WINTER • 1999 - 2000 A QUARTERLY RESEARCH & … · Research, writing, and design by Technically...

WINTER • 1999 - 2000

ALSO:BUILDING A BETTER MICROSYSTEM

SPRAY IT AGAIN, SAMVR Tool Helps Prepare for Terrorist Attacks

A QUARTERLY RESEARCH & DEVELOPMENT JOURNALVOLUME 1, NO. 4

Z MACHINEProviding Clues toAstronomical Mysteries

S A N D I A T E C H N O L O G Y

[From atop a 25-foot ladder, Sandian Larry Shipers examines a pair of camerasthat provide feedback to a computer system that controls Sandia’s automatedpainting system for the F-117 Nighthawk, also known as Stealth. The system spraysa thin radar-absorbent coating on the surfaces of the fighters. To test the system,this cardboard mock-up of some of the aircraft’s more hard-to-reach surfaces wasconstructed in a high bay.

ON THE COVER: Sandia National Laboratories’ Z Machine, the mostpowerful X-ray generator on Earth, is helping scientists better understandastronomical phenomena such as black holes, neutron stars, and the evolutionand eventual expiration of the universe.

(Photo by Randy Montoya)

Sandia Technology is a quarterly journal published bySandia National Laboratories. Sandia is a multiprogramengineering and science laboratory operated by SandiaCorporation, a Lockheed Martin company, for theDepartment of Energy. With main facilities in Albuquerque,New Mexico, and Livermore, California, Sandia has broad-based research and development responsibilities for nuclearweapons, arms control, energy, the environment, economiccompetitiveness, and other areas of importance to the needsof the nation. The Laboratories’ principal mission is tosupport national defense policy, by ensuring that the nuclearweapons stockpile meets the highest standards of safety,reliability, security, use control, and military performance.For more information on Sandia, see our Web site athttp://www.sandia.gov.

To request additional copies or to be addedto our mailing list, contact:Connie MyersMarketing Communications Dept.MS 0129Sandia National LaboratoriesP.O. Box 5800Albuquerque, NM 87185-0129Voice: (505) 844-4902Fax: (505) 844-1392e-mail: [email protected]

Sandia Technology Staff:Laboratory Directed Research & Development ProgramManager: Chuck Meyers, Sandia National LaboratoriesEditor: Chris Miller, Sandia National LaboratoriesResearch, writing, and design by Technically Write

“Creating a neutron star or black holeon Earth for scientists to understanddistant reactions is something of a

problem, since there would be no moreEarth. But because “neutron starsand black holes emanate X-rays

similar in effect to those emanated byZ, we realized we have a chance to

test astrophysical theoretical modelsthat have never been tested

experimentally,”

Mark Foord, Physicist

[

1


Dear Readers: This issue of Sandia Technologycontains articles about a variety of recentSandia developments in diverse areasof the weapons program. The Labs’ Zmachine — the most powerful X-raygenerator on Earth — continues to makenews, this time by helping physicistsexamine what happens to iron in thegrip of black holes and neutron stars.The work could even give scientists amuch better understanding of theformation and expiration of the universe.Several stories focus on Sandia’sgrowing field of work in microsystems.A major development is the creation ofa new, advanced, five-level polysiliconsurface micromachining process thatwill permit microelectromechanicalsystems (MEMS) to become morereliable and sophisticated. MEMS,complex machines with micro-sizefeatures, can be found in a variety ofproducts, including optical devices,computer game joy sticks, car airbagsensors, ink-jet printers, and projectiondisplays. Another article describes the creationof a synthetic product whose need wouldescape the understanding of most sanepeople — until you learn it could savelives as well as millions of taxpayerdollars. Synthetic sludge is the product,and it successfully mimics the deadlysludge found in underground nuclearwaste storage tanks. Working with thesafe sludge will give workers cleaningup nuclear waste sites a betterunderstanding of how best to deal withthe product’s deadly cousin. I hope you will take a look at the“Insights” column at the back of thisissue. The column is a reprint of amessage written by Sandia Presidentand Laboratories Director C. PaulRobinson from the Sandia Lab News –Sandia’s in-house newspaper . Thecolumn was published on Sandia’s 50thanniversary, an occasion observed byemployees as well as by many VIPswho joined us at the Labs for a day ofreflection and celebration.

Chris MillerEditor

FROM THEE d i t o r

TABLE OFC o n t e n t s

2

4

7

8

10

13

15

Z Provides Clues toAstronomical Mysteries

VR Tool Helps Preparefor Terrorist Attacks

A Dirty TaskGets Cleaner

Building a BetterMicrosystem

Remote SensorsAnalyze Gases From Afar

ControllingQuantum Dots

Fine Resolution,Real Time Images

by C. Paul RobinsonPresident and Laboratories DirectorSandia National Laboratories

I N S I G H T S

2


Black Holes, Neutron Stars, Evolution of the Universe:

And could

Sandia, charged

with U.S. nuclear

stockpile steward-

ship, confidently

certify the safety

and reliability of

every component

in every weapon

every year?

Experiments onSandia NationalLabora to r ie s ’Z machine — themost powerful X-raygenerator on Earth —usually focus on thedefense of the UnitedStates and harnessingnuclear fusion forelectr ical power. But data from recenttests undertaken thereby researchers fromthe Department ofEnergy’s LawrenceLivermore and Sandianational laboratoriesshould help astron-omers t ry ing tointerpret images fromthe billion-dollar Chandra X-ray obser-vatory now orbiting Earth. (Alsobenefiting will be two billion-dollarX-ray orbiting observatories expectedto be sent aloft from Europe and Japanin the next year.)

The results will further humanunderstanding of black holes, neutronstars, and the evolution and eventualexpiration of the universe, predictsLivermore physicist Mark Foord, oneof the leaders of the joint Sandia-Livermore project. The methods devel-oped in the work also can be used inweapons physics, says Sandia projectcollaborator and physicist Jim Bailey.

Apprentice cooks inempty kitchens

Iron is of interest to astronomersbecause it is among the most com-plicated of elements widespread in theuniverse, and, therefore, among the

hardest to understand.Several explanations arepossible for the effectsit creates on imagestaken by the recentlylaunched Chandraorbiting telescope. The problem forastronomers, who nowhold bleacher seats towatch titanic energiesthat are transformingelements on a scalenever before seen, issomewhat similar to thatfaced by apprenticecooks in empty kitchenswho watch master chefscook on TV: Withoutbeing able to cookalong, many fine points

of the process can only be guessed at. Creating a neutron star or black holeon Earth for scientists to understanddistant reactions is something of aproblem since there would be no moreEarth. But because “neutron stars andblack holes emanate X-rays similar ineffect to those emanated by Z, werealized we have a chance to test astro-physical theoretical models that havenever been tested experimentally,” saysFoord. Says Bailey, “We’re looking withspectroscopic eyes at the atomicphysics of ionized iron so that thesecan be compared with theoreticalcalculations. Astrophysicists will haveto consider what implications ourfigures have for their models.”

3


[ Creating a neutron staror black hole on Earth

for scientists tounderstand distant

reactions is something ofa problem since there

would be no more Earth.[

In front of a containment can at the center of Sandia’s Z machine, a small window-

like frame holds a shining foil containing a layer of mixed iron and sodium fluoride

only 500 angstroms thick, sandwiched between micron-thick plastic substrates.

Continued on page 6


4

— whenrescue personnel may need to treatmass casualties following the releaseof a nerve agent in a shopping mall,theme park, or subway — there willbe no second chances. Rescuers whobecome victims of a terrorist attackcan’t save lives. Soon emergency medical technicians(EMTs) and firefighters may be able

to practice responding to such attacksusing a virtual reality (VR) trainingtool under development at SandiaNational Laboratories. Sandia computer scientists havecombined seven years of virtual realityresearch into BioSimMER, a VRapplication that immerses first respond-ers in a 3-D, computer-simulatedemergency setting — a small airport

in which a biological warfare agenthas been dispersed following a terroristbombing. Simulated casualties with avariety of symptoms are foundthroughout the airport. BioSimMERcan help emergency personnel makebetter decisions if they are ever calledto respond in a real chem-bio attack,says project leader Sharon Stansfield. “With virtual reality, you can practice

“With virtual reality, you can practice over and over again, like in a videogame,” Stansfield said. “You make mistakes; you learn. If someone

dies, you can hit the reset button.”[

VR Tool Helps Prepare for Terrorist Attacks

SPRAY ITAGAIN, SAM

[Lydia Tapia of Sandia National Laboratories demonstrates BioSimMER, a virtual reality application that allows rescue personnelto practice responding to a terrorist attack involving release of a biological agent.

5


over and over again, like in a videogame,” she said. “You make mistakes;you learn. If someone dies, you can hitthe reset button.”

Saving ‘cybercasualties’

The computer simulation engagesthe rescuer’s eyes, ears, and decision-making abilities through goggles thatdisplay the scene’s images. The rescuerwears sensors on the arms, legs, andwaist, allowing the player’s motionsto be fed back into the simulation. The researchers worked closely withDr. Annette Sobel, a Sandia physicianand researcher, to create “cyber casu-alties” with realistic symptoms andreal-time changes in their conditions.One virtual casualty has a visible chestwound. Another has symptoms thatindicate head trauma. Another suffersfrom inhalation of staphylococcalenterotoxin B (SEB), the airborne bioagent used in the simulation. And afourth appears to exhibit the symptomsof SEB exposure, but closer examina-tion shows him to be suffering frompsychological shock. During a simulation, the player musttriage, diagnose, and attend to themedical needs of each casualty. Visualindicators, such as a victim’s move-ments, labored breathing, skin color,and vital signs, give clues to eachvictim’s condition. If the rescuer iswrong or not fast enough, the victimor patient could die. After making a diagnosis, a playercan administer medical treatment byreaching for and using tools in a virtualmedical kit. Players may need to attendto initial decontamination procedures,place masks over a patient’s nose andmouth, or place sensors or other mon-itoring equipment near the patient.Most important, they need to learn toprotect themselves. “A player who dies a quick cyberdeath will not likely forget theimportance of personal protective

equipment in the future,” Stansfieldsaid. Although textbook-style preparationand live-training exercises are valuablepursuits as the nation’s emergencypersonnel prepare for future emer-gencies, BioSimMER offers someadvantages. “We tend to understand what we seewith our eyes and do with our hands.The strength of VR is being able totrain on things you can’t do otherwise,particularly in highly contaminated orhighly stressful situations,” Stansfieldsaid. “We don’t think BioSimMERshould replace live exercises, but it canprovide an inexpensive way foremergency personnel to practice.”

Achieving a suspensionof disbelief

Although BioSimMER’s graphicsaren’t as refined as the latest 3-D videogames, the VR application does offersome realism that video games can’t. “In video games, the world is imag-inary,” Stansfield said. “But a VR worldis a representation of a real place withrepresentations of real people movingin real-time. Everything in that worldmust move and respond as if it werereal and bound by the laws of physics.” Creating a virtual world with suchphysical realism does require sometrade-offs, she said, but the ultimategoal is to create a suspension ofdisbelief. “You’ve got to make the

player believe, at least temporarily,that he or she is in the situation youare presenting,” she said. The airport used in BioSimMER isa fictitious one-story, three-gate airportbased loosely on a real airport in centralNew York. The research team modeledhow the airborne biological agentwould spread through the airportfollowing an explosion that dispersedthe agent. More than 20 first responders gottheir first chance in July to test driveBioSimMER at the National Emergen-cy Response and Rescue TrainingCenter at Texas A&M University . “Those people out there — thepolice, the firefighters, the EMTs —they do important jobs. They are tellingus they want this. If we can do some-thing to help them, then we are usingtechnology to make a real contributionto society,” Stansfield said. Development of BioSimMER wasfunded by the Defense AdvancedResearch Projects Agency. Bio-SimMER also builds on previousSandia VR work, including appli-cations for training battlefield medicsand for law enforcement small-teamstactical training. Although BioSimMERis a research prototype, the researchershope to continue development andrefinement of the system and scenarios,with the goal of making a version ofBioSimMER available to users in thenot-too-distant future. Currently, work is focusing onmaking the user’s interaction with theBioSimMER virtual world easier andmore realistic with funding from DOE’sOffice of Science and Technology PilotProjects in Biomedical EngineeringProgram.

Technical contact:Sharon Stansfield,(505) [email protected]

“We tend to understandwhat we see with our eyesand do with our hands.The strength of VR isbeing able to train onthings you can’t do

otherwise, particularly inhighly contaminated or

highly stressfulsituations.”

[ [

“We have a collaboration with fouror five groups around the world whosemain job is to analyze data fromChandra using their own codes,” saysFoord. “They have made some predic-tions that we’re going to compare toour data.” If necessary, the codes willbe modified. The results will be compared withastrophysical calculations embeddedin computer codes of how neutron starsaffect the widespread element.

Interpreting data fromthe stars

The experiments proceed by placingsquare centimeters of iron, a few hun-dred angstroms thick, close to the Z-pinch at the heart of Z. (A Z-pinchachieves its output by generating apowerful magnetic field that collidesions at an appreciable fraction of thespeed of light.)

This exposes the metal to tempera-tures of more than one million degreesfor a few billionths of a second, ioniz-ing the metal. “Iron has thousands of spectral lines,”says Foord. “We know their positions(on a spectroscopic scale), but theydiffer in intensity depending on whichelectrons, or how many, have beenstripped away in the ionizing heat ofa neutron star. Calculating these rela-tive intensities is uncertain. This makesit difficult to predict how highly ionizediron is, or should be, in other star

systems. If we can measure in the labwhat the actual figures are, we learnhow to interpret our data from thestars.” A Sandia instrument that takesseven images temporally and aLivermore instrument that takes oneimage in time and two images spatiallywill help determine effects of theintense X-ray pulse from Z on ironsamples in terms of spectrum, temp-erature, density, and ionization. “Our first two shots were in lateOctober and we are now analyzing thedata,” says Foord. “It appears that wewere successful at producing highlyionized iron and were able to obtainan accurate measurement of the radi-ation produced from the Z-pinch. Wehope to be back in a few months to dofollow up experiments and then returnagain in the summer to exploredifferent regimes.” Other collaborators in the experi-ments are Livermore’s Bob Heeter,Bob Thoe, and Jim Emig. Heeter, arecent Princeton graduate and Liver-more postdoctoral fellow, is the leadexperimentalist for the project. Sandia’s Tom Nash, GordonChandler, Dan Nielsen, Dan Jobe, andPat Ryan also contributed to theexperiments.

Technical contacts:Jim Bailey,(505) [email protected]

Mark Foord,(925) [email protected]

Bob Heeter,(925) [email protected]


A machine that creates temperatures rivaling those of the sun is helping physicsts examineup-close what happens to iron in the grip of black holes and neutron stars.

“It appears that we weresuccessful at producing highlyionized iron and were able to

obtain an accurate measurementof the radiation produced fromthe Z-pinch. We hope to be backin a few months to do follow-up

experiments and then returnagain in the summer to explore

different regimes.”

[ [

6

Continued from page 3

7


Synthetic goods are generally modeledon scarce but desirable materials —diamonds, fine wools, even fruit juices. Jim Krumhansl is offering the worldsomething a bit different: synthetic sludge.The unusual product, which harmlesslymimics the deadly sludge found in under-ground nuclear waste storage tanks, couldsave taxpayers hundreds of millions ofdollars in cleanup costs around the UnitedStates. It will provide researchers with asafer and less costly way to determine whichradioactive wastes will remain embeddedin, and which will migrate from, the sludgefound in waste storage tanks. This should permit easier and more cost-effective decommissioning of some tanksat a cost of $10 million each, rather thanworst-case disposal, with maximumsafeguards for every tank of $65 millioneach. There are approximately 180 suchtanks at Washington state’s HanfordReservation alone. Far fewer are at theSavannah River Site in South Carolina. The work is an outgrowth of the February1995 Galvin Report’s suggestion thatnational laboratories find new science tocut down on the very large costs of projectedenvironmental clean-up bills. The artificial sludges consist of “non-radioactive representations of screaminglyradioactive materials” that precipitate as amatter of course during storage in giantunderground tanks that may contain up toa million gallons of nuclear waste, saidKrumhansl, a Sandia researcher. The sludges created by Krumhansl’sresearch group are part of a joint projectamong the Department of Energy’s SandiaNational Laboratories, Pacific NorthwestNational Laboratory (PNNL), and theUniversity of Colorado.

Sludge that won’t budge

Naturally occurring sludge sticks to thewalls of tanks and could serve as a long-term source of radioactive contaminationin the environment. “When the tanks are emptied, some ofthe sludge won’t budge,” said Krumhansl.“The tanks will be sluiced, sloshed, and

squirted, but people won’t be sent inside toclean them up.” While purists might prefer that scientistsexperiment on real sludge, workers feeldifferently. Many of these materials are soradioactive that working with them is bothcostly and requires taking extreme measuresto protect the health of researchers per-forming the work. Synthetic sludges, while chemicallysimilar to radioactive ones, are not radio-active and can be handled without dangerby lab workers attempting to quantify howmuch radioactive material sludges will storeor release, and for how long. Costs of decommissioning tanks, treatingevery material as a worst case in its potentialto escape into the environment, couldamount to hundreds of millions of dollars.“The decision how to treat these tanksultimately depends on how much hazardthere is from their residual radioactivitybeing able to move about. If virtually noneof it goes any place, then you’re a lot freerto do simple decommissioning techniques,”Krumhansl said. “The question is what fraction of radio-nuclides in the sludge will stay there inde-finitely and what fraction could becomemobile and enter the groundwater.” The worst-case alternative is removingto a new location potentially corrodedradioactive tanks that are 70 feet across and

55 feet high, with tops seven feet below theground. The simpler alternative is to leavethe tanks in place and seal them off, saidKrumhansl.

Wanted: the sludgehammer

“Imagine being the one to dig up sevenfeet of radioactive earth just to reach thetop of the tank,” said Krumhansl, describingthe worst-case ordeal. “Where do you putthe dirt? How do you cut the tank into smallenough pieces to move? Working in a lead-lined (hazardous materials) suit is likeworking in a sauna. It’s awkward, uncom-fortable, you need a respirator, you need acutting torch, you have to make sure thepieces you’re cutting don’t fall down andhit you. The job may take robots becausethe radiation levels are so high. It’s a difficultand expensive task to do.” To judge whether the synthetic sludgesare realistic representations of their morelethal counterparts, group member Jun Liuof PNNL has done transmission electronmicroscope (TEM) work comparing theatomic structure of a small number of actualradioactive sludges with the group’s syn-thetic version. (The TEM uses such a smallsample that the radioactivity constitutes nodanger.) In some cases, the synthetic sludgescontain nonradioactive elements from theperiodic table that “weren’t exact matchesfor radioisotopes, but provide useful insightsinto the way these things work,” saidKrumhansl. Examples of this includesubstituting rhenium for technetium andneodymium for americium. Other elements,like strontium, cesium, and selenium havenonradioactive isotopes that behave iden-tically to the radioactive isotopes in theactual waste. Finally, the artificial sludgesalso contain nonradioactive components,such as lead and cadmium present in theactual wastes, that are of concern becauseof their heavy-metal toxicity.

Technical contact:Jim Krumhansl,(505) [email protected]

A Dirty Task Gets CleanerSynthetic Sludge Permits Safer, Less Expensive Cleanup

Sandia researcher Jim Krumhansl believesthe use of synthetic sludge could reduce thecost of decommissioning some radioactivewaste storage tanks at several DOE facilities.

A new, advanced, five-levelpolysilicon surface micromachiningprocess pioneered at Sandia NationalLaboratories promises that microelec-tromechanical systems (MEMS) of thefuture will be more reliable and capableof doing increasingly complex tasks. “This five-level polysilicon surfacemicromachining technology has thepotential of becoming the industrystandard, replacing the more commonly

used two- or three-level polysiliconsurface micromachining approaches,”said Sandia engineer Steve Rodgers,who with colleague Jeff Sniegowskihas spent the past several years proto-typing designs and developing theinnovative process. “We have beenworking hard to baseline the techno-logy as a reproducible manufacturingprocess, and we’re getting there.”The new technology was developed at

Sandia’s Microelectronics Develop-ment Laboratory. MEMS devices madefrom the five-level process eventuallywill be manufactured in the new Micro-systems and Engineering SciencesApplication (MESA) facility beingplanned for construction at Sandia. MEMS, complex machines with

micron-size features, can be found ina variety of products, including opticaldevices, computer-game joy sticks,car-airbag sensors, ink-jet printers, pro-jection displays, and more. They are sosmall that they are almost imperceptibleto the human eye and have movingparts no bigger than a red blood cell. Sandia will begin offering the five-level technology next spring to externalcustomers for prototyping purposes.Information about all MEMS coursesis available on Sandia’s micromachineWeb page located at http://www.mdl.sandia.gov/micromachine. Almost all of today’s surface micro-machine components are designed forand fabricated in technologies thatincorporate three or fewer levels ofstructural materials. The levels aretypically deposited as thin films ofpolysilicon that are about one to twomicrons thick. These films are separ-ated by air gaps that initially are defined


8

This image shows a ratcheting system fabricated in the five-level technology. Twenty ofthese gears fit on a period in a newspaper sentence.

BUILDING A BETTERMICROSYSTEM

Five-Level Process Could Become Industry Standard

They are so small thatthey are almost

imperceptible to thehuman eye and have

moving parts no biggerthan a red blood cell.

[ [

by layers of sacrificial silicon dioxide(sacrificial because they will eventuallybe eliminated) of about the samethickness. Processes with thickerpolysilicon film — up to tens ofmicrons — exist, but are typicallylimited to only that layer. “In general, the more layers ofstructural material that a designer hasto work with, the more complicatedthe device that can be fabricated,” saidSniegowski, who developed the fab-rication technology for the five-levellayering method. “Therefore, surfacemicromachined components havegreater functionality than bulk micro-machined parts.” That’s why, he added, the five-levelpolysilicon surface micromachineprocess that incorporates four layersof structural films, plus an electricalinterconnect layer, is so attractive. “This technology permits mechanicalfunctionality that only a five- or morelayer process could offer,” Sniegowskisaid. “It provides a base for designingtruly sophisticated multilevel micro-electromechanical systems, whilesimultaneously offering much of theyield and robustness that typically isassociated with single-level micro-machining technology.” A two-level polysilicon process hasonly one layer of structural material,

with the other level defining the groundplane. Such a technology is useful forfabricating simple sensors and

actuators. With three levels, it is pos-sible to create gears with hubs, whilethe four-level technology provides anadditional layer of material that canbe used to define linkage arms thatmove above the plane of the gears,enabling a continuous 360-degreerotation. The five-level technologyexpands on this to permit complex

interacting mechanisms to be fab-ricated on moving platforms. But to do this, several challengeshad to be overcome, which Sniegowskiand Rodgers have met. For example,as additional layers are added, moretexture appears on the surface. Thisoccurs because the top layer acquiresthe characteristics of all the lowerlayers, including high and low spots.The result is the creation of protrusions,called mechanical parasitics, thatextend from the upper mechanical layerto lower areas. They can interfere withoperation. These parasitics, Sniegowski said,can significantly “constrain thedesign.” If the design doesn’t takethem into account, they could easily“collide with the teeth and preventrotation of the gear.” To eliminate this problem, a Sandiateam led by process engineer DaleHetherington modified and patented aprocess commonly used in manu-facturing integrated circuits —chemical mechanical polishing — toplanarize (flatten) the surface. Highspots are eliminated after a very thicklayer of sacrificial oxide covers allprevious layers. Through chemicalmechanical polishing, the high spotsare eroded, producing a uniform flatsurface. Rodgers said the new five-layerprocess provides a foundation forfabricating components that offer highperformance, reliability, and robust-ness. However, work continues toimprove the process even further. “Weare now adding the final touches beforeoffering it for widespread use,” he said.

Technical contacts:Steve Rodgers,(505) [email protected]

Jeff Sniegowski,(505) [email protected]

9


“This technology permitsmechanical functionalitythat only a five- or more

layer process couldoffer,” Sniegowski said.

“It provides a basefor designing truly

sophisticated multilevelmicroelectromechanical

systems, whilesimultaneously offeringmuch of the yield and

robustness that istypically associated with

single-level micro-machining technology.”

[[Sandia’s Steve Rodgers (left) and Jeff Sniegowski look over a computer schematicof a five-level polysilicon surface micromachine.

A new remote sensor the sizeof a dime being developed by theDepartment of Energy’s SandiaNational Laboratories should allowusers to rapidly detect dangerous gasesfrom up to two miles away. Called Polychromator, the devicewill use a combination of optics andmicroelectromechanical systems(MEMS) to determine gas types. It isa joint effort of Honeywell, Massachu-setts Institute of Technology (MIT),and Sandia. MIT is designing theMEMS structures that are beingfabricated by Honeywell. “Imagine a soldier on a battlefieldlooking through a pair of binocularswith the Polychromator chip built into

it and detecting from afar the natureof gas being emitted in a smoke cloud,”says Sandia researcher Mike Butler,one of Polychromator’s inventors.“There’s no need to obtain a sampleof the gas or even get close to it.Instead, the detection is made from asafe one or two miles away.” He adds, “It promises that in a matterof seconds you can tell the nature of agas and whether it’s toxic withouthaving an actual sample. It has allkinds of possibilities.” Butler, together with Mike Sinclair,also a Sandia researcher; formerSandian Tony Ricco; and MIT pro-fessor Steve Senturia have obtained

two patents on the device, the mostrecent one in May. The researchersexpect to have a functioning prototypeby 2001, the end of a four-and-a-half-year funding period for the project bythe Defense Advanced ResearchProjects Agency. The Polychromator is a variation ofa conventional gas-analysis technique,correlation spectroscopy. Infraredradiation is passed through a samplegas that imparts a spectral pattern tothe radiation. This pattern then iscorrelated with the spectrum of areference gas to provide the identi-fication. Each gas has its own distinctspectral pattern, which can be used toidentify the gas. Butler says the standard method islimited because it requires a referencecell of the gas and the testing equip-ment is bulky. The Polychromator chip, however,is small — about the size of a dime —and does not need a reference cell. Itconsists of thousands of microscopicMEMS grating elements on a siliconwafer manufactured with standardmicrolithographic “mask and etch”techniques. Each element is 10 micronswide and one-centimeter long, looks

like a long flat beam, and is designedto move up and down. Light from thegas cloud shines on the chip after beingcollected by an optical device, such asa telescope or binoculars. As a soldier or other user peersthrough the binoculars, with lightstriking the Polychromator chip, amicroprocessor inside would direct ina matter of seconds the MEMS gratingthrough a series of movements . Each

10


This computer image simulates a soldier on a battlefield looking remotely at a gascloud through a pair of binoculars containing the Polychromator chip. It promises that in a

matter of seconds youcan tell the nature of a

gas and whether it’stoxic without having an

actual sample.[

Functioning PrototypeExpected by 2001

TINY SENSOR TOANALYZE GASES

FROM AFAR

[

movement would result in a differentgrating designed to detect a specificgas. When the grating matches andidentifies the gas, the device wouldalert the soldier. One of the developments makingthe Polychromator possible was acomputer program created by Sinclairto design the gratings. He came upwith the program in 1994, shortly afterhe and the other future patent holdersbegan talking about the possibility ofcombining a hologram with a spectro-meter to remotely analyze gas clouds.Senturia, the MIT professor, was aSandia consultant at the time. “The infrared spectra of variousgases have been known for years,”Sinclair says. “My computer programuses these known spectra to calculatethe grating profiles for each of the gases.” Another computer program thenuses these grating profiles to controlthe movement of the MEMS grating

elements. So far, Sinclair has generatedgrating profiles for about 10 gases, butmore profiles are on the way. While the team is still perfecting themovable MEMS grating, they alreadyhave completed silicon wafers contain-ing fixed gratings. Sandia’s CompoundSemiconductor Research Laboratoryfabricated the fixed gratings that candetect 10 different gases, mostly thoseused in chemical warfare. Unlikethe Polychromator, these gratingsdon’t move. Sinclair notes that remote chemicalsensors are nothing new — some haveeven been developed at Sandia in thepast. This sensor, however, promisesto have several unique characteristics. “Unlike conventional spectrometers,the Polychromator will do a lot of thespectral processing up front, makingit work faster that other types,” Sinclairsays. “In addition, it will be verysensitive and selective — able to

discern quickly chemicals that havesimilar infrared spectra.” The researchers are already on theroad to a functioning Polychromatorprototype. They have an optics pack-age in-hand and are waiting on theproduction of a refined prototype chipfrom Honeywell and electronics froman outside contractor. Both are expect-ed in the next few months. “All we’ll have to do then is integrateeverything, test, and see how well itworks,” Sinclair says.

Technical contacts:Mike Butler,(505) [email protected]

Mike Sinclair,(505) [email protected]

11


Sandia’s Mike Sinclair (left) and Mike Butler prepare to put the dime-size Polychromator remote sensor chip intoa testing unit.

12


The Department of Energy has givenSandia National Laboratories the go-ahead to develop a conceptual designfor a $300 million Microsystems andEngineering Sciences Application(MESA) facility. The purpose of theproject is to join Sandia’s expertise inweapons design, very fast computing,and microsystems into an immersiveenvironment in which scientists canimagine, design, and create the 21stcentury’s best non-nuclear componentsfor nuclear weapons. Other benefits include offeringAmerican businesses and universitiesnew opportunities to advance throughuse of the cutting-edge science that thefacility and its programs will providein microsystems and computer-aideddesign and simulations. One military advantage of the newmicrosystems will be that they areexpected to reduce the likelihood ofmalfunction in the event of an unex-pected event, like a crash involvingnuclear weapons. “A big arming or safety deviceinvolved in an accident severe enoughto destroy or fire off a missile wouldmost likely be damaged as much asthe explosive device,” said DavidPlummer, nuclear weapons represen-tative for MESA. Microsystems designed by high-speed computers are smaller thanpostage stamps. Their componentshave little mass, so they have little

inertia — the force thatkeeps an object movingonce it is put in motion.Inertia throws pas-sengers forward, for

example, when a carcomes to a sudden stop.

The amount of inertia anobject possesses is a keyfactor in how much

damage it may suffer.Devices of extremely low

mass, having almost noinertia, have no problemstopping very suddenly andthus are difficult to destroy

impact. “As the magnitudes of potentialaccidents increase — as jet fuel getshotter, as air speeds get faster — so dothe possibilities of accidents exceedingour current controls. We need morerugged safety devices,” said Plummer.“MESA will produce them.” The primary function of the facilitywill be the defense of the United States,said Tom Hunter, senior vice-presidentof nuclear weapons. The facility alsowill include weapon component mod-eling and simulation, weapon certifi-cation, and embedded microsystems. However, MESA also representsthe leading edge of industry’s interestin intelligent microsystems, said DanHartley, vice president for laboratorydevelopment. “The demand for lessexpensive, more intelligent, smallersystems is ubiquitous in industry, andwe believe that what we’re doing hererepresents the birth of a new field ofbusiness endeavor.” Al Romig, Sandia vice president forscience, technology, and components,said, “This will create a capability todo our business in a new way. It’llrevolutionize the way we’ll refurbishthe nuclear stockpile and build newspace satellites. It also will be anopportunity to drive the creation ofentirely new industries in the UnitedStates, using our Science andTechnology Park.” “Conceptually, MESA is a new wayof designing intelligent, autonomous,reliable microsystems from the outsetand will be the wave of the future forall complex commercial products thatwill incorporate microsystems, such

as cars, planes, and computers,” saidDavid Goldheim, director for Sandia’scorporate business development andpartnerships. “At an appropriate time,we plan to ask businesses to becomeinvolved in the design of portions ofthe program. Industry will help designand use parts of the new facility.” MESA is also of interest to univer-sities “because in our meetings withdeans, microsystems are a high prioritywith them as well, and they want tohave more to do with us because wewill be defining the leading edge,”added Hartley. Regarding students, Don Cook,director of the MESA project said,“Students in school are being trainedin new technologies. We can’t expectto train the brightest graduates on oldtechnology. We want them to advancenew technology even further. That’swhy we’re a national lab.” Approximately $95 million is slatedto be funneled into new equipment forMESA, said Cook. Said John Stichman, director ofweapons systems engineering in NewMexico for Sandia, “As we refurbishweapons, we need to know that thoseweapons we return to the stockpilemeet our safety and capabilityrequirements. These requirementspertain to the next several decades oflife these weapons would have in thestockpile. Microsystems hold thepromise to provide the most capableimplementation of the safety of refur-bished weapons. The nuclear weaponsystems organization is working closelywith the MESA planning staff to assurethat MESA is well-aligned with wea-pons systems design and development.” The tentative schedule is to beginengineering design in fiscal year 2000and construction in 2001. The projectis scheduled for completion by 2003,and to be fully operational by 2004.MESA eventually will house 600employees.

Technical contact:Don Cook(505) [email protected]

Planned MESAFacility GetsDOE Nod


The term quantum dots has a nice ringto it, like or a special version ofconnect-the-dots games that absorb theattention of children. But while the tinyentities may sound amiable, they findeach other repulsive, and in that just-discovered knowledge may lie thesecret of controlling their formation. Effective control would turnassemblages of dots containing a fewthousand atoms (by comparison apencil dot contains millions of atoms)into the world’s most effective solid-state lasers, said principal investigatorJerry Floro. He leads a research teamusing probes developed at SandiaNational Laboratories. “We developed novel probes thatuncovered a repulsion effect betweenquantum dots. This effect may com-pletely govern the way they organizethemselves. Understanding this self-organization is critical if we are tocontrol dot characteristics for lasingdevices,” Floro said. The probes, one of which wasrecently patented, have made uniquereal-time measurements of atomsclustering to form relatively large three-dimensional dots, called islands. Scien-tists for the first time observed the roleof mutual repulsion in causing dots tochange shape and self-organize as theygrow. Understanding these factors is animportant step in basic science. Thesmaller the dot, the shorter the emissionwavelength; the more tightly the dotsare packed, the more intense the beam;and the more uniform their size, themore uniform the frequencies. The ongoing work was done incollaboration with researchers now atBrown University and the Universityof Illinois at Urbana-Champaign.

Buckling upto relieve stress

Quantum dots form when very thinsemiconductor films buckle due to thestress of having lattice structuresslightly different in size from those ofthe material upon which the films aregrown. Just a few percent differencein lattice size creates stresses (orpressures) in a film that are 10 timeslarger than those present in the deepestoceans of Earth. These huge pressures,as new layers are deposited, force theinitially flat film to separate into dotsand then pop up into the third dimen-sion to relieve stress, rather than con-tinue to grow against resistance in twodimensions. This extra dimension,combined with the extremely smallsize of the dots, gives them differentproperties from when the material wasin its original flat-film shape. Semiconductor quantum dots havethe potential to produce laser-lightoutput at wavelengths where, in amanner of speaking, no flat film has

gone before, depending on the size ofthe dots. But while crude collectionsof quantum dots have been grown andset lasing, knowledge of the conditionsneeded to influence the dots so thatthey form more regularly in size, shape,and pattern — thus improving controlof their lasing frequency and intensity— has been only a dream. Techniques of physical etching —even nanolithography — can’t be usedto make them because the dots are sosmall that they do not manifest thecontinuous nature of a solid. “Peoplehave made crude devices,” said Floro.“But to make them reliably, we need to understand the ground rules.” The collaboration between Sandiaand Brown University makes use ofoptical and stress measurements toobserve dot formation as it happenson silicon germanium. Stress in the filmcauses the substrate to bend, which theresearchers measure by bouncing laserbeams off the sample. When the dotsform and change shape, the stresschanges and so does the amount of bendin the substrate. So, mapping the sub-

13

Doting Dots These Are NotControlling Quantum Dots Key to

Producing Effective Solid-state Lasers

continued on page 14

Sandia’s Jerry Floro (left) and John Hunter examine an atomic force micrograph image ofsemiconductor quantum dots.

strate as it bends reveals when dots firstform and how their shapes evolve. Said Floro, “Tiny dots cause detec-table bending in a substrate that is10,000 times thicker than the dotsthemselves.” The conventional laboratory ap-proach, by contrast, has been to art-ificially stutter dot formation intoseparate time intervals and bring inter-mediate results to powerful microscopesto observe formations at each stage. As more film layers are deposited,the dots grow closer and closer, and,because they repel each other, they areforced to become more uniform in size,line up in orderly fashion, and changetheir shape. Some dots are even “eaten”by their neighbors in an attempt toreduce overcrowding, a process knownas coarsening.

Working withlarger entities

But how does one “see” dots so small?The researchers were clever and usedbigger dots, called islands, thousandsrather than hundreds of angstroms insize and made of silicon germanium,since the larger ones could be moreeasily examined. The researchers earlierhad shown that groups of larger semi-conductor entities interact the same wayas smaller semiconductor entities. According to Dennis Deppe, anelectrical engineering professor at theUniversity of Texas at Austin workingin the field but unconnected with theSandia-Brown effort: “Many of thesame basic growth phenomena are seenin different material systems. So it ispossible to learn some important phys-

ical principles concerning nucleationand dot formation in one semiconductorsystem (and have some of them) carryover to another.” “We directly measure the kineticsof nucleation, coarsening, self-organization, and phase transformationswithin growing island arrays,” saidFloro. “All these processes areexplained within a unified model thatworks with ensembles of islands ratherthan individual islands in isolation.” In terms of actual formation, theprocess characteristically went like this:10 atomic layers of film would formsmoothly. As more layers were depos-ited, the film broke up into tiny pyramid-shaped islands. With more layers, thepyramids self-organized and coarsened,and then became dome-shaped islands.

Quantum dots asdiffraction gratings

But there’s more. The researchers,not content with one novel tool toexamine dots, realized they had another. Floro—along with Bob Hwang (aCalifornia-based Sandia researcher),Ray Twesten at the University of Illinoisat Urbana-Champaign, Eric Chason (aformer Sandia researcher), and BenFreund of Brown University—mademeasurements that treat dots as theoriginators of light-interference patterns.

Since light’s direction and intensityvaries depending on the size, shape,and spacing of the islands, the resultsoffer information in real-time to deter-mine what is happening to the tinyislands as temperatures, material com-positions, and stresses change. “We realized that if we could produceislands more than 1,000 angstromsacross, the spacing between islands waslike that of a diffraction grating,” saidFloro. “Combined with our real-timestress observations, this allowed us tomeasure stress, shape, and size simul-taneously instead of having to stop theprocess, take the dots out, and measurethem. A key ingredient was our abilityto show that the basic physics of thelarge islands mimics that of the muchsmaller dots.” Observing the process of dots goingfrom one shape to another to relievestress has provided insight into thephysics governing island formation.“It showed us what controls dotevolution and how process conditionslike temperature and strain enhance orsuppress dots.” Silicon germanium is not a good laseremitter, but it is simple enough to derivethe applicable physics. “We need to findout next how much of the physicslearned in silicon germanium will applyto real laser materials like indium gal-lium arsenide,” said Floro. “If we canunderstand the physics, we can makebetter quantum dots.” The Department of Energy’s Officeof Basic Energy Sciences, Division ofMaterials Sciences, is funding this work.

Technical contact:Jerry Floro,(505) [email protected]


14

Quantum Dots (continued)

Understanding this self-organization is critical if we are

to control dot characteristics for lasing devices.

Observing the process ofdots going from oneshape to another to

relieve stress has providedinsight into the physics

governing islandformation.

[ [

Lynx, a new fine-resolution, real-time synthetic-aperture radar (SAR)system, has been unveiled by SandiaNational Laboratories and GeneralAtomics of San Diego. Designed to be mounted on bothmanned aircraft and unmanned aerialvehicles (UAVs), the 115-pound SARis a sophisticated all-weather sensorcapable of providing photographic-likeimages through clouds, rain or fog and

in daytime or nighttime conditions, allin real-time. The SAR produces imagesof extremely fine resolution, far sur-passing current industry standards forsynthetic-aperture radar resolution.Depending on weather conditions andimaging resolution, the sensor canoperate at a range of up to 85kilometers. “The Lynx represents a breakthroughon many fronts,” said Bill Hensley,

Sandia project leader. “Because theimage resolution is so fine and theinstrument itself is so lightweight, itrepresents a technology breakthrough.The real-time, interactive nature of theradar and the innovative operatorinterface make it a breakthrough formeeting the ease-of-use needs of front-line military users. And because Sandiadeveloped the technology and suc-cessfully transferred it to General


15

Sandia researcher Bill Hensley checks the Lynx synthetic-aperture radar (SAR) installed on a General Atomics I-GNAT unmanned aerial vehicle.

Fine-Resolution, Real-time Images20/20 Vision Even On a Cloudy Day

Atomics, the Lynx radar also is atechnology-transfer success story.” Mike Reed, Lynx program managerat General Atomics, said that Sandiaand General Atomics joined forces in1996 when the San Diego-based com-pany (whose Aeronautical Systems, Inc.affiliate builds a line of unmanned aerialvehicles) wanted to develop an advanc-ed, lightweight SAR system. GeneralAtomics provided the funds to Sandia,which already had a sophisticated SAR,to implement an enhanced design as acommercial product and deliver twoprototype units together with licensesand manufacturing information toproduce the unit. General Atomics and Sandia spentthe next three years working togetherto refine and enhance the SAR into alightweight, user-friendly system withextended range and much higher reso-lution. General Atomics has commencedproduction of subsequent units forcommercial sales. The new SAR will enhance thesurveillance capability of the GeneralAtomics Aeronautical Systems UAVsand other reconnaissance aircraft, whichpreviously were equipped only withcameras, infrared sensors, and older-generation SAR equipment. “Camerasprovided good data, but they don’t workat night or in rainy, foggy, and cloudysituations,” Hensley said. “Fine-resolution-image SAR radar is perfectfor these circumstances because it can‘see’ in the dark and peer through cloudsand fog.” Flying at an altitude of 25,000 feet,the Lynx SAR can produce one-foot-resolution imagery at standoff distancesof up to 55 kilometers. At a resolutionof four inches, the radar can makeimages of scenes 25 kilometers away(about 16 miles) even through cloudsand light rain. The radar operates in Ku band witha center frequency of about 16.7 GHz,although the precise value can be tunedto prevent interference with otheremitters. Lynx introduces several new charac-

teristics and functions. In addition tobeing very lightweight, the radar candetect very small changes in a scene byusing a technique called coherent changedetection. It also will be able to detectmoving targets. Sandia and General Atomics workedto make Lynx as much like an opticalsystem to use as possible. The radarforms an image covering an area largerthan that displayed, storing it in cachememory. This allows the operator topan around within the total scene toconcentrate on a particular area ofinterest. The radar’s fine resolutionallows it to detect small surface penetra-tions — even footprints in a soft terrain. Lynx has been flown successfully formore than 140 hours on a Departmentof Energy plane and on the GeneralAtomics I-GNAT. In all testing, the SARworked with the precision expected.

Sandia and General Atomics continueto explore ways to improve the Lynx.Future upgrades could include an inverseSAR mode for imaging of seabornetargets, interferometric SAR (requiringthe use of two antennas) for three-dimensional imaging, the ability to cueother sensors, and radio-frequencytagging — both for combat identificationand for precision-strike applications.Additional “cognitive” enhancementsare planned to make interpretation ofthe radar image more user friendly. General Atomics, founded in 1955,is involved in high-technology nuclearenergy, commercial, and defense-relatedresearch and development. Affiliatedmanufacturing and commercial servicecompanies include General AtomicsAeronautical Systems Inc., which buildsthe family of Predator, GNAT, Prowler,and Altus UAVs.

Sandia technical contact:Bill Hensley, (505) [email protected]

General Atomics contact:Mike Reed, (858) [email protected]


16

Flying at an altitude of25,000 feet, the Lynx SAR

can produce one-footresolution imagery at

standoff distances of upto 55 kilometers.

[ [The Lynx radar image shows the Rio Grande Valley .

Although Sandia began as a part of theLos Alamos laboratory during the ManhattanProject to develop the first atomic weapons,the decision to make it a stand-alonelaboratory — on Nov. 1, 1949 — launchedan exciting endeavor whose history is stillbeing written after 50 years. Sandia has grown to be the nation’s largestscience and engineering laboratory, withimpressive facilities in New Mexico andCalifornia, and at smaller sites scatteredaround the United States. Yet the labora-tory, then as now, is most distinguished forour staff. Indeed, Sandians’ sense of serviceto the nation has always been a major partof our laboratory’s character. We have a deepfaith in the power of technology — appliedwith wisdom — to change the world for thebetter, and we are continuing to build animpressive record of success through fund-amental inventions, scientific discoveries,applications of technology, and majornational projects. Our primary mission has always been theordnance engineering of all U.S. nuclearweapons, including system safety, security,and use-control for the weapons systems.We have achieved extraordinary levels ofreliability and safety for U.S. nuclearweapons, while in storage or in militarydeployment. Over the years, we have put theskills and capabilities developed for theweapons mission to good use in a widespectrum of other fields.

Achievements span many domains Sandia’s achievements span many domains— in the basic sciences of fission and fusion,in the creation of leading-edge electronicand microelectronic devices, in the creationand flight of advanced experimental satellites,in the pioneering of new sources of energyand greater efficiency in its use, in the devel-opment of some of the world’s fastestcomputers, in the development of new med-ical instruments, and through the creation ofenvironmentally conscious manufacturingtechniques now adopted by industries aroundthe world. Most of Sandia’s major advances couldlikely never have been achieved if approa-ched within the confines of a single scientificor engineering discipline. If there is a “secret

of success” for Sandia’s past 50 years, it isthis coming-together of world-class expertsfrom a variety of training and experience tomerge their unique contributions in tech-nically challenging endeavors.

Credit for the entire laboratory We began as a more narrowly focusedcopy of Bell Labs and have emerged sinceas one of the world’s most diverse scienceand engineering laboratories, numberingmore than 1,500 PhD researchers in a totalstaff of about 7,500. Although Sandians makeextraordinary individual contributions, ourculture is primarily world-wise technologistsworking together and understanding that “thejob gets done better if you don’t care whogets the credit.” Thus, over the years, thename most often credited and rememberedby the world for our major advances hasbeen simply “Sandia.” Our history, like all history, moves withfits and starts, and today’s challenges for thelaboratories are no less daunting than ever.The tensions of the Cold War have nowreceded, with many Americans hardlyremembering our role in helping to ensurea peaceful outcome. The current period ofgreater peace and stability has led to theinevitable questioning of the value of main-taining such a large laboratory for nationalsecurity. To Sandians, the greatest lesson of historyis that pressures for wars of national ag-gression are inevitable, until such time aswe can change the very nature of man — adoubtful and certainly long-term aspiration.Until that daunting challenge is met, if ever,it is only through maintaining strength thatthe United States and its allies can hope todeter aggression and defend their vitalinterests. Therefore, we must be conscien-tiously pursuing the advancement of defensescience and the perpetual strengthening ofthe U.S. deterrent — the best hope to preventmajor wars in the future.

Saying what we believe... regardless It is thus that the nearly 30,000 Sandiaemployees over these 50 years have foundtheir place — working in comfortable har-mony on larger-than-life national securitymissions. Not surprisingly, Sandians are

profoundly patriotic people. There is atradition at our laboratories that even thoughwe serve the national government, we alwaysplace service to the nation above service toany political party. Thus, we provide eachadministration and each Congress the verybest advice of which we are capable. Simi-larly, we must always say what we believeto be right, regardless of what potentialconsequences may result from it. Recently,a Sandian on assignment in Washington toldme that a senior government official, whomhe described as one of the real “movers andshakers,” said he believed Sandians couldbe counted on to give the best advice, evenif that advice could hurt the laboratory. Idon’t know of a higher compliment we couldbe paid.

Truman’s phrase still motivates Indeed, Sandia National Laboratoriesbegan with a request by President HarryTruman which contained the phrase “youhave here an opportunity to render anexceptional service in the national interest.”That simple but very profound phrase hasbeen taken to heart by every Sandian, andthey have used it as a catechism to motivatethemselves and each other ever since. Tounderstand this simple phrase in all of itscomplexity, its challenges, and its oppor-tunities — from inventing a new technologyto figuring out how to counter major threatsto our people and our way of life — is tofully appreciate the call we have eachanswered in serving the nation throughservice in this laboratory.

Nov. 1, 1999


INSIGHTS by C. Paul Robinson

President and Laboratories DirectorSandia National Laboratories

Sandia National LaboratoriesCelebrates 50 Years of“Exceptional Service”

Sandia is a multiprogram laboratory operated by Sandia Corporation, a LockheedMartin Company, for the United States Department of Energy

under contract DE-AC04-94AL85000

SAND 99-3107

Marketing Communications Dept.MS 0129Sandia National LaboratoriesP.O. Box 5800Albuquerque, NM 87185-0129

Bulk RateU.S. PostagePAID

Albuquerque, NMPermit No. 232

If there is a “secret of success”for Sandia’s past 50 years, it isthis coming-together of world-class experts from a variety of

training and experience to mergetheir unique contributions in

technically challengingendeavors.

C. Paul RobinsonPresident and Laboratories Director

Sandia National Laboratories

WINTER • 1999 - 2000 A QUARTERLY RESEARCH & … · Research, writing, and design by Technically...

Documents

Transcript of WINTER • 1999 - 2000 A QUARTERLY RESEARCH & … · Research, writing, and design by Technically...