16th Project Integration Meeting Handout PIM... · 5101-157 Department of Energy Low-Cost Solar Array Project 16th Project Integration Meeting Handout Pasadena Center September 24-25,

5101-157

Department of Energy

Low-Cost Solar Array Project

16th Project Integration Meeting

Handout

Pasadena Center September 24-25, 1980

Jet Propulsion Laboratory

California Institute of Technology

Pasadena. Cal ifornia

Pre pared by the .let Propulsion Laboratory, Ca li forn ia In st itute of Ted1 nology. for the Department of Energy through an agree me nt with the Nat io nal

Aero nautics and Space Ad minis tration.

T he JPL Low-Cost So lar Array Projec t is spomored by the Departmen t of Fnergy (DOE) and fo rms part of the Solar Photovoltaic Conversion Progra m lo initiate a major effort toward the development or low-cost solar arrays.

This report was prepared as an account of work spomorcd by the United States Government. Neither the Unitt:d States nor the United States Department of Energy, nor any of then employees, nor any of their cont ractors. subcontractors, or their employees. makes any warrant}. c:--press or implied. or assumes any kgal liability or responsibilit} for tht: accuracy, completeness or usefulness of any information, apparatu s. product or process d1 scloscd , o r represents that it s use would no t in fri nge privately o wned rights.

REMINDER :

Please bring this Handout

with you to the PIM !!

5101-157

Department of Energy

Low-Cost Solar Array Project

16th Project Integration Meeting

Handout

Pasadena Center September 24-25, 1980


California Institute of Technology

Pasadena, California

I TECHNOLOGY

DEVELOPMENT AREA

K.M. KOLIWAD, MGR M. LEIPOLD, DEP.MSR.

SILICON MATERIAL TASK -

R. LUTWACK, MGR

LARGE AREA Si SHEET TASK -

J.K. LIU, MGR

ENCAPSULATION

-- TASK

C. COULBERT, MGR

ADVANCED PHOTOVOLT AICS - TASK R,J, STIRN, MGR

lf-j £--1

LOW-COST SOLAR ARRAY PROJECT

STAFF

E. CHRISTENSEN C. T. HANSEN

I PRODUCTION

PROCESS AND EQUIP. AREA

D, B. BICKLER, MGR

- PROCESS AUTOMATION

- NEWLY DEVELOPED PROCESSES

- TECHNOLOGY TRANSFER

- ASSESSMENT OF EMERGING TECHNOLOGY

- NEAR-TER¥ C03T REDUCTION

rr-· .. -. 12. ]I

PROJECT MANAGER W. T. CALLAGHAN

DEPUTY MANAGER R.R. McDONALD

SECY: M,J, PHILLIPS

ANALYSIS AND INTEGRATION

AREA

P.K. HENRY, MGR

- PROJECT INTEGRATION

- ARRAY TECHNOLOGY C03TS

._ ECONOMICS/ INDUSTRIALIZATION

"- ARRAY LIFE CYCLE ANALYSIS

FINANCIAL B. S. LENCK, MGR

PROCUREMENT P. S. RYKEN

QUALITY ASSURANCE K.J. ANHALT

I ENGINEERING

AREA

R.G. ROSS, MGR

- ARRAY ENGINEERING

- ADVANCED ARRAY REQUIREMENTS

-ARRAY DEVELOPMENT

~ ENVIR01'1MENTAL TESTING R&D

- ENVIRONMENTAL ASSESSMENT

-

I OPERATIONS

AREA

L.N. DUMAS, MGR

LARGE SCALE PRODUCTION

TASK L.D. RUNKLE

- FAILURE ANALYSIS AN;:> REPORTING

- ENVIR/FIELD TESTS

i,.... PERFORMANCE MEASUREMENTS

.__ MODULE INTERFACE AND CONTROL

CONTENTS

MEETING OBJECTIVES------------------------------------------------- 1

MAP: Meeting Locations-------------------------------------------- 2

PIM AGENDAS-------------------------------------------------------- 3

WORKSHOP AGENDA---------------------------------------------------- 10

ADVANCED PHOTOVOLTAICS TASK---------------------------------------- 11

STATUS OF TECHNOLOGY TRANSFER-------------------------------------- 13

TECHNICAL SUMMARIES

Silicon Material--------------------------------------------- 21

Large-Area Silicon Sheet------------------------------------- 31

Encapsulation------------------------------------------------ 47

Production Process and Equipment----------------------------- 59

Engineering---·---------------------------------------------- 77

Operations--------------------------------------------------- 87

Large-Scale Production--------------------------------------- 89

LSA PROJECT ACTIVE CONTRACTS--------------------------------------- 97

LSA PROJECT PUBLISHED DOCUMENTS------------------------------------ 101

MAPS: Pasadena Area; Pasadena Center---------------------------

For Your Information

Inside Back

Cover

Check-in: Please check in at the registration desk on the lower level of the conference building before the start of the meeting on Wednesday morning.

Telephone Messages: Incoming calls will be received at JPL on (213). 577-9520. Constant coverage of this phone will be provided and messages transmitted.

Badges: We will appreciate your returning your badges at the end of the meeting.

iii

w

F"""?-1

MEETING OBJECTIVES

The LSA Project will convene its 16th Project Integration Meeting on September 24 and 25, 1980, at the Pasadena Center, Pasadena, California. Registration begins at 7:30 a.m. on September 24 on the lower level of the Center.

The PIM meetings are conducted to continue the exchange of information necessary for the assessment of progress in the Project.

The objectives of the 16th PIM are the review of silicon-ribbon and HEM casting; module durability and life testing, and the status of module and array technology. An assessment of the status of cell, module and array technology will be presented using estimated prices of modules and arrays if produced in large quantities in factories using today's technology.

Rather than breaking into the technology sessions, the plenary session will extend through Wednesday afternoon with panel discussions on a variety of subjects, such as the Florida and Phoenix photovoltaic homes and the status of CdS cells and modules. The technology sessions will begin on Thursday morning.

We are sure that you will find the themes for this PIM timely and interesting.

A map, facing the agenda, shows the room numbers for the sessions you are interested in attending.

1

MEETING LOCATIONS

C 314 C 304 C 306

C 305 0307

~ I i

-~

CONFERENCE BUILDING PLAZA LEVEL

• iii &ii iii'i

C 101 C 102 C 103 C 104 C 105

ill R:I IU! lb

IJI D EXHIBIT AREA 0 I[ EXIT ~~~ a t1

C 124 C 112

LITTLE TtEAmE

[=E:J -1: FREIGHT DOOR

~ '

' .

LCMDING DOCK

CONFERENCE BUILDING LOWER LEVEL

2

MAYCoc.l(...

0 '8 / J3L>.:>Gcii.

I tf 1?$'t.io'- Fa.e:s. p-,<>=1 C,.,. WEDNESDAY : Sept ember 24 , 1980 t ~ 100 1'10(.

a1~ll1Dc.1-1c°:~s="''"'- 7:30 VRegi s t ration

LJi·~.:,;ir1mt~a.y e.,,.,._,,,._.,5 '""'.,. 1)-r,.__ 8: 30 .. Welcome and PIM Overview M,u.~T -PAY IUW.6 f=

8. 40 f\l G<AJ , THA'"' ~i<NU-<>D • ., Announcement s

roa.. !!>AU.'-'P rJ • re

AGENDA

Lowe r Lobby of Pasadena Center

W. Callaghan

Maycock/Magid/Forney

Liu

12:05 Discussion

} '101

C. Coulbert

P. Henry

W. Callaghan

10 min

20 min

55 min

20 min

55 min

55 min

10 min

!Zez_,44 1'1 T'7" ?e;n.PO.rul't /}-?I,: C

,i>l)J!. .<} 7'],tJ,./

FA , '- ..;rl\,c i'-JPfr~T /?.4>-l Do ,v\.

o S1 t.1Co..:> RFP l-11,>'\R. 12:15 Lunch

Panel :

(Copies of morning viewgra phs should be posted ) Cl04 D~R.A.-::..4.< •0N o Coe@. t.1'-!=0'LT .- S.C)l,(IIJI)(.

C!>Y<c"T"ALL•,-.J e; s;+=t l : 30

2:20 Panel :

3 : 10 Coffee

3:40

~ otovoltaic Houses Florida House (Florida Sola r Cent er ) Phoenix Photovolta i c System ( ARCO) Phoeni x Utility (Salt River Project) Phoenix Cont ractor (John Long Homes) Other houses (MIT/LL)

CdS Cell/Module Progress and Prognosis Photbn Power SES Univ . of Delaware - !EC SER I

Pane±:, ,_..Photovoltaic Market Strategi es Unl imited·· ~M,u,.'1 &,:,;A.CH Motorola Vti.V1'i!?w.£- ,c/MP--7 h&,0y Solar Power llss1st ,.., TJ<a J:b:va.o,:, tJ> f'V /Ji Ml T /El k3/D, · ·/r,777V()e Sr.,,.A..<y

Indus try ' s Perspective of 4:30 Panel : .. and Role In Meeting ODE ' S PV Goals ARCO Sol ar Photowatt Solarex Spire

5: 20 Social Hour

THURSDAY : September 25 , 1980

8:00 Technology Sessions (Simultaneous) Silicon Material Large Area Silicon Sleet Encapsulation Production Process and Equipment Engineeri ng/Operati ons

9 :30-11 :00 Coffee available

12 :15 Lunch

1: 30 Cont inuation of Technol ogy Sessions Silicon Ma terial Engineeringl (beratiPDS ProJec t ilnal ysis and Integrati on

Discussi ons at Poster Areas

2: 45 Coffee

3:15 Interim Performance Criteria

3:70 Summaries LSA Lead Center DOE

5:0IJ End of Meeting

3

}

ClOl

J . Hesse (Moderator) A. Li tka W. Hawley G. Peoples R. Kayes ~ C. (!a ;,<.

50 mi n

K. Koliwad (Moderator) 50 min G. Roder ick S. DiZio F. Russell s Deb

I

f {.,, ,t,J ,tvs~' 30 mi n

P. Maycock (Moderator) 50 min r,-;._~---t~~. !'N_rray /4MIII J:J.X,tflY.o~9sdal e . PM' ~ ·'.n ri~ .. ...,_~ "'-'t)IT, in CC,UU//..JtCA71~ ,_,

G~ il len

ClOl M. Alper (Moderator) 50 min

J . " I . (si,) IC? <eS Cl\4:(.c1,:. f>W R.~\\)1, .... , s

J .. t;;,\lldsmith ~'tittle

4 hr C310 R. Lut wack C3l4 J . Liu C301 c. Coulbert C312 D. Bickler C316 R. Ross/L . Dumas

1-1/4 hr

1-1/4 hr C310 R. Lutwack C316 Ross/Dumas C312 P. Henry

Cl 04

30 min

ClOl G. N.Jss - SERI 05 min

1-3/4 hr

ClOl

SILICON MATERIAL

CHAIRMAN: RALPH LUTWACK

THURSDAY - 8:00 a.m. - 3:15 p.m. (Room C310)

8:00 Silane/Silicon Process

9:00 Hydrochlorination of Silicon Tetrachloride and Metallurgical-Grade Silicon

9:20 Gaseous Melt Replenishment System

9:45 Coffee

10:05 Zinc Reduction of Silicon Tetrachloride

10:45 Dichlorosilane CVD Process

11:25

11:45

12:15

1:30

2:10

2:25

2:50

Process Analysis

Silicon Halide/Alkali Metal Process

Lunch

Definition of Purity Requirements

Effects of Impurities on Solar Cell Performance

In-House Silicon Program

Coffee

L.hion Carbide Corp.

Mass. Inst. of Tech.

Energy Materials Corp.

Battelle Columbus Labs

Hemlock Semiconductor

Lamar University

AeroChem Research Labs

Westinghouse R&D Center

c. T. Sah Associates

JPL

NOTE: Time allocated includes discussion period after each presentation.

4

C

C

D '

.

C C '

u

w

LARGE AREA SILICON SHEET

CHAIRMAN: Jim Liu

THURSDAY - 8:00 a.m. - Noon (Room C314)

8:00 Advanced czochralski

8:35 Advanced Czochralski

9:00

9:25

9:45

10:10

10:35

11:00

11:25

11:50

Semicrystalline Process

Coffee

Enhanced ID Slicing

Multiblace Slurry Slicing

Fixed Abrasive Slicing Technique

Partial Pressure of Reactant Gases

Cell Fabrication

Cell Fabrication

Hamco/Kayex

Siltec

SEMIX

Siltec

Hoffman (Norlin)

Crystal Systems

Univ. of Missouri

Applied Solar Energy

Spectrolab

NOTE: Time allocated includes discussion period after each presentation.

5

ENCAPSULATION

CHAIRMAN: CLIFF COULBERT

(Materials and Process Development)

THURSDAY - 8:00 a.m. - f\tlon (Room C301)

8:00 overview

8:20 Low-Cost Materials

9:00 Electrostatic Bonding

9:30 Ion Plating

10:00 Coffee

10:30 UV Screening Films

11:00 Module Design Analysis

12:00 Discussion

6

Cliff Coulbert

Springborn Labs

Spire

Illinois Tool Works

University of Toronto

Spectrolab

F' ~

F1 ~

lr1 ~

tJ I

'

pf{ ~

~ ~

~ . '

.

~ ' .

I I .

~ . '

I '

.

iJ I .

~ ' '

'

~ LhV

~ I I rm ~

PRODUCTION PROCESS ANO EQUIPMENT

CHAIRMAN: DON BICKLER

THURSDAY - 8:00 a.m. - Noon (Room C312)

8:00

8:15

8:55

10:00

10:30

10:50

11:30

12:00

Introduction

Junction Formation

Metallization

Coffee

Assembly

Sequences

Assessment

MEPSDU Status

7

D. Bickler

Spire Lockheed

Bernd Ross Assoc Solarex Spectrolab

MB Associates

Motorola Westinghouse

Univ. of Pennsylvania Science Applications

D. Bickler

PROJECT ANALYSIS AND INTEGRATION

CHAIRMAN: P. HENRY

THURSDAY - 1:30 p.m. - 3:00 p.m. (Room C312)

1:30 IPEG4 Development

2:00

2:15

Energy Payback Analysis

Module Efficiency vs Price Goal Tradeoff

8

R. Chamberlain

E. Muha

R. Aster

I I I I I

I I I I I. ti 11

11 II

I 11

11

I

(;./Zoc~.5 ---&.aa ~ ... s;;?Z, {Z,JPtAt7>"-

,5,r,,C'ALt' )/,S.' /.IA /OC,CI.J,

~;,.7/,r;,I'

ENGINEERING/OPERATIONS AREAS AGENDA

CHAIRMEN: ROSS/DUMAS

JI /. n:t7tA-(;fiU.J ,

vn1.. , n THURSDAY - 8:00 a .m. - 3:00 p.m. (Room C316) D4!X, ~~~~~~~~~~~~~-'--~-'-~~~~~

~o.(.AG~

t-C!<B t?I I 8: 00

8:30 Al M5 tl (!f.rl 17V1 t.fZ>,i.

d,,.,,v,1.....,Z. e;-,./';,so. 8: 45 ~ ;,,,,,,,.,,,= • ..;;

XOKlv ,j.::l~J,..',. /)(J c,./L;,'.

L471t'c. ekTC-r /.,?/W,nnJ ,t),'-?_

;po.Cw 9 : 45 ~7/~IJ,o

t/77 <- w .n:,v.>"7·-.,, 10 : 15 t>e:tC.

10:30

.M/J/IPM/1 11 ·. 00 I ~o,cw/ J i'):;),

t,;:;,7,,:-, 0 t.

/?o"l/¥#t'O /J.wow,•.ue~ 11 : 20 -;,~; t o . Ii 271 Ir".~ 11: 50

JI Ii I /,0. >-4 ..... •

.5B.C'c,.,.,.. 12 : 15 S;7>,J 8ot,<h'PVJ,-.;:,

3CiC4.J .SeJt.4,z,~ 1 : 15 /. ::XVq'/ drt,/,v~i-)V" .s:~ <- 1: 45 St.,N'V~ ,R/M~S 2 :05 ~/y/

2:30

Environmental Testing

vPield Test Update

.lfest and Application Pro jects: ~emote Stand Alone ..PROA 38 -~j: . Laguna 'Natural Bridges ~0'11~u

vtailure Analysis

Coffee

Low-cost Array Structures

J. Griffith

P. Jaffee

LeRC §() !Wuuuy -A. Baisley C.'n ... ~"45

R. Baisley S. Forman

S. Sollock

A. Wilson

Array Wind Tunnel Test Results Boeing /2 /ll, LL. e1<.

Photovoltaic/Thermal Module Development A. Wen/S. Gasner

Array Electrical Safety Guidelines Underwriters Lab

Lunch

tvbdule Hot- Spot Testing Results J . Arnett

Electrical Insulation Study Results G. Mon

Solar Cell Environmental Testing Cl emson Uliversity

Next Meeting

9

TIME

7:30 - 8:00

MODULE DURABILITY AND LIFE TESTING WORKSHOP

TUESDAY, SEPTEMBER 23, 1980 - 8:00 a.m. - 5:30 p.m.

VON KARMAN AUDITORIUM, JET PROPULSION LABORATORY

SPEAKER SUBJECT

REGISTRATION AND COFFEE (Registration will be in the Visitors Lobby)

8:00 Coulbert INTRODUCTION

8:30

9:15

10:00

10:15

10:30

11:15

11:45

12:45

1:10

1:40

2:40

3:00

3:15

5:30

Ross

[)Jmas

o Cl:>jectives, scope of workshop, definitions

MODULE DURABILITY GOALS

o Quantification of durability o Atlowable failure levels

MODULE DURABILITY EXPERIENCE

o o o

COFFEE BREAK

Field exposure/application sites Failure experience versus goals Key failure mechanisms

MODULE DURABILITY DESIGN TECHNIQUES

Coulbert INTRODUCTION

Arnett o 0

Moore o

Moore

Mon

Gupta

LUNCH

0

o 0

o 0 0

Kaelble o

COFFEE BREAK

Soiling Cell cracking/hot spots

Interconnect fatigue

Structural failure/glass breakage

Electrical terminal failure Electrical insulation breakdown

Encapsulant thermal degradation Encapsulant photodegradation Delamination

Corrosion

APPROACHES TO IMPROVED RELIABILITY ANO LIFE INDUSTRY PANEL Ross - Moderator

Adjourn

10

ri bi

Fl ~

F1 ~

C

im,,, I

u

'L..J

L.J

ADVANCED PHOTOVOLTAICS TASK

The overall objective of the Advanced Photovoltaics Task (APT) is to identify, develop and demonstrate new solar cell materials and process technologies capable of producing photovoltaic modules having cost and performance characteristics meeting the Multi-Year Program Plan goal of $0.15 - $0.40/Wp.

D.Jring FY81, technology development assessment studies will be performed with emphasis on amorphous Si and CdS technologies. Such factors as cost potential, scalability in area and production through-put, materials utilization, and module/encapsulation requirements will be addressed. Elements of advanced solar cells which need to be considered for these studies include substrates, semiconductor deposition techniques, barrier formation, metallization and interconnections, antireflection coatings, and encapsulation materials and methods.

Materials and state-of-the art devices being developed in various laboratories, most of which are under contract to the Solar Energy Research Institute, will be obtained in order to provide the LSA Project with the means to independently and continuously assess the technology development prospects of the leading advanded photovoltaic approaches.

The in-house cell and material analysis facility is now being assembled. The facilities presently include: (1) automated data acquisition system for current voltage characteristics, both dark and illuminated, and spectral response, (2) scanning electron microscope with semiquantitative energy dispersive analysis of elements by x-rays, with beam-induced currents, and with cathode-luminescence, (3) photoluminescence, (4) transmittance and reflectance spectra-photometry, (5) high resolution profilornetry, (6) Hall and van der Pauw measurements, (7) scanning laser beam system, and (8) advanced photovoltaic material fabrication equipment. Additional apparatus to become operational during FY81 includes a scanning ion-microprobe for thin-film profiling.

11

w

w

'w

LJ

STATUS OF TECHNOLOGY TRANSFER

The Production Process and Equipment Area is concerned with the transfer of technologies developed by the LSA Project for manufacturers of solar modules. The status of this transfer activity is reported periodically to provide an overview of processes available from JPL, and to promote experimentation, modification, and application. Inquiries and requests for process specifications should be directed to the LSA Project PP&E Area at (213) 577-9225.

Surface Preparation

Process

Texture Etching

Spray A-R

Etching Damage Removal

Text4re Etching

CVD Si3

N4

Plasma Damage Etching

Wax-Masking

Plasma Metal Pattern Diffusion

Spray A-R

Evaluated

Contractor

Lockheed

Lockheed

MBAssociates

MBAssociates

Motorola

Motorola

Motorola

Motorola

RCA

Evaluation in Process

Available

13

25

61

Under Development 3

Suspended 2

Status Surveyed By Comments

E

E:P

EP

EP

EP

A

E

EP

EP

EP

E

EP

General Electric Confirmed

Solar Power

Univ. of PA

Solar Power

Spec trolab

S.olar Power

Solar Power

Westinghouse

Solar Power

App. Solar Energy

Spectrolab

Univ. of PA

Solar Power

Westinghouse

Univ. of PA

App. Solar Energy

Photowatt

Solar Power

Spectrolab

Westinghouse

Univ. of PA

Economic only

Confirmed

Economic only

Economic only

Confirmed

Economic only

Status Code: E - Evaluated; EP - Evaluation in process; A - Available; D - Under development; S - Suspended.

13

Process

si3N4 A-R Coating

Spray A-R

Texture Etching

Wafer Surface Preparation

Plasma CVD

Surface Preparation

Polish Etching

A-R Coating

Proximity Texturing

Dip A-R

Surface Preparation (Cont)

Contractor

Sensor Tech

Sensor Tech

Sensor Tech

Sensor Tech

Sensor Tech

Solarex

Spectrolab

Spectrolab

Texas Inst.

Westinghouse

Status Surveyed By Comments

A

D

E

EP

EP

A

A

A

A

D

E

EP

Lockheed

Solar Power

Univ. of PA

Solar Power

Univ. of PA

Confirmed; req. mod.

Economic only

Economic only

App. Solar Energy Confirmed by Spin-on

General Electric

Solar power


14

C

_,.,.....

L.J

Li

,-,

LJ

Junction Formation

Process Contractor Status Surveyed By Comments

Ion Implant Lockheed EP General Electric

Univ. of PA Economic only

Laser Annealing Lockheed A

Aluminum 'BSF MBAssociates A

Edge Etching MBAssociates A

POC13

Diffusion MBAssociates EP Solar Power

Ion Implantation Motorola EP App. Solar Energy


POC13

Diffusion RCA EP General Electric

Solar Power

Laser Scribing Sensor Tech E Mobil Tyco Confirmed

Spec trolab Confirmed

Westinghouse Confirmed

EP Univ. of PA Economic only

Spray-on Dopants Sensor Tech A

Post-Diffusion Solarex A Cleaning

Al Printing Paste ?pee trolab E Photowatt Confirmed

Laser Scribing Spectrolab A

Polymer Diff. Wafer Spec trolab A

Print and Fire BSF Spectrolab E App. Solar Energy Confirmed

RCA Confirmed

Westinghouse Confirmed

EP General Electric

Photowatt

Remove Oxide and Spectrolab E Photowatt Confirmed Clean Al Rack

EP Univ. of PA Economic only


15

Junction Formation (Cont)


Ion Implantation Spire E Motorola Confirmed Furnace Anneal

RCA Confirmed

EP General Electric

Univ. of PA Economic

Spin-on Polymer Texas Inst. EP Univ. of PA Economic Dopants CVD p+ Westinghouse A


16

only

only

F ~

C ' '

L:.J

,-,

A.

Metallization

Process Contractor Status Surveyed By

Aluminum Back Contacts ARCO Solar

Thick Film

Front Contact Formation

Electroless Pd/Ni

Plate Copper

Thick Film

Electroless Ni

Lockheed

MBAssociates

Motorola


RCA

Sensor Tech

E

EP

EP

EP

E

EP

D

EP

EP

Westinghouse

General Electric

Solar Power

Spire

Univ. of PA

General Electric

Univ. of PA

Solar Power


Photowatt

Solamat

Solarex

Solar Power

Westinghouse

ARCO Solar

NASA Lewis

Spire

Univ. of DE

Univ. of PA

General Electric

Spectrolab

Univ. of PA

Solar Power

Spire

Univ. of DE

Univ. of PA

Comments

Confirmed

Economic only

Economic only

Confirmed

Confirmed

Confirmed

Confirmed

Confirmed

Confirmed (Mod.)

Economic only

Economic only

Economic only

Economic only

Economic only


17

Metallization (Cont)


Electroless Ni Solarex EP Motorola

Solar Power

Spire

Univ. of DE Economic only


Negative Silk Solarex A Screening

Wave Soldering Solarex A

Mo/Sn Metallization SOL/LOS EP Bernd Ross System

Solar Power

Print and Fire Spectrolab E Photowatt Confirmed Front Contact

EP General Electric


Wrap-around Spectrolab s Not cost-eff. at this time


18

I

tJii!d

LJ

LJ

Module Assembly


Elec. Test and Sort MBAssociates A

Framing MBAssocia tes EP Solar Power

Lamination MBAssociates EP Solar Power

Module Layout and MBAssociates EP Solar Power Interconnect

Gap Weld RCA s Not cos t-ef f. at this

Double Glass RCA A

Mass Soldering RCA EP Solar Power

Interconnect and Solarex EP Solar Power Encapsulation

Applying Interconnects Spec trolab A

Circuit Assembly Spectrolab A

Final Test Spec trolab A

Frame Module Spec trolab A

Lamination Spring born A

Laminate Circuit Spectrolab A

Lead Cell Spec trolab A

String Assembly Spec trolab A


19

time

TECHNICAL SUMMARIES

L.J

A

Li

w

n

r,

LJ

In-House Program

SILICON MATERIALS TASK Jet Propulsion Laboratory


Silicon Processing

The objective of this effort is to provide support to the Silicon Material Task in selected areas.

Data from fluidized bed reactor (FBR) experiments were presented to Union Carbide Corporation (UCC) and AeroChem representatives, consultants, and JPL personnel at a review. The experimental results were considered encouraging and useful in that they indicate a suitable range of FBR operation.

Experiments were conducted in the two-inch-diameter FBR. Tests with no agglomeration at entry gas concentrations up to 14 mole% silane (SiH4) at 700°C and at V/Vmf of about 8 to 10 have been confirmed with less than 6% Si dust forming.

A six-inch~diameter FBR experimental system was designed for the JPL in-house program, and procurement was initiated. The system will be used to complement the UCC FBR program in the areas of heating, Si particle handling, seed production, process monitoring, and fundamental understanding. It was designed with the maximum of versatility to study alternate processing techniques if needed during the FBR development period.

In the program on converting SiH4 to Si powder in free space, the continuous-flow pyrolyzer was modified to include a scraper to prevent powder accumulation in the reactor during long-term operation. Studies are continuing to provide information on the cost-effective design of a free-space reactor.

In the Silane-to-Molten Silicon effort, wherein the conversion of SiH4 to molten Si in a single step is being investigated, the reactor and ancillary apparatus were installed, and experimentation has begun.

A method is being developed in the Silicon Material Research Laboratory to consolidate the sub-micron Si powder produced by the free-space reactor of the UCC process. The method consists of melting the powder on top of a pedestal, followed by undirectional solidification. A test apparatus was constructed using the high-frequency generator of a Lepel float-zone apparatus as the source of heat that will be applied to produce a stable melt on top of the pedestal. In the area of analysis for impurities in Si by the TSCAP (Thermally Stimulated Capacitance) measurement apparatus, the facilities and equipment required to fabricate Schottky diodes for these measurements were completed, and diodes are being prepared from n-type Si obtained from the Westinghouse R&D Center program on impurity studies.

Date

21

SILICON MATERIAL TASK

AeroChem Research Laboratories, Inc. Princeton, New Jersey

Contract Title: Development of Processes for the Production of Solar

Grade Silicon from Halides and Alkali Metals

Contract No.: 955491

The objective of this program is to c~aracterize the kinetics and mechanism of the formation and growth of silicon particles from the decomposition of silane at high temperatures. The experiments are aimed at determining the rates at which gas-phase species form silicon particle precursors, the time required for silane decomposition to produce silicon particles, and the competing rate of growth of silicon seed particles injected into a decomposing silane environment.

In this reporting period, the 'AeroChem high-temperature fast-flow reactor (HTFFR) was modified to study the decomposition of silane. The HTFFR consists of a 2.5 cm i.d. alumina reactor tube wound with resistance heating wire in three contiguous independently powered zones. The useful length (for kinetic measurements) of the reactor is the uniform temperature zone extending about 30 cm upstream from the optical observation plane. Experiments will be performed at pressures from 5-760 Torr, at temperatures from 600-1300°C and average gas velocities of 1-100 m s- 1 (i.e., reaction times from% 3-30 ms). A cooled inlet will be used to inject room temperature silane into any part of the uniform temperature zone of the HTFFR, and a fluidized bed feed system will be used to inject silicon seed particles.

The basic experiment consists of varying the position of the cooled inlet and measuring scattered light intensity in the observation plane to determine the particle size as a function of time available for silane decomposition. The amount of silane entering the HTFFR will be known from the flow rate through a precision orifice and the amount of silane leaving the HTFFR will be measured by the IR absorption cell. In addition, simultaneous probing for the Si atoms (not particles) and the SiH (and SiH2) molecular species will be made at the observation plane.

The optical diagnostics consist of absorption and fluorescence spectroscopic techniques to measure gaseous species concentrations and light scattering techniques to monitor particle formation and growth. Silicon atoms (if present) will be detected by atomic absorption, SiH (and SiH2) will be detected using laser-induced fluorescence, and SiH4 will be detected by IR absorption. The techniques under consiµeration to determine average particle size or particle size distribution are turbidity measurements, polarization techniques, forward scattering, and Mie resonance.

~") . .' / . ' ,,·. (_ '-{. ··-..7 •• ~- ·',,,~ f /~ ~t.,· ..

- Approval Signature

22

27 August 1980

Date

C C

F1 bJ

b.J

A

w

L.i

L.J

u

L.J

LJ

Contract Title:

Contract No. :


BATTELLE'S COLUMBUS LABORATORIES

Columbus, Ohio 43201

EVALUATION OF SELECTED CHEMICAL PROCESSES

FOR PRODUCTION OF LOW-COST SILICON

954339

Battelle's Columbus Laboratories (BCL) is developing the zinc vapor reduction of silicon tetrachloride in a fluidized bed of seed particles as a promising process for producing low-cost high-purity free-flowing silicon granules.

Current activity relates to the Process Development Unit (PDU) which consists of four critical units from ~he 50,000 kg/year full-scale EPSDU design, i.e., the zinc vaporizer, fluidized-bed reactor (one of two 25 MT/year units), by-product condenser, and electrolytic cell, with auxilliary equipment to permit operation in an 8-hour batch mode.

Since the 15th PIM in March of 1980, two runs of short duration were made in the PDU, feeding reactants at about one half of the 25 MT per year design rate for the single 7-inch (~18 cm) diameter fluidized-bed reactor. One of these runs was terminated prematurely due to development of a constriction in the downstream section of the equipment. The second was terminated as the result of misalignment and subsequent binding of a chain drive in the displacement system that meters liquid zinc to the zinc vaporizer.

Other attempts to operate the PDU have been thwarted by a series of problems which although minor in origin have been major in effect. At the writing of this summary, it is believed that most of the "bugs" have been worked out and that runs of full 8-hour duration can be anticipata:l.

It appears that a predicted ~100 ppmw of residual zinc in the anticipated lOOOµm silicon granular product of the projected EPSDU operation can be removed (to <10 ppm) by vacuum heat treatment at 1100 C for several days, or over a shorter time if a higher temperature can be sustained without sintering. However, it may be better to avoid the cost of such a treatment, and to accommodate the zinc removal in the subsequent melting of the granules to form sheet. The volume of the condensed zinc that would be handled turns out to be a fraction (probably <5%) of that of the SiO(g) that normally evolves from contact of molten silicon with silica crucibles over a period of a few hours.

Experiments with a direct-coupled induction-heated zinc vaporizer indicate that, by using saturable-core-reactor control, one can avoid the formation of the plasma in the zinc vapor that had plagued initial operation of the PDU and forced the interim adoption of a less controllable vaporizer having graphite-tray susceptors.

Approv~I Signature ' ; I

Date

23


Energy Materials Corporation

Harvard, Massachusetts

Contract Title: Gaseous Melt Replenishment --------955269

The objective of this program is to develop an improved silicon production reactor with periodic batch delivery of product to either a casting or shotting process or through a liquid silicon transfer system directly to a crystal growth system.

The chemical reactions, H2

reduction of SiHC1 3 , are those in commercial use for poly formation. The major innovation is in reactor design which allows a high productivity of silicon. Calculations based on epitaxial deposition rates indicate that a reasonable sized system can produce materials rapidly enough to keep pace with either 10 cm or 12 cm diameter Czochralski crystal growth operating in a semi-continuous mode.

Polycrystalline silicon will be deposited on the inside walls of a resistively heated, multi-walled fused silica reaction chamber by H2 reduction of SiHCl3. After sufficient silicon has been produced, the reactor is flushed with argon and the silicon melted out of the reactor into a Czochralski crystal growth crucible. The reactor is then returned to the deposition stage. The reaction chamber and a heated delivery tube to the crystal growth system are separated by a "U" tube which acts as a valve by adjusting the temperature above or below the m~lting point of silicon contained in the U-bend.

The first phase of the program comprises development of a prototype system capable of a production rate of 0.5 kg/hr. During the last five months we have completed 10 reactor tests. Six reactor tests yielded silicon deposits. A deposition rate of 6 microns/minute and a production rate of 235 grams/hr. at 20% conversion were the best results of the six runs. These six runs used scaled down cylindrical reactors. We plan to test a larger reactor this month in order to produce 500 grams/hr. at 18% conversion~

~gnatllre

24

F1 ~

~ bl

C

F kl

C

F ~

F1 ~

LJ

L.J

Silicon Materials Task

Hemlock Semiconductor Corporation Hemlock, Michigan

Contract Title: Development of a Polysilicon Process Based on Chemical Vapor Deposition

Contract No. : 955533

The objective of this program is to demonstrate the feasibility of a chlorosilane based chemical vapor deposition process for the production of a low cost-high purity polysilicon. Efforts are currently being expended in the following technical areas:

• Dichlorosilane (DCS) Reactor Feasibility/Optimization

• Intermediate Sized DCS Reactor Development

• DCS Process/Product Evaluation

• Preliminary EPSDU Design/Integration

During this reporting period over 30 experimental reactor runs have been successfully completed. Highlights of reactor performance include growth of rods up to 53 mm in diameter ( the largest ever in this reactor) and molar conversions into silicon in excess of 54%. In several cases, silicon deposition rates have been more than a factor of two greater than corresponding runs made with trichlorosilane, with power consumption per kg silicon produced comparably lower. The surface quality of the silicon generated from DCS has been uniformly good. With a single exception, no operational problems unique to DCS have been encountered.

An optimization program for the experimental reactor is nearly completed. It is an experimental design whose purpose is to provide detailed correlations between operating parameters, such as rod temperatures, and system responses, such as deposition rates. A laboratory scale rearranger system has been utilized to determine kinetic parameters for the liquid phase redistribution of trichlorosilane or trichloro-silane/silicon tetrachloride mittures by Dowex ion exchange resin MWA-1. Catalyst lifetime behavior and conditioning requirements have also been established.

Construction of the PDU has been delayed pending receipt of safety-related information about combustion and hydrolytic properties of DCS from experimental work performed by Hazards Research Corp.

Process flow diagrams have been prepared for the EPSDU, a 100- to 200-ton facility. Material and energy balances have also been constructed for the EPSDU.

A preliminary design and economic analysis of a 1000 tonne facility was completed during this reporting period. The 1000 tonne plant includes silicon tetrachloride hydrogenation, DCS generation and purification, silicon production from DCS, and recovery/recycle of chlorosilane byproducts.

Date

25

Contract Title:

Contract No. :


LAMAR UNIVERSI1Y Beaumont, TX 77710

PROCESS FEASIBILITY STUDY IN SUPPORT OF SILICON MATERIAL TASK

954343

Major efforts in chemical engineering analysis centered on the DCS process - Case A which involves production of dichlorosilane (DCS) as a silicon source material for polysilicon production in the Hemlock Semiconductor Corporation program. The preliminary process design of a plant to produce DCS was completed including process flowsheet (100%), base case conditions (100%), reaction chemistry (100%), raw materials (100%), utilities (100%), major process equipment (100%) and production labor (100%). The process design package was forwarded for economic analysis.

Economic analysis of the DCS process - Case A was completed using detailed data from the process design package. The economic analysis indicated a total product cost without profit of 1.29$/kg of DCS (1980 dollars). This product cost without profit includes direct manufacturing cost, indirect manufacturing cost, plant overhead and general expeneses. The sales price of DCS at 15% DCF rate of return on investment is 1.47 $/kg (1980 dollars). Additional results are reported for various profitability levels as measured by ROI (return on original investment) and DCF (discounted cash flow rate of return).

Analyses of process system properties were continued for chemical materials important in the production of silicon including compilation and collection activities of the property data for use in the final report. Progress and status are reported for critical constants, vapor pressure, heat of vaporization, heat capacity, density, surface tension, viscosity, thermal conductivity, heat of formation and Gibb's free energy of formation. Property data compilation and collection activities are about 25% complete in -regards to the final report.

Approve~ Date

26

C

r, ~

C

C

C C

C

k..l

Contract Title:

Contract No. :

Silicon Materials Task

Massachusetts Institute of Technology Cambridge, Massachusetts

Investigation of the Hydrogenation of Sicl4

955382 (subcontract under NASA Contract NAS7-100, Task Order No. Rd-152)

A laboratory research and development program is carried out to study the hydrochlorination of SiC14 and m.g. Si to SiHC1

3,

3 SiC14 + 2 H2 +Sit 4 SiHC13

To continue the copper catalyst studies, the effect of copper concentration on the reaction rate is investigated. CuCl is an effective catalyst for this reaction at concentration as low as 2 wt% based on Si.

The effect of Si particle size distributions is also investigated. Reaction kinetic measurements are made on two particle size of m.g. Si, 32x65 mesh (500x210 µm) and 150x400 mesh (105 x 34µm) in addition to the standard 65x150 mesh Si used in previous studies. Results of this study show that the reaction rate is essentially independent of Si particle size, Thus, the reaction occurringat the Si metal surface is the rate-determining step. Mass transfer via diffusion is not rate-limiting.

The effect of impurities in the 98.5% m.g. Si metal on the reaction rate is also studied. An ultra high purity Si mass (32x400 mesh) is prepared by grinding a sample of electronic grade, polycrystalline Si. Reaction kinetic measurements on this high purity Si mass show that the rate of the hydrochlorination of SiC14 to SiHC13 is about one order of magnitude slower than that of the m.g.·Si under similar reaction conditions. The impurities in the m.g. Si appear to act as a catalyst. Addition of CuCl to this high purity Si mass greatly increases the reaction rate. Thus, copper catalyst provides a convenient means to recycle ·off-specification solar grade Si metal. Plausible reacti.on mechanism on the hydrochlorination of SiC14 is discussed.

~7:6:~i ~I Signature

27

Contract Tit I e :

Contract No. :


UNION CARBIDE CORPORATION

Tonawanda, New York 14150

Silane-to-Silicon EPSDU

954334

Preparation of the site for the 100 MT/yr EPSDU at E.Chic., Ind. was started in July, marking the beginning of installation subcontracting which will continue through 1981. The site preparation work included grading, fencing and roadbase, and was completed in Aug. The civil installation pkg. covering equipment foundations, structures, and utilities has been issued for bids. This work is scheduled to start in late September, 1980.

All of the basic EPSDU facility design requirements have been obtained. Engineering drawings which show the location of each piece of process equipment have been finalized and are now being detailed to show piping, valves, elec. connections and instrumentation. Supporting design documents such as process control diagrams and wiring diagrams are being reviewed and finalized. Environmental permits have been obtained and design safety reviews are ongoing.

Orders have been placed for all major pieces of equipment required for the M.G. silicon to silane portion of the process. Equipment items on order include hydrogenation and redistribution reactors, distillation columns, heat exchangers, pumps, compressors, storage tanks and bins. Equipment has been ordered for delivery starting in Dec., 1980 for installation at EPSDU during first and second quarter, 1981.

Based on the EPSDU design, a process design for a conunercial lOOOMt/yr facility was completed, including a flowsheet, process description, mass balance, facility layout and equipment functional specifications. This package will form the basis for the upcoming commercial economic assessment.

The free-space reactor PDU work entered into a new phase to demonstrate purity and operability. The reactor wall temperature profile was modified and a quartz liner was installed. Five consecutive runs of two hours each were conducted at the design throughput of 5 lb/hr. A long duration (12 hr) run will be conducted shortly. Design of an alternate silane pyrolysis PDU using a fluidized bed was completed, and fabrication of the unit is underway.

Melter subcontract work with Kayex has proceeded on schedule. Design of a prototype melter system is complete, except for the powder feeding equipment. Small scale melting and shotting feasibility tests were conducted in parallel with the design effort.

A slim-rod reactor and an epitaxy reactor for silane purity analysis were fabricated, and silicon deposition testing has started. Both units are ready for Q.C. testing method development.

~~ Approval Signature

~2-z,lf?o Date

28

r bJ

bJ

LJ

i

bl

SILICON MATERIALS TASK (Part 2

Westinghouse R&D Center/Dow Corning Corp.

Pittsburgh, PA 15235

Contract Title: INVESTIGATION OF THE EFFECTS OF IMPURITIES AND PROCESSING ON SILICON SOLAR CELL PERFORMANCE


The objective of this program is to determine the effects of impurities, processing and impurity-process interactions on the properties of silicon and silicon solar cells so that impurity limits for solar grades of silicon can be defined and cost-benefit tradeoff can be made by the users of this cheaper, less pure silicon.

The phase IV effort now underway includes five major topics of study: (1) evaluation of experimental silicon materials, (2) investigation of impurity effects in polycrystalline devices (3) identification of impurity thresholds for high efficiency cells (4) assessment of process effects such as ion implanting on impurity-doped devices and (5) an extension of studies to identify long term impurity effects. The dominant activities so far have been in the areas of polycrystalline cells, impurity aging affects and high efficiency cell modeling.

We have found that the threshold for ingot structural breakdown is lower during polycrystalline silicon growth than when growing single crystals, at least for the impurities Fe, Ti, V, Cr and Mo. At impurity concentrations for which single crystal can be grown, polycrystalline ingots develop metal-rich inclusions. The effect of the inclusions is to shunt solar cells producing very low efficiencies. When the melt impurity concentration is reduced by 30 to 50%,inclusion incorporation is generally eliminated. Polycrystalline cells doped with lower levels of Ti and V (~ 1013 cm-3) show little indication of impurity segregation to grain boundaries.

Further studies of accelerated aging effects under thermal stress indicate that while solar cells bearing Ti or Mo would show essentially no performance reduction due to impurities after 20 years, cells doped with Cr, and Ag degrade much more rapidly apparently in keeping with their expected high diffusion constants. Nb doped cells fall somewhat between these pairs.

We have initiated the development of a model to depict the functional relationships betwee~ cell performance and impurity content for high efficiency devices including back surface field cells. Qualitatively the model predicts lower impurity thresholds for performance reduction in high efficiency cells that .in our standard devices.

Approval Signature Dote

29

LJ

bl

LJ

LJ

LJ

In-House Program

LARGE AREA SILICON SHEET TASK



(1) Silicon Crystal Growth Laboratory:

Crystal growth work on silicon bicrystals with varying grain orientation is continuing. Experiments on crystal growth with impure silicon and crystal growth in low-cost crucibles have been initiated.

(2) Photovoltaic Materials and Device Testing Laboratory:

The following work is in progress:

(a) Preliminary experiments with bicrystals using static and transient capacitance measurements.

(b) Minority car~ier lifetime and surface recombination velocity measurements using a SEM with beam blanking.

(c) Minority carrier diffusion length and optical absorption coefficient measurement and analysis of low-cost sheet.

(d) Recombination mechanism studies utilizing dark IV and EBIC results on various silicon sheet material.

(3) Solar Cell Prototype Fab~ication Laboratory:

Routine fabrication and testing of solar cells made from different low-cost sheets is continuing.

~--Date

31

LARGE AREA SILICON SOLAR CELLS

Applied Solar Energy Corp. City of Industry, Calif.

Contract Title: Silicon Solar Cell Process Development, Fabrication, and Analysis


The objective of this program is to investigate, develop, and utilize technologies appropriate and necessary for improving the efficiency of solar cells made from various unconventional silicon sheets. Silicon sheets processed included EFG ribbons, dendritic Web, SOC and wafers from HEM cast ingots and ingots from semi-continuous CZ growth techniques.

Solar cells were fabricated using a baseline process. Other process variations, such as formation of shallow junction, fine grid lines, BSF, better AR coating and application of gettering, etc. Performance was evaluated under both AMO and AMl illumination conditions. Comparison was made with conventional CZ silicon slices processed with the sheets. In addition, back-up measurements were made of minority carrier diffusion length,spectral response, dark diode I-V characteristics and small light spot scanning. Good agreement was found between these back-up measurements and the cell performance. In particular, minority carrier diffusion length was still seen to be a dominant factor in determining cell efficiency.

Discussion is given on the effect of grain sizes and BSF on solar cell parameters. For completeness, performance summaries of all sheet cells evaluated to date will be included.

K.S. LING 8/27/80 Approval Signature Date

32

C

C

C ri ~

181 b±J

C

C

C

C C

LJ

L.J

Pl i


CORNELL UNIVERSITY

ITHACA, NY

Contract Title: Characterizationof Structural, Electrical and Chemical Properties of Silicon Sheet Material


Progress in the period 05/15/80 to 08/28/80 can be summarized as follows: EFG and Web material was analysed by optical microscopy and etching, EBIC (electron beam included current microscopy}, TEM (transmission electron microscopy}, and HVTEM (high voltage transmission electron microscopy at 1.2 MeV). The results are:

(a) Web material

Our observations on the Web material agree with previous published findings which report that:

(1) The major structural defect in the Web material is a single or multiple twin in the central plane of the ribbon.

(2) The dislocation density varies over _the cross section but is generally relatively low (-105 cm-2 ).

(3) The dislocat!~ns have burgers vectors <110> and line direction [211] and <110>.

New features, not previously reported, are:

(1) The central twinning region may be a microtwin or twins. (2) Dislocations with burgers vectors of the <211> type acconnnodate

small tilt components between the twin planes. (3) Hexagonal partial dislocations arrays acconnnodate small twist

components between the twin planes.

(b) EFG material

The new findings are:

(1) The electrical activity of twin boundaries exhibiting dotted EBIC contract is associated with the presence of partial dislocations in the boundary.

(2) A significant fraction of the straight twin boundaries present are secondary twins of the (111)-(115) type. These secondary twins contain a high density of dislocations and are strongly electrically active.

The latter finding is particularly interesting, since it was previously assumed by all investigators of EFG material that the straight twin boundaries are all of the same type, i.e., coherent twins.


33

Contract Tit I e :

Contract No. :


CRYSTAL SYSTEMS, INC.

SALEM, MA 01970

SILICON INGOT CASTING--HEAT EXCHANGER METHOD (HEM)/ MULTI-WIRE SLICING--FIXED ABRASIVE SLICING TECHNIQUE (PHASE III)

954373

This contract is for casting silicon ingots by the Heat ExchangerMethod (HEM) and slicing by multi-wire Fixed Abrasive Slicing Technique (FAST).

Silicon ingot size cast by HEM has been extended to 34 cm x 34 cm x 20cm weighing 45 kg. This is the largest silicon ingot cast. The first ingot of 34 cm x 34 cm cross-section was 10 cm high weighing 20 kg. Solidification rates of 3 kg/hr were achieved and there was no problem of crucible attachment or ingot cracking. This was followed by casting a 26 kg ingot and later two 45 kg ingots.

One of the problems encountered in ingot casting was cracking of the crucible. During heat treatment to develop a graded structure, cracking occurred or it cracked while loading. The cracking during heat treatment was eliminated by reducing the gradients and cracking under load was minimized. The crucible, therefore needs to be annealed to remove most stresses prior to loading and supported to create uniform loading conditions. Crucibles have been fabricated to achieve a squarer shape. The 32 cm x 32 cm x 40 cm crucibles have flat sides and bottom with minimum taper and curvature in the corners. The crystals' shape will be similar to that achieved by welding flat plates together.

A new slicing head was designed and fabricated. The salient features of this bladehead are a very high degree of rigidity and accurate alignment. The bladehead has been enlarged to accommodate 750 wires (25 wires/cm). Initial testing with the new bladehead has shown that 500 feet per minute surface speeds can be achieved. Increased speed is limited by the drive unit rather than the bladehead. Slicing tests using the bladehead with electroplated wires has resulted in average slicing rates of 5.1 mils/minute (0.13 mm/min) with 83% yield.

Blade development has been continued. Along with the 45 µm diamonds used for slicing, smaller filler diamonds were used to prevent erosion of the matrix. It has been demonstrated that the diamond pull-out problem is thus minimized. So far only impregnated blades had diamonds in the cutting edge only. It has been shown that it is possible to electroplate diamonds also in the cutting edge. This will give reduced kerf, improved accuracy and less costs.

34

August 26, 1980 Date

F1 bl

D . '

Fl kJ

~ '

'

F1 b:d

Fl I : Ld

C ' I

'

Fl IJ

r1 bJ

bl Contract Title:

Contract No. :


P.R. HOFFMAN COMPANY

Carlisle, Pennsylvania 17013

Slicing of Single Crystal and Polycrystalline

Silicon Ingots Using Multi-Blade Saws

955563

Work on this contract was completed, and a final report submitted, on May 23, 1980.

The objective of this program was to slice several silicon ingots to yield 8 and 10 mil thick wafers at a rate of 18 to 25 wafers per centimeter of ingot length. Sufficient evaluation of dimensional parameters, yields and production rates was to be performed to provide for optimization of the wafering process.

Results of the tests performed indicate that current state-ofthe-art of Multi-Blade Slurry Wafering does not provide for successful wafering of 1 m2 per kilogram of 10 cm diameter silicon ingot. The major problems to be overcome are related directly to blade wear, feed force control, and abrasive slurry characteristics. Other major factors in accomplishing the goals of the Silicon Sheet Task are the cost of consumables and wafer cleaning and handling.

A proposal for investigation of all the above problem areas is currently being developed for submission to J.P.L.

; 7 Date

35

Contract Title:

Contract No. :


HONEYWELL TECHNOLOGY CENTER BLOOMINGTON, MINNESOTA 55420

Silicon-On-Ceramic Process

954356

SCIM-coating of wide substrates (10cm x 100cm) has been investigated over a range of substrate velocities (4-30cm/min). At high speeds, the coating process works very well, but the layers have been too thin ( <SOµm). At the speeds required for adequate thickness (3-Scm/min) there have been problems with substrate buckling and breakage due to thermal stress developed on cooling. Achievement of the desired linear longitudinal temperature profile in the cooling zone has been slow due to difficulties in measuring the temperature of moving substrates. Thermal modeling has been helpful in quantitative design and in qualitative understanding of the various temperature readings.

Achievement of uniform transverse temperature profiles in the coating trough has also been more difficult than expected because of the low emissivity of liquid silicon compared to mullite and carbon. With adequate temperature uniformity (10-20°C across the substrate) and with proper substrate temperature on entering the coating zone, smooth layers without dendritic structure are obtained. To date, coating of fully slotted substrates has not been attempted, and cell fabrication has been only on dipcoated material.

Cell efficiencies have been significantly increased by using a slow cooldown after the phosP.horus diffusion. The best SOC cell had a total area conversion efficiency of 10.5%, (AMl,AR) for a cell area of 5 cm2. For 29 recent cells, the average efficiency was 9.9% with a standard deviation of 0.3%. For most of these cells, the cooling rate was approximately 5°C/min for temperatures between 850°C and 700°C. In addition to the slow-cooldown experiments, we have investigated a number of other experiments to improve the performance of SOC cells. These experiments include two-step diffusions and KOH etching.

The LBIC technique was used to investigate grain boundary passivation and to determine the effects of processing on diffusion length. For one group of SOC cells, very encouraging results were obtained using material that was treated in a hydrogen plasma at Sandia. This treatment not only reduces recombination at grain boundaries to a considerable extent, but also increases the bulk-diffusion length within grains. For another group of SOC cells, the LBIC response indicated that highly twinned regions with no subsurface structure, tend to have low dislocation densities and good LBIC responses.

Approval Signature

36

August 29, 1980

Dote

Fl w F1 bu

Fl ~

F1 ~

F1 Lilll

F1 ~

F1 lad

F1' ~

~ kJ

F1 ~

Fl kl

Fl ~

Fl ~

F Ld

P'l

L,J


HAMCO DIVISION OF KAYEX CORPORATION

ROCHESTER, NY

Contract Title: Low Cost Czochralski Crystal Growing Technology


The program objective is to demonstrate the growth of 150 kg of 6" diameter silicon crystal from one crucible utilizing the Czochralski technique. The process developments deal with improvements aimed at lowering crystal growth costs.

Process automation techniques utilizing microprocessor controlled crystal growth are also under development and are aimed at reducing cost and improving process yield.

Development priorities have been issued by JPL as follows:

Priority 1: Priority 2: Priority 3:

Microprocessor controls Accelerated growth Accelerated meltback - chunk material utilizing cold crucible premelter.

The accelerated meltback of polycrystalline rods utilizing RF heating has been de-emphasized.

A series of single batch crystal growth runs have been made utilizing microprocessor control. The runs have been demonstrated utilizing 12 inch diameter crucibles and 4" diameter crystal growth.

A molybdenum heat sink arrangement has been designed and fabricated. Several crystal growth runs have been made, i.e. batch and recharge using this heat sink.

A successful 150 kg 6" diameter crystal growth run has also been demonstrated.

The cold crucible premelter system has been assembled and successful melting trials have been undertaken on a bench scale. Interfacing of the equipment with the crystal grower is ongoing.

A further 150 kg run is planned utilizing microprocessor control of the growth process for 6" diameter.

Contract 954888 has been extended for the purpose of designing a crystal growth facility based on the Kayex semicontinuous crystal growth process. The objective is to satisfy TR82.

E. J. Roberts 09/02/80

Approval Signature Date

37

Contract Title:

Contract No. :


MATERIALS RESEARCH, INC.

CENTERVILLE, UT

Analysis of Defect Structure in Silicon

955676

Materials Research, Inc. has recently entered into a new 12 month LSA Task II support contract (july 1980) entitled Analysis of Defect Structure in Silicon, Contract No. 955676. Quantitative analysis of silicon sheet will involve characterization of grain size, dislocation density, twin boundary (spacing and density), and impurity precipitates. Defect analysis will be performed on JPL-provided samples of Edge-defined Film-fed Growth (EFG), Silicon-On-Ceramic (SOC), cast ingot by Heat Exchanger Method (HEM), Dendritic web, and Czochralski sliced ingots.

A quantimet 720 Image Analysis Unit is used with computer programming to scan a sample. A predetermined number of data taking fields of view per sample is used. The usual output is a computer printout listing dislocation and twin density values. Grain size and impurity precipitate values are usually determined on an optical microscope.


38

ri ~

C .

.

Fl w

~ I I .

F1 t±d

F1 kl

F1 bJ

F1 kl

r w

. C

r bJ

Ld

\,' =

Contract Tit I e :

Contract No. :


Mobil Tyco Solar Energy Corporation

Waltham, Massachusetts 02154

LARGE AREA SILICON SHEET BY EFG

954355

In the time period since the last PIM, the contract was to achieve several significant milestones, culminating in a full-scale "Technical Features Demonstration" in which mulple (three) growth of 10 cm wide ribbon was to be achieved at high speeds (4.5 cm/min) for eight hours. A mean solar cell efficiency of 10.2% was also to be obtainable from ribbon produced during this demonstration.

There was, indeed, significant progress toward all goals. Firstly, solar cell efficiencies over 13% were demonstrated to be achievable from resistance machine-produced EFG material using small (6 cm2) cell areas. Secondly, in non-continuous growth of single 10 cm wide ribbons, speeds up to 4.5 cm/min were demonstrated and cell efficiencies over 10% on large area cells (N50 cm2) were also reached in such ribbon.

Also, in single, but continuously melt-replenished cartridges, 10 cm wide ribbons were grown over eight to nine hours at speeds around 3o5 cm/min under fully automatic control, and all automatic control systems for the multiple ribbon equipment were built, assembled, and tested well before the planned completion date.

However, the "Technical Features Demonstration" cannot at this time be considered to have been successful. Even though in all four full-scale multiple runs significant lengths of ribbons were grown from some of the cartridges, the length of time in which the cartridges all grew full width ribbon stably was much too short. In fact, we must at this point conclude that a 4" multiple furnace needs significant further engineering development and some redesign before definite conclusions can be drawn about the detailed design features of a future full-scale production unit.

Fritz V. Wald 8/22/80 Approval Signature · Date

39

Contract Title:

Contract No. :


SEMIX INCORPORATED

GAITHERSBURG, MD

Semicrystalline Casting Process Development and Verification

DE-FC01-80ET 23197

"Semicrystalline Casting Process Development and Verification" is a three year Cooperative Agreement between Semix Incorporated and the United States Department of Energy. The goals of this Agreement are to demonstrate the commercial readiness of a silicon sheet manufacturing process compatible with the 1982 price goal of $2.80 per peak watt and to demonstrate the technology readiness to meet the 1986 price goal of $.70/Watt.

The initial effort is aimed at economic evaluation of the projected 1982 and 1986 technologies in order to pinpoint critical process subsystems and set specific technical objectives for achieving the price goals. Equipment and process development will be carried out to meet the yields, throughput, productivity and other process parameters necessary to support program goals. A continuous verification procedure will be maintained to insure technical and economic viability of each development change. Current SAMICS analyses show that projected Semix semicrystalline technology can produce sheet material to meet 1982 and 1986 price goals.

The first phase (June 1980 - June 1981) has been organized to define, develop, and assess critical subsystems to improve their capability. During the second phase (June 1981 - June 1982) these individual critical subsystems will be assembled into operational units and evaluated for improvement in process technology. Modifications will be made as required. The third phase (June 1982 - June 1983) will encompass the combination of these subsystems into an Experimental Process System Development Unit (EPSDU), which will be operated with minor modification, to demonstrate connnercial readiness. In all three phases, work will be carried out in parallel to demonstrate the technology readiness for meeting the 1986 goals.

September 2, 1980

Date

F71 kBJ

r, ~

Fl ~

C

C

C

w

LJ

Contract Title:

Contract No. :


Siltec Corporation Menlo Park, California

CONTINUOUS LIQUID FEED CZOCHRALSKI GROWTH

DOE/JPL-954886

This project is directed toward the design and development of equipment and processes to demonstrate the continuous growth of crystals, by use of the Czochralski method, suitable for producing single-crystal silicon for use in so 1 a r ce 11 s. 11 Continuous is defined as the growth of at least 150 kg in monocrystalline silicon ingots, 150 rrun in diameter, obtained from one growth crucible.

Our approach in meeting this goal is to develop a furnace with continuous liquid replenishment of the growth crucible. This has been accomplished through the use of a meltdown system with a continuous solid silicon feeder and a melt transfer system, with associated automatic feedback controls. During the past months, several demonstration runs incorporating continuous melt replenishment were performed, growing individual crystals of 150mm in diameter and with weights of 40, 52 and 65 kg per ingot.

We experienced problems in handling ingots of this size during some of the demonstration runs. One of the three 50-kg ingots to be pulled for the 150-kg goal fell into the melt during the process of removing it from the furnace. This was caused by damage introduced into the seed through the seed holder. The problem was corrected by using a new design for the seed holder when pulling ingots with weights >50 kg. Growth conditions were extremely stable while growing these large ingots with continuous melt replenishment. Average growth velocity deviations were only ±0.25 in./hr. Typical solidif~cation rates during these runs were determined to be 3.5 to 4.0 kg/hr. Structural problems occurred typically about 12 to 15 in. below the ingot shoulder, which was attributed to silicon monoxide particles interfering with crystal growth. However, it was possible to grow monocrystalline, dislocation-free material after several ingots had already been pulled and half the melt had already been in the crucible for more than 60 hrs. This is a significant result, for it shows that it is possible to grow large portions of the 150 kg material monocrystalline, provided the silicon monoxide level in the furnace interior is kept to a minimum. This is usualli accomplished when leak rates of the total system are kept below 10- Torr liter/second. Additional work was done to increase the amount of monocrystalline material yield by carefully controlling the argon flow inside the furnace interior, thereby reducing SiO buildup in critical areas.

./ App ~val Sig[ature

41

6/19 /Bo r 7

Date

Contract Title:

Contract No. :


Siltec Corporation Menlo Park, California

ENHANCED I.D. SLICING TECHNOLOGY

DOE/JPL-955282

The purpose of this program is to develop and demonstrate enhanced I.D. slicing technoloqy that will siqnificantlv increase the number of usable slices per inch of ingot over industry practice. This method requires a reduction of both blade and slice thickness and will be achieved through a combination of three key slicing technology elements: ingot rotation with minimum exposed blade area, dynamic cutting edge control, and the use of prefabricated insert blades. Experimentation with ingot rotation and minimum exposed blade area continued during the past months. Although average cutting feed rates of 13 to 15 mm per minute for slices 100 mm in diameter, 250 µm thick, with kerfs of 152 µm were produced, these results could not be demonstrated on a consistent basis. Problems usually occurred after the cutting edge had penetrated 0.7 in. into the ingot, in the form of fracture lines, following the curvature of the cutting edge. This problem persisted for a wide range of cutting parameters and was identified as the result of high frequency vibrations of the cutting edge. The effectiveness of the cutting edge position control system was further evaluated. Blade deflection values for 250 µm wafers, cut with 152 µm kerfs, were typically reduced by one order of magnitude. The impact on dampening vibrations of the blade cutting edge through the deflection control mechanism was minimal, but cutting rates could be increased from 15 to 25 mm/min in the first 0.5 to 0.7 in. of radial cutting edge penetration. Further work was performed to manufacture blades with prefabricated cutting edge inserts. Several samples utilizing electron beam welding were prepared with good bonding results. Sample preparation with bonding the entire ring to the core are presently underway. Difficulties arose during the welding process through the presence of diamonds in the nickel/ stainless steel bond interface. Exclusion of the diamonds through an additional plating process eliminated this problem. The current main thrust of this effort is to eliminate the slice fracturing problem by eliminating high frequency vibrations. Current slicing tests are performed with a 16 5/8-in. cutting head with a lower overall natural frequency.

o' Aproval sfsnature 8/2.<!/BO 1 Dfte

42

fl bttJ

r9 ~

C

C

L:..J

Contract Title:

Contract No. :


Spectrolab, Inc.

Sylmar, California

Silicon Solar Cell Process Development

Fabrication and Analysis

955055

The objective of this contract is the fabrication of solar cells from the various unconventional silicon materials by a baseline process, to fabricate solar cells from these same materials by methods intended to obtain improved conversion efficiency by known techniques, to fabricate solar cells by low cost methods, to measure the solar cell characteristics and provide an analysis of the results.

At this PIM we intend to report on gettering experiments with Hem material, BSF on EFG material, BSF on Web and shallow junctions in SOC material fabricated into solar cells.

A comparison will be given between AMO and AM1 measurements. A general review of the materials studied throughout the contract will also be given.

43

27 August 1980

Date

Contract Title:

Contract No. :


University of Missouri - Rolla

ROLLA, MO

Determination of Effect of Varying Partial Pressures of Reactant Gasses, Primarily Oxygen & Nitrogen, in a Furnace Atmosphere where Molten Silicon is in Contact with Die & Container Material. 955415

Of all the materials studied under this program to date silicon carbide is the most attractive since it experiences the least degradation of its polished surface while at the same time being most wetted by molten silicon. Current research being carried out on General Atomic silicon carbide coated graphite conforms these conclusions. In addition, the importance of the removal of adsorbed oxygen from the silicon and substrate surfaces prior to melting the silicon has been demonstrated. Continued decrease of sessile drop contact angle after long times appears to be correlated with interaction between the silicon and the substrate, and occurs when the surfaces have not been allowed to reach an equilibrium with the gaseous environment which must have a P02 below the Si-02 equilibrium value. When such equilibriation is permitted to occur, the sessile drop contact angle quickly reaches a very stable value, which is then maintained for the duration of the experiment.

Several other activities in connection with the silicon carbide sessile drop experiments are being pursued. Experiments designed to verify the previously reported method for obtaining the surface energy of die and container materials are continuing. The formation of a silicon nitride coating on silicon sessile drops when nitrogen (PN2 ~ 0.5 atm) is introduced into the H2 buffer gas atmosphere prevents the formation of an equilibrium contact angle.

Analysis of data obtained on the oxygen content of the silicon furnace purge gases in the JPL and Mobil-Tyco facilities indicates that equilibrium conditions do not exist between their purge gases and the molten silicon. For this reason it is possible to use inert gases containing levels of oxygen much higher than the equilibrium oxygen partial pressure without seriously contaminating the molten silicon with oxide. This is true for several reasons. First, much of the purge gas never reaches the temperature of the molten silicon, and thus the oxygen has insufficient time to react before being exhausted from the sytem. Second, the portion of the oxygen that thermally accommodates with the 1700K graphite surfaces in the furnace is quickly converted to CO. Finally, in the case of the Mobil-Tyco ribbon-pulling system, the oxides that do form on the surface of the silicon reservoir remain there as a skin, while the silicon used in the formation of the ribbon is drawn from below this floating oxide skin through the die where it is exposed only to graphite and not to the surrounding atmosphere, maintaining an extremely low oxygen activity until it emerges from the top of the die where it very quickly freezes before oxidation can occur.

P. Darrell Ownby

Approval Signature

44

August 29, 1980

Date

f8l ~

119 bd

fl ~

F' bcl

w I

.

Fl. w F ~

F' bJ

,,,,,..,

L.J

LI

Pl

Contract Title:

Contract No.:


Westinghouse R&D Center


Silicon Web Process Development

DOE-JPL-954654

Phase III of this program was completed July 23, 1980. The highlight of work since the April 1980 Project Integration Meeting was the successful demonstration of semi-automatic web growth, including demonstration of the contract goal of eight hours. Development of fully automatic melt level control was the final requirement necessary for achievement of semi-automatic growth.

The melt level control system developed for this program consists of a three component control loop comprising: a) a melt replenishment system, b) a melt level sensing system and c) a circuit which closes the loop with components a and b. During the previous reporting period long term manually controlled melt replenishment was demonstrated for 17 hours which constitutes the growth period for a one day (24 hour) growth cycle. In the same period a melt level sensing system was installed and operated successfully. Closing of the loop provided fully automatic control of melt replenishment and, in so doing, provided se~ automatic control of web growth. The semi-automatic growth mode is very cost effective because operator action is drastically reduced and the permissable length of growth run is extended to the desired order of three days or gre~ter.

In addition to providing semi-automatic growth, automatic melt level control permits web growth to occur under optimum conditions of speed, width and minimized thermally generated stress. Control of these parameters is very cost effective because they govern the area throughput rate and affect the quality of the web.

Evaluation of the economics of dendrite recycling was completed. At the most recent Project Integration Meeting the economic significance of three options for dendrite recycling was reported. The assumed high quality of web grown from melts containing re-cycled dendrites has now been verified thus confirming the economic projection that recycling of dendrites provides a small but significant cost saving.

Phase III of this program is now completed and all critical goals have been reached. The reaching of these goals is essentially the pre-requisite for future work to demonstrate technology readiness.

Approval Signature ~2S,l1Rh

Date

45

'L:...J

w

L.J

LJ

In-House Program

Encapsulation Task



Life Prediction Modeling

Developing quantitative relationships that relate environmental stress such as solar ultraviolet, wind, temperature extremes, and moisture to the rate of degradation of module performance and structural integrity are objectives of the Encapsulation Task in-house efforts. These activities are integrated with contractual activities to develop an over-all module life prediction method.

Photodegradation rates and mechanisms and ultraviolet absorption characteristics of polymeric encapsulants are being measured as a function of polymer composition and test exposure conditions. Data are being obtained for silicones, EVA, P-nBA, PVB, ·and polyurethane. Additional materials will be characterized during the coming year.

Modeling of the photodegradation of UV screening acrylic outer cover films has yielded rates of degradation of the material constituents and of the total system. These data have been used to provide material composition criteria for the achievement of optimum low-cost long-life cover films.

Encapsulation material degradation data for low-cost advanced encapsulant systems is being gathered using various test hardware such as mini-modules (12" x 16"), two-cell modules and individual material samples. Exposure facilities include JPL laboratory test chambers and selected California field test sites at Point Vicente, JPL, Goldstone, and Table Mountain.

A structural computer model has been formulated to study failure modes associated with temperature and moisture expansion stresses within the module encapsulation system. The purpose of this study is to identify areas of potential cracking and delamination and evaluate the possible propagation of these failures.

A long term accelerated module life test has been implemented to evaluate the validity of a life testing plan developed by Battelle. A closely controlled and monitored module degradation rate experiment with accelerated temperature cycling, high humidity, so2 gas and applied current flow is being conducted with ten prototype modules simultaneously over a four to six month test period.

Dote

47

ENCAPSULATION TASK

CASE WESTERN RESERVE UNIVERSITY

CLEVELAND, OHIO

Contract Title: Systems Studies of Basic Aging and Diffusion


The objective of this study is to establish the kinetics and mechanism of thermal and photo-oxidative degradation of candidate encapsulation materials for the LSA Project. That information is to be used to develop accelerated testing and life performance prediction methods and to guide the continuing selection of optimum performance encapsulation material systems. The study is being carried out in cooperation with related studies at JPL.

The present work is aimed towards establishing the modes of degradation of poly(n-butylacrylate)(PNBA) and their effects on selected chemical and physical properties. It has been found that for PNBA: 1. Extensive, nearly complete, gel formation is obtained after 22 hours of exposure to a medium pressure mercury lamp. Changes in the molecular weight of the residual soluble fraction and the amount of gel formed as a function of exposure time have been established. 2. Spectroscopic analysis of the system indicates that only traces of new chemical groups are formed during the exposure time. There is no evidence for the evolution of significant amounts of volatiles during that period. 3. Thus, it is apparent that the primary effect of exposure is the formation of a crosslinked gel structure with only traces of other chemical reaction products. A crosslink density of less than 1% can account for the formation of insoluble gel without producing a significant concentration of other chemical groups, including volatiles. 4. A study of dynamic-mechanical properties by torsion-braid analysis shows essentially no change in behavior for samples exposed up to 42 hours. The material retains its desirable properties at least to that exposure level.

Continuing studies include the determination of: 1. Changes in polymer mass, molecular weight and its distribution (including network structure) and chemical composition after very long term exposure. 2. The quantum efficiency of the reactions. 3. Changes in polymer flow (creep), dynamic-mechanical properties, and other relevant physical properties related to continuing crosslink formation (and other degradation modes) upon exposure. · 4. The validity of predictions of long term exposure performance based on short term testing.

Approval Sig 8/25/80

Date

48

r ~

1189 ~

Fl ~

r9 ~

C

F9 ~

LJ

w

LJ

Ld

Contract Title:

Contract No.:

ENCAPSULATION TASK

Illinois Tool Works Inc., Venture Group

Elgin, Illinois

Ion Plating of Solar Cell Arrays

955506

Summary of Progress from March 1, 1980 to August 1, 1980

The ITW contract was initiated in December 1979 to investigate, develop and demonstrate the capability to produce operational solar cells having metallizations and AR coatings deposited by gasless ion plating, which will separately and/or in combination with a low cost encapsulation system meet the LSA project life, cost and performance goals.

Functional solar cells have been produced with ion plated silver metallizations on both front and back surfaces. Three of these cells were brought to the 15th PIM. Since that time work has been started to produce solar cells with more economical metallization materials such as nickel, copper, titanium and aluminum. Investigations into providing a good ohmic contact to silicon have resulted in a titaniumcopper composite which when applied to n-t~e silicon has demonstrated an a~ceptable contact resistance (~1100•cm ). Efforts are continuing to develop a metallization system which will yield a good ohmic contact top-type silicon. Studies in this area have been focused on aluminum and aluminum-titanium composites.

Approval Signature 8-'l.',- 80

Date

49

ENCAPSULATION TASK

MOTOROLA INC., SEMICONDUCTOR GROUP

PHOENIX, ARIZONA

Contract Tit I e : ANTI-REFLECTION COATINGS ON LARGE AREA GLASS SHEETS


Summary of Progress for Period of March I, 1980 to September 1, 1980.

This project was performed in order to develop a method of producing inexpensive and durable antireflective coatings for use on solar module cover panels. The method researched employs aqueous solutions of sodium silicate which are applied to cover glass panels in a thin, uniform film by constant-speed withdrawal of the panels from the solution. Hardening of the sodium silicate film is then accomplished by exposure to mineral acid. The project is now completed.

~

The practicality of the method of producing coatings has been demonstrated, and correlations of production variations with optical performance have been made. Durability testing by means of abrasion, stain and dye exposure, glass cleanser exposure, water blast, and corrosive gas exposure have shown the antireflective coatings produced to possess hardness and chemical resistance close to that of glass itself.

The optical efficiency of the coatings is fair, with peak transmission values of 96.7% (for a wavelength of 0.54 µM) typical, as contrasted with a peak value of 92.7% transmission (at 0.46 µM) for uncoated glass.

8-22-80 Approval Signature Date

50

F1 ~

r, 6ul

r1 ~

F L±ll

179 Ldl1

,....,

ENCAPSULATION TASK


PHOENIX, ARIZONA

Contract Title: ANTI-REFLECTION COATINGS APPLIED BY ACID LEACHING PROCESS


Summary of Progress for Period of March 1, 1980 to September 1, 1980

This project had as its purpose the research of an established process for producing antireflective films on soda lime glass surfaces by means of a process of soaking the glass in fluosilicic acid solutions which produce a skeletonized silica layer on the glass which serves as an antireflective coating.

The project is now complete. Characterization of coating optical performance as a function of process variations was completed, and durability tests to determine chemical and physical ruggedness were also performed.

Optical performance was exceptional. 99.8% transmission of normally incident light (of 0.46 µM wavelength) was achieved on some samples, as compared to a transmission of 92.7% for light of the same wavelength through uncoated glass. Chemical durability was also very good, due to the inert silica composition of the optically active layer.

Two defects of the process were found to exist. The coatings lack abrasion resistance, and they are very difficult to produce reproducably in a commercial process. The lack of abrasion resistance makes the coating susceptable to damage not only from airborne dust and hail, but also to that caused by routine cleaning. It is possible to use the coating on the inner surfaces of concentrator units, but use on exposed surfaces is not practical. The process is very sensitive to glass and acid composition so that formation of a suitable coating is as much a matter of chance as of design. Analytical methods of controlling acid bath chemistry are either too insensitive or too expensive and timeconsuming to afford practical methods of process control.

Approval Signature

51

8-22-80 Date

Contract Title:

Contract No. :

ENCAPSULATION TASK

ROCKWELL INTERNATIONAL SCIENCE CENTER

THOUSAND OAKS, CALIFORNIA

Study Program for Encapsulation Materials for Low

Cost Solar Arrays

JPL Subcontract No. 954739

The major objectives of this study are to conduct a physical/chemical study of surface and interface degradation in solar cell encapsulant systems induced by moisture, temperature, and UV radiation. One current effort is focussed on developing corrosion monitors as nondestructive evaluation (NDE) tools for LSA life predictions. A second effort is directed toward developing corrosion models and materials selection criteria for environment and corrosion resistant interfaces.

Analytical and experimental studies are now underway to develop AC impedance measurements for nondestructive evaluation (NDE) of LSA module performance and life prediction. AC impedance measurements can isolate the series Rs and shunt RsH resistance and capacitance C of single solar cells. For multiple solar cells interconnected in modules or branch circuits the AC impedance measurement evaluates the average values and distribution functions of Rs, RsH and C within the solar array. Experimental studies have shown that AC impedance measurement can detect small reductions in Cell RsH which are predictors for longer term aging effects on photovoltaic performance as measured by photovoltaic current-voltage (IV) measurement. Analytic models for IV performance are now being combined with AC impedance modeling and aging studies. This study is leading toward new methods for correlating the frequency dependence of AC impedance with IV response.

An exploratory study has been undertaken to verify that a hydrothermal stress analysis (HTSA) computer program developed for composite laminates is applicable to solar cells. This program combines a molecular diffusion model for multilayer media with laminate calculations for internal stress. The present study shows that the HTSA computation provides reasonable predictions of cell life and mechanical failure mechanism under environmental (hydrothermal) cycling. More extensive calculations now include new material selections intended to reduce environmental effects on internal stress and thereby extend computed cell life. These computations indicate that moisture barrier coatings can significantly extend solar cell life and provide an effective basis for material selections and cell design modifications.

Approval Signature '1¥ ~'1; Jf8[)

Date

52

C

Fl ~

r:

C

LJ

Contract Title:

Contract No. :

ENCAPSULATION TASK

Spectrolab, Inc.

Sylmar, California

The Design, Analysis, and Test Verification

of Advanced Encapsulation Systems

955567

The objective of this program is to develop analytical methodology for advanced encapsulation systems which will aid in the determination of optimum systems for meeting the Low-Cost Solar Array Project goals.

During this period, optical, thermal, electrical and structural analytical models have been developed and implemented into computing algorithms. A given encapsulation system can be analyzed as to its ability to meet LSA structural and electrical specifications. Additionally, the NOCT and power of the system are calculated.

An algorithm has been developed which will use power data combined with manufacturing costs and expected durability to generate a total system life-cycle energy cost. This cost can then be used to rank encapsulation system's life cycle cost effectiveness.

53

A. Garcia

22 August 1980

Dote

ENCAPSULATION TASK

Spire Corporation Bedford, Massachusetts 01730

Contract Title: INTEGRAL GLASS ENCAPSULATION FOR SOLAR ARRAYS


This program is aimed at the development of electrostatic bonding (ESB) as an advanced encapsulation technique for terrestrial solar arrays. The electrostatic bonding process is used to join cells directly to the module front glass without use of adhesives or other organic materials. A variety of module designs combining this integral front structure with either electrostatic bonding or conventional processing for back encapsulation have been developed.

Three types of modules being developed under this program are:

• Integral front - Conventional cells are electrostatically bonded to glass sheets to provide hermetic and permanent sealing of the cells to module fronts. Ordinary encapsulating procedures are used to protect the module back.

• Preformed contacts - Silicon wafers containing p-n junctions but no front contact metal are electrostatically bonded to glass fronts. As part of this process wire screens are trapped between the glass and wafers to provide the front contact and eventually the cell to cell interconnects. Conventional back encapsulation is employed.

• Low temperature ESB Planar front surface cells are electrostatically bonded to large area glass sheets on a hot plate type of bonder. Capital equipment requirements for this process are negligible.

During the current period ESB minimodules have been fabricated. A total of 7 integral front ESB minimodules, each containing 24 series connected cells, has been made. Cell efficiency in the completed modules have consistently been above 12.5%.

One 24 cell preformed contact minimodule has been fabricated. Efficiency of multiple cell assemblies fabricated in this manner exceeds 996, and individual cell performance is significantly higher.

Samples demonstrating the feasibility of low temperature ESB have been fabricated.

54

Cl

. C C

I

LJ

Contract Title:

Contract No.:

ENCAPSULATION

SPRINGBORN LABORATORIES, INC.

ENFIELD, CT

Investigation of Test Methods, Material Properties and Processes for Solar Cell Encapsulants

954527

This program involves the evaluation of materials and processes for the encapsulation of solar ce1ls. Material selections are based on the 1986 cost objective of $3.78/M (1980 dollars).

During this quarter development efforts on solar module potting compounds were continued. Emphasis is now being placed on two materials to be used as lamination compounds (EVA and EMA) and two materials intended for use as casting syrups (polyurethane and polybutyl acrylate). The EMA (ethylene methyl acrylate) is a newly discovered resin that appears to process as easily as the EVA but may offer the advantage of higher heat stability. This resin also has inherent anti-block resistance. The greatest emphasis has been placed on the development of the butyl acrylate -casting syrup and an intermediate stage formulation is currently under evaluation in a commercial fabrication process. The gel content of the cured system has been raised, the cure temperature lowered from 90°C to 60°c: In terms of properties and processing, this syrup may become a direct replacement for the expensive silicones at very much l

1ower cost.

. All the current candidate pottants were evaluated for thermal stability using thermogravimetric analysis (TGA) and differential thermal analysis (DTA). The results indicate that the highest thermal stability is found in the EVA and EMA pottants. Test samples of all the pottants are also under continuing long term exposure to RS/4 radiation to provide an assessment of the material performance under accelerated ultraviolet exposure.

Low cost outer covers for substrate design modules are still being explored and a new biaxially oriented pure acrylic film shows promise at this time.

Adhesives and primers are also under development for the high reliability bonding of module components. A recently developed primer, the best to date, gives an EVA/glass ~eel strength of 40 pounds per inch and an EVA/steel strength of 50 pounds per inch. The primer is also effective when blended into the resin, thereby eliminating the priming step.

Bernard Baum September 2, 1980


55

ENCAPSULATION TASK

W. Pradellok and O. Vogl University of Massachusetts, Amherst, MA 01003

Contract Title: Polymeric UV Absorbers


One of the presently most important ultraviolet absorbing classes of compounds has the 2-(2-hydroxyphenyl)benzotriazole group as part of the molecule. A number of low molecular weight additives of this type, whose purpose is to arrest or retard photodegradation in polymers, are now commercially available. In particular, loss of the stabilizers over

long exposure periods by physical leaching, migration or volatilization from the polymer, decreases the effectiveness of such stabilizers and leads to more rapid deterioration of the mechanical properties of the polymers.

Recently, interest in polymeric UV stabilizers has concentrated on the synthesis of UV-absorbing functional monomers which can be readily polymerized, copolymerized or graft copolymerized to achieve a protective UV-stabilizing species with built-in polymer permanency. Such systems have been found to be effective in providing long-life UV protection for polymeric systems.

We have undertaken an extensive study with scale-up to a one pound quantity of the synthesis of 2( 5-vinyl-2-hydroxyphenyl) ben2Ptriazole and succeeded in a sequence of seven steps and an overall yield of approximately 15%. 0-nitroaniline was diazotized and coupled with LJ-ethylphenol in good yield to 2-nitro-2'-hydroxy-5'-ethylazobenzene (I). Reduction of I with Zn dust and NaOH gave 2( 5-ethyl-2-hydroxyphenyl) benzotriazole which was acetylated to the corresponding 0-acetyl compound (Ila) with acetic anhydride and pyridine. N-bromosuccinimide bromination gave in high yield the 1-bromoethyl compound ( 111) • Dehydrobromination of ( 111) was carried out with triethylamine in acetonitrile to 2{ 5-acetoxy-2-hydroxy) benzotriazole ( I Va). Finally ( I Va) was purified by short-path distillation and hydrolyzed with NaH co

3 in aqueous methanol to 2(5-vinyl-

2-hydroxyphenyl) benzotriazole (IV).

The new monomer is a powerful UV absorber. Polymerization, copolymerization and graft copolymerization with common monomers and polymers were investigated.

JPL was supplied with 285g of IV. In our efforts to transfer the technology and know-how that we have developed for the scale--up synthesis of IV, we have prepared and transmitted to JPL clear descriptions of the individual steps of the synthesis.

f!!:llfg:JL~ Date

56

~ I .

1771 ~

r ~

C

C

Fl

Ld

LJ

w

LJ

ENCAPSULATION TASK

University of Toronto Toronto, Canada

Contract Title: Modeling of Polymer Photooxidation


The objective of this study is to develop a new approach to long-term lifetime predictions of hydrocarbon polymeric materials in an outdoor environment. The method basically utilizes computer techniques to generate the time evolution of chemical change using as input a detailed photooxidation mechanism .

To this end an efficient and accurate computer simulation package has been developed which facilitates the long-term prediction of chemical change utilizing relatively small amounts of computer (real) time. Also, a generalized photooxidation mechanism (model) has been formulated based on an extensive literature search.

The accuracy of this method is limited only by the accuracy of the model used in the calculations . Accordingly it will be necessary to refine the model based on experimentally obtained data. An experimental validation program has been planned and will soon be undertaken to obtain data on the photooxidation of simple liquid alkanes. Validation of a solid phase (polymer) model similarly requires experimental data. A subcontract with EcoPlastics Limited has been let to develop the sensitive analytical probes necessary to obtain such data after short-term outdoor exposure. This includes development of a laser photolysis - gas chromatograph instrument and an automatic viscometry apparatus, both currently under construction.

/t~Jlit (~J q,r 0 Date

57

r,

In-House Program

Production, Process and Equipment Area

Jet Propulsion Laboratory Pasadena, California

The PP&E lab has continued to process four inch diameter Cz wafers for use as experimental control cells. Repeatability is being stressed. Process variability is being reduced by way of better techniques and process optimization. Several pieces of equipment which appeared to function initially have required "debugging" after a few hours of operation.

New pieces of equipment are still arriving and being set-up. The vacuum laminating equipment from ARCO Solar has been operated and is being debugged. The Quantronix laser has ariived and is being refurbished. Our Yasunaga wire saw has been operated and suffered wire breakage (three times on the same ingot). SEM investigation of the wire indicates that it was kinked during rewinding and broke due to tensipn before it entered into the abrasive silicon cutting action.

A model of the PP&E lab has been constructed at a scale of 1/4 inch to the foot. The model is being used for planning floor space for future machines and for routing facilities in a more efficient manner.

Design work has begun on a 4 ft. by 4 ft. glass module utilizing 5 inch diameter round cells. Equations are being developed to optimize the spacing of three interconnecting copper strips. The 12R losses due to the cell grid line lengths are determined by the spacing. Consideration is being given to spreading the interconnecting strips beyond electrical optimum in order to increase cracked cell tolerance as pointed out by the Engineering Area.

-~' GJ o'W:J.-, D. B. Bickler Approval Signature

59

4 September 1980 Date

Contract Title:

Contract No. :

· .. PRODUCTI.ON PROCESSES AND EQUIPMENT

APPLIED SOLAR ENERGY CORPORATION CITY OF INDUSTRY, CA.

Development of High Efficiency (14%) solar cell Array Module

955217

The objective of this program was to design and develop a large area (3"diameter) P/N solar cells with conversion efficiency of 16.5% or better at AMI and 280C. In addition tooling was designed and fabricated to produce the cells and six (6) modules. The design goal was a module with a minimum output of 90 watts at AMI and 28 C and an overall efficiency of 14%.

The objective of developing a 311 diameter, P/N solar cell with an average efficiency of 16.5% was ntore difficult than expected. Although some cells did fall within the 16.5% efficiency area, the average of the 1,112 cells fabricated was 13.5%. Considerable time was spent during this contract on the improvement of cell efficiency, but it was decided to adjust our goals and build the six (6) modules using cells in this average efficiency of 13.5% range.

The module fabrication and testing has been completed with an average power output of 75.3 watts and an overall module efficiency of 11%. This module design used one hundred twenty (120) 311 diameter P/N solar cells connected eight (8) in parallel by fifteen (15) in series. In addition there were parallel connections on the back (N) side which divided the module into three (3) series blocks. The construction was an aluminum welded "C" channel frame with riveted hold down angles to hold the encapsulated assembly in place. Tedlar, Sunadex glass, and PVB made up the encapsulated assembly.

Among the tooling developed during this contract was a semiautomatic soldering machine, a cell test fixture, and a vacuum pick up fixture. The soldering machine has proven to be most successful and saves considerable time.

The draft of the final report has been submitted to JPL. Work is continuing on the SAMIS costing and any required revisions to the final report will be made.

K.S. LING 8/27/80 Approval Signature Date

60

C

C

Ld

Ld

Contract Title:

Contract No. :

PRODUCTION PROCESS AND EQUIPMENT

APPLIED SOLAR ENERGY CORPORATION CITY OF INDUSTRY, CALIFORNIA 91745

Development of Low Cost Contacts to Silicon Solar Cells

955244

The goal of the contract is to test the technical feasibility and effective cost of a copper plating system for the production of low cost contacts in the manufacturing of solar cells of high volume.

The contract has been completed and a final plating sequence was established. The process first defines the contact pattern by a printon plating resist. Next 50A of palladium is electrolessly plated followed by a sintering at 4oooc for 10 minutes. Next nickel is electrolessly plated ( 2000A) which serves as a barrier to copper migration. This step is followed by an electrolytic copper plating (4-6 microns) and a sintering at 30ooc for 5 minutes.

One hundred and fifty five cells were delivered to JPL utilizing this contact system. One hundred and nine were 3 inches in diameter while another 46 were 2~11 diameter cells. Forty cells had a electrolessly plated tin layer to protect the cells from corrosion problems. In sintering studies these cells showed no degradation at temperatures as high as 4oo0 c for 5 minutes.

The first draft of the final report has been submitted to JPL and approved along with a separate document on the process specifications. The SAMICS format A's have been completed and submitted as a separate document. IPEG calculations show a process cost of 11.8¢/watt in 1980 do 11 ars.

K. S. LING

Approval Signature

61

August 26, 1980

Date

Contract Title:

Contract No. :

PRODUCTION PROCESS AND EQUIPMENT AREA

Bernd Ross Associates

San Diego, CA 92109

DEVELOPMENT OF AN ALL-METAL THICK FILM COST EFFECTIVE

METALLIZATION SYSTEM FOR SOLAR CELLS

955688

This summary covers work done from May 1980 through August 1980. The objective of this investigation is to study economical thick film solar cell electrodes based on screenable base metal compositions.

During the initial phase of this contract materials were ordered for base metal paste experiments on solar cell front surfaces.

Cross sections of previously reported solar cell electrodes were examined optically and by scanning electron microscopy (SEM). A new feature was found consisting of a spike with the broad base on the silicon surface, extending into and occassionally .through .. the metallization. Several attempts were made to identify the material of the spikes by SEM X-ray fluorescence. By utilization of an electron microprobe a definitive analysis of the spike was obtained, indicating the material to be silicon-copper eutectic (m.p.802°c). The metallization was fired at 550°c, far from the pure eutectic formation temperature, however the role of the lead and silver additions, depressing the temperature of formation,remains to be assessed.

Reproducibility problems have arisen in the duplication of previously reported copper pastes. Neither the rheology of the green pastes nor their fired metallurgical properties were similar to previously manufactured pastes. This is presently under investigation.


62

r, ~

F9 6d

F1 ~

w

Contract Title:

Contract No. :


KULICKE AND SOFFA INDUSTRIES, INC.

HORSHAM, PA 19044

Automated Solar .. Ma>dule Assembly Line

955287

The solar module assembly machine which Kulicke and Soffa delivered under this contract is a cell tabbing and stringing machine, flexible in design, and capable of handling a variety of cells and assembling strings up to 4 feet long which then can be placed into a module array up to 2 feet by 4 feet in a series or parallel arrangement, and in a straight or interdigitated array format. The machine cycle is 5 seconds per solar cell. This machine is primarily adapted to 3 inch diameter round cells with two tabs between cells. Pulsed heat is used as the bond technique for solar cell interconnects.

The solar module assembly machine unloads solar cells from a cassette, automatically orients them, applies flux and solders interconnect ribbons onto the cells. It then inverts the tabbed cells, connects them into cell strings, and delivers them into a module array format using a track mounted vacuum lance, from which they are taken to test and cleaning benches prior to final encapsulation into finished solar modules. Throughout the machine the solar cell is handled very carefully, and any contact with the collector side of the cell is avoided or minimized.

A lamp simulator has been used to test bonded solar cells to determine if the bonding operation had any degrading effect on the cell. I-V profile curves taken of these sample cells, before and after the bonding operation indicate no apparent effect on the electrical characteristics of the solar cell by the bonding operation.

.4v1r LS, I tr() Dote

63

Contract Title:

Contract No. :


LOCKHEED MISSILES AND SPACE CO., INC.

Sunnyvale, CA.

EVALUATION OF LASER ANNEALING FOR SOLAR CELL

JUNCTION FORMATION

DOE/JPL-955696

The goal of this contract is to evaluate the merits of high power lasers in annealing ion implantation induced damage in silicon wafers. The main thrust of the program is to carry the pulsed laser annealing technology to a system capable of single pulse annealing of 3-inch diameter or larger wafers.

For this contract, a Q-switched Nd:Glass laser capable of greater than 30 joules output is used. The laser is equipped with a second harmonic generator which yields a dual wavelength capability at A =1064nm and 532nm. The pulse width of the laser is variable between 20-50ns with a 25nnn raw beam diameter.

An optical system to homogenize the laser beam has been developed. The system minimizes hot spots in the beam, and transforms the spatial beam distribution from a gaussian to a nearly flat top profile. With the use of the homogenizing system, the anneal area has been increased from the 25mm raw beam diameter to 30mm which adequately accommodates fabrication of a 2 x 2cm cell with a single pulse.

Analysis of ion implanted/laser annealed substrates has been initiated. Rutherford Backscattering (RBS) and Transmission Electron Microscopy (TEM) data revealed colllplete silicon recrystallization and recovery from the amorphous state during laser annealing with no evidence remaining of the implant induced damage. Secondary Ion Mass Spectrometry (SIMS) data has shown that the implant profile is changed to a desirable flat top distribution while still maintaining shallow junction requirements.

Activity is underway to fabricate functional 2 x 2 cm cells using varying laser energy densities and their respective correlation with cell efficiencies.

,.,4. ·. ·) M. i~P~ · / J. s.~tzeff

Approval Signature

64

8/21/80 Date

Fl ~

Fl bid

Fl ~

F1 ~

F1 kJ

F9 ~

C Fl hd

F1 bJ

~ kJ

F1 6d

fl w F1 ~

Fl ~

LJ

LJ

PRODUCTION PROCESSES AND EQUIPMENT

MBAssociates

San Ramon, California

Contract Title: Automated Solar Module Production


This contract is a direct follow-on to contract No.954882: Automated Cell Layup and Interconnect Using an Industrial Robot. Work is well underway on this new contract which is broken in five phases.

Phase one is to improve the previously developed hardware to both increase performance and decrease cycle time to 10 sec/cell.

Phase two concerns expanding the capabilities of the existing system to detect and dispose of broken cells before preparation and to make postsolder electrical tests.

Phase three is actually the bulk of the program as it involves the automated encapsulation of solar modules. To do this we are developing three distinct pieces of equipment.

First, and most complex, is the Encapsulation Preparation Station. This machine measures and cuts to length, from roll storage, all the various encapsulation materials except the cover glass. It also guides the materials as they are laid up into the second piece of equipment, the Automated Lamination Chamber. This is a vacuum chamber with a built-in heat source to cure the encapsulants. It is of modular design in order to interface with the Encapsulation Preparation Station in a manner that would allow many chambers to be cycled simultaneously. These two machines are both microprocessor controlled using the same computer that controls the previously developed cell preparation station.

The third piece of equipment to be developed during phase three is an end effector that will enable the robot to pick up the 35, loose, inter-connected cells and place them in the chamber. It also allows the robot to handle the cover glass as well as the completed modules.

Phase four involves all remaining work (sealing, framing, etc.) required to complete the module. Phase five is the fabrication of six l'x4' modules using the above developed equipment,

Work to date (late August) has involved mainly phase one. However, we are well along with the proof-of-principle mockups for phase three.

65

2 / Au'j us(: 80 Date

Contract Tit I e :

Contract No. :


MOTOROLA INC., SEMICONDUCTOR GROUP PHOENIX, ARIZONA

PHASE 2 OF THE AUTOMATED ARRAY ASSEMBLY TASK OF THE LOW-COST SILICON SOLAR-ARRAY PROJECT

954847

Summary of Progress for Period Starting March, 1980.

This portion of the contract is concerned with specification of process control parameters and limits which will allow progress toward automation of the process sequence. The main objective of this contract is sufficient process control limit definition to permit advanced equipment prototypes to be designed for incorporation into an advance pilot line facility.

Motorola is developing a process sequence which is capable of utilizing both sliced ingot wafers or directly grown sheet. The sequence incorporates texture etching, ion implantation, LPCVD silicon nitride (for an antireflection coating, surface passivation, and as a plating mask), mechanically masked plasma patterning of the nitride, and plated metallization.

All process steps have been demonstrated, and the final report draft has been submitted. A SAMICS analysis projects a process sequence cost of less than $0.70/W utilizing RTR material as the starting material.

p

Approval Signature

66

8-22-80 Date

Fl w F1 bl

F ~

ri ~

C F bJ

F1 l@J

Fl k!

r ~

Fl

Fl

Ld

P'1

LJ

PRODUCTION PROCESS AND. EQUIPMENT AREA


PHOENIX, ARIZONA

Contract Title: THE DEVELOPMENT OF A METHOD OF PRODUCING ETCH RESISTANT WAX PATTERNS ON SOLAR CELLS


Summary of Progress for Period of March 1, 1980 to September 1, 1980

This project was performed to determine the feasibility of using recoverable wax for producing etch resistant images on silicon wafers during ~rocessing. The image production was achieved by using a letterpress printing technique as an alternative to conventional photoresist processes.

Despite attempts at many variations in printing processes - including the use of both molten wax and solvent-wax mixtures, and the use of both flat and cylindrical printing plates - the technique of letterpress printing failed to provide the consistently high image quality necessary for a successful process. The lack of image quality was found to be a consequence of the hard and non-absorbant nature of the silicon substrates, which caused masking wax to spread from its area of application as it was squeezed between the contacting wafer and printing plate.


67


MOTOROLA INC., SEMICONDUCTOR GROUP PHOENIX, ARIZONA

Contract Title: THE ESTABLISHMENT OF A PRODUCTION-READY MANUFACTURING PROCESS UTILIZING THIN SILICON SUBSTRATES FOR SOLAR CELLS


Sunnnary of Progress through September 1980.

This program called for investigation, development, and characterization of methods to establish a production-ready manufacturing process which utilizes thin silicon substrates for solar cells. These thin substrates were prepared by sawing directly to thicknesses of 8 mils and 5 mils.

A pilot line process using ion implanted phosphorus and boron layers has been specified.

Saw damage removal studies have confirmed that etching 0.0005 inch of silicon from each side of a wire-sawed wafer is sufficient to guarantee the absence of saw damage.

Pilot production lots have been processed. Over 400 wafers were run. The pilot runs confirmed the ability to process this material, provided that a compatible process sequence has been chosen.

8-22-80 Approval Signatfu.e Date

68

rm] ksJ

F1 LBJ

Fl 6J

F' ~

w l '

F1 6d

F1 ~

Fl btd

LJ

L.J

LJ

L.J

L.J

Contract Title:

Contract No. :


PHOTOWATT, INC.

Tempe, AZ

Phase II Production Process

954865

Work was performed under two separate tasks. The spray-on aluminum process was further modified to obtain cost reductions and uniformity. Initial experiments were conducted with 25 to 50 micron aluminum particles. However, this solution was found to precipitate in the spray booth and the aluminum container. Various other available aluminum sizes were investigated. The powder, with 6 to 8 micron particle size, provided the most reproducible results. Cost reduction in the aluminum spray sequence is being investigated by developing a mixture of aluminum and silicon.

Initial experiments showed no indication of warpage or aluminum spiking. In addition, the excess aluminum removal step is not needed when this technique is used.

Additional work on microwave heating has enabled us to achieve the temperature of 1100°c with +5% variations in uniformity. New concepts of multiwafer heating in a diffusion boat have provided excellent drive-in with leaky wave guide technology. Conveyerized boat diffusion is therefore considered feasible. Additional work must be done in order to verify batch-to-batch consistency.

Clay Olson 09/02/80


69


RCA Laboratories

Princeton, New Jersey 08540

Contract Title: Development of Megasonic Cleaning of Silicon Wafers


The major goal, to develop and operate a continous Megasonic cleaning and cold air drying system for processing 2500 wafers/hour and to thereby reduce the cost of solar device fabrication was achieved. The new system consists of an ammonia-hydrogen-peroxide bath ultrasonically agitated at 900 kilohertz. Wafers, loaded into 3/32-inch spaced quartz carriers, are moved on a belt past two pairs of Megasonic transducers, thereby cleaning both wafer surfaces as well as the quartz carrier. The wafers are dried in a novel, room-temperature, high-velocity air dryer in the same quartz carriers also used for subsequent annealing. Cold air drying requires about the same energy as the centrifical drying process but is gentle enough to allow wafer drying within the quartz boats. A laser scanner effectively monitored cleaning ability on a sampling basis. Particle counts established that Megasonic cleaning is approximately twice as effective as hot hydrogen-peroxidesulfuric acid ("Z") cleaning. The solar device efficiency of Megasonically cleaned devices are slightly more efficient than "Z" cleaned devices. Interestingly, analyses of processed solar cells suggest that Megasonically cleaned cells have a normal efficiency distribution while "Z" cleaning may result in a bimodal efficiency distribution. This bimodality is believed to be due to residual particulate contamination which results in significant efficiency degradation.

The following-factors, contributing to the improved cost effectiveness of the process, have been demonstrated: (1) By recirculating and filtering, the cleaning solution can be used for 100,000 wafers with only a small chemical makeup; (2) Megasonic scrubbing uniformly cleans both sides of the wafer; (3) Wafer drying by a high-velocity, roomtemperature air stream consumes little energy; (4) No transfers from plastic to quartz carriers prior to the high temperature anneal are needed, thereby reducing loss due to wafer breakage by handling; and (5) The personnel safety of the system is excellent and waste disposal has no adverse ecological impact.

With the addition of mechanical transfer arms, two Megasonic systems similar to the one developed will produce enough clean wafers for a 30 lfM/year factory. A system with about five times the wafer throughput is well within the existing teclmology; about eleven such systems per cleaning process would be sufficient for a 500 MW/year factory.

A.H. Firester a~ ~r Approval Signature

70

8/28/80 Date

F1 ~

r bJ

F1 ~

F1 bci

L.J

L..I

w

L:J

,....,

L...J

L..J

L..J

Production Process and Equipment

Science Applications, Inc. (SAi)

McLean, Virigina

Contract Title: Analysis of Cost Effective Photovoltaic Panel

Design Concepts Using Light Trapping


The objectives of this task is to summarize the results of a detailed Monte Carlo analysis of light trapping photovoltaic module designs. The data will be presented as a series of design rules relating to the physical~ optical, mechanical and thermal properties of module, particularly to their encapsulation. The contract was initiated in late June and initial work has involved documenting the basic properties of encapsuJants commonly used and those projected to be used in the future.

Light trapping is a relatively well known effect in flat plate photovoltaic panels. Science Applications, Inc. has been conducting in-house studies of trapping for some time and has evolved several novel ways to use trapping. As an outgrowth of that work SAi developed a 3D Monte Carlo ray tracing analysis for the flat plate geometry. This allows the evaluation of many design options quickly. The computer program has been validated by experiments conducted in the SAi Energy Systems Laboratory in Mclean, VA.

Trapping normally takes place due to diffuse reflection from regions between cells and total internal reflection from a single index encapsulant. Two novel trapping approaches will be studied as part of this effort:

• Intercell trapping via a high index layer between cells

• Intracell trapping from the cell grid itself

Both these and the trapping produced in customary designs will be reviewed and design rules evolved to optimize performance. Realistic assessment of factors such as optical absorption, thermal related degradation of performance, etc. will be made also in this task.

After the first task is completed, an analysis of the cost/benefit of light trapping will be made.

Appr~u Dote

71

Contract Title:

Contract No. :

PRODUCTION PROCESS & EQUIPMENT AREA

SOLAREX CORPORATION

Rockville, Maryland

Nickel/Solder Metallizati.on

954854

This program is concerned with nickel/solder metallization for low cost silicon solar cells. It comprises a technical comparison of the complex: Motorola process (involving several steps of palladium plating and heat treating prior to electroless nickel plating) with simple electroless nickel plating together with studies of some of the underlying processes which are important to the understanding of electroless nickel plating on silicon.

It has been demonstrated that nickel/solder metallization survives l,OCD hours at 85° C, 85% RH and 0.45 volt bias. Cells do not perform well under thermal cycle and thermal shock conditions from -65° to +150° C, but they do tolerate thermal shocks from -40°to +100° C.

It has been shown that nickel metallization on silicon solar cells can be sintered for at least thirty minutes at 300° C without damaging cell properties.

It was discovered that the electroless plating solution dissolves silicon dioxide and also some silicon prior to depositing nickel.

The Motorola process was found to be extremely cumbersome, and both electrical and adhesion· properties were found to be virtually identical to those obtained with single step electroless nickel plating.

~~ L. Oe&£.vv : Approval Signature

72

August 25, 1980 Dote

C

C

w.J

L.J

L...J

L.J

r,

L....I


Spectrolab, Inc.

Sylmar, California

Contract Title: Investigation of Proposed Process Sequence for the Array Automated Assembly Task - Phase II


The scope of the contract covers the investigation of technology readiness of a process sequence for the low cost fabrication of photovoltaic modules. The process sequence includes a sodium hydroxide surface preparation followed by an N diffusion using a polymeric source. The back surface field and contact are formed using screen printed aluminum paste with an HF etch and mechanical brushing step. The wafers are cleaned and screen printed on the front with silver. AR coating is applied with evaporated SiOx. Circuits are then fabricated using ultrasonic soldering and laminated in EVA with a double vacuum bag technique.

Verification runs using this process indicate that a better than 90% yield can be achieved with an average efficiency greater than 14% (AMl). Circuits have been laminated using these verification cells to produce high efficiency modules.

The contract has been completed, and verification cells and modules have been delivered to JPL for evaluation.

73

22 August 1980

Dote

Contract Tit I e :

Contract No. :

PROCESS & EQUIPMENT TASK

Spectrolab, Inc.

Sylmar, California

High Resolution, Low Cost Solar Cell

Contact Development

955725

The scope of the contract covers the development and evaluatl_Qn of forming solar cell collector grid contacts by the MIDFILMQ9 process. This is a proprietary process developed by the Ferro Corporation which is a subcontractor for the program.

The MIDFILM process attains line resolution characteristics of photoresist methods with processing related to screen printing. The surface to be processed is first coated with a thin layer of photoresist material. Upon exposure to ultraviolet light through a suitable mask, the resist in the non-pattern area cross-links and becomes hard. The unexposed pattern areas remain tacky. The conductor material is applied in the form of a dry mixture of metal and frit particles which adhere to the tacky pattern area. The assemblage is then fired to ash the photo-polymer and sinter the fritted conductor powder.

Experimental work to optimize the MIDFILM process as a means of applying solar cell collector metallization has continued. Preliminary collector grid patterns have been selected for 29 cm2 cells, suitable silver and frit candidates have been selected. Cells are being fabricated and evaluated.

N. Mardesich

74

22 August 1980

Date

PRODUCTION PROCESS & EQUIPMENT DEVELOPMENT (TASK 4)

Spire Corporation Bedford, Massachusetts O 1730

Contract Title: DEVELOPMENT AND FABRICATION OF A SOLAR CELL JUNCTION PROCESSING SYSTEM

Contract No.: DOE/JPL 955640

7 March - 22 August 1980

The objectives of this program are to design and build a single piece of equipment capable of ion implanting and pulse annealing junctions for 4-inch-diameter solar cells. Wafers will first pass under the phosphorus ion beam and then under the electron beam at the rate of 1800 wafers/hour in a cassette-to-cassette mode. The wafers are transported in vacuum by means of a "walking beam" which uses no oils and has no rubbing parts in the vacuum, thus providing a contamination-free environment.

In July 1980, detail design of all components of the electron beam annealer were completed and reviewed by JPL. The pulser consists of a vertical energy storage tank set over a processing vacuum chamber which contains the wafer transport system ang electron beam emission cathode. As of this writing, all major pieces of hardware are almost completed and final assembly will begin in late September.

The energy storage capacitors, which are high-energy-density solid dielectric tubes, are manufactured at Spire. The prototype capacitor has been completed. The pulser requires 13 units, 7 of which will be completed in mid-September.

The wafer transport system consists of a section of · straight track which contains a "Y" fork on each end. Wafers are alternately fed in on each leg of the "Y", so that vacuum lock pumpdown times do not interrupt the flow of wafers. For interim testing of the pulser, the straight section and one "Y" track are now being fabricated. This interim system will only have a throughput capability of 300 wafers per hour. The remaining "Y" and the alternately pumped vacuum locks which will permit full 1800 wafers per hour operation will be built in fiscal '81. This interim hardware will be ready for installation in the vacuum chamber in September and will be used for the anneal optimization tests.

Latest calculation of the electron beam output with a full complement of 19 capacitors is 2.9 joules/cm2 (maximum) over either a round or square area of 100 cm2. Only 1.5 joules/cm2 are required for silicon annealing, so that the pulser will be capable of doing other pulse processing such as contact sintering, pulsed diffusion and pulsed recrystallization.

Since the hardware is now all determined, IPEG estimates have been able to be made for . a production version of this machine having a double track system with a throughput capability of 30 MW /yr of wafers. This estimate is 1.8 cents per watt (1980$) for junction formation.

r Approval Signature

g/1z/~ I l Date

75

Contract Tit I e :

Contract No. :

PRODUCTION AND PROCESS ENGINEERING

Westinghouse Electric Corporation


SILICON DENDRITIC WEB MATERIAL PROCESS DEVELOPMENT

955624

Ultrasonic seam bonding tests have been carried out on dendritic web solar cells with aluminum back s urface fields. The front contact for all cells was evaporated TiPd with electroplated Cu. Two t ypes of back contact structures were t ested. In the first, the AQ, back was brushed to remove surface oxides followed by evaporation of AQ,. In the second, the excess (AQ,Si) was removed by etching and an evaporated TiPd electroplated Cu contact applied. Bond parameters have been determined and bond s trengths .have been measured using a 90° pull testo Once the bond parameters were established, no cells were broken during bondingo A statistical study has been carried out on the second type of back contact to determine bond reproducibility. Based on preliminary cost and yield estimates, the second type of back contact (TiPdCu) is preferred.

Suitable techniques and curing cycles have been developed for laminating solar panels f rom dendritic web silicon solar cells using ethylene vinyl acetate or polyvinyl butyral as the laminate material . The resulting laminations were free of bubbles and the dendritic web material (about 150 µm thick) was not cracked. A cost analysis (SAMICS) has been carried out on a 25 MW facility using ultrasonic bonding and EVA lamination incorporated into our process sequence. This analysis shows a selling price of $0.66 in 1986 (1980 $).

-~~L 1 Dafe R. B. Campbell

76

I

I I

I

LJ

LJ

LJ

b.J

LJ

LJ

LJ

In-House Program

ENGINEERING AREA


Pasadena, CA

During this reporting period the activities within the Engineering Area were reorganized for improved visibility and increased capability for technology transfer to the photovoltaic community. Emphasis this reporting period has been placed on array requirements generation, array subsystem development, array component optimization, and performance criteria and standards development.

Array Requirements

Activities in the array requirements task consisted principally of monitoring and coordination of the contracts investigating array fire resistance and electrical safety, !LC building codes, and array wind loading levels. Specific details of these contracts are covered in the individual contractors' reports. Additionally, a second PV circuit design optimization workshop was conducted at JPL on May 19-20, for photovoltaic program participants who had missed the workshop held in conjunction with the 15th PIM.

Array Subsystem Development

Work on optimum ground-mounted arrays continued with present emphasis on detailed design features including module edge treatment and gasketing, ground handling provisions, and aesthetics. A status update together with a new full-scale prototype 8 feet high by 20 feet long currenttechnology ground-mounted array is schedul~d for this PIM. The new array will demonstrate the use of the JPL optimized low-cost structure concept in the context of present intermediate load center applications and present day Block IV modules. Present day costs are being generated for quantities as small as one or two units and as large as several thousand units. Bids are presently being obtained for fabricating one of the structural beam sections using actual high speed mass-production rolling-mill tooling. The particular section is difficult to fabricate using "hand" techniques and offers the opportunity of obtaining real mass production costs for a one-mile-long minimum order.

In the area of residential array designs LSA Engineering Area personnel are completing evaluation of proposals for the integrated residential photovoltaic array development effort. Contract award is anticipated in September 1980.

77

I Date

ENGINEERING AREA IN-HOUSE SUMMARY (Continued)

Array Component Optimization

Array component optimization continued in a number of areas including module electrical insulation, hot-spot testing, array circuit design, cell environmental testing, cell fracture strength testing, encapsulant soiling, module environmental testing, and PV/Thermal module development.

In the area of module electrical insulation, an extensive series of breakdown tests on .48-mil Mylar have been completed; tests have begun on 1.42-mill Mylar. A computer code is being created to compute flaw density vs. voltage stress, intrinsic breakdown probability, and module breakdown probability.

Other activities include continued measurement of Block II and III minimodules in an attempt to define a procedure suitable for assessing the voltage breakdown probability statistics for the Block IV modules to be delivered this Fall. Past Hi-Pot testing only provided a go/no-go binary measurement.

The series/parallel effort is now focused on developing tests for determining the reverse-bias characteristics of individual cells which are shadowed or cracked in a module which is operating in the short circuit mode. Selected cells in each module are being subjected to the 100 hour hot-spot endurance test. The series/parallel final report is also in preparation.

Work continued at Clemson University on environmental testing of various solar cell types. An important fallout of the workshop held at Clemson last May is added interest shown by several of the cell manufacturers. One cell manufacturer has now considered engaging Clemson in a special cell test program on development cells, whereby some costs may be shared by the cell manufacturer. A document which will contain the Proceedings of the Cell Re'iiability Workshop is presently in preparation at JPL for distribution to LSA and the photovoltaic conununities.

LSA Document 5101-137 "Fracture Strength of Silicon Solar Cells," October 15, 1980 was distributed to the photovoltaic conununity. Fracture mechanics testing of Applied Solar Energy Corporation solar cells was completed. The data are being analyzed.

In the module soiling task, deployment of material samples at the California exposure sites for the second year of the modules soiling investigations was completed. Retrieval of samples will occur on 90 day centers. Borosilicate glass (#7809) samples are being added to all sites.

In the area of PV/T module development, several performance test methods have been identified and iterated with members of the PV/T standards subgroup. A general purpose test collector has been completed and the PV/T test site at JPL is in operation. Verification testing of the proposed test methods was initiated in August.

78

F1 ~

C

r 6:1

r bJ

ri lkJ

LJ

LJ

LJ

LJ

LJ

r,

Ld

LI

w

LJ

LJ

LJ

LJ

ENGINEERING AREA IN-HOUSE SUMMARY (Continued)

Performance Criteria and Standards

Comments on the draft version of the Interim Performance Criteria document have been forwarded to SERI. Initial industry comments ranged the full spectrum from acceptable as is, to constructive critique, to unacceptable at this time because it would adversely affect the photovoltaic industry's ability to reduce cost. This principal issue for the Array2Subsystem are the Standard Reporting Conditions, specifically 800 W/M standard irradiance condition.

The Electrical Performance Subgroup of the Array Subsystem Task Group met at Sandia on July 15-16, 1980. Draft test methods for the IV and thermal characteristics of actively and passively cooled concentrator modules was reviewed. Several of these proposed methods were reviewed by the task group during the annual meeting of the Performance Criteria and Test Standards Project in Colorado in August.

79

Contract Title:

Contract No. :

Engineering Area

Boeing Engineering and Construction

Seattle, ·wA

Wind Loads on Flat Plate Photovoltaic Array Fields

954833

The loads due to wind on an array and on its support structure strongly influence the design and ultimately the cost of the photovoltaic panels, panel and array support structure, and foundations. This contract consists of an experimental boundary layer wind tunnel test, using 1/24-scale models, of the wind forces on 8 foo·t chord flat plate photovoltaic arrays. Local pressure coefficient distributions and normal force coefficients were obtained on the arrays for a range of various parameters, including tilt angle, array separation, ground clearance, and protective wind barriers. Test data were compared to theoretical results previously reported.

The most significant result from the test is the large reduction in the aerodynamic forces on arrays interior to the array field. The array on the outer boundary of the array field protects the interior arrays from the wind. Fences, in turn, can be used to protect the arrays on the outer boundary. Other results showed that the smaller the ground clearance of the arrays, the lower the aerodynamic load. Array spacing had very little effect. Array tilt angle variation showed larger loads on outer boundary arrays with increasing tilt angle. However, the larger array tilt angles produced smaller aerodynamic loads on the arrays interior to the field.

The aerodynamic loads on the array side edges due to oblique wind are higher by several orders of magnitude than the aerodynamic loads at locations interior from the edges. Attempts to reduce these edge loads by modifying the fence and array edges are presently being conducted.

P-l'I-~ Date

80

C

Fl

LJ

Ld

LJ

Contract Title:

Contract No. :

ENGINEERING AREA

Clemson University

Clemson, SC 29631

INVESTIGATION OF RELIABILITY ATTRIBUTES AND ACCELERATED STRESS FACTORS ON TERRESTRIAL SOLAR CELLS

954929

The objective of this study is to develop test methods for evaluation of solar cells, perform investigations of factors involved in the reliability of terrestrial solar cells and develop specifications for the accelerated stress testing of solar cells. The overall program approach involves determining the reliability characteristics of currently available commercial cells by accelerated stress testing. The third year's effort also entails studying methods of second-quadrant characterization, and some preliminary experimental work on second-quadrant effects.

A microcomputer controlled tester has been constructed and is now being used routinely to measure both encapsulated and unencapsulated cells. Operation involves a shuttered ELH simulator which permits the lamps to stabilize, yet because the measurement interval is less than 1 second does not allow the cell temperature to rise above ambient. Use of the tester has uncovered previously unnoticed time dependent phenomenon associated with some cell types.

Shifting of the operating point of a cell from the first to the second quadrant of the IV characteristic results in the generation of an appreciable amount of heat. This can lead to second breakdown with resultant destruction of the solar cell and/or its encapsulation. A mathematical model has been developed which describes the experimentally observed breakdown characteristics of an unencapsulated solar cell. Work is currently underway extending the model to encapsulated cells. Experimentation with single cell minimodulesshows destruction starting at 10 v and with no modules surviving 20 v reverse bias.

Routine first quadrant stressing is continuing on 8 cell types,.

-Y /2. 2. /g-0 Date

83

ENGINEERING AREA

DSET LABORATORIES, INC. PHOENIX, ARIZONA 85029

Contract Title: SUNLIGHT AGING TESTS OF SOLAR CELL MODULES

Contract No. : BQ-713131

The accelerated aging of mini-modules was continued using DSET's Super-Maq® Fresnel-concentrating accelerated weathering machine. Through August 24, 1980, the two Block II modules have been subjected to 2,770,780 langleys of radiation, and the six Block III modules have been subjected to 1,352,920 langleys of radiation. The Block II and Block III modules have been exposed to an equivalent of 14 1/2 and 7 years, respectively, of outdoor weathering in an "average" southwestern environment.

Weekly visual inspections, monthly 35nnn slide photos, and monthly I-V measurements are used in monitoring the physical and electrical characteristics of the modules. Failure modes such as cell cracking, delamination, carbonation, and contact corrosion, as well as max power losses, non-ohmic contact, and series resistance changes have been observed during the Super-Maq® exposure program. In several cases, early detection of such failures has accurately predicted similar field failures in block series modules employed in DOE demonstration programs.

A total of 27 new mini ·and sub-mini-modules have been received for sunlight aging tests. All of the new modules have been visually inspected and photographed, and initial I-V measurements have been made. Accelerated and real-time exposure testing of these modules will begin next month.

~.L f° .-!« Aull,u_ Approval Signature

84


C r1 ~

ri b±d

w

LJ

L.J

LJ

LJ

Contract Title:

Contract No. :

ENGINEERING AREA

DSET LABORATORIES, INC. Phoenix, Arizona 85029

NATURAL SUNLIGHT SPECTRAL MEASUREMENTS

BQ-713137

DSET's Solar Spectroradiometer has been field-checked, and calibration and operational procedures have been developed. Also completed are the formatting and computer software development required to interface the 16-bit BCD output of the radiometer with the Data General Nova 3D Computer (DSET's main frame) and a "transportable" Data General MicroNova MClOO (for field operations).

All hardware and the rights to the design and construction are the property of DSET Laboratories. The computer software program required to make the data acquisition/format compatible with JPL data analysis capabilities have been developed under the auspices of this program.

The spectroradiometer is based on the following components: (1) source optics comprising a 6-in integrating .sphere and a detachable/ couple pyrheliometer with a 6° field of view, (2) a double quartz prism high resolution monochromator manufactured by Carl Leiss (Berlin), a Princeton Applied Research synchronous motor-chopper assembly, and a PAR Model 286 Synchro-Het(erodyne) phase-locked amplifier, (3) thermo-electrically cooled lead sulfide infrared and UV-enhanced silicon photodiode UV-VIS detectors, and (4) associated calibration and control equipment.

When fully field operational, the spec~roradiometer is capable of making complete wavelength continuous solar spectral measurements in the 290- to 2500-nm wavelength region for both the direct beam and global conditions--the latter at any angle of tilt from horizontal and 0° horizontal.

The spectroradiometer will be employed to measure both the direct beam and global spectrum all day on approximately the 15th of each month for two years, and all day for a complete week at each of the two annual solstices and two annual equinoxes.

85


Engineering Area

Underwriters Laboratories, Inc. Melville, New York

Contract Title: Development of Photovoltaic Array and Module Safety Requirements


Testing of a limited number of modules by the procedures described in Underwriters Laboratories {UL) Standard 790, "Tests for Fi re Resistance of Roof Covering Materials 11 has been conducted. The results obtained are being factored into a proposed safety standard for photovoltaic modules and panels now in development.

The 1978 National Electric Code {NEC) has been reviewed with regard to existing specifications that may be applied to photovoltaic arrays. Limited suggestions have been made concerning changes in the NEC that may be considered for photovoltaic arrays. Included in this work is a discussion of connections for proper functioning of ground fault detection equipment.

Sample modules have been examined with regard to conformance with UL proposals now under consideration.

,4,j L~ If ID Date

86

~ ~

F9 ~

r ~

l:l r ~

C

Fl

L.t

r

w

In-House Program

OPERATIONS AR.EA

JPL

Pasadena, CA.

Sheet 1 of 2

The Block III modules have all been delivered.

Block IV modules have been received at JPL from General Electric, Motorola, Spire, Applied Solar Energy, Solarex and Solar Power Corporation. Modules from General Electric and Spire have satisfactorily completed the qualification test program. As of August 28, 1980 no modules have been received from ARCO Solar or Photowatt. A purchase order has been issued to Spire for the procurement of additional Block IV modules.

Qualification testing of Block IV modules has proceeded during this interim. Problems in design or processing have been manifested in delaminations, cracked cells and hi-pot failures. Tests have started on one module provided by the Halcrow/World Bank project. One type of PRDA module has been retested after failing the first round. The PRDA modules suffered noticeable degradation.

The twelve continental remote sites were visited during the AprilSeptember period, and the condition and performance of the modules noted. Approximately 1000 sets of I-V data were obtained using the portable I-V data logger on the 167 modules. Three new module failures were discovered, and six modules showed new evidence of performance degradation. A reduction in the fill-factor of 5% or greater from the pre-installation characteristic is a primary criterion for designating a module as having performance degradation. To date eighteen modules have been stolen from these sites, some of which are accessible only with difficulty.

At the JPL site, fifteen modules, classed last April as badly degraded, have been removed from the field and five additional modules have been reclassified as degraded. Daily monitoring of modules at the JPL site classes the modules as satisfactory, suspicious, or degraded depending upon a comparison of the I-V curves.

Failure analysis efforts since April have centered mainly on modules returned from application experiment sites.

(continued on sheet 2)

Approval Signature

87

2-fl ~I-~ /?l?e..)

Dote

In-House Program

OPERATIONS AREA

JPL

Pasadena, CA.

Sheet 2 of 2

Modules returned from Natural Bridge National Monument include (1) Motorola, Block III, which have experienced voltage breakdown primarily between cell interconnect foil and the metal substrate, and (2) Spectrolab, Bl ock II, which have suffered glass superstrate fracture. Detailed analysis of both of these failures is still underway.

Solarex Block II modules returned from the LeRC stand-alone applications experiments at Schuchuli2 Airzona and Upper Volta in Afr ica show module failure resulting from fatigue of the metallic cell interconnects. Results of stress analysis to date support the fat i gue hypothesis, however, the failure analysis continues -in order to establish conclusively the cause.

The performance of the Solarex and Solar Power modules at Mt. Laguna Air Force Station is being closely monitored. The I-V characteristics of the individual strings and modules have been taken and modules returned to JPL for failure analysis .

One ARCO Solar residential module from the John F. Long residential installation has been provided to JPL for failure analysis. The failure mode has been reproduced; however, further work is in pro gress.

Approval Signature

88

z. 9 cl7 I 'f'cfc..>

Date

I

I

I

L...J

w

LJ

w

LJ

L.J

w

L.J

LARGE SCALE PRODUCTION TASK

APPLIED SOLAR ENERGY CORPORATION CITY OF INDUSTRY, CALIFORNIA 91745

Contract Title: Third Generation Design Solar Cell Module


The objective of this contract was to design fabricate, acceptance test, and evaluate pre-production modules complying with the requirements of JPL Document No. 5101-16, Revision A, entitled "Block IV Solar Cell Module Design and Test Specifications for Intermediate Load Center Applications", dated 1 November 1978. In addition ASEC also prepared a standardized price estimate using SAMICS for 10,100 and· 1,000 Kilowatts of solar modules. The modules were 27.38 11 x 47.24" and contained one hundred thirty six (136) silicon solar cells (3.05 11 DIAMETER). These were connected electrically, thirty four (34) in series by four (4) in parallel. This design incorporated a four (4) piece anodized ~luminum frame with a special spring loaded press fit corner fastening system. The encapsulated cell assembly was made up of 11 Sunadex 11 glass, PVB and white Tedlar. Cells were interconnected with solder clad copper mesh using a solder reflow technique for attachment. Module reliability has been improved by using four (4) additional parallel connections on the back (P) side of the cells to create a total of six (6) series blocks within the module. In addition a by-pass diode was placed across the output leads for module protection. All the modules have been delivered to JPL for qualification testing. The initial testing has been completed which established NOCT at 54.5°c and average power at 15 volts and NOCT to be 79 watts. This was somewhat lower than expected but with the fixed voltage requirement (15V) and a higher than anticipated NOCT the power is no longer taken at the maximum power point. The final design review was held on 19 August 1980. A few minor corrections have since been made to our Inspection Procedure Plan and final module testing is near completion at JPL. In addition the initial SAMICS printout has been reviewed and corrections are being made. The final engineering documentation, drawings and the final report has been submitted for JPL review and approval.

K. S. LING

Approval Signature

89


Large Scale Production

Motoro 1 a, Inc. Phoenix, Arizona

Contract Title: Design, Fabrication, Test, Qualification and Price Analysis of "Third Generation" Design Solar Cell Modules


The final design review for the Block IV intermediate load module was conducted at JPL on 25 August 1980. Module delivery requirements have been met. Final design and documentation approval awaits satisfactory completion of qualification testing at JPL.

Bruce Larson


90

C

C

C

C

.n

'L.J

Contract Title:

Contract No.:

ABSTRACT


PHOTON POWER, INC.

EL PASO, TEXAS

Sprayed CdS Backwall Cells/Panels

No. LK-720306

This paper discusses the general status of Photon Power's activities in the research, development and manufacturing fields.

In the research area the cell efficiencies (7%) and individual parameters (current, voltage and fill factor) are listed. Slides showing S.E.M. pictures of the various crystal structures for different films indicate the physical characteristics which appear to control our cell behavior.

In the development section the efficiencies obtained with our pilot line are given (ranges from 3% for large areas to 6% for 1 cm2 cells). A brief discussion of the process is given together with slides which show the scale up achieved since December 1976.

In the manufacturing field the basic concept of the new factory is discussed.

The life of the CdS/CuxS backwall cell is discussed and an indication of our results so far is given.

For the future, the scale of our operations for 1981 is outlined and a by-product of our process is revealed.

91

Contract Title:

Contract No. :


PHOTOWATT, INC.

TEMPE, AZ

Third Generation Solar Cell Module

955410

The objective of this contract is to design, fabricate, acceptance test, evaluate and provide standard costing for 30 pre-production solar PV modules which comply with JPL Document 5101-16. The accumulated output power of these 30 modules shall be in excess of 900 watts.

Block IV module construction consists of a laminate of low iron glass/ PVP/Tedlar. It's electrical subassembly is made up of 72 3-inch diameter solar cells. They are arranged in 6 parallel rows of 12 series connected cells, making the nominal electrical output (SOC) of the module 6.6 amperes at 5.0 volts. The module design also incorporates two paralleled by-pass diodes to enhance long term power capability and limit "hot-spot" problems when the module is utilized in high voltage applications. Redundant output terminations are provided via an AMP Corporation connector design.

All piece parts, documents, and initial process specifications are complete. Pre-production versions of CM-1 and QA-1 have been approved by JPL personnel. Photowatt is currently fabricating the modules. A module has been shipped to JPL for preliminary evaluation.

Clay Olson 9/2/80


92

Fl ~

C

C

LJ

w

LJ

LJ

LJ

Contract Title:

Contract No. :

Large Scale Production

SES, Incorporated

Newark, Delaware

Exchange of Information

LK-694034

SES, Incorporated has changed the design of its product to utilize larger 7.80 inch square cells. These large cadmium sulfide/copper sulfide cells are individually encapsulated in 8 inch square hermetic modules. The major features of this product are:

o Lower cost-eliminates handling and interconnection of individual small cells.

o Redundant+ and - terminals for interconnect reliability.

o Elimination of front side contacts - all connections are made to rear of cell.

Presently specified module efficiency is 3 to 4% for commercial product.

Panels for nominal 12 volt operation are fabricated with extruded aluminum support structures from 48 series cells i~ individu~l modules. The nominal peak power rating (100 mw/cm, AMD, 28 C) for these panels is 65 watts (4.1 amps@ 16.0V). Panel size is 3~" X 103".

Super Cell-Internal Circuit (Rear View) ,-wire Gnd----,

8 Positive Copper Bus ·

Pos1t1ve T erm1nal 8

Negative Terminal

0

Continuous Sheet Negative Electrode

Positive Copper Bus

L_wire Grid__J

Negative Terminal

Positive Terminal

Rear View of 7 .8" square cell

Approval Signature R. O. Johnson

93

August 11, 1980

Date

Contract Title:

Contract No. :

LARGE-SCALE PRODUCTION

Solar Power Corporation

Woburn, Massachusetts

Design, fabrication, test qualification and price analysis

of "Third Generation" design solar cell modules

955403

This contract objective is to develop a module for use in intermediate load applications. It must comply with requirements specified in JPL Document 5101-16 Rev. A. The module design incoporates a 0.72 m2 porcelainized steel substrate to promote weatherability, durability, and thermal emissivity. The cell string is composed of 72 100 nun cells with redundant integral copper clad invar buss bars to promote peak power reliability over its life. The cells are suspended in a lamination composed of EVA, fiberglass scrim and a llumar top cover to promote transmittance and protect against cell corrosion.

All modules under this contract have been fabricated. Thirteen (13) have been delivered to JPL for qualification and evaluation testing. Four (4) are undergoing qualification testing by Solar Power.

Preliminary Samis study is complete.


94

Fl ~

F1 ~

~ . '

F1 ~

C Fl ~

~ I .

r1 bd

LJ

LJ

w

LJ

Contract Title:

Contract No. :


Solarex Corporation

Rockville, Maryland

Design, Fabrication, Test, Qualification and Price Analysis of "Third Generation" Design Solar Cell Modules

955404

Solarex has built 36 modules, 18 for intermediate load center application and 18 for residential applications. Features include:

Semicrystalline silicon as the basic cell material.

Outside envelope dimensions of 63.5 cm x 120 cm.

72 9.5 cm x 9.5 cm cells arranged in a high density pattern.

Fault tolerant cell-interconnect design which allows for some cell damage with little resulting performance· degradation.

3/16" tempered Sunadex Superstrate Ethylene Vinyl Acetate (EVA) Pottant, White Tedlar moisture barrier.

Manufacturing is complete and the last lot of modules is being shipped to JPL.

The first group of modules have been measured by JPL. The electrical output average was 60.6 watts (intermediate load) and 59.5 watts (residential load) respectively, referenced to NOCT and design voltage.

The preliminary SAMICS analysis has been completed and submitted to JPL.

Approval Signature

95

August 27, 1980

Date


Spire Corporation Bedford, Massachusetts 01730

Contract Title: DESIGN, FABRICATION, TEST, QUALIFICATION AND PRICE ANALYSIS OF THIRD GENERATION DESIGN SOLAR CELL MODULES


This program was aimed at the development of a high performance Block IV module. Tasks included module design, development and fabrication, performance and qualification testing and SAMICS price analysis. Twenty modules, producing a peak power in excess of 1 kW, were delivered to JPL.

Morlule design features include:

• Dimensions, 40 x 120 cm • 152 rectangular ion-implanted cells • 97% local cell packing density • Low iron glass cover • EV A encapsulant • Stainless steel frame

This program was completed successfully with the final design review that was held on 21 August 1980. Highlights of the program included:

• Average module efficiency of 12.3% at 2soc • Average module efficiency of 10. 7% at NOCT • Spread in module performance of only 2% • Qualification of the module per JPL specifications

96

F9 ~

F1 ~

Fl ~

Pl

Ld

r,

LSA PROJECT ACTIVE CONTRACTS w

r,

w START TERM. CONTRACT TASK NAME DATE DATE NUMBER DESCRIPTION

ri SILICON MATERIAL TASK:

LJ AEROCHEM RESEARCH LABORATORY 10/79 11/80 955491 SILICON HALIDE--ALKALI METAL

r'9 PRINCETON, N.J. FLAMES

BATTELLE MEMORIAL INSTITUTE 10/75 06/80 954339 EVALUATION OF Si PRODUCTION Ld COLUMBUS, OH

I""'\ DOW CORNING 07176 10/78 954559 SOLAR CELL-GRADE Si PROCESS, HEMLOCK, MI ARC FURNACE

LJ ENERGY MATERIALS CORP. 04/79 09/80 955269 GASEOUS MELT REPLENISHMENT

n HARVARD, MA SYSTEM

HEMLOCK SEMICONDUCTOR CORP. 10/79 09/80 955533 DEVELOPMENT OF A POLYSILICON LJ HEMLOCK, MI PROCESS

A LAMAR UNIVERSITY 10/75 09/80 954343 PROCESS FEASIBILITY STUDY BEAUMONT, TX

bi MASSACHUSETTS INST. OF TECH. 04/79 03/81 955382 INVESTIGATION OF THE

r=, CAMBRIDGE, MA HYDROGENATION OF SiCl4

C. T. SAH ASSOCIATES 02/77 04/80 954685 STUDY OF EFFECTS OF IMPURITIES ~ URBANA, IL IN SILICON MATERIALS

r, SOLAREX CORPORATION 01/79 02/80 955307 ANALYSIS OF THE EFFECTS OF ROCKVILLE, MD IMPURITIES IN SILICON

bl UNION CARBIDE CORPORATION 10/75 12/82 954334 SILANE TO SILICON EPSDU

"'1 NEW YORK, NY

l=.I WESTINGHOUSE ELECTRIC CORP. 10/75 07/81 954331 SOLAR CELL GRADE SILICON

PITTSBURGH, PA DEFINITION

n LARGE-AREA SILICON SHEET:

LaJ

APPLIED SOLAR ENERGY CORP. 05/78 08/81 955089 SOLAR CELL PROCESS DEVELOPMENT r, CITY OF INDUSTRY, CA

'-=.I GORNELL UNIVERSITY 03/78 08/80 954852 CHARACTERIZATION OF STRUCTURAL,

ITHACA, NV AND CHEMICAL PROPERTIES OF SILICON

r,

CRYSTAL SYSTEMS 11/75 12/80 954373 HEM SLICING PROCESS LJ SALEM, MA

F'1 HONEYWELL, Ir,,£. 10/75 12/80 954356 DIP COATING PROCESS BLOCMINGTON, MN

LJ KAYEX CORPORATION 12/77 09/80 954888 DEVELOPMENT OF ADVANCED METHODS

ROCHESTER, NV FOR CONTINUOUS CZ GROWTH r,

KAYEX CORPORATION 03/79 09/80 955270 LOW COST CZOCHRALSKI CRYSTAL L.J ROCHESTER, NY GROWING TECHNOLOGY

r, MATERIALS RESEARCH, INC. 06/80 07/81 955676 ANALYSIS OF DEFECTIVE STRUCTURE CENTERVILLE, UT IN SILICON

u MOBIL-TYCO SOLAR ENERGY CORP. 10/75 12/80 954355 E~E-DEFINED FILM-FED GROWTH FOR

WALTHAM, MA SILICON GROWTH DEVELOPMENT r,

Ld

r, 97

L.d

. C F\ ~

START TERM. CONTRACT TASK Fl NAME DATE DATE NUMBER DESCRIPTION ~

NORLIN INDUSTRIES 01/80 07/80 955563 SLICING OF SINGLE CRYSTAL ANO CARLISLE, PA POLYCRYSTALLINE SILICON INGOTS

CJ USING MULTIBLADE SAW .

SILICON TECHNOLOGY CORP. 12/78 10/79 955131 INGOT SLICING BY I.D. SAWS OAKLAND, NJ

0 SILTEC CORPORATION 12/77 05/80 954886 DEVELOPMENT OF ADVANCED METHODS .

MENLO PARK, CA FOR CONTINUOUS CZ GROWTH

SILTEC CORPORATION 12/79 07/80 955282 ENHANCED 1.0. SLICING - SILICON f9 MENLO PARK, CA INGOTS : I

~ SPECTROLAB, INC. 08/78 01/81 955055 SILICON SOLAR CELL DEVELOPMENT,

SYLMAR, CA FABRICATION AND ANALYSIS F1

UNIVERSITY OF SOUTHERN CALIF. 01/80 07/80 955612 MEASUREMENT OF ABSORPTION u LOS ANGELES, CA COEFFICIENTS OF SILICON

UNIVERSITY OF MISSOURI 05/79 08/80 955415 DETERMINE EFFECTS OF PRESSURES

C COLUMBIA, MO OF REACTANT GASSES .

WESTINGHOUSE ELECTRIC CORP. 04/77 07/80 954654 WEB DENDRITIC PROCESS DEVELOPMENT PITTSBURGH, PA

Fl

ENCAPSULATION: bJ

CASE WESTERN UNIVERSITY 03/77 12/81 954738 SYSTEM STUDIES--BASIC AGING ANO

CJ CLEVELAND, OH DIFFUSION '

ILLINOIS TOOL WORKS 09/79 08/81 955506 DEPOSIT OF ANTI-REFLECTIVE ELGIN, IL COATING BY ION PLATING

~ MOTOROLA, INC. 02/79 04/80 955339 ANTI-REFLECTIVE COATING ~

PHOENIX, AZ

ROCKWELL SCIEN:E CENTER 03/77 12/81 954739 MATERIALS INTERFACE PROBLEM STUDY fTT!l THOUSAND OAKS, CA kJ

SPIRE CORPORATION 05/76 06/80 954521 ENCAPSULATION--MATERIALS BEDFORD, MA PROPERTIES AND PROCESSING

A SPRINGBORN LABORATORIES, INC. 05/76 08/80 954527 METHODS AND MATERIAL PROPERTIES bl

ENFIELD, CT EVALUATION

UNIVERSITY OF MASSACHUSETTS 08/79 12/81 955531 PROCESS FOR POLYMERIC ULTRAVIOLET ~ AMHERST, MA STABILIZERS AND ABSORBERS I

~ UNIVERSITY OF TORONTO 01/80 12/80 955591 PHOTOOEGRADATION MODELING

TORONTO, ONTARIO, CANADA ~ ~

PRODUCTION PROCESS AND EQUIPMENT:

APPLIED SOLAR ENERGY CORP. 11/78 07/80 955217 HIGH-EFFICIENCY SOLAR MODULE r: CITY OF INDUSTRY, CA I

APPLIED SOLAR ENERGY CORP. 10/78 04/80 955244 LOW-COST CONTACTS CITY OF INDUSTRY, CA

C} APPLIED SOLAR ENERGY CORP. 03/79 09/81 955423 LABORATORY SERVICES .

.

CITY OF INDUSTRY, CA

ARCO SOLAR, INC. 01/79 11/80 955278 AUTOMATED SOLAR PANEL ASSEMBLY f7l CHATSWORTH, CA ~

98 F1 ~

r,

L;J

r,

Ld

START TERM. CONTRACT TASK r, NAME DATE DATE NU-1BER DESCRIPTION

w BERNO ROSS ASSOCIATES 05/80 05/81 955688 ECONOMICAL IMPROVED THICK FILM

SAN DIEGO, CA SCl..AR CELL r=i

KULICKE & SOFFA, INC. 12/78 03/80 955287 AUTOMATED SOLAR MODULE ASSEMBLY L.d HORSHAM, PA

r, LOCKHEED MISSILE AND SPACE CO. SUNNYVALE,. CA

03/80 03/81 955696 LASER ANNEALING

w MB ASSOC IA TES 07/80 09/81 955699 AUTOMATION EQUIPMENT OEVELOPMEN!

SAN RAMON, CA AND MODIFICATION r')

MOTOROLA, INC. 09/77 05/80 954847 PHASE II, PROCESS DEVELOPMENT L;J PHOENIX, AZ

r"I MOTOROLA, INC. 01/79 05/80 955324 ETCH-RESISTANT WAX PATTERNS PHOENIX, AZ

LJ

MOTOROLA, INC. 01/79 01/80 955328 THIN-SUBSTRATE CELLS

r, PHOENIX, AZ

RCA CORPORATION 03/77 08/80 954868 PHASE II, PROCESS DEVELOPMENT LJ PRINCETON, NJ

,..., RCA CORPORATION 03/79 03/80 955342 MEGASONIC CLEANING PRit,.CETON, NJ

L..J

SCIEIICE APPLICATIONS, INC. 06/80 01/81 955787 ANALYSIS OF SOLAR CELL MODULE

r, t-tLEAN, VA EFFICIENCIES

SENSOR TECHNOLOGY (PHOTOWATT) 12/78 03/80 955265 POLYSILICON SOLAR CELL LJ CHATSWORTH, CA

r, SOLAREX CORPORATION 09/77 08/80 954854 PHASE II, PROCESS DEVELOPMENT ROCKVILLE, MD

t=J SPECTROLAB, INC. 09/77 05/80 954853 PHASE II, PROCESS DEVELOPMENT

r, SYLMAR, CA

SPIRE CORPORATION 01/80 02/82 955640 SOLAR CELL JUNCTION LaJ BEDFORD, MA

r, UNIVERSITY OF PENNSYLVANIA 07/79 11/80. 954796 AUTOMATED ARRAY PHILADELPHIA, PA

L..a..J

WESTINGHOUSE ELECTRIC CORP. 03/80 09/80 955624 SILICON DENDRITIC WEB PROCESS r, PITTSBURGH, PA

LJ ENGINEERING:

F"'I BOEING C0-1PANY 08/77 09/80 954833 FEASIBILITY STUDV--SOLAR DOME SEATTLE, WA Ef\CAPSULATI ON

L.J

BURT HILL KOSAR RITTELMANN 10/79 02/80 955614 OPERATION-MAINTENANCE COST DATA r, BUTLER, PA FOR RESIDEN!IAL PHOTOVOLTAICS

LJ BURT HILL KOSAR RITTELMANN 04/80 03/81 955698 COMMERCIAL/INDUSTRIAL

BUTLER, PA PHOTOVOLTAIC REQUIREMENTS

r, CLEMSON UNIVERSITY 12/77 10/80 954929 SOLAR CELL RELIABILITY TEST CLEMSON, SC

L;J

r,

Ld

l"9 99

Ld

START NAME DATE

!IT RESEARCH INSTITUTE 03/80 CHICAGO, ILL

MOTOROLA, INC. 02/79 PHOENIX, AZ

UNDERWRITERS LABORATORIES, INC. 05/79 MELVILLE, NY

LARGE-SCALE PRODUCTION:

APPLIED SOLAR ENERGY CORP. 05/79 CITY OF INDUSTRY, CA

ARCO SOLAR, INC. 07/79 CITY OF INDUSTRY, CA

GENERAL ELECTRIC CO. 05/79 . PHILADELPHIA, PA

MOTOROLA, INC. 05/79 PHOENIX, AZ

SENSOR TECHNOLOGY (PHOTOWATT) 04/79 PHOENIX, AZ

SOLAR POWER 06/79 WOBURN, MA .

SOLAREX CORPORATION 05/79 ROCKVILLE, MO

SPIRE CORPORATION 05/79 BEDFORD, MA

TERM. CONTRACT DATE NUMBER

01/81 955720

09/79 955367

10/80 955392

09/80 955409

08/80 955402

08/80 955401

03/80 955406

09/80 955403

02/80 955403

02/80 955404

02/80 955405

100

TASK DESCRIPTION

RELIABILITY ENGINEERING TECHNICAL Sl,JPPORT

STUDY OF TERMINATION DESIGN REQUIREMENTS

SOLAR ARRAY ANO MODULE SAFETY REQUIREMENTS

BLOCK IV

BLOCK IV

BLOCK IV

BLOCK IV

BLOCK IV

BLOCK IV

BLOCK IV

BLOCK IV

F1 ~

LJ

LJ

,...,

LSA PROJECT PUBLISHED DOCUMENTS*

Document No.

5040-29

Author

DOANE, J. W. June, 1976

Title

The Cost of Energy From Utility-owned Solar Electric Systems

5101-7 PROJECT OFFICE LSA First Quarterly Report - April to June 1976 ERDA/JPL-1012-76/6 October, 8, 1976

5101-10 PROJECT OFFICE ERDA/JPL-1012-77/1

5101-12

5101-13

5101-14

ZOUTENDYK, J. October 28, 1976

GONZALEZ, C. C. February 14, 1977

EDELSON, E. January 26, 1977

LSA Second Quarterly Report - July to September 1976

Progress in Silicon for Terrestrial Photovoltaic Crystal Technology'Solar Energy Conversion

Availability of Ultraviolet Radiation Data (for Encapsulation System Design)

Preliminary Analysis of Industrial Growth and the Factors that Affect Growth Rate

5101-15 CHAMBERLAIN, R. G. SAMICS (Solar Array Manufacturing Industry Costing September 1977 Standards) Workbook

5101-16 LSA ENGR. AREA Rev. A November 1, 1978 DOE/JPL-1012-78/10

5101-19 MOORE, D. February 28, 1977

5101-20 CANTU, A.H. February 28, 1977

5101-21 BISHOP/ANHALT Rev. B November 3, 1978

5101-24 PROJECT OFFICE ERDA/JPL-1012-77/2

5101-31 STULTZ/WEN July 29, 1977

5101-32 PROJECT OFFICE DOE/JPL-1012-77/3

Block IV Solar Cell Module Design and Test Specification for Intermediate Load Center Applications

Cyclic Pressure-Load Developmental Testing of Solar Panels

Test Program on Low-Cost Connector For Solar Array Modules

Acceptance/Rejection Criteria for JPL/LSA Modules

Project Quarterly Report-3 for the Period October 1976 to December 1976

Thermal Performance Testing and Analysis of Photovoltaic Modules in Natural Sunlight

Quarterly Report-4 for the Period January 1977 to March 1977

5101-33 CHAMBERLAIN/ASTER Interim Price Estimation Guidelines: A Precursor and September 10, 1977 an Adjunct to SAMIS III, Version One

5101-36

5101-39

5101-40

5101-43

SMOKLER, M. October 15, 1977

JAFFE, f. August 3, 1977

COULBERT, C. D. June 8, 1977

GRIPPI, R. A. October 7, 1977

User Handbook for Block II Silicon Solar Cell Modules

LSA Field Test Activity System Description

Development & Validation of A Life-Prediction Methodology for LSA Encapsulated Modules

Module Efficiency Definitions, Characteristics and Exijmples

* Documents with DOE/JPL numbers are available from: Technical Information Center P.O. Box 62 Oak Ridge, TN 37830 Phone: (615) 576-1304

101

Document No.

5101-44 Rev. B

5101-45

5101-46 DOE/JPL-1012-77/4

5101-51

5101-53 DOE/JPL-1012-77/6

5101-54 Vol. I DOE/JPL-1012-78/1

5101-54 Vol. II DOE/JPL-1012-78/1

5101-55 DOE/JPL-1012-78/2

5101-56 DOE/JPL-1012-78/3

5101-57 DOE/JPL-1012-78/7

5101-58

5101-59

5101-60 Rev. B

5101-61

5101-62 DOE/JPL-1012-78/6

5101-65 DOE/ JPL--1012/ 78/7A

5101-68

5101-69

5101-70 Rev. B

5101-71 Rev. B

Author

CHAMBERLAIN/ASTER Apr i 1 21, 1980

GONZALEZ, C. C. December 6, 1977

PROJECT OFFICE June 1977

PRATURI/LUTWACK/ HSU July 17, 1977

O'DONNELL/LEIPOLD/ HAGAN March 1, 1978

SMITH, J. L. April 1978

SMITH, J. L. April 1978

PROJECT OFFICE

TURNER, G. B. March 1, 1978

CHEN, C. P. February 2L, 1978

ESTEY, R. S. March 15, 1978

CHAMBERLAIN, R. G. February l, 1978

· FIRNETT, P .J. April 21, 1980

CUDDIHY, E. April 13, 1978

MOORE/WILSON

LSA ENGR. AREA March 24, 1978

ASTER, R. W. May 12, 1978

DAUD/KOLIWAD June 15, 1978

Title

SAMICS Input Data Preparation

Environmental Hail Model for Assessing Risk to Solar Collectors

Project Quarterly Report-5 for the Period April 1977 to June 1977

Chemical Vapor Deposition of Silicon from Silane Pyrolysis

Compatability Studies of Various Refractory Materials in Contact with Molten Silicon

Historical Evidence of Importance to the Industrialization of Flat-Plate Silicon Photovoltaic Systems: Executive Surmnary

Historical Evidence of Importance to the Industrialization of Flat-Plate Silicon Photovoltaic Systems

Project Quarterly Report-6 for the Period July 1977 to September 1977

Structure of Deformed Silicon and Implications for Low-Cost Solar Cells

Multi-Wire Slurry Wafering Demonstrations

Measurement of Solar and Simulator Ultraviolet Spectral Irradiance

SAMICS Usage No. 1

Standard Assembly-Line Manufacturing Industry Simulation (SAMIS) Computer Program User's Guide -Release 3

Encapsulation Material Trends Reliability 1986 Cost Goals

Photovoltaic Solar Panel Resist - Simulated Hail

Photovoltaic Module Design, Qualification and Testing Specification

Price Allocation Guidelines

Effect of Grain Boundary in Silicon Sheet on Minority Carrier Diffusion Length and Solar Cell Efficiency

CHAMBERLAIN/FIRNETT Standard Assembly-Line Manufacturing Industry /HORTON Simulation (SAMIS) Design Document - Release 3 April 21, 1980

CHAMBERLAIN/FIRNETT Standard Assembly Line Manufacturing Industry /HORTON Simulation (SAMIS) Computer Program Source Code -April 21, 1980 Re]ease 3

102

C

C

A

Document No.

5101-72

510i-73 DOE/JPL-1012-78/8

5101-75

5101-76 DOE/JPL-1012-78/9

5101-77

5101-79

5101-81 DOE/JPL-1012-78/13

5101-82 DOE/ JPL-1012- 79 /6

5101-83 DOE/JPL-1012-78/14

5101-84 DOE/JPL-1012-78/11

5101-85 DOE/JPL-1012-78/12

5101-88 JPL Pub 1. 79-14

5101-91, Vols. I-III DOE/JPL-101:.:-25 JPL Publ. 79-103

5101-93 DOE/JPL-1012-79/5

5101-94 DOE/JPL-1012-78/17

5101-98 DOE/JPL-1012-79/1

5101-99 DOE/JPL-1012-3

Author

MAXWELL, H. June 15, 1978

VON ROOS, O. May 31, 1978

SMITH, J. L. May 30, 1978

STULTZ, J. W. July 31, 1978

GUPTA, A. August 10, 1978

GUPTA, A. August 18, 1978

PROJECT OFFICE November 15, 1978

SMOKLER, M. I. November 15, 1979

LSA ENGR. AREA November 1, 1978

HOFFMAN/MILLER October 15, 1978

JAFFE, P. September 15, 1978

PROJECT OFFICE

SMITH, J. H

CHA~IBERLAIN, R. G. January 15, 1979

ASTER, R. December 1, 1978

GRIFFITH, J. S. January 1, 1979

PROJECT OFFICF.

Title

Encapsulant Candidate Materials for 1982 Cost Goals

Determination of Bulk Diffusion Lengths for Angle-Lapped Semiconductor Material via the Scanning Electron Microscope - A Theoretical Analysis

The Penetration of the International Market by Domestically Produced Photovoltaic Power Systems: A Survey of Recent Estimates

Thermal and Other Tests of Photovoltaic Modules Performed in Natural Sunlight

Photodegradation of Polymeric Encapsulants of Solar Cell Modules

Effect of Photodegradation on Chemical Structure and Surface Characteristics of Silicon Pottants Used in Solar Cell Modules

Project Quarterly Report-7 for the Period October 1977 to December 1977

User Handbook for Block III Silicon Solar Cell Modules

Block IV Solar Cell Module Design and Test Specification for Residential Applications

Bias-Humidity Testing of Solar Modules

LSA Field Test Annual Report August 1977 to August 1978

Project Quarterly Report-8 for the Period January-March 1978

Handbook of Solar Energy Data for South-Facing Sur(aces in the United States

Volume I: An Insolation, Array Shadowing, and Reflector Augmentation Model Volume II: Average Hourly and Total Daily Insolation Data for 235 Localities (Alaska-Montana) Volume Ill: Average Hourly and Total Daily Insolation Data for 235 Localities (North Carolina-Wyoming)

A Normative Price for a Manufactured Product: The SAMICS Methodology

Volume I: Executive Summary Volume II: Analysis

Economic Analysis of a Candidate 50¢/Wpk Flat-Plate Photovoltaic Manufacturing Technology

Environmental Testing of Block II Solar Cell Modules

Project Quarterly Report-9 for the Period April-June 1978

103

Document No.

5101-100 DOE/JPL-1012-4 JPL Publ. 79-16

5101-102

5101-103 D0E/JPL-l0l2-79/8A

5101-104 DOE/JPL-1012-79/1

5101-105 DOE/ JPL-1012- 20

5101-106 DOE/JPL-1012-21

5101-107 DOE/JPL-1012-18

5101-108 DOE/ JPL-1012-19

5101-109

5101-112 DOE/ JPL 1012-27

5101-133 DOE/JPL-1012-29 JPL Puhl. 79-88

5101-134 DOE/ JPL-1012-30 JPL Puhl. 79-96

5101-135 DOE/JPL-1012-31 JPL Publ. 79-92

5101-137 DOE/ JPL-1012-32 JPL Puhl. 79-102

5101-138 DOE/ JPL-1012-36

5101-139 DOE/ JPL-1012-34 JPL Publ. 79-116

5101-141 DOE/ JPL-1012-38 JPL Pub 1. 80-5

5101-142 DOE/ JPL-1012-42 JPL Pub l. 80-21

Author

PROJECT OFFICE

SLONSKI, M. L. February 15, 1979

REPAR, J. January l, 1979

GRIFFITH, J. S. January 1, 1979

PRATURI, A. K. April 15, 1979

PRATURI, A. K. April 1, 1979

RHEIN, R. A. April 15, 1979

RHEIN, R. A. April 15, 1979

PROJECT OFFICE

PROJECT OFFICE

PROJECT OFFICE

GRIFFITH, J. S. September 1, 1979

LAUE/GUPTA September 21, 1979

CHEN, C. P. Oc to b er 15 , 1 9 79

LSA ENGINEERING January 15, 1980

SALAMA, A. M. November 1, 1979

JAFFE, Peter December 15, 1979

PROJECT OFFICE

Title

Project Quarterly Report-10 for the Period July-September 1978

Energy Systems Economic Analysis (ESEA) Methodology & User's Guide

Experience with Silicones in Photovoltaic Modules

Environmental Testing of Block II Solar Cell Modules

Modeling of Silicon Particle Growth; a Progress Report

On the Modeling of Silane Pyrolysis in a Continuous Flow Reactor

Purification of Silicon by the Silicon Fluoride Transport Process - A Thermochemical Study

Silicon Preparation and Purity from the Reaction of Sodium with Silicon Tetrafluoride and Silicon Tetrachloride - A Thermochemical Study

Quarterly Report-11 for the Period October 1978-December 1978 and Proceedings of the 11th Project Integration Meeting

Progress Report 12 for the Period January 1979 to April 1979 and Proceedings of the 12th Project Integration Meeting

Progress Report 13 for the Period April 1979 to August 1979 and Proceedings of the 13th Project Integration Meeting

Environmental Testing of Block Ill Solar Cell Modules - Part 1: Qualification Testing of Standard Production Modules

Reactor for Simulation and Acceleration of Solar Ultraviolet Damage

Fracture Strength of Silicon Solar Cells

1982 Technical Readiness Module Design and Test Specification - Intermediate Load Applications

Characterization of Deliberately Nickel-Doped Silicon Wafers and Solar Cells

LSA Field Test Annual Report August 1978-August 1979

Progress Report 14 for the Period August 1979 to December 1979 and Proceedings of the 14th Project Integration Meeting

104

Fl ~

r1 ~

17'1 bBJ

,....,

,......,

,......,

..--,

L.J

Document No.

5101-143

5101-144

5101-147 DOE/JPL-1012-40 JPL Publ. 80-12

5101-148 DOE/ JPL-1012-41 JPL Pub 1. 80-34

5101-150

5101-151 DOE/ JPL-1012-44 JPL Pub 1. 80-27

5104-154

5101-155

5101-156

5101-158

5101-159

Author

PROJECT OFFICE January 1980

CUDDIHY, E. F. January 15, 1980

BOUQUET, F. L. February 1, 1980

MOORE, D. M. March 1, 1980

CHRISTENSEN, E.

PROJECT OFFICE April 1980

Tit le

Electricity from Photovoltaic Solar Cells -Status of Low-Cost Solar Array Project

Encapsulation Materials Status to December 1979

Glass for Low-Cost Photovoltaic Solar Arrays

Proposed Method for Determining the Thickness of Glass in Solar Collector Panels

Electricity from Photovoltaic Solar Cells (Black & White Update of 5101-143)

Progress Report 15 for the Period December 1979 to April 1980 and Procedings of the 15th Project Integration Meeting

CHAMBERLAIN/ASTER/ SAMICS Cost Accountable Catalog - Version 4 FIRNETT April 21, 1980

CHRISTENSEN, E. June 1980

FIRNETT, P .J. July 21, 1980

ASTER/CHAMBERLAIN/ MILLER/FIRNETT July 21, 1980

Electrical Power from Solar Cells (Update of 5101-150)

Improved Price Estimation Guidelines (IPEG) Computer Program User's Guide

Improved Price Estimation Guidelines (IPEG) Design Document

CHAMBERLAIN/FIRNETT Improved Price Estimation Guidelines (IPEG) /MILLER Computer Program Source Code July 21, 1980

105

PASAD ENA CENTER

>,

"' 3: ~ "' c..

" >, 0 :: <

Unde r -

u.J

0 z <{

"" 0

ai > < 0 en C Q)

:0 ~

gr ound + Parking

Enlrancet

PASADENA AREA

BLDG . 512 ld5 NO. ALTADENA D~.

V> 0 <{ ~

0 z <{

0 z a w w z a

BLDG 502 "" "' ....I

u.J <{ u ....I u.J ,_ ~ <{

::J ~ ....I ....I ....I 2500A E. FOOTHILL BLVD ~ <{ ....I u.. w

....I :i: <{ <{

BLDG 507 GRE N 2500B E. FOOTHILL BLVD

r---, CALIF ORNIA I I

____..

Exhibilio n

B ldg.

8 Tennis

~

Apr~

CALIFORNIA INSTITUTE

OF TECHNOLOGY

PASADENA CENTER

(; ~lo r ,11.l ,. B lvd

Gree n St. ---Conference

Civic ! B ldg. / .,.-.; >

Aud. X

~ "'-=

8 Ice Chalet

-

H o l1da y Inn G; -

C 0 rd o v ,1 S1

Del Ma· B.,d

BLDG. 506 133 NO. ALTAD ENA DR ,

L ~-

~ Q)

ai > > < < "' ~ "O ,:, u 0 ::, a: w

"' 0 ....I

• P ark 1n g

,Entr anc e

R"g1s1ra11on L ,,bby, low e1 lloor

E.ntrances

Und e1 ground parkin g h a5 en t rance lo c o nference building