AKELEUTEl3[ABBEEVSr(lTED TEST METHODS, STUDY 4 OF TASK 3 (ENCAP!SULisliTlON) OF TW E-LOW-COST SILICON SOLAR ARRAY PROJECT

U.S. Department of Energy

DISCLAIMER

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

DISCLAIMER

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

NOTICE

Available from the National Tedmbd laFomtion Senica, U. S. m-ttryt af C?9mwm, S@iqfi~td, Virg@a 221 6 1,

, - , ' I ,

- - - L L

DOE/J PL/954458-7 Distribution Category UC-63b

SEVl:WT11 QIIAII'I'ER1,Y I'ROC;RI':SS REPORT

(Covering October t l~rough Dcccml~cr 1977)

on

ACCELEMED/N<HW,VTATED TEST bETI1ODS,

snm 4 OF T A ~ K 3 (ENCAPSULATION) OF TIE .

LOW- COST ST LICON SOUR ARRAY PROJECT

JPL CONTRACT NO. 954458

This work was perfonned fo r the J e t Propulsion Laboratory, California I n s t i t u t e of Teclmology, under NASA Contract NAS7-100 f o r the U.S. Energy Research and Development Administration, Division of Solar Energy.

The JPL Low-Cost Si l icon Solar Array Project is funded by ERDA and forms pa r t of the ERDA Photovoltaic Conversion Program t o i n i t i a t e a major e f f o r t toward the development of low-cost s o l a r a r rays .

Prepared January 3 , 1978 ,__ _ _ _ _ _ _ -- . -

by

J. M. Kolyer, Principal Invest igator

and

NOTICE

sponsored by the United StstesCovcmmcnt. Neither the United Slates nor the United Stater Deparlmnt of Energy. nor m y of their cmployeec, nor nny of their ~mn~nac~uts, rubuttui~ctuna, ut ~ltclr ctt~pluyecr, 41tab.n any warranty, exprcu or implied, or assumes any legal liability or responsibility for the accuracy, completcneu or vseru!nert of any information, apparatus, product or prosen didclosed, or represcnts that t u urc would not

- - 7- -- . - - - .~ - - J

N. R. Mann, S t a t i s t i c i a n

Rockwell International \

Autonmk. Strategic Syrtmr Dlvlrion

-- Elabonlc Syotmr Group 3370 M~ra lo l i~a Avel~ut!

P.O. Box 4192 A 1 1 ~ I l ~ l n 1 . (;A 92800

DISTRIBUTION O F THIS DOCUMENT I S UNLIMITED

THIS PAGE

WAS INTENTIONALLY i

LEFT BLANK

TAl3I. E 01: CONTENTS

Page

A . F i r s t Year Study (Inherent Weatherability . of Transparent Encnpsulants) . . . . . . . . . . . . . . . 2

1 . Transparent P l a s t i c Films . . . . . . . . . . . . . . . . 2

2 . Universal Test Specimens (UTS's) . . . . . . . . . . . 3

B . SecondYearStudy . . . . . . . . . . . . . . . . . . . . . . 4

1 . Inherent Weatherability of UTS's with Nine DifferentEncapsulant-Substrate . . .

. . . . . . . . . . . . . . . . . . . . . . . . . Combinat ions . . 4

2 . Iiyperacceleration by the W Component of Highly-Concentrated Natural 'Sunlight . . . . . . . . . . . 5

IV . CONCLUSIONS AND REC8MENDATIONS . . . . . . . . . . . . . . . . . . . 5

. . . . . . . . . . . . . . . . . . . . . . V . PLANS FOR NEXT QUARTER 6

A . F i r s t Y e a r S t u d y . . . . . . . . . . . . . . . . . . . . . 6

. . . . . . . . . . . . . . . . . . . . . B . Second Year Study 6

. . . . . . . . . . . . . . . . . . . . . . . . . . . VI . REFERENCES 6

LIST OF FIGURES

Page

1. Yellowing of Lexan in Phoenix (4S0S) . . . . . . . . . . . : . . . 7

. . . . . . . . . . . . . . 2. Yellowing of Lexan in Miami (4S0S) 8 ' * '

Page

I . Retention of b1aximum Power by Solar Cells in UTS1s Exposed Outdoors . . . . . . . . . . . . . . . . . . . . 9

11. Conditions for Accelerated Exposure of UTS1s with,Nine Different Encapsulant-Substrate Combinations . . . . . 10

I I I . Electrical Data on bloisture-~e~raded solar' Cells (after 72 .days at 80°C and 100% Relative Humidity) . . . . . . . 11

IV. Conditions for Exposures Planned . . . . . . . . . . . . . . . . . . . . . . with the Solar Furnace 12

V. Transmission of Schott UG- 11 uV-~ransmitting Black Glass, 3 mm Thick . . . . . . . . . . . . . . . . . . . . . . . . . 12

To meet .the goals of the LSSA progr'am, solar cell encapsulants must provide protection for 20 years. Consequently, the objective of the present program is to develop methodology for making confident pre- dictions of encapsulant performance at any exposure site in the .Il.S.A.

During the first year of the program, inherent weatherability was studied. Inherent weatherability is controlled by the three weather factors common to all exposure sites: insolation, temperature, and humidity. Emphasis was focused on the transparent encapsulant portion of miniature solar cell arrays by eliminating weathering effects on the substrate and circuitry (which are also parts of -the encapsulant system). The most extensive data were for yellowing, which was measured con- viently and precisely. Considerable data also was obtained on tensile

I strength. Changes in these two properties after outdoor exposure were predicted very well from accelerated exposure data. This is remarkable considering that outdoor W intensity data is very limited. In addition, the feasibility of predicting an important but difficultly-measured property by correlation with an easily-measured property was demonstrated. Although more outdoor exposure data will be received, this first part of the program can be said to be successfully, concluded.

In continuation of the study, the power output of solar cells is being monitored in accelerated. test conditions and in outdoor exposures. For this purpose, Universal Test Specimens (UTS's) with nine different sub- strateltransparent encapsulant combinati'ons were prepared. Again, the - objective is to predict outdoor performance from accelerated exposure data with photochemical stresses of 8 times normal. Continuous accelerated exposure'under 8 key combinations of ultraviolet (W) light intensity, temperature, and humidity for 2 months has been completed. Solar cell per- formance, which degrades due to moisture , now ,will be forced to the failure point by exposure of the UTSts to 100 percent relative humidity at 100' C.

A subsequent objective is to accelerate degradation rates by a factor o f 100 or more. This includes the purely thermal reactions, such as hydrolysis, as' well as the photochemical reactions. The photochemical

' acceleration is the more difficultproblein. The use 'of natural sunlight avoids the problem of imperfect matching of the solar spectrum by lamps. However, it must be established how much the W component of sunlight can be concentrated on samples without changing , the degradation mechanism. Exposure of plastic films 'is planned at the Army's solar furnace .at White Sands in February 1978.

The first.yea.r study (inherent weatherability of transparent cncap- sul.kts) is nearing completibn. Successful predictions for the rate of loss of. properties. on outdoor exposure were made for plastic films based upon accelerated data'. T~CSC predictions iare summarized' in the last (sixth) Quarterly Progress Report. Degradation data for samples exposed beginning in winter, summer, or fall are discussed below. The electrical per.fonance od solar cells in UTS's exposed outdoors is also reported.

The second year study involves IJTS's with several' transparcnt encapsulant -substrate systems which are relatively rapidly degraded on outdoor exposure. The key property being followed is maximum power out- put of the solar cells. Exposure conditions and the status of the test program are discussed below.

. Of special interest is the possibility of hyperacceleration of photo- , ' chemical degradation, e.g., by 100 times, by exposing samples to the W

component of mirror-concentrated sunlight in a solar furnace. Plans for . . a trial of this principle are described.

111. DISCUSSION OF RESULTS THIS QUARTER

' . A. First Year Study (Inherent Weatherability of Transparent F~capsulants) . . ,

1;- Transparent Plastic Films ' - . _ _ . .

.- Data are .complete for yellowness (optical absorbance at 360 m)

. . and tensile strength of plastic films exposed outdoors for up to 300 days. Electrical properties.of similarly-exposed Universal Test Specimens (UTS's) also have been determined. All of these results

. . are tabulated'in the lnterim Keport, prepared on October 24, 1977.

Data from 24-condition accelerated exposure were correlated successfully with abbreviated (outdoor) data. 'Predictions are sum- . . .

: . 'mrized in the Sixth Quarterly Progress Report (for July through September 1977), and full details of the mathematical procedures were pres'ented to JPL in a letter of October 21, 1977.

Outdoor exposure of Lexan and polystyrene films was initiated in the fall of 1976. These samples are now deteriorating badly. In Miami, pol.ystyrenc exposed at 45% for 300 days (started 10-20-76) was so badly embrittled that the test was terminated. The same was true for Lexan at 379 days (exposure started 9-1 -76) or. 340 days (exposure started 12-22-76). In Phoenix, M Q U A exposure of Lexan ended due to embrittlement at 300 days and MlA exposure ended at 420 days.

l n t c res t now ccntcrs on. thc r e l a t i v c r a t e s of degradation of Lcxan exposcd beginning in f a l l o r on the winter and summcr s o l s t i c e s . Degradation rn tcs arc much higher in .summer than in winter bccause of grcntcr W intcns i ty . Thcrcfore,, seasonal data a r c necessary t o r~ Cine our ma thcma t i ca l modcl s .

Outdoor exposurc of Lcxan f i lm samples a t 45O~ ( t i l t e d 4s0, facing south) was bcgun in Phoenix and Miami on the wintcr and summer s o l - s t i c e s (December 2 2 , 1976 and June 2 1 , 1977) t o obtain seasonal values f o r modcl equation parameters. Data a re complcte for the absorbance. of thcse films a t 360 nanometers: up t o 300 days exposurc f o r those s t a r t e d on the winter s o l s t i c e and 150 days fo r those s t a r t e d on the .summer s o l s t i c e . Results a re shown graphical ly i n Figurcs 1 and 2 .

The "exposure factor" i s based upon temperature levels experienced by thc samples and, more importantly, the cumulative u l t r a v i o l e t l i g h t energy received. Several assumptions were necessary in calcu1atin.g the exposure . factor . . These a re given, along with monthly values f o r the factor. a t Phoenix and Miami,,in .the l e t t e r of October 2 1 , 1977 from J. M. Kolyer t o H. E . Marsh of JPL. .

There appears t o be a tendency f o r the absorbance values t o con-, verge a f t e r about 300 days whether exposure began i n f a l l , midsummer, o r midwinter. The data points of a Weibull p lo t i n general suggest an exponential model (slope = . I ) , but a rigorous analysis remains t o be made. An exponential model is a t t r a c t i v e because it represents a simple photochemical react ion. However, some other mode1,such a s the lognormal, may eventually give the best f i t .

2. Universal Test Specimens (UTS's)

Power loss of the s o l a r c e l l s in the s i l i cone rubber-encapsulated UTS's of the f i r s t year study was a t t r ibu ted t o moisture-induced l i f t i n g of the contacts (grid l ines ) from the s i l i c o n surface (Reference 1 ) . Degradation was severe (average re tent ion of o r ig ina l power f o r the s i x c e l l s of a UTS = 57 percent) a f t e r exposure i n the dark a t 80°C and 100 percent r e l a t i v e humidity f o r 72 days. Table I shows t h a t a t 300 days the c c l l s i n UTS's exposed outdoors a l so have shown s ign i f i can t power l o s s , especia l ly i n Miami (natural exposure t i l t e d a t 4S0 facing south) and i n Phoenix on the EMMA sunlight- concentrator. Faster degradation i n bliami than Phoenix was expected because of Miami's moist climate. However, E3BIAQUA exposure, i n which' samples a r e sprayed with water 8 minutes each hour, would have been expected t o be more severe than EMMA exposure, which involves no spraying. The UTS's exposed outdoors f o r 420 days now have been returned from the t e s t sites but must be e l e c t r i c a l l y t e s t ed by the s o l a r c e l l manufacturer.

.. . B. Second Year Study

1. Inherent Weatherabil i t y of' lJTS1 s with Nine Different Encapsulant-Substratc Combinations

The nine ar ray systems are described i n our Sixth Quarterly Progress Report.

Outdoor exposure of UTS's beg& a t Florida (4S0S) on October 31, 1977 and a t Phoenix (450S, W, M Q U A ) on October 23, 1977.

. . Accelerated (xenon lamp) exposure was continued f o r '61 days. The

e ight exposure conditions a r e given in,Table 11. The property monitored was current ( in milliamperes) produced by the s o l a r c e l l s a t 0.350 vo l t s . The "power point", which i s t h a t point on the IV curve a t which the prod- uc t of current and voltage is a t a maxirm, 'occurs f o r our s o l a r c e l l s i n t h e - v i c i n i t y of 0.350 vo l t s . Therefore, current a t 0.350 vo l t s is a good r e l a t i v e measure of the m a x i m power produced by the c e l l . That t h i s statement is t rue was demonstrated by examining IV curve data obtained by Optical Coating Laboratory, Inc. on s o l a r c e l l s before and a f t e r moisture-induced degradation of the contacts . Table I11 shows t h a t the percent of o r ig ina l power calculated from data f o r current a t 0.350 v o 1 . t ~ , . is 'very close t o t h a t found from the ac tual power point da ta .

The measurement of a s ing le point on the IV curve without locat ing the exact power point s implif ied the experimental procedure so t h a t we ~ o u l d obtain several exposure time points f o r mathematical modeling. Tape cables from the UTS's i n the xenon lamp cabinet were successively plugged, by means of an edge connector, i n t o a ro tary switch, allowing convenient measurement of the s i x s o l a r c e l l s on each UTS. A l l da ta were expressed in terns of a reference c e l l a t a f ixed distance from the 'lamp.

For UTS's ocposcd outdoors, both mi l l iwat ts n t thc poticr point nnd milliamperes a t 0.350 vo l t s w i l l be monitored.

As done previously f o r p l a s t i c f i l m proper t ies , s o l a r cell perfor- mance a f t e r outdoor exposure w i l l be predicted on the bas is of 'accelerated data. Also, observation o f failure'modes w i l l be of specia l i n t e r e s t because the ar ray systems were designed with several possible "weak links".

A d i f f i c u l t y i s t h a t the c e l l s used in t h i s test, unlike the c e l l s .,attached t o the e a r l i e r ( s i l icone n1b)ner enc.apsulated) UTS s , have a

' uniform coating of solder on the Ti-Ag gr id lines (contacts). Therefore, . moisture penetrat ion w i l l take considerably longer. I t is moisture pene-

t r a t i o n t h a t causes hydrogen evolution by the Ti-Ag couple, with accom- panying separat ion a t the Ti-Si in ter face . The r e s u l t is l i f t i n g of the contacts and. decreased power (Reference 1) .

IV.

The current at 0.350 volts for each of the 144 solar cells in the accelerated exposure (6 cells on each of 24 UTS1s) was determined after 0, 2, 5, 9, 19, 40 and 61 days exposure. Results are being calculated and will be given in the January 1978 Monthly Progress Report. Obser- vations on encapsulant degradation also will be reported. To accelerate failure by moisture-induced degradation of the contacts, which are relatively resistant in this test as mentioned above, UTS1s now are being exposed to steam at 1000C. Resulting data will indicate whether or not ionic degradation products from the deteriorated encapsulants can accelerate corrosion of the solar cell contacts.

2. Hyperacceleration by the W Component of Highly-Concentrated Natural Sunlieht

Experimental work is planned for the week,of February 13, 1978 at the Army's solar furnace at White Sands, .New Mexico.

It is planned to expose polystyrene and Lexan films, along with novel actinometer films provided by A. Gupta of JPL. Conditions are listed in Table IV.

The visible/infrared regions of the solar spectrum must be removed to avoid overheating the samples. Schott W-transmitting black glass filters have been purchased for this purpose. The transmission at several wavelengths was checked and found to.be close to typical values given by the manufacturer (Table .

At White Sands, it is planned to cool the heat shock-sensitive Schott glass with flowing water. However, if this glass should break, an aqueous solution of nickelous and cobaltous sulfates (Reference 2) in a Vycor cell will be used as the filter.

The objective is to obtain precise data on absorbance at 360 nun for the films which will make possible graphical plots. These plots will disclose the intensity level at which the mathematical model for degradation changes. This level will be the upper limit at which the hyperaccelerated method represents normal exposure.

CONCLUS IONS AND REC-TIONS

1. Conclusions regarding prediction of outdoor exposure data from accelerated exposure data were given in the last (Sixth) Quarterly Report. These conclusions were drawn from the first year study on inherent weatherability of transparent encapsulants. The second year study is in progress and no conclusions are possible at this point.

2 . It is recommended that multicondition exposure, as used successfully in the predictions of the first year study, be applied in the final evaluation of candidate encapsulants. Relatively precise lifetime prediction will be required to single out the most cost-effective materials.

V. : PLANS FOR NEXT X A R m , - . . , . . .

. A. ..First Year Study , . . . . . . . . . , .

I :

.-.. .:. :. .I.. Continue to monitor outdoor .exposures of UTS1s .and films. . : .Estjmate seasonal degradation .rates.and refine mathematical models

. . ' 2. Examine weathered ~edlar and polystyrene films by Fourier transform -. : ' .

. - attenuated total .reflectance (ATR) infrared spectroscopy. The . . . . quantification.of subtle changes in a weather-resistant plastic such as Tedlar is essential to prediction of its lifetime.

B. ' Second Year Study

, : ., . 1. .Moni$or .t,he maximum power of solar cells on UTS's expo'sed to steam '

at 100°C in the accelerated %test. . Also monitor maximum power of soi.ar cell s i,n I ~ S 's exposed o~mloors. Predict. ret.ent.i.on of power

. . 0utput.b~ the cells on UTS's exposed outdoors from the accelerated , - power data. : Detcminc whe~her,corrosivc products from cncapsulernt

degradation accelerate the rate of moisture..attack on metallization (grid lines) on the cell.

, , .: . , . . 2.. txpose films to the',, w c6mponent of suniight at high con-

centrations using the Army's solar furnace at White Sands, New Mexico. .. By mathematical .modeling of data, determine the.upper light intensity limit at which the degradation mechanism remains unchanged and normal exposure is validly represented. If successful,

. . this technique could be.of.wide application in the testing of materials. such as plastics and cpatings.

t .

V I . R E E M a S

1 Interim Report, Thi+s Contract (.JPL Contract No. 954458) , October 24, 1977, page 10. . ' ,

2. L. 'R. Kollcr, Ultraviolet Radiation; Second Ed'ition, Wiley and Sons, 1965, page 174.

1:igurc 1 . Yellowing of I,cxan in I'hocnix (45"s)

log 1 0 f lixposurc ]:actor)

J.'igurc 2 . Ycl lowing of I.cxrln i n Miami (4S0S)

0 lixposurc st;^ rtctl !)/I /70

, lixposurc s t ; ~ r t e d 0/21/77

TABLE I

Retention o f b4aximum Power by Solar Cel l s

i n UTS's Exposed Outdoors

EMb/IAQllA

9 7

101

-

100

-

97

9 9

9 6

92

EMMA

101

9 9

-

9 5

100

97

8 8

83

Exposure Time , Days

5

10

15

30

60

90

150

210

300

Miami, 45's

100

100

9 7

98

93

9 9

9 9

85

7 8

Phoenix, 45's

103

108

102

103

87

9 0

94

9 0

9 5

,Conditions f o r Accelerated Exposure -

of UTS's with Nine Dif fe ren t . s . . .

. . . .

Enca~su1ai t ' -Subs t ra te ~ombinat. ions

Condition ,

No. . .

1 . . .

2

3 , ,

4'

5 . .

6 .

7

8 ' 0 . .

W In tens i ty Relat ive t o

.. Noon 'sunnier :

Sunlight

0'. 66

K.UU

1.00

1.00

1 .oo

0.66

0

Encapsulant . Temperature,

O c . .

2 8

4 3

4 3

7 2

7 2

6 4

72

--

Relat ive 1-tumidity ,

. . %

100

0

100

0

100

'100

0

. .. l'Al3Ll! 1 I I'

, . ~. ' ' . .

l l c c t r i c a l Data on b!oisture::~e~raded ~ o l a r ' c e l l s

(After 72 Days a t 80°C and 100% ~ e l a t i v e Humidity)

-

Cell No.

1

2

3

4

5

6

Cell No.

1

2

3

4

5

6

Power Point Af te r Exposure,

% of Or ig ina l

5 3

56

4 1

89

46

4 7

Current a t 0.350 v o l t s

Af t e r Exposure, : of Or ig ina l

51

56

4 1

8 7

4 3

4 3

Power Po in t , m i l l i w a t t s

Or ig ina l

17.7

17.7

17.9

17.4

15.1

18.7

Af t e r Exposure

9.33

9.98

7.38

15.4

6.92

8.72

Current , mill iamperes, a t 0.350 v o l t s

Or ig ina l

50.5

51.0

50.5

43.4

42.3

54.3

After Exposure

25.7

28.4

20.7

42.8

18.3

23.5

TABLE IV

Conditions f o r ~ x p o s u r e s Planned - .

" with t h e So la r Furnace .

*A f i l t e r w i l l t ransmi t only the UV component t o prevent overheating o f samples by v i s i b l e / i n f r a r e d l i g h t .

Sunl ight Concentration, suns *

25 '

50

100

' 200

300 - 400

TABLE V'

Exposure Time, hours

0 .2 , 0 .4, 0 .8, 1 .6 , 3..2

I I

I t

1 1

1 1

T I - a n s ~ ~ ~ i s s i u r i of bchott UG-11 -

W-Transmit t ing Black Glass. 3 m Thick

-12-

*U .S. GOVERNMENT PRINTING OFFICE a 1978-740-306/94,24. Region 4..

Wavelength, nm

300

320

360

500

800

1000

Percent External Transmission

Schott ' s Data, Typical - ----

65

7 3

63

<o. 001

0.001

(0.001

Found f o r Melt No. 745922 --

6 8

7 4

6 1

0 .1

0.2 -