january february 1996 - s&tr | september 2019 · 2019. 5. 29. · january/february 1996 this...

Science and Technology R

eviewLaw

rence Livermore N

ational LaboratoryP.O

.Box 808, L-664

Livermore, C

alifornia 94551

JanuaryFebruary 1996

Lawrence

Livermore

National

Laboratory

Printed on recycled paper.

Nonprofit O

rg.U

. S. P

ostage

PAID

Livermore, C

AP

ermit N

o. 154

GroundwaterCleanup withHydrostratigraphicAnalysis


Also in this issue: Micropower Impulse RadarAlso in this issue: Micropower Impulse Radar

Digital Mammography

SCIENTIFIC EDITOR

Becky Failor

PUBLICATION EDITOR

Sue Stull

WRITERS

Arnie Heller, Robert D. Kirvel, Dale Sprouse, and Katie Walter

ART DIRECTOR

Kathryn Tinsley

DESIGNERS

George Kitrinos and Kathryn Tinsley

GRAPHIC ARTIST

Treva Carey

COMPOSITOR

Louisa Cardoza

PROOFREADER

Catherine M. Williams

S&TR is produced by the TechnicalInformation Department as a service for the Director’s Office.

2 The Laboratory in the News

4 Patents and Awards

5 Commentary on Environmental Restoration

Features6 Groundwater Cleanup Using Hydrostratigraphic Analysis

Effective site cleanup of underground contaminants is achieved ahead of schedule by using hydrostratigraphic analysis.

16 Micropower Impulse RadarInvented and developed at LLNL, this inexpensive and highly sensitive radar system produces and samples extremely short pulses of energy. This novel technology is finding dozens of new uses in Laboratory programs and in sensor devices for homes, automobiles, factories, and hospitals.

Research Highlights30 Probing with Synchrotron-Radiation-Based Spectroscopies32 Operating a Tokamak from Across the Country

36 Abstracts

S&TR Staff January/February 1996

LawrenceLivermoreNationalLaboratory

Printed in the United States of America

Available fromNational Technical Information ServiceU.S. Department of Commerce5285 Port Royal RoadSpringfield, Virginia 22161

UCRL-52000-96-1/2Distribution Category UC-700January/February 1996

This publication is a continuation ofEnergy and Technology Review.

Page 6

Page 16

About the Review

About the Cover JanuaryFebruary 1996

Lawrence

Livermore

National

Laboratory



Also in this issue: Micropower Impulse RadarAlso in this issue: Micropower Impulse Radar

Returning groundwater to its natural state is high on priority lists at Lawrence LivermoreNational Laboratory. Our feature article thismonth describes how groundwater cleanup atthe Laboratory is progressing faster thananticipated, thanks to new developments inhydrostratigraphic analysis. This multidisciplinary“smart pump-and-treat” approach maximizesextraction as it links data on physical propertiesof sediments, groundwater, and contaminants.This all translates into fewer wells, less time,and lower cost. We will be sharing ourcomprehensive cleanup know-how with otherenvironmental restoration projects.

The Lawrence Livermore National Laboratory, operated by the University of California for theUnited States Department of Energy, was established in 1952 to do research on nuclear weapons andmagnetic fusion energy. Science and Technology Review (formerly Energy and Technology Review) ispublished ten times a year to communicate, to a broad audience, the Laboratory’s scientific and technologicalaccomplishments, particularly in the Laboratory’s core mission areas—global security, energy and theenvironment, and bioscience and biotechnology. The publication’s goal is to help readers understandthese accomplishments and appreciate their value to the individual citizen, the nation, and the world.

Please address any correspondence (including name and address changes) to S&TR, Mail Stop L-664,Lawrence Livermore National Laboratory, P.O. Box 808, Livermore, California 94551, or telephone(510) 422-8961. Our electronic mail address is [email protected].

Prepared by LLNL under contractNo. W-7405-Eng-48

• •

Cov

er p

hoto

s: J

ames

Sto

ots

S&TR is available on the Internet athttp://www.llnl.gov/str/str.html. As referencesbecome available on the Internet, they will beinteractively linked to the footnote references at the end of each article. If you desire moredetailed information about an article, click on anyreference that is in color at the end of the article,and you will connect automatically with thereference.

Electronic Access

We want to know what you think of ourpublication. Please use the enclosed survey form to give us your feedback.

What Do You Think?

3

Science & Technology Review January/February 1996

2 The Laboratory in the News


Collaboration for advanced tissue welding systemLivermore and Conversion Energy Enterprises (CEE) of

Spring Valley, New York, have embarked on a cooperativeventure aimed at developing a prototype automatic medicalsystem for laser welding of tissue. LLNL will lend its expertisein lasers, computers, optics, and microtool development.

Tissue welding would be faster than traditional methods,could be used in areas difficult to reach, would make betterjoints, speed healing, and decrease risk of complications.

Current methods for binding tissue together are stitching andstapling. While functional, both can allow puckering aroundthe wound and seepage of fluids. For some difficult-to-get-atlocations, stitching or stapling is just not possible.Contacts: Dennis Matthews (510) 422-5360 ([email protected])or Luiz Da Silva (510) 423-9867 ([email protected]).

Blood-gas monitoring system under developmentThe Laboratory and Novametrix Medical Systems Inc. of

Wallingford, Connecticut, are seeking to develop a quick, easy,noninvasive way to monitor blood for substances such as oxygen,carbon dioxide, anesthesia, or alcohol.

Important for monitoring the condition of patients,measurements of such gases in the bloodstream are usedduring surgery or routinely throughout the day.

The objective of the collaboration is to develop a way to uselight projected through skin to read blood gases more accurately.In conducting the work, LLNL expertise in optics, fiber optics,and signal processing will be combined with Novametrix’sexperience in developing blood gas monitoring instruments.Contact: Howard Nathel (510) 423-3262 ([email protected]).

Optical system for carpal tunnel surgeryThe Laboratory has teamed with Envision Medical Corp., a

manufacturer of medical video camera systems, located in SantaBarbara, California, to develop technology that will help surgeonsto perform the endoscopic carpal tunnel release procedure.

So far, endoscopic surgery to correct hand problems due torepetitive stress has had limited success. One problem is thatthe surgeon cannot get an adequate view of the interior of thehand. The joint project seeks to produce a better endoscopiccameravideo system for minimally invasive surgeries such as carpaltunnel surgery.

The Cooperative Research and Development Agreement(CRADA) will make use of LLNL’s knowledge of optics, sensors,computer imaging, and microtools to improve surgical successand to reduce both post-operative pain and the current national

yearly price tag for direct and indirect costs of carpal tunnelsyndrome—estimated at $10 to $15 billion. Contacts: Robert Van Vorhis (510) 423-1693 ([email protected]) orSteve Burastero (510) 424-4506 ([email protected]).

Mammoth Mountain mystery unravelingWorking with the U.S. Geological Survey (USGS),

Laboratory scientists have been helping unravel the mystery ofthe death of trees in California’s Mammoth Mountain area andthe near-asphyxiation of a Forest Service Ranger in 1990.

In the August 24, 1995, issue of Nature magazine, membersof the USGS group and Lab scientist John Southon wrote thatthey believe seismic activity deep inside the mountain is causingemissions of magmatic carbon dioxide (CO2) similar to thoseobserved in other volcanic areas like Mt. Etna and Mt. Vesuvius.

Southon, a member of Livermore’s Center for MassSpectrometry, said his principal role in the investigation hasbeen to provide the expertise of LLNL’s radiocarbon group toconfirm the conclusions reached by the primary USGS researchers.

Lab scientists, for example, aided the USGS team bydistinguishing emissions of magmatic CO2 from those resultingfrom naturally occurring biological activity. Recently LauraHainsworth, a post-doctoral fellow in the group, has been usingcarbon-14 measurements on individual rings in the tree remainsto determine whether there is a correlation between CO2 emissionsand seismic activity at Mammoth. She plans to extend this workto other volcanic sites like Mt. St. Helens.Contact: John Southon (510) 423-4226 ([email protected]).

CD-ROM package aids breast cancer researchersThe Laboratory has assembled the first CD-ROM library of

mammogram images that is designed to help researchers perfectcomputer software for detecting one early sign of breast cancer.

The 12-CD package contains digitized images of mammogramsexhibiting microcalcifications, an early cancer indicator that canbe very small and difficult to see. The Laboratory’s high-enddigitizer was able to produce digital images at a resolution of35 micrometers.

The mammograms include documented conditions ranging fromharmless cysts to malignant tumors. Because the CD images are ofknown diagnosed conditions, breast cancer researchers can processthe images with their computers to see whether their programscan detect the actual calcifications in the CD mammograms.

Each image in the CD-ROM library, developed in collaborationwith the University of California Medical Center in San Francisco,is accompanied by a matching image that marks the exact locationand extent of any lesions.Contact: Laura Mascio (510) 422-0924 ([email protected]).

True 3-D motion computer imaging developedLivermore computer scientists have developed the next step

in computer imaging—true 3-D motion imaging. The system,dubbed “CyberSight,” can digitally capture and display moving3-D subjects to a degree of realism never before achieved.

CyberSight works by first capturing a subject on video using astereo camera system. Instead of placing markers on the subject asreference points (as is done in current motion-capturing systems),line patterns are projected onto the subject. The pattern data ispicked up by the cameras and fed to a computer, which transformsthe data into complete surface reconstructions, in motion andhigh resolution, to exact measurements.

The Laboratory is seeking opportunities to commercialize theCyberSight technology and to expand it into different applicationsthrough licensing or joint development. Possible uses forCyberSight range from greater cinematic realism to improvedcontrol of industrial robots. CyberSight has potential applicationsin the medical area, for example, to analyze the movement ofpatients with cerebral palsy or to assist surgeons in the effects ofplastic surgery. In security applications, CyberSight mightprovide a reliable facial recognition system. Defense applicationsmight involve 3-D modeling of material deformation under stressor determining proper fit for military equipment, such as gas masks.Contact: Shin-yee Lu (510) 422-6882 ([email protected]).

Subatomic solution sought for “dark matter” mysteryA search is under way at Livermore for the answer to one

of the most profound mysteries of the universe: the nature ofdark matter. Dark matter constitutes as much as 90 to 99% of the universe. Astrophysicists know it is there because itsgravitational pull keeps our own Milky Way galaxy fromflying apart. But positively identifying the invisible material isanother story. Theories range from large planet-like objects tomicroscopic axions.

At Livermore, a powerful electromagnet is being used tosearch for axions. If the existence of axions is confirmed, thefinding would have a sudden and dramatic impact on howastrophysicists worldwide view the basic construct of the universe.

If axions do exist, they likely weigh a billion and perhaps a trillion times less than the lightest known particle, the electron.Furthermore, their interaction with matter and radiation isexpected to be extraordinarily weak, which has posed a majorchallenge in even conceiving of an experiment that could detectthem, let alone carry out such a search.

The research team’s solution is a specially made 4-m-high,12-ton magnet that has been lowered into a cylindrical hole inthe floor of a secluded building at the Laboratory. The strongmagnetic field will stimulate the axions to convert into a veryweak microwave signal in a tunable microwave cavity resonator

with state-of-the-art cryogenic amplifiers. Searching for the axionis very similar to tuning one’s car radio very slowly, lookingfor a weak station.

The experiment is a collaboration of Livermore,Massachusetts Institute of Technology, University of Florida,UC Berkeley, Lawrence Berkeley National Laboratory,University of Chicago, Fermi National Accelerator Laboratory,and Institute for Nuclear Research of the Russian Academy ofSciences in Moscow. Approved for construction in January1993, the $1.4-million experiment took more than two years toassemble. Researchers began taking data in November 1995.Contact: Karl Van Bibber (510) 434-8949 ([email protected]).

Allenby, Dimolitsas join LLNL management teamBraden Allenby has joined the Laboratory in a two-year

term appointment to lead LLNL’s new strategic initiative oflong-range unified energy and environmental programs.Allenby is a leading researcher on industrial ecology, thescience of balancing ecosystems with industrial systems.

Director Bruce Tarter said that Allenby will help theLaboratory create a “strategic, integrating framework for long-range program and resource development” for the Laboratory’senergy and environmental programs.

Allenby comes to Livermore from AT&T, where he wasvice president of research for technology and environment.After his two-year tenure at Livermore as director of energyand environmental systems, he will return to AT&T’s globalmanufacturing and engineering organization.

Spiros Dimolitsas has joined the Laboratory as AssociateDirector for Engineering. Dimolitsas, formerly with COMSAT,held a series of positions of increasing responsibility in areas ofsignal processing and noise filtering; speech, image, and datatransmission; network quality; and communications systemsengineering. His most recent COMSAT assignment wasdirector of program development.

In announcing the appointment, Director Tarter citedDimolitsas’ COMSAT experience as particularly relevant tothe Laboratory. Tarter commented, “His company, likeLivermore and institutions elsewhere, is undergoing seriousself-examination of mission and methods.

“He has had experience with significant changes in fundingexpectations and resource allocations. He has responded withdecisions and practices that strengthened the company andraised revenue while maintaining a focus on the value of theindividual employee.”

The Laboratory in the News

4 5


Each month in this space we report on the patents issued to and/orthe awards received by Laboratory employees. Our goal is toshowcase the distinguished scientific and technical achievements ofour employees as well as to indicate the scale and scope of thework done at the Laboratory.

Patents and Awards

Patent issued to

Joe W. GrayDaniel Pinkel

Daniel L. BirxPhillip A. ArnoldDon G. BallEdward G. Cook

William D. DailyAbelardo L. RamirezRobin L. NewmarkKent UdellHarley M. BuettnerRoger D. Aines

Daniel D. DietrichRobert F. Keville

Alan D. ConderRonald E. HaighKeith F. Hugenberg

Patent title, number, and date of issue

Methods for Chromosome-Specific Staining

U.S. Patent 5,447,841; September 5, 1995

Air and Water Cooled Modulator


Dynamic Underground Stripping: Steam andElectric Heating for In Situ Decontaminationof Soils and Groundwater


Mini Ion Trap Mass Spectrometer


Triggerable Electro-Optic AmplitudeModulator Bias Stabilizer for IntegratedOptical Devices


Summary of disclosure

A method of staining target chromosomal DNA, employing labeled nucleicacid, blocking nucleic acid and chromosomal DNA in in-situ hybridization.Labeled repetitive segments are substantially blocked from binding to thechromosomal DNA, while hybridization of unique segments within thelabeled nucleic acid to the chromosomal DNA is allowed, permittingdetection of hybridized labeled nucleic acid containing unique segments.

A compact, reliable, high-average-power magnetic compression circuithaving a lifetime greater than 5,000 hours, with an average power greaterthan 8 kW, which can deliver at least 20-kV pulses of less than a 200-nsduration at a repetition rate greater than 4 kHz and peak power output of at least 5 MW. A solid-state switched magnetic compression circuit usesprinted circuit board technology with a physical layout that providessufficient air and water cooling.

A process for removing localized underground contamination of volatileorganic compounds by heating a contaminated area using steam injectionand electric currents to vaporize the contaminants, and then removing themigrating subsurface fluids and vapor by vacuum extraction and liquidpumping. Injection and extraction wells are constructed within or aroundthe periphery of the contaminated area.

A miniature ion cyclotron resonance (ICR) mass spectrometer thatcombines a unique electron source and mass analyzer/detector in a singledevice, and is mounted into a hollow permanent magnet, for detectingenvironmental pollutants or illicit substances. Its low power consumptionmakes portability possible.

An apparatus for applying a DC bias modulation to a conventionalintegrated optical Mach-Zehnder electro-optic modulator. A DC bias boxcontains four basic interrelated circuits: a trigger circuit, a ramp and holdcircuit, a negative peak detection circuit, and an adjustable delay circuit.The DC bias box ramps the DC bias along a transfer function curve to anydesired phase or point of operation.


INCE contaminants were first detected in groundwaterand soil at the Laboratory, we have been working to find

the best ways to mitigate the problem, always with the goalof doing the job in the shortest time possible in the mostcost-effective way possible. We are fortunate to have at theLaboratory a multidisciplinary staff accustomed not only totaking a science-based approach to problem solving but alsoto being innovative in finding and pursuing creative solutionsto difficult problems.

We began using standard environmental restorationindustry methods to attack the groundwater contamination atLivermore. But in our search for more cost-effective solutionsto accomplish the cleanup, we have developed and are applyingseveral new and modified tools to attack the problem. Mostof these tools can be applied to remediation projects anywhere.

Hydrostratigraphic analysis, derived from practices in theoil and gas industry, is a relatively new tool in groundwaterapplications. We have found it useful for characterizing thesite—for organizing our accumulated data on geology,hydraulics, and chemistry into a three-dimensional,conceptual model of our subsurface. Computerizedgroundwater modeling is another tool, which helps us to forecast how the subsurface will respond to variousremediation treatments. It also helps us to optimize cleanupefforts—with the goal of reducing total cleanup costs. Whilegroundwater modeling is by no means new, we have developedseveral new models with interactive visualization software,which can be used by groundwater remediation planners atvirtually any site.

The standard method for removing contaminants fromgroundwater is “pump-and-treat,” which involves pumpingcontaminated groundwater to the subsurface and then treatingit to remove the contaminants. By applying our modelingcapabilities and by managing our extraction well field efficiently,we have developed a “smart” pump-and-treat system that canclean groundwater to below regulatory-mandated concentrationssignificantly faster than the conventional application of thepump-and-treat method.

Congressional budgetary pressure plus the necessity toremediate environmental contamination require that innovative,cost-effective methods for environmental restoration be developedto address environmental problems effectively and affordably.One such method is dynamic underground stripping (see Energy& Technology Review, April 1994), which we developed incollaboration with UC Berkeley. We used it to remove gasolinefrom groundwater and soil. Electrical heating and injectedsteam mobilize the gasoline contaminants for extraction andsubsequent above-ground treatment. This demonstration projecttook less than a year to remove approximately 7,600 gallonsof gasoline that would have taken decades to remove at asignificantly higher cost with conventional technology. Weare also evaluating other in-situ bioremediation and enhancedthermal treatment methods for cleanup operations here.

As this work has progressed over the 13 years sincecontamination was first discovered at Livermore, we havebeen learning. We know much more about the site than we didin 1983, plus we also have learned about the various steps inthe planning and cleanup process. The characterization phase—during which we drilled many dozens of monitor wells, tookhundreds of soil and groundwater samples, and collected a varietyof other valuable data—would today proceed differently becausewe learned how to more effectively characterize a contaminatedsite. We have advanced the state of the art of groundwatermodeling and use it to greatly shorten cleanup times, makingboth conventional and innovative groundwater remediationmethods less costly. We have developed a better understandingof and techniques for removing volatile organic compounds,fuel hydrocarbons, and other hazardous materials from soil andgroundwater.

Our dedicated team of scientists, engineers, and techniciansis successfully pursuing a complicated groundwater remediationprogram at Livermore. We want to give this country a return onits investment in the Laboratory by bringing the site cleanupsto completion as cost effectively and as quickly as possible,and then transferring our accumulated know-how to industryso it can be put to use on other environmental restoration projects.

Commentary on Environmental Restoration

S

Patents

AwardsThe American Physical Society has bestowed two awards for physicsexcellence to groups of Laboratory employees. The first, for excellencein plasma physics is shared by a nine-person team that includes SteveHaan, Dave Munro, and Steve Weber from X Division in the Physicsand Space Technology Directorate; Russell Wallace of the Chemistryand Materials Science Directorate; Gail Glendinning, JosephKilkenny, and Bruce Remington of the Inertial Confinement Fusion(ICF) Program in the Lasers Directorate; and two University of Rochesterscientists. The second, the Simon Ramo Award for outstandingdoctoral thesis research, has been awarded to Chris Decker, a post-doctorate staff member of X Division.

For his work in providing a means for more reliable mammography, JoséHernández has been cited for Outstanding Technical Contributionfrom the Hispanic Engineer National Achievement Awards.Hernández, an electrical engineer at the Laboratory, was instrumental inthe development of a full-field digital mammography system.

Brian MacGowan and Bruce Remington have been made fellowsof the American Physical Society in recognition of their outstandingcontributions to physics. Their fellowships come from the Division ofPlasma Physics and cite MacGowan’s work in short-wavelength x-raylasers and Remington’s experiments in Rayleigh–Taylor instability.

The American Physical Society also honored Luiz Da Silva with theNuclear and Plasma Sciences Society Early Achievement Awardfrom the Institute of Electrical and Electronics Engineers. Da Silvawas cited for his work in the development of x-ray lasers and theirapplications to probing plasmas and biological imaging.

Marriann Silveira, Tom Boock, and Al Huntley of the Laboratory’sEnergy Management Program and Clark Scott, Water Shopsupervisor, journeyed to Washington, D.C., recently to receiveFederal Energy and Water Management Awards for theirrespective groups.

Harry GallesHead, Environmental ProtectionDepartment

7

integrate chemical, hydraulic, andgeologic data into a detailed, three-dimensional model of the subsurfaceenvironment.1 As an integral part ofsmart pump-and-treat for the past two years, the process demonstrates thatthe Livermore groundwater cleanupefforts are being conducted in acomprehensive and cost-effectivemanner. The success of the process isdrawing inquiries from federal agenciesand other national laboratories eager formore information on better ways tocharacterize, monitor, and clean upgroundwater contamination atSuperfund sites across the U.S.

Hydrostratigraphic analysis isproving to be an effective managementtool for making better-informed andmore timely decisions regardinggroundwater cleanup. These decisionsinclude positioning and designingextraction and monitoring wells,prioritizing the construction and phasedstartup of remediation systems,managing the extraction of subsurfacecontaminants, finding the sources ofpast contaminant releases, andevaluating the effectiveness of theremediation systems.

The technique is also an effectivevisualization tool for presenting

LLNL.4,5 The contaminants of greatestconcern are two VOCs dissolved inwater some 50 to 200 feet* below theground. They are primarily theindustrial solvents perchloroethylene(PCE) and trichloroethylene (TCE).

Pump-and-Treat

Although the estimated volume ofthe solvents in the subsurface is onlyabout 200 gallons, the concentrations ingroundwater range up to about 10 partsper million in some areas (state and

Individual contaminant plumes are effectively targeted for hydraulic capture and cleanup

by a Livermore team of researchers. Their use of hydrostratigraphic analysis integrates

chemical, hydraulic, geologic, and geophysical data, which results in a three-dimensional

conceptual model of the subsurface area.

6

Groundwater Cleanup Using Hydrostratigraphic Analysis


world-renowned center ofapplied scientific research,

Lawrence Livermore NationalLaboratory is the source of some of themost complete characterizations ofnature, from the human chromosome tothe atom. Yet few people realize thatthe Livermore site itself is also one ofthe most well-characterized sites in theU.S., if not the world.

Some 30 hydrogeological crosssections for the site recently have beenprepared using a newly developedmethodology that incorporates datafrom over 500 boreholes drilled as partof the Laboratory’s groundwatercleanup program. The cross sectionsreveal a complex maze of geologicalzones underlying the Livermore site and

complex geologic and groundwaterremediation issues to the Department ofEnergy, federal and state regulatoryagencies, and the local Tri-Valleycommunity. In addition, hydrostrati-graphic analysis forms the basis of two-and three-dimensional computersimulations of groundwater contaminanttransport using advanced physics codesto estimate cleanup times, costs, anddesign parameters. Finally, it shouldprove to be a valuable method toevaluate the effectiveness of innovativecleanup technologies, such as dynamicunderground stripping.

Before implementing hydrostrati-graphic analysis in early 1994,Laboratory environmental experts hadconstructed numerous maps and crosssections showing the distribution ofhazardous materials known to beresiding in some of the complexgeological strata underlying the site2,3

(see sidebar on p. 10). Although thesemaps formed a solid basis for planningthe groundwater cleanup at LLNL, theycould not be directly used to implementcleanup because the subsurfacecontaminant pathways were not shownor well understood. Hydrostratigraphicanalysis is an extension of this previoushydrogeologic work performed at


beyond. In depicting variousunderground geologic strata, the crosssections are of more than passingacademic interest. They show thelocation of underground contaminantsand the distribution of extraction andmonitoring wells constructed tomonitor, remove, and treat thosecontaminants. The overall data also showthe nature of the interconnectedness ofthe strata and migration of contaminantswithin them.

What’s more, thanks to aninnovative groundwater cleanupstrategy known as “smart pump-and-treat,” the cross sections measured over the past seven years reveal theindisputable shrinkage and hydrauliccontrol of plumes of contaminantscalled volatile organic compounds(VOCs) that once posed a risk to localmunicipal water supplies (Figure 1).

Decisions and Visualizations

The cross sections are the result ofa process, called hydrostratigraphicanalysis, that allows scientists to

A

Jim Chiu samples

extraction-well

groundwater to

measure progress

in the Laboratory’s

groundwater

cleanup efforts.

*Hydrogeologic measurements useU.S. units rather than metric.

9


Hydrostratigraphic Analysis

to be cleaned up, not necessarily on thelogistics of how to clean them up,” notesLLNL hydrogeology team leader RickBlake. “As we planned cleanup andbegan implementation, we realized thatmore needed to be understood abouthow the contaminated strata areinterconnected. What we lacked was asite-wide road map of the subsurface,which would allow us to target specificcontaminated zones and enable us toplace our extraction wells at optimumlocations to meet our cleanup objectives.”

In the initial phases of the cleanupproject, Livermore experts used two-dimensional maps and cross sectionsplus scientific and engineering estimatesas they placed wells throughout the siteand off site (Figure 2). Although thesehydrogeologic cross sections showedindividual permeable zones in detail

(see Figure 3), “we realized that if wecould group these zones into units thatare hydraulically interconnected, wecould simplify implementation,”explained Blake.

The Livermore site and surroundingvicinity are underlain by a complexnetwork of alluvial silts, clays, sands,and gravels that are tens to hundredsof feet thick. Known as the Upper andLower Livermore Formations, theseformations were deposited by multiplestreams within the past 2 million years.The sand and gravel channels withinthese formations serve as migrationpathways for various contaminants.Compounding this complexity, theshifting of tectonic plates underlyingCalifornia has tilted the geological strata,causing groundwater and any dissolvedcontaminants to travel westward at arate of about 70 feet per year.

Multidisciplinary Approach

Because of its complexity, theLivermore subsurface cannot be fullycharacterized using geology alone.“We knew we had to take a moremultidisciplinary approach to get thejob done,” notes Blake.

Together with Weiss Associateshydrogeologists Michael Maley andCharles Noyes, Blake developed thehydrostratigraphic framework for thesite. A geologist with 15 years ofexperience in the oil and gas industry,Blake was familiar with stratigraphicanalysis used by oil and gas companiesfor decades in their search for newaccumulations of petroleum and naturalgas. “The key to unlocking the detailsof the strata underneath the Laboratory

8



up the plumes. Minimizing the cost ofgroundwater cleanup using pump-and-treat technology requires a thoroughunderstanding of the hydrogeologicfactors that control the flow and transportof contaminants in the heterogeneoussubsurface.6

Improving the Maps

“In the early years of theenvironmental restoration program,emphasis was placed on locating sourceareas and identifying strata that needed

Vas

co R

d.

100200

5

5

15

15

Arroyo Seco

Arroyo Seco

Arroyo Seco

Arroyo Seco

100

55

5

5

15

1515

15

Vas

co R

d.

East Ave.

100

5

Vas

co R

d.

East Ave.

100

100

5

5

15

15

Vas

co R

d.

East Ave. East Ave.

NO

RT

H

5

5

1515

15

15

LLNL

(a) October 1988 pre-pumping (b) April 1993

(c) April 1994 (d) February 1995

Scale: feet

0 500 1000

PCE isoconcentration contour (ppb)

Extraction well

5

Mapped area

Gre

envi

lle R

d.

Patterson Pass Rd.

East Ave.

Vas

co R

d.

LLNL

Hazard Ranking Scoring System anddecided to list the site on the NationalPriorities List (Superfund). As a result,the Laboratory was charged by state andfederal environmental regulatory agencieswith cleaning up the contaminatedgroundwater and stopping the westwardmigration of the plumes.

The primary remediation technologyin use at the Livermore site isgroundwater pump-and-treat. Thistechnique uses a network of extractionwells that pump contaminatedgroundwater to the surface for treatmentto remove contaminants. A network ofmonitoring wells is used to track theeffects of groundwater extraction andmeasure how effectively the Laboratoryis hydraulically capturing and cleaning

federal maximum contaminant levelsfor the two compounds are 5 parts perbillion). All together, past discharges ofthese materials from a number of areas atthe site have contaminated approximately3 billion gallons of groundwater belowthe Laboratory and immediate vicinitycovering approximately a square mile.

The VOCs formed severalunderground plumes below theLaboratory, some of which have traveledslowly off site underneath Vasco Roadand beyond the Laboratory’s westernboundary. Because of estimatedmigration paths and flow rates, theplumes were judged to pose a potentialrisk to municipal water supply wellslocated about 1.5 miles to the west. TheU.S. Environmental Protection Agencyevaluated the contamination using a

Figure 1. Concentration maps of the western edge of LLNL illustrate the

decrease in PCE concentrations over six and a half years resulting from

hydraulic capture and PCE mass removal at extraction well locations.

Gre

envi

lle R

d.

Vas

co R

d.

East Ave.

Patterson Pass Rd.

LLNL

NO

RT

H

Arroyo Seco

1000

1,000

100

100

100

100

100

100

100

10

10

10

10

10

10

10

1

D

E

F

GA

B

CScale : feet

0 500 1000

Total VOC isoconcentration contour (ppb)

Groundwater flow direction

Existing Treatment Facility

Planned Treatment Facility

10

Figure 2. This type of two-dimensional contaminant

plume map was used initially for preliminary planning

of LLNL pump-and-treat groundwater cleanup.

11



ensure that they honor all of theindependent data sets used to developthe final interpretation. “This is themost laborious and tedious part of theprocess, but one that is absolutelycritical for developing a technicallydefensible, comprehensive interpretation,”says Blake. “As you can imagine, with 30 interconnected cross sections, onceyou change interpretations on one crosssection, changes cascade through mostof the others, requiring many hours ofcareful, painstaking revisions.”

Pathways Revealed

Careful analysis of hydraulic datafrom monitoring wells revealed thatmany of the underlying strata, oncebelieved to be geologically andhydraulically separate, are actuallyinterconnected. This interconnectednessbecame evident when active pumpingat one well resulted in the water levelbeing drawn down in other wells aroundthe site. These wells, although drilledto varying depths in what appear to beseparate permeable zones, as shown inFigure 4, actually tap into a singlehydraulically interconnected unit.

Using data and observations such asthese, Blake and coworkers found thatthe underlying Livermore site andneighboring area just west of theLaboratory are divided into seven layers,called hydrostratigraphic units (HSUs),each stacked on top of the other with low-permeability sediments separating oneHSU from another. Figure 5 is a crosssection showing VOCs above cleanuplevels in each hydrostratigraphic unit.Because low-permeability boundarieslimit water moving between HSUs,contaminants mostly travel withinindividual HSUs during their slowmigration westward.

10



was to integrate the oil and gas industrytechnology with hydrogeologicapproaches used in the environmentalcleanup industry,” he says.

Hydrostratigraphic analysisaccomplishes this task by linking dataon physical properties of the sediments,groundwater, and contaminants fromextraction and monitoring wells. Thisinformation includes:• Descriptions of geologic formationsand their physical properties, includingborehole geophysical logs, in whichelectrical currents and gamma radiationdetectors create distinct signaturesrevealing the nature of subsurface rockand sediment types and their containedfluids.• Hydraulic test data, including evaluationof the response of observation wellsduring aquifer pumping tests todetermine the extent of hydrauliccommunication, or water movement,from one geologic stratum to another.• Groundwater elevation data measuredin monitoring wells for evaluatinggroundwater flow directions.• VOC concentrations in soil andgroundwater for mapping contaminantdistributions.7• Plume “signatures” based on the ratiosof VOC concentrations, which can beused to trace an individual contaminantplume back to its source area.

Since 1984, these data have beencollected at LLNL by staff andenvironmental consultants using rigorousstandard operating procedures as well asquality assurance and control protocols.Strict adherence to these procedures hasresulted in an enormous database of high-quality environmental data concerningthe site.

Although Livermore-developedsoftware is used to display much of thissubsurface data on maps and crosssections (Figure 3), the actual process ofdata integration is performed manuallyby the geologists and hydrogeologists.During this process, subsequent revisionsof the working maps and cross sections

600

500

400

300

200

100

0

East Ave.

*

***

*

*

*

*

*

*

***

**

**

*

**

**

**

*

**

**

**

*

**

***

** *

**

**

**

** *

***

**

**

*

**

**

*

**

**

*

**

**

**

**

**

**

**

**

**

**

N

Ele

vatio

n, fe

et a

bove

mea

n se

a le

vel ?

?

?

?

?

?

?

???

?

??

?

?

?

????

?

??

?

?

?

?

?

?

? ?

?

??

?

?

??

?

2500

0

40

Approximate scale in feet

6.3 × vertical exaggeration

S

Vasco Rd.

Arroyo Seco

Location of cross section

LLNL

Moderate- to high-permeability sediments

Potentiometric head (6/88)

Sediments analyzed for VOCs

Contact; dashed where inferred, queried where uncertain or diagrammatic

VOCs

*?

Facing east

Figure 3. Hydrogeologic cross sections like this one show the detailed geology and

contaminants, but do not show how the strata are interconnected.

Origins of Cleanup Efforts

Much of the groundwater contamination underlying the Livermore site originated inthe early 1940s when the U.S. Navy converted some 640 acres from agricultural useinto a flight-training base and aircraft assembly and repair facility. Most of thecontaminant releases from this time are believed to have been solvents used to cleanairplanes, their engines, and associated parts. Smaller releases of gasoline, diesel fuel,and other compounds are also known to have occurred.

From 1950 to 1954, California Research and Development Co., a subsidiary ofStandard Oil, occupied the southeastern portion of the site. This marked the beginningof testing with radioactive materials at the site and probably the first releases of smallamounts of tritium (a radioactive isotope of hydrogen) to the environment.

Since the Laboratory’s founding in 1952, additional releases are attributed tolocalized spills, landfills, disposal pits, broken sewer lines and pipes, and leaking tanks.Releases of solvents were the most prevalent, although small releases of polychlorinatedbiphenyls (PCBs), metals, radionuclides (primarily tritium), gasoline, and pesticidesalso occurred.

In 1983, LLNL personnel detected VOCs (volatile organic compounds) on site andin domestic water supply wells just west of the site that were in concentrations abovemaximum contaminant levels (MCLs). The Laboratory immediately informed theregulatory agencies and owners of private wells nearby and provided city water hookupsto affected residences. The State of California issued a regulatory order in 1984 toinvestigate groundwater quality underlying LLNL and off site, ultimately leading toinvestigation of more than 350 potential release sites.

Because the VOC concentrations exceed drinking water standards and are ingroundwater within 1.5 miles of a municipal water supply, the U.S. EnvironmentalProtection Agency placed LLNL on the National Priorities List (Superfund) in 1987 forcleanup. Other environmental problems, such as leaking underground tanks and closureof hazardous waste management facilities, are managed under this program, too. Inresponse, the Laboratory drew up a comprehensive remedial action plan, which wasreviewed by regulatory agencies, the DOE, and the public. Where MCLs vary betweenstate and federal regulations, the Laboratory must observe the stricter level.

Pilot groundwater cleanup began in fiscal year 1989. To date, more than 200 milliongallons of VOC-contaminated groundwater have been extracted, treated, and then eitherput into a recharge basin or reused.

The Laboratory’s Environmental Restoration Division, part of the EnvironmentalProtection Department, is the focal point for the development of restoration and wastetreatment techniques needed for environmental cleanup on site and off site. For efficientsoil and groundwater cleanup, program scientists and engineers have developed and areusing advanced sampling, monitoring, and two- and three-dimensional modelingtechniques for underground cleanup operations. Hydrostratigraphic analysis providesthe necessary framework to successfully carry out this effort. This so-called “smart”approach saves time and money compared to conventional pump-and-treat groundwatercleanup programs. The Laboratory also has evaluated treatment methods for VOCs ingroundwater, including ultraviolet light/hydrogen peroxide oxidation, air stripping, andsolar detoxification. Steam flooding and soil heating were conducted to remediate agasoline spill from the 1970s.

13



the units are indistinct or missing. Thus,it may be necessary to redefine HSUboundaries as new field data are collected.Because of the general westward dip ofthe geological strata, VOCs initiallypresent near the surface are found atincreasing depths from east to west withina unit. In addition, once VOCs migratevertically downward from source areasthrough the vadose zone (the area abovethe water table) and encounter the saturatedzone, their transport becomes primarilylateral and follows the groundwater flowdirection within the HSU.

HSU-5, in the eastern portion ofFigure 5, shows this relationship. Asource near an old salvage yard in thesoutheastern quadrant of the siteintroduced VOCs into HSU-5 several

12



HSU-6

HSU-5

HSU-4HSU-3B

HSU-1B

Water table

HSU-1A

HSU-3A

HSU-2

NorthA'South

A

W-311W-360W-352W-277W-292W-276

W-370

W-370

285.0

226.5

260.0

260.0

254.0267.0247.5

235.0

Ground surface

Saturated zone

0

50

100

150

200

250

0

50

100

150

200

250

0

50

100

150

200

250

0

50

100

150

200

250

0

50

100

150

200

250

0

50

100

150

200

250

0

50

100

150

200

Unsaturated zone

Gre

envi

lle R

d.

Patterson Pass Rd.

East Ave.

Vas

co R

d.

NO

RT

H

A'

A

LLNL

Cross-section location

Facing west

Well name

HSU boundary

Screened interval

Total well depth, feet

Gamma

Point resistivity

Gravel and sand

Silt and clay

Geological

Geophysical

The seven HSUs that form thehydrostratigraphic framework are thekey feature of 30 cross sections theteam developed to cover the site andsurrounding areas. The cross sectionsrepresent a working hypothesis of thesubsurface structure that continues togain definition as more field data aregathered and analyzed.

Together the HSUs containinformation on the geology of the site,the present location of contaminants andtheir migration pathways, the three-dimensional geometries of individualplumes, and the relationship betweenplumes and their sources at the surface.

These cross sections and maps areused to optimize the locations ofextraction and monitor wells, to ensureadequate hydraulic control of plumes,and to maximize contaminant removal.They also show the relationshipbetween contaminant source areas andVOC plumes within HSUs.

Controlling Migration

Figure 5 also shows that VOCs tendto migrate within the confines ofindividual HSUs. Transport betweenHSUs, however, may occur where thelow-permeability sediments separating

decades ago. Once the VOCs migratedinto the saturated zone, groundwatercarried them laterally westward withinHSU-5. Figure 5 shows how extractionwells have been placed to target VOCsin HSU-5 close to the source area aswell as near the higher concentrationson the western edge of the plume.

Figure 6 is a north–south crosssection along the western edge of the site,looking eastward. Similar to Figure 5,Figure 6 shows that VOCs originatingfrom sources to the east in this area arefound mainly in HSUs 1B and 2, withminor contamination in HSU-3A.Plumes in these three units are beingremediated by extraction wells locatedat two treatment facilities in theLaboratory’s southwest corner.

Making Steady Progress

Maps such as Figures 5 and 6 areallowing the Laboratory to minimize thenumber of wells needed for site cleanupand compliance monitoring, and in turnreduce expenditures for wells andpipelines. At Treatment Facility A alone,the number of extraction and monitorwells necessary for cleanup wasdecreased by about 20% by usinghydrostratigraphic analysis. Togetherwith accompanying pipelines and otherinfrastructure, that reduction translatesinto significant cost savings, on theorder of $500,000 per treatment facility.

HSU methodology also is allowingoverall cleanup to progress faster thanexpected because Laboratory staff now

Figure 4. This hydrostratigraphic

cross section shows geologic and

geophysical borehole data that

researchers gathered for correlating

subsurface hydrogeology.

Hydrostratigraphic unit (HSU)

interpretations were cross checked

and refined using hydraulic and

chemical data.

NORT

H

Patterson Pass Rd.

Gre

envi

lle R

d.

Vasc

o R

d.

East Ave.

Rhonewood SubdivisionArroyo Seco

Southwest area

Projection of intersection of water table and HSU boundary


HSU-1AHSU-1B

HSU-3BHSU-2

HSU-5

HSU-3A

HSU-7HSU-6

HSU-5HSU-4

HSU-1B

HSU-2

HSU-3B

HSU-1A

�

Saturated zone

Unsaturated zone

Water table

��

��

��

��

��

��

��

��

Location of Figure 1

Total VOCs above cleanup levels

Known hazardous waste releases

Suspected contaminated areas

Elevated vadose zone total VOC soil concentrations

100 to 999 ppb

10 to 99 ppb

1 to 9 ppb

Area where total VOCs are below cleanup levels

��

Isoconcentration contour (ppb)

Extraction well screened interval

Subsurface features

�

Surface features

Facing north

Figure 5. The relationship is shown between source areas

and VOC groundwater plumes within HSUs. Extraction wells

are positioned to target individual VOC plumes.

15



that the Laboratory’s groundwater cleanupefforts are being conducted in the best,most cost-effective manner possible.Stresses Blake, “Our ultimate goals areto accelerate the VOC plume cleanupprocess, demonstrate that the plumesno longer pose an environmental threat,and be the first federal environmentalsite in the U.S. to be removed from theSuperfund list.”

Key Words: hydrostratigraphic analysis,groundwater, groundwater restoration andremediation, hydrostratigraphic unit (HSU),pump-and-treat.

References1. R. G. Blake, M. P. Maley, and C. M.

Noyes, Hydrostratigraphic Analysis:The Key to Cost-Effective Ground WaterCleanup at LLNL, LLNL, Livermore,CA, UCRL-JC-120614 (1995).

2. W. F. Isherwood, C. H. Hall, and M. D.Dresen (Eds.), CERCLA FeasibilityStudy for the LLNL Livermore Site,LLNL, Livermore, CA, UCRL-AR-104040 (1990).

3. R. K. Thorpe, W. F. Isherwood, M. D.Dresen, and C. P. Webster-Scholten(Eds.), CERCLA Remedial InvestigationsReport for the LLNL Livermore Site,LLNL, Livermore, CA, UCAR-10299(1990).

4. M. D. Dresen, J. P. Ziagos, A. J. Boegel,and E. M. Nichols (Eds.), Remedial

Action Implementation Plan for theLLNL Livermore Site, LLNL, Livermore,CA, UCRL-AR-110532 (1993).

5. U.S. Department of Energy, Record ofDecision for the Lawrence LivermoreNational Laboratory, Livermore Site,LLNL, Livermore, CA, UCRL-AR-109105 (1992).

6. F. Hoffman, “Ground-WaterRemediation Using ‘Smart Pump and Treat,’” Ground Water 31 (1), pp. 98–106 (1993).

7. F. Hoffman and M. D. Dresen, “AMethod to Evaluate the VerticalDistribution of VOCs in Ground Waterin a Single Borehole,” Ground WaterMonitoring Review 10 (2) (1990).

14



public as well as with regulators. Incommunity and regulatory-agencymeetings, researchers have usedgraphics depicting hydrostratigraphicanalysis to clearly convey the extent ofthe Laboratory’s commitment to cleanup contaminated groundwater and toillustrate the success of the efforts todecrease VOC concentrations andthereby reduce any potential dangers tomunicipal water supplies.

In a time of shrinking federal budgets,a very important goal is to usehydrostratigraphic analysis to demonstrate

have even better information for placingthe extraction wells for maximum effect.Through smart pump-and-treat, PCEconcentrations in HSU-2 along thewestern margin of the site have beenreduced from over 1,000 parts per billionto less than 100 parts per billion overthe last seven years. The newly installedextraction wells and associated pipelinesthat were designed using the HSUmethodology have accelerated thecleanup and may allow cleanup objectivesto be reached in another 10 to 15 years

100

06000

Scale: feet

��

��

��

��

��

��

��

��

��

��

��

Unsaturated zoneWater table

Saturated zone

LLN

L

NORTH

Patte

rson

Pas

s R

d.

East

Ave

.

Vasco Rd.

Greenville Rd.


10

10

10

10

100

100

100 10

0

100

100

1000

1

1

1

1

1

1

HSU-1B

HSU-2

HSU-3AHSU 3B

HSU-4HSU-5

HSU-6

HSU-7

HSU-1ASurface features

Subsurface features

Active extraction well

Planned extraction well

Groundwater flow direction

Isoconcentration contour (ppb)

>999 ppb

100 to 999 ppb

10 to 99 ppb

1 to 9 ppb

Areas where total VOCs are below cleanup levels

Extraction well screened interval

Water table

��

Facing east

Figure 6. This block diagram shows both cross-

sectional and map views of HSU-2 VOC plumes that

are targeted for cleanup.

rather than the Laboratory’s originalestimate of 50 years.

The steady progress of the Lab’sgroundwater cleanup effort can be seenback in Figure 1, the sequence of fourmaps covering the time period October1988 to February 1995. Together themaps show a dramatic decrease in VOCconcentrations in HSU-2, illustrating thecapacity of hydrostratigraphic analysisto effectively monitor plume changesas remediation work proceeds.

The maps have proven to be aneffective communication tool with the

About the Scientist

For further information contact Richard G. Blake (510) 422-9910([email protected]); or Charles Noyes orMichael Maley (510) 422-8820.

RICHARD G. BLAKE joined the Laboratory in 1992 as anenvironmental scientist/hydrogeologist. He graduated with aB.S. (1977) and an M.S. (1983) in geology from CaliforniaState University, Los Angeles. He is currently theHydrogeology Team Leader in the Environmental RestorationDivision of the Environmental Protection Department. Blakealso spent twelve years in the California oil and gas explorationindustry: seven years for PG&E’s subsidiary Natural GasCorporation of California, and five years as Fleet Oil Co.’s vice

president. His publications include papers about California oil and gas exploration,stratigraphy and sedimentation, engineering geology and geologic hazards,hydrogeology, and environmental restoration.

17


ADIO detection and ranging(radar) was first developed in the

1920s. Most of us associate radar withcombat scenes in movies or an occasionalspeeding ticket. Conventional radar usesbeamed and reflected microwave energyto detect, locate, and track objects overdistances of many miles. Almost alltypes of radar were developed fordefense applications, and they continueto be used by the military and a fewcivilian organizations. Commercial usehas been limited primarily because mostradar systems are large, and they can becomplex and cost $40,000 or more. Adramatic change in radar use is imminent,resulting from work done at LLNL.

We have invented and patented a fundamentally different type ofcompact, low-power radar system calledmicropower impulse radar (MIR), whichis orders of magnitude less expensive toproduce than other conventional radars.Unlike conventional radar, which sendsout continuous waves in bursts, MIRuses very short electromagnetic pulsesand can detect objects at much shorterrange. The new technology has become

MicropowerImpulse Radar

R

16

A new pocket-size radar that operates

up to several years on AA batteries and

costs only a few dollars is stimulating

Laboratory research efforts and a

variety of industrial products. Its many

potential uses include security, rescue

operations, and health monitoring.


Figure 2. In an MIR motion sensor,

a transmitting antenna radiates a

pulse that is about 0.2 nanoseconds

long. Reflections from targets return

a complex series of echoes to the

receiving antenna. The return signal

is sampled at one range-gate time

by an impulse receiver containing

a voltage sampler along with an

averaging circuit and amplifier. The

detector listens at the appropriate

time for an echo. For an object about

3 m from the MIR, the sampled

gate at 20 nanoseconds after

transmission would just

capture it.

19


Micropower Impulse Radar18


Micropower Impulse Radar

beam pulsed Nova laser generatessubnanosecond events that must beaccurately recorded. In the late 1980s,Laboratory engineers began to developa new high-speed data acquisitionsystem to capture the data generated byNova and the next-generation lasersystem, the National Ignition Facility.The result was a single-shot transientdigitizer—a 1993 R&D 100 Awardwinner described in the April 1994issue of Energy and Technology Review.1

The LLNL transient digitizer, whichis the world’s fastest, functions as ahigh-speed oscilloscope combined witha digital-readout device. The instrumentrecords many samples from singleelectrical events (a brief signal called a “transient”), each lasting only 5 nanoseconds (5 billionths of a second).Compared to competitive products, suchas the best oscilloscopes, the transientdigitizer is much smaller and morerobust, consumes less power, and costsfar less.

While developing the transientdigitizer, project engineer McEwan hadan important insight. The samplingcircuits developed for it could form thebasis of a sensitive receiver for anextremely small, low-power radarsystem (Figure 1).

MIR Components

The principal MIR components areshown in Figure 2: a transmitter with apulse generator, a receiver with a pulsedetector, timing circuitry, a signalprocessor, and antennas. The MIRtransmitter emits rapid, wideband radarpulses at a nominal rate of 2 million persecond. This rate is randomizedintentionally by a noise circuit. Thecomponents making up the transmittercan send out shortened and sharpenedelectrical pulses with rise times asshort as 50 trillionths of a second (50 picoseconds). The receiver, whichuses a pulse-detector circuit, only acceptsechoes from objects within a preset

Figure 1. The micropower impulse radar

(MIR) proximity sensor board. distance (round-trip delay time)—from afew centimeters to many tens of meters.

The MIR antenna determines muchof the device’s operating characteristics.A single-wire monopole antenna only 4 cm long is used for standard MIRmotion sensors, but larger antennasystems can provide a longer range,greater directionality, and betterpenetration of some materials such aswater, ice, and mud. Currently, themaximum range in air for these low-power devices is about 50 m. With anomnidirectional antenna, MIR can lookfor echoes in an invisible radar bubbleof adjustable radius surrounding the unit(Figure 2). Directional antennas can aimpulses in a specific direction and addgain to the signals. We can separate thetransmitter and receiver antennas, forexample, to establish an electronic“trip-line” so that targets or intruderscrossing the line will trigger a warning.We are also exploring other geometrieswith multiple sensors and overlappingregions of coverage.

Behind MIR Technology

Impulse RadarConventional radar sends out short

bursts of single-frequency (narrow-band) electromagnetic energy in themicrowave frequency range. Otherradars step through multiple (wide-band) frequencies to obtain moreinformation about a scene. An impulse,or ultrawide-band, radar such as MIRsends individual pulses that containenergy over a very wide band offrequencies. The shorter the pulse, thewider the band, thereby generating evengreater information about reflectedobjects. Because the pulse is soshort, very little power isneeded to generate the signal.MIR is unique because itinexpensively generatesand detects very fast(subnanosecond) pulses.The drawback of usingshort, low-power

NoisePulse-repetition

interval

Impulse generator

Delay

Motion processor

Sensitivity control

Impulse receiver

Alarm

Transmitting antenna

Receiving antenna

Range control

Range gate

Target

0.2-nanosecond impulse

Radar echoes

Radar bubble

Wall

government sponsors are interested inlow-cost, lightweight MIR sensors inareas of defense, law enforcement,transportation infrastructure, and theenvironment. Envisioning perhapshundreds of other uses, Tom McEwan—the electrical engineer who inventedMIR—has compared the newtechnology to the Swiss Army knife.

The Genesis of MIR

MIR, with origins in LawrenceLivermore’s Laser ProgramsDirectorate, is now being developed bythat directorate’s Imaging and DetectionProgram. The Laboratory is home to the100-trillion-watt Nova laser. Developedfor nuclear fusion research, the ten-

LLNL’s fastest growing technology transfer activity

primarily because of its low costand extraordinary range of applications.

Among the scores of uses underinvestigation for MIR are new securityand border-surveillance systems;underground, through-wall, and oceanimaging; fluid-level sensing; automotivesafety, including collision-avoidanceand intelligent cruise-control systems;“smart” devices such as lights, heaters,and tools that automatically turn on oroff; and medical diagnostics.

The technology has potential use infinding earthquake survivors underrubble and in monitoring for sudden-infant-death syndrome. Various

21



after each transmitted pulse, called arange gate. If we choose a delay timeafter each transmitted pulse correspondingto a range in space, then we can openthe receiver “gate” after that delay andclose it an instant later. In this way, weavoid receiving unwanted signals.

The MIR receiver has a very fastsampler that measures only one delaytime or range gate per transmitted pulse,as shown in Figure 3a. In fact, we usecircuitry that is similar to the transmitimpulse generator for this range-gatedmeasurement, another unique featureof our device. Only those returnpulses within the small range gate—corresponding to a fixed distance fromdevice to target—are measured. The gatewidth (the sampling time) is always fixedbased on the length of the pulse; but thedelay time (the range) is adjustable, asis the detection sensitivity. Averagingthousands of pulses improves the signal-to-noise ratio for a single measurement;i.e., noise is reduced, which increasessensitivity. A selected threshold on theaveraged signal senses any motion andcan trigger a switch, such as an alarm.

Randomized Pulse RepetitionAs mentioned earlier, a noise source

is intentionally added to the timingcircuitry so that the amount of timebetween pulses varies randomly around2 MHz. There are three reasons forrandomizing the pulse repetition rateand averaging thousands of samples atthose random times. First, interferencefrom radio and TV station harmonicscan trigger false alarms; but withrandomizing, interference is effectivelyaveraged to zero. Second, multiple MIRunits can be activated in one vicinitywithout interfering with each other ifthe operation of each unit is randomlycoded and unique. Each unit creates a pattern recognizable only by theoriginating MIR. Third, randomizingspreads the sensor’s emission spectrumso the MIR signals resemble backgroundnoise, which is difficult for other sensors

to detect. Emissions from an MIR sensorare virtually undetectable with aconventional radio-frequency receiverand antenna only 3 m away. In otherwords, randomizing makes the MIRstealthy.

Equivalent-Time SamplingMore sophisticated MIR sensors,

such as our MIR Rangefinder, cyclethrough many range gates. As shownin Figure 3b, the delay time is swept, orvaried, slowly with each received pulse(about 40 sweeps per second) toeffectively fill in the detection bubblewith a continuous trace of radarinformation. In essence, we are takingsamples at different times, thus differentdistances, away from the device. Theresult is an “equivalent-time” record ofall return pulses that can be correlatedto object distance. The equivalent-time

echo pattern exactly matches theoriginal “real-time” pattern, except thatit occurs on a time scale slowed by 106.We can easily display the equivalent-timeecho pattern on an oscilloscope or readthe data into a computer. We are applyingthis sampling technique to many short-range applications, such as lightweightaltimeters or reservoir-level measurement,as well as all MIR imaging applications.

Forming ImagesWith equivalent-time sampling,

we can form images by moving theRangefinder in front of a target areaor by using a stationary array ofRangefinders. Figure 4a showsunprocessed radar information weobtained along a concrete floor in theNova facility. Each vertical trace is areturn signal from a different position

20



pulses is that less energy can bemeasured on the radar returns. Wesolved this problem by transmittingmany pulses rapidly and averaging allreturns.

The advantages of producing anddetecting very brief radar impulses areconsiderable:• The target echoes return muchinformation. With short pulses, thesystem operates across a wide band offrequencies, giving high resolution andaccuracy. The system is also lesssusceptible to interference from otherradars.• Battery current is drawn only duringthe short time the system is pulsed, sopower requirements are extremely low(microamperes). One type of MIR unitoperates for several years on two AAbatteries.• The microwave power associated withpulsed transmission is exceedingly low(averaging tens of microwatts) and ismedically safe. MIR emits less thanone-thousandth the power of a cellulartelephone.

Range-Gated RadarTransmitted energy from any radar

is diffracted and scattered by objects inthe field of view, such as cars, trees, orpeople. Larger and more conductiveobjects generally produce larger returns.Because the wavelength of MIR signalsin air is currently about 15 cm, we caneasily detect objects of that size orlarger at distances of about 15 cm orgreater. Distorted, low-amplitudereflections of the transmitted pulse arepicked up by the receiving antenna inthe time it takes for light to travel fromthe MIR to the object and back again.

The operating principle of MIRmotion sensors is based on therelatively straightforward principle ofrange gating. In looking for the returnsignals, MIR samples only those signalsoccurring in a narrow time window

Range gate

Fixed round-trip delay time

Transmitted MIR impulses (about 2 million/second)

Radar echoes received

500 nanoseconds

200 picoseconds

Distance

Time

25 milliseconds

Incr

easi

ng r

ange

Swept range delay

Time-equivalent record of returns

Figure 3. (a) Following

an impulse transmitted by

MIR, a range gate opens

briefly after a fixed delay

time to sample the

received radar echoes.

(b) To obtain a more

complete record of returns

for more sophisticated

applications, we sweep

the range delay over

various delay times to

obtain target information

at different distances. We

have effectively slowed

down the radar signal by

about a factor of 1 million

to get an “equivalent-time”

record of radar returns

that can be correlated to

object distances. (Pulses

pictured here are not to

scale.)

(a)

(b)

(a) Unprocessed data

(b) Reconstructed image

7.5 ns

33 cm

Figure 4. (a) Unprocessed radar information we obtained along a concrete floor in the Nova facility.

(b) After applying a specialized image-reconstruction algorithm to the unprocessed MIR data, buried

rebar and conduit shown in a cross section become clear.

Rebar

Conduit

23



Commercially Ready MIR

Table 1 lists some of the commercialapplications of MIR. One key factor invirtually all commercial markets for MIRis cost. Most of our sensor units can bemanufactured at a fraction of the costof existing technology—indeed, theyare typically hundreds of times lessexpensive. In many cases, there simplyis no practical alternative technologyon the market that is as robust, accurate,and inexpensive.

Security SystemsHome security systems now on the

market can cost thousands of dollars,require regular maintenance, and bedisrupted by interference from aneighbor’s system. At a projected costof $20, an MIR sensor (Figure 6),powered by AA batteries, operateswithout frequency channels or wiring

and is simple to install. Motion sensorscan be adjusted for sensitivity and rangeso that a pet, for example, would nottrigger an alarm and could roam freelyanywhere below a ceiling-mounted MIRsensor. Installations of MIR sensorsalready exist in the DOE nuclear weapons

22



system and its intended purpose, thefollowing features are common to mostunits:• Low cost, using off-the-shelfcomponents.• Very small size (circuit board is about4 cm2).• Excellent signal penetration throughmost low-conductivity materials, so it isable to “see through” walls, concrete, andother barriers, including human tissue.• A sharply defined and adjustable rangeof operation, which reduces false alarms.• Long battery life, typically severalyears, because of micropower operation.• Simultaneous operation of many unitswithout interference.• Randomized emissions, making thesensor difficult to detect.

Current MIR prototype units atLLNL are made with low-cost, discretecomponents. In the planning stages aresingle chips—application-specificintegrated circuits (ASICs)—that willreplace most of the discrete parts andresult in even lower cost and smaller size.

One limitation is that the penetrationof MIR signals through a materialdecreases as that material’s electricalconductivity increases. Thus, the

technology cannot see through thickmetal, such as a ship’s hull, or sea water,but it still can penetrate substances withmoderate electrical conductivity, suchas the human body.

MIR as a Sensor Technology

MIR technology opens up manypossible low-cost sensor systems formotion detection or proximity, distancemeasurement, microwave imageformation, or even communications. Forexample, in some cases it has advantagesover many kinds of conventionalproximity and motion sensors, such aspassive infrared (heat sensors), activebeam-interruption infrared, ultrasound,seismic, and microwave Doppler devices.Many of these sensors are adverselyaffected by temperature, weather, andother environmental conditions, makingthem prone to false alarms. Passiveinfrared sensors can be triggered by lightand heat, and their detection range is notwell defined. Even a thin sheet of paperblocks both infrared and ultrasoundsignals.Similarly, ultrasound motion andDoppler microwave sensors interferewith one another when several units areco-located. Without range gates, thesesensors can trigger as easily on distantobjects as on nearby insects. They canalso have limited material penetration,detectable emissions, and expensivecomponents. MIR technology providesan attractive alternative to these devices.

We are following two paths indeveloping and applying MIR technology.For well-developed products, weencourage commercial applications,and we are licensing the technology toqualified manufacturers in the U.S.using a procedure that ensures fairnessof opportunity. For ideas that requiremore research and systems development,we are continuing to explore electronics,antennas, signal processing, and imagingconcepts as we develop programs thatwill apply MIR technology to supportLaboratory missions and addressproblems of national interest.

Table 1. Some commercial applications of MIR.

CommercialSector Application of MIR

Automotive Parking assistance; backup warning; precollision detection;cruise control; airbag deployment; electronic dipstick for all fluid levels

Security Home intrusion and motion sensor; keyless locks, automatic doors;child monitoring; vehicle theft alarm; radar trip wire; perimeter surveillance

Appliances Stud finder; laser tape measure; wireless thermostat; automatic dispenser;automatic tool shutoff; toys, games, and virtual reality

Manufacturing Fluid-level, proximity, and harsh-environment sensing;robotic sensor; industrial automation

along the floor. When many individualvertical views into the floor are stackedside-by-side, resembling slices of breadmaking up a loaf, we can reconstruct across section of the floor. As expected,features are obscured by the clutterinherent in all radar measurements. Toresolve the locations of buried objects,such as rebar and conduit shown inFigure 4b, we apply a specializedimage-reconstruction algorithm usingdiffraction tomography.2

Many such slices stacked togetherform a full 3-D view of the subfloor orother concrete structure (Figure 5).This unique combination of the MIRsensors and imaging software isspurring new, low-cost nondestructiveinspection methods.

Summary of Features

As MIR technology has evolved, aunique combination of featuresresulted. Although certainspecifications—signal strength,operating range, and directionality—can vary depending on the type of

MIR Recognition and Awards• Thirty U.S. patent applications.

• Twelve industry licensees and many more expected.

• Popular Science, cover story March 1995 and Best of What’s New Award 1994.

• New Scientist, cover story August 1995.

• Electronic Design News, 100 Hottest Products of 1994.

• Intellectual Property Owners, Distinguished Inventor of 1994.

• Federal Laboratory Consortium Award for Excellence inTechnology Transfer 1995.

Figure 5. Imaging steel

in concrete with MIR.

(a) The internal elements

of a concrete slab before

pouring. (b) Reconstructed

3-D MIR image of the

elements embedded in

the finished, 30-cm-thick

concrete slab.

(a)

(b)

25



imaging, highway and bridge-deckinspection, and hand-held wall surveying.In defense and law enforcement, we areexploring MIR in scenarios such asborder control and surveillance systems,mine and ordnance imaging, the imagingof individuals behind walls, and proximityfuses. We also envision potential usesin environmental and medical research.

Many of these applications ofnational interest are government-sponsored, involving signal processing,computations, and communicationsexpertise along with hardwaredevelopment, and we draw onLaboratory experts in all those fields.Following are a few areas in which wehave already made substantial progress.

Border SurveillanceBorder and perimeter surveillance

pose serious technical problems for thenation and many industries and agencies,including drug enforcement, landmanagement, and military security.Among many other issues, visible devicessuch as antennas or cameras are often

targets for vandalism or attack. TheLaboratory is working with the U.S.Border Patrol to demonstrate anautomated, covert surveillance systemfor international borders as well as formilitary sites and police boundaries.

By combining an array of concealableMIR units with advanced, low-costcomputation and communicationtechnologies, we plan to deploy anautomated surveillance method. Wecan monitor a localized area or establishan unattended, electronic trip-line thatwould cover a few kilometers andeventually extend across perhapshundreds of kilometers. MIR modulesplaced up to 100 m apart would measurehuman movement—discriminatingbetween people and other sources ofmotion—and rapidly communicate anintrusion down the chain of modules tothe nearest base station. We now havesensor units in place at the BorderPatrol station in El Centro, California,at an International Atomic EnergyAgency facility, and at Sandia NationalLaboratories, Albuquerque.

Detecting MinesLandmine detection is a serious

military and humanitarian problem.One thousand people are killed ormaimed every week worldwide bymines left from previous wars. MIRcan detect both plastic and metallicland mines buried in most soils. Ourtechnology is attractive because itssmall size and low cost allow eitherhand-held or vehicle-mounted arraysand because images formed by an arrayaid in discriminating mines from groundclutter.3,4 Currently, a laptop computercan reconstruct an image in less than10 seconds, but much higher speeds arefeasible. Field tests at the Nevada TestSite show conclusively that the MIRsensor readily detects buried minesthrough 2-D imaging, but full 3-Dimaging (Figure 8) may be necessary tomore reliably discriminate between amine and other buried features, likerocks of similar size and shape. A lineararray of MIR modules mounted on thefront of a remote-controlled vehicle, oron a boom extending beyond the vehicle,

can detect antitankand antipersonnelmines. Even inareas of roughterrain or densefoliage, portablemine-detectionsystems operatingin the look-aheadmode are feasiblewith currenttechnology.

24



complex, DoD Special Forces, U.S.Border Patrol, and the intelligencecommunity. At Livermore, an MIRsecurity system is now being installedin the lobby of the Nova building.

Automotive SensorsMIR motion sensors placed on the

side of vehicles can alert drivers aboutother cars in blind spots, warn whenanother vehicle is too close, and activateside air bags. A sensor placed on therear bumper, for example, providesparking assistance or warns when a curbis very close. In one test, a sensor unitinstalled in a car’s taillight sectionfunctioned perfectly even when wesmeared mud over the taillight or placed

30 cm of ice in front of the sensor unit.One licensee is expected to equip carswith MIR proximity sensors by the1997 or 1998 model year. Otherautomotive uses include securitysystems, traffic flow sensors, distanceand speed indicators, and dipsticks(described below).

Tools to ManufacturingDo-it-yourself tools based on MIR

can locate wooden or steel studs in awall, steel within concrete (Figure 5),plumbing lines, or electrical wiring. We envision electronic tape measures,automatic thermostats, automaticdispensers, games, and toys thatincorporate the new MIR technology. Inmanufacturing, we are exploring roboticsensors, harsh-environment sensors, andindustrial automation equipment basedon MIR.

One application in particular, the“electronic dipstick,” has the potentialto revolutionize the way fluid levels aremeasured in virtually every industry.The electronic dipstick, a low-cost,solid-state sensor that has no movingparts, is impervious to wetting, corrosion,sludge, and condensation. The deviceshown in Figure 7 launches a signalalong a single metal wire, rather thanthrough air, and measures the transittime of reflected electromagnetic pulsesfrom the top of the dipstick down to aliquid surface. Our tests show that theelectronic dipstick can resolve fluid-level changes smaller than a millimeterand is accurate to within 0.1% of itsmaximum length. The dipstick candetect all fluid levels in a car, measureoil levels in supertankers, and remotelymonitor water levels in reservoirs,among many other uses.

Projects in the Works

Some of our ongoing projectsinclude specialized motion sensors,short-range altimeters, radar ocean Figure 7. The electronic dipstick is a metal wire connected by cable to an MIR electronic circuit.

As a highly accurate fluid-level sensor with no moving parts, this device has myriad applications

in manufacturing and is significantly lower in cost than laboratory equipment performing the

same task.

Figure 6. An MIR concealable

intrusion sensor detects

intruders at ranges up to 6 m.

Units can be mounted on the

ceiling, located behind objects,

or hidden in shelves, closets, or

drawers. The system detects

motion by repeatedly monitoring

the echo pattern to see if it

changes. A change signifies

that an intruder has penetrated

the invisible radar bubble.

27


Micropower Impulse Radar26



Inspecting the Infrastructure More than 40% of the 578,000

highway bridges in the U.S. havestructural deficiencies or are obsolete.Corrosion of steel reinforcing bars(rebar), hidden by concrete and asphaltlayers, leads to fracturing anddelamination, which can result infailure. Visualizing the details of manystructures such as bridge decks hasrequired destructive techniques, such ascoring.

We are developing MIR devices tonondestructively image bridges androadbeds, evaluate civil structures,inspect power poles, and locate buriedpipes. We received funding from the

Asphalt surface

Delamination

Corrosion products

Concrete deck

Front-mounted transceiver array

Survey wheel

Rebar

Figure 8. A typical plastic

antitank mine is shown (top)

before burial at the Nevada

Test Site. MIR technology

was used (bottom) to image

the mine at three depths, or

horizontal “slices.”

Federal Highway Administration tobuild a vehicle for highway and bridgedeck inspection. In that project, we havedesigned a prototype vehicle-mountedinspection system (Figure 9) thatacquires data at speeds approachingthe normal flow of traffic. Speed isimportant because a large portion ofinspection costs arise from trafficcontrols. We envision three modes ofinspection: a quick mode at the highestvehicle speeds for preliminaryassessments, a limited-depth mode forhigher-resolution data, and a detailedmode at slower speed to inspect theentire deck thickness (up to 40 cm).Deployment of the full system isscheduled for fall 1996.

Medical ApplicationsOur radar’s average emission level

is about a microwatt—about 3 orders of magnitude lower than mostinternational standards for continuoushuman exposure to microwaves. Thus,MIR is a medically harmless diagnostictool. In addition, the sensors we aretesting remotely measure human vitalsigns much like the medical tricorderenvisioned in Star Trek, withoutinterfering with computers, digitalwatches, FM radio, or television.

Our MIR heart monitor (Figure 10a)measures muscle contractions(responses of the heart) rather than theelectrical impulses (stimuli) measuredwith an electrocardiogram (EKG).Figure 10b shows the output waveformof a prototype heart monitor comparedto that obtained from a standard EKG.The MIR output is complex and rich in detailed information, and we areactively working with physicians tounderstand its significance.

As a medical monitor, a very smallMIR unit built into a single chip couldsubstitute for a stethoscope. The U.S.Army is interested in a portable devicethat could be worn inside clothing sothat a soldier’s vital signs can be relayedfrom the field to a medical command post.

Depth of 4 cm

Depth of 5 cm

Depth of 6 cm

An MIR-based breathing monitor(Figure 11) does not have to makecontact with a person’s body, and it can operate through a mattress, wall, or other barriers. The detection ofbreathing motion can be a valuableasset in hospitals and homes, couldguard against sudden-infant-deathsyndrome, and might be used by peoplewith breathing disorders such as sleepapnea, in which the affected individualoccasionally stops breathing.

We are exploring the use of MIR foradditional medical devices, includingspeech-sensing devices and a polygraphsensor. Devices for the blind could warnof obstacles and variations in terrainand help to train individuals in usingcanes. We are initiating clinical studiesto optimize medical radars for heart,respiration, and speech applications.The potential payoffs are enormousnot only in financial terms but also in benefits to society.

Rescue OperationsCameras, dogs, and acoustic

equipment tuned to signs of lifecurrently help rescuers to findsurvivors buried after an earthquake,avalanche, or other disaster. Soon,wall- and rubble-penetrating portableMIR devices could assist in search-and-rescue operations. We have testedunits that detect respiration andheartbeats at a range of about 3 m. Inthe midst of wreckage too unstable tosupport rescuers, miniature radardevices could be tossed into the debrisfrom a safe distance and signalpersonnel when physiological signs aredetected. We are working with the U.S.Army Corps of Engineers EarthquakePreparedness Center (San Francisco)and with the NASA Disaster ResponseTeam (Ames) to develop such devices.

Figure 9. Vehicle-mounted

radar imaging for bridge-deck

inspection. Arrays of MIR

modules mounted on the front

and rear allow the vehicle to

cover a 2-m-wide swath with

each pass. Radar images are

reconstructed and processed

at an on-board workstation.

The current maximum range forMIR is about 100 m using high-gainantennas. Our intent is to extend therange to about half a kilometer. Longerrange will require an improved signal-to-noise ratio. We are looking at higher-power systems and improved antennadesigns to extend the range and atbetter signal and image processing toreduce noise.

Key Words: electronic dipstick;micropower impulse radar (MIR); radarheart monitor; ultrawide-band radar, radarimaging, microwave sensors.

References1. T. E. McEwan, J. D. Kilkenny, and

G. Dallum, “World’s Fastest Solid-StateDigitizer,” Energy and TechnologyReview, UCRL-52000-94-4, pp. 1–6(April 1994).

2. J. E. Mast and E. M. Johansson, “Three-dimensional Ground-penetratingRadar Imaging using Multi-frequencyDiffraction Tomography,” SPIE Vol.2275: Advanced Microwave andMillimeter Wave Detectors, pp. 25–26(1994).

3. D. T. Gavel, J. E. Mast, J. Warhus, andS. G. Azevedo, “An Impulse RadarArray for Detecting Land Mines,”Proceedings of the AutonomousVehicles in Mine CountermeasuresSymposium, Monterey, California, April4–7, 1995, Section 6: 112–120 (1995).

4. S. G. Azevedo, D. T. Gavel, J. E. Mast,and J. P. Warhus, “Landmine Detectionand Imaging using Micropower ImpulseRadar (MIR),” Proceedings of theWorkshop on Anti-personnel MineDetection and Removal, July 1, 1995,Lausanne, Switzerland, pp. 48–51 (1995).

29



Looking to the Future

We continue to develop MIR for a variety of applications, and we are exploring ways to increase itsperformance in difficult situations.Even though the new radar technologyperforms very well, we still need toaddress issues such as reduced clutter,enhanced resolution and contrast,electromagnetic attenuation by differentmedia, multiple scattering, shadowing,dispersion, real-time operation, andfull 3-D imaging speed.

For some applications, we want toextend MIR from the centimeter-waveregion into the higher-frequency,millimeter-wave region. Higher-frequency MIR will provide betterresolution and give greater signaldirectionality with divergence of onlya few degrees. Higher frequency willalso mean that MIR could detectsmaller objects of 2 cm or smallerdiameter, such as concealed weaponsor bullets, and even tiny asteroidsapproaching a spacecraft. MIR usingmillimeter waves could replaceultrasound motion sensors, such asthose used for automatic door openers.

28



About the Researchers

–0.75

–0.50

–0.25

0

0.25

0.50

0.75

Res

pons

e, V

0 5 10 15 20 25 30 35 40Time, s

–2.0

–1.5

–1.0

–0.5

0

0.5

1.0

1.5

2.0

Res

pons

e, V

0 1 2 3 4 5Time, s

MIR

EKG

(b)

(a)

Figure 10. (a) An MIR

cardiac monitor.

(b) Its output (upper

trace) is distinctly different

from that obtained by a

conventional EKG (lower

trace). We are working

with physicians to

correlate the radar signals

with physiological

functions.

Figure 11. An MIR breathing monitor detects the respiratory cycle through a 10-cm-thick chairback. The higher-frequency waveforms are cardiac activity.

For further information contact Stephen Azevedo (510) 422-8538([email protected]).

For licensing and MIR partneringinformation contact (510) 422-6935([email protected]). Also see ourhomepage (http://www-lasers.llnl.gov/ lasers/idp/mir/mir.html).

STEPHEN AZEVEDO, currently the Program Group Leader ofthe Microradar Project in the Laser Programs Directorate, has abackground in digital signal and image processing. Concentrating inelectrical engineering, he received a B.S. (1977) from the Universityof California at Berkeley; an M.S. (1978) from Carnegie-MellonUniversity; and a Ph.D. (1991) from the University of California atDavis. Azevedo joined LLNL in 1979 and since has been a principalinvestigator in computed tomography research and radar remotesensing; he also has done work in signal processing, modal analysis,

x-ray inspection, nondestructive evaluation, and imaging. He is the author or coauthor ofover 40 publications on these subjects.

THOMAS E. MCEWAN has been a member of the Laser ProgramsDirectorate in the Imaging and Detection Program since joining theLaboratory in 1990. His accomplishments here include inventing themicropower impulse radar (MIR) and developing the world’s fastestsolid-state transient digitizer and a palm-size impulse generator. Hereceived his B.S. (1970) and M.S. (1971) degrees from the Universityof Illinois (Chicago Campus) in electrical engineering. From 1970 to1985, he was a design engineer at Nanofast Inc. From 1986 to 1989,McEwan led the design of high-speed microelectronics at NorthropCorporation, where he supported programs in radar jamming,electronic countermeasures, and computer-chip development. In

addition to the MIR recognitions listed on p. 23, he has six patents in widebandelectronics.

31


Livermore’s ALS researchers are seeking to understandon an atomic scale the structure of materials such as thin films,multilayers, diamond films, novel semiconductors, and epitaxialoverlays on single-crystal substrates, and then relate thatknowledge to the properties that would govern the materials’performance in a variety of applications.

Seeing the TheoryThe ALS experimentalists have been closely coupling

their atomic-level measurements to work done by Livermore’stheoretical scientists, who are forging new pathways in thecomputational study of novel materials and their interactions(see Energy & Technology Review, August–September 1994).Part of the structural information gathered by the ALS group,can be compared directly to elements of theoretical modelsthat groups such as Livermore’s H-Division have developed.Says Terminello, “What that comparison does is allow us tohave greater confidence in the theoretical models for use in

predicting the structure and behavior of new materials— so in thefuture we can achieve ‘materials by design.’”

One of the Livermore team’s principal tools to characterizeinterfaces is a specially developed electron spectrometer, basedon a design originally created at IBM in the early 1980s. Whilean electron energy analyzer is a fairly common piece of researchequipment, the device built for the ALS beamline permits moredetailed and streamlined analysis of atomic and electronicstructures. For example, it records how many electrons are beingemitted from interface atoms in the sample under examination,and it simultaneously preserves the electron trajectories as theyare emitted from the surface. Furthermore, the device collectsall the angular information at one shot. The bottom line is thatthe unit allows researchers to efficiently do the angle mappingrequired to get the complete picture of the physics governingthe emitted electrons.

The unit recorded its first photoemission in October 1994and has been in use ever since. The device’s angle-resolving

capabilities became operational in July 1995.Terminello predicts that the analyzer will developinto a “workhorse” in the coming months asthe team continues its nanoscale look at thefundamental properties of a variety of novelmaterials and observing the phenomena that aretaking place within them.

Key Words: synchrotron radiation, spectroscopy,Advanced Light Source, Beamline 8.0, materialinterfaces.

References1. J. A. Carlisle, et al., “Probing the Graphite Band

Structure with Resonant Soft X-ray Fluorescence,”Physical Review Letters 74 (February 13, 1995).

2. J. J. Jia, et al., “First Experimental Results from the IBM/TENN/TULANE/LLNL/LBL UndulatorBeamline at the Advanced Light Source,” Review of Scientific Instruments, 66 (February 1995).

3. J. A. Carlisle, et al., “Characterization of Buried Thin Films with Resonant Soft X-ray Fluorescence,”Applied Physics Letters 67 (July 3, 1995).

30


Research Highlights

AKING things smaller isn’t as easy as it looks! Using novel materials to manufacture atomic-scale

microelectronic, electro-optic, and other devices requires aclear understanding of the materials’ atomic, electronic, andbonding structures. Through studies they are conducting at theAdvanced Light Source (ALS), a $100-million synchrotron-radiation user facility, researchers from Lawrence LivermoreNational Laboratory are helping to provide that understanding.

Headed by Lou Terminello of the Chemistry and MaterialsScience Directorate, the Livermore group is part of a largeresearch team. It includes researchers from IBM, the Universityof Wisconsin, the University of Tennessee, Tulane University,and the Lawrence Berkeley National Laboratory, where thetwo-year-old ALS is housed. The team was formed to poolresources to build and operate a state-of-the-art, soft x-rayand vacuum ultraviolet beamline at the ALS. The 33-m-longbeamline—called Beamline 8.0—was designed especially toprovide the brightest and highest resolution photon flux inan energy range from 40 to 1,500 eV.

Using techniques such as photoemission, photoabsorption,soft x-ray fluorescence (SXF) spectroscopy, and photoelectronholography, the researchers conduct their experiments toessentially “capture” atomic-level images from the materialsample they are probing.

Enhanced SXFThe initial experiments demonstrated the unique capability

of the Advanced Light Source to enhance SXF spectroscopy inorder to probe systems that are difficult if not impossible usingother techniques. Last year, the Beamline 8.0 team became thefirst of the ALS users to publish its experimental results inrefereed journals, with papers in Physical Review Letters,1Review of Scientific Instruments,2 and Applied PhysicsLetters.3 The reports focused on results using SXFspectroscopy to reach deep into the interior of materials tolook at the atomic and electronic structures of graphite andtitanium oxide. Resulting images showed dispersive featuresin fluorescence spectra.

While the collaborators share a technical affinity in advancingnew characterization techniques, each affiliated organizationhas its own specific interest. For example, Tulane and the

University of Tennessee are chiefly focused on SXF to examinethe electronic structures of solids. On the other hand, IBM,which provided most of the funding for construction of theshared beamline, is interested in basic and applied researchon microelectronic materials and on characterizing materialinterfaces.

Robust InterfacesInterfaces are also a central thrust for the Livermore team’s

research. With funding from the Department of Energy’s Officeof Basic Energy Sciences, the team is studying the structure ofheterogeneous interfaces, an important aspect of microelectronicsor other advanced materials. Explains Terminello, “Becausemicroelectronic devices are getting smaller and smaller, aninterface is becoming a more important constituent of theoverall device. Let’s say you join dissimilar materials andwant to traverse the interface with electrons for your electronicdevice. The interface very definitely has an impact on how,and whether, the device performs.”

The team’s materials science work at the ALS has involvedmaking interfaces as well as characterizing them. In oneexperiment, the researchers grew oxides on silicon usingnitrous oxide. “We found we could get a robust interface usingthat growth method because we are essentially growing a thinnitride film right at the interface,” said Terminello. “That makesthe interface less susceptible to breakdown when high-voltagefields are applied to very small devices.”

M

Soft X-Ray Fluorescence

This Livermore team of researchers helped to buildand begin operation of Beamline 8.0, a soft x-ray andvacuum ultraviolet beamline, at Lawrence BerkeleyNational Laboratory’s Advanced Light Source.

Holograms

For further information contact Louis J. Terminello (510) 423-7956([email protected]).

These examples show the atom-imaging technique that the Livermore team isdeveloping with the ALS. Top row: photoelectron holograms. Middle row: real-spaceatomic images that were recovered from holograms above. Bottom row: 3-D renderingsmade from above images (center one is an ideal model of the atoms).

Probing withSynchrotron-Radiation-Based Spectroscopies

Atom reconstruction slices

3-D renderings

33

through sound and visuals to scientists and techniciansresident at the experiment. The recent explosive growth ofhigh-speed, wide-area computer networks like the Internet ismaking remote operation possible.

Working Toward Remote OperationSince 1991, scientists at Livermore have been developing

capabilities for remote operation of a tokamak. In a collaborationwith General Atomics in San Diego, the Livermore project

Remote Operation of Tokamak32


Operating a Tokamak from Across the Country

HYSICISTS often share the use of large, expensive,experimental facilities to study the actions and reactions of

minute particles. But demands upon existing facilities are high.Lawrence Livermore National Laboratory is pioneering thedevelopment of technology for remotely conducting magneticfusion experiments as a way to maximize the use of experimental

tokamak facilities.Fusion experts liveall over the world,so their ability toconduct experimentsfrom multiplelocations will enablemany new scientificcollaborations.

P

Research Highlights

Real-time data crosses thecountry within a tenth of asecond via the Internet.

Split-second flash of light frominside the tokamak at MIT

indicates a plasma has formedand the shot is successful.

Real-time video ofresearchers in the

control room at MIT

From this control room in Livermore, a team of LLNL and MITresearchers successfully tested the technology for controlling fusionexperiments from a distance via the Internet—in this case, from3,000 miles across the country. During the first full day of thedemonstration, 21 of 35 “shots” on the tokamak at MIT werecontrolled from computers in Livermore.


Greater demand also is being made for more efficient useof funds to construct and use such facilities. Like otherelectronic operations, remotely operating experiments willalso significantly cut the time and cost of travel to thesefacilities as well. Scientists at one laboratory could controlall phases of a physics experiment on a device at anotherlocation, while simultaneously conferring with colleagues atother laboratories and universities who are obtaining real-timedata from the experiments in process.

While remote control of machines is nothing new, remoteoperation of a complex, experimental magnetic fusion reactorfrom the other side of thecountry is a different matter.LLNL researchers arefinding that successrequires not only real-time access to controlsand diagnosticequipment but also direct access

Internet ties to ITER database

noitarepokamakotetomeR

Distributed computing and data analysis

35


electronic communications will bring substantially increasedeffectiveness to doing science and spur applications far beyondoperating a physics experiment—to environmental monitoring,engineering design, medical triage, and remote diagnosis.

Key Words: Magnetic fusion, remote operation, distributed computing.

countries. ITER may be operating as early as 2007, by whichtime the participants expect to implement full remote operation.

With both a domestic tokamak and ITER in mind,Livermore recently began work on a DOE-fundedDistributed, Collaboratory Experiment EnvironmentsProgram to develop remote operation capabilitiesfurther. This two-year project involves eightnational laboratories and universities working infour groups to develop testbeds for remote accessto various kinds of expensive, hard-to-duplicatephysics facilities—from an electron microscope toLawrence Berkeley National Laboratory’sAdvanced Light Source to a tokamak. By buildingthese testbeds and using them for real-worldexperiments, the groups are studying the technicaland interactive aspects of controlling apparatus,taking data, and interacting with colleagues overwide-area networks. The goal is more than anincremental change in today’s use of computersand local-area networks; rather, it is the introductionof a new realm.

Livermore is leading one of the four groups in acollaboration that includes Princeton Plasma PhysicsLaboratory, Oak Ridge National Laboratory, and GeneralAtomics. This group is working to integrate an industry-standard distributed computing environment with Internet-based audio-visual communications to enhance remotecollaborations on General Atomics’ DIII-D tokamak. Theproject demands real-time synchronization and exchange ofdata among multiple computer networks, as well as thepresentation of enough auditory and visual informationassociated with the control room environment so that remotestaff are fully integrated in operations.

The vision is for a scientist thousands of kilometers awayto get the same sense of presence and control as at theexperiment site. The end result of this project should bedistributed environments that provide location-independentaccess to instruments, data handling and analysis resources,and fellow collaborators. This merging of computers and

34


staff has developed software to remotely access General Atomics’DIII-D tokamak. Today, when experiments are in process,researchers in Livermore can control diagnostic equipment,operate data acquisition systems, and obtain and view results.They have access to all computer-based information at theDIII-D. They also have created integrated, network-based,high-performance computing and data storage facilities.However, the main controls for actual remote operation ofthe DIII-D are not on the network.

Unlike the DIII-D, the Alcator C-Mod tokamak atMassachusetts Institute of Technology’s Plasma FusionCenter, the newest tokamak in the U.S., was designed withnetwork-based control and data acquisition systems. Althoughthe Alcator C-Mod hardware was not designed specifically forremote operation, its systems are compatible with this option.Until recently, however, only a few instruments had beenoperated remotely.

In March 1995, scientists in Livermore conducted fusionexperiments on the Alcator C-Mod device in Cambridge,Massachusetts, in the first transcontinental operation of atokamak. The Livermore team and MIT researchers workedtogether to operate the tokamak using a part of the Internetcalled the Energy Sciences Network, or ESNet, which ismanaged by Livermore. The plasma shape, the particle fuelingsource, the radio frequency heating, and a reciprocating probewere all controlled in real time from Livermore over ESNet.

Scientists also exchanged a variety of data betweenLivermore and Cambridge: video images, experimental datafrom the diagnostic equipment inside the tokamak, and videoand audio communications between researchers at each end(see figure pp. 32–33). Data and signals crossed the country inabout 100 milliseconds. Multiple video cameras capturedimages of the control room in Cambridge, of the exterior andinterior of the tokamak, and of researchers in Livermore. Aflash of light inside the tokamak indicated each successfulpulse, or shot. These real-time visuals of the people andequipment at Cambridge helped bring the experiment to thescientists in Livermore.

The ability to control the tokamak’s systems fromLivermore did not eliminate the need for a local staff atCambridge. Engineers on site at the tokamak monitored allsystems to assure the safety of the equipment and local personnel.

This demonstration was the definitive test for controlling alarge, complex physics experiment from a remote location,and MIT and Livermore scientists learned much about thepossibilities for remote collaboration. They also learned thatwork remains to be done to make the remote researcher morea part of the experiment.

Bringing Remote Operation Fully On LineRemote operation is an integral part of the design of the

huge International Thermonuclear Experimental Reactor(ITER), a magnetic fusion collaboration among the EuropeanCommunity, Japan, the Russian Federation, and the U.S.ITER is planned to operate in steady state with controlledignition and steady burn. Design is under way at severalinternational sites, although a location for the reactor itselfhas not yet been determined. Current planning includes a network of control room facilities in each of the partners’

Remote Operation of Tokamak Remote Operation of Tokamak

How Magnetic Fusion Works

Magnetic fusion scientists use a tokamak (the word is a Russian acronym) to duplicate the Sun’s process of creatingenergy through fusion. The goal is to create a commerciallyviable energy source without contributing to global warmingor acid rain and without producing toxic wastes.

In the doughnut-shaped tokamak, powerful magnets areused to confine plasma—a highly ionized gas like the materialat the Sun’s surface. Enough energy must be used to heat theplasma to a temperature sufficient to produce ion velocitieshigh enough to react and fuse—at temperatures in the range of20 million to 100 million degrees kelvin. Current tokamakscreate plasma in bursts a few seconds long, but scientists areworking toward steady-state operation, which is moreadvantageous for power generation and easier on the materialsin the system.

Fusion energy has not found its way into our electricalsockets because confining and heating the plasma are verydifficult. However, recent experiments give every indicationthat ignition for controlled fusion power applications will beachieved in the next 10 to 15 years.

For further information contact Tom Casper (510) 422-0787([email protected]).

Tokamaks use magnetic fields fromvariously shaped magnets (orange, red)to contain a plasma (yellow) of hot gases.

36


Groundwater Cleanup withHydrostratigraphic Analysis

The Lawrence Livermore National Laboratory’sgroundwater cleanup program has made dramatic strides inremoving contaminants underneath the Laboratory and nearbyarea. Individual contaminant plumes are effectively targetedfor hydraulic capture and cleanup by a Livermore team ofresearchers. Their use of hydrostratigraphic analysis integrateschemical hydraulic, geologic, and geophysical data, whichresults in a three-dimensional model of the subsurface area.The resulting hydrostratigraphic framework developed atLivermore is proving to be a highly useful management tool toplan, budget, implement, and monitor the groundwater cleanupeffort. Researchers expect to transfer their knowledge to otherremediation sites. Dollar savings so far have been in optimallyplacing extraction wells, thereby maximizing contaminantremoval and saving time.■ Contact: Richard G. Blake (510) 422-9910 ([email protected]), or Charles Noyes or Michael Maley (510) 422-8820.


Invented and developed at Lawrence Livermore NationalLaboratory is an inexpensive and highly sensitive, low-power radar system that produces and samples extremelyshort pulses of energy at the rate of 2 million per second.Called micropower impulse radar (MIR), it can detectobjects at a greater variety of distances with greatersensitivity than conventional radar. Its origins in theLaboratory’s Laser Directorate stem from Nova’s transientdigitizer. The MIR’s extraordinary range of applicationsinclude security, search and rescue, life support,nondestructive evaluation, and transportation.■ Contact: Stephen Azevedo (510) 422-8538 ([email protected]). Forlicensing and MIR partnering information (510) 422-6935([email protected]) or our homepage (http://www-lasers.llnl.gov/lasers/idp/mir/mir.html).

Abstracts

This document was prepared as an account of work sponsored by an agency of the United States Government. Neither the United States Government nor the University of Californianor any of their employees makes any warranty, express or implied, or assumes any legal liability or responsibility for the accuracy, completeness, or usefulness of any information,apparatus, product, or process disclosed, or represents that its use would not infringe privately owned rights. Reference herein to any specific commercial product, process, or service bytrade name, trademark, manufacturer, or otherwise, does not necessarily constitute or imply its endorsement, recommendation, or favoring by the United States Government or the Universityof California. The views and opinions of authors expressed herein do not necessarily state or reflect those of the United States Government or the University of California and shall not beused for advertising or product endorsement purposes.

U.S. Government Printing Office: 1996/683-074-30008

january february 1996 - s&tr | september 2019 · 2019. 5. 29. · january/february 1996 this...

Documents