II

NASA CONTRACTOR REPORT

40 h ,-. ..../_,.,_..,.. - . . ;:" ;'; _...,: -I e U

PHASE CHANGE INDICATORS FOR SUBAMBIENT TEMPERATURES

by D. R. Kasanof and E. Kimmel

Prepared by i TEMPILO CORPORATION New York, N. Y. for LatzgZey Research Center

NATIONALAERONAUTICS AND SPACE ADMINISTRATION WASHINGTON, D. C. MARCH 1970 ._

TECH LIBRARY KAFB, NM

00b0899 c-w, J NASA CR-1576

/PHASE CHANGE INDICATORS FOR SUBAMBIENT TEMPERATURES

bL&~Q- By D. R. Kasanof and E. Kimmel

Distribution of this report is provided in the interest of informationexchange. Responsibility for thecontents resides in the author or organization that prepared it.

\ p- ' '\.prepared under ContractNO. L/.@&J31.-566by Luc.g9 "-.._--. TEMPILO CORPORATION New York, N.Y.

for Langley Research Center

NATIONAL AERONAUTICS AND SPACE ADMINISTRATION

For sale by the Clearinghouse for Federal Scientific ond Technical Information Springfield,Virginia 22151 - CFSTI price $3.00 PHASE CIIANGE INDICATORS FOR SUBAMBIENTTEMPERATURES

By D. R. Kasanof and E. Kinnnel TEMPIL" CORPORATION

Forty candidate organic compounds were evaluated as possible fusible temperature indicators for use in the range -50°F to -150°F. Of these, thirty with melting points ranging from -37°F to -162°F were found to be useful as temperature indicators. Nine candidates were rejected because they did not yield opaque, crystalline coatings by any of the application procedures attempted. One candidate material was rejected because it proved to be unstable. Phase studies of several binary systems were also carried out.

A number of application procedures were evaluated. An aerosol-spray procedure was adopted because it consistently yielded opaque coatings, was relatively simple to employ and could probably be adapted for use in the Langley Field test apparatus.

The indicators were foundto signal properly at a reduced pressure of 2mmercury, as well as at sea-level barometric pressure.

INTRODUCTION

In the July 1964 issue of the Journal of the American Institute of Aeronautics and Astronautics, Messrs. RobertA. Jones and JamesL. Hunt of the NASA Langley Research Center published a paper captionedof "Use Temperature-Sensitive Coatings for Obtaining Quantitative Aerodynamic Heat Trans fer Data".

This paper described the use of Tempilaq" phase-change temperature indicators for obtaining accurate heat-transfer data on models of complex shape in wind'tunnel tests. All of the work described was carried out using temperature indicators having ratings above 100"F, the lowest temperature for which these materials are commercially available.

In a letter of March 4, 1965 Tempil" Corporation proposed to under- take the development of phase-change indicators which could be used to extend this technique to temperatures well below the normal climatic range. NASA Contract NAS1-5667, dated December 9, 1965, authorized an investigation to develop materials and corresponding application techniques that couldbe used as phase changetemperature indicators in the temperaturerange of -50°F to -150"F, andwould be suitablefor aero- dynamic heat transfer testing by the method described in the Jones and Hunt paperreferred to above. The indicators wouldbe used at environ- mentalpressures ranging from 2mm Hg to oneatmosphere.

The followingdesired characteristics of the indicators wereoutlined in the contract to serve as developmentguidelines:

a.me phase change of theindicators should be clearly visible so that it canbe recorded with black and white film.

b. The indicatingtemperature should be unaffected by ambient pressures from 2mm Hg to oneatmosphere.

c. The indicatingcoating should have a veryshort response time so thatthe phase change temperature will beunaffected by the heating rate.

d. The method forapplying the indicator should besuch that repeatability of results canbe obtained.

e. The indicatorsshould have a low order of toxicity and shouldbe chemically inert.

LITERATURE EXPLORATION - The targetproperties -- meltingpoint range, low vaporpressure, long-term stability, inertness andlow order of toxicity -- appearto have fairly well restricted the chemical compounds availableas candidate temperature indicators. Examination of the literaturerevealed that two classes oforganic compounds, and aromatichydrocarbons, offered the most promising, readily available candidates. Our own experiencewith esters at elevated temperatures (over 100°F) led us tobelieve that this class of compounds mightalso havethe necessary crystallization properties, i.e., little or no tendencyto supercool below the freezing point and a strongtendency to formopaque, clearlyvisible crystals -- asopposed tothe formation of glasses which would be undesirable.

Aromaticethers, cyclic aliphatic ketones and aliphaticalcohols were otherclasses of organic compounds which offeredseveral additional candidate materials.

SOURCE OF CANDIDATE MATERIALS - All candidatesfor fusible temperature indicators actually used in -this work project were the best grades available from DistillationProducts Industries, Eastman OrganicChemicals Department, Rochester, N.Y. 14603. All are listed inthe current D.P.I. catalog, and were evaluated without further purification or treatment.

LOW-TEMPERATURETEST CHAMBER - The test chamberused throughout this work program was aneight cubic foot (2' x 2' x 2') capacity Model WF-8-175V

2 ~ "

Low Temperature Environmental Chamber manufactured by the Webber Manufac- turing Company, Inc.,Indianapolis, Indiana.

Air in this chamber was circulated by a fan and air temperatures were adjustable fromambient to -175°F. Provisions were availablefor evacuat- ingthe chamber topressures of less than 2mHg. On eachof the two sides of the chamber was a 2-inchservice port through which electrical leads, tubesor probes could be introduced.

The side-hinged,overlapping door on the front of the chamber was suppliedwith two 6-inchport holes for glove and/or hand insertion and manipulation of articles within the chamber underatmospheric pressure conditions. These ports were flanged and could be sealed if the chamber was tobe evacuated. The door was alsoequipped with a 12'' x 12" observa- tion window consisting of a special hermetically sealed 6-panethermal glass assembly.

SCREENINGOF CANDIDATE MATERIALS - Initialattempts to evaluate candidate materials involvedobservation of small quantities of liquid in glass test tubes and incapillaries. This method was found unsatisfactorybecause it was difficult to observe changes of state whichmight have taken place and becausethe length of time necessaryfor the relatively large quantities of liquidto reach thermal equilibrium with the air was found tobe excessively long.

After some experimentation,the following procedure was adopted and used in the initial screening of mostof the candidate materials:

A dropor two of fluid was placedin the center of a 75x25mm glassmicro- scope slide and coveredwith an 18mm coverslide, thus sandwiching a thin layer of material between glasssurfaces. This glass sandwich was then placed on awooden rackin an almost vertical position. The rack was paintedblack to form a darkbackground. Illumination was provided by four 73 watt bulbs, two on eachside of therack. The chamber was thencooled until crystallization of thecandidate occured or until a temperature of -175°F was reached.

Using this set-up, results were easilyobserved andphase-changes from liquid to solid, when accompanied by theformation of opaque white crystals, was verydramatic against the black background.

The abovemethod was relatively simple and enabledus to evaluate several samples at once. Results obtained were thought to correlate well withthose which would beobtained under the anticipated end-use conditions.

AEROSOLAPPLICATION OF INDICATOR MATERIALS - The followingaerosol application method was developed as a consequenceof our search for a final form forthe temperature indicators and correspondingpractical method for applying them. It was usedprincipally for the final temperature calibra- tion of the indicators, although it would alsohave been an excellent means

3 for initially evaluating the crystallization properties of eachmaterial -- the evaluation carried out on almost all of thecandidates by themicro- scope-slide method described in the previous section.

In employing thisprocedure, the substrate was cooledto a temperature below themelting point of theindicator. In the case of readilycrystallized materials,cooling the substrate to approximately thirty degrees Fahrenheit below themelting point of theindicator was sufficient.In other cases, coolingwith liquid nitrogen to temperatures approaching -300'F was found necessary.

A six-ounceaerosol container containing 60 gramsof theindicator liquid and 75 gramsof Freon 12 propellant -- no solvent -- was then inserted into the test chamber through oneof theport holes anda spray was directedonto the cooled test piece from a distance of six to eight inches. The aerosolcontainers were stored under ambient conditions prior touse, but apparently cooling of the indicator particles by evaporation of theFreon 12, by passagethrough the cold atmosphere of the chamber,and, finally, by contact with the cold substrate was sufficient to yield a white, opaque,adherant, solid coating several thousandths ofan inch thick.

All applications of indicatorcoatings by this methodwere carried outunder atmospheric pressure conditions. When calibrations were to be carried out in avacuum, application was followed by sealing of the chamber and evacuation.

UNSUCCESSFUL APPLICATION FORMS AND PROCEDURES

Suspensionsof materials in a volatilevehicle (Tempilaq') - Most of the indicatorcandidates studied are excellent solvents for organic compounds. Becauseof this,the choice of fluidsavailable as insoluble, volatile fluidsfor dispersion of thecandidates was severelylimited. After discardingwater because of its highmelting point, the remaining possi- bilities werenarrowed down toaliphatic hydrocarbons or some of their fluoro- and chloro-substitutedderivatives (Freons).

Severalsuspensions were madeup in hexane,the most volatile aliphatic hydrocarbonavailable which was a liquid understandard atmospheric tempera- ture and pressureconditions. The most seriousproblem encountered was the lack of volatility of the vehicle at the low temperature employed in this investigation. At a temperature of -lOO°F, hexane showedno tendency to evaporate. Results werethe same whetherapplication was by brush oraerosol spray. This was notcompletely unexpected, and confirmedexpectations that it might be necessaryto employ compressedgases such as methane, ethane or oneof the more volatile Freonsas a vehicle.This wouldhave involved some means ofsuspension of the liquid at ambient temperatures, followed by a procedurefor converting this suspension to a dispersion of fine crystals on coolingto below theindicator's melting point. Alternatively, developing a low-temperaturemilling procedure for forming the dispersions directly from thesolidified indicator and solvent wouldhave been necessitated. No work beyond theseveral hexane dispersions was carriedout.

4 i; Brushing of Super-Cooled Pure Liquids - By carefully cooling some ' indicator liauids to temperatures below their melting points,it was possible to brush-apply these super-cooled indicators onto a cooled substrate. The distukbance caused by brushing the indicators onto the surface was usually sufficient to cause instantaneous crystallization on the surface. The coatings which were opaque were also extremely uneven, and made up of rather large crystals and lumps. Furthermore, premature crystallization of the bulk prevented coating more than one or two square inchesof surface during any one experiment.

Aerosol Application of Pure Indicators Cooled to below their Melting Points - Prior to the use of pure indicators in aerosol containers stored at room temperaturesby the method described above, cooling and storage of aerosol containers at temperatures which ensured crystallization of the indicator was attempted. Excessive hardeningof the spray valves, severe reduction of Freon12 pressure and the formationof excessively large indica- tor crystals in the container-- all causedby cooling to the low temperatures of our experiments -- posed problems which were resolved by resortingto room temperature storage of the aerosol containers.

TENPERATURE MEASUREMENT - The Webber test chamber, described above, was used with a Model 104NY platinum resistance thermometer(.084 inch diameter) Purchased from the Rosemount Engineering Company, Minneapolis, Minnesota. Connecting this probe to the SpeedomaxH controller, after making the appro- priate change's in the circuitry of the controller, provided a low-lag, sensi- tive temperature measuring system.

Accuracy of the L&N Speedomax H recorder is2 0.3% of the temperature span; for our recorder this would be 0.83"F. The readability of this instru- ment is approximately 0.5"F. Accuracy of the platinum resistance thermometer, calibrated by Rosemount Engineering Co., was better than0.28"F. All calibra- tions carried out by Rosemount are traceable to N.B.S. standards.

"" OF INDICATORS - Candidate materials which froze CALIBRATION ~ TEMPERATURE " . rexly, i.e., supercooled less than 30°F below their melting points, were calibrated by the following procedure: A one-inch square of heavy gauge aluminum foil was perforated and slipped over the tip of the horizontally oriented resistance thermometer. With the tip of the thermometer pointing toward the observer, the offace the aluminum test piece hanging from it also faced the observer. (See sketch in AppendixF.) After cooling the aluminum test piece and thermometer to below the anticipated melting point, a coatingof the indicator material was sprayed onto the aluminum by the aerosol procedure described above. The environmental chamber was then allowed to warmslowly up until melting was visually observed.

The heating rate of the chamber was less than one degree Fahrenheit per minute. This low rate of temperature rise ensured thermal equilibrium

5 between the air, test piece andtemperature probe, and avoidedany dependence of the observed melting points on the heating rate. . Consideringthe accuracy of theinstrumentation and the reproducibility ofthe recorded melting points, the average (2 to 4 determinations)values obtained by this method,and tabulated in Appendix A for seventeen indicator materials, are probablyaccurate to within plus or minus 2°F.

Candidatematerial-s which did not readily form opaque coatings by sprayingonto a surface at -175°F orhigher were calibrated by thefollowing procedure: A circularphenolic specimen (diameter I%", weight 15 grams) was cooled to a temperatureapproximately 50°F below theexpected melting point (or to -175°F) and thendipped into liquid nitrogen and held submerged forseveral minutes.Immediately uponremoving from theliquid nitrogen, the specimen was sprayedwith the indicator. After allowing approximately fifteen minutes forthe temperature of the test piece to come to equilibrium with the air in the chamber, thetemperature was allowedto rise slowlyuntil melting occured.

The phenolicspecimen was suspendedapproximately 3 inches below the platinum-resistancetemperature probe during the calibration procedure. By means of a calibratedcopper-constantan thermocouple, it was determinedthat the air in the vicinity of the test piece was within onedegree of the temperatureof the platinum-resistance probe. Because of this slight possibletemperature variation, a possibletemperature lag of uncertain proportionscaused by thebulk of thephenolic specimen and a slightly higherrate oftemperature rise, it is estimatedthat the accuracy of the valuesobtained by this methodand tabulated in Appendix B for thirteen indicator materials is plusor minus 5°F.

EFFECTS OF VACUUM - The contractstipulated that the calibration of the temperatureindicators was to beGarried out under vacuum conditions as low as 2mm Hg as well as underambient pressure conditions.

Thermodynamic considerations(Clapeyron equation) and actualexperience withhundreds of organic and inorganic compounds have shown, however, that themelting point change resulting from a one-atmospherepressure change, the maximum possible in going fromatmospheric conditions to a complete vacuum, is of theorder of hundredths of a degree -- a negligible consideration in viewof the discussions of theexisting experimental errors in the various sections above.

Nevertheless, the seventeen indicators listed in Appendix A werere- calibrated at a pressure of less than 2mm Hg. The resultsare tabulated in Appendix C. Consideringthe normal experimental error under atmospheric conditions and theabsence of an air-bath to facilitate thermal equilibrium

6 J betweenthe aluminum test slide and thetemperature probe, the observed 1 differencesbetween the values in Appendixes A and C are notsignificant.

Of significance are the facts that noneof the indicators sublimed beforemelting, and thatthey all appearedquite stable, exhibiting no signs of boilingor gas evolution, upon melting.

All of the temperature indicators listed in Appendix B were also evaluatedunder vacuum conditions for sublimation before melting and stabilityafter melting. None of these indicatorssublimed before melting or showed signs ofdecomposition or evaporation upon melting. Determination of melting points was not attempted because there was no possibility of thermalequilibrium between the temperature probe and the plastic test piece underthe existing vacuum conditions.

BINARY MIXTURES - Some work was carriedout to determine the feasibility of using mixtures of candidate materials with satisfactory supercooling, freezing and crystal-opacity properties in order to obtain lower melting indicatorswith similar satisfactoryproperties. Threebinary systems were evaluated.

Two estermixtures, /Methyl Octanoate andHexyl Acetate/ HeptylAcetate, were found to yield easily frozen systems with melting points closelyfollowing a classicalphase-rule curve for a binarysystem. Melting pointsvalues for thesesystems were tabulated in the progress report ofJune 26,1967, but they are notrepeated here because the temperature measurement methodsused during these evaluations were less refinedthan the onedes- cribed above forthe calibration of thesingle component indicators and the recordedmelting points are undoubtedly in error. The eutecticmixture of theOctanoate system melted approximately 15°F below the melting point of the lowermelting component whilethe Acetate system eutectic melted approximately 9°Fbelow the melting point of thelower-melting component.

A system of aromatichydrocarbons, p-Cymene/m-Xylene, was found to yield mixtureswhich could not be frozen byany of the methodsused.

REJECTED CANDIDATE MATERIALS - Of theforty candidate materials evaluated, ten were rejected. Nineof thesecould not be made toyield opaque, solid coatings by anyof the methods described above. One, Dimethoxybenzene, appearedto be unstable. It frozereadily, but its meltingpoint decreased duringstorage, indicating the formation of a contaminatingdiluent, and its coatingsbubbled vigorously upon melting,contrary to the behavior of the thirtyuseable candidates. The tenrejected materials are listed in

Appendix D. '

REFRACTIVE INDEX MEASUREMENTS - At theoutset of this work program, it appearedthat refractive index measurements might provide a simple, room- temperature method fordetermining the suitability of a new lot of some particularindicator material. A Bausch & Lomb, Model9935-A20, Abbe-3L Refractometerthermostatted at 77°F (25°C) was used toobtain the refractive indexof each of the indicators listed in Appendixes A and B. These

7 refractive indexes are tabulated in Appendix E.

While therehas been no justification to date for carrying out a work programdesigned to correlate refractive index values with melting point valuesfor anyof the indicators, interest in oneor moreof them in the future might justify additional w6rk alongthese lines.

CONCLUSIONS

1. Fusibletemperature indicators for the temperature range -50°F to -150°F having the target properties outlined in the introductory section can be obtained from severalclasses of organic compounds -- mainly,however, from organic esters.

2. Packaging in anaerasol spray container in conjunction with one of the applicationprocedures outlined in the report appears to offer the simplest, most useful and most versatile application form of these indicators.

8 APPENDIX A

TEMPERATUREINDICATORS-MELTING POINT DATA

MELTING POINT (OF)

Anisole -37

EthylBenzoate - 38

MethylOctanoate -38

EthylOctanoate -52 m-Xylene -56 n-Heptyl Acetate - 60

PentylHexanoate - 62

Diethyl Malonate -65

Cy c 1open t anone - 67 ter t. -Butylbenzene -78

Hexyl Acetate - 89

HexylFormate -91 p - Cymene -97 n-PentylPropionate -102

PentylButyrate -103 MethylHexanoate - 104

PentylFormate -105

9 APPENDIX B

TEMPERATUREINDICATORS REQUIRING COOLING OF SUBSTRATE WITH LIQUID NITRdGEN

MELTING POINT DATA

MELTING POINT (OF)

1 2 y4- Trimethylbenzene -59

Mesitylene -76

EthylPropionate - 107

sec.-Butylbenzene -114

PropylPropionate -114

IsobutylIsobutyrate -118

n-Butylbenzene - 132

n-Butylalcohol -139

ButylButyrate - 139

MethylValerate -139 Ethylbenzene - 143 Ethyl Valerate - 147

EthylButyrate -161

10 APPENDIX C

TEMPERATURE INDICATORS MELTINGPOINT DATA IN VACUUM

MELTZNG POINT (OF)

Anisole -36

EthylBenzoate -38

Methyl Octanoate -39

EthylOctanoate -52 m-Xy 1ene -56 n-Hep tyl Acetate - 60 Pentyl Hexanoate - 64 Diethyl Malonate - 65

Cyclopentanone -66 ter t. -Butylbenzene -75

Hexyl Acetate -88

Hexyl Formate - 88 p-Cymene -97 n-PentylPropionate - 102

Pentyl Butyrate -101

Methyl Hexanoate -102

Pentyl Formate -102

!

11 I APPENDIX D

REJECTED CANDIDATE MATERIALS

n-Butyl Formate

n-Butyl Propionate

Cumene

m-Dimethoxybenzene

Ethyl Isovalerate

Isobutyl Propionate

Methyl Cyclohexane

3-Methyl Cyclohexanone

iso-Pentyl Alcohol

Propyl Butyrate

12 APPENDIX E

REFRACTIVE INDEX OF INDICATORS @77"F (25OC)

RJ3FRACTIVE INDM

Anisole 1.5155 n-Butyl Alcohol 1.3978 Butyl Benzene 1.4882 sec.-Butyl Benzene 1.4884 ter t .-Butyl Benzene 1.4907 1.4046 iso-Butyliso-Butyrate 1.3975 Cyclopentanone 1.4353 p-Cymene 1.4886 Diethyl Malonate 1.4117 Ethyl Benzene 1.4938 EthylBenzoate I. 5032 1.3906 Ethyl Octanoate 1.4165 1.3822 Ethyl Valerate 1.3987 n-Heptyl Acetate 1.4130 Hexyl Acetate 1.4077 n-Hexyl Formate 1.4056 Mesitylene 1.4975 MethylHexanoate 1.4042 Methyl Octanoate 1.4153 1 Methyl Valerate 1.3958 1.4104 Pentyl Formate I. 3972 Pentyl Hexanoate 1.4196 Pentyl Propionate 1.4066 PropylPropionate 1.3912 1,2,4-Trimethylbenzene 1.5025 m-Xylene 1.4957

13 ... . . ""

APPENDIX F

TEMPERATURE INDICATORCALIBRATION UNIT

*To L&N Recorder

f" Temperatu Probe

e Coated Test Slide

Front of Chamber

14