DTIC ADA532486: Influence of Aluminum Passivation on the Reaction Mechanism: Flame Propagation Studies

1.3 MB·English

by  Defense Technical Information Center

FDrm AppnM!d
REPORT DOCUMENTATION PAGE OMS NO. 07GHH88
I'ublil: Ilc(IOrling burden for lhl, CQIJCtliOIl of infol'1lla,ioll is csllnulloo 10 avcragc I hour pcr rCSjllll15C. iDcIudillg !he time far ~ ~ ~ ""~ ..." "_ns.
galllcring and Inalmalnlng thl! d~m "ccded. nnd colllpll'lins rulU ,.lviuwiDg Ihe collection of Infnnn.allOll. Scod CtlIIIIIII:'III ~ Illis btlAlnI <'liIiInIII:s ... a:!} ..... iIIIpIXIilf'." ~
"f Illrurmntion. Including ~ull8"S!lans for reduclnll ,bl~ burden. 10 Washin!ll0n HcalkluanclS Scr\-1<:es. Dlrrc10mll for hlformatloft ()pcr.o:irJM am Rq:>oIa. IllS Jcl'"_ IRri!. ~.
Sulle 12.001. ArlllllllOn, VA 22202·4302. alld IU 11m OffiCI! of IMa na~clll"Rl ant! flIIIl'CI. I'llJlMWOl'k Rcductiolll'ml<!C1 (071)10188.) W~DII. DC Z05III.
I. AGENCY USE ONLY ( Leave Blank) J. REPORT TYPE AND OATF.5 COVERED
:/=~~OATE
Energy & Fuels 23 (2009) pp4231-4235
4. TiTlE AND SU8TITLH S. FUNDING NUMBERS
Influence of aluminum passivation on the reaction mechanism: flame W911NF...()4..l-0217
propagation studies
6. AUTHOR(S)
B. Dikici. SW Denn, ML Pl'll1toya. VI Levitas, RJ. Jouet
8.
7. PERFORMING ORGANIZATION NAME(S) AND ADDRESS(ES) PERFORMING ORGA"IIZA TlON
REPORT NUMBER
Texas Tech University, Mechanical Engineering Department. lubboCK. TX
79409
9. SPONSORING 11',,10NITORING AGENCY NAME(S) AND ADDRESS(ES) 10. SPONSORING I MONlTORING
AGENCY REPORT ""'UMBER
U. S. Army Research Office
P.O. Box 12211
Research Triangle Park, NC 27709-221 I
II. SUPPLBMENTAItY NOTES
The views, opinions and/or findings contained in this report nre those of the Iluthor(s) and should nUl be construed as an official
Depl.lI'Lmenlllf'the Army position, policy or decision. unless so designated by other documentation.
12 a. DISTRIBUTION J AVAILABILITY STATEMENT 12 b. DISTRIBUTION CODE
Approved ror public relcosefederal purpose rights
13. ABSTRACT (Ma1>il1lUm 200 lVords)
Currently, two main known mechanisms of aluminum (AI) nanopartic1e reaction are discussed in the
literature, namely those based on diffusion through an oxide shell and mell-dispersion. The two
mechanisms lead to opposite predictions in nanoparticle design. The diffusion mechanism suggests that
the reduction Dr complete elimination of the oxide shell will increase AI reactivity, whereas the meltdispersion
mechanism suggests an increase in initial oxide thickness up to an optimal value. The goal of this
study is to perform critical experiments in a confined flame tube apparatus to compare these two
predictions. Specifically, the flame propagation rates ofperfluoroalkyl carboxylic acid (CI3F27COOH)-
treated Al nanoparticles with and without an alumina shell were measured. Results show that when there is
no alumina passivation shell encasing the Al core, the flame rate decreases by II factor of22-95 and peak
pressure deceases by 3 orders of magnitude, in comparison with the AI particles with an oxide shell. These
results imply that the melt-dispersion reaction mechanism is responsible For high flame propagation rates
observed in these confined tube experiments.
14. SUBJECT TERMS 15. NUMBER OF PAGES
Rt'Rclion Mechanisms, Thermite Combustion, Aluminum Particles, Passivation of Aluminum 5
Particles, AI synthesis techniques
16. PRICECODE
118.
17. SECURITY CLASSIFICATION SECURITY CLASSIFICATION 119. SECURITY ClASSIFICAllO!-1 20. UMITATIONOF ABSTRACT
OR REPORT ON TI·HS PAGE OF ABSTRACT
unclassified
UNCLASSIFIED UNCLASSII'JED lll'lCLASSIFIED
NSN 7540-01·280-5500 SUllu:llani Fonn 298 (Rtv..l..",
I'I5:7ibed hv ANSI SId. 239.111
298-102 •
Enclosure 1
:
EnergyFuels2009,23,4231–4235 DOI:10.1021/ef801116x
PublishedonWeb08/06/2009
Influence of Aluminum Passivation on the Reaction Mechanism:
Flame Propagation Studies
BirceDikici,†StevenW.Dean,†MichelleL.Pantoya,*,†ValeryI.Levitas,*,‡andR.JasonJouet§
†MechanicalEngineeringDepartment,TexasTechUniversity,Lubbock,Texas79409,USA,‡DepartmentofMechanical
Engineering,DepartmentofAerospaceEngineering,andDepartmentofMaterialScienceandEngineering,
2025BlackEngineeringBuilding,IowaStateUniversity,Ames,Iowa50011-2161,USA,and§ResearchandTechnology
Department,IndianHeadDivision-NavalSurfaceWarfareCenter,IndianHead,Maryland20640,USA
ReceivedJanuary11,2009.RevisedManuscriptReceivedJune24,2009
Currently, two main known mechanisms of aluminum (Al) nanoparticle reaction are discussed in the
literature, namely those based on diffusion through an oxide shell and melt-dispersion. The two
g mechanismsleadtooppositepredictionsinnanoparticledesign.Thediffusionmechanismsuggeststhat
or the reduction or complete elimination of the oxide shell will increase Al reactivity, whereas the melt-
s.
ac dispersionmechanismsuggestsanincreaseininitialoxidethicknessuptoanoptimalvalue.Thegoalofthis
s.
b study is to perform critical experiments in a confined flame tube apparatus to compare these two
u
009 | http://pef801116x pptnrrroeeeaadstslieuucdtmrieAoidnnleasn.capeSnaapossesepcisavibfraitytciicao3llenloysrs,wdhtieehtrlehlseafonlnafcdmmawseainigptnghriootthpuueadtgaeA,anlitnaicolocunromerm,aitnptheaaesrsfihoslaefolmnplweewrerfiratluehteomtrhdeoeeaacAslrukelryapelsadecr.satRrbicbyeloesausxlyfwtalsiictcsthhoaoarcwniodfoth(2xCa2idt1-3ewF9sh52he7eaCnlnlOt.dhTOephrHeeeas)ikes-
er 18, 20.1021/ roebssuelrtvseidmipnlyththesaetcthoenfmineeltd-dtuisbpeeresxiopnerrimeaecntitos.nmechanismisresponsibleforhighflamepropagationrates
b1
BRARY on Septemugust 6, 2009 | doi: csoolnUidvseepnoatrifotnincaalnelomfmuiecelrtoarinmcdientogexrIri-endsdtiirzizoeeenddrtuspfcooatrirfomftneicurlsleamstiiaonnntysh.aeSdcivonamcnebtnaugasnetisooopnvaoerrf- ecmxooaFmlmoypbriondseneitantehnuseo.mpGeaftfrrreaitcoinctxileeoidrsf,eept(taMharetlori.ceOfloe3hu)sa.ni3zvdeesobinmeefillnaamrmreeassnpuyeltessdtfuotdhrireAosulgtahhnadat
LIA mixture. One of the first investigations by Bockmon et al.4
NIV Web): tpioclienstshabveetwheigenherresapcetacniftiscscuormfapcaeraedretao,tmheicrreoamreetmero-rseizceodnptaacrt- examinedthecombustionbehaviorofnanocompositepowder
U mixtures of Al and MoO as a function of Al particle size.
by TEXAS TECH Publication Date ( tsrGaaipencsareldheeccaostiinfa.triencRtcrertsreesuhadtrsaouifmenmcasedcetodehsgicaepenarafncellrreeaateimit(acdyisl.eeeeecs.iss,rmpie1eztap0eehsrdeeaoμ.snvsTh)tdehhoscteemohrtdmeeohifgefpcofeoaurneerrsfere,ifeoidritscenyptiaeoodcnotnitcisfhodytaenitnnhsomgceofeiiccnmbrcrceuoeiraxtrmewctoatueeisvtoreeeeenndrr. Fcf1oro2bloan1smmeftironev7ee58sdd00pbeftnoueomdrr9nm.5aiCn0nixgdomtcnupo/srsrinedewsfseisgwhruaueirbtnreha3lettmphicoaeehnrapatssian.cuTrglrteeheiscsmelbieaneevsnloifetzlrswaeamwgw8eee0arsssnepprmeeaeeedn.d4duailcwnyeczedreredefarnosfeoomdrt
d There are currently two main mechanisms of aluminum
e particulatereactions(i.e.,10ms).
d nanoparticle oxidation applicable to the high heating rate
oa Aumann et al.1 examined the oxidation behavior of
nl conditionsconsistentwithflamepropagationinatube.Each
w ultrafine grain aluminum powder using Rutherford back-
o mechanismissummarizedbelow.
D scattering spectrometry, thermogravimetric analysis, and
Diffusion Mechanism. Conventional metallic Al fuel na-
high resolution transmission electron microscopy. They
noparticlesarecoveredbyathin(1.7-6nm)aluminashell.
suggestthatAlpowdermixtureswith20-50nmdiameters
Duringcombustionthereactioniscontrolledbydiffusionof
can react 1000 times faster than micrometer particle mix-
theoxidizerand/orfuelthroughthegrowingoxideshell.5,6
tures because of the reduced diffusion distances between
Thismechanismissimilartothatformicrometerparticles7,8
individualreactants.1Hunt etal.2studiedtheignitionand
but also adds diffusion of aluminum toward oxygen. The
combustionbehaviorsofnanometricnickelandaluminum
diffusionmechanismwasjustifiedinref5experimentallyfor
intermetallic mixtures. Laser ignition experiments on
theheatingrateof103K/s.Thismechanismrequiresatleast
pressedNi/Alpelletsshowthatnanometriccompositeshave
1 sfor complete oxidationto occur, whereas theestimated
significantly reduced ignition times over micrometer-scale
(4)Bockmon,B.S.;Pantoya,M.L.;Son,S.S.;Asay,B.W.;Mang,J.
*Towhomcorrespondenceshouldbeaddressed.E-mail:michelle. T.J.Appl.Phys.2005,98(6),064903.
[email protected]. (5)Park,K.;Lee,D.;Rai,A.;Mukherjee,D.;Zachariah,M.J.Phys.
(1)Aumann,C.E.;Skofronic,G.L.;Martin,J.A.J.ofVac.Sci. Chem.B2005,109,7290.
Technol.,B1995,13(3),1178–1183. (6)Rai, A.; Park, K.; Zhou, L.; Zachariah, M. Combust. Theory
(2)Hunt,E.M.;Plantier,K.B.;Pantoya,M.L.ActaMater.2004,52 Model.2006,10,843.
(11),3183–3191. (7)Rosenband,V.Combust.Flame2004,137,366.
(3)Granier,J.J.;Pantoya,M.L.Combust.Flame2004,138(4),373– (8)Trunov,M.A.;Schoenitz,M.;Zhu,X.;Dreizin,E.L.Combust.
383. Flame2005,140,310.
r2009AmericanChemicalSociety 4231 pubs.acs.org/EF
:
EnergyFuels2009,23,4231–4235 DOI:10.1021/ef801116x Dikicietal.
Table1.MaterialPropertiesofReactantPowdersa
material manufacturer particlesize(nm) oxidethickness acidcontent(basedonmass) activeAl(basedonmass)
Al-80 NovaCentrix 80 2 71.6% 23.8%
(Al-C F )cd IndianHead,NSW 100 noshell 75% 25%
13 27
(Al-C F )th IndianHead,NSW 100 noshell 75% 25%
13 27
MoO3 Mach-I 44 n/a n/a n/a
aNotethatC F representsC F COOH.
13 27 13 27
reactiontimeinflamepropagationexperiments4is4orders forbothcompositesandbothmixtureswerepreparedwith
ofmagnitudesmallerfornanoparticles.Theeffectofphase similarstoichiometry.Theexperiments wereperformedto
transformations in the alumina shell from amorphous to measure the pressure output and flame speed of loose
crystalline phases, which leads to fracture of a shell and powdersreactinginaconfinedconfiguration.
acceleration of oxidation, was studied in refs 9 and 10. It is noted that there are conditions under which the
Propagationratesforthediffusivemechanismofoxidation perfluorinatedcarboxylicacid-passivatedAlparticlesexhi-
couldbeincreasedbyreducingorcompletelyeliminatingthe bit improved performance over the alumina passivated
initialoxideshellthicknesses. Alparticles.Forexample,inrefs15and16acid-passivated
Melt-DispersionMechanism.Levitasetal.11,12proposethe Alparticlescombinedwithplasticbondedexplosive(PBX)
melt-dispersion mechanism for fast reaction of aluminum showedtwiceasgreatdentvolumecomparedtostandardAl
nanoparticlesatfastheatingrate.ThemeltingoftheAlcore formulations in small-scale shock reactivity and internal
g
or causesa6%volumeincrease,whichleadstoapressurebuild- blaststests.Also,inrefs16and17theevolutionanddecay
cs. upwithintheparticle(0.1-4GPa).Thishighpressurecauses ofaluminumandaluminummonoxideweremonitoredfrom
a
bs. the oxide shell to spallate, after which the melted Al core laser ablation studies. Both studies show that the reaction
u
http://p116x blaercgoemteensseixlepporseesdsuarnedaanndudnilsopaedrsinesgmproelstseunreAwlcalvuestperrosdinucaelsl rtiaotneooffathciedo-pxaidsseivcaotaetdinAg.lparticlesincreasedbytheelimina-
009 | ef801 dcliuresctetirosnasreanndotaltimhiigthedvbelyocdiitfyfu.Rsioeanctthiornouogfhsuacnhosxmidaellsbhaerlle, ExperimentalSection
er 18, 20.1021/ wsehrvicehd.exTphlaeinmsetlht-edihspigehrsifolanmme escpheaendissmexpiserismcheenmtaalltyicaolbly- stuSdaiemdp:lweiPthreapnadrawtiitohno.uTtwaonadliuffmeriennatptayspseivsaotfioAnlsphaerltl.icTlehsewtyepree
Septemb9 | doi: 1 iAllluTnshtarinasotmepdaercitnhicarlneefiss1cm2o.malblouwstsioonntehtaotcraensonlovtebveaerxiopulasinpeudzzblyesthine ouinfsfiAonrglmwtaiwtthiooonudtfifoafnreraeealnuctmhsisnyaanmtphpaelssesiisivsatlteiicsothnendsiqhinuelelTsw.aabPslhefuy1sr.itchTaelhrepprrfeiorppsaterrAetdyl
ARY on st 6, 200 Adilffpuasriotinclemreacdhiuans,issmhe.llItthiisckfnoeusns,doirnalruemfsin1u1m-1c3onttheanttathree stoeafcm8h0pnlonelmgwie(aps)sroInpvcuid.r,ecAdhaubssyetidtnh,efTrmoXma,nwuNiftahocvataunCraeervnewtrrhaixgoeu(pfsoaedrrmtiBceElrelTyd)i.aNOmaxenitdoeer-
BRugu notcontrollingparametersindividually.Instead,thereexists shell thickness was determined by the manufacturer. Detailed
UNIV LIWeb): A saMpre,eelaadntsdiownflesarhmeipeobbspetatewiendeee.dnF,AowrlhcMeorre<easr1af9do,irumsMatxo>imsh1ue9mllrtahptioecssksnidbeelscesrferlaaatsmieode, adTnihaaemlsyeestepesraorotfifpcla8er0stiwnclmeerseifzrteohmednisNttrrieobavutaetCidoenwnstitrfhoixrtphisaerpCtricelseFesnwteCidtOhiOanvHreerafacg1ied0d..
CH ate ( with increasing M. In fact, an increase in shell thickness ThissamplewillbedesignatedasAl-80. 13 27
nloaded by TEXAS TE Publication D rsiMmfsleiaomdeImnwcun.hcemaLaestanesshetpciieMxshsermtea,e1dsnasn4tinudastoddhmerfedyeiA,nbcsfucarolbirmtrtoepii1acentaha-srrseltelu3siyltcpahμfleetlpemaixsoomardnwpmteisaifohtifrrnhfuiatpeiastcdaieblmosneefn,sdeut.iwlrntawt-ehndceieotdinshrnpfobtmleruayraemtsslsittoate-unntdsodpimsayeaplieeuneddcrmghisafaiitfonnhunndae-- ntCppwfrraaeieoertrTrmfettFelhhiducmeyloceal)aasartseotdhlhacute.(oalemAyustnAtrsildin-avincnduetoCgrydema1cpeta3gawceFlenouoto2ohmm7pfdie)cpaiicArnfrmodftaleas;iorsicctaeallsiiyednhnoms,tednisplstileClz,toeceh1fahbwe3nwnuFAadsita2qsle7HscuiCsisoye3sdOn*snei:NdntOtdthesh(eHrewtMesmesia.aifgzieisdnnre)TsPedapthpdtyrwseeaerudsupsaecsasashiinrvpanpedatsrgdahteerdapB(tdfteaAirEcsortilblehTgem-dyes-.
Dow (Al2O3) shell. For this reason, two types of nano-Al/metal Th13ea2c7idcontentforallsampleswasdeterminedbasedonthe
oxide composites are examined to determine the shell’s specificmassesofAlandacid.
influence on flame speed. Thefirst Al sample is passivated SEMmicrographsofthecompositesofAl-80withMoO and
3
withAl O andtreatedwithperfluoroalkylcarboxylicacid (Al-C F )th with MoO are presented in Figures 1 and 2.
2 3 13 27 3
(C F COOH). The second Al sample is devoid of an Figure3isaschematicdiagramillustratingthetwotypesofAl
13 27
alumina shell and instead passivated with perfluoroalkyl surface passivation examined in this study. Each type of Al
carboxylic acid (C13F27COOH). Elimination of alumina particlewascombinedwithnanoscaleMoO3,whichhasarod-
likemorphologyandwaspurchasedfromMach-I.TheactiveAl
shellshouldimprovediffusionbutmakethemelt-dispersion
content for Al-80 in Table1 refersto Alin the zero oxidation
mechanismimpossible.Thesameoxidizer(MoO )wasused
3 state. It is also noted that the alumina content for the Al-80
sampleis4.6%onamassbasis.
(9)Trunov,M.A.;Schoenitz,M.;Dreizin,E.L.Combust.Theory
Model.2006,10(4),603–623.
(10)Trunov,M.A.;Umbrajkar,S.M.;Schoenitz,M.;Mang,J.T.; (15)Warren,A.D.;Lawrence,G.W.;Jouet,R.J.SockCompression
Dreizin,E.L.J.Phys.Chem.B2006,110,13094–13099. ofCondensedMatterInAIPConferenceProceedings;Elert,M.,Furnish,
(11)Levitas,V.I.;Asay,B.W.;Son,S.F.;Pantoya,M.L.Appl.Phys. M.D.,Chau,R.Holmes,N.C.,Nguyen,J.Eds.;2007;Vol955,pp1018-
Lett.2006,89,071909. 1021.
(12)Levitas,V.I.;Asay,B.W.;Son,S.F.;Pantoya,M.J.Appl.Phys. (16)Jouet,R.J.;Carney,J.R.;Granholm,R.H.;Sandusky,H.W.;
2007,101,083524. Warren,A.D.Mat.Sci.Technol.2006,22(4),422–429.
(13)Levitas,V.I.;Pantoya,M.L.;Dikici,B.Appl.Phys.Lett.2008, (17)Jouet, R. J.; Carney, J. R.; Lightstone, J. M.; Warren, A. D.
92,011921. ShockCompressionofCondensedMatterInAIPConferenceProceed-
(14)Levitas, V. I.; Pantoya, M.; Watson, K. W. Appl. Phys. Lett. ings;Elert,M.Furnish,M.D.,Chau,R.,Holmes,N.C.,Nguyen,J.Eds.;
2008,92,201917. 2007;Vol955,pp1247-1250.
4232
:
EnergyFuels2009,23,4231–4235 DOI:10.1021/ef801116x Dikicietal.
Figure4.SchematicdiagramillustratingtheAcrylictubeusedfor
flamespeedmeasurements.
Figure1.SEMmicrographsofthepostmixedcompositesofAl-80
withMoO3:(A)80000(cid:1)magnification;(B)40000(cid:1)magnification. Al as C13F27COOH on a per mole basis but is only approxi-
mately33%Albymassduetothehighmolecularweightofthe
acid.
Adetaileddescriptionofthesynthesisandcharacterizationof
Alnanoparticleswithoutanoxideshellispresentedinrefs15-
g 18.Insummary,SEM,thermogravimetricanalysis(TGA)and
or
s. attenuatedtotalreflectanceFouriertransforminfraredspectro-
c
s.a metry indicate that the C13F27COOH molecule binds to the
ub surfaceoftheAlparticle,protectingthesurfacefromoxidation
p
p://6x inambientair.
htt11 For all Al particles reacting with MoO3, it was decided
009 | ef801 nboectatuosedetshceribcoemthpeetcionmgpooxsiidtaiotinoninatnerdmfsluoofrienqautiivoanlernecaectriaotnios
ber 18, 210.1021/ wmThiatehteslAyamlthaperleesanAmol-et8mu0naadslessropstecoroconednt.atiaAngslelsaAnolfasAdalmdaiptnilodensaaccloid4n.t(6aseimneaTaspasbp%lreox1oi)-f.
ARY on Septemst 6, 2009 | doi: mFCi1ag3guFnr2ei7f)i2tc.haStiwEoiMnt.hmMicorOog3r:a(pAh)so8f0t0h0e0(cid:1)posmtmagixneifdiccaotimonp;os(iBte)s4o0f0(A00l-(cid:1) AaFcoodxolr2dirdrOeeidza3s.pelalTrononAoddrelevacdacsoahrmimteoodbfpitl7nfher0aseo.,ts6mieothAnmaelatcshslmaesemarae%rxploiylmfe,sfMuu,thameolimOslefa3tlar.asansmRnpteseoelgrapcfatueerrneosdltplteoraosifgcsaMhaotfimofouOniewxl3thrurwaiirccatehhes.
BRugu mixture.
LIA Tocombinereactants,themixturesaresuspendedinhexane,
UNIV Web): aspnededasMisiussoendixtoSiomnpicraotvoerm30ix0t0urperhoobme ovgibernaetiitnyg.Tahtisuldtreavsicoeniics
H e ( programmedtoproducetheultrasonicwavesfor60swith10s
ECDat on/off intervals.Thisintervalpreventsatemperatureincrease
S Ton andpossiblethermaldamagetothesample.Itisnotedthatthe
XAcati influenceofthehexaneonthereactantsisnotknown.However,
y TEPubli itnhethseaemxepewriamyesnutcshdetshcartibtehdeheeffreec,tasllorfeathcteanmtsixwinegresporlueptiaornedairne
b
d maintainedasacontrolledparameter.
e
d ThemixturesarepouredintoaPyrexcontainerandplacedon
a
wnlo Figure3.SchematicdiagramillustratingAlsurfaceandpassivation ahotplateat50(cid:1)Cfor10mintoevaporatethehexane.Afterthe
Do types. evaporation,thepowderisreclaimedforexperimentation.
Roughly 450 mg of powder mixture prepared for each mix-
ture composition and placed into an acrylic tube shown in the
SurfacepassivationofAlnanoparticlesbyC13F27COOHis Figure 4. The tube is open at both ends, and acrylic is chosen
based on a wet chemistry method. This method involves pre- because of its transparency and strength. Transparency allows
paring and passivating the oxide-free Al.18,19 Oxide-free Al flamepropagationinsidethetubetobeseenduringcombustion.
particlescanbepassivatedbyexposureofthemetalsurfaceto Moredetailsonthisapparatuscanbefoundinref6.
C13F27COOH. This technique has been applied to several Thetubesarefilledwithpowder,thenplacedontoavibrating
systems, but for the purposes of this paper the source of Al block and vibrated for approximately 5 s to reduce density
wasthecatalyticdecompositionofH3Al3N(Me)Pyrorthermo- gradients, after which more powder is added. This process
lyticdecompositionof(CH3CH2)3Al.Passivationoftheparti- results in a bulk density of 8% of the theoretical maximum
clesoccursatroomtemperaturebyexposureoftheasformed density(TMD).TheTMDiscalculatedastheratiooftheactual
Al0 particles to a solution of C13F27COOH in dry diethyl mass of reactants divided by sample volume to the weighted
ether. The preparations are done under inert atmosphere and averageofpuresoliddensitiesofeachreactantinthematrix.In
indrysolventtopreventcontaminationbyoxygen.Thehighest thisway,aTMDof10%describesasamplethatconsistsof10%
Alcontentmaterialpreparedtodatecontains13timesasmuch solidparticlesand90%airvoids.
Themassinthetubesvariedbecauseofthedensityrelatedwith
different compositions and equivalence ratios. A bent piece of
(18)Jouet,R.J.;Stern,A.G.;Rosenberg,D.M.USPatentNo.US
Nichromewireisplacedononesideofthetubeandusedasthe
7,192,649B1;March20,2007.
(19)Jouet,R.J.;Warren,A.D.;Rosenberg,D.M.;Bellitto,V.J.; ignition source. The tube is placed into the testing block with
Park,K;Zachariah,M.R.Chem.Mater.2005,17(11),2987–2996. ports for 4 PCB piezo-crystal pressure sensors used to collect
4233
:
EnergyFuels2009,23,4231–4235 DOI:10.1021/ef801116x Dikicietal.
Figure5.Consecutiveframeimages(A)Al-80/MoO ;(B)(Al-C F )cd/MoO ;(C)(Al-C F )th/MoO .
3 13 27 3 13 27 3
pressuredata.Themixtureisignitedatoneendofthefilledtube
viaresistanceheatingoftheNichromewire.
The test block was placed in a stainless steel combustion
chamber for safety purposes. The combustion chamber has an
g acrylic window (139.7 (cid:1) 25.4 mm dimensions) that is used to
or obtainhigh-speedvideoofthereaction.Thechamberisconnected
s.
c toavacuumpumptobringthechambertoadesiredpressure.A
a
bs. pressuregaugeonthetopofthechamberisusedtomeasurethe
u
p pressure. The product gases are evacuated from the chamber
p://6x during and after the reaction. All experiments were performed
er 18, 2009 | htt0.1021/ef80111 apaTtnohdA1seitawcitaoVammnsiseueiidnorsanepadcneaRrtapopeietrsruneeerdacneorivsccriuthdrhloafenPlravmhtmioaseiennbttthpl.oeermoedmpira7iegs.cs1atitoiiohnoningliohgrf-ahstftpelaaesmnewddeitthpcaairknomeptseharaegasateutwriboiaeenss.
b1
Septem9 | doi: opTfrhopisphaofgrtaoamtgiroeanprahwtseearatelalrohewcigoehrddseaadm1ap6tliμn5sg1tf0er0em0qpuforearnamclye.rseIgsponelirutitsoieoncnoannbddetf(wlfapemes)ne.
LIBRARY on August 6, 200 imwmiitPanhgereertseshs.seupPrfehlecaatwmntaoteosstmmpimeedeeada.stubaryeadtcrquaucsikinsiigntigofonthuseroPpfoCtwsBiatir1oe1n3wAoaf2st2huepseifedlazotmo-ecdrfyerstoetnart-l
NIV Web): purseesdsutorecotrnavnesrdtutcheerst.raAnsdPuCcBer4s8ig2nAa2l2instiognaaclacloibnrdaitteidonoeurtpwuats. Figure6.Pressuremeasurementsfor(A)Al-80/MoO3reactionand
H Ue ( ANationalInstrumentsdataacquisitionboardwasconnected (B)(Al-C13F27)th/MoO3.
Cat to the signal conditioner and controlled by Labview Version
ED
by TEXAS T Publication 81.000skoHftzw.are. Pressure dataRweasusltcsollected at a sample rate of inTflhaemreesuplrtosptaabgautliaotnedriensDTuilastcbinulegss2ifoirnnodmicattweoexdtrifefmereednitffpearernticceles
d assemblieseventhoughthechemistryofbothmixturesisthe
e
d Figure 5 show consecutive still frame images recorded at same (i.e., same fuel-rich mixture composition). Promotion
a
o
wnl 51000 fps. Images for mixtures of Al-80 with MoO3 were ofdiffusionandeliminationofthemelt-dispersionmechanism
o recordedat62500fpsaregiveninref4. byreplacingtheoxideshellwithaperfluoroorganicpassiva-
D
Pressure measurements for the Al-80 and (Al-C F )th tionlayerreducedflamerateandpressuregenerationdrama-
13 27
mixturesaregiveninpanelsAandBofFigure6,respectively. tically. This result implies that among the two alternatives,
ForAl-80mixturesthedatafromthefourthpressuresensor the melt-dispersion mechanism is responsible for the high
(farthest from ignition point) is used for characterizing the flame rate of Al nanoparticles. Despite the fact that the
steady-state propagation behavior. For the (Al-C F )th aluminashellinhibitsdiffusion,doesnotparticipateinreac-
13 27
mixturesthedatafromonlythefirstpressuresensorisreported tion(i.e.,isadeadweight)andactsasaheatsink,thepresence
becausetheflamepropagatedonlytothepointbetweenfirst of the alumina shell is a necessary condition for the melt-
and second pressure channels. For (Al-C F )cd pressure, dispersionmechanismthatisproposedastheprinciplereason
13 27
thepressurehistorywasnotcapturedsincetheflamequenched for the high flame propagation rates observed in reactions
beforereachingthefirstpressurechannel. involving nanometric Al particles. Although these experi-
For the combustion of Al-80 with MoO , higher flame mentsdonotprovedirectlythemelt-dispersionmechanism,
3
speeds (95 times higher flame speeds compared to (Al- they are fully consistent with predictions based on this
C F )cd, 38 times higher flame speeds compared to (Al- mechanism.Theseresultsarealsoconsistentwithexperiments
13 27
C F )th, and 2607 times higher peak pressures compared in ref 13 that show that a decrease in oxide shell thickness
13 27
with(Al-C F )th)aremeasured.Pressuredatacorrespond- reduces the flame rate in experiments similar to the one
13 27
ingtobothAl-C F sampleswithMoO showaninsigni- presentedhere.
13 27 3
ficant pressurerise compared toAl-80sampleswithMoO . In general, flame rate depends on various parameters,
3
Table2summarizestheseresults. includingmechanismofflamepropagation,Alparticlesize,
4234
:
EnergyFuels2009,23,4231–4235 DOI:10.1021/ef801116x Dikicietal.
Table2.SummaryofResultsfromHigh-SpeedCameraandPressureSensorsforOptimalMixtureCompositiona
reactants Al-80 (Al-C F )cd (Al-C F )th
13 27 13 27
mass(mg) 480 420 420
flamespeed(m/s)(fromhighspeedcamera) 285.2(10 3(0.5 7.5(0.5
peakpressure(kPa) 1676.670 N/A 0.643
risetime(μs) 228.571 N/A 682
pressurizationrate(MPa/sec) 6601.89 N/A 0.10185
propagationrate(frompressuredata)(m/sec) 178.6 N/A 8.13
samplerate(pps) 100000 10000 10000
aN/A:notapplicablebasedonquenchingofthereaction.
size distribution of fuel and oxidizer particles, oxide shell Conclusion
thickness, the morphology of the mixture, and others. It is
Flamepropagationinaconfinedtubeapparatuswasmea-
impossible,atleastinthegivencase,tokeepallparametersthe
sured for C F COOH-treated Al nanoparticles with and
same, excluding oxide thickness. Thus, active Al content is 13 27
without an alumina shell. Removing the oxide shell should
slightly smaller for the Al-80 sample with an oxide shell
promote a diffusion reaction mechanism but make the melt
(Table1), butthispresenceofaluminashouldreduceflame
dispersionmechanisminoperative.Thealumina-passivatedAl
speedforthemaccordingtothetraditionaldiffusionmechan-
particlesexhibitedflamespeedsontheorderof285m/sandthe
ism.ThediameterofAlnanoparticleswithashellisslightly
acid-passivated Al particles propagated at roughly 7 m/s or
g smallerthanthatwithoutashell.However,theflamespeedis
bs.acs.or tohxeidseamsheeflol.r13boTthhepapratrictilcelesizseizseifdthisetyriabrueticoonvemreadybbyeaspirgonpifeir- wmisoaansgnwqiutiutehdnetchhfeoedrm.AixTlthuperaerpstiewcaliektshwpArietlhspsoauurrtteiacdlneesoctexhaidasetesilnabcyyleurd3ienotchrodemeorpxsiadore-f
u cantlydifferentforsampleswithandwithoutanoxidelayer.
http://p116x Hofomwiecvreorm,eevteern-sacnalaedpdaitritoicnletsodAidlnnaontocphaarntigceletshoeffulapmteov3e0l%o- smheecllh.aTnhisemseisrerseuslptsonimsibplleyftohrathitghhe fmlaemlte-dpisrpoeprsaigoantiorenacrtaitoens
er 18, 2009 | 0.1021/ef801 cipAsirotloynpp.2awa0rgatFaisctolmierosantidhsreeas.stoeTedhearxenapdpsetorgiicienmnttheeinarsatttsthae,aldltthtpthehreeecssledseuiaifnrnfedeeresfettonerprcmtoehsbiesneitabtwwcleioeecestonycmpaflnepasnmaooretf- rsoeepxbdetsrueeedcrmvt,ieieondncaainicsnectoohorxefdiocdaouennrcsfpheinrewelelvidttihohtuuitcbhskeeenxemepxseesplreltie-rmddimietsnoeptneattrhlssef.iioTnrnedhdimenuygecrctsei,ho1pa3nrnweiisnshemenfnltaattmnhhdeee
Septemb9 | doi: 1 udnradsetricmailnlyereoduurcecsopnacrlutisciloenreathcatitvitryeminotvhinegflatmheetouxbiedeexpshereil-l ictniovnec,dorinteidtorunacssttwiothnoeontfhttheheedmoifxefliutds-edioissnhpeerlrlessaihocotniuomlndelcmehaeadcnhtisoamntihsiemsin.nocFtreooarpseetrhaine-
ARY on st 6, 200 amodfevtnahtne.tTmahgaiestserreoisafullwrteiimsthnooovutitninganctohonextridaoedxiiscdhteieollns,h1w6e-lilt1.h9Ipwnrhdeievceihodu,sshfosotwrudttihheees reactionrate,accordingtoexperimentsinrefs16-18.
BRugu conditionswhenthemelt-dispersionmechanismisnotopera- Acknowledgment. The authors gratefully acknowledge sup-
NIV LIWeb): A ttihvee,otxhideedsifhfeulsliwonillrpearoctmioontemreecahctainoins.mappliesandremoving p1-o0r2t17fro(mprotghreamArmdiyrecRtoerseaDrcrh. ROaflfpiche AconntthreancitenW),91O1fNficFe-0o4-f
CH Uate ( (20)Moore,K.;Pantoya,M.L.;Son,S.F.J.Propul.Power2007,23, NDar.vCallifRfoersdeaBrcehdfocrodn)t,raacntdNN0a0ti0o1n4a0l8S1c0i1e0n4ce(Fproougnrdaamtiodnirgercatonrt
S TEon D 181–187. CBET-0755236(programdirectorDr.PhillipWestmoreland).
XAcati
y TEPubli
b
d
e
d
a
o
nl
w
o
D
4235