allowing for diagnostics andheating during experiments.

We customize the surface shapes ofthe diamonds for the pressures atwhich we perform experiments. Forexperiments at pressures below500,000 atmospheres (50 GPa), eachdiamond is ground to have a flat facethat ranges from 100 µm(micrometers) to 500 µm in diameter;for experiments at still higher

pressures, we use beveled diamondshaving a 7- to 8.5-degree bevel on a 300-µm culet with a 30- to 75-µmflat face. (As points of reference, astandard sheet of paper is about 50 µmthick and a human hair is about 100 µm in diameter.)

Once the diamonds are perfectlyaligned, we remove the tight-fittingpiston that holds one of the twodiamonds in place. Between the culetsof the anvils, we place a 250-µm thickgasket (a strip or circular metal disc oftool steel or rhenium) and apply a smallforce to indent or prepress its surface.Then we drill a hole that is 30 to 150 µm in diameter in the center of theindented area. Into that hole we placethe sample with a pressure medium—liquid, gas, or solid—which helps todistribute the compressive force of thediamond faces.

To calibrate pressure during theexperiment, we add a pressure marker,such as a small ruby chip or platinumpowder. Under illumination of ahelium–cadmium laser, the ruby chipemits fluorescent light at characteristicfrequencies (spectral lines), the

wavelengths of which are calibrated asa function of pressure against a knownmarker material. The volume of theplatinum under pressure can becalculated from the x-ray latticeparameters and compared with theknown pressure–volume relationshipfrom shock-wave data in order toascertain the sample pressure. (Thepressure marker acts as a pressuresensor and also indicates when theapplied stress becomes nonuniform.)When the pressure is no longerhydrostatic, because, say, a fluidpressure medium has become a solid orhas become very viscous, the resultingnonuniform stress broadens the rubyfluorescent peaks.

The DiagnosticsA significant advantage of the DAC

is that diamonds are transparent tox rays and visible light. We exploit thisfeature when we watch the changes inthe material as the pressure andtemperature are changed. To determinethe sample material’s crystal structureduring an experiment, we collimate thex-ray beam, selecting rays nearly

19

Science & Technology Review March 1996

Diamond Anvil Cell

Until roughly two decades ago, theonly alternative to nuclear tests formeasuring the properties of materialsat ultrahigh pressures and temperatureswas shock experiments—shock waveswere driven through the material ofinterest while changes in the materialproperties during passage of the shockfront were measured. However,because shock techniques are dynamic,precise material properties are difficultto measure directly. Instead, thediagnostics were focused onmeasurements that could be capturedin such brief durations; then, usinglarge-scale numerical simulations thatincorporated data from theexperiments, researchers inferred theproperties of interest.

The Diamond Anvil Cell

The diamond anvil cell (DAC) haschanged these circumstances becauseof the pressure and temperatureregimes to which a sample can besubjected. It joined shock experimentsand tests driven by high explosives asmeans of providing the experimentaldata that are important starting pointsfor science-based stockpilestewardship. This apparatus enablesLawrence Livermore researchers tomeasure many of the properties ofinterest directly under static pressureconditions (instead of indirectly as indynamic shock-wave experiments). Theuse of static pressure means thatultrahigh pressures can be maintainedfor significantly longer times than inshock experiments, allowing moreaccurate measurements to be takendirectly. Pressures within the diamondanvil cell can approach 350 gigapascals(1 GPa = ~ 10,000 atmospheres*) and

temperatures can approach 6,273 kelvin(10,832°F, 6,000°C)—that is, pressureand temperature equal to those at thecenter of the Earth.

The DAC is also more cost effectivethan shock-wave experiments. Instead ofproviding only one volume–densitynumber at a given pressure perexperiment, it provides a range of dataacross the pressure spectrum of theexperiment and thus more informationfor fewer experiments. Anotheradvantage of the DAC is the smallsample size needed. Each experimentrequires about a microgram of anelement, significantly less than in aweapon. The small samples presentminimal possibility of radioactivecontamination, and containment of thesmall amount of radiation is assured.

The DAC’s capabilities areparticularly important for weaponsphysicists now that the United States isno longer conducting nuclear tests. Thesafety and reliability of nuclearweapons must now be maintained withindirect experimental techniques andlarge-scale computations. In particular,the DAC enables direct measurementsof changes in volume and density, as afunction of changes in the material’scrystal structure and of melting underhigh pressure, that strongly influencethe hydrodynamic stability ofimploding systems. Fifty years ago,instability was an intractable problemfor the designers of the first nuclearweapons. Despite major advances inscience and technology, ourunderstanding of instability remainslimited because the actual physicalstate of the material experiencing thesechanges in volume and density oftencould not be measured. The DAC nowchanges this situation. It can provide

some of the data required to accuratelypredict the yield and performance ofnuclear weapons—and thus their safetyand reliability—without nuclearweapons tests.

LLNL’s physicists also use theDAC data to interpret the datacollected from earlier shock-waveexperiments. Shock waves passingthrough a material raise its pressureand temperature simultaneously,making it difficult for researchers toidentify with certainty the separateeffects of pressure and temperaturealone from the data. By staticallycompressing the same type of materialat room temperature in the DAC, wecan isolate the effects of pressure onthe changing pressure–densityrelationship (i.e., equation of state) ofthe material. Physicists then use thesedata to calculate the temperaturecomponent from the shock data andthus derive separate pressure andtemperature values for those data.They thus deduce further informationabout the high-temperature equation ofstate and phase stabilities useful toweapons physicists in confirming ormodifying the complex theoreticalcalculations upon which weaponscomputer codes are based.

Figure 1 shows a comparison ofDAC data with data from shockexperiments recalculated using atheoretical equation-of-state model foruranium. The slight discrepancybetween the DAC and theoreticalequations of state suggests that theparameters chosen for the theoreticalcalculations may need further minormodifications that could lead to moreaccurate predictions of weapon safetyand yield. The DAC is thus animportant tool that provides weapons

18

Science & Technology Review March 1996

Diamond Anvil Cell

* 1 atmosphere = the ambient air pressure at sea level.

0

20

40

60

80

100

20 22 24 26Crystal density, g/cm3

Pre

ssur

e, G

Pa

Figure 1. Comparisonof diamond anvil celldata (•) with valuescalculated from atheoretical equation-of-state model for uraniumusing data from shockexperiments (▲). Theplot shows that the dataderived from theory andshock experimentswarrant correction bythe data from thediamond anvil cell.

means of “seeing” the changes in thecrystal structure of the sample andcollecting data about its changingequation of state under the intensepressure of the DAC.

Commonly, phase transformationsare thought of as those from a solid toa liquid to a gas. However, there aretransformations from one solid toanother, and these are the structuraltransformations generally studied usingthe DAC. In solid-to-solid structuralchanges, the atoms of an elementrearrange themselves in response tochanging pressure, changingtemperature, or both to newconfigurations. The shape of theatomic structural “cages” changes bythe rearrangement of the atoms.Structural changes can be accompaniedby a sudden volume change. However,

the volume change can be smallenough not to be recognized or to beable to be accounted for by a normalmargin of experimental error. It canalso be smooth and gradual and notexhibit the spikes associated withlarge, sudden changes. Whether subtleor sharply defined, these are thestructural transformations of interest inDAC experiments.

The diagnostic x rays used to recordthese data in our DAC experiments atultrahigh pressures are not like thosefrom medical or conventionallaboratory x-ray units, which are tooweak to yield data in a reasonable timeand cannot be collimated sufficiently tocollect accurate data. Rather, we use thevery bright, highly coherent x rays froma synchrotron source such as the one atthe National Synchrotron Light Source

at Brookhaven National Laboratory inNew York State and collimate them to 5 to 10 µm in diameter. A combinationof high beam intensity and excellentcollimation is essential to reduce thetime required for data collection (10 to30 minutes at each pressure, rather thantens of days) as well as to reduce theeffects that the pressure gradient acrossthe sample has on the data.

When pressures exceed 40 gigapascals (GPa), we use theapparatus shown schematically in Figure 4 to record the diffraction pattern.First, we use a pair of adjustable slits tocollimate the beam from the synchrotronx-ray source to a diameter of less than10 µm. Then we clamp the DAC, withsample and ruby-chip pressure marker inplace, to a four-circle goniometer* inorder to align the DAC with respect to

21

Science & Technology Review March 1996

Diamond Anvil Cell

parallel to one another with a slitsystem.2 We pass the well-collimatedbeam of monochromatic (single-energy)x rays from a rotating anode generatorthrough the sample and both diamondsand record the resulting diffraction

pattern on x-ray film (see Figure 3).Efficient computer programs interpretthe resulting patterns, which consist of acomplex series of concentric arcs orreflections in a spectrum. These x-raydiffraction patterns thus become the

20

Science & Technology Review March 1996

Diamond Anvil Cell

~715,000 atmospheres

Figure 3. The experimentally interpreted structural sequence for uranium at increasing pressures recorded as an x-ray diffraction patternby a collimated x-radiation experiment. It is x-ray diffraction patterns like these, but at much higher pressures, that provide the diamondanvil cell data of use to weapons physicists.

1 atmosphere

* A goniometer is an instrument with a number of degrees of freedom to move a crystal inspace and uses x-ray diffraction to measure the angular positions of the axes of a crystal.

Hardenedsteel piston

Zirconium pad

Belville spring

Tungstencarbide rocker

Ruby

Metal gasket

Diamond

Sample

Pressure medium

Belvillespring

Press

Piston

Cylinder

Piston diamondand culet

Tungstencarbide

rocker Gasket toolsteel or rhenium

3-micrometer hole for sample and

pressure medium

Figure 2. (a) Schematic of thediamond anvil cell. (b) The celldisassembled, showing themajor optical and mechanicalcomponents. (c) Close-up of theapparatus for holding andcompressing the sample.

(a)

(b)

(c)

three categories, we are in search ofdata about the stability—or lack of it—ineach crystal structure and the equationof state under ultrahigh pressures. Weare looking for structural changes as afunction of pressure and temperature,changes in volume (density) due to thestructural changes, the ultimatestructural form that is stable for theseelements, and the similarities betweenlanthanides and actinides. These are thedata that physicists require incombination with shock-wave-deriveddata to confirm or modify the theoryconcerning the high pressure behaviorof these metals upon which weaponscode calculations for uranium andplutonium are based. These DAC datacan improve the precision of the

computer codes for the behavior ofweapons materials and thereby improvethe predictability of their structuralbehavior in the weapons regime.

The LanthanidesThe lanthanides, or rare-earth series

of elements (elements 57 through 71 ofthe periodic table—lanthanum throughlutetium), are nearly indistinguishable intheir chemical behavior. Although theyall have the same outer electronic shellconfiguration, each element has onemore electron than its next lighterneighbor. This additional electron islocated deep within the atom’s electronstructure. This configuration causes asmooth progression of physicalproperties across the series but has little

effect on chemical properties. Thenormal (unpressurized) crystal structuresof these elements (Figure 6a) show aregular progression across the series.

We studied the lanthanides in theDAC primarily to confirmexperimentally the broadly relatedpattern of the elements’ crystalstructure across the series predicted bytheoreticians.

Our detailed studies of some rare-earth elements have experimentallyconfirmed the existence of thestructural sequence predicted bytheoreticians. As pressure increased,the lanthanides transformed to face-centered cubic and a six-layeredstructure (Figure 6b). Under increasingpressure, the lanthanides follow the

23

Science & Technology Review March 1996

Diamond Anvil Cell

the x ray. Aided by the ion chamber, wealign the DAC so the 10-µm-diameter x-ray beam probes the center of the sample(the area of greatest pressure and theleast pressure gradient). The penta prismis essentially a microscope that directslight so that we can see the sample priorto experiments. The x-ray beam from thesynchrotron source passes though thediamonds, diffracts from the sample, andpasses through the pinhole collector inthe upper part of the apparatus. It thenenters the germanium–lithium detector,which records the energy diffractionpattern from the sample, data essential toidentify the changing crystal structure.The optical multichannel analyzer in thelower part of the apparatus determinesthe pressure at which the crystal changestake place by measuring the laser-induced fluorescent light from the ruby-chip pressure marker. Thus, the changingvolume and density of the sample aremeasured as a function of pressure.

Heavy Metals Experiments

As part of our continuinginvestigation into the high-pressureproperties of metals, we have used the diamond anvil cell to determine the pressure–volume relationship andany possible changes in the crystalstructure for some actinide andlanthanide metals to approximately325-GPa pressure at room temperature.Figure 5 shows the lanthanide andactinide series from the periodic tableof the elements; shading highlightsthose elements we have studied in somedepth. The Laboratory is the worldleader in the study of lanthanides andactinides under extreme static pressureand temperature conditions.

One purpose of these investigationsis to obtain consistent, thorough data ofgeneral scientific interest about theproperties of these metals underpressure. Another is to study the

behavior of the actinide weapons metalsuranium and plutonium under pressuresapproaching those in imploding nuclearweapons. These purposes, however, arenot separate. In theory and reality, thereare connections between the high-pressure behavior of elements in bothseries that are of particular relevance tothe high-pressure behavior of theactinide weapons metals uranium andplutonium.3 Representative DACfindings about lanthanides and actinidesillustrate how DAC research works ingeneral and how it contributes toweapons safety in the absence ofnuclear testing.

Our findings concerning thelanthanides and actinides to date fallinto three categories: those concerningthe lanthanides, those concerning theheavy actinides (americium through theend of the series), and those concerningthe lighter actinides (thorium, uranium,neptunium, and plutonium).* In all

22

Science & Technology Review March 1996

Diamond Anvil Cell

* We have avoided DAC experiments with protactinium because it is too radioactive even in the small quantities needed for our work.

IA

IIA 4

Be

IIIB

IVB VB VIB VIIB VIIIB

IB IIB

IIIA IVA VA VIA VIIA

0

3 Li

1 H

11 Na

19 K

37

Rb

55 Cs

87 Fr

88 Ra

39 Y

21 Sc

104 (Rf) (261)

72 Hf

178.49

40 Zr

22 Ti

105 Unp

107 Uns

73 Ta

41 Nb

23 V

106 Unh

74 W

42 Mo

24 Cr

75 Re

43 Tc

25 Mn

76 Os

44 Ru

26 Fe

77 Ir

45 Rh

27 Co

78 Pt

46 Pd

28 Ni

79 Au

47 Ag

29 Cu

80 Hg

48 Cd

30 Zn

81 Tl

49 In

31 Ga

82 Pb

50 Sn

32 Ge

83 Bi

51 Sb

33 As

84 Po

52 Te

34 Se

85 At

53 I

35 Br

86 Rn

54 Xe

36 Kr

13 Al

14 Si

15 P

16 S

17 Cl

18 Ar

5 B

6 C

7 N

8 O

9 F

10 Ne

2 He

56 Ba

38 Sr

58 Ce

140.12

90 Th

232.04

92 U

238.03

60 Nd

144.24 93

Np (237)

61 Pm (145)

94 Pu (244)

62 Sm 150.4 95

Am (243)

63 Eu

151.96 96

Cm (247)

64 Gd

157.25 97

Bk (247)

65 Tb

158.93 98

Cf (251)

66 Dy

162.50 99

Es (252)

67 Ho

164.93 100

Fm (257)

68 Er

167.26 101

Md (258)

69 Tm

168.93 102

No (259)

70 Yb

173.04 103

Lr (260)

71 Lu

174.97 91

Pa (231)

59 Pr

140.91

20 Ca

12 Mg

Lanthanides

Actinides

Cerium

Praseodym

ium

Neodymium

Promethium

Samarium

Europium

Gadolinium

Terbium

Dysprosiu

m

Holmium

ErbiumThulium

Ytterbium

Lutetium

Thorium

Protactinium

Uranium

Neptunium

Plutonium

Americium

CuriumBerkelium

Californium

Einsteinium

Fermium

Mendelevium

Nobelium

Lawrencium

57 La

138.91 89

Ac (227)

Figure 5. All of the elementsin the lanthanide and actinideseries have a complex atomicstructure. Elements shadedyellow are those we haveexperimented with most usingthe diamond anvil cell.

Diffractedbeam

Sample

Front slit(10 micrometers)

Goniometer head

Penta prism

Optical multichannel analyzer Monochromator

Beam stop

Germanium–lithiumdetector

Shield

Synchrotron source Ion chamber

Figure 4. The configuration of our static high-pressure experiments using synchrotronradiation, the means of recording accurate diffraction patterns of materials in areasonable time when pressures exceed 40 gigapascals.

reverse of the normal, unpressuredprogression pictured in Figure 6a.When we increased the pressurebeyond 100 GPa, we observed that thesix-layered structure further transformsto a body-centered tetragonal structure.However, we did not see any bigvolume changes when the lanthanidestransformed from one structure toanother as the pressure increased. Thisbehavior is contrary to what otherexperimenters have conjectured. Thus,our data suggest that the volume ofthese metals changes rather smoothlyas a function of pressure without big,sudden changes.

The Actinides In theory and in experiments, the

actinides, especially the lighter onesearly in the series, are less consistent intheir behavior at high pressures than thelanthanides. The heavier actinides(americium through the end of theseries) are predicted to behave underincreasing pressures like trivalentlanthanides such as samarium andgadolinium. Our DAC experimentsgenerally agree with theory for theheavier actinides. Thus at roomtemperature and pressures to 20 GPa,the trivalent lanthanides and heavyactinides studied exhibit similarities.

On the other hand, the lightactinides, which include the weaponsmetals uranium and plutonium, arebelieved to behave less symmetricallyand predictably under intense pressurethan the lanthanides and heavyactinides. We are therefore studyingthem in the DAC in order to comparethe electron behavior deep within themwith similar behavior in the lanthanidesand other actinides so that we can makecritical conclusions about their high-pressure behavior.

Our findings concerning the otherearly actinides we have studied indepth (thorium and neptunium)illustrate the methodology andpotential uses of DAC experiments forthe study of uranium4 and plutonium.

At room pressure and temperature,thorium has a face-centered cubicstructure. In previous experimentalstudies to pressures below 100 GPa,we studied thorium with gold as apressure marker. Because of theinterference of the thorium and thegold diffraction lines, we did notidentify phase changes in theseexperiments. However, our detailedinvestigation of thorium to 300 GPawith platinum as an internal pressuremarker showed that indeed thoriumgoes through a structural change from face-centered cubic to a body-centered tetragonal at about 72.6 GPawith no further transformations evenup to the highest pressure. Our studies also suggested significanttransfer of electrons from outer shellsto those deep within the atoms as thepressure increased.

Because thorium has a stable body-centered tetragonal structure even at300 GPa and similar body-centeredtetragonal structures are stable, as wehave seen, for some lanthanides,LLNL researchers have asked whetherthe body-centered tetragonal phase isthe ultimate high-pressure stablestructure at room temperature forthese metals. Answers to suchquestions are essential if theoreticiansare to fine-tune their computer-generated models and codecalculations.

Recent studies on the next actinidemetal, neptunium, have, providedanswers to the question. As pressure is

increased, the orthorhombic crystalstructure of neptunium at ambientconditions transforms to a body-centered tetragonal structure and thento a body-centered cubic structure thatis stable to the highest pressure (seeFigure 7).5 This suggests that we mightsee similar body-centered tetragonal tobody-centered cubic structuraltransformation in the other actinidesand rare-earth metals. Thus, theultimate stable structure of the trivalentlanthanides, the heavy actinides, andsome light actinides may be body-centered cubic, not body-centeredtetragonal. We also hypothesized fromthese studies that neptunium shouldhave two body-centered cubicstructures, one at low pressures andhigh temperatures before melting andanother at high pressures and lowtemperatures. Both hypotheses providenew input to theory that can improvethe precision of computer weaponscode calculations.

Our classified DAC research on thelight actinides uranium and plutoniumhas provided vital information thatallows us to revise the computermodeling of the behavior of plutoniumduring nuclear explosions. In theabsence of testing, this data is vital inassuring weapon safety, reliability, andpredictability. To a lesser but equallyvital extent, our DAC work on thelanthanides and other actinides relatedto weapons materials has contributedto those refined codes. It allowsconfirmation or revision ofcalculations derived from theory anddynamic experiments with accuratedata that we can “see” from a high-pressure spectrum captured in staticDAC experiments.

25

Science & Technology Review March 1996

Diamond Anvil Cell24

Science & Technology Review March 1996

Diamond Anvil Cell

Figure 6. (a) The basic crystal structures found in lanthanide solids at room temperature and normal pressure across theseries beginning with the lightest elements and moving to the heaviest. Our diamond anvil cell experiments have confirmedtheoretical predictions that under increasing pressures approaching 100 gigapascals (GPa), the crystal lattice structure ofthe lanthanides follows the reverse of this sequence. In recent experiments, we saw a further transformation to a six-layeredstructure (b), which transforms to a body-centered tetragonal structure at pressures beyond 100 GPa but without the majorvolume changes predicted by some researchers.The crystal structures in (a) were first drawn by C. J. Alstetter, MetalsTransactions, 4, 2723 (1973).

Triple-hexagonal close-packed or a

six-layered structure

A

C

A

B

C

B

A

Face-centered cubic

A

C

B

A

Samarium-type close-packed

A

C

A

C

B

C

B

A

B

A

Hexagonal close-packed

A

B

A

Double-hexagonal close-packed

A

C

A

B

A

This close-packed phase sequence in the lanthanides can be observed by decreasing the atomic number Z at normal pressure or by compressing a single element at room temperature.

The observed crystal structure transformations of the lanthanides under increasing pressure across the series.

(a) (b)

not been thoroughly investigated atelevated pressures and temperatures.The DAC should allow researchers tocollect such critical data under staticconditions. We recently embarked onan exploratory study of equations ofstate and structural changes in highexplosives using the DAC andsynchrotron radiation. These studieswill also inform us whether crystalstructural changes in a high explosivesuch as triaminotrinitrobenzene(TATB) under pressure could causechanges in burn rates.

Key Words: actinides, diamond anvil cell,equations of state, lanthanides, science-based stockpile stewardship, shock-waveexperiments, x-ray diffraction.

References1. For more information on the

components of the diamond anvil cell,see “Diamonds: Powerful Tools forHigh-Pressure Physics,” Energy andTechnology Review, UCRL-52000-83-2(February 1983), pp. 11–19.

2. For further information on the use of x-ray diffraction to investigate crystalstructure, see “X-Ray DiffractionApplications,” Energy and TechnologyReview UCRL-52000-87-11/12(November/December 1987), pp. 23–28.

3. J. Akella, “Application of Diamond-Anvil-Cell Technique to the Study of f Electron Metals and Some MaterialsRelevant to Planetary Interiors,” inFrom Mantle to Meteorites, ed. K. Gopalan, et al. for the IndianAcademy of Sciences Press (1990), pp. 249–261.

4. Uranium has given us no help answeringquestions about the ultimate stable crystalstructure of the early actinides under highpressure. In our diamond-anvil-cellexperiments, it does not go through thephase changes at high pressures and roomtemperature that its neighboring elementsdo, and we do not yet know why.

5. Recent theoretical calculations of thestructural transformations in neptuniumhave shown a structural sequencesimilar to that we found experimentally.P. Söderlind, J. M. Wills, A. M. Boring,B. Johansson, and O. Eriksson,“Theoretical Study of theCrystallographic Structures inNeptunium,” Journal of Physics:Condensed Matter 6, 6573–6580(1994), and P. Söderlind, B. Johansson,and O. Eriksson, “TheoreticalZerotemperature Phase Diagram forNeptunium,” Physical Review (January11, 1995) (in press).

27

Science & Technology Review March 1996

Diamond Anvil Cell

Future Directions

The DAC has enabled us to obtainphase stability information thatdynamic techniques such as shock-wave methods could not supply and toincorporate that information into ourtheoretical models. Our scientistsconstantly endeavor to improve DACexperimental techniques in order toobtain better data and to obtain furtherinformation about the physicalproperties of any material, includingweapons-related materials. With theaddition of a laser or a resistanceheater or with cryogenic cooling, wecan also use the DAC to explore thepressure–volume–temperaturerelationship and the resulting structuralchanges of any material—its equationof state and phase diagram. Higherpressure and increased temperaturemay force further structural changes,until the material loses its crystalstructure entirely—that is, it melts.

An area of new technology forobtaining high pressure andtemperature data using x-raydiagnostics is electrical transport

experiments such as ohmic heatingbased on resistance to the current. Sofar, these experiments have beenamong the most difficult to performwith diamond anvil cells. Specialpreparation of the sample, anvils, andcell is required, and electricalconnections fail easily under the highstresses present in the diamond anvilcell. Consequently, electrical transportexperiments have been very difficult toperform beyond several tens ofgigapascals. Our scientists havedeveloped techniques to overcomethese problems and will embark onfurther studies of the weapons materialsto still higher pressures andtemperatures.

In any nuclear weapon, highexplosives play a pivotal role at thetime of detonation. These energeticmaterials generally have complexcrystal structures with low symmetryand are poor x-ray diffractors.Consequently, properties that arecrucial to performance—such as howthe behavior of high explosivesdepends on increased pressure and thuson changes in crystal structure—have

26

Science & Technology Review March 1996

Diamond Anvil Cell

For further information contact Jagan Akella (510) 422-7097(akella1@llnl.gov), Bruce Goodwin (510) 423-7736(goodwin2@llnl.gov), or Samuel T. Weir (510) 422-2462(weir3@llnl.gov).

JAGANNADHAM (JAGAN) AKELLA, who holds a Ph.D.in geochemistry, is the principal investigator of a LawrenceLivermore project that uses the diamond anvil cell (DAC) tostudy the f-electron metals and related weapons materials underultrahigh pressures and temperatures.

Akella has successfully used DAC technology inconjunction with synchrotron x-radiation to investigate undercontrolled laboratory conditions the physical and chemicalbehavior of highly toxic radioactive samples under ultrahigh

pressures and temperatures. These investigations have improved understanding ofthe fission triggers in nuclear weapons. Akella is also interested in developing newDAC techniques to investigate a wide array of nuclear and nonnuclear materials.Such work is relevant to a better scientific understanding of science-based stockpilestewardship issues as well as geological phenomena.

Akella joined the Laboratory in 1977 and has published more than 80 paperscovering geological sciences, condensed matter physics, and materials science. He isan elected fellow of the Mineralogical Society of America and the IndianMineralogical Association and is one of the recipients of the Department of Energy’s1994 Weapons Recognition of Excellence awards of the nuclear weapons program.

About the ScientistFigure 7. Schematic of the transformationof (a) orthorhombic neptunium at ambienttemperature and pressure to (b) body-centered tetragonal to (c) body-centeredcubic. The green circles in (a) representone type of neptunium atoms, located nearthe corners of an orthorhombic subcell; thegray circles represent another type locatednear the centers of two subcells (the arrowsindicate they are offset from the centers).Applying pressure moves the atoms to theirnearest corner and the center of eachsubcell and results in (c) a smaller, moresymmetric body-centered cubic form.

(a) Orthorhombic (b) Body-centered tetragonal (c) Body-centered cubic