Top Banner
? MANUFACTURING IMPROVEMENTS IN THE PHOTOVOLTAIC MANUFACTURXNG TECHNOLOGY (PWT) PROJECT C. Edwin Witt, Richard L Mitchell, Holly Thomas, Maaha I. Symko National Renewable Energy Laboratory M.S. 3221,1617 Cole W., Golden, CO, 80601, USA Telephone: 303B84-6402; Fax: 303B84-6481 8svI Richsrd w, U.S. Depaament Of Energy, EE-131 afi F- CI 8 07 3 5 -- 1000 Xndepndence Avenue SW, Wash. DC, 20585, USA Doughs S. Ruby, Sandia National Laboratories Albuquerque, New Mdco 891854752, USA ABSTRAm The Photovoltaic Manufacturing T&ology Project (PVMaT) is a govunmenrlindustry research and development (R&D) partnership between the U.S. federal government (through the U.S. Department of Energy ma) and members of the U.S. PV industry. The goals of PVMaT are to help the U.S. PV industry improve module manufacturing processes and equipmen% accederate manufacturing cost reductions for PV modules, balanca-of-systems components, and integrated systems; inc- commercial product performance and dabfity; and enhance the investment oppoWties for substantial scale-ups of U.S.-based PV manufacturing plant capacities. The qptich fm PVMaT har been to &&a risk taking by industry as it qlm new manuhtwing options and ideas for improved PV mdes and other componenN advances system and product integratiq and develops new system d e dl of which will lead to overall I.gduced system M&cycle costs for diable PV end pmducts. The PVMaT Fhse 4A module manufacturing RgtD projects am just being completed and initial results fbr the work diaccbd primarily to module manufactum are reported in this paper. FouzSen (14) new Phase SA subcontracts have also just been awarded and planned R&D areas for the ten (10) bussed on module manufacture are described. Fially, government funding, subcontractor cost sharing, and a comparison of the relative efhrts by PV teshnology throughout the PVMaT project are presented. Keywords: Manufacturing and procesSing - 1: Module Manufacturing - 2: PV Module -3 1. INTRODUCTION The phorovoltaic Manufacturing Technology Project (PVMaT) is a governmenVdustry resear& and development (R&D) -hip between the U.S. Energy [DOE]) and members of the U.S. PV industry. Its goals are to help the U.S. PV industry improve module manufacturing prosesses and equipmen% acderab manufacturing cost reductions for PV modules, brhce-of-systems components, and integratd systems; increase c o m m d product performance and reliability; and enhance the investment opp~xtunities for substantial sde-ups of U.S.-based PV manufacturing plant capacities. Md government (through the U.S. Department of PVMaT was initiated in 1990. Its concept, general management, and procurement appm6hes were dcvelopad by a team consisting of repmeatatives fiom year projests are being carried out through c o s t - s h d awards resulting from competitive solioitation. Each proposal is evaluated by a panel of experts sekctcd from technology, manufacturing, business planning, and applications (including utilit~) areas. To date them have bttn five phm: Phase 1, iden- the industry's industry, W vmt, and national laboratorieS. Multi- problems on a company by company basis; Phase 2 (i two paas, A and B), supporthg research and development @it&) to resolve module manufacturing problems and reduce manufacturing oosts; Phase 3, supporting R&D to resolve generic issues in module manufacturing; Phase 4A, supporting R&D to facGtate productdriven manufacturing advances involving balanceof-system components as well as modules; and Phase SA, continuing efforts similar to those addressed Phase 4. This paper descrii activities primarily related to module manufacture and includes Phase 4A2 R&D asults and newly awarded Phase 5A subcontract plans. Phase 4Al and SA1 subcontracs, addassing systems and system components other than modules, are descni in a separate paper by H. Thomas, et. al., from the National Renewable Energy Laboratory. Phases 1,2, and 3 of PWaT have been describad in detail in earlier p a p [l],R]J3]. Finally, in this paper, we present some summary information on government hding, subcontractor cost sharing, and a comparison of the relative efforts by PV tachnology throughout the PVMaT project.
7

EE-131 afi CI 8 3

Dec 28, 2021

Download

Documents

dariahiddleston
Welcome message from author
This document is posted to help you gain knowledge. Please leave a comment to let me know what you think about it! Share it to your friends and learn new things together.
Transcript
Page 1: EE-131 afi CI 8 3

?

MANUFACTURING IMPROVEMENTS IN THE PHOTOVOLTAIC MANUFACTURXNG TECHNOLOGY (PWT) PROJECT

C. Edwin Witt, Richard L Mitchell, Holly Thomas, Maaha I. Symko National Renewable Energy Laboratory

M.S. 3221,1617 Cole W., Golden, CO, 80601, USA Telephone: 303B84-6402; Fax: 303B84-6481

8 s v I Richsrd w, U.S. Depaament Of Energy, EE-131 afi F- CI 8 07 3 5 --

1000 Xndepndence Avenue SW, Wash. DC, 20585, USA

Doughs S. Ruby, Sandia National Laboratories Albuquerque, New M d c o 891854752, USA

ABSTRAm The Photovoltaic Manufacturing T&ology Project (PVMaT) is a govunmenrlindustry research and development (R&D) partnership between the U.S. federal government (through the U.S. Department of Energy ma) and members of the U.S. PV industry. The goals of PVMaT are to help the U.S. PV industry improve module manufacturing processes and equipmen% accederate manufacturing cost reductions for PV modules, balanca-of-systems components, and integrated systems; inc- commercial product performance and dabfity; and enhance the investment oppoWties for substantial scale-ups of U.S.-based PV manufacturing plant capacities. The q p t i c h fm PVMaT har been to &&a risk taking by industry as it qlm new manuhtwing options and ideas for improved PV m d e s and other componenN advances system and product integratiq and develops new system d e dl of which will lead to overall I.gduced system M&cycle costs for diable PV end pmducts. The PVMaT Fhse 4A module manufacturing RgtD projects am just being completed and initial results fbr the work diaccbd primarily to module manufactum are reported in this paper. FouzSen (14) new Phase SA subcontracts have also just been awarded and planned R&D areas for the ten (10) bussed on module manufacture are described. Fially, government funding, subcontractor cost sharing, and a comparison of the relative efhrts by PV teshnology throughout the PVMaT project are presented. Keywords: Manufacturing and procesSing - 1: Module Manufacturing - 2: PV Module -3

1. INTRODUCTION

The phorovoltaic Manufacturing Technology Project (PVMaT) is a governmenVdustry resear& and development (R&D) -hip between the U.S.

Energy [DOE]) and members of the U.S. PV industry. Its goals are to help the U.S. PV industry improve module manufacturing prosesses and equipmen% acderab manufacturing cost reductions for PV modules, brhce-of-systems components, and integratd systems; increase c o m m d product performance and reliability; and enhance the investment opp~xtunities for substantial sde-ups of U.S.-based PV manufacturing plant capacities.

M d government (through the U.S. Department of

PVMaT was initiated in 1990. Its concept, general management, and procurement appm6hes were dcvelopad by a team consisting of repmeatatives fiom

year projests are being carried out through cost-shd awards resulting from competitive solioitation. Each proposal is evaluated by a panel of experts sekctcd from technology, manufacturing, business planning, and applications (including utilit~) areas. To date them have bttn five phm: Phase 1, iden- the industry's

industry, W v m t , and national laboratorieS. Multi-

problems on a company by company basis; Phase 2 (i two paas, A and B), supporthg research and development @it&) to resolve module manufacturing problems and reduce manufacturing oosts; Phase 3, supporting R&D to resolve generic issues in module manufacturing; Phase 4A, supporting R&D to facGtate productdriven manufacturing advances involving balanceof-system components as well as modules; and Phase SA, continuing efforts similar to those addressed Phase 4.

This paper descr i i activities primarily related to module manufacture and includes Phase 4A2 R&D asults and newly awarded Phase 5A subcontract plans. Phase 4Al and SA1 subcontracs, addassing systems and system components other than modules, are d e s c n i in a separate paper by H. Thomas, et. al., from the National Renewable Energy Laboratory. Phases 1,2, and 3 of PWaT have been describad in detail in earlier p a p [l],R]J3]. Finally, in this paper, we present some summary information on government hding, subcontractor cost sharing, and a comparison of the relative efforts by PV tachnology throughout the PVMaT project.

Page 2: EE-131 afi CI 8 3

DISCLAIMER

This report was prepared as an account of work sponsored by an agency of the United States Government. Neither the United States Government nor any agency thereof, nor any of their empioyees, makes any warranty, express or implied, or assumes any legal liability or responsibility for the accuracy, completeness, or use- fulness of any information, apparatus, product, or process disclosed, or represents that its use would not infringe privately owned rights. Reference hmin to any spe- cific commercial product, process, or %Nice by trade name, trademark, manufac- turer, or otherwise dots not necessarily constitute or imply its endorsement. recom- mendation, or favoring by the United States Government or any agency thereof. The views and opinions of authors expressed herein do not necessarily state or reflect those of the United States Government or any agency thereof.

Page 3: EE-131 afi CI 8 3

DISCLAIMER

Portions of this document may be illegible electronic image products. Images are produced from the best available original document.

Page 4: EE-131 afi CI 8 3

2. PHASE4A2

P W a T 4A2 R&D results are presentad in the following paragraphs.

AS3 Americas, Inc. is advancing its technology through manufacturing line improvements which consist of (1) a fuaher decrease in the EFG wafer thickness to 250 micromeiers in mass produstion and an increase in the n u m h of wafers produced from one crucible; (2) an

material to meet the demands for h ighc r -eWiy solar cdls, (3) an improved s o M Wrica?ion technology; (4) dtveloping a -on glass removal process that is

developing an in- interconnect, hination, and Rnbricotion mdod. R-t accomplishments include construction and demonstration of prototype equipment for removing phosphorous glass from diffused wafers whicb can d u o s acid consumption and fluoride emissions by 9S%, impbentation of the first phase of a Statistical h.ocess Control program and the development of a new module diode housing.

increase in the elecbonio quality of as-pwn EFG

en*nmentally spfe and d u d in cost; and (5)

Astrohwer, Inc. is addassing the development of a iow-cost manuficturing wity for Sicon-FilmTM solar cells and panels by using the continuous processing capnbility ofthe Silioon-FilmTM technology. The focus of their efirts has been (1) decrease the consumption of costly, highquality silicon, (2) eliminate wafer sawing steps, (3) develop a high-yield, continuous manufachving technology, and (4) increase solarcell size. he APZS S i c o n - F i m cell is in production and efficiencies exceeding 12% have beem measured on this 240 cmz d. Small-scale devices have dunonstratbd efficiencies of 16.6%. Prooessing advanses inchde the development of a continuous dithsion prooess and a continuous antireflection coating process-

IO- Thin Taohnologies h6. is b m the throughput of their metallization, a-Si deposition, lascr- scriiing, and welding prooesses, with the goal of ducing the ovedl module manuhchuing costs on the ITF production line by 68%. T h y haw improved laser rtgisCration and substrate throughput by 30% through the development of a new position d&e&or and alignment system and the instdiation of a new laser s c n i , increased throughput in the printer by 70% with an active screen alignment that allows a ten-micron reprodusl‘bility, identified a wakebased ink which withstands the subsequent processing temperatures, and improved the throughput of the ZnO deposition process step by 50%. Photovoltaics International, U C is establishing a b w cost manuhcturing capability for linear concentrator modules by using their continuous processing qmbiity. Under this efht to manuh&m, they will take advantage of the continuous processing capability of their lens and sidepanel extrusion technology. They have completed the tooling and testing of the new 2OX

lens wttrusion ptocess equipmen& compkccd the design and initial hbrication of the &ver assembly Station, completed design and prototyping of the toll formed W e prosess, and completed development of the self j i i low V, linear concentrator module.

Simens Solar Industries is addressing improvements in their Czdra l& (Cz) silicon module manuhcturing technology to reduce module cod per watt by 18%. These gods are being addressed by identifying b a t i v e Cz module designs, material sources, and ptocesses that lower module component costs, and by improving manuhcturing ptocess yields, raducing labor

productivity. n e y haw implemented production of the lSO-mm cell and module product line, whicb leverages the use of silicon by over 30% in the pmduction of Cz solar cells; continued to improve manufacturing productivity and yield by over 10%; implemented Statistical Prooess Control in their diffusion and cell printing lines to improve capability and ddcal yields; and implemented polrsiIiQon preparation techniques to mitigate silicon supply variation on yields.

costs in cz module manufacturing, and increasing

The reader is refbrd to several other papers being p r e s e n t e d at this conference for more detail on the individual company accomplishments in PVMaT.

3. PHASESA

Subcontracts for the PVMaT SA competition were awarded d y in 1998 with the last of the 14 a 4 being placed in June. The 14 subcontracts arc wtpectad to total about $6OM over a three year period with a 48% subcontmctor wst-shiuing. ( --one o b r s responded to the PVMaT SA RFP. They proposed activities totding over $97 million. Of that amount, approximately 49% was o f k d as subcontractors’ cost- share.)

Table 1 presents the SA winners and their resp66tive areas of manuhcturing R&D. Subcontractors addressing primarily their own ptocess-spasifac module manuhchuing problems are designated SA2. Subcontractors addressing primarily generic R&D of inwest to broad portions of the industry or system and other non-module component manuhcturing problems am designated SAL

TABLE 1. PVMaT PEASE SA hfANUFACTURING AREAS

SA1 W o n Manufactwe of the Technology, Inc. ~dvaaced Sunsinem 325

5A2 ASE Americas Inc. The EM; High Volume PV

SA2 AstroPower Silicon-FilmTM Solar Cclls by a Flexible Manufacturing SY - AC Module

Man&turing Lm

Page 5: EE-131 afi CI 8 3

5Al

5A2

SA2

5A2

SA1

SA1

Production of Solar Grade Silicon by Refining of Liquid Metallurgical Grade Silicon Efficiency and Throughput Advrnces in Continuous Roll-t+RoU a-Si AUoy PV Manufacturing Technology Continuous, Automated Maaufa&~ring of String Ribbon Si PV Moddts Throughput Improvtmtots for Thin-Film based CIGS Modults Manufacturing and System InrpFovemeato for one and Two k W Invatera A d v d Powqua& ManuhcturiW

Crystal Systems, InC.

EnergyConvcrsion Devices, Inc.

EVagrtenSOlar, Inc.

GlobalSolar Energy, L.L.C.

<)moionPowcr Ensiwering Corporation PoWtrLight Comration

5A2 Solar cells, Inc.

5A2 Solarex

R&D on CdTe Product Manufacturing I Implpvemtntsin Polycrystalline Silicon PV

P O S t - M W M~ufa~turing Process Automation for Photovoltaic

I Dcvclopment of a M y - Integrated PV System for

Sitmens Solar, Inc. RBtD on Siemens Cz Sicon ~ ~ C t M a n O o h e t u r i n g I

ASE Americlls Inc. will fixus their ttsearoh on improvements in prosess integmtion, Satistical procesS Control implementation, data systems implementation, and IS09000 and ISO14OOO implementation. The improvements will be dire&& at a reducing yield lossts in areas of el&d and mechanical performance and reducing chemical waste. ASE will also develop processes which can be scaled to high volumes in the

productivity, as well as the production of solar oells from much thinner wafers. This efirt includes work in

w a h s and R&D to ensue a stronger EFG wafer and improve cell processing to achieve 15% solar cell C W i .

growth of thin EFG cylinders with improved

Cutting tachnOlOgY b in6aaSe @ for cutting

Astrohwer, Inc. is working on improvements for a flexiile muhkuing system for their Sin-Film'" Solar Cells. Their mearch during this b y e a r ef i r t will foous on several areas of the Sicon-Filmm production line, including the development of a continuous, in-line prooessing technology for largagna ( 3 k m wide by 1.2-m long) planks with higher throughputs, r e d u d material 00~5, and increased safety. As paxt of this subcontract, AstroFower will develop a continuous, high-speed, largxnx contact meEallization processes for largwma Sicon-Filmm. This will significantly increase their throughpub improve front contact line width and contact tesishurce, and raduoe Oost. Achieving the wtpected improvements in the Silioon-Filmm sheet growth p~ooesses would result in hrgewea (900 cm2), l2%cfficient (10.8W) solar &. Crystal Systems, Inc. will bus on the production of solar grade silicon by refinement of liquid MG silicon at low cost, in abundant supply to meet the PV industry's rapidly expanding needs. The appmch proposed is based on t h - d e m i d re techniques and the

of this work shall be equipment and processes adqua& for use in an MG silicon production plant utilidng 500 kg cbarges as well as demonstration of the removal of impurities, including boron and phosphorous, to less

Heat Exchanger Method (HEM) %. The ultimate b u s

than 1 ppma after refinement at the 500 kg cluuge kvd. The development and imphentation of this upgraded solar grade silicon Eaadstook is expeckd to result in significant labor wst savings and increased throughput, with the production cost goal of solar grade silicon to be less than $mkg.

Energy Conversion Devices, In0 (ECD) will perform manuhcturing R&D to: develop a new substrate heating system and tunperaaue sensor system designed to achieve more accurate temperature controlling and monitoring for the production of highcffickncy solar &; develop a set of in-line real-time mated quality monitoring systems for production machines; demonstrate the feasibility of using Zn m d targets in the DC sputtering process to pnpare 2110 layers for

internal hardware for aSi intrinsic layer @layer) deposition chambers.

high WfOnnanSe ba&-d&n; and redesign the

hrergreen Solar, Inc., will bus on improving their string nibon crystal p w t h process through reductions in labor and materid costs, and capital costs of addjtional futrrses, through increased automation as well as increased efficiency. T h y will also focus on developing high-throughput automated Qtll and moduk manuhcturing line processes such as: a continuous glass etch; high-speed drying and application of decals; automation in diffusion, glass etch, high-speed drying, printing and application of decals; and molecular hydrogen passivation. Additionally, Evergreen will increase their manuhcturing automation and throughput as well as decrease their labor costs through the deploymeat of a patterned backskin, an in-line tester, and an automated shinging operation.

Global Solar Energy, L.L.C. focus is performing R&D on the following: refining an all-laser, multipleAcam, high-@ scribing method for all CIGS PV layers; developing a process for inl-jet printing as a replacanent to screen printing for insulating s c n i areas, and integration of this process, along with the laser scniing methods, into the manuhcturing production lim, developing and integrating a high-rate

Page 6: EE-131 afi CI 8 3

?

GIGS deposition prootss for a moving f l a i l e substrate onto their manufachving line, designing, assembling, and opthking a Parallel Detector Spectroscopic EllipsomezU (PDSE) for transfer onto the thin-film CIGS production line; and developing an alternative back contact marial that is compatiile with CIGS prooessing on f l a i l e substrate.

Siemens SoIar Industries will focus their rtstaroh on the development and integmtion of ne.w opthized cell fabrication processes into their manuhtuing line for the production of 17% &bt, l S g n cells. Additionally, they will develop hge-uca cell proddon capabiity for 2OOmtn diameter, 4.5-watt prdotrpe solar cells and low-cost prototype modules. S i Environmental, SIlfetY and Health activities willbeditaotbdtowad reducingtheirhazardous~ by over 50% through re~ycling and muse of slurry materials in their wire saw process and an over 70% reduction in caustic waste.

Solar Cells, Inc. (SCI) is developing, design& and implementing an improved M A lamination process, an impved potting proQedure, and an improved s c n i i technique for the SCI CdTe module production line. They will also complete qualification testing of their fr9mekss, 6oom x l2Ocm module with pigtails as well as o t h advanced module designs such as: 1) investigating a junction box instead of pigtails; 2) evaluating module sizes other than 6oom x l2Ocm; 3) developing alternative module voltage configurations; 4) and developing alternative cmapulation marials or processes. Additionally, SCI will wrk toward refining and improving their environmeatrl, safety, and health prognuns throughout their hcjlitia and initiating activities dated to obtaining ISO14OOO oertificaton.

Solatwt, a Busintss: Unit of AMOCO/ENRON Solar will focus on s e m d parallel e&&. The first area involves Solarex working with a 10- subcontractor to develop a prooess to produce silicon fbadstooL h m COmmWhl p& H2siF6. Ah, Solarex will refine produaion line p m s areas Tor impmved product and materials handling to increase production line yield and reduce labor costs. And they will improve process measulltment and control in their production line and reduce rework through the implementation of an improved information system. Solarex will also malo3 improvemats in their casting and wafer sawing processes while developing mslluhloturing techniques which arc emenhmtally more benign. In another area of h u s , Solatwt will to demonstrate and implement a oost-effactive, d u s t cell prooess that produce a minimum avvage cell efficiency of 15% and improves their Oell line eleslricrl yield. F i y , they will develop and qualify an encapsulation system that meets their technisal and reliability requirements and can be laminated and cured more quickly in the present S o k laminatots, ' hbk 3 shows the dative cost sharing by phase for the PVMaT project to date. The Subcontracts awarded in

Spire Corporation research will focus on automation of photovoltaic module assembly. They will develop a series of automated, flexible systems, including (1) an intcgmted module edge processing system, combining automated edge trimraing, edge sealing, and ftaming prooesses, (a) an automated junction box instahtion system, (3) a final module test system combining high voltage isolation testing and perhrmance testing in a

buikstoragesystem. Solarselllamhabsandotha module m e prov ided by the industry will be used by spite to evaluate the automated procmses developed in the pgram. The system will be capabk of assembling modules made with d k CrySEalline silicon solar cells or thin film solar Oen Isminate.

SPI-SUN SMULATORTM, and (4) an automated

4. PVMAT - TECHNOLOGIES AND RESOURCES TO DATE

'hbk 2 ShsWS, by techlogy area, the total fuading for PVMaT for Phases 2 through 4 (the problem solution phases). The more ma- crystalline silicon caohnologies have the largest share of the funding. The amorphous silicon area is second in size and probably benefitted from the large U.S.A. Festptcb h d i n g of the 1980s in its apparent initial lead in thin film ef€orts. But we are now seeing copper indium diselenide (CIS) join cadmium telluride as a promising candidate for manufacturing improvements leading to viable products. See global Solar En- in saction 3. above Tot a description of the first CIS subcontract in the problem solving stage of PVMaT.

TABLE 2 PVMaT FUNDING BY TECHNOLOGY (through Phase 4)

I I

Cadmium Telluride 8206 13

I I 6162 I 9 1 co-nmofs

I I Crystalline Silicon 26588 41

3 and 4A1 are generally of the more generic or system component and integration type with less emphasis on

Page 7: EE-131 afi CI 8 3

process-spaoifio module manufacture. thest awards wen constrained to less

In addition, effort, both in

murots and time, The o-rs in these areas have also been smaller companies, generally maetiag the "small business" criterk ibr PVMaT paIticiion. ConsequentIy, they have not been required to cost share as heavily and the data show they have not.

T-3. PVMaTFuNDINGANDcosTSHARE (BY

solicitation /I PVUPT 2A - 21316

procuU, Specific

PV mant&mring prooesses, CechnOlogY, and cast reduction have been substantially advanced by the US PV industry. PV manuhcturing improvements continue to be seen in both the mort mature crystalline silicon approaches and the newer thin film technologits. FinaIly, based on the response to the the most recent P W a T solicitation, Phase SA, industry

q u i r e conhued or new RBtD and the responding companies find it ustfulto sharewith the govtrnrntnl program in supporting these R&D efforts.

has identified many areas of PV manufawring which

I PVUPT2B- I 13384 I 14557 I 5 2 1 Pmccss specific PVMaT 3 - G t n U i C

PWaT 4A1- product Driven BOS cud Systems P m T 4A2 - P d u c t Driven Module Manufacturing TOTALS

5. CONCLUSIONS

6. ACKNOWLEDGMENTS

This work is supaoagd under DOE contract n u m h AC36-83CH10093 with NREL, a national laboratory managed by Midwest Research InstiUte. Many people have contributed to the development and implementation of the Photovoltaic Manuficturing Technology project and to the R&D effo* carried out in this program. The authors thank each of them and reQognizethispaper~~tstheirworlt.

REFERENCES

[l] C.E.

@] C.E.

(31 R.L.

Witt, L.O. Herwig, R. Mitchell, and G.D. Money, "status of the Photovoltaic Manuhcturing Technology (FVMaTj Project," A.ocescEings of th8 22d lEEE

Vegas, Nevada, oocober, 1991. Wilt, L.O. Herwig, R.L. Mitchell, H.P. Thomas, R. sellus, and D.S. Ruby, " R m t Progrrss in the Photovoltaic Man&-

M O W h b @8Cw &@lW8, h S

T h l O g Y Prow (3pvMaT)," -8-S

O f th8 WOYbd cO@Y8RCU OR PhOtOWbkS, Waikoloa, Hawaii, Dcoember, 1994. Mitchell, C.E. Witt, H.P. Thomas, D.S. Ruby, R. King, and C.C. Aldrich, "Progress Update on the U.S. Photovoltaio manuh6h.uign Technology Project"

@8cW& Co@~mce, Anaheim, CA., October, 1997.

h0ce-s of th8 26di lEEE P h o t o W h b

Sandia is a multiprogram laboratory operated by Sandia Corporation, a Lockheed Martin Company, for the United States Department of Energy under contract DE-AC04-94AL85000.