AD-753 410

Degradation Mechanismsof Pigmented Coatings

Ohio State University

prepared for

Air Force Materials Laboratory

OCTOBER 1972

Distributed By:

National Technical Information ServiceU. S. DEPARTMENT OF COMMERCE52'05 Port Royal Road, Springfield Va. 22151

DE1GRADATtOt4 WCI1AMSK4S Of ýPl GEN~tTD C1AYMNg~

0xi

Appr'-*ved for, public rel.ease; distribution unlimited.

NATIONA'LL FECHNICAL

Air foarce tKsterials L~aboratoryAir Force Zystems Command

lgi-ii1-1,Patterson. Air Force B e., Ohio )4.5433

When Governume*r& 2'27 'i'i. roh&-:&t;

Supi. ed~ ~ ':, 4CCfi(tA~2' or isc~ &J, & ot to be4Jpre'Aitn or tlr40an i-n awy marmer iteenz±ug : te

rtidtl~c r ;persaonr o)r ýcorpo r, tdr, or contIng axW righso4 r' T t~~r'' l4l~r. ' ýr r Zell0 r3 .e'any.+aAte 4nve in a4& tlz may"

'Ide op r~ tbi c i.!r'led si 'l ne~ o t ber re

ýje-. 4, cu)s-drt os mtulolg t O'oieo

UNCLASSIFIEDSecuntY Classification

DOCUMENT CONTROL DATA - R & D(Stcutity classiflication of tfltl, body of abhtract and indexing annotation muxt be entered when the overmji report is classilled)

SORIGIN ATING ^CTVITY (Colpoeai. &,thot) iZi. REPORT SECUJRITY CLASSIrICATION

Ohio State University Research Foundation U ;NCLASSIFIED1314 Kinnear Rd. 2b. GRo.P

Columbus, Ohio 43212 N/A3 IREPORT TITLE

DEGRADATION M4ECHANISMS OF PIGME1NED COATINGS

4 DESCRIPTIVE NOTES (~'pt o lreport and inclusive da:ea)

Final I January 1971 - '0 Auimst 1972S. AU THORIS) (•irst no", middle initieS. least name)

William Be CampbellJoe K. Cochran

4. REPORT OATE 7I. TOTAL NO. OF P %GES ]7b. NO. OF RVFS

D M.CONTRACT OR GRANT NO 9,.ORiGINATOW'S REPDORT NUMBE'R($|October, 1972 88 97OTNUBRS

F33615-71-C-1257Pb.PROJECT NO.

7342C. -b. OTHER REPORT NO(S) (Any other numbers that may be assigned

Task No. this report)

d. 734202 A,•4L-TR-71-42, Pt II10. DISTRIBUTION STATEMENT

Approved for public release; distribution unlimited.

%I. SUPPLEMENTARY NOTES $2 SPONSORING MILITARY ACTIVITY

Air Force Materials Laboratory (MBE)TECH, OYHER Air Force Systems Command (AFSC)

A Wright-Patterson ArB Ohio 4541313 ABSTRACT Oxygen transport in polymonomethylsiloxane was investigated and compared topolydimethylsiloxane properties. The effects of rutile pigmentation on the permeabil-ity, diffusion, and solubility of oxygen through polymonomethylsiloxane were investiga-ted. Permeability and diffusion constants decreased with increasing pigment concentra-tion and there was no evidence of oxygen sorption on the pigment.

Relative adhesion of polydimethylsiloxane and polymonomethylsiloxane onrutile was predicted from wate: contact angles. Polymonomethylsiloxane was proposed thave the greater adhesion but was small in either case.rui..adThe stability of dimethyl and monomethyl polysiloxanes pigmented withfutile and zinc oxide was evaluated in a simulated solar ultraviolet environment.Oxygen transport properties of the selected polymers were not observed to limit damage

of eith pigment. Degradation of silicone/rutile paints was attributahce to independentloptical danage in the pigment and polymer. The silicate-coated rutile exhibited smallreflectance changes as reported for other silicate coated pigments.

Ultraviolet degradation of zinc orthotitanate powders exhibited no depend-ence on either pigment particle size or quenching temperature. The initial solarabsorptance increased with increased quenching temperature.

FORM A*DD INOV ,',473 UNCASSIFIED, ,,,o,, 0%, UC•/ a -nolz' i oD

1Jraif p'Secur~ty Classification

14. KEY WORDS LINK A LINK 1 LINK C __

ROLE WT ROLE WT ROLE WT

U.V. Space Degradation -

Thermal Control Coatings

Zinc OxidePolymonome thylsiloxane

Polydimethylsi loxane

Diffusion in Polymeric Coatings

Silicone Rutile Paints

Zinc Orthotitanate

Rutile

Solar Absorptance j

Unclassif iedSecuraty Clasification ,

AFML-TR-71-42

PART II

DEGRADATION MECHANISMS OF PIGMENTED COATINGS

William B. Campbell, Ph.D.Joe K. Cochran, Jr., Ph.D.

H[i

44

[IEl[

* N o

'i Approved Cor public release; distribution unlimited.

4.

44•

,14

FOREWORD

This report was prepared by Ohio State University Research Foundationunder Air Force Contract F33615-71-C-1257. The contract was initiatedunder Project No. 7342, "Fundamental Research on Macromolecular Materialsand Lubrication Phenomena," Task No. 734002, "Studies on the Structure-Property Relationship of Polymer Materials." The work was administeredunder the direction of the Elastomers and Coatings Branch, NonmetallicMaterials Division, Air Force Materials Laboratory, with Dr. William L. Lehnserving as Project Engineer. This report describes work conducted betweenI January 1971 and 30 August 1972. This report was released by the authorsin September 1972 for publication as a Technical Report.

This technical report has been reviewed and is approved.

'WARREN P. JuVlWS, ChiefElastomers and Coatings BranchFcametallic Materials DivisionAi. Force Materials Laboratory

Al,!1

ABSTRACCT

j Oxygen transport in po]ymonorae-tlylsiloxane was investigatedand compared to polydimethylsiloxane properties. The effects ofrutile pigmentation on the permeability, diffusion, and solubil-ity of o4gen through polymonomethylsiloxane were investigated.Permeability and diffusion constants decreased with increasing,pigment concentration and there was no evidence of oxygen sorp-tion on the pigment.

Relative cdhesion of po:lydimethylsiloxane and polymonome-thylsiloAane on futile was predicted from water contact angles.Polymonomethylsiloxane was proposed to have the greater adhesionbut was small in either case.

II The stability of dimethyl and monomethyl poJysiloxaxespigmented with rutile and zinc oxide was evaluated in a simu-lated solar ultraviolet environment. Oxygen transport proper.ties of the selected polymers were not observed to limit dBmageof either pigment. Degradation of silicone/rutile paints wasattributable to independent optical damage in the pigment andL polymer. The silicate-coated rutile exhibited small reflectance

changes as reported for other silicate coated pigments.

Ultraviolet degradation of zinc orthotitanate powdersexhibited no dependence on either pigment particle size orTuenchimg. temperature. The initial solar absorptance increasedwith increased quenching temperature.

LI

:i

iii

Hff TABLE OF COWTETS

SECTION PAGE

I Introduction 1

II Technical Discussion 3

1. Solar Reflectance and Absorptance 32. Flight Evaluation of Ccatings 33. Degradation and Damage Mechanisms 5

a. Zinc Oxide 6Sb. Rutile 7

c. Zinc Orthotitanate 7d. Polymethylsiloxanes 8

A 4. Pigment Surface Stabilization 85. Mass Transport in Polymers 9

a. Polydimethylsiloxane 12b. Effect oif Pigmentation 13

F1 6. Surface Energies, Contact Angle, and Adhesion 15

III Experimental Procedure 19

1. Materials 19

a. Rutile Pigment 19ID b. Zinc Oxide Pigment 19c. Polydimethylsiloxane (PIMS) 19d. Poblymonomethylsiloxane (RMMS) 19

2. Vxygen Transport 20

a. Measurement 20t b. Sample Preparation 20

3. Contact Angle 21

S[j a. Measurement 21b. Sample Preparation 21

4 t. In Situ Reflectance 23

a. Measurements 23b. Rutile Pignent and Paints Preparation 23c. Zinc Oxide Pigment and Paints Preparation 24

Preceding page bilankv

: 1'

TABLE OF CONTENTfS (Continued)

SECTION PAGE

5. Zinc Orthotitanate Pigment 24

a. Pigment Formvlation 24b. Puticle Size Reduction 26c. Heat Treatment 28d. irr',diation Powder Samples 28e. In Situ Reflectance 28

IV Results and Discussion 31

L. Oxygen Transport 31I a. Polymethyloiloxanes 31b. Effect of Pigmentation 31

2. Contact Angle

3. Silicone/RutiJe Optical Properties 48

a. Solar Reflectance 48b. Induced Solar Absorptance 54

4. Silicone/Zinc Oxide Degradation 54

5. Zinc Orthotitanate 58

V Conclusions 67

VI References 69

APPENDIX 77

vi

j [LIST OF ILLUSTRATIONS

Figtue Page

1 Typical Ultraviolet Degradation of (a) Zinc Oxide,(b) Rutile, and (c) Zinc Orthotitanat, 4.t FiveSuns In' ensity 4

2 Typical Pressure-Time Curve for Measurement of

Gas Permeability and Diffusion 11

3 Surface Energy/Contact Angle 16

4 Contact Angle Measurement Apparatls 22

5 Flow Chart for Preparation of Zn 2 TiO4 Pigments 25

6 Relationship between Particle Size and Ball Milling Time 27

I 7 Annealing Curve for Zinc orthotitanate Powders 29

8 Oxygen Sorption Isotherms at 250 C 32

9 Temperature Dependence of Permeability Constants forOxygen in Polymethylsiloxanes 34

10 Temperature Dependence of Diffusion Coefficients forOxygen in Polymethylsiloxanes 35

11 Temperature Dependence of Solubility Constants forOxygen in Polymethylsiloxanes 36

12 Temperature Dependence of Permeability Constants forSPol~monomethylsiloxane/Rutile Compositions 38

13 Temperature Dependence of Diffusion Coefficients forf Oxygen in P io3 omethylsiloxane/Rutile Compositions 39

14 Temperature Dependence of Solubility Constants for[ Oxygen in Polymonomethylsiloxane/Rutile Compositions 40

15 Effect of Pigmentation on Oxygen Solubility in Polymono-methylsiloxane 43

16 Effect of Pigmentation on Oxygen Permeability in Poly-

monomethyls iloxane 44

17 Effect of Pigmentation on Oxygen Diffusion in Poly-monoraethylsiiloxane. Dashed lines are equation (35) 45

vii

LIST OF ILLUSTRATIONS (Continued)

Figure Page

18 Normal and Inverted Water Drops 47

19 Polished Rutile 49

20 Reflectance Spectrum of Silicate-Coated Rutile 50

21 Reflectance Spectrum of Polydmmethylsiloxane/20 PerCent Rutile 51

22 Reflectance Spectrum of Polymonomethylsiloxane/20 PerCent Rutile 52 •

23 Effect of Pigment Concentration on Solar Reflectanceof Rutile/Silicone Paints 53

24 Effect of Pigment Concentration on Induced SolarAbsorptance in Rutile/Silicone Paints 56

25 Ultraviolet Degradation Rates of Zinc Oxide Paintsand Pigment 59

26 Rate of Re3lectanaee Decrease at 2.05 Microns for Ultra-violet Irradiated Zinc Oxide Paints and Pigment 60

27 Rate of Reflectance Decrease at 550 nm for UltravioletIrradiated Zinc Oxide Paints and Pigment 61

28 Effect of Temperature on Particle Size of ZnTiO4Powders 63

29 In Situ Solar Absorptance as a Function of Temperature 65for ZnTiO4 65

viii

J LIST OF TABLES

Table Page

I Oxygen Transport Constants for Polydimethylsiloxane/jSilica 12

II Rutile Pigment Composition 19

III Compositions of Polymonometbylsiloxane Pigmentedwith Rutile 20

IV Compositions of Polydimethylsiloxane Pigmented withRutile 24

V Surface Areas and Particle Size of Zn 2 TiO4 Pigments 26

VI Oxygen Transport in rolymethylsiloxanes 33

VII Arrhenius Constants for Oxygen Transport in Poly-methylsiloxane 37

VIII Arrhenius Constants for Oxygen in Polymonomethyl-siloxane/Rutile 41

IX Water Contact Angles 46X Silicone/Rutile Solar Absorptances 55

XI Solar Absorptance of Zinc Oxide Paint and Pigments 57XII Zinc Orthotitanate Surface Areas 62

SXIII Zinc Orthotitanate Particle Sizes 62

XIV Solar Absorptance of Zinc Orthotitanate Pigments 64

XV Effect of Pressure on Oxygen Transport Constants inPolydimethylsiloxane at 25°C 76

1 XVI Effect of Pressure on Oxygen Tran,•:,rt Constants inPolymonomethylsiloxane at 250 C 77

SXVII Oxygen Transport Constants for Oxygen in Polymono-methylsiloxane/Rutile Compositions 78

ix

I DEGRADATION MErCHANISMS OF PIGMENTED COATINGS

I. INTRODUCTION

Stringent temperature control for spacecraft is necessary to pro-tect man and temperature-sensitive electronic equipment. The exchangeof radiant energy between the vehicle and space, the principal meansfor temperature regulation, is controlled by the optical properties of

3 the surface of the spacecraft. For space travel near the earth, coat-ings have been applied with a low solar absorptance (a.) to infraredemittance (e) ratio. Thermal control coatings are white and consist ofoxide pigments of high index of refraction dispersed in a polymericbinder of low refractive index. Unfortunately, coatings of this typeare damaged when exposed to high energy bombardment (solar ultravioletradiation, protons, and electrons) in space. The damage produces anincrease in the solar absorptance of the paint.

Much work has been directed toward identification of the degrada-tion mechanisms. If the causes of degradation were understood, coatingswhich are stable or, at least, degrade in a predictable manner, mightbe developed. Degradation has been observed in the pigment and thepolymeric binder; however, binders now exhibit minimal degradation and

Sthe major goal remains to develop a stable pigment.

Radiation produced defects which absorb photon energies in the solarspectrum are the cause of optical degradation in oxide pigments. Theidentification of the absorbing defects has received considerable atten-tion. For the purpose of this investigation, the exact absorbing speciesis unimportant but in all cases the formation of absorbing species isii accompanied by the photodesorption of oxygen. Thus, the oxygen trans-port properties of polymer encapsulants will influence pigment stability.

f In some cases, polymeric binders have substantially influenced thedegradation of thermal control coatings but the interaction of the pig-ment and the polymer has not been established. Suggestions of pigmentsurface passivation by the encapsulating medium has warranted considera-tion of the pigment-polymer interface stability. The surface conditionof the pigment has been regarded as being a prime consideration in under-

Hl standing degradation.

The purposes of this investigation are to determine pigment degra-dation dependence on (1) oxygen transport in typical polymeric binders,(2) pigment/polymer surface compatibility, (3) pigment surface area, and(4) heat treatment.

It

1 II. TECHNIML DISCUSSION

1. Solar Reflectance and Absorptance

The effect of optical parameters on solar reflectance, R., andsolar absorptance, as, ,may be defined by.measurement principles. Solarreflectance is measured for calculation of solar absorptance and theirsum is unity, Rs + -1 (1)

Reflectance is measured as a function of wavelength and relative to asuitable white standard such as MgO, CO3 , or Ba90 4 . The reflectanceof the standard is assumed to be. one, 1RStd = 1. The subscript denotes

[ a sample sufficiently thick to prevent transmission; in practice, typicalstandards have a reflectance of 0.98 to 0.99.' The reflectance of typi-cal theiml control pigments is shown in Fig. 1.

[ Solar reflectance over thc solar spectrum2 may be calculated from

Rs= JI.R& .(2)

where Ia is the solar radiation intensity at a specific wavelength and

[j R• is the resulting reflectance.3 At'sar wavelength;" solar reflectance

is proportional to solar radiation intensity. In practice, the solar.spectrum is incremented into n equal energy bands, and solar reflectanceis calculated from

Rs Ri/n (3)

where Ri is the average reflectance.

2. Flight Evaluations of Coatings

Until 1965, the solar absorptance of oxide pigments, such as ZnOand TiO2 , was thought to increase only slightly by exposure to ultra-violet radiation. In the laboratory, the wost stable thermal control

Ii coating was polydimethylsiloxane pigmented with zinc oxide which laterproved extremely unstable. Before 1965, optical stability was deter-mined by measuring the solar reflectance before and after ultravioletirradiation in a vacuum but with post-irradiation reflectance measure-ments in air.

The first serious discrepancies were reported by Pearson' for datacollected on Orbiting Solar Observatoy II. Solar absorptances of

Preceding page blank 3

100 -

60 -40^

20 1025 E55

O.,

.60 •" . • Pre-Irradiation

/40 Post-Irradiation

0 20 1280 Esil

100 (c).1 --

4% 100 I

60

40 -

20 1000 ESU

01--2 1 - - a,_ I a ,, I 1

0.2 0.6 1.0 1.4 1.8 2.2 2.6

Wavelength (microns)

Figure 1. Typical Ultraviolet Degradation of (a) Zinc Oxide,

(b) Rutile, and (c) Zinc Orthotitanate at Five Suns

Intensity

4

titanium dioxide and zinc oxide in silicone binders increased at ratesgreater tban laboratory predictions. Confiming data were obtaied onthe Hulner IV flight where z-in oxide 2n poASSiUM s32iat degrededat ten times the laboratory rates. 5 The flight-laboratory diseancieswere Am -t-rated by HacfilLan et al. by decreased reflectance iiter

vacut and ultraviolet irradition.! Subsequent ueastaremzts of reflec-tance u-der vacu were referred to as in situ

Zinc oxide and titanidm dioide, both in pollailoxene,exiited inrae of 0.10 and 0.15, respectively., in solar absorpac,

L&ac, at 133D equivalent sunlight hours, SH. ND further absorptancecaes occurred on term•i•tion of ultraviolet irr'Aiation, but -. ithinten minutes of air admission, inuced solar absorptance decreased to0.01 for zinc oxide and to 0.03 for t-itanium dioxide. The solar absorp_-tance recovery to near pre-irradiation values invalidated mamy previousradiation daage concepts. A description of in situ reflectance appa-

S[ratu7 was reported by Zerlaut and Courtney.

Since radiation-induced solar absorptance bleaches r=_idly byexrowsre to air. a nmer of fl!ght experiments were designed to testlaboratcry in situ measurements. Data frcm Fegsus I and SO-I (bothnear-earth orbit spacecraft) agreed well with laboratory results for azinc oxide and polydimet=lylsiboxane coating.e Flight-laboratory correla-tion was shown to be highly dependent on the laboratory ultravioletradiation source. Samnles frow the near-earth orbit of OSO-III corre-lated well for zinc oxide degraded with a mercury arc lamp while xenonradiation produced much higher optical damage than fo-ml in flight.9

This was surprising because xenon radiation matches the solar spectrumbetter than a iaercury arc, and induced absorptance is highly dependenton the spectrnm of the radiation source. 1 0

For deep space flights (the Mariner and Lunar Orbital series),

solar absorptance increases were much greater. Increased degradation~ Hwas attributed to increased pu-ticulate radiation (solar electrons and

protons) found in deep space. Particulate radiation is negligible fornear-earth flights and prediction of thermal control responses fraulaboratory testing is reliable. Deep space simulation using combina-tions of electrons, protons, and solar ultraviolet radiation has beenpoor due to the inability to duplicate the solar radiation erairommentand the synergistic effects of combined irradiation. 11112

3. Degradatian and Damage Mechanisms

H IThe evolution of oxygen from oxide pigments by the combination withphoto-induced defects is necessary for most proposed mechanisms of ultra-violet degradation. The strongest evidence for relating ultraviolet-induced reflectance decreases to oxygen desorption has been the recoveryof reflectance to pre-irradiation values with oxygen admission.

5

a. Zinc Oxide

optical dmage in zinc oxide occurs = two distinct regions,the visible (0.4-0. microns) and the infrared (2abe 0.8 micron).

ammage in the infrared region rapidly and completely bleaches upon exc•-sure to air while damage in the visible shows very little recover; - Theinfrared d~amage spectra can be r~e!1eneted by -vacwai and may be observedfor several cycles of air and vaczmm expsure. it is related to the lossand gmin of byjsica1y adsorbed o.-ygen.

P-hotdeorpion of o•'gen ha been rel.ated tDpoteconductivrit,

in zinc oxide. Increases in systen ressiure and conductivity occurredsimultaneously with tungsten irradiation. Desorbed o.ygen was presunedresp•onsible for uress-te imcreases. I O`-gen ras sbhin to be _hoto-desorbed f--m irradiated zinc oxide, but anomwtric t n-e-vented fps species identification. 0he absorird oxygen species were

ostualated to be 0, 0-, and O with the relative ouantities dependent

on tenneratture. 1-5 3_y electrical confuactivity and pressure nathe uredominant a-.sorbed species a teaterau was de ed to be

Zinc oxide has a negati-v' surface charge produced by the cheei-sorption of molecauar oxvgen on the surface as depicted in eTuation (41)-,

92 (g)+e -+ 0.

Oxygen adsorption concentrates excess electrons at the surface and pro-duces an electron deficient region extending a-p ro-irnate]-yr 100 to 1000 .f.:: the surface.!-8 Further evidence for surface degradation was thelar-c- of detectable darae in single crystal zinc oxide.19

Radiation induced holes, attracted to the surface by the nega-tive charge, combine with chemisorbed oxygen to form physically adsorbedoxygen. Samll attractive forces (0.05 eV) permit thermal oxygen desorp-tion. Under vacuum, desorption is irreversible and excess zinc is de-posited at the surface to reverse the ch-_rge. Holes must overcome aretarding potential before the defect concentration reaches equilibrium.Thus, induced absorptance reaches a constant -value for a specific set ofconditions. Degradation rate may seem to be initially limited by oxygendesorption and later by hole arrival at the surface.' 9

Blakemore assumed that interstitial zinc was responsible forincreased solar absorptance.;'° Sklensxy, et al. proposed induced infra-red absorption by either free carrier electrons or uxygen vacancies (F-centers). Free carrier absorption was eliminated because free electronswere not detected by electron spin resonance and absorption did notincrease as the third power of wavelength. Visible abtorption wasattributed to either lattice distortion steaming from excess zinc orhole traps. Dack of visible spectrum recovery was in agreement withdiffusion of zinc to produce lattice strain.2 1

dislocations- The resultimg lattice strain womi cae toe absorution

tion -Y a comEaex zinc-dislocation color center w=M occur. 7bey sup-xported the disl~cation theory by inducig surface catio'Is in singe

crystal zinc oxi to prodmce a brad edge shift i•_d•t INids -re fr-the puheodegradtion shift.22

b. _•'-ile

Siiautile degretion .ini• r to zinc oxid with abrtionfrc. the Ibsorption edge into the infrared air a r- fo a.oroi-mtely 0.8 to Z.0 mxron. Fecovery of rutile is inw• Iete with cg ter

damage rmining in the visible than in the infrared.33 Nomeasurable absorption increse for single crystal rutile is in accordwith a calculated electron depletion denth of 2000 L..2,

I oxygen adsorption-desorption degradation control --s been sug-gested for rutile. Correlatton has been estblihed beween increasedS01-2-lar bsorptance andA increases in c-arbon dioxi-d mrsr e-,-olye fvm]Jrutile containing in excess of 100 rm carbon.--- An increase in oxygenpressure was observed during the peik degradation r-ate of r-tile, butthe quantity of oxygen was too msall to clearly di•tigsh betweenadsorbed and desorbed oxygen.V

A large nuber of defects bate been suggested; among thrn areE F-centers. interstitial titanium, and Ti72.27 Greenberg, et al. suW-gested thit the broadness of th- absorption band resulted from seve.-acodleined defects such as (e/Ov-0-,'-i+3)0, (O(->2 /-i+j 3 ), or(Ti--'/0V T-+3)O22 Coufova and A end studied the o tical abso_ tionvspectrum by annealing rutile in hydrogen and attributed absorptionpeaks at 0.5 and 0.66 micron to electrons at oxygen vacancies.2Von Hipple, et al. observed absorption at 1.5 microns in oxygen-deficient rutile and corclud-d it was caused by lattice collapse fromoxygen loss, a g T to trap electrons and form Ti3 near oxygenvacancies. ° Yahia measured electrical conductivity of rutile as afunction of oxygen pressure at 10000K. 31 For pure nutile, an oxygenvacancy mechanism was predominant above ID rmTg and titaniu intersti-tials dominated below 10 nmHg. For Al-doped ruti:e, both -echan:sms

were present but could not be separated.

c. Zinc Orthotitanate

Zinc orthotitanate exhibits ultraviolet-induced absorptionsimil-r to rutile except the peak degradation occurs at 0.85 micron.Through electron paramagnetic resonance studies, Ashford and Zerlautconc±udel that Ti+' was created by photodesorption of oxygen and isresponsD-le for induced absorption in zinc orthotitanate., 2

!7

S•os have received videspread use as & biader in tbeimlcontrol coatuqp. 3FuJy&U.L=IdWiJ3aC V [(C43)2 SIiiij, an poqrnow-Met1wlsjlaxaM, [HSiOt.]n, bave absorption eges neer 0.2 micron adare tia parmet in the rkinider of the ultraviolet and the visibleapctru. Bth silicones exh•,bt absorption incsreses in the ultra-violet and cortvave portion of the visible spectrtm after exposure tosolar envJroPt. _Pboumoaret- lmsiloxe is fternfly cured and ismore stable. !he greater ation of the dibylsilomne resultsf--= damme to the cta.lvst used in cmrh.4 Recent evidence indicatesthxtt the instability of the dineftl form is not caused by curing agentsbut by silicone decoupositior or trace contamdmtion.35

Ionsilicone is r coapred to pigmentdegradation. Pimentation with ultraviolet absoxbant uterials providesadditiaonl pretection for the binder by rei. Tbus, the najor

effort in ilprovizng the solar stability of tbermsl control coatings hasbeen directed tord pigments.

Caqprisson of binderless pigments with paints indicateAir d ration resistance from the silicone binders. A rutileyolydimetbylsibxame want showed (1) infrared absorptance increasesto be lover and slover for the paint than the pigment, (2) paint degra-dation to increase with temeratare, eud (3) inereasing the ultravoletintensity ineed pigment ation while paint degration rmainedinsensitir.e. 1 These results suggested that oygen transport thrmgh

Sthe binder my n-dit degradation. In a similar investigation of rutile-silicone paints ard binderlers pigments, recovery :ates suggesteddiffusion-limited kinetics for the paints. Also the binder appeared tohave a passivatirg effect on the pigment.Y7

&. inent Surface Stabilization

Single crystals exhibit no increase in solar absorptance, whilecorresponding pigne.ts with surface areas four orders of magnitudegreater degrade severely. Similarly, larger particles exhibit lessdegradation for the same volume concentration than smaller ones.38

Increased degradation rates my be attributed to increased latticestrains and increased surface defect states produced in grindingA.

The radiation induced defects may be confined to the electron depletedregion near the surface of each particle. Pigments with a radiussmaller than the depletion depth degrade in bulk, while the solarabsorptance of larger perticies remain umebanged at a core.

Tihe importance of improving pigment surface stability led to vari-ous passivation processes. The most widely referenced passivatlon waszinc oxide with potassium silicate in aqueous solution. 4 0 Silicate-coated zinc oxide became one of the most ultraviolet stable thermalcontrol coatings produced to date.41 Zerlaut, et al. reported that

8

Sthe si I Jcatreplce surface caq~ and prevented oace pbatodesorp-

tion.42 Thi nehni suggeatd cbnia passiation but Sklesk,

et al. conluded that silicate coating my 1ysically inbiit oxygentransport.45 A siuilar silicate coating is cvmaunly used on terrestrialrutil pigment to isprme the resistance to ultraviolet discolorationand fading.4 4 7he coating is wecipitated from aqeous sodium silicateand calcired at 4OO-6O0C.45 Apuroiiutely 1$ of zinc oxide is alsoprecipitated from a zinc sulfate solution during the coating process.__

7itania-opacified porcelain en lr bave extreme stability in anultraviolet el-lV NI -eIt, and ev hiig. 4egrable antim -y oxideeh3ibits surprising .-tability in an et.. . The stabilization of an

Sennel *y involve either chemical passivation and/or physical inhibi-tion of oxygen piotodesorption. low oGgen diffusion in glasses promte

y sical. sivation even iu the absence of cemical stabilization.

5. Wss Tramnport i;- folyners

For steasy state 'fow of a gas fhlrogt a menbrane of thickness. Z,the permeability coefficient, P, is defined as

Sj = P (p1-p.)/t. (5)

where j is the fbu Der unit membrane' area wa-oemdicular to the directionof flov, p1 is the gas pressure on the high pressure side of the membraneand p is the pressure on the_ fi pressure side. 6 For experimentalsiEplicity P2 = 0 and P2 , << -,, so that - = p, . If Henry's law isH valid over the pressure region, the so; ilt cotificiene., S3, is

c = I %6)

where C iz concentration.4 9 For conditions of equilibrium at the gaspolymer interface, the steady state diffusion ccefficient, D, is givenby Fick's first law.

J = DCl(

and by combining eqamtions (5)- , e relation between P, D, and Sis

P = DS. (8)

For diffusion in a pure polymer the diffusion coefficient is equal[] to the interdiffusion coefficient, Di2 , for the gas-polymer mixture.

If Henm-j's law is valid for the gas-polymer solution, there is no chemi-cal activity correction, and if the gas solubility is dilute withrespect to the diffusion gases, then

11

29

ixere k1 and D,* are the intrinsic and sef-diffusi coe cients ofthe gas in the pas-poli mixture.5 ° ihen the hilh pressure is main-tained cosdtant, the Mas coaeetration in the medrane is initiallyzero and the pas permeates into an evacuated costant Volme, the lowprssure wil increase with a ont Inersig rate until a steedystate Pssure rate is reabed, Fig. 2. Exro nof the steacbstate pressure-ttme carve to the time axis estabidlsh- the lag time, L,'Wich is used to cala-late the diffai coe z5u n,-

D = 26L. (10)

fte stead state pressure- rate, dP/dt, is used in deteming the pews..-ability constaut, P. Detemimtions of the diffusion coefficiet fr~Mtime lag and the permability constant from htendy statehave been used extensively to detemime Pas permeability, diffusion,and o3hbility in polymers.

Pezmesbility, diffusion and soliiility are exponentially relatedto perature accordiig to the hrrheniUs relationshis,

P = P0 exp( ()

D = Do exp-(ED/IR), (32)

and S = So exp-(AHs/-), (13)

ihere Po, Do, and So are pre-exponential factors, Ep aid ED are activa-

tion energies for permeability and diffusion, respectively, and AHs isthe heat of solUt:!on.52 Substitution of equations (11) to (13) intoequation (8) gives the relation between Ep, ED, and bHs,

Ep= ED +tMs- (14&)

Equations (11) to (13) have been shown to be valid for simple gases inmany polymers over limited "inperature ranges.53254

Activation energies for diffusion are dependent on the degree ofcrosslinkig in polymers. As crosslinking increases in natural rubberby additions of sulfiur, diffusion activation energies increase. Thechanges may be attributed to a reduction in the mobility of the polymerchains as crosslinking increased.5s

Heats of solution for gasea in polymers include the heat of mixingand heat of condensation. Both are small with the heat of mixingusuaMly positive and the heat of condensation always negative. Thus,heats of solution may be positive or negative and are usually small.56

10

I

80!9

60

0 dP

CO40

20

0

0 100 200 300 400

Time, min.

Figure 2. Typical Pressure-Time Curve for Measurezent ofGas Permeability and Diffusion

1U

PbydiUwt ls7Iloxam (FIKS) corsists of ]inear •lecules basedon the (CH3)2 SIo 'it shown belm.--

I I j-Si-O--i-- 0-Si- 0 -

I I I

Ybis elastaur r•mi aNorpbms to approximte-Iy -50C where Cystal-lization begins and continues to below -60C.5 6 A ow energy barrierallws free rotation of CH3 groups about siloxane bonzxs5 and resultsin bigh as mobilities in the est•mer.-,c 0 1 Solubilities of gases inFOIS are slm2ar to tbose in mtura. rubber although diffusion and perme-ability coefficients are at least an order of mguitude greater.6 2

Barrer and Chio neasured oxygen perneability, diffusion, andiraubmdty constants in oldimettwlsilmane filled with silica overthe temrature range -WO to O°C, Table 1. The low activJation energiesfor diffusion are associated wrish free rotation of etbYl groups.

Table I - Oxygen Transport Coustants forPolydiuetlsiloxane/Silica6 3

T Volume Per Cent Silica(c) 5.54 18.2

p X 1083(cc(STP)//seccan/cm/cm Hg) 0 4.89 3.85

D x 108(cm=/sec) 0 12.0 10.6

S(cc(STP)/cc/atm) 0 O.311 0.277

-4o/ne) -Io to o 207& 1870 a

(Ealmoe -40 to 0 216o 3060

1is

(cal/moie) -4o to • o -9oa -11o

aCalculated

LCcpmrison of RD and Do to those for self-diffusion in liquids indicatedthat pol]ydimet1lsiloxane is more comparable to liquids than otherelastmers. Higher activation energy for diffusion at the higher fillerconcentration was attributed to the filler particles acting as crosslink-ing agents and producing a tighter chain network.6 3

b. Effect of Pignentation

Solubiity--Where the binder and the pigment in thermal con-trol coatings can be considered separate, noninteracting phases, the gas

Li_ solubility coefficient of the paint in the absence of pores can beexpressed as

S = VBSB + VpSp (15)

where VB rnd Vp are the volume fractions of binder and pigment, respec-

tively, -B is: the solubility coefficient of binder, and Sp is the sur-

face adsorption coefficient of the pigment. Should porosity be preseniin the pigmamted polymer, the quantity of gas occupying the pore wouldbe added to equation (15), giving,

S =VBS + VpSp +VvSv

where Vv is the volume fraction of pores or voids and Sv is the solhu-bility coefficient. If there are no voids and the polymer wetr thepigment to the exclusion of gas adsorption, the solubility of the com-posite will be represented by equation (15) with So equal to zero.04If the pigment is not completely wet by the polymer, an extensive inter-face is created for gas adsorption. For natural rubber filled withcarbon black, the hydrogen, nitrogen, and oxygen solubilities in thecomposite were greater than predicted for rubber alone. 6 5

Barrer, et al. investigated solubility of propane at 400C innatural rubber filled with 0 to 40 volume per cent zinc oxide. At lowfiller concentrations, propane solution occurred in the rubber only; aspigment concentration increased, filler adsorption and solubility waswell represented by equation (15). The poro0sity at higher pigment con-centrations was associated with aggregates which prevented wetting by

j the polymer.8e

Solubility of C4 and C5 paraffins in silicone rubber filledwith 0 to 19 volume per cent silica was predicted by equation (15) withsilica adsorption equal to zero. An investigation of temperature effectsshowed no variation in heats of solution with filler concentration. 6 7

Diffusion--Diffusion in a heterogeneous medium is a complexsituation for which diffusion equations have not been derived. Instead,existing equations have been modified to take into account, among otherthings, increased path length due to the presence of the filler. Barrer,et al. reported for tran'ient state flow,

13

p, 1 QiC CiM O

-J

2C )2CB (17)ýt = VBKDB- •X- !J

which can be compared to Fick's Second Law where D is not a function ofposition,

i 8C D 2c 'X- 2 (18)

where C and CB are concentration in the pigmented polymer and polymer,

respectively; t is time; K is a structure factor; D and DB are the dif-

fusion coefficients in the pigmented polymer and polymer, respectively.The concentration of gas in the pigmented polymer excluding the porephase is

C = VBCB + VpCp (19)

and if Henry's law holds for the pigment and the polymer,

Cp/CB = Sp/'SB. (20)

Rearrangement of equation (19) and substitution of equation (20) gives,

= . (21)VB + Vp(Sp/SB)

Therefore,) 2 CB 1 C_ (22)

ýx2 VB + Vp(Sp/SB) ýx2

andVB

D =KDB VB (23)VB + Vp(Sp/SB)

If no adsorption occurs on the pigment,

D = K DB (24)

Equations (17) through (24) are based on the assumptions that diffusionproceeds only in the polymer phase and diffusion in the polymer isunchanged by the filler. 68

Rayleigh6 9 in 1892, derived a structure factor predicting theinfluence of a cubic lattice of spheres on the properties of a medium.His factor has been successful 0 ,7 in explaining diffusion in hetero-geneous media and can be expressed as

K= (l2 + VP ~Vp 10/3) (25)

I K=14

- I

Equation (25) makes no assuption on the size of particles present butonly upon the volume fraction.

Permeability--Changes in permeability constants caused by pig-Li mentation have been expressed using the structure factor of equation(25) .72 Since D = P/S, substitution into equation (23) gives,

-K __ (26)"S SB VB + Vp(Sp/SB)

and on rearrangement,

P =KVj3 s (27)P = K VBPB VBSB + VpSp

Substitution of equation (15) into (27) gives,

P= K VBPB, (28)

where PB is the permeability constant for the polymer. Equation (28) isLi valid whether or not adsorption occurs on the pigment.

I 6. Surface Energies, Contact Angle, and Adhesion

Equilibrium between the three surface energies in a system composedof a drop on a plane-solid surface is given by Young's equation.

7SV = 7SL + 7LV cos e (29)

where ySV, 7 SL, and 7LV are the surface free energies of the solid-vapor,

h j solid-liquid, and liquid-vapor interfaces, respectively, and 0 is the

s,)lid-liquid-vapor contact angle measured through the liquid phase,Fig. 3. Of the four parameters only 7 LV and 0 are experimentally deter-

minable. The surface free energy of the liquid-vapor interface is thesurface tension of the liquid in equilibrium with its vapor and the con-tact angle can be measured by a variety of techniques.7

S!The wettability of a solid by a liquid is indicated by contactangle. If a solid is not completely wet, the liquid contact angle isgreater than zero and if complete wetting or spreading occurs, the con-tact angle is equal to zero. Conditions for spreading are given by

S = 7 Sv - (7SL + 7SV) (30)

"15

4

Vapor

YLV

LLiquid

'44r8

Figure 3. Surface Energy/Contact Angle

where S is the spreading coefficient. For spreading to occur, S mustbe greater than zero. This simply means that the combined free energiesof the solid-liquid and liquid-vapor interfaces mist be less than thesolid-vapor free energy. Under this condition the systems total freeenergy is reduced by eliminating as much of the solid-vapor interfaceas possible.

7 4

The reversible work of separating a liquid from a solid is equalto the free energy change of the system,

WA = YSV + 7LV - 7SL (31)

where WA is the work of adhesion. On substitution for ySV from Young's

equation, the work of adhesion is

WA 7LV (1 + cos 0). (32)

16

-I

LiZisman presents a compilation of work of adhesion values and shows thatthe work of adhesion is a parabolic Duriction of -surface tension.7'5

Contact angles of ýliquids advancing over previously nomietted Sur-faces may be very different from the receding angle. This has beenattributed to the surface roughness of the solid. If gas is not trappedat the solid-liquid interface, contact angles less than 90° decreasewith increasing surface rouSbness'and angles greater than 900 increase. 76

Strain in glass has been observed to have a sizeable effect on watercontact angles. The contact angle of water on a clean glass is normallyzero, but Bartell, et al. found water contact angles as high as 800 on

I glass that had been rapidly cooled. Examination of the glass -with polar-ized light revealed intense strain. After strain removal by annealing,the contact angle returned to zero. Induced mechanical compression alsocaused an increase in contact angle.7 7 The effect of gravity on con-tact angle is considered negligible and is generally ignored; however,differences in contact angles as much as 120 have been caused by gravityfor a lucite-water-air system. 7 8

Surfaces containing exposed methyl groups ,are hydrophobic andexhibit large contact angles with water. Highly condensed packing ofmethyl groups in crystallized paraffins have water contact angles aslarge as 1110, while the less densely packed methyl groups of amorphousisraffins have contact angles of 101%. n angle of 1010 was found forpolydimethylsiloxane which i#dicated that the methyl groups at the sur-face efficiently shielded the subsurface Si-O linkages. 7 9

From studies of adhesives, it is generally f6und that polar surfaceswill bond to polar surfaces and nonpolar to nonpolar. Glass with a nega-

L. tive surface charge has a polar surface and, consequently, it is wet bywater. When bonding glass to glass, polymeric materials with polar func-

Stional groups such as-OH,-qOl, -C = 0, and -COOCH 3 produce good adher-[J ence. For bonding polar surfaces to nonpolar surfaces, an intermediatecoupling agent containing both polar and nonpolar functional groups isused. The polar groups of the coupling agent are oriented to the polarsurface and nonpolar to the nonpolar. 8 0

Moser has shown that adhesion between two materials can be predictedIi by determining the polarity of the surfaces using a water contact angle

as an indicator. Surfaces wet by water were considered polar. As con-tact angles increased, the surfaces were considered more nonpolar andpolarity ranges were arbitrarily defined as: 0 to 2U', polar; 21 to 45°,slightly nonpolar; 46 to 65', moderately nonpolar; and 66 to 990 nonpolar.In general, similar polarity was necessary for good bonding. Where goodbonding was obtained between surfaces of opposite polarity, a reorienta-tion of the functional groups was evidenced by changes in water contactangles. 81-8,1

S[I

3i. MAterials

a. Rutile Pigment

The pigment for confact angle studies and solar stabilityree nrts was Ti-Pure R-9W Rutile, lot 9980, from E. I. pu_%t

demnours and Co3upy. This pigent was made by the chloride processand encapsulated vith a silicate-zinc oxide coating from a sodiumsilicate-zinc sulfate aqueous solution before themal treatment atWoO-6O°C. Typical chemical analysis of the final mroduct is given i-Table II.s4 The A120 3 originated by oxidation of TiC14 and A-lCl, -aspresent in ,Lhe pigment, not in the pigment coating.

f Table II. Rutile Pigment Ccmposition

SOxide Weight Per Cent

Ti02 89.5

A1o3 1.01

SiO2 5.8: ZnO 1.2

b. Zinc Oxide Pigment

New Jersey SP-500 zinc oxide was chosen for degradation rateA studies.

c. Polydimethylsiloxane (PDM.S)

~l The polydimethylsiloxane was RTV-602 from General ElectricCompany and was received with a typical viscosity of ten poises. Aftercrosslinking with 1,1,3,3,tetramethylquanideve, a transparent elastomer

was formed.

d. Polymonomethylsiloxane (R4MS)

An organopolysiloxane resin, 01-650 from Owens-Illinois, Inc.,was dissolved in ethanol for paint preparation. Crosslinking by thermal

-, treatment produced a brittle transparent solid. Ethanol and water werevolatilized during curing, leaving pol~ymonomethylsiloxane. 8 5

jI Preceding page blankII1

2. Oxj&en ?zanspcrti

Fer ezbility ccnstants wt.-:- -ft-erune& try the stecIy statemetbod and diffusion -eefficien-tz bj U= !:-. Solubility cotnts

were calculated, (S = P/D)1 , whbere S., aA itD arre solbihlity, peruieebil-it" and diffbsion constants, respec•.ive-iy. Apparatus and Procedure hivebeen described earlier.g

b. Seaple Prepe-ration

%o]ymonmtbyl si~ozane (MM) was dissolved in absoluteetbhaol at 43 per cent solids content. Citric acid was added at aconcentration of one weight per cent, of CWS to reduce brittlenessin the cured zol r.n3 7 PMS/-•-u&•-e comositions, Table I-!, were

formla÷ed using a polymer density of 1.298 and a pigment density of4.o66. Etrinol required to aissolve .r.- and thin the paints "msadded aid paint voiumes of 150 cc were dispersed by milling for 2A hours.

Samples were cast in 100-m diameter glass rings on flat andlevel 3.013-cm Teflon sheets -And covered with filter paper. Self-supporting nembranes with ,.1ass rings attached were removed from theTeflon and cured vertically at IC'C for 48 hours. The glass ringssupported the membranes during curing. Membrane thicknesses ofapproximately 0.02 cm had variations of six to seven per cent.

Table III. Compositions of PblymononethylsiloxanePigmented with Rutile

Weight Per Cent Volume Per Centa

Composition Rutile P14S Rutile PWMXS Ethanolb

11 85.3 14.7 91.8 5.2 20012 73.3 26.7 89.6 lO.4 200

13 55.0 45.0 79.3 20.7 200

14 41.6 58.4 69.0 31.0 200avolume per cents were calculated from measured densities of 1.298 and

4.066 for po].ymonometbylsiloxane and rutile, respectively.

bEthanol volume percentages are based on rutile plus T114S volumes of

100 per cent.

20

I

3. Contact Angle

a. Y m u

7be wtting !c F P-1 ietics were determined by contact anglemminmrints of uater drops. A cell with glass windws momuted to a

bsr~italjpositionfi metallurgical mIcr 1-sac gPe permitted vertical andoMIrlzntal positioniM. Drops were Ml- rated with colJimated U&A

and recorded with a 35 1 Pa.

Gravity effets were elimited by neasuring contact angesof a drop on the substrate (normal) and suspen.ed from the zubstzrt[1(Inverted), Fig. It. Th substrate was mouted off-center and rotated

Lon imined ball berings. The sifstrate position as fixed by a weighttbat could be adjusted to _oduce an exact 180" rotation.

SThe stage was laraflel. with the rotation axis. A bubble levelmounted to the sge pemitted leveling in the inverted position.Parallrel surfaced substrates were positioned with spring loaded clipsand cotid be rapidly rotated from the rnomal to the inverted positionvhdl waintaining a lee. sawe surface.

A glass pipette containing distilled d-inieralized water wasinsertWd through a cvmpression O-ring seal in the top of the ceil andlowered the substrate surface. A container of water was placed inthe cell and 30 minutes was allowed fcr atmosphere saturation. A dropwas placed on the substrate and the pipette was retracted to preventinterference with sample inversion. Drops were allowed five minutes toequilibrate, photogralihed, then inverted repbotograthed after fiveminutes, returned to normla, ana phc,')gratphed again after five minutes.Three drops on each of two samples of each naterial were measured.Drop reflection on the substrate surfaces gave a sharp delineation ofthe liquid-solid interface except for contact angles near90. Contactangles were measured with a gonioeter on the eplarged projection ofS~photograph negatives.

b. Sample Preparation

Water contact angles were measured on po•,ycrystalline rutile,free-formed surfaces of PINS and !W4S, rutile surfaces that were sela-rated from cast P11S and H9S, and PH1S and PM4 surfaces separatedfrom rutile. Free-formee. surfaces were the air exposed surfaces ofpolymers cast on the low energy surface of Teflon and cured, as describedfor oxygen transport samples.

Silicate-coated, chloride-processed rutile pressed at 5000psi as 2.8-cm discs was fired at 11150C for 168 hours. Density of95 + 0.1 per cent theoretical was obtained for six samples. Disc sur-faces were diamond ground parallel to a thickness of 0.4 cm and one sur-U• face was polished to a l41m diamond finish. Samples were cleaned by

- 21

VT

Mrs, a

I -

S'•

li or L

22

Iioilim in 2:1e eit Ic-1ft=Ic acd, risiot with li at.cm=titiez ofI ~distilef- Water ant &71i6 at nrk: for -five =Lmztez. The =q1-t~e

-sresitzg of Watnr o a glass sao et clea t z a-ac e 1;M=

PC •and otiG so ut2= Prepared for mo-szr we--*

Cast an cleaned rutiLe s--stratez aod. c•T-ed. A-• b"see of ivr heziena~llowd easy separatica for measurmset of the ratielesiU~icaca itter-_faces. An samples ilWIOrigi silieomes wert e eiantns tey at est e to@e

A~j. for 2k bmrs to pvrmte Solvrent remomla.

A;.0 In Sif - 3sns ct on ata esueetes • tMea e effect of ac6m ultraviolet orrediation on the solmr re-felctoancero pcsiethysilozar- /zime oiide paints was e e asurt e at thevlzsacum rland Coatings Branch, Irnigt-rfttecrSOn Ai Force Base.Saaples were irradiated rit n a 2.5 re oenon source (Spectrolab MoadedE-25) at an ultran ol let ane,+ sofr -tis oot e sun in- a VWsarof 5 x rLt tor-. Origicnla ultraviolet intensity ars estimted to be3.0 + 0 - 0.3 suns but on actual m~easurem~ent was found to have the

lower value because of severe degntdation of the optical surfaces inthe source unit. Refsectance was measured in sita with a Becr lun Dr-y Aref]-hctodeter using a four-inch integrating sphere in the vaclum chamber.

i1Reflectance froum 0.35 to 2.5- microns was measured in air before vu--down. The samples were e eld in vacuum for cne week and pre-irrayiation

vacuum reflectance was measured. In situ reflectance was easuwredperiodically during irradiation and a recovery scan was iade after twodays in air. Solar reflectance, solar absorptance, opi induced solarabsorptance were calculated. Sample temperature varied ftro 60 to 8.3 c

during irradiation.

b. Rutile -Pigment and Faints Preparation] ~A i-utile standard was prepared by spraying a water slurry ofchloride-processed, sill.-.ate-coated i-utile onto a heated aluminum sub-strate of 2.4 cm diameter and 0.88 cm. thickness. Multiple layers pro-duced a coating 0.0075 cm (3 mils) thick. All-miinum substrattes werecleaned by abrading wdith 24 0-grit silicon carbide, rinsing in tolueneand acetone, and drying. Rutile/silicone paint compcsitions 1-4,Table IV, and compositions 11-14, Table JII, were cast to 0.003 cm(12 mils) wnd cured en the aluminum substrates. Paint preparation andand curing were the same as described for oxygen transport samples."'•,"Citric acid vas omitted from the PMMS paints.

I2

[ 23

Sable 1W.. t .am of l.• 3npii Orted with qriH ti

J ey P- iper oi deer ronrte

Con a "eclao e prO PMsrae Ro • si"•-0 Pamd sc eno-

1 13.72 US5-36 9.h

2 31-97 56-03 20-:6 89.643

3 52.22 47-78 21.17 '18.83 50)

4 46. 35.59 V-.78 69.22 66

atolueover cents were pfrmted f s eatmold densities of 0999 and4A.O6 for pobydimetIVy~sI]Dxame and rutile,, respectively.

cast volueme percetages are based ratile and IDS voures h of 100per cent.

c. Zinc Oxide Pigment and Paints Preparation

aNe Jersey SP-50F zinc oxide was cbosen for degradation ratestudies. A zinc oxide standeard was prepered by spray ing a upter sliyon a trecltaned alt am owdserbstrate. aoaydiVrethyosi Xatie and ao3Yznd-metrlhsisoxane were pigmlented with 20 volume per cent zinc oxide andcast of arueeim substrates, Paint prepraation and curing were the sameas described for odygen transport samples.

5. Zinc ortbotitanate Pinoent

a. Pigment Forha ilation

Zinc orthotitanate pigments were prepared by ball -J3in andheat treatment to obtain powders with a variety of particle sizes andthermal histories. A flow char-t of the pigment processing is presentedin Fig.. 5.

In a five-gallon polyethylene jar, 3600 gm of a 2:1 molarratio of reagent grade ZnO and tei02 (anatase) were ball-milxed with9000 cc of deineralized, double-distilled water for 46 hours. Theslurry was dried, crushed to 60 mesh, and reacted for 24, hours at 9250Cin an alumina crucible. The resulting material, Sample A-1, containedzinc orthotitanate and zinc oxide, as identified by x-ray diffracto-meter, and had a particle size of 0.91 micron.

The reacted ZnTiO4 Was slurried with 4500 cc of 10% aceticacid for 24 hours using a polyethylene stirrer. After settling, 4000 ccof the sample was decanted, and the insoluble material was mixed with

24

1- 0 aa£0 0 0 to

00

SEp 4P4

£A '- so a

m a týa 6 a1

s~o of^

%0~ ~ ~ *0 :% 9%9 & % %

fn = 04, 0 a-I 00 m 1

govU 0 0

9:42

cu~~ a a -

*d a 0 0 0

E00 0 0 0 0 I~ ~4'0 0 0 00 zP o~ o~ o oZ

(D r.

H 4, )4-. 0 0 0ý

rIOI

25

I

AM c f isne =er 7e ecntzWprces asrepeated eia Iti" and f bateh, S•We A-2, ws dried at IWC. eit losan lmebl S 1kJ p 7 )".0%) and the mnly x-ray diffractmeter detecta-ble peak other •f ••! bad an intensity ess; tan :K-. The particle Lsize of Slam A-2 wa reduced to O.8 pga by tie leaching process.

b. Particle Siz eduction j

Sample A-2 was split into five batches, one 8m0• batch andfor 650-9p batches. 7be 8W-p batch (Sample A-3) ws slurried with135D cc of distilled water and milled with almi balls in a one-gallonpol1eftlee Jer for 100 bours. Vi 20 cc samples were perodical

tr during milg (sampl A-3-1 to A-3-9). Surface area andequivalent spberical particle size for each sawm represent the extentof particle size redaction, Meble V. 7he particle size depmedence onbanl-•miing tine, Fig. 6, provided the control necessary to achieve as fic particle size. X-ray investigation showed a detectable quan-tity of aIam after 52.3 bours with the alteini content i 2creasi

with time. To minimize contamition from will balls, samples for ultra-violet degradation were liited to a inxiz millin time of 36 hours.

Table V. Surface Areas and Particle Size of Z•n 2TiO4 Piients

A-i Surface Areaa Particle Sizea

A-i 1.26 0.91 Reacted at 950oC

A-2 1.41 0.80 Acid ueacbedb

A-3-1 1.47 0.77 0.5 hr mill c

A-3-2 1.54 0.73 1.5 hr mille

A-3-3 1.73 0.66 5.0 hr millc

A-3-4 1.88 o.60 10.0 hr millC

A-3-5 2.09 0.54 27.7 hr mj11lc

A-3-6 2.62 0.4.3 52.3 hr millc

A-3-7 3.24 0.35 73.8 hr ,,i11c

A-3-8 3.45 0.33 97.3 hr rillc

A-3-9 3.70 0.31 100.0 hr millc

a Measurement of surface area and calculation of particle

size has been described previouslyAsob Reacted at 9500C prior to acid leach.c Reacted at 950°C and acid leached prior to ball milling.

26

11 £m/Im) V3HW 33Y-J1

El a

43

'4

PPd

CD,

IllJ.lw WI- 313V8

27

Three of the four 6 50-gram batches, Samples A-4 to A-6, wereslurried with 1050 cc of distilled water and ball-milled for 36.8 and2 hours, respectively. Resultant equivalent particle sizes were 0.4,0.58, and 0.61 gm for Samples 4, 5, and 6, respectively. The fourthbatch, Sample A-7, was not ball-milled and had the same particle sizeas Sample A-2, 0.80 pm. All samples were dried at 1100C and crushedto 60 mesh.

c. Heat Treatment

Samples A-4 to A-7 were split into five 125-gram batches forheat treatment. Four batches from each sample were quench heated to850, 950, 1050, and 1150 0 C, held for one hour, and quench cooled in500 cc of distilled water. The fifth batch was quench heated to 850 0C,held for one hour, and annealed to room temperature over three days,Fig. 7. All batches were fired as uncompacted powders in platinum cruci-bles. Temperature was controlled to ± 5oc in a silicon carbide resis-tance furnace. Each heat-treated batch was dried and ball-milled with200 cc of distilled water for two hours. Twenty cc of each slurry weredried for surface area determinations.

d. Irradiation Powder Samples

Water slurries of twenty heat-treated pigments were sprayedonto heated aluminum substrates, 2.4-cm diameter by 0.08 cm thick.Multiple layers were applied under a dry nitrogen pressure of 16 psi.Two coated substrates of each sample were delivered to Mr. C. P. Boebel,Elastomers and Coatings Branch, Air Force Materials Laboratory, onAugust 25, 1971.

e. In Situ Reflectance

Ultraviolet exposure testing of the 20 Zn 2 TiO4 samples was

accomplished in the equipment described in Section III. 4.a. In situpretest reflectance was measured in air and in a racuum (4 x 106-O-r).Reflectance was recorded at 18, 45, 1i0, and 230 equivalent ultravioletsun hours. A final reflectance scan was made after 48 hours air expo-sure in the dark. Temperature of the specimen back was 490C duringirradiation.

28

LI

800

700-

600

500-

S? 0-W

[I FURNACE OFF.H 300

100-

Fl 0 10 20 30 40 50 60 70

TIME (hr)

Figure 7. Annealing Curve for Zinc Orthotitanate Powders

T 29

hI]

1 IV. RESULTS AND DISCUSSION

1. oxygen Transport

Oxygen tra.,,_srt properties of pol•ydimethylsiloxane/rutile compo-11 sitions were reportel in AFML-TR-71-42, Part 1.88 For comparison with

polymonometbylsiloxane properties, a port'on of the polydimethylsiloxanedata has been reproduced.

L" a. Poy]metbylsiloxanes

Oxygen sorption isotherms at 250 C for P1MS and PMMS, Fig. 8,show that Henry's law held for the pressure range investigated. Con-stant oxygen diffusion coefficients of 18.9 ± 0.3 x 10-0 and 5.02 ± 0.09x 1O-7 m2 /sec at 25'C for PDMS and H4MS, respectively, for the pressurerange 0 to 60 cm Hg established that diffusion coefficients were inde-

L. pendent of concentration.

Oxygen transport properties in the polymethylsiloxanes areshown in Table VI. Permeability and diffusion coefficients of PDMS werean order of magnitude greater than values for RMS and the solubilityof PEMS was approximately 30 per cent lower. The lower diffusion coef-I] ficients in Pff4S resulted from greater crosslinking and a lower methylgroiup concentration.

The temperature dependence of oxygen transport constants inthe siloxanes conformed to the Arrhenius relawionship within limits ofexperimental precision, Figs. 9-11. Least-squares activation energy fordiffusion, Table VII, was 1300 cal/mole greater for the monomethyl-siloxane and is attributed to the greater crosslinking in BMMS whichlimits chain flexibility. Permeability values and temperature depend-ence in both silicones were controlled by diffusion properties because

solubility values and temperature dependence were similar.

b. Effect of Pigmentation

1 Temperature Dependence--After complete polymer curing, thetemperature dependence of oxygen transport in the pigmented polymono-metbylsiloxanes was exponentially related to temperature by theArrhenius relationship within limits of experimental precision, Figs.12-14. The least-square temperature coefficients are presented inTable VIII.

Solubility in PMMS Compositions--Where there are no voids

and the polymer wets the pigment to the exclusion of gas adsorption,1 the solubility of the composite will be represented by

S =VBSB (33)I Preceding page blank

31

0.35 -"

Polymonomethylsiloxane

0.30 -( Polydimethylsiloxane

0.25

02 /

/ or~0.20L

0ca 0

US0.15

0110O0 10 0/E

0.05

0 I A I I

0 20 40 60

p2 (cm Hg)

Figure 8. Oxygen Sorption Isotherms at 25°C

32

Table VI. Oiygen Transport in Polymet•ylsiloxanes

P S

T (cc.(STP)/sec/ ;D(c6(STP)/(0c) cm1/cm/cm Hg) (""/ see) cc/atm).

0 5.28 x 10-8" 12.8 x 10"6 0.31325 7.41 19.1 0.29432 8.06 21.1 0.28942 9.04 24.2 0.283

B50 9.86 26.8 0.278

0 1.68 x 1o-9 2.68 x 10-7 0.47625 2.83 5.02 0.428

] 32 3.29 5.55 0.45050 4.72 7.92. 0.45360 5.95 10.2 o.443

aLeast-squares values for average 6f three samples.

, II

i,

T (C)

-6.8 80 60 40 20 0

-7.0 Polydisethylsiloxane

-7.4

0-8.2

Polymonomethylsiloxane

-8.4

-8.6

-8. * 8_2.8 3.0 3.2 3.4 3.6 3.8

10 3/T (T in oK)

Figure 9. Temperature Dependence of Permeability Constants

for Oxygen in Polymethylsiloxanes

34

U

T (0C)

-4.4 60 60 40 20 0I i i I!i

§1

-4.6 N4k

N0

-4.8 *

Polydisethylsiloxane

-5.0

-J -6.0

SPolymonomethylsiloxane

0A• -6.2

J -6.4

-6.6

'1 -6.8 I I

2.8 3.0 3.2 3.4 3.6 3.8

103/T (T in OK)

Figure 10. Temperature Dependence of Diffusion Coeffi-cients for Oxygen in Polymethylsiloxane

1 35

T (0C)

-0.3 So 69) 40 20 0Si I I II

Polymnoaethylsiloxane

Polydinethylsiloxane

. -0.56

A E

-0.*6-

0

I 12.8 3.0 3.2 3.4 3.6 3.8

16 3 /T (T in OK()

Figure 11. Temperature Dependence of Solubility Constantsfor Oxygen in Polymethylsiloxanes

36

II Table V 11. 1Urrmiius Cowt=U fl or Oxygen Tmrportin DbyetbVlsiIDLxne

Po(cc(S7!?)/sec/ Std. Error of F,

HSample &Cz/u/cn Hig) (cal/wle) (cal/role)

FIUM-' 2.25 z 10-6 2019 67PIMS-2 2.64 2138 99*1 1US-3 4.45 21497 107PMS-Ave. 2.98 2188

41 RS 1.71 3774 136

Do ED. St1. Error Of MEDSample (cz?/sec) (cal./M0le) (cal/role)

PJI4-1 2A.6 x l0-3 2837 1140PM~S-2 1.07 2418 207PIE4S-3 1.39 2522 L46PhIKS-Ave. 1.54 2602

LiPMMS 0.38 3936 150

SO S0 Std. Error o"e

Sample (cc(STP)cc/atrm) (cal/mole) (cal/mole)

PtD4S-l 0.068 -829 145PI!4S-2 0.186 -280 208I]IMS4-3 0.247 -134 100PIN4S-Ave. 0.247 -414

F] M4s 0.3146 -159 145

37

T (OC)

so2-S.2 -'r- iI

0

0

-8. -; ~-8. •

10.4 V

o -8.]

Per Cent Rutile

o -0o

-_8.(" 5.2 -9 EK "

20.7 - E

i ~-9.€

-'-9. ! a

2.8 3.0 3.2 3.4 3.6 3.8

10 3 /T (T in OK)

Figure 12. Temperature Dependence of Permeability Constantsfor Polymonamethylsiloyane/Rutile Compositions

38

'U- i °ci

53 60 200

-6.0

-6.1

-6.2[

-6.3N

-6.4 Per Cent Ruti.e

0 -0E

-_ 6.5- 10. V20. 7-

-6.

-6.2.8 3.0 3.2 3.4 J.o 3.8

103/T (T in OK)

Figure 13. Temperature Dependence of Diffusion Coefficientsfor Oxygen in Polymonomethylsiloxane/Rutile

Compositions

39

-0.2 b 9 20 0I II I

Per Cent Stut!-'e0 -o5.2-0

10.421.7 -o

-0.3-

0

0-4

-. 0,0

11 El

-0.5 E '

-0.612.8 3.0 3.2 3.4 3.6 3.8

10 3/T (T in OK)

Figure 14. Temperature Dependence of Solubility Constantsfor Oxygen in Polymonomethylsiloxane/RutileCompositions

110

,~UI t

.Tabe VIII. Arzbius Constants for OxWen in POIYmono-Metylwsiloanw/Rutile-St.Erro

Po x 106cc (.TfP) /Sec/ ISt.Erro

Saqle (2/CR/C, Hg) (cal/1ole) (cal/1ole)

10 1.71 3774 136U 1.06 3624- 10912 1.47 3874 10813 1.96 4136 116

SDo X 104 ED Std. Error of EDSample (ca 2 /sec) (cal/moie) (cal/mole)

10 3.77 3936 1503 2.24 3685 15412 2.05 3694 20213 3.57 4043 124

so(cc(STP)/ As Std. Error of AHS

Sample cc/ata) (cal/mole) (cal/mole)

10 o.346 -159 14511 0.359 -60 15312 0.5w0 197 145

13 0.417 94 169

41

where VB is the volume fraction of the polymer and SB is the satubility Iconstant of the polymer.88 There "was no indication of oxygen adsorptionon rutile in the ROM compositions. The concentration dependence of the jsolubility constants was well represented by equation (33), dashed linesin Fig. 15. Scatter in solubility values was greater than found inPUMS ccWpositions due to the variations in P14S mmbranes. Comparisonof observed and calculated densities indicated negligible porosity.Heats of solution were constant within limits of experimental precisionfor all rutile concentrations, Table 8.

Adsorption on the rutile may occur in FWS compositions since iit is comon for PFHS ccupositions above 20 per cent. A BMMS composi-tion with 30 per cent rutile was investigated but oxygen flow throughthe membrane was too rapid for determination of permeability and diffu-sion. Thermal expansion mismatch between the brittle 1M9S and rutileresulted in fine cracks after thermal curing. Critic acid additionsincreased membrane flexibility but the samples remained very brittle andrequired careful handling.

Permeability in PMMS C-mpositions--Permeability in a hetero-geneous medium such as a pigmented polymer may be described by

P = K VBPB, (34)

where K is a structure factor and P is the permeability constant of thepolymer. Rayleigh's structure factor for a cubic lattice of sphereswas used to explain the pigment concentration dependence of permeability Jconstants. Values of K were reported in AFML-TR-71-42, Pt. 1.88

Permeability constants for P4MS compositions decreased withincreasing pigment concentration; however, permeability values werelower than predicted by equation (34), Fig. 16. Considering the excel-lent agreement for PF1S compositions formulated with the same pigment,the discrepancy is most likely caused by thickness variation of thePWS membranes. There is no reason to think that permeability constantsfor pigment-free PMMS are more accurate than for pigmented compositions,and it is probable that pigment-free constants were high.

Diffusion in R4MS Compositions--Diffusion coefficients de-creased as pigmentation increased, Fig. 17. These data were comparedto valu-s predicted by Rayleigh's structure factor for the case of anonadsorbing filler

D = K DB (35)

where DB is the diffusion coefficient of the polymer. Diffusion coef-ficients were lower than predicted by equation (35). This may resultby either sorption on the pigment or thickness variation. Thicknessvariations will cause the permeability and diffusion constants of a

42

~Elo

0/E

43

00 4 1

ot 4.1E) 0

-0

E) El -- -o , I, ,

0 C )j

o'

0 0)

/

S I 0 ;1-- 7, I

!0.d i o

0 °"

-9.

5.0

8 -50 0C0;0 - 25 0 C

4.5 0

00.4.0-

'3 .5 - [3

3.0

0

X

2.5-

2.0-i 0

2.01

10tA

0.51

SI-; ~1.0 -

0 5 10 15 2J

Volume % Rutile

Figure 16. Effect of Pigmentation on Oxygen Permeabilityin Polymonomethylsiloxane. Dashed lines areequation (34)

44

0-500C

250

00

IA 4

'0 Vot une Pe r Cent xIitile

-~

* ffect of p1 iienttiO~ on Q ygef Dif fvlsU in

Pol.PiOflOmetbhr1silo~xae . Daslid )~ e ~

eliatiOn (35)

4~5

77

sample to deviate consistently from predicted values. Comparison ofFigs. 16 and 17 show this. Thus, there is nD basis for concludingoxygcn sorption on the pigment in IEMS compositions.

2. Contact Angle

The reflecilons of normal and inverted water drops, Fig. 18, deline-ated water-soLd interfaces except at angles near 900. Contact angledifferences due to gravity, A g, were 1 to 40. For example, the height

to radius ratio, h/r, of a water drop on PDMS, Fig. 18, increased from1.16 to 1.32 on the inversion but the contact angle decreased only 10.Calculations of contact angles on PDMS, using the eciiation for sphericalgeometry,,9

tan 0/2 = h/r, (36)

resulted in 98 and 1060 for the normal and inverted positions, respec-tively, an average of 1020 which agrees with the 1010 measured averagevalue. Gravity has a larger affect on drop shape than on contact angle;however, in all cases, inversion decreased contact angles, Table IX.

Table IX. Water Contact Angles

Contact AngleSample Normal Inverted Average Aeg

Rutile - 1 42± 2 38± 3 40 3Rutile - 2 29± 2 27± 4 28 2

PDMS 102 ± 3 101 ± 2 101 1

MS 9o ±1 89_±1 90 1

PMES/Rutile 92 ± 3 89 ± 2 90 3

PMMS/Rutile 80± 1 76 ± 1 78 4

Rutile/PIrMS-l 80 ± 2 78 ± 1 79 3Rutile/PLrMS-2 73 ± 2 70 ± 3 72 3

Rutile/PMMS 61± 5 58± 4 60 3

46

0 nj

4-)

Il ~4)p

4-,

4 Ea,r- "pI~ ~A

d43 PI

Mz 0-

'- 0)

0

H m

4-)

a)0 'd

qJ C!)

Cd 4 U-'

447

Complete wetting had been expected on rutile but average contactangles of 28 and 400 were observed. Scanning electron micrographs,Fig. 19, showed submicron scratches and a maximum pore diameter of 20microns with the majority below 10 microns. Such microscopic roughnesswas too small to produce the high contact angles. 9 0 The surface con-sisted of rutile and a glassy phase which occupied 21 per cent of thesurface, Fig. 19. The contact angle of a composite may be calculatedfromp•

cos Oc = f cos0 + f 2 cos 02 (37)

where 0., 0,, and O2 are the covtact angles of the composite, phase 1and phase 2, respectively, and f, and f 2 are surface fractions. Bothrutile fand glass should be completely wet by water and ec should bezero. Strain can increase water contact angles on glass to 800 and mayaccount for the high contact angles. 92

The average contact angle of 1010 for PIDS agreed with Shafrin andZisman. 9 -3 A lower methyl group surface concentration in RFIS resultedin a lower contact angle of 900. Curing polymetbylsiloxanes on rutileresulted in contact angles U and 120 lower than angles on free-formedsurfaces. Usiig Moser's hypothesis, 9 4 both results predicted betteradhesion for EIMS than for FIMS. The ease of stripping the siloxanesfrom rutile showed the adhesion to be small.

Rutile surfaces separated from cured siloxanes have high contactangles. The increase may have resulted from incomplete solvent renovaland/or a residue of methyl groups.

3. Silicone/Rutile Optical Properties

a. Solar Reflectance

The reflectance spectra for the water sprayed rutile, PDMS/20per cent rutile, and NMMS/20 per cent rutile are shown in Figs. 20through 22. Reflectance of the paints was higher in the visible and

lower in the infrared than for rutile powder. Characteristic siliconeabsorption lowered infrared reflectance, and specular reflection fromt6e smooth silicone surfaces increased visible reflectance. The lowerrefractive index of PDMS caused a larger relative refractive index forPDMS/rutile paints. The initial solar reflectance of PDMS paintsexceeded PMMS paints, Fig. 23. Solar reflectance maximized between 10and 20 per cent rutile. Increased scattering from low pigment concen-trations were offset by loss of surface gloss at higher pigment concen-trations.

48

(a) (b)

J 0

`40'-

(c) (d)

Figure 19. Polished Rutile: (a) and (b), Scaining ElectronMicrographs, 500X and 2000X, Respectively;' (c)and (d), Reflected Light Micrographs of UnetchedSurface, 500X and 2000X, Respectively

; v i49

0dcii

Lj (D

'4)

1 0 4,

('jj4)

0t 00

0. OD 0

050

C" 4y

4,

00

U)

p

0 0D

0 ,

4-4

CfP4,

C~C,

rt4

511

9r4

0 0

04

o b,

0

tO E 0

*14

-c 04ro>

0

:3 ci)

C)

-- 74

0l 00

52

flil

~I

84

IOOZ Rutile

iii ~~~~83 - - - - - - -

"Polydimethylsiloxane6 82

C,i

-• '. 81

e Polyeonomethylsiloxane44

u 80

0

78 051 1 0 2 30

Volume Per Cent Rutile

Figure 23. Effect of Pigment Concentration on Solar Reflectanceof Rutile/Silicone Paints

53HJ

b. Induced Solar Absorptame

7he stability of the silicate-coated rutile produced anindiced absorptance of O.009 in 230 M, Table X. Silicate coatingproduced degradation resistance simila to that of silicate-coated zincoxide9s and zinc orthotitamteo and was the first time the effect hasbeen observed for rutile.

Mhe recovery bleaching of rutile supported the oxygen depend-ence of the irradiation damage. Differences between induced absorptaicesat 230 ESH and air recovery values for the paints, T.able X, were aboutequal to that of the binderless pigeent. Bleaching was attributable tothe pigment and permanent damage represented polymer degradation. Sili-cone permeability was too high to produce observable differences in pig-nert degradation.

Induced solar absorptance of WI.S paints was constant, Fig. 24,and indicated equal degradation from !MM and rutille. For PMS paints,induced solar absorptances in the pigment increased with concentrationincreases to 20 per cent and continued to decrease 'o 30 per cent.

The pigment degraded approxi-rately the same in both PDMS andHOS. Thus, surface interaction between the pigment and the polymer didnot measurably affect degradation.

Water associated with the pigment was indicated by the charac-teristic absorption band at 1.9% which increased in intensity withincreasing pigment concentration. For the PDMS series no absorptionoccurred for samples 1 (5% TiO2 ) and 2 (10% TiO-,), slight absorption forsample 3 (20% TiC2 ) and strong absorption for sample 4 (30% TiO2 ). Theseresults support the oxygen solubility data which indicated oxygen sorp-tion on the pigment starting at 20 volume per cent and increasing withconcentration.P' In contrast to oxygen solubility data which did notshow sorption on the pigment, B.MS paints had strong water desorptionthat increased with increasing pigment concentration. Citric acid, pre-sent in If4S oxygen transport samples but not in UV irradiated paints,may have adsorbed on the pigment preventing oxygen sorption. Waterabsorption disappeared in vacuum and partialLy reappeared when returnedto ambient conditions.

14. Silicone/Zinc Oxide Degradation

'F Infrared reflectance decreases of 4 to 10 per cent were observed

between initial in-air scans and pre-irradiation vacuum scans for allsamples. Characteristic water absorption was absent and the reflectancedecreases demonstrate the extreme surface oxygen sensitivity )f infraredspectrum of zinc oxide. Pre-irradiation vacuum solar absorpt 'ces andultraviolet induced solar absorptance are reported in Table X7

54~

0 000 ooo 8- - uc uc0 0000 0000 0

* go*

C) x.

A'.4 _Dzto 07 d ago 0000 0 4) I

Sl 0

44>

8 ot'00 0000 1 0

000 000

0En\. h o o0 cm I'D E _:t 9~-1 O\~ 4) im

4' r, HO'N O (cn cu J VAa E- a)Ciu-C~i Cli ('u uCtj Cd j r-I $4 04)

0 3.3-. . te 4)M 0000 000 0 a)pv

0 0

4J) 4

wLqi\\W ~ 00\0 0 0 4'

0 ~ ~~~ ~ ~ ~ ~ 00 - - i ~ ~ ~j I - od

0000 000

Xl 0 ) +

0 a)0-W X- Hr y Y ) 0() y 0 0) a 11

H 4,

COO

Ea Cl) H

0- 0ý(L 0 C.W H 1

'ti Lf\000 0 UO0 00 )

g u Y H-s HC'jc ( 'ti 0d 0

55

3.5 1 1 1 10

3.0

2.5-10

S2.0

1.0

0.5- 6R45 ES- 230 ESH -

0 & I I5 10 15 20 25 30

Volume Per Cent Rutile

Figure 24. Effect of Pigment Concentration on Induced SolarAbsorptance in Rutile/Sillcone Paints

56

.1

i.

0L NrOC (Y 0 I0 0 0

I) 0000

-- H

4-)0

c00 C' CYi 0 L00 0 00

Li0 0 0 0l 0>0

4' ý

2~ 8)

0 ~0 0 0 0 0 PD0 0 *0 a-

0 41]j H 0 0 0 m

0)

rIo to u k

'fi 0H

0) d) ()

ca Hn U U) u 4).

H - 0 0 0 0 0

aS ,O057

Comparison of degradation rates, Fig. 25, indicate that oxygentransport through the polymethylsiloxanes had little effect on pigmentstability. If oxygen transport in the binders had limited degradation,the order of increasing stability would have been ZnO< H!IS/ZnO. Solarabsorptance increases for PE1S paints were greater than for PMMS paintsas would be predicted by oxygen diffusion coefficients, but binderlesszinc oxide stability was between that of PEMS and PMMS paints.

The smaller induced absorptances of BIMS paint compared to ZnOpigment may seem to indicate encapsulation protection. This is not sub-stantiated by the infrared reflectance changes, Fig. 26. Reflectancechanges at 2.05 mn were almost identical for ZnO pigment and PMMSpaints. The greater reflectance decreases of PDMS paints in the infra-red and in the visible at longer exposure times, Fig. 27, reflect insta-bility associated with the polymer and/or pigment/polymer interactionin addition to pigment damage. All samples bleached to within 1% ofthe initial in-air reflectance values.

5. Zinc Orthotitanate

Surface areas and equivalent spherical particle sizes for the zincorthotitanate powders are presented in Tables XII and XIII. Particlesize of most samples increased with increasing temperature, Fig. 28.The most surprising results were the smaller particle sizes of thell50°C-quench compared to 1050°C-quench for Samples A-5, A-6, and A-7.

Exponential size increases with increasing temperature were notexpected in the loose powders. The final two-hour ball milling mayhave been responsible for the anomolous particle size at 1150 0 C. Ballmilling was necessary to produce sprayable slurries of the lightly sin-tered powders.

Solar absorptance data, Table XIV, for the heat treated powders didnot show significant UV damage trends for either quenching temperatureor pigment surface area. EYcept for samples quenched from 1150%C andA-5, 950°C-quench, all powders had initial vacuum solar absorptances of0.09 4 0.01 and were fairly stable with increases in solar absorptance,Ias, of 0.02 ± 0.005 after 230 EUVSH. Sample A-5, 950°C-quench wasthin and permitted "shine through" and absorption.

The dependence of solar absorptance on temperature was apparent forpre-exposure and post-exposure reflectance studies, Fig. 29. The great-est increase in absorptance with temperature was between the IDSD°C andllSD°C-quench samples. The visible region exhibited the greatestincreases in absorption with temperature. Solar absorptance was identi-cal for The 8500c.-quench and anneal samples.

Ultraviolet irradiation damage to all samples was greatest in thevisible region (0.45pm) and the visible damage occurred primarily in thefirst 20 EUVSH. Visible damage was approximately 50%0 recoverable while

58

:1111

Iid14ý

C~Co

0w 0

Q.1

2x 0w *r-

w Wi

0 0 4-)

N N H

(I) (I) .1*4

~ 4-,H

o0 :1

LC~

(1N30 83d) SDV

59

0Q

4.')

.4-

CH

L 00

LA I

0 0i

CL C)

0 cm x f

LA Cu

U I

N t4

4-.1

a. N .

CH 4L'

0 o 0

(1N30 83d) 7G-~~

6o

00

4)V

U) W

0 kl

4x

0) I0

*t-1

t-4 N 0ý O

0,

CCl)

61~

IiTable XII. Zinc Orthotitanate Surface Areas

Heat Treatment Surface Area (m2/en)

( 0 c) A-4 A-5 A-6 A-7j

Before Heating 2.56 1.97 1.86 1.41

850 - Anneal 1.76 1.53 1.49 1.44

850 - Quench 2.29 1.67 1.60 1.58

950 - Quench 1.67 1.53 i.46 1.45

1050 - Quench 1.57 1.37 1.29 1.40

1150 - Quench 1.414 1.42 1.32 1.43

Table XIII. Zinc Orthotitanate Particle Sizes

Heat Treatment Particle Size* (nm)

( 0 c) A-4 A-5 A-6 A-7

Before Heating 0.44 0.58 o.61 080

850 - Anneal 0.64 0.74 0.76 0.78

850 - Quench 0.50 0.68 0.71 0.72

950 - Quench 0.68 0.74 0.78 0.78

1050 - Quench 0.72 0.83 0.87 o.81

31150 - Quench 0.78 0.80 0.85 0.79

*Calculated from surface area.

62

0.90

0.85

0.80

0

00

0.70 0

S0.65-(..)

0.60-•ii °°

[I 0.55

0.50-

0.45I I I I I I850 950 1050 1150

TEMPERATURE (oC)

Figure 28. Effect of Temperatuxe on Particle 3ize of Zn2 TiO4

Powders

63

- CO

HE- w ONmI - N U oa N n 0 Y<0000.-4 00 UOOOr-r-l4 WOr

'4 u000000 00000 000006 00000 3

c30

H.4- coc -- N yx % -- , 0U' 0 -, Y O \ OD0t\ * G 00H~P -H OOHHH 00 00 -z , -0) ()0000 o0\< -1 0 8900000I 00000 00000 C X

r4 0 000C;;6 6 6

4.34.

0 .4-0

toC-) r c'H W H C\j CuJC Halr C\jHC\JC~J C\ r ciH CMjC\j rjj00000 00000 00000 00000 V3CD

4.)' (3ý 8 00 00 0 00 00 0 00 0 4.)..')S

04.4)

0 WId

~0 04 OOdCH r-r- HH;6 HHHHý 66 6H 6666 H 4 a)

0 0 T3-04. Ca P

qS 04.)

0) g0- 0Q)0

4. 8 '0 o0 00)i0 ý~riC3 ,y~ C)

wo00000000 0)0666 'ci660S ci- 0) 0 ) ) 0)

aSC U C HU- Hr-ri I Hr- Id Q H+4C10 (L)___ _ . - .: Z -.--- 0 00).D ) C0

j0j ;0)4--4C

0.) r- y4;-4 oz 'Co (I4.)1

H- 000 000 000 00

() a

0 0 0

[1 0 ) 6.

fi o

U) In 1J 4-

0 0d W o W

< 4 0SN 00 m-

o~ 0 coa

d ad

HSAfJ 0OZ (OVA) SD

Cd-Cd~

z 0I A Cd*r0~~ oU 7d -

I (#2 Cd

IoZ

0o 0

2 w0"I Z) 0

00~C ci ci

0)

0 I CI co 11 r

o 000a0

0

65

damage in the infrared (0.8 to 1.0,im) recovered completely on air erpo-sure.

Lack of damage dependence on ball milling time (A-4 to A-7 series)was surprising. Both grinding-induced strain and chemical contaminationincreased with milling time and would be expected to contribute todegradation cf the pigments.

06

SV. CONCLUSIONS

A Investigation of thermal control coatings composed of polymethyl-

siloxanes pigmented with rutile and zinc oxide resulted in the followingS]conclusions:

1. Oxygen adsorption on rutile was present in polydimethyl-siloxane and not in polymonomethylsiloxane.

2. For gas solution in the polymer and on the pigment, heatsof solution may be predicted by assuming separate non-

jji interacting phases.

3. Rayleigh's structure factor successfully predicted theeffect of pigmentation on permeability and diffusion con-

11stants.

4. Pigment-polymer surface interaction did not measurablyaffect ultraviolet induced solar absorptance.

5. Oxygen transport in the polymers was too high to limiti induced solar absorptance of the pigment.

Ultraviolet irradiation of zinc orthotitanate powders of varyingparticle size and quenched from various temperatures resulted in thefollowing conclusions:

1. Particle size and heat treatment temperature had noobservable affect on ultraviolet degradation of thepigments.

1 2. Pre-irradiation and post-irradiation solar absorptanceincreased with quenching temperature.

67

L

L VI. REFRMCES

L 1. Wendlant, W. W., and Hecht, H. G., Reflectance Spectroscopy:Chemical Anaýlsis, Vol. 21, Interscience Publishers, New York,

2. Johnson, F. S., "The Solar Constant," Journal of Meteorology, 11,pp. 431-439, (1954).

3. Fogdall, L. B., and Cannaday, S. S., "Ultraviolet and ElectronRadiation Effects on Reflectance and Emittance Properties of ThermalControl CoatingS," Technical Report AFML-TR-70-156, Air ForceContract F33615-69-C-1238, The Boeing Company, Seattle, Washington,p. 27, (July, 1970).

4. Pearson, B. D., "Preliminary Results from the Ames EmissivityExperiment on OSO-II," Pro_0ess in Astronautics and Aeronautics:Thermophysics and Temperature Control of Spacecraft and EntryVehicles, Vol. 18, Academic Press, Inc., New York, pp. 459-472.

5. Lewis, D. W. and Thostesen, T. 0., "Mariner-Mars Absorptance Experi-I ment," Progress in Astronautics and Aeronautics: Thermophysics and

Temperature Control of Spacecraft and Entry Vehicles, Vol. 18,Academic Press, Inc., New York, pp. 441-457, (1965).

6. Macmillan, H. F., Sklensky, A. F., and McKellar, L. A., "Apparatusfor Spectral Bidirectional Reflectance Measurements during Ultra-violet Irradiation in Vacuum," Progress in Astronautics and Aero-nautics: Thermophysics and Temperature Control of Spacecraft andEntry Vehicles, Vol. 18, Academic Press, Inc., New York, pp. 129-149, (1965).

S7. Zerlaut, G. A. and Courtney, W. J., "Space-Simulation Facility forIn Situ Reflectance Measurements," Progress in Astronautics hndAeronautics: Thermophysics of Spacecraft and Planetary Bodies,

IVol. 20, Academic Press, New York, pp. 349-368, (1967).

8. Neel, C. B., "Role of Flight Experiments in the Study of Thermal-Control Coatings for Spacecraft," Progress in Astronautics andAeronautics: Thermophysics of Spacecraft and Planetary Bodies,Vol. 20, Academic Press, New York, pp. 411-438, (1967).

H 9. Millard, J. P., "Results from the Thermal Control Coatings Experi-ment on OSO-III," Progress in Astronautics and Aeronautics: ThermalDesign Principles of Space and Entry Bodies, Vol. 21, AcademicPress, New York, pp. 769-795, (1969).

10. Arvesen, J. C., "Spectral Dependence of Ultraviolet-Induced Degrada-f] tion of Coatings for Spacecraft Thermal Control," Progress in Astro-

nautics and Aeronautics: Thermophysics of Spacecraft and PlanetaryBodies, Vol. 20, Academic Press, New York, pp. 265-280, (1967).

Preceding page blank 69

Ui. Slemp, W. S. and :Hankinson, T. W. E., "Environmental Studies ofThermal Control Coatings for Lu-nar Orbiter," Progress in Astro-nautics and Aeronautics: Thermal Design PF-inciples of Space andEntry Bodies, Vol. 21, Academic Press, New York, pp. 797-817, i(1969).

12. Caldwell, C. R. and Nelson, P. A., "Thermal Control Experiments

on the Lunar Orbiter Spacecraft," Ilogress in Astronautics andAeronautics: Thermal Design Principles of Space and EntmyBodies,Vol. 21, Academic Press, New York, pp. 819-852, (1969).

13. Gilligan, J. E., "The Optical Properties Inducible in Zinc Oxide,"Progess in Astronautics and Aeronautics: Thermophysics of Space-craft and Planeet a Bodies, Vol. 20, Academic Press, New York,pp. 329-361, (1967).

146D. B., "Photodescription in Zinc Oxide Semiconductor,"The Journal of Chemical Physics, C8, PP. 870-873, (1958).

15. Baery, T. I S., "Solar-Raa"The Reactions of Oxygen at Darkand Irradiated Zinc Oxide Surfaces," Proceedings of the RoyalSociety: Series A, 255, pp. 124-144, (1968).

16. Glemza, R. and Kokes, R. J., "Transient Species in Oxygen Take-Upby Zinc Oxide," Journal of Physical Chemistry, C6, PP. 566-568,(1962).

17. Ibid.

18. Blakemore, J. S., "ASolar-Radiation-Induced Damage to OpticalProperties of ZnO-Type Pigments," Lockheed Missiles and SpaceCompany, Contract No. NAS8-11266, pp. 6-8, (September, 1965).

19. Sklensky, A. F., MacMillan, H. F., and Greenburg, S. A., "Solar-Radiation-Induced Damage to Optical Properties of ZnO-Type Pig-ments," Lockheed Missiles and Space Company, Contract No. NAS8-18114, pp. 1-11-1-15, 3-21-3-24, 4-1-4-7, (February, 1968).

20. Blakemore, J. S., "A Model for Extraterrestial Solar Degradation

95152, Interi~m Report No. 1, pp. 16-19, (September, 1967).

70

*iit

HL 23. Greenburg, S. A., MacMiU.an, H. F., and Sklerisky, A. F., "DamageMechanisms in Thermal Control Coating Systems," Lockheed Missilesand Space Company Technical Report AFTL-TR-67-294, pp. 92-94,(September, 1967).

24. Ibid, pp. 78, 92.

25. Compton, D. M. J., Firle, T. E., and Neu, J. T., "Mechanisms of

Degradation of Polymeric Thermal Control Coating.s," Technical! Report AFML-TR-68-334, Part 1, pp. 48-50, (January, 1969).

26. Greenburg, op cit, pp. 66-68.

f I 27. Greenburg, op cit, pp. 92-94.

i 28. Thid.

Li 29. Coufova, P. and Arend, H., "The Defect Structure of Oxygen Deficient

Rutile," Czechoslovak Journal of Physics, 11, pp. 845-847, (1961).

L 30. VonHippel, A., Kalnajo, J., and Westpahl, W. B., "Protons, Dipoles,and Charge Carriers in Rutile," Jotxnal of Physics and Chemistry of

VjSolids, 23, pp. 779-799, (1962).

31. Yahia, J., "Dependence of the Electrical Conductivity and Thermo-electric Power of Pure and Aluminum-doped Rutile and EquilibriumOxygen Pressure," Physical Review, 130, pp. 1711-1719, (1963).

32. Ashford, N. A., and Zerlaut, G. A., "Development of Space-StableThermal-Control Coatings," lIT Research Institute Report No.

I U6002-77, pp. 45-52, (July, 1969).

34. Zerlaut, G. A., "Development of Space-Stable Thermal-Control Coat-ings," lIT Research Institute Report No. U6002-94, Contract No.NAS8-5379, p. 70, (November, 1970).

35. Personal Communication from AFML/MBE (March 1970).

36. Greenburg, op cit, pp. 55, 90-91.

37. Firle, T. E. and Flanagan, T. M., "Mechanisms of Degradation ofPolymeric Thermal Control Coatings: Effects of Radiation onSelected Pigments," Gulf General Atomic Incorporated, Report No.AFML-TR-68-334, Part II, pp. 91-93, (March, 1970).

71

38. Zerlaut, G. A., Firestone, R. F., Rayiunas, V., Serway, R., andRubin, G. A., "Pivelopment of Space-Stable Thermal Control Coatings,"IIT Research Institute Report No. U6003-31, Contract No. NAS 8-5379,p. 7, (November, 1965).

39. Zerlaut, G. A., Noble, G., and Rodgers, F. 0., "Development of

Space-Stable Thermal Control Coatings," IIT Research InstituteReport No. U6002-55, Contract No. NAS8-5379, pp. 65-67, (September,1967).

40. Zerlaut, G. A. and Rubin, G. A., "Development of Space-StableThermal Control Coatings," IIT Research Institute Report No. U6002-36, Contract No. NAS8-5379, pp. 42-43, (February, 1966).

41. Millard, op cit, pp. 771-778.

42. Zerlaut, G. A., Noble, G., and Rcgers, F. D., "Development ofSpace-Stable Thermal Control Coatings," IIT Research InstituteReport No. U6002-59, Contract No. NAS8-4379, pp. 51-52, (January,1968).j

43. Sklensky, op cit, pp. 5-3-5-4.

44. McGinnis, W. J., U. S. Patent 3,410,708, (Novermber 12, 1968). .145. McGinnis, W. J., U. S. Patent 3,510,335, (May 5, 1970).

46. McGinnis, W. J., E. I. duPont de Nemours and Company, PigmentsDivision, Wilmington, Delaware, Personal Communication, (March 18,1971).

47. Zerlaut, Fireston, Rayuinas, Serway, and Rubin, op cit, p. 6.

48. Crank, J. and Park, G. S., Diffusion in Polymers, Academic Press,New York, pp. 4-5, (1968).

49. Ibid.

50. Barrer, R. M. and Chio, 11. T., "Solution and Diffusion of Gases andVapors in Silicone Rubber Membranes," Journal of Polymer Science, 10,pp. 111-138, (1965).

51. Barrer, R. M., "Permeation, Diffusion, and Solution of Gases inOrganic Polymers," Transactions of the Faraday Society, 35, Pp.628-643, (1939).

52. Stannett, V., "Simple Gases," Diffusion in Polymers, edited byCrank, J. and Park, G. S., Academic Press, New York, pp. 41-73,(1968).

72

"U

53. Ibid.

54. Barrer, R. M., "Some Properties of Diffusion Coefficients in Poly-[ mers," Journal of Chemical Physics, 61, pp. 178-189, (1957).

55. Barrer, R. M. and Skirrow, G., "Transport and Equilibrium Phenomenain Gas-Elastomer Systems. I. Kinetic Phenomena," Journal of PolymerScience, 3, PP. 549-563, (1948).

I 56. Stannett, loc cit.

57. Scott, D. W., "Osmotic Pressure Measurements with Polydimethylsil-oxane Fractions," Journal of the American Chemical Society, 68, pp.H 1877-1879, (1946).

58. Ohlberg, S. M., Alexander, L. E., and Warrick, E. L., "CrystallinityH and Orientation in Silicone Rubber," Journal of Polymer Science, 27,

pp. 1-17, (1958).

59. Roth, W. L. and Harker, D., "The Crystal Structure of Octamethyl-spiro(5,5)Pentasiloxane: Rotation about the Ionic Silicon-OxygenBond," Acta Crystallographica, l, pp. 34-42, (1948).

[J 60. Kammermeyer, Karl, "Silicone Rubber as a Selective Baxrier," Indus-trial and Engineering Chemistry, 49, pp. 1685-1686, (1957).

I 61. Barrer, R. M., Barrie, J. A., and Raman, N. K., "Solution and Diff'-sion in Silicone Rubber, II-The Influence of Fillers," Polymer, 3,pp. 605-614, (1962).

! 62. Barrer, R. M., Batrie, J. A., and Raman, N. K., "Solution and D]i.ffu-

sion in Silicone Rubber, I-A Comparison with Natural Rubber,"Polymer, 3, pp. 595-703, (1962).fL

63. Barrer and Chio, loc cit.

64. Barrer, R. M., Barrie, 3. A., and Rodgers, M. G., "HeterogeneousMembranes: Diffusion in Filled Rubber," Journal of Polymer Science,1, pp. 2565-2586, (1963).

I 6. VanAmerongen, G. J., "The Effect of Fillers on the Permeability ofRubbers to Gases," Rubber Chemistry and Technology, 28, pp. 823-826,(1955).

66. Barrer, Barrie, and Rodgers, loc cit.

67. Barrer, Barrie, and Raman, "Solution and Diffusion in SiliconeRubber, II, loc cit.

68. Barrer, Barrie, and Rodgers, loc cit.

73

69. Rayleigha, lord, Philosophical Y-gazine, _34., pp. 481-5022 (1892).

70. EFrrer, Bari-c n oigrb t

71. P-rrer, N-rrik', zazl ftaman' GC~ cit.

722. R~rrr, EPArric, :~!am yl rv9erz, foe cit.

73. Zin-"-., W. A., 'R]inof Eouilibriumm Cont,-;2'. Aal-;e to isiauid andSolid COmstitmtioan,' Contct Angle, Wettabili!,y, and Adhx-sion,,American Cbeizical Society, Wiashington, D.C., pp. 2, 46, (1963).

7k.. zktmt op cii., PP. 4-5.

7z. 75."rnn, op cit, Pp- 3-4, 3b-4-.1

710. Adamson, A. W., ?hclCezStry of Siurfaces, :;econ-d E!itein,

Interscierr -e %~bJ shers, N~ew York, ppv. 357-363, (19967).

o'% the Free Surface E-mergy- of SolidS T," Joairna1 of 7-ksical i-

istr-, !!2, nmp. aft-W8, (1936).

7/8. Awa.-Onv op cit, Pp. 370.

79. ShXrn E. G., an Zi=-n Wo A., "Uier !,irmitz to the ContactAngles of Liud to Solids,- !%ntact Anlwettabiliit andAdhesion, AnrCenCemcall Socielty, WashiigtoEn, -)C. pp145-

4 1~57, U963).ao. ?olani, I .Te Proverties of Glass Sur--aces, Wile-Y, !iew oQ* P. 1434-.4307 UIj).

81. R{oser-, F., -Approxi-mating time, Attractive Forces of Adhesion fo;ýLss arnd OVher ;urfaf-es,' AMl "Alletin, U-24:1, vp. 62-66, (1950).

42e. M~oser, F., _'ettinrW Fhenoneria-A Prans &,r y~ee mi!irF Alhesio,'The Glass Industry, pp. 201-203, 228, 232. (April, 19a56,.

83. Mb~er. F., -Pollmeric Adhzesi-;es for Glzzs,ý Plasticz ?ez1rimlig,--:) 7T,,805, (095b.

Divisieý;I, Personal Cb= axication, (March 29,-, §'r.

85 . - ., an:; Maovrtin, . .ý'. 'cajxr ~

- 7-

I 87. Gagnon, D. W., Chemical & Polymer Development, Owens-Illinois,Toledo, Ohio, Pecrsonal Comaxdeation, (August, 1970).

S88. Campbell, op cit, pp. 28-37.

r 89. Bartell, F. E., and Zuidema, H. H., "Wetting Characteristics of

Low Surface Tension Such as Talc, Waxes, and Resins," Journal ofAmerican Chemical Society, 58, pp. 1449-1454, (1942).

90. Dettre, R. H., and Johnson, R. E., "Contact Angle Hysteresis, II.Contact Angle Measur-ements on Roughi Surfaces," Contact Angle,Wettability, and Adhesion, American Mhemical Society, Washington,

"":-•, pp. 136-144, (1964).

91. Adamson, A. w., Physical Chemistry of Surfaces, Second Edition,[1 .Interscience Publishers, New York, pp. 35-363', (196!).

| 92. Bartell, F. E., Culbertson, J. L., and Miller, M. A., "Altcrationof the Free Surface Energy of Solids I," Journal of PhysicalChemistry, 40, pp. 881-889, (1936).

93. Shafrin. E. G., and Zisman, W. A., -'Upper Limits to the ContactF!nles of Liwuids to Solids," Contact Angle, Wettability. andAdhesion, American Chemical Society, Washington, D.C., pp. 145-157, (1963).

94. Moser, F., "Approximating the Attractive Forces of Adhesion for

Glass and Other Surfaces," ASTM %-lletin, TP248, pp. 62-66, (1950).

- 95. Zerlaut, G. A., and Rubin, G. A., "DeveloDment of Space-Stable1 Th- mal Control Coatings," IIT &-!search Inzt.tute Report No. U6002-

" NAS8-5379, pp. 42-43, (February, 1966).- -- - - -'_..

96. Zeraut, G. A., "Develops et of Space-Stble Thernal Control Coat-ings," lIT Research Institute Report No. U6002-9o4, !iS8-5379, pp.

70-71, (Foveember, 1970).I•l ~97. Canibell.. A. n-T?.-7!-b?, Pt. i, op cit, pp. 2b-32.

:1-

LII

ILII I

I

�LII

I]LI APPENDJJC

LI OXYGE4 TRANSPORT DATA

�LILI[I

'LILI

�Li[1U

II

* preceimg page blank

:1

0\ UN Er, n U,t-- u \0 0 _:0 -1 w1c

U:) 0) vi \4) 0IC7 0 H CU C

00

CU

E-44 0 HHHH

004cm

to -Ir ri Hi- -01

00

0- 0 e *

Ca tz

ciiE-4 Ei

05 0

'-4

CP 04 "0 O a\- 0'

0 0

L0

'0 oU

A4 ~ cn cri cl\ cu ,E- H p-t n c

0C

0

*1P4~~ '~~

0

0c'0

0 ( 0 c

x ~ -- l. x0i El 00 c- If -I-

0 0O 04. H 4.r:

r40

4) 0-

L0 cfli

E-4l

.C1

79

Li

Table XVII - Oxygen Transport Constants for Oxygen inPolymonomethylsiloxane/Rutile Compositions

TP x i0 ST (cc(STP)sec/ D x 10' (cc(STP)/

cm2 /cm/cm Hg) (cmw"/sec) cc/atm)

Sample 10-0% Rutile

o 1.68 2.68 0.47625 2.83 5.02 0. 42832 3.29 5.55 o.J45050 4.72 7.92 0.45360 5.95 10.2 0.443

Sample 11-5.2% Rutile

0 1.34 2.49 o.40925 2.26 h. 56 0.37632 2.70 5.02 O.40950 3.86 7.52 0,39060 4.35 8.25 0.401

Sample 12-10.4% Rutile

25 2.13 4.o6 0.39932 2. h4 4.51 o.41150 3.54 6.66 O. 4o460 4.22 7.62 o.421

S-mpDle I1-90-74, Rut.ilc

0 0.95 2.09 0.34825 1.87 3.78 0.37832 2.08 4.65 0.34150 3.05 6.36 o. 36460 3.86 8.13 0.366