NA sA-r/_-/oogs7

NASA Technical Memorandum 100957 NASA-TM-100957 IAIAA-88-3300 !

°.

,6

Catalytic Ignition of Hydrogen andOxygen Propellants

Robert L. ZurawskiLewis Research CenterCleveland, Ohio

and

James M. GreenSverdrup Technology, Inc.NASA Lewis Research Center Group _ r;: ...._,:,-_,_ €.,:.,_,,,,:_Cleveland, Ohio

OCTI 6 19_gLA_,_,cv R_.c/:p,CH cENTER

tlt;-:_'_'t f': :-"L,_I iAfdPTOtl, Vl_:'S,ta,,,

Prepared for the., 24th Joint Propulsion Conference

cosponsored by the AIAA, ASME, SAE, and ASEEBoston, Massachusetts; July 11-13, 1988€

IJ/ A

https://ntrs.nasa.gov/search.jsp?R=19880015305 2020-05-01T23:01:09+00:00Z

AIAA-88-3300CatalyticIgnitionofHydrogenandOxygenPropellantsRobertL. Zurawski,NASALewisResearchCenter,Cleveland,OH;andJamesM, Green,.Sverdrup-Technology,Inc.,NASALewisResearchCenterGroup,Cleveland,OH

OCTI g 1988LAN,3LEYRE3EARCHCENTER

LiEp,e,R_' t-i,_.SAI-I/',Mi'qOr|. Vd-:GINIA

AIAAIASMEISAEIASEE24thJOINTPROPULSIONCONFERENCE

JULY11-13,1988/Boston,Massachusetts

Forpermissiontocopyorrepublish,contacttheAmericanInstituteofAeronauticsandAstronautics370L'EnfantPromenade,S.W.,Washington,D.C.20024 "

CATALYTICIGNITIONOFHYDROGENAND OXYGENPROPELLANTS

RobertL. ZurawsklNationalAeronauticsand SpaceAdmlnlstration

LewisResearchCenterCleveland,Ohio44135 -,:

and

JamesM. Green ,o

SverdrupTechnology,Inc.NASA LewisResearchCenterGrOup

Cleveland,Ohio44135

SUMMARY

An experimentalprogramwas conductedto evaluatethe catalyticIgnltlonof gaseoushydrogenandoxygenpropellants.Shell405 granularcatalystand amonolithicspongecatalystweretested. Mixtureratio,massflow rate,propel-lanttemperature,and backpressurewere varledparametricallyIn testingtodeterminetheoperationallimitsof the catalyticigniter.The testresults

o showthatthe gaseoushydrogenandoxygenpropellantcombinationcan be ignited_ catalyticallyusingShell405 catalystover a wlde rangeof mixtureratios,I

"' massflow rates,and propellantinjectiontemperatures.Theseoperatlngcondl-tlonsmust be optlmlzedto ensurereliableignitionfor an extendedperiodoftime. A cyclicllfeof nearly2000,2-secpulsesat nomlnaloperatingcondl-tlonswas demonstratedwith the catalyticigniter. The resultsof theexperlmentalprogramand the establishedoperationallimitsfor a catalyticigniterusingthe Shell405 catalystare presented,

INTRODUCTION

The hydrogen/oxygenpropellantcombinationIs commonlyused for space pro-pulslonsystemsbecauseof its high specific Impulse,fast reactionrate, lackof toxicity,and excellentregenerativecoolingcapability; However,thlsblpropellantcombinationis not hypergollc. Ignitionof liquidpropellantrocket englnesfueled by hydrogenand oxygen must be accomplishedby an Ignlterdevice which releasesheat and initiatesreactionof the maln propellants. Anumber of ignitiontechniqueshave been investigatedfor liquidpropellantrocket enginesand combustiondeviceswlth varyingdegreesof success. Theseignitiontechniqueshave included;pyrotechnic,hypergollc,electrlc,hot'gas

" tapoff,ionic, passivethermal,dynamic thermal,and catalytlcignition(ref. 1). Catalytlc ignltlonis a promisingtechniqueand has severalpotentlaladvantagescomparedto the Other ignitiontechniques. This report presentstheresultsof an experlmentalprogramconductedto investigatethe operatlonalcharacteristlcsof a catalytlcigniterfor the gaseous hydrogenand oxygen pro-pellantcomblnatlon.

CatalytlcIgnltlonis a simple conceptWhich has flown In space for tlmeperiodsover 8 years, performingthousandsof quallfledcycleswith monopropel-lant hydrazlne. CatalytlcIgnltershave a mlnimumcomponentcount whlch makesthem slmple, safe, llghtwelght,and Inexpenslve. The simpllcltyof the conceptis demonstratedby the fact that it requlresno externalenergy source. In

addition, catalytic igniters are passive in operation and present no radio fre-quency interference problems. These attributes make catalytic ignition compet-itlve with conventional igniters such as the spark torch and hypergoltc systemfor use in liquid bipropellant rocket engines.

The feasibility of using catalysts to promote the reaction of hydrogen andoxygen was demonstrated in several research programs in the 1960's and early1970's. Early investigations by the NASALewis Research Center (ref. 2) study-lng catalytic ignition identified several granular catalysts such as Shell 405and Englehard MFSAthat are reactive with hydrogen and oxygen. Shell 405 usesan iridium catalyst agent on a porous aluminum oxide substrate. Shell 405,which Is very effective for the decomposition of monopropellant hydraztne(ref. 3), became a preferred catalyst for the ignition of hydrogen and oxygen.An extensive experlmental data base was created for catalytic ignition ofhydrogen and oxygen propellants using Shell 405 catalyst (ref. 4). However,the longevity and reliability required for a catalytic igniter to be competl-tlve with spark torch igniters for use with hydrogen and oxygen propellants wasnot demonstrated.

Granular catalysts such as Shell 405 possess a high specific surface areaof metal but also have shortcomings for catalytlc ignition of hydrogen and oxy'gen propellants. These include the Inherently high pressure drop of a tightlypacked granular catalyst bed and the attrition due to thermal and mechanlca(loads on the catalyst particles. Monolithic catalysts are unitary structureswhich offer the potential advantages of very low pressure drop, attritionresistance for longer life and design flexibility. The NASAAmesResearchCenter investigated monolithic catalyst beds for hydrazlne reactors in theearly 1970's and demonstrated a monolithic catalyst wlth activity and perform-ance comparable to Shell 405 catalyst (ref. 5). A monolithic sponge catalystwas tested in addition to the Shell 405 granular catalyst in the present pro-gram to explore these potential benefits with hydrogen and oxygen. The mono-llthlc catalyst consisted of a carbon sponge substrate coated with rhenium togive it structural integrity and iridium as the catalyst agent.

The goal of the present experimental program was to build on the estab-ltshed experimental data base for catalytic ignition, exploring areas such aslow temperature ignition, monolithic catalyst beds, and propellant injectiontechniques, and to improve the llfe and rellability of the catalytic igniter.Catalytic igniter hardware was designed using generalized design guidelines andscaling criteria for catalytic igniters that were developed through analysis ofexperimental data from reference 4. A facility was constructed to test thecatalytic igniter under space-simulated conditions. Finally, testing was con-ducted to establish the operational limits of the catalytic igniter over a wlderange of operating conditions. ,.

In thts report the results of the experimental program are presented.The catalytlcignitertest hardware,test faclllty,and test proceduresaredescribed. The igniteroperatlonalcharacteristicsfor both the granularShell 405 catalystand monollthlcsponge catalystare discussed. Finally,theresultsof tests to characterizethe performance,rellabillty,and ilfe of thecatalytlcigniterusing the Shell 405 granular catalystare presented.

:,APPARATUS

Test Hardware

A schematicof the catalyticIgnlterassemblyis shownin figure1. Thehardwareconsistsof an upstreaminjector,spoolplece,downstreaminjector,

: and nozzle. The reactorof the catalyticigniteris containedin the spoolpieceand containsa mIxlngchamber(to enhancemixingof thehydrogenandoxy-gen)and thecatalystbed. Bothof the injectorunitsare madeof type304stainlesssteel,whilethe spoolpieceand nozzleare madeof HastelloyX.Hydrogengas andthe primaryoxygengas are introducedintothereactorbymeansof a cross-drilledinjectionschemein theupstreaminjector.The hydro-gen gas flowsaxiallyfromthe fuel inletintothe reactorthroughtwelveO.OBl-cm(0.032-In.)diameterinjectorholes. The oxygengas flowsfromtheoxidizerInletin throughthe annulusof the injectorintoO.157-cm(0.062-In.)diameterholesthatintersectwiththe nine0.053-cm(O.021-1n.)oxygenor|f-ices,and thenflowsaxiallyintothe reactor. Figure2 showsa schematlc.ofboththe upstreamanddownstreaminjectors.TheoxygeninletisIsolatedfromthe catalystbed and fromtheoutsideusingrubberO-rlngseals. The upstreamInjectoris connectedto the spoolpiecewith boltsthatpassthroughcompres-sionsprings.Withthesespringsin place,the upperinjectionfacepIatekeepsthe catalystbed underconstantcompressionso thatthe catalystremainstightlypackedif it attrlts.The reactorhas a chamberdiameterof 1.270cm(0.500in.)and a nomlnallengthof 5.080cm (2.000in.).

The hydrogenand primaryoxygenare Injectedintoa mixingchamberwherethe gasesare premIxed,flowthroughthe catalystbedwheretheyignite,andthenflow throughthe downstreamInjectorintothe nozzle. The nozzlehas athroatdiameterof 0.508cm (0.200in.)and an exitdlameterof1.270cm(0.500in.). Thisgeometryyieldsa valueof 6.25for both the contractionandexpansionratios. The mixingchamberis 1.270cm (0.500in.)longand consistsof an inertmaterialseparatedfromthe catalystbed by a sta|nlesssteelscreen. The catalystbed Is nominally3.810cm (1.500in.)long.

Downstreamoxygeninjectionincreasesigniterllfeby allowlngthe cata-lystto operateat lowertemperaturesand mixtureratiosthana designwhereall theoxygenis passedthroughthe catalystbed. The downstreaminjectorusesa cross-drilledinjectionscheme. The combustiongasesfromthe catalystbed passaxiallythroughtwelve0.102-cm(O.040-1n.)orificesIntothe down-streamchamber.The secondaryoxygenentersfromthe annulusthroughnine0.064-cm(0.025-In.)diameterorlflces. The hardwarecan thusbe operatedwitha lowmixtureratioin the catalystbed and a highermixtureratioIn thedownstreamchamber.Figure3 showsan explodedviewof the catalytlcigniter

.. hardware.

Two differentcatalystmaterlalswere tested,alongwith a numberof dlf-ferentmixingchambermaterlaIs.The majorityof the testingwas donewlth14 to 18mesh Shell405 catalyst,whichconsistsof iridiumcatalyston an alu-minumoxide(alumina)substrate.Specificationsfor theShell405 catalystaregivenin tableI. A monollthlccatalystwas alsotested,whichconsistsof acarbonspongesubstratewlthrheniumdepositedon the carbonfor strengthusingchemlcalvapordeposition(CVD). The activemetal,iridium,was thendepositedon the spongeusingCVD. Characteristicsof the monolithicspongecatalystaregivenin tableII. The mixingchambermaterlalstestedwere sillcasand,fusedzlrconlalsillca,zlrconla/magnesium,0.238cm (0.094in.)diameter,highcarbon

chrome steel balls, and 0.238 cm (0.094 In.) dlameter 440 stalnless steel balls.Figure 4 shows the Shell 405 catalyst and someof the mixing chambermaterialstested. Figure 5 shows the monollthlc sponge catalyst.

Temperature measurementswere taken inside the catalyst bed and on thebackwa11 at the locations shown tn figure 1. The backwa11 thermocouples wereused for diagnostic purposes and to collect heat transfer data from the tgnlter.Type K (chromel constantan) thermocouples were used for the temperature measure-ments. The bed thermocouples were labeled T1, T2, T3, and T4, with T1 locatednearest the upstream Injector In the mixing chamber and T4 farthest downstreamin the catalyst bed. These thermocouples were located at the inner radius ofthe reactor. Pressure measurements were taken Inside the catalyst bed (IPC)and in the downstream injector (PC) using strain gauge type pressure trans-ducers. The pressure transducers for these measurements were located justdownstreamof the primary and secondary injectors, respectively.

Test Facility

Testing of the catalytic Igniters was conducted In the Combustion ResearchLaboratories at NASALewis. The test facility, which was designed and builtfor the testing of Ignltlon systems and small gaseous hydrogen/oxygen rockets(up to 50 lbf thrust), is capable of testing at sea level or space simulatedaltitude, with ambient or chtlled propellants. Figure 6 showsa photograph ofthe test stand In the test facility.

A schematic of the propellant feed system for the test facility ts shownin figure 7. The feed system consists of two gaseous oxygen feed llnes and onegaseous hydrogen feed llne. Each oxygen line can deliver a maximummass flowrate of 0.050 kg/sec (0.110 lb/sec), and the hydrogen line can deliver a maxi-mumflow rate of 0.012 kg/sec (0.026 lb/sec). Sonic orifices are used to con-trol the mass flow rates In all propellant lines. These orifices deliver aconstant mass flow rate under choked flow condltlons. The massflow rate Isbased on the upstream temperature and pressure of the propellant. The upstreampressure Is regulated while the upstream temperature is ambient. Differentdiameter orifices can be Installed In the feed lines to deliver the mass flowrates necessary for testing. A gaseous nitrogen system purges all the propel-lant lines and the vacuumchamber. Propellants and gaseous nitrogen are dellv-ered from remote 16.55 MN/m2 (2400 pslg) trailers.

The altltude system consists of a 0.61 m (2.0 ft) diameter, 1.22 m(4.0 ft) long vacuum chamber driven by a two stage ejector system mounted onthe roof of the test cell. The vacuumchamber Is mounted on tracks for easy ..removal for hardware access or sea level testing. The ejectors consume3.63 kg/sec (8.0 lb/sec) of combustion air to draw a vacuumof less than0.0017 MN/m_ (0.25 psla) in the chamber.

The propellant inlet temperature to the igniter Is controlled by passingthe gaseous propellant through a boiling liquid nitrogen heat exchanger. Thepropellant lines are coiled inside the heat exchanger to provide more surfacearea for heat transfer. A partition separates the fuel and oxidizer coils inthe heat exchanger as a safety precaution. Liquid nitrogen Is fed from aremotely located dewar into the compartments of the heat exchanger. The levelof liquid nitrogen In the heat exchanger is automatically controlled by level

4

sensing devices that enable more liquid nitrogen to be addedautomatlcally whenthe levelis too low.

•, The controlsystemIn the cellconsistsof bothmanualand automatlccon-tro1. A programmablecontrolsystemIs usedto openand closevalvesduringthe run basedon manuallyenteredtimingvalues. Automaticandmanualabortsare avallableif a problemshouldoccurduringtesting.

The testcellhas bothhigh-speedand low-speeddataacquisitionsystems.The hlgh-speeddataacqulsltlonsystemrecordslO0readlngs/sec,averagesevery10 readings,andoutputsa datapointeveryI/I0sec. A hlgh-speeddata.pro-gramon themalnframecomputerperformscalculatlonswlththe raw data. Aminicomputeranda stripchartrecorderare usedto monltorthe fac111tyandobtalnlocal,immediatedataacqulsitlonduringtesting.

Test Procedure

Pulsemodetesting,whichsimulatestherepeateduseof the Ignlterwlth areusableengine,was conductedwiththe catalyticIgniter.Eachhotflre pulsewas 2 or 3 sec In duration,precededby a hydrogenleadto ensuresmoothstart-upandfollowedby a hydrogenlagto insuresmoothshut-down.The cata-lystbedwas thenpurgedwith nitrogengas to removeallof the residualpro-pellants.Newcatalystbedswereconditionedby flowinghydrogenthroughthebed for approximatelyI mln priorto the test. Thlshydrogenpurgeactivatedthe catalystbed by removingresidualoxygenadsorbedon the surface.

Catalyt!cIgnitertestswere conductedovera widerangeof operatingcon-dltlons. The mixtureratio(ratioof oxidizerto fuel)In the catalystbedwas varledfrom0.3 to 1.2,wlth the nominaloperatingvaluebeing1.0. Thetotalmassflowratethroughthe bed was variedfromO.O00gto 0.002?kg/sec(0.002to 0.006Ib/sec),with0.0018kg/sec(0.004lb/sec)beingthe nominaloperatingvalue. WhendownstreamoxygenInjectionwas employed,theoverallmixtureratiowas variedfrom2.0 to 12.0and theoverallmassflow ratewasvariedfrom0.0027to 0.0215kg/sec(0.006to 0.0473lb/sec).The initialtem-peratureof thecatalystbed was variedfromambienttemperaturesto -164.3°C(-263.7°F). Testingwas doneat ambientbackpressure(sea-level)as well asundera vacuum. Thetestingpresentedin the Resultsand Discussionsectionofthlsreportwas conductedwith 14 to 18 mesh Shell405 catalyst,with theexceptionof themonoI|thlccatalysttesting.

RESULTSAND DISCUSSION,.

IgnlterOperationalCharacterlstlcs

" The Ignlterwas InltlallycharacterizedwithoutdownstreamInjectlonInorderto protectthe nozzlefromthe hightemperaturesresultlngwhendown-streamoxygeninjectionwas employed. For thesecases,themixtureratioInthe catalystbedwas;

(_)b H°a- Nfa

5

where

(O/F)b catalyst bed mixture ratioa"

Hoa primaryoxygenmassflow rate,kg/sec

Wfa hydrogenmassflow rate,kg/sec

Severaltestswere initiallyconductedin the programwithonly Shell405catalystin thereactor. The inefficientmixingof the primaryInjectorresultedIn damaginghot spotsIn thecatalystbed. The hot spotswere due toaxialinjectionof a nonuniformmixtureof hydrogenandoxygenIntothe cata-lystbed. Thesetestsemphasizethe criticalityof the propellantinjectionconditionsto theoperationof the catalyticIgniter.A highefficiencypri-mary injectorwhichcan delivera uniformmixtureof hydrogenand oxygentothe catalystbed Is essentialto extended,rellableoperationof the catalyticIgniter.In subsequenttestsa mixingchamberwas placeddownstreamof theprimaryinjectorto premixthe propellantspriorto theirinjectionIntothecatalystbed.

Screenlngtestswere conductedto evaluatevariousmixingchambermaterl-alsfor use in the reactor. Thematerialstestedwere silicasand,fusedzlrconta/sIlica,fusedzlrconla/magneslum,highcarbonchromesteelballs,and440 stainlesssteelballs. Bothtypesof steelballswereableto withstandthe hlgh-pressure,hlgh-velocltygasesand thermalshockInsldethe reactor,whiletheothermaterialsquicklydeterioratedto varylngdegreeswlth use.Thlsdeteriorationincreasedthe pressuredropacrossthe reactorwlth tlmeandalsoadverselyaffectedtheflow characteristicsand mixingefficiencyof thediffusionbed. FlameflashbackIntothemixingchamberoccurredfrequently.The steelballsdid not deterioratewith tlmeas dld theothermaterials,buttheyexhibitedlocalizedmeltingandfusionif a flashbackoccurred. Thiswasmoreof a problemwiththe carbonsteelballsthanwith the stainlesssteelballs. Consequently,0.238cm (o.og4in.)diameter440 stainlesssteelballswereselectedfor use in the mixingchamber. The stainlesssteelballspro-vldedadequatepropellantmixingto preventdamaginghot spotsIn the catalystbed.

Testswerenextconductedto selecta nominalmassflowrate throughthecatalystbed. The massflowrate throughthe reactoraffectsthe pressuredropand the externalparticlemassand heat transferresistances,whichInfluenceconversionin the reactor. The pressuredropthroughthe catalystbed and tem-peratureweremeasuredas a functionof reactortotalmassflowrate. Theresultsof thesetestsare shownIn figures8 and 9, respectively.A nominalmassflowrateof 0.0018kg/sec(0.004lb/sec)was chosento maintaina reason- 'ablepressuredropthroughthe reactoras shownin figureB. Also,reactortemperatureis roughlyconstantat massflowratesabove0.0018kg/sec(0.004Ib/sec),suggestingthatexternalparticleheatand masstransferresis-tancesaremlnlmaIbeyondthispoint. It shouldbe notedthatgranularcata-lysttype,catalystparticlesize(whichlargelyaffectsthe internalheatandmasstransferresistances),and catalystbed volumewerenot variedduringthetestprogram. ResultsreportedherecouldchangeIf the catalystmetalIoad-ing,typeof catalyst,or the catalystparticlesizewerechanged.

A phenomenawhlch critically affects the operation of the Igniter Isflashback. Flashback is the suddenpropagation of the flame tn the upstreamdlrectton from the catalyst bed Into the mixing,chamberor to the Injector.

,. This phenomenoncan severely damagethe mixing chamberand Injector andadversely affects the performanceof the tgnlter. Figure 9 showsno Incidentsof flashback on any of the test flrlngs (flashback would be detected by largerlsesIn temperatureat T1 and T2).However,earlytestlngIndlcated_thatflashbackmay occurfor massflowrateslowerthan0.0016kg/sec(0.0035Ib/sec).

The hardwareused In thls testprogramwouldnotpermlta thoroughstudyof theflashbackphenomena.The hardwareemployeda showerheadtype Injectorfor boththe fueland oxidlzersuchthatthe propellantswere Injectedax1allyIn discretestreams.Thls led to regionsof nonuniformmlxtureratioIn thereactor.A mIxlngchamberwas thereforenecessaryto enhancemIxlngof thepropellants.However,meltlngandfusionof themIxlngchambermaterlalwlthflashbackalteredthe flow characterlstlcsIn thereactor,whlchmadethe studyof theflashbackphenomenadifflcult.A moreefflclentInjector(a premIxlngtypeInjectorwltha largenumberof elementsper unltarea,forexample)thatsuppllesa uniformmlxtureof hydrogenandoxygento the catalystbed wouldsubstantlallyImprovethe performanceof the Ignlterandmakeflashbackpre-dlctableas a functionof the flowcondltlonsIn the reactor. Nhllethe flash-backphenomenawas not thoroughlystudledIn thlsprogram,the effectsofvariationsIn flowveloclty,mixtureratlo,and IgnlterInternalgeometryonIgnlterflashbackcharacteristicshave beencharacterizedIn the past(ref.4).

Testingwas nextconductedto determlnea nomlnalmixtureratlofor thereactor.Thls operatlngmixtureratloshouldbe hlghenoughfor Ignltlon,whlleremalnlnglow enoughto prolongthe 11feof the catalystbed. Figure10showsthe effectof mixtureratioon the temperaturedistributionat the end ofa 3-secflrlng. Basedon thlsdata,a nomlnaImlxtureratioof 1.0was chosen.Operatingat reactortemperaturesabove648.9°C (1200°F) (a mixtureratioof1.0)11mltsthe 11feof the catalystbed. TI,whichwas locatedIn the mixingchamber,remalnedat amblenttemperature,Indlcatlngthatflashbackdld notoccur. T2 Increasedonly s11ghtlydue to axlalconduction,whlleT3 and T4Increased11nearlywlthmixtureratio.

Throughoutthe courseof testlngIt was observedthat the 1ocatlonof theflamefrontchangeddue to a varietyof factors. NotlceIn a comparisonofflgures9 and 10 at the sameoperatingconditions(mlxtureratlo= 1.0 andmassflowrate= 0.0018kg/sec(0.004Ib/sec))thatthe 1ocatlonof the flamefrontmoveddownstreamfromT3 (flg.9) to T4 (flg.10). The flowcharacteristicsofthemlxlngchamberand catalystbedmay havechangedwlth use due to channellng

•. of theflow,1ocallzedhot spots,or flashback,or the actlvltyof the catalystbed may havedegraded.Figure11 showsa transienttemperaturedlstrlbutlonfor a hot firetestat nomlnalflowcondltlons.The largemagnitudeof T2 (andT3) for thisflrlng,whlchwas theflrstflrlngof a freshcatalystbed, Indl-catesthattheflamefrontwas locatednearT2 and thereforethe catalystwasmuchmoreactivethanthe flrlngat a mlxtureratloof l.O In figure10. Thlsemphaslzesthefactthat the temperaturedistributiondependson the natureandamountof prevloustesting.

A factorthathad a majorImpacton the temperaturedlstrlbutlonand the1ocatlonof the flamefrontwas the propellantInlettemperature.A comparison

7

of flgures11 and 12 showsthe effectof Inlettemperatureon the translenttemperaturedistributionin the reactor.Ambienttemperaturepropellantswereusedfor the testfiringof figure11,whllechllledhydrogenand a prechlI1edcatalystbedwere usedfor the testfiringof figure12. As shownin fig, ..ure 12,whenthe catalystbedwas prechllledto -97.7°C (-143.9°F)andch111edhydrogenwas usedtheflamefrontmoveddownstreamto T4 in thereac-tor. Thlswas theonly locationin the reactorwherethe temperaturereachedavaluehlgherthan-17.78°C (0 °F).

Testingwas nextconductedto determinetheoveralleffectof Inltlalbedtemperatureon Ignitlonof hydrogenandoxygenuslngShell405 catalyst.Thebedtemperaturewas adjustedby flowlngchilledhydrogengas throughthe bedprlorto the testflrlng. The chllledhydrogengas and amblenttemperatureoxygengas werethenusedfor the testflring. An ignltionboundaryfor theShell405 catalystwas deflnedby varyingthe Initialbed temperatureandmix-tureratioIn thereactor. Theresultsare shownin flgure13. It wasobservedthatas temperaturedecreasesmlxtureratlomust be increasedtoachleveIgnltlonof the hydrogenand oxygenpropellants.A modelproposedbyKesten(ref.6) suggeststhatfor low-temperature,hydrogen-richsystemstheoverallreactlonrate stronglydependson the concentratlonof oxygenand thetemperaturewlthinthe catalystparticle.It Is suggestedthatcatalystactlv-Ity is temporarilyreducedby capillarycondensatlonor freezlngof waterIntheco]dcatalystpartlcles.If slgnlflcantquantitiesof waterblockenoughporesto Isolatelargesegmentsof the porousparticlefrom thereactants,thecatalystbedwouldbe renderedentlrelyineffective.Thlswouldexplalntheeffectsof catalystbed temperatureandmixtureratioon ignitionas shownInfigure13. Ignltlonof hydrogenand oxygenin thecatalytlcignlterwas demon-stratedwith inltialShell405 catalysttemperaturesas lowas -112.8°C(-171.1°F)and fuel inlettemperaturesas lowas -165.2°C (-265.4°F). Ignl-tlonis possibleat lowerinitlalcatalystbed temperatures,but the resultlngignitiontemperatureat thehighermixtureratiowouldbe detrimentalto thelifeof the catalystbed.

To characterizetheoveralloperatlonof the igniter,testlngwasconducteduslngdownstreamoxygenInjectlon.For thesecases,the totalmix-tureratlofor the ignlterwas;

I_ W°a+ Hobt Hfa

where

(O/F)t totalmlxtureratlo --

Hoa primaryoxygenmassflowrate,kg/sec

Hob secondaryoxygenmassflowrate,kglsec

Hfa hydrogenmassflow rate,kg/sec

For optlmumignlteroperatlon,the totalmixtureratioshouldbe nearstoichlometrlcIn orderto releasethemaximumamountof energyfor main pro-pellantIgnltlon,whllethereactormixtureratioremalnslow to Increasethe11feof the Ignlter.To determlnetheoptlmumoperatingcondltionsfor the

Igniter,the catalystbed and totalmixtureratioswere varied,whlleholdlngthe massflowratethroughthe reactorconstantat0.0027Ib/sec(0.006Ib/sec).

•. The resultsof thistesting,plottedin figure14,showthata boundaryexistsbetweenignitionand no Ignitionin the downstreamchamber. Fuelrlchmixturesin the downstreamchamberweredifficultto ignite. Ignitionin the downstreamchambercouldnot be achievedat totalmlxtureratioslessthan5. Oxidizerrichmixturesin thedownstreamchamberwereeasilyignited,evenat catalystbed mixtureratiosas lowas 0.4.

The igniterwas testedwithoutdownstreamInjectionat sea level(ambientpressure)and at spacesimulatedconditions.Theignitioncharacteristicsinthe reactorwerenot responsiveto variationsin nozzlebackpressure.Theflow throughthedownstreaminjectorwas chokedfor the rangeof reactorflowstested. Testswerenot conductedto determinetheeffectof nozzlebackpres-sureon the ignitioncharacteristicswhendownstreamoxygenwas employed.

MonolithicSpongeCatalystTests

Ignitertestswere conductedIn whlcha monollthlcspongecatalystreplacedthe Shell405 granularcatalyst.The uniquefeatureof thlsmonollthlccatalystwas thatthe activecatalystmaterialwas deposlteddlrectlyonto themonollthlcspongesupportby chemlcalvapordepositionwithoutthe useof a porous,highsurfaceareawashcoat.Porouswashcoatsare usedwlth noblemetalcatalyststoIncreasethe surfaceareafor dlsperslonof the catalystagentwhen the surfaceareaof the supportItselfIs low. However,washcoatscan be llfe-llmltlngtothe catalystbecausethelrcoefflclentof thermalexpansionor thermalconduc,tlvltymay be differentfrom thatof themonollthlcsupportor becausethewashcoatmay not adherewell to themonolithlcsupport. In addltlon,spongemonolithsare dlfflcultto washcoatbecausethewashcoattendsto depositInand clogthe poresof the sponge.

The monolithicspongecatalystwas not successfulfor ignitionof the gas-eoushydrogenand oxygenpropellantcombination. At nominaligniteroperatingconditions,themonollthlccatalystbed was onlyableto elevatethe tempera-turein the reactorslightly.This inabilityto Ignitethe propellantswas dueto the lowavailableactivesurfaceareaof the monollth.The porousShell,405granularcatalystpossessesa much largeractivesurfaceareaper unitvol-ume thanthe monollthlcspongecatalyst,whichhad only itsgeometricsurfaceareaavailablefor catalystdlsperslon.

The performanceof the monollthlcspongecatalystat nomlnalIgniteroper-.. atlngcondltlonsIs shownIn figures15 and 16. Figure15 showsthe effectof

massflow rateon exlttemperaturefor a mixtureratioof l.O. It was observedthatreducingthemassflow rateand,hence,the velocityof gasesthroughthemonolithicspongecatalystbed increasedthe exittemperature.Figure16 showstheeffectof mixtureratioon theexlt temperaturefor a massflow rateof0.0018kg/sec(0.004lb/sec).As:expected,the exlttemperatureincreasedwithincreasingmlxtureratlo. Ignitioncouldnot be achievedfor mixtureratiosbelow3.0 In themonoIIthlccatalystbed. At a mixtureratioof 3.0, Ignitionwas achievedwhichdestroyedthe monollthdue to theextremetemperature.Thepoorperformanceof the monollthlcspongecatalystin thesetestscan beattributedto two factors. The lowactivesurfaceareaof thecatalystpre-ventedIgnltlonat a nominalreactormixtureratioof 1.0,and the structural

durability of the catalyst was inadequate due to the use of the carbon sub-strate. Uponignition, the monolithic spongecatalyst was immediatelydestroyed.

.4

IgniterPerformance

In orderto obtaina measureof theoverallperformanceof the catalyticIgnlterusingShell405 catalyst,the C* efficiencywas calculatedfor therunswithdownstreamoxygeninjection.Thefollowingrelationwas used toestimateC* efficiency.

PcAt

Htot• Ceff - C* (lO0)th

where

C* experimentalC* efficiencyof the igniterelf

Pc pressure In the downstream chamber, MN/m2

At cross-sectlonalthroatarea,m2

Ntot total mass flow rate, Noa + Nfa + Nob, kg/sec

C* theoreticalcharacteristicvelocityfor igniter,m/secth

Figure 17 shows the varlatlonin C* efficiencywith the total mixtureratio of the catalyticigniter. The C* efficiencyreachesa peak value of79.2 percentat the stolchlometrlcmixtureratio for hydrogen/oxygen,8.0. Thelow valuesof C* efficiencycan be attributedprlmarllyto thermallosses andInefficientmixing of propellantsin the downstreamchamber.

PulseMode LlfeTests

Testingwas alsoconducted(withoutdownstreamInjection)to determinethelifeof theShell405 catalystbed. The testingconsistedof 2-secpulsesfol-lowedby a ]5-seccoo]downperiod. Afterthefirst]lO0pulses,an attemptwasmade to shortenthe cooldownperiodbetweenrunsby convectlvelycoollngtheexteriorsurfaceof the reactorwlthwater,as opposedto the normalcooldown .-by forcedconvectionof air overthe exteriorsurfaceof the igniter. Thedecreasedtimebetweenrunswas not sufficientto allowthe catalystbed to bepurgedof all the reactedgasesbetweenruns,andflashbackoccurred. Thisflashbackalteredtheflow characteristicsin themixingchamber,whichin turnalteredthe ignitioncharacteristicsof the igniter. Becauseof the adverse•effectsit caused,thewatercoollngwas stoppedafter10 pulses. Testingthenresumedas before,with the catalystbed sustainingignitionfor a totalof1980pulses.

lO

Figures 18 and 19 showthe characteristics of the igniter during thepulse-modelife tests. Only data from the first 1100 pulses are reported herebecausebf the changeIn performance after water coollng was attempted. Fig-

. ure 18 showsthe variationIn pressuredropacrossthe catalystbed and down-streaminjectoras the numberof pulsesincrease.The pressure-dropIncreasethroughoutthe testingcanbe.attributedto degradationof the mIxlngchamber•

.. materla]and catalystbed and to smallcatalystparticlescloggingthe down-streaminjectororifices.Figure19 showsthe varlatlonIn the temperaturedistributionIn thebed as the numberof pulsesincrease.The largevaluesofT2 andT3 indicatethatafterapproximately500pulses,thetemperaturedlstrl-butionreacheda steadyoperatinglimit,afterwhichT3 andT4 decreasedonlysllghtlyas the numberof pulsesincreased.

CONCLUDINGREMARKS

An experlmentalprogramwas conductedat NASA Lewisto investigatethecatalytlcIgnltlonof gaseoushydrogenand oxygenpropellants.Theobjectivesof the programwereto establlshdesignguidelinesfor catalytlcIgniters,developan experlmentaldata basefor catalytlcignitionof hydrogenandoxygenpropellants,and demonstratethe 11reand rellabilltyof the catalyticigniterfor rocketpropulslonsystems.Contributionsto an existingexperlmentaldatabasefor catalytlcignitionof hydrogenand oxygenpropellantsweremade in theareasof low-temperatureignition,monollthlccatalystbeds,and propellantinjection and mixing techniques.

The feaslbllity of catalytic ignition for reusable propulsion systemsusing hydrogen andoxygen propellants was demonstratedwith a cyclic ]ire testof nearly 2000.consecutive 2-sec pulses at nominal operating conditions, withno maintenanceto the igniter required. In other tests mixture ratio, mass

, flowrate,propellanttemperature,and backpressurewerevariedparametrically, to establishtheoperationallimitsat whicha catalytlcigniterusingShell

405 granularcatalystcan be rellablyoperated.The resultsof thesetestsshowthatthe gaseoushydrogenandoxygenpropellantcombinationcan be ignitedcatalytlcallyovera wide rangeof operatingconditions,but the ignitermustbe operatedwithintheseestablishedconditionsto ensurerellableignitionforan extendedperiodof time.•Ignitionwas demonstratedwith InltlalShell405catalystbed temperaturesas lowas -112.8°C (-171.1°F) and fuel Inlettem-peraturesas 1owas -165.2°C (-265.4°F). The ignitiontemperaturethresholddecreasesas themixtureratioIn the catalystbed is increasedand ignitionispossibleat lowerInltlalcatalystbed temperatures,but the resultingtempera-tureat the highermixtureratiowouldbe detrlmentalto the lifeof the cata-lystbed.

J.

TestswerealsoconductedUsinga monolithicspongecatalystwhichcon-sistedof a carbonspongesubstratecoatedwlthrheniumto give It structuralintegrityand iridiumas the catalystagent. Althoughthistypeof monoilthlccatalystdid notperformwell becauseItsactivesurfaceareaand structuraldurabllltywere inadequate,It Is believedthatothertypesof monoilthlccat-alystsholdpromisefor improvedlifeand reliabilltyof catalytlcignitersbecauseof theirlowerpressuredrop,improvedattritionresistance,andgreaterdesignflexibilitycomparedto granularcatalysts.

11

REFERENCES

I. Pollard,F.B.; and Arnold, J.H., AerospaceOrdnance Handbook,Prentice-HallInc., EnglewoodCliffs, NJ, 1966, Chapter3, j-

2. Roberts,R.N.,Burge,H.L.,and Ladackl,M., "Investigationof CatalyticIgnitionof Oxygen/HydrogenSystems,"R-6303,RockwellInternationalCorp.,CanogaPark,CA, Dec. 1965,NASACR-54657. '"

3. Russl,M.J.,"A Surveyof MonopropellantHydrazlneThrusterTechnology,"AIAA Paper 73-1263,Nov. 1973.

4. Johnson,R.J., "Hydrogen-OxygenCatalyticIgnitionand Thruster Investiga-tion; Volume I - CatalyticIgnitionand Low Pressure ThrusterEvaluations,"TRW-14549-600I-RO-OO-VOL-I,TRN SystemsGroup,RedondoBeach,CA, Nov.1972, NASA CR-120869.

5. "MonolithicCatalystBeds for HydrazlneReactors,"REPT-73-R-360,RocketResearchCorp.,Redmond,WA, May 1973,NASACR-132983.

6. Kesten,A.S.,and Sanglovanni,J.J.,"TransientModelof A HydrogenlOxygenReactor,"UARL-K910962-12,UnitedAircraftCorp.,EastHartford,CT, Feb.1971, NASA CR-120799.

TABLE I. - SHELL405 CATALYSTSPECIFICATIONS

Catalystcarrier:

Sizeand shape ............ 14 to 18 mesh rounded granules

Type ....................... Alumina, type RA-I

Surfacearea, m2/g (ft2/Ib) .............. 234 (I.14xi06)

Bulk density, g/cm3 (Ib/in.3) ............. 1.01 (0.0365)

Porevolume, cm3/g (in.3/Ib) ............... 0.25 (6.92)

Finished catalyst:

Catalyst type ........................ Iridium

Active catalystmaterial,wt % ................. 32.2

Totalsurface area, m2/g (ft2/Ib) .......... 118 (5.7613xi05)

Active surface area, pmol H2 adsorbed/g ............. 413

Active surface area, m_molH2 adsorbed/Ib .......... 1.87x105

TABLE II. - MONOLITHICSPONGE CATALYST CHARACTERISTICS

Catalystcarrier:

Sizeand shape, in...... 0.5 by 2.0 cylindricalvitreous foam plug

Type ............................. Carbon

Bulk density,g/cm3 (Ib/ft3) ........... . . . 0.06 (3.7467

Porosity,pores/cm (pores/in.t;percent ........ 39.4 (100); 96

Finishedcatalyst:

Catalysttype ........................ Iridium

Rheniumweight, g (Ib) ................ 2.345 (0.0052)

Iridiumweight, g (Ib) ................ 1.626 (0.0036)

12

(_ THERMOCOUPLE

(_) PRESSURETRANSDUCER

" SECONDARYOXYGENPRIMARYOXYGEN INLET

INLET _4.45

/-DOWNSTREAM" O-RING / INJECTIONUNIT

SEALS_ / ,\

rTEFLON O-RINGUPSTREAMIN- ,/ SEALS

JECTIONUNIT-, j-NOZZLE .

FUELINLET-,,

• 1.27

I

l L NIXINGI lL L-SPOOL CHAMBER

COMPRESSIONPIECESPRING _ 5.08

14.63

FIGURE1. - SCHEMATICOF CATALYTICIGNITERHARDWARE. [ALLDIMENSIONSINCENTIMETERS.]

UPSTREAM DOWNSTREAMINJECTIONUNIT INJECTIONUNIT

!iJj-SECONDARYl,IIr OXYGENINLET

_--.9 PRIMARY /-/12 COMBUSTION J/9 SECONDARYL 12 FUEL OXYGEN GASORIFICES OXYGEN

ORIFICES ORIFICES 0.102 CM DIAM ORIFICES0.081CM 0.053 CM 0.064 CM DI4_J,tDIAM DIAM

•FIGURE2. - SCHEMATICOF INJECTORUNITS.

13

SPOOLPIECE THERMOCOUPL

FIGURE3. - EXPLODEDVIEWOF CATALYTICIGNITERHARDWARE, FIGURE4. - CATALYSTAND DIFFUSIONBED MATERIAL.

FIGURE5. - MONOLITHICSPONGECATALYST, FIGURE6. - TESTSTAND,

14

SECONDARY -- FIREOXYGENLINE-I '('_) VALVE ,(T_

• r---n I• .. I I I _ I'"' SONIC ,-VACUUM

JGo_i__ ORIFICE , C,A_ERI

"" L_._IPRI{_IARYoxYGEN..... LINEII-_I_ _'_ J_1"_F_ S'O_ICoRIFICEFIRE _ I• LVIEWING J I'm"_"

I I _ EXC_HANGER ORIFICEt

HYDROGENLINEJ

FIGURE7. - PROPELLANTFEEDSYSTEMSCHEMATIC,

MIXING•CHAMBER CATALYSTBED

3 -- _ HYDROGENCOLDFLOW -- 1600----"{D'--- HOTFIRE TESTING -- 400 800 --

- ,2oo

600-- ou-

2- ,p - 300_. _ _ _ ..._"'"_ =_" °_....°/

-2oo_= _oo- ="--_ ,,, _ TIT2

--100 _" 200-- --'_3-_ T3400

_" "-_-'-- T4

0"1 I I I o o 0_?- -'?--L'_-_}-1 - -_- - J --'_'_0 .001 .002 .003 .OOq .005 .0005 .0010 - .0015 .0020 .0025 .0"030

TOTALMASSFLOWRATE,KG/SEC TOTALMASSFLOWRATE,KG/SEC

I I I I I I .I I I I I0 .002 .004 .006 .008 .010 .002 .003 .00q .005 .006

TOTALMASSFLOWRATE,LB/SEC TOTALMASSFLOWRATE,LB/SEC.

FIGURE8. - EFFECTOF MASSFLOWRATEON REACTOR FIGURE9. - EFFECTOF MASSFLOWRATEON REACTORPRESSUREDROP. [MIXTURERATIO= 1.0"3 SEC TEMPERATUREDISTRIBUTION.[MIXTURERATIO=RUN DURATIONFOR NOT FIRETESTS.] 1.0"3 SEC RUN DURATION.]

15

MIXINGCHAMBER CATALYSTBED

0 IGNITION- 1200 • NO IGNITION

600 -- 50 ITI 100T2 _ 0 0 0 0

._._" o_ 0T3 - 800 0 --,,oo T,4 ./ _ o:

./ _ _ o o=_ / _ _ -_o- . @ o200- / _,oo) ) m m _ -1oo_=

• ,oo o_" "/ _ " I -2oo

-2oo I I I I I I -2oo I I I I -soo .-0 .5 1.0 1,5 2,0 2.5 3,_ -400 ._ .6 .8 1.0 1.2

TIME,SEC MIXTURERATIO,O/F

FIGURE12. - EFFECTOF INLETTEMPERTUREON REACTOR FIGURE13. - EFFECTOF INITIALBED TEMPERATUREAND 'TEMPERATURETRANSIENT. [TOTALMASSFLOWRATE= MIXTURERATIOON IGNITION. [TOTALMASSFLOW RATE0.0018 KG/SEC(0.004 LB/SEC): MIXTURERATIO= 1.0" = 0.0018 KG/SEC(0,004 LB/SEC)" _ SECRUNDURA-3 SEC RUN DURATION.] TION.]

16

9 o

• _ _ = 0_18 KG/SEC(0_q LB/SEC)12 O M = 00014 K_SEC (0_3 LB/SEC)

% _"_-_ M = 0.0009 KG/SEC(0._2 LB/SEC)0 IGNITION qO

_ 10 . O . NOIGNITION _ _ _ D (_ _ [_ --D-- E]q 1_

- t

= 2Oloy I I I i1

_ _ 60N o

4 _ • • o

_ 402 I I I ' I

2 4 G 8 10 0 2 q 6 8T_AL MI_URE R_IO, (0_) t : TI_, SEC

FIGURE14. - EFFECTOFMI_URE RATIOON00_REAM IGNI- FIGURE 15. - EFFECTOFMASSFL_ R_E ONREACTORTION [T_AL MASSFL_ _TE = 0.0027 K_SEC (0006 TENPERATUREFORTHE_NOLITHIC SPONGE_T_YSTLB/SEC)_ 2 SECRUN00_TION] [MI_URE RATIO= 10]

O/F = 1.0---0--- O/F = 1.5

150 -- _'-O'_ O/F = 2.0 -- 300--"-,_-'-- O/F = 2.5

F..__._.._...__..._.-.._---_-.._ 80-0" loo-X" ...u--o--u_ 2oo_ ig. ,,=,,o..v--!'°"°'-"°" ,00_8

Gs I I I I I I0 2 q G 8 5 G 7 8 9 10

• TIME,SEC -- 0 TOTALMIXTURERATIO, (O/F)tFIGURE16. - EFFECTOF MIXTURERATIOON REACTORTEM- FIGUREi7. - EFFECTOF TOTALMIXTURERATIOON IGNITER

PERATUREFOR THE MONOLITHICSPONGECATALYST. C* EFFICIENCY.[REACTORMASSFLOWRATE= 0.0027[TOTALMASSFLOWRATE= 0.0018KG/SEC(O.O(}tl KG/SEC(0.006LB/SEC)-2 SEC RUN DURATION.]LB/SEC). ]

17

lzlO0 -- -- 200

o .,x,

• ' o - 180_=

1200 _ ! ,.

0,.

1100 -- 160

), I I I I I I

10000 200 400 600 800 1000 1200 ,-.NUMBEROF CYCLES,N

FIGURE18. :- VARIATIONOF REACTORPRESSUREDROPFORCYCLICLIFE TESTING. [TOTALMASSFLOWRATE=0.0018 KG/SEC(O.OOtlLB/SEC)"MIXTURERATIO= 1.0"2 SECRUNDURATION,] :_

MIXING

•CHAMBER CATALYSTBED

_,@ @ @--1600

800--

_c_, - 120o6oo _-_-o- c_-..n-- __ c_ °="

-800 _400- _ T1 ..

---o---T2_- 0 --_-- T3 --200 -- 400

---.-L_-----T4I_

_0 200 400 600 800 1000 1200 ""

NUMBEROF CYCLES,N

FIGURE19, - VARIATIONOF REACTORTEMPERATUREDISTRIBUTIONFOR CYCLICLIFETESTING. [TOTALMASSFLOWRATE= 0.0018KG/SEC(O._LlLB/SEC),MIXTURERATIO= 1.0"2 SEC RUN DURATION.]

18

i

NASA Report DocumentationPageNational.Aero,tlauticsandSpaceAdministration

1. Report No. NASA TM-100957 2. GovernmentAccessionNo. 3. Recipient'sCatalog No.

AIAA-88-3300

4. Title and Subtitle 5. ReportDate "

CatalyticIgnitionof Hydrogenand OxygenPrope11ant s B.PerformingOrganizationCode ';

7. Author(s) 8. PerformingOrganizationReport No.

Robert L. Zurawskiand ,lamesM. Green E-4104

10. Work Unit No.

506-42-219. Performing OrganizationName and Address

National Aeronautics and Space Administration 11.ContractorGrantNo.Lewis Research Center

Cleveland, 0hlo 44135-3191 13.Type of Report and Pedod Covered

12. SponsoringAgencyName and Address TechnI cal MemorandumNati onal Aeronauti cs and Space Admini stratl on 14.Sponsoring Agency CodeWashington, D.C. 20546-0001

15. Supplementaw Notes

Preparedfo_ the 24th 3olnt PropulsionConferencecosponsoredby the AIAA, ASME,SAE, and ASEE, Boston,Massachusetts,July If-13, 1988. Robert L. Zurawskl,NASALewis ResearchCenter;,lamesH. Green, SverdrupTechnology,Inc., NASA LewisResearchCenter Group, Cleveland,Ohio 44135.

16. Abstract

An experimentalprogramwas conductedto evaluatethe catalytlcignitlonof gase-ous hydrogenand oxygen propellants. Shell 405 granularcatalystand a mono-lithicsponge catalystwere tested. Mixtureratio,mass flow rate, propellanttemperature,and back pressurewere varied parametricallyIn testing to determinethe operationallimitsof the catalyticigniter. The test results show that thegaseoushydrogenand oxygen propellantcombinationcan be Ignltedcatalyticallyusing Shell 405 catalystover a wide range of mixture ratios,mass flow rates,and propellantinjectiontemperatures. These operatingconditionsmust be optl-mized to ensure reliable ignitionfor an extendedperiod of time. A cycllc lifeof nearly 2000, 2-sec pulses at nominaloperatingconditionswas demonstratedwith the catalyticigniter. The resultsof the experimentalprogram and theestablishedoperationallimitsfor a catalyticIgnlterusing the Shell 405 cata- 'lyst are presented.

,0

17. Key Words (Suggestedby Author(s)) 18. DistributionStatement

Catalytic ignition; Rocket engine Unclassified - UnlimitedIgnltlon;Igniters;Catalysls; i SubjectCategory20Monolithiccatalyst.

19. SecurityClassif. (of this report) 20. SecurityClassif.(of thispage) 21. No of pages 22. Price"

Unclassifled Unclasslfied 20 A02

NASAFORM1626OCT86 *For sale by the NationalTechnicalInformationService,Springfield,Virginia22161

Nationa,Aeronauticsan° OO.T.C..SSM.,.IIIIIISpaceAdministration

Lewis ResearchCenter ADDRESSCORRECTIONREQUESTEDCleveland,Ohio44135

Official Business

Penalty for Private Use $300 Postage and Fees PardNahonal Aeronauhcs and

Space Admmtstrat_onNASA 451 _-

NASA