Table Of Content

The Possibility of Forming Propargyl Alcohol in the Interstellar Medium PrasantaGorai1,AnkanDas1∗,LitonMajumdar2,1,SandipKumarChakrabarti3,1,BhalamuruganSivaraman4,EricHerbst5 1IndianCentreforSpacePhysics,Chalantika43,GariaStationRd.,Kolkata,700084,India. 2Laboratoired’astrophysiquedeBordeaux,Univ.Bordeaux,CNRS,B18N,alleGeoffroySaint-Hilaire,33615Pessac,France. 3S.N.BoseNationalCentreforBasicSciences,SaltLake,Kolkata,700098,India. 4AtomicMolecularandOpticalPhysicsDivision,PhysicalResearchLaboratory,Ahmedabad,380009,India. 5DepartmentsofChemistryandAstronomy,UniversityofVirginia,Charlottesville,VA22904,USA. 7 1 0Abstract 2 Propargyl alcohol (HC CH OH, PA) has yet to be observed in the interstellar medium (ISM) although one of its stable isomers, n 2 2 apropenal(CH2CHCHO),hasalreadybeendetectedinSagittariusB2(N)withthe100-meterGreenBankTelescopeinthefrequency Jrange18−26GHz. Inthispaper,weinvestigatetheformationofpropargylalcoholalongwithoneofitsdeuteratedisotopomers, 3HC CH OD (OD-PA), in a dense molecular cloud. Various pathways for the formation of PA in the gas and on ice mantles 2 2 2surroundingdustparticlesarediscussed. Weusealargegas-grainchemicalnetworktostudythechemicalevolutionofPAandits ] deuteratedisotopomer. OurresultssuggestthatgaseousHC2CH2OHcanmostlikelybedetectedinhotcoresorincollectionsof Ahot cores such as the star-forming region Sgr B2(N). A simple LTE (Local thermodynamic equilibrium) radiative transfer model isemployedtocheckthepossibilityofdetectingPAandOD-PAinthemillimeter-waveregime. Inaddition, wehavecarriedout G quantumchemicalcalculationstocomputethevibrationaltransitionfrequenciesandintensitiesofthesespeciesintheinfraredfor . hperhapsfutureuseinstudieswiththeJamesWebbSpaceTelescope(JWST). p -Keywords: Astrochemistry,ISM:molecules,ISM:abundances,ISM:evolution,method: numerical o r t s a1. Introduction hypothesis that the carriers of these bands are aromatic in na- [ ture,consistingmostprobablyoffreemolecularpolycyclicaro- The discovery of large numbers of interstellar and circum- matic hydrocarbons (PAHs), possibly with other heavy atoms 1 stellar species regularly refreshes our understanding of the v such as nitrogen (Salama & Ehrenfreund, 2014; Noble et al., physical conditions of the sources of astrochemical interest 9 2015). Othersuggestionsincludesurfacefunctionalgroupson 0(Herbst, 2006). Astronomical observations along with labora- small grains, quenched carbonaceous composites, amorphous 4tory investigations of various meteoritic samples have discov- carbon,hydrogenatedamorphouscarbonandcondensedphase 6eredthepresenceofnumerousorganicmoleculesofbiological PAHs (Brenner & Barker, 1992; Ja¨ger et al., 2009). Recently, 0 interest (Cronin & Chang, 1993). It is also believed that the . two fullerenes, C60 and C70, have been discovered in infrared 1production of such molecules in star- and planet-forming re- emission in post-stellar objects (Cami et al., 2010) while the 0gionsofinterstellarclouds,whichtendtobepartiallysaturated cationC+ hasbeenconfirmedinnear-infraredabsorptionina 17species containing the elements nitrogen and/or oxygen in ad- diffusec6lo0ud(Walkeretal.,2015). dition to carbon and hydrogen, should be connected in some : vmannerwiththeproductionofterrestrialbio-molecules. Other In order to understand the synthesis of PAHs, either in in- i Xtypes of organic molecules are also present in the ISM. There terstellar or circumstellar regions, it is essential to understand ris strong evidence for species astronomers refer to as “carbon the formation of the six-member aromatic species, benzene achains” in cold and dense interstellar clouds. These carbon (C H ). Sofarthereareonlytwoexperimentallystudiedpath- 6 6 chains are unsaturated and linear species, which can be sim- ways that might result in the synthesis of interstellar or cir- ple hydrocarbons or species with other heavy atoms such as cumstellarbenzene. Thefirstistheadditionofthreeacetylene cyanopolyynes, which contain a terminal cyano (CN) group. (C H )molecules(Zhouetal.,2010)andthesecondisthere- 2 2 Variousinfrared(UIR)emissionbandsinthe3−15µmrange combination of two propargyl (C H ) radicals (Wilson et al., 3 3 have been observed in different astrophysical sources (Alla- 2003). The formation of these radicals could occur in a num- mandola et al., 1985; Tielens & Allamandola, 1987). Labora- berofways. Sharathetal.(2014)carriedoutanexperimentto toryinvestigationsalongwiththeoreticalcalculationsledtothe studythethermaldecompositionofPA,andfoundtheproducts toincludeOHandC H ,suggestingthatPAcouldbeaprecur- 3 3 ∗Correspondence author: Ankan Das; Electronic sortobenzeneformation. Inaddition, Sivaramanetal.(2015) mail:[email protected] found that benzene is the major product from PA irradiation, PreprintsubmittedtoMolecularAstrophysics January24,2017 andsuggestedthatthedissociationofPAplaysakeyroleinthe 2. Chemicalnetwork synthesizeofbenzeneininterstellaricymantles. 2.1. Formationpathways InTable1,allformationanddestructionpathwaysofPAuti- Since PA might play a crucial role in the formation of PAH lizedarepresentedwithratecoefficients,ifapplicable,inboth molecules, it is of interest to explore various aspects of its in- thegasanddustphases.Theratecoefficientsareshownfortwo terstellar chemistry and spectroscopy in detail. Although PA temperatures(T = 10Kand100K)torepresentthetempera- has not been detected unambiguously in the ISM, propenal ture dependency (if any). The determination of the rate coef- (CH CHCHO), one of it isomers, has been detected (Hollis 2 ficients actually used is discussed in the next few subsections. et al., 2004) towards the star-forming region Sgr B2(N). Re- MostratecoefficientsforthecaseofdeuteratedPAarenotvery quenaetal.(2008)estimatedtheabundanceofCH2CHCHOto different, and are not tabulated. Reaction numbers R1-R6 of bearound0.3−2.3×10−9withrespecttotheH moleculeinthe 2 Table 1 represent various possible pathways for the formation galacticcenter.Moreover,PAhasawell-knownrotationalspec- ofPA(HC CH OH). ReactionnumbersR1-R5arefoundtobe 2 2 trum.DependingupontheinternalmotionoftheOHgroup,PA exothermicinbothphasesandareincludedinournetwork.The couldpossesstwostableconformers,namedgaucheandtrans. reactionexothermicities orendothermicities forall thesereac- However,microwavestudiesofPAshowthatthemoleculeex- tionshavebeencalculatedbyusingtheGaussian09(Frischet ists only as the gauche isomer, in which the hydroxyl H atom al.,2009)programwithaB3LYPfunctional(Becke,1988;Lee lies∼ 60◦ outoftheH−C≡C−C−Oplane(Hirota,1968). et al., 1988) and basis set 6-311g++(d,p). Note that reaction Pearson & Drouin (2005) summarized other rotational stud- exothermicities or endothermicities do not differ significantly ies of PA, extended the experimental work of Hirota (1968) betweenthegaseousandicemantlephases. Wecalculatedthe through600GHzandobtainedrotationalanddistortionalcon- endothermicity/exothermicity(∆H)ofareactionbytakingthe stants for the gauche form of PA and its -OD singly deuter- differencebetweenthetotaloptimizedenthalpyincludingzero ated isotopomer. According to their studies, the gauche state point corrections of the products and reactants. If ∆H is posi- is split by inversion into two states, separated by 652.4 GHz tive, we label the reaction endothermic and if ∆H is negative, for normal PA and 213.5 GHz for the -OD isotopomer. Other welabelthereactionexothermic. Anotherformationreaction, spectroscopicworkonPAandrelatedspecieshasalsobeenun- R6,isconsideredonlyinthegasphase. Theindividualforma- dertaken. Nyquist (1971) recorded and assigned Infrared and tionreactionsarediscussedinthefollowingparagraphs. Raman spectra for PA, and its deuterated isotopomers, while Reaction R1 (O+C H ) in the gas phase was studied by 3 3 Devendra & Arunan (2013) carried out experiments to deter- Kwon et al. (2006), who carried out an experiment as well as minethestructureoftheAr..PAcomplexanditstwodeuterated abinitiostatisticalcalculations. Theyfoundthatthereactionis isotopologues. TheyfoundPAtohaveagauchestructure,with barrier-less and can produce propynal (HC CHO) and H. The 2 Arlocatedinbetweenthe-OHand-C≡C-Hgroups. Inanother conversionofpropynalintoPAthencanoccurviatwoassocia- study, Devendra & Arunan (2014) carried out experiments for tionreactions(R3,R4)withatomichydrogen. Inthegasphase, the pure rotational spectra of the PA dimer and its three deu- theprocessoccursviaradiativeassociation,inwhichemission teriumisotopologues. of a photon stabilizes the intermediate reaction complex. Lee etal.(2006)predictedthatthebarrier-lessadditionofO(3P)to propargylradical(C H )onthelowestdoubletpotentialenergy In this paper, we report the use of our interstellar chemical 3 3 surfacecouldproduceseveralenergy-richintermediates,which model to explore various pathways for the formation and de- undergo subsequent isomerization and decomposition steps to struction of PA (gauche form), and to estimate the possibility generatevariousexothermicreactionproducts. Theirstatistical of detecting this molecule in a dense molecular cloud. Since calculationalsosuggeststhattheprimaryreactionchannelleads there are some existing laboratory results for the spectrum of to the formation of propynal. Reaction R2, in which the radi- the -OD deuterated form of PA and since some observational calCCHandformaldehydeproducepropynal+H,wasstudied evidence for deuterium fractionation of large complex species byDongetal.(2005)andbyPetrie(1995). Dongetal.(2005) exists (see for instance DCOOCH /HCOOCH , Demyk et al. 3 3 calculated a very small barrier of 2.1 kcal/mol at the highest (2010)), we also consider the -OD isotopomer of PA. Various leveloftheory, whilePetrie(1995)assumedthechanneltobe vibrationaltransitionsofPAarecomputedandcomparedwith barrier-less based on similar reactions. The propynal product theexistingexperimentalresults. canthenalsobehydrogenatedtoPAviaR3andR4. AsanalternativetotwosuccessiveH-atomassociationreac- Theremainderofthispaperisorganizedasfollows. InSec- tions involving atomic hydrogen, we checked the reaction of tion2,wediscussvariousreactionpathwaysandtheirrateco- H with propynal (HC CHO) to form PA but found it to be 2 2 efficientsfortheformationanddestructionofPAandOD-PA. highly endothermic. We tried a few other pathways for the InSection3.1,modelingdetailsarepresentedwhileinSection formation of PA via single step reactions, sometimes involv- 3.2wediscussmodelingresults.. LTEradiativetransferresults ingaradical. Inthiseffort,weconsideredthereactionbetween arepresentedinSection4.1,whilecomputedvibrationalspec- C H andCO,thereactionbetweenpropynalandH O,andthe 2 4 2 traforPAandOD-PAarediscussedinSection4.2. Finally,in reactionbetweenthepropargylradicalandOH(reactionnum- Section5,wedrawourconclusions. ber R5). The reactions between C H and CO and between 2 4 2 Table1:FormationanddestructionpathwaysofPAanditsrelatedspecies. Reactionnumber Reaction α β γ Gasphase Icephase (type) ratecoefficient ratecoefficient atT=10K(100K) atT=10K(100K) Formationpathways R1(NR) O(3P)+C3H3→HC2CHO+H(−252.30a,−231.11b) 2.3×10−10 0.0 0.0 2.3×10−10(2.3×10−10)x 2.83×10−7(2.51×104)y R2(NR) C2H+H2CO→HC2CHO+H(−109.17a,−106.307b) 1.00×10−10 0.0 0.0 1.00×10−10(1.00×10−10)x 2.84×10−17(3.03×103)y R3(NR) HC2CHO+H→HC2CH2O(−86.69a,−86.23b) - - - 0(1.96×10−15)x 6.24×10−9(8.94×10−3)y R4(RR) HC2CH2O+H→HC2CH2OH(−420.35a,−421.67b) 1.00×10−10 0.0 0.0 1.00×10−10(1.00×10−10)x 1.77×10−1(2.54×105)y RR56((RDRR)) CH3CH23C+H2OOHH→2++HCe−2C→H2HOCH2C(−H320O1H.35+aH,−298.97b) 12..0607××1100−−180 0−.00.59 00..00 21..0000××1100−−170(5(1.1.060××1100−−81)0x)x 7-.86×10−29(2.19×102)y Destructionpathways RRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRR789111111111122222222223333333333444444444455555555556666666666777777012345678901234567890123456789012345678901234567890123456789012345(((((((((((NNI((((((((((((((((((((((((((((((((((((((((((((((((((((((((((DDDDDDDDNIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIPPPPPPPPCCCCCCCCDRRNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNRRRRRRRR)HHHHHHHHRRRRRRRRR)))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))) HHHHHHHHHHHHHHHHHHHHHHHHHHHHHHHHHHHHHHHHHHHHHHHHHHHHHHHHHHHHHHHHHHHHHCCCCCCCCCCCCCCCCCCCCCCCCCCCCCCCCCCCCCCCCCCCCCCCCCCOOCCCCCCOOCCCCCCCCC22222222222222222222222222222222222222222222222222222222222222222CCCCCCCCCCCCCCCCCCCCCCCCCCCCCCCCCCCCCCCCCCCCCCCCCCCCCCCCCCCCCCCCCCCCC2222HHHHHHHHHHHHHHHHHHHHHHHHHHHHHHHHHHHHHHHHHHHHHHHHHHHHHHHHHHHHHHHHHHHHHOOOOOOOOOOOOOOOO2222222222222222222222222222222222222222222222222HHHHOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOCCCC+++++++++++++++++++HHHHHHHHHHHHHHHHHHHHHHHHHH++++++++++++++++++++HHHHCCHHHHHHCHHHHHPC2222+++++++++++++++++++++++++OOOOCCHHHHHHCHHHHHPPPCCC++++HHR++3+3+3++++2Cee++eeeHHHHDHD++HRRROOCCHHHHHHCHHHHHOD3+3+3++++2PPCCCee++++++OPO+3−−−→→→→→→→→OD++eeHHOOOPPP+++3HRRO3+3+3++++2++HDCeeH→→→→→→→→T→→++++eeee+−−→→→→OD++++HHHTTT→→OOPPO+3+−−−−OO→→→→→→→→→→→HCCHCCHC→→PPCC→→→OOO++HHOOO→→TT→→HCCHCCHHCC+NT→→→→→→33333HCCHHHHCCRRCCOONNNTTTOOH3→→HHHO333HCHC33CHCHHHHHCCC→3OHHH23C2CHCOO→PPH→HHHHH34HHCCNNTTO+3333++2+HCOH233CCCCCCH→→→CCCCCC→→→HHHH23OOOTTCCHO+2HH+2+22COO33++HHCC+++2C2H333HO+H22322+2CC2+++→→C→→OOHH+OOC22CC++C+3+CHHHH++3+3CCOCCC2++2C++++CCCHHCHCHH2C223C+C+2OHHNNCC+33TT2HHH++22HHCOC2HHHHH332O++2+2++++H3OHHCH2C23HHH+HC2HHHHCH2C2H+OHHeCCCC222H+HCHH+++CHHH2O2→→CC22222+→→+32HeHCCC232HHOO22HH+e22HHC+O+OOOH+OOOOO2+2HH+2H2COH22HOH2COOHeH2O+2O+HOH+HO2+OOC+222H22CCHCH+OO+2+2++HH++HH+HHO+3CHHe+HH+3COOe2+++22H+O2OO++++++H+HHO22OO++++2++++CC+CHOOO+HCC+CHHOH+OH+OHHC+HH++++2COOHHCHHH+HHH++22O+222+OCC22HHH2OHOHO++2O2H+CCC22D+HDOCC+2HH2OHH2HOHH+CHH+DH22HHH+HHO2HH+H+HHHH22HO2D+H3+C++HO22H+22+2O4OO+HOOHHCHHHHHH2O2O++++OOOOHDHD 992211333316666333327766311115111122221444426666666611111111222841833.....................................................................2200747733844442881700114222294410663324444200000000333333330000589000033068811633331007433777000081182220096666314700×××××××××××××××××××××××××××××××××××××××××××××××××××××××××××××××××××××111111111111111111110000000000000000000011111111111111111111111111111111111111111111111110000000000000000000000000000000000000000000000000−−−−−−−−−−−−−−−−−−−−11111111111111117778−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−0000000077777777110000000000000000000000000000000000000000008777722999999999999999999999998888999999999999999 00------------------------------------------0000000000000000---------000000000000000000000000000000000000000000000000000..................000000000000000000...................................................5555555555555555555555555555555555555555555555555559999 000000000000000000000000000000000000000000001111111177777777000000000....................................................55555555.........000000000000000000000000000000000000000000008888888800000000022222222 991118221113333122114433266663887611118222219999999922222222111531611.....................................................................22220722991888891185005508888488436644088884111111114444444411193466600222977992666639938445509999466084444900000444444444444444400054144×××××××××××××××××××××××××××××××××××××××××××××××××××××××××××××××××××××111111111111111111111011111111111111111111111111111111111111111111111100000000000000000000000000000000000000000000000000000000000000000000−−−−−−−−−−−−−−−−−−−−−7−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−0000(00000000000000001100000000000000000000001111111111111111666766668888188999988889888882288898888888888888888888888888844444444((((((((.33318355((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((79944338877411117776511119999999922222222........2133226655311115622244456622........................................×................2266835555244442663544221111111144444444....666230042559566882111169444×0031526611166669448111444444444444444444107728220063333318111×××××××××××××××××××××××××××××××××××××××××××××××01××××××××××××××××××××11111110−11111111111111111111111111111111111111117111111111111111111110000000−0000000000000000000000000000000000000000)700000000000000000000−−−−−−−x−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−)7777777−−−−−−−−−−−−−−−−−−−−x1100000000000000000000001111111111111111)))))))000000000000000000002299999999988889999988888888888844444444xxxxxx98888899999999988889x))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))yyyyyyyyyyyyyyyyxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxx33------------------------------------------9999999922222222---------..................541111111144444444884444444444444444××××××××××××××××××111111111111111111000000000000000000−−−−−−−−−−−−−−−−−−331111111111111111888888888844444444((((((((((((((((((339999999922222222..65................1111111144444444444444444444444444××××××××××××××××××11111111111111111100000000000000000011−−−−−−−−−−−−−−−−))1111111111111111yy8888888844444444))))))))))))))))yyyyyyyyyyyyyyyy Notes:Forthetwo-bodyreactions(R1-R2,R4-R8,R67-R75),ratecoefficientsaretabulatedasα( T )βexp(−γ).Forthephoto-dissociationreactionswith rateecxoteeffirncaileinnttsearsrteeltlaabruplahtoetdonassα(R5γ1(cid:48)-R.5F8o)r,trhaeteractoeecffioceiffienctisenatrseotafbiounla-tpeodlaarsnαeuetxrpal(−reγaAcVti)o.nFso(rRp9h-Rot5o0-3d)0,0iswsoecuiasetiothneTbryelcaotisomnidcirsacyusisnedducinedSpuh&otoCnhses(Rna5v9i-cRh6(61)9,82). Tabulatedratecoeffic1−ieωntsfortheseion-neutralreactionsintermsofα( T )βexp(−γ)arevalidforthelowtemperatureregimeonly. agasphaseenthalpyinkJ/mol 300 T bicephaseenthalpyinkJ/mol xratecoefficientincm3s−1 yratecoefficientins−1 3 propynalandH Oarefoundtobehighlyendothermicinnature (1995) estimated a rate coefficient of 1.0×10−10 cm3 s−1 for 2 whereasreactionR5ishighlyexothermicandislikelybarrier- reactionR2withtheassumptionthatitoccurswithoutabarrier less,sincethereactantsarebothradicals. Sincebothreactants inthegasphase. Althoughthisassumptioncontradictsthecal- inR5(C H andOH)arereasonablyabundantintheISM,we culationofasmallbarrierbyDongetal.(2005),weassumethe 3 3 thinkthatthisreactioncancontributetowardstheformationof reactiontobebarrier-less,andusetheestimatedratecoefficient interstellarPA,although, inthegasphase, itmustproceedvia ofPetrie(1995). a radiative route, so that it must be looked at closely. On ice As can be seen in Table 1, reactions R3, R4, and R5 are mantles,however,radical-radicalassociationreactionsarenor- highly exothermic in nature. Based on the high exothermicity mallyquiteefficient. AscanbeseeninTable1,R4andR5are of these reactions, one might assume that these reactions can thesoleradical-radicalreactionsleadingdirectlytotheforma- processwithoutbarriers. tionofPA,bothinthegasandontheice. Agasphasedissocia- The Hydrogen addition reaction of HC CHO (reaction R3) 2 tiverecombination(DR)reaction(R6),inwhichprotonatedPA may occur in two ways. First, H addition may occur with andanelectronrecombinetoformsmallerneutralproducts,is the O atom of the CHO group and produce HC CHOH and 2 alsoincluded. Thisreactionmaycontributesignificantlytothe secondly H addition may occur with the C atom of the gasphaseformationofPAunlessthetwo-bodyproductchannel CHO group and produce HC CH O. Our quantum chem- 2 2 shownisunimportantifion-neutralprocessescanproducepro- ical calculation found that the hydrogen addition to carbon tonatedPAefficiently. Similarpathwaystoallthoseconsidered is more favourable than the hydrogen addition to oxygen in for the synthesis of PA are assumed to be responsible for the the CHO group. Using the DFT/6-31+G(d,p) method, we productionofOD-PA. found that the gas-phase activation barrier (∆E‡) and Gibbs energy of activation (∆G‡) for the second possibility of re- 2.2. Destructionpathways action R3 (H+HC CHO→HC CH O) is 3.74 kcal/mol and 2 2 2 As shown in Table 1, the destruction of gaseous PA occurs 9.63kcal/molrespectively. via ion-neutral (IN) and photo-dissociative (PH & CR) path- Forthecomputationofthegasphaseratecoefficientforre- ways, as well as via two radical-neutral (NR) reactions. The action R3, we use transition state theory, which leads to the destructionofgasphasePAalsooccursviaadsorptionontoice, Eyringequation(Eyring,1935): but the reverse process of desorption also occurs. These processes are in our network, but not listed in Table 1. The rate k=(kBT/hc)exp(−∆G‡/RT). (1) coefficientforthegasphaseNRreactionbetweenthehydroxyl Theratecoefficientcalculatedbytheaboveequationthushasa radical(OH)andPAwasmeasuredbyUpadhyayaetal.(2001), strong temperature dependence. In Table 1, we have included whousedlaserphotolysiscombinedwiththelaserinducedflu- the rate coefficient for two temperatures; 10 K and 100 K. At orescence technique at room temperature. According to their 10 K, rate coefficient is ∼ 0 and at 100 K it becomes 1.96× study,thisreaction(R7)producesanadduct,HOC HHCHOH. 2 10−15 cm3 s−1. Thus in the low temperature regime, R3 does SincetheabundancesofOHandODarecomparableinadense notcontributeatalltothegasphaseformationofPA,whileit cloud (Millar et al., 1989), a similar destruction mechanism couldplayarolefortheformationofPAinthehightemperature with OD (R8) is also considered here. In addition to the gas- regime. phase, reactions R7 and R8 are included in the ice phase as InthecaseofreactionsR4andR5,wewereunabletolocate well. Fortheion-neutral(IN)destructionofgasphasePA,we anysuitabletransitionstateandassumethatthesetworeactions include reactions R9-R22 by following the similar gas phase are barrier-less, as is customary for radical-radical reactions. ion-neutral(IN)destructionpathwaysavailableformethanolin The rate coefficients of these two reactions are assumed to be Woodall et al. (2007). Similar IN destruction reactions (R23- 10−10 cm3s−1. TheformationofOD-PAistreatedwithsimilar R50)forHC CHOandHC CH Oarealsoconsidered. Photo- 2 2 2 pathwaysandratecoefficients. dissociationreactions(directorcosmicrayinduced)(R51-R66) For the formation of PA by the DR mechanism, which in- are also responsible for the destruction of PA and its asso- + volvesthedestructionofHC CH OH (R6,R67-R75),wefol- ciated species in both phases. In Table 1, DR reactions R6 2 2 2 + and R67-R75 involve the ions (HC CH O+, HC CH OH + lowthedestructionofCH3OH2 fromWoodalletal.(2007)for andHC CH OH+); theseionsarepro2duce2dbyvari2ous2form2a- theratecoefficientsandproductchannels. Ratecoefficientsbe- 2 2 tion/destructionpathwaysofPAanditsrelatedspecies. Forthe tweentwospeciesarestandardlyparameterizedasafunctionof temperaturebytheequation destructionofOD-PAandotherassociatedspeciessimilarde- structionpathwaystoallthoseconsideredforPAareincluded. k=α(T/300)βexp−γ/T. (2) 2.3. Ratecoefficients Inthisparticularcase,thevaluesoftheparametersforthefor- 2.3.1. Gasphaseratecoefficients mationofPAareα=2.67×10−8cm3s−1,β=−0.59andγ=0. Slagleetal.(1991)experimentallyobtainedatemperaturein- Similar product channels and rate coefficients are used for the dependentratecoefficientof∼2.31×10−10cm3s−1forreaction formationofOD-PAbyaDRreaction. R1. Accordingtotheirstudy, thisreactionproceedsthrougha Now let us consider the destruction of gaseous PA. Upad- fastandirreversibleassociation-fragmentationprocess. Weuti- hyaya et al. (2001) studied the rate coefficient of the NR re- lizetheratecoefficientobtainedbySlagleetal.(1991). Petrie action between PA and the OH radical (R7). According to 4 their study, it produces an adduct with a rate coefficient of Temperature (K) (9.2±1.4)×10−12 cm3s−1,whichweusewithnotemperature 1e+0710.0 57.5 105.0 152.5 200.0 dependence. Herealso,weassumeasimilarratecoefficientfor reactionR8, whichinvolvesOD.Similardestructionreactions andratecoefficientsareadoptedforOD-PAaswell. Ion Neutral (IN) reactions are the dominant means for the -3m)1e+06 destructionofneutralinterstellarspecies. Iftheneutralspecies y/ c is non-polar, we use the Langevin collision rate coefficient nsit e (isHpeorblasrt,,w20e0e6m; pWloayketlhaemtraejteaclt.o,ry20s1c0a)l.ingIfrethlaetionneuftoraulndspienciSeus log (D1e+05 & Chesnavich (1982) and Woon & Herbst (2009). From our quantum chemical calculations, we find that the polarizability α = 5.62×10−24 cm3 anddipolemomentµ =1.6548Debye d D 10000 forPA.Forthedestructionreactionsoftheothertwoassociated 1e+06 1.025e+06 1.05e+06 1.075e+06 1.1e+06 log (Time/year) species(HC CHO, HC CH O),inreactionnumbersR23-R36, 2 2 2 weuseαd =5.26×10−24cm3andµD=3.08Debye,andforR37- Figure1: Adoptedphysicalconditionsforthewarm-upphaseinwhichthe R50weuseα =5.93×10−24cm3andµ =1.1480Debye. For densityincreasesasthetemperatureincreases.. d D deuteratedPA,theion-neutraldestructionreactionshavesimi- larratecoefficientstothoseofnormalPA,theonlydifferences becauseoftheincreasingoverlapbetweenthecrosssectionsfor beingduetothereducedmass,whicharerathersmall. Deuter- photo-dissociation and the cosmic-ray-induced emission spec- atedreactionsandtheirratecoefficientsarenottabulatedhere. trum (Gredel et al., 1989), we use γ(cid:48) = 752.0 (in Table 1, we Forthephoto-dissociationreactionsofPAanditsassociated presented it under the column head marked γ) for our calcu- speciesbyexternalinterstellarphotonsandcosmicray-induced lationsofthecosmic-ray-inducedphoto-dissociativereactions. photons, we use analogous products and the same first-order The same photo-dissociation rate coefficients are adopted for rate coefficients (s−1) as the case of CH OH (Woodall et al., 3 thedestructionofOD-PAanditsassociatedspecies. 2007). For the case of external photons, we use the following Dissociative recombination (DR) reactions and rate coeffi- relationfortheratecoefficients: cients for some of the associated ions of PA are shown in re- k=αexp(−γA ) (3) actionsR67-R75. Pathwaysandratecoefficientsofthesereac- V tionsareadoptedbyfollowingthesimilarDRpathwaysavail- where α represents the rate coefficient (s−1) in the unshielded able for CH O+, CH OH +, and CH OH+ in Woodall et al. 3 3 2 3 interstellar ultraviolet radiation field, A is the visual extinc- (2007). Since for all these reactions, β (cid:44) 0, the reactions are V tion,forwhichweuseavalueof10,andγisusedtotakeinto temperaturedependent. ThesameDRratecoefficientsareused account the increased extinction of dust in the UV. Here, fol- fortheassociatedionsofOD-PA. lowingWoodalletal.(2007),weuseα = 6.0×10−10 s−1,and For two or three-body gas-phase reactions in Table 1, the γ = 1.8inourmodel. Incorporatingalltheparametersintothe rate coefficients are represented in terms of the three rate co- aboveequation,weobtainaphoto-dissociationratecoefficient efficients, α, β and γ. Most of the gas phase rate coefficients ofabout9.14×10−18s−1. adoptedhereareeitherestimatedortakenfromsimilarkindof Forcosmic-ray-inducedphoto-reactions, weusethefollow- reactions. For reactions R1-R2, R4-R5 and R7-R8, we assign ingrelation, whichwasoriginallydevelopedby(Gredeletal., β and γ to be zero. For the dissociative recombination reac- 1989): tions(R6andR68-R75), thesethreecoefficientsareestimated based on similar reactions listed in Woodall et al. (2007). For k=αγ(cid:48)/(1−ω) (4) reaction R3, we calculate the gas phase rate coefficient by us- where α is the cosmic-ray ionization rate (s−1), γ(cid:48) is the num- ing transition state theory, which leads to the Eyring equation (Eyring, 1935). Thus for reaction R3, these three parameters berofphoto-dissociativeeventsthattakeplacepercosmic-ray are not shown. For destruction by photo-reactions, we supply ionization and ω is the dust grain albedo in the far ultraviolet. Here, weuseω=0.6, α = 1.3×10−17 s−1, andγ(cid:48) = 752.0by these three coefficients following the similar type of reactions availableinWoodalletal.(2007). Forthedestructionofpolar following the cosmic-ray-induced photo-reactions of CH OH 3 neutral species by ions, we use the two relations discussed in inWoodalletal.(2007).Byincludingtheseparametersintothe aboveequation,weobtainaratecoefficientof2.44×10−14s−1, Su&Chesnavich(1982).Theserelationscannotberepresented over the whole temperature range in terms of one set of three whichisroughly4ordershigherthantherateofexternalphoto- coefficients. However,wecantabulateα,βandγforreactions dissociativereactions. Cosmicrayinducedphoto-reactionscan R9-R50inthelowtemperatureregime. playanimportantroleininterstellarchemistry. Thechoicesof theseparametersarehighlyreactionspecificandawrongesti- mation might lead to misleading results. In the UMIST 2006 2.3.2. Icephaseratecoefficients database,γ(cid:48)rangesashighas5290andaslowas25.0. Though Chemical enrichment of interstellar grain mantles depends thehighervaluesofγ(cid:48)aremorereliableforthelargermolecules on the binding energies (E ) and barriers against diffusion d 5 1e-09 1e-09 (a) Isothermal phase ) ce1e-12 n a d n u b1e-15 PA (gas) 1e-12 A OD-PA (gas) og ( POAD -(PicAe) (ice) ce) l1e-18 n a d n 1e-08 u 1e-10 (b) Ab1e-15 CH (gas) ( nce)1e-12 HH3CC2CC3HHOO ( g(gaass)) log a1e-14 2 2 d CH a=0.00 un1e-16 H3CC3 (HiceO) (ice) a=0.03 b 2 a=0.30 A1e-18 HC2CH2O (ice) 1e-18 ( og 1e-20 l 1e-22 1e-24 1e+03 1e+04 1e+05 1e+06 1e+03 1e+04 1e+05 1e+06 log (Time/year) log (Time/year) Figure2: Chemicalevolutionof(a)PAandOD-PAand(b)theirrelatedspecies Figure3: ChemicalevolutionofPAduringtheconstantdensityisothermal duringtheisothermalphaseatnH=104cm−3withaconstantvisualextinction phaseatnH = 104cm−3withaconstantvisualextinctionof10anddifferent of10. valuesofthenon-thermalreactivedesorptionparametera. (E ) of the adsorbed species. The binding energies of these b 2004; Lattelais et al., 2011; Garrod, 2013). Since PA has an - species are often available from past studies (Allen & Robin- OHgroup,creatinghydrogenbondswithawatersubstrate,this son,1977;Tielens&Allamandola,1987;Hasegawa&Herbst, moleculeisthereforeexpectedtohaveahigherbindingenergy, 1993; Hasegawa et al., 1992). But these binding energies close to that of water. Here, we assume the binding energy of mostly pertain to silicates. Binding energies of the most im- bothPAandOD-PAtobethesameasmethanol(5530K).Since portantsurfacespeciesonice,whichmostlycontrolthechemi- HC CHOforms by the reaction between C H and O, we add calcompositionofinterstellargrainmantles,areavailablefrom 2 3 3 thebindingenergiesofC H (2220K)andO(800K)todeter- some recent studies (Cuppen & Herbst, 2007; Garrod, 2013). 3 3 mineabindingenergyvalue3020K.Inthecaseofthebinding We use these latter energies in our model. For the rest of the energyofHC2CH2O,weaddthebindingenergiesofHC2CHO speciesforwhichbindingenergiesarestillunavailable,weuse (3020 K) and H (450 K) to obtain 3470 K. For other binding thesamevaluesasinpaststudiesorestimatenewvalues. For energiesforspeciesinreactionR1,R2,R3,R4andR5,weuse barriersagainstdiffusion,whicharepoorlyknown,weuseval- E (C H) = 1460K,E (H CO) = 2050KandE (OH) = 2850 uesequalto0.35E (Garrod,2013). d 2 d 2 d d K.OurtransitionstatetheorycalculationfoundthatreactionR3 For the formation of PA in the ice phase, we include sur- (hydrogenadditiontothecarbonatomoftheCHOgroup)con- facereactionsR1-R5. Theratecoefficients(Rdiff)ofthesere- tains an activation barrier of about 3.59 kcal/mol (1807 K) in actions were calculated by the method described in Hasegawa theicephase. et al. (1992), which is based on thermal diffusion. They de- finedaprobabilityfactorκtodifferentiatebetweenexothermic Destruction of ice phase PA occurs by various means; thermal desorption, non-thermal desorption from interstellar dust reactionswithoutactivationenergybarriersandreactionswith activation energy barriers (E ) in such a way that the effective grains via exothermic surface reactions, cosmic ray induced a rate coefficient becomes R = κ×R . Thefactor κ is unity desorption, photo-dissociation, and reaction with OH or OD ij diff radicals. Intactphoto-desorptionisnotincluded. Thermaldes- intheabsenceofabarrier. Forreactionswithactivationenergy orptionplaysacrucialroleonlyathightemperatures,depend- barriers,κisdefinedasthequantummechanicalprobabilityfor tunneling through a rectangular barrier of thickness a(= 1 Å), ingupontheadsorbedspecies. Forthermaldesorption,weuse therelationprescribedbyHasegawaetal.(1992). Atlowtem- whichiscalculatedfromtheequation peratures, non-thermal desorption by reaction exothermicity, κ=exp[−2(a/(cid:126))(2µE )1/2]. (5) photons, and cosmic rays serve as important means for trans- a ferringthesurfacemoleculestothegasphase. Fornon-thermal Binding energies to the surface for some complex organics reactivedesorption(Garrod&Herbst,2006), weconsiderthat with a hydroxyl group are normally considered to be higher allsurfacereactionswhichresultinasingleproductbreakthe due to the phenomenon of hydrogen bonding (Collings et al., surface-molecule bond with some fraction a. Here, we use a 6 Temperature (K) 6384 reactions involving 641 gas phase species and the sur- 10.0 57.5 105.0 152.5 200.0 facechemicalnetworkconsistsof358reactionsinvolving291 1e-08 ) surface species. The gas phase chemical network is mainly nce1e-10 Collapsing and warm-up phase (a) adoptedfromtheUMIST2006database(Woodalletal.,2007), a d1e-12 but contains some deuterated reactions as well. To minimize bun1e-14 POAD -(PgaAs )(gas) thehugecomputationaltimeandavoiddifficultyinhandlinga A PA (ice) large chemical network, we have considered only some dom- g (1e-16 OD-PA (ice) inant deuterated reaction pathways. Deuteration reactions im- o l1e-08 portantfortheincreaseofdeuterationinthecoldgasphaseare ) e included in our network. Our selection was based on some anc1e-12 C3H3 (gas) (b) earlier studies by Rodgers & Millar (1996); Roberts & Mil- bund1e-16 HHC3CCH22CC3 (HHicO2eO) ( g(gaass)) lwarhi(c2h0w00e)r;eRaodbdeerdt/supetdaatle.d(2a0n0d3i)d.eAntlilfitehdeadsetuhteerdaotemdinreaancttpioanths-, A HC2CHO (ice) g (1e-20 HC2CH2O (ice) waysfortheformationofessentialdeuteratedspeciesintheta- o bles(Table13-18)ofAlbertssonetal.(2013)arealsoincluded l1e-04 here. Forthegrainsurfacereactionnetwork,weprimarilyfol- e) (c) c lowHasegawaetal.(1992)andCuppen&Herbst(2007),while n a1e-08 for the ice phase deuterium fractionation reactions, we follow nd H2O (gas) u CH3OH (gas) Caselli (2002) and Cazaux et al. (2012). In addition to this, Ab1e-12 CHH2O4( (giacse)) weconsidersomereactionsinvolvingPAandcoupledspecies, log (1e-16 CCHH34O (iHce )(ice) whIincihtiaalreeldeemscernitbaeldabinunSdeacnticoens2w.ithrespecttototalHnucleiin 1.000e+06 1.025e+06 1.050e+06 1.075e+06 1.100e+06 allformsareshowninTable2. These“lowmetal”abundances Time (year) are often used for dense clouds, where hydrogen is mostly in theformofmolecularhydrogenandionizationofthemediumis Figure4: Chemicalevolutionof(a)PAandOD-PA(b)C3H3,HC2CHOand mainlygovernedbycosmicrays.Weadoptedthesevaluesfrom HC2CH2O(c)H2O,CH3OHandCH4 forthewarm-upphase,duringwhich bothdensityandtemperatureincrease. Leungetal.(1984). Itisassumedthatinitiallyalldeuteriumis locked up in the form of HD. The initial abundance of HD is assumedtobe1.6×10−5(Roberts&Millar,2000)withrespect fiducialvalueforthefractionofa = 0.03. Forthecosmicray tothetotalnumberofhydrogennuclei. induced desorption rates, we follow the expression developed In order to consider suitable physical conditions for a star by Hasegawa & Herbst (1993). Photo-dissociation by direct formingregion,weadoptamodelwithaninitialphaseofcon- (R51-R58) as well as cosmic ray induced photons (R59-R66) stantdensity(n =1.0×104cm−3)andtemperature(T = 10K) H are considered for the destruction of ice phase PA and its two followedbyawarm-upphase(temperatureincreasesfrom10K associatedspecies,andtheratecoefficientswereassumedsimi- to 200 K). The warm-up phase is assumed to be accompanied lartotheirgasphaseratecoefficients. Thedissociatedproducts by an increase of density from n = 104 cm−3 to n = 107 H H remainonthegrainsurfaceandarethusabletoreactanddesorb cm−3, relevant when the material approaches the inner proto- dependingontheirbindingenergies. Forthedestructionofice stellar regions and appropriate for hot-core conditions. Both phasePAbyreactionswithOHandOD(R7andR8),wecon- phases have a visual extinction of 10. The initial phase is as- sider the destruction pathways proposed by Upadhyaya et al. sumed to be sustained for 106 years and the subsequent phase (2001)forthegas-phase, butwecalculatetheratecoefficients lasts for another 105 years (a typical lifetime for a young em- of these ice phase reaction pathways by using diffusive reac- beddedstage(Evansetal.,2009)). InFig. 1,theadoptedtime- tions with barriers against diffusion, which have already been dependentphysicalconditionsforthesecondstageofevolution discussed. SelectedicephaseratecoefficientsusedforPAare areshown. shown in Table 1 for T = 10 K and 100 K. Desorption and Fortheriseindensityandtemperature,weassumeconstant accretionratecoefficientsarenotincluded. slopesm andm respectivelyforthedensityandtempera- den temp tureasdeterminedfromthefollowingequations: 3. Chemicalmodeling ρ −ρ m = max min =99.9cm−3yr−1, (6) den Time −Time 3.1. Model f i We use our large gas-grain chemical network (Das et al., T −T m = max min =1.9×10−3Kyr−1. (7) 2015a,b) for the purpose of chemical modeling. We assume temp Time −Time f i thatthegasandgrainsarecoupledthroughaccretionandther- mal and non-thermal desorption processes. A visual extinction of 10 and a cosmic ray ionization rate of 1.3×10−17 s−1 3.2. ModelingResults are used. Including the deuterated reactions and deuterated Figure2a,bshowsthechemicalevolutionofPA,OD-PAand species, our present gas phase chemical network consists of their related species in the gaseous (solid line) and ice phase 7 Table2:Initialelementalabundanceswithrespecttototalhydrogennuclei. Species Abundance H2 5.00×10−01 He 1.40×10−01 N 2.14×10−05 O 1.76×10−04 e− 7.31×10−05 C+ 7.30×10−05 S+ 8.00×10−08 Si+ 8.00×10−09 Fe+ 3.00×10−09 Na+ 2.00×10−09 Mg+ 7.00×10−09 HD 1.6×10−05 (dashed line) during the cold isothermal phase. The ice phase K). productionofPAisdominatedbythesuccessiveHassociation Ascanbeseen,thegasphaseproductionofPAandOD-PA reactions (R3 and R4). Similarly, for the case of OD-PA, the is found to be favourable around the high temperature region. associationreactionsarefoundtobedominant. Thepeakabun- This happens due to the increase in the rate coefficient of re- dances of PA and OD-PA and their related species are listed action R3 with the temperature. For the ice phase reactions, inTable3,forbothphasesofourcalculation. Forexample,in HorDadditionreactionsbecomeirrelevantbeyond40K,be- theisothermalcoldphase,thepeakfractionalabundancesofPA yond which the C H and OH or OD radicals become mobile 3 3 and OD-PA in the ice p are 5.03×10−10 and 1.33×10−11 re- enoughtoreactwitheachotherongrainsurfaces. Atthesame spectivelyandforthegasphase,thesevaluesare1.07×10−12 time, the destruction of PA and OD-PA by reactions with OH and1.32×10−13 Notethattheiceandgasphasepeaksdonot and OD increases due to the increase in the destruction rates occuratthesametime. with increasing temperature. Above 100 K, C H and OH or 3 3 Non-thermaldesorptionprocessessignificantlycontributeto ODradicalsroughlydisappearandtheonlysourceoficephase themaintenanceofthegas-phaseabundancesofPAandOD-PA PAorOD-PAisaccretionfromthegasphase.Sincethedensity withourfiducialvalueaof0.03forreactivedesorption. Fig. 3 increaseswithtime,accretionofthegasphasespeciesbecomes show the gas phase abundance of PA versus time with a = 0, more favourable. Above 140 K, the gas phase abundances of a = 0.03anda = 0.3intheisothermalphase. Forthecaseof PAandOD-PAareroughlyinvariant. Fig. 3,throughoutsomebutnotallofthetimeoftheevolution- In Table 3, we list the peak abundances of PA, OD-PA and ary stage, the PA abundance for the a = 0.03 case is slightly their related species for the warm-up phase along with their higherthaninthecasewithoutreactivedesorption(a=0). For correspondingtimesandtemperatures. Inthisphase, thepeak the highest ‘a’ value (a = 0.3), with the most rapid reactive abundances of ice phase PA and OD-PA are found to be de- desorption,thegasphasepeakabundanceissignificantlylarger creased in comparison with the isothermal phase. The peak thanthatwitha = 0.03andalsowitha = 0. Intheisothermal fractional abundances of ice phase PA and OD-PA are found phase,thepeakabundanceofgasphasePAforthesethreeval- to be 7.09×10−11 at 32.2 K, and 7.23×10−12 at 47.3 K, re- uesofaisfoundtobe8.69×10−13,1.07×10−12and1.12×10−10 spectively, whereas, forthegasphase, thecorrespondingpeak fora=0.00, 0.03and0.30respectively. Attheinitialstagesof abundancesare1.20×10−9and2.14×10−12respectively. evolution,themodelswitha=0anda=0.03giveverysimilar Although reality is more complex due to the consideration gasphaseabundancesThismeansthatthegasphaseproduction of variable density and temperature in the second phase, the ofPAinourfiducialmodelisactuallydominatedbygasphase basicexplanationforthehigherabundancesofPAandOD-PA chemistry(byreactionR5)aslongastheaparameterisnottoo inthegasphaseisthatinsteadofchemicaldesorption,thermal high. desorption becomes much larger at higher temperatures, and Figure4a,b,creferstothewarm-upportionofoursimulation. indeedtheiceabundancesbecomeinfinitesimalattemperatures Fig. 4arepresentstheevolutionofPAandOD-PAwhereasFig. above100K.Inthefirstphaseofoursimulation(Fig.2),which 4brepresentstheevolutionofPArelatedspeciessuchasC H , occurs while the temperature is 10 K, only non-thermal slow 3 3 HC CHO andHC CH O andFig. 4crepresents theevolution desorptionoccurs,andindeedtheiceabundancesarehigher. 2 2 2 of the major icy species water, methane and methanol for the For the gas-phase at higher temperatures, at which there is gas and ice phases. The temporal evolution of gas phase and verylittleice,fractionationoccursviagas-phaseformationand icespeciesshowninFig. 4a,b,cappearstodifferslightlywith destruction reactions, which lead to low fractionation at tem- similarmodelsfoundinGarrod,Weaver&Herbst (2008).This peraturesabove50Kduetoendothermicreactionsbetweense- differenceissolelyduetotheadoptedphysicalconditionmen- lecteddeuteratedionswithH whichreturndeuteratedspecies 2 tioned in section 3.1 in comparison to that in Garrod, Weaver to normal ones. After all, the high temperature limit of OD- &Herbst (2008). FromFig. 4citisclearthatamongthemain PA/PA is 10−4, which is close to the D/H elemental value of icyspecies,methanedisappearsfirstfromthegrainmantledue 10−5. At temperatures under 50 K, the abundance ratio of to its low adsorption energy (1300 K) followed by methanol gaseous OD-PA/PA approaches unity, a very high degree of (adsorptionenergy5530K)andwater(adsorptionenergy5700 fractionation,whiletheiceratiois10−2.Thesehighratiosprob- 8 ablystemfromlargevaluesoftheatomicabundanceratioD/H – since the observed rotational spectra for PA and OD-PA are inthegasandonthegrainsatlowtemperatures. expectedtobecleanerinthesefrequencyrangesgiventhatlow In the warm-up case, the predicted fractional abundance of energy transitions of light molecules fall at much higher fre- PA under hot-core conditions lies at approximately 10−9. Our quencies.Thisiswhywehavelistedonlytheintensetransitions resultsarguestronglyforthepossibilityofdetection,especially forPAalongwithODPAinTable4thatfallinALMABands attemperaturesabove150K.Interestingly, thispredictionim- 1,2and3. pliesthatPAdoesnotoccupythesameregionsasdoesitsiso- CurrentlyALMABand3isinoperationwhileBands1and merpropenal.(Requenaetal.(2008)measureditsabundanceto 2arenotyetavailable. Thestrongesttransitionspointedoutin ∼2.30×10−9withrespecttoH ),butthespeciesisseeninab- Table4forPA(at116.23546GHzand116.23586GHz)arejust 2 sorptionagainstSgrB2(N),andsoismorelikelytobepresent beyondALMABand3andarenotavailableintheALMAtime incolderforegroundgas(Hollisetal.,2004). ForOD-PA,the estimator. Thus, we have focused on the next strongest tran- predictedhot-coreabundanceis≈ 10−12 attemperaturesabove sitionsofBand3at89.39184GHzand89.39212GHz, which 150. Forthecoldcorecase,eventhepeakfractionalabundance alsohaveanoverlapwithBand2. UsingBand3,wecandetect ofgasphasePAwiththefiducialvalueofaissufficientlylow these transitions of PA with approximately 7.5 hours of inte- toruleoutdetectionunlessthehighestvalueoftheparametera gration time by assuming a spectral resolution of 0.5 km/s, a forreactivedesorptionisused. WeconcludethatPAandpos- sensitivity of 17 mK, a signal-to-noise ratio of 6, an angular siblyOD-PAcanmostlikelybedetectedinwarm-upregionsat resolution 3 arcseconds and 40 antennas of 12 meter array in temperaturesabove100K. the ALMA time estimator. But it is not possible to detect any linesforOD-PAusingALMABand3withanyreasonablein- tegrationtime. 4. AstronomicalSpectroscopy 4.2. Vibrationalspectra 4.1. Detectability of PA and OD-PA in the millimeter-wave HereweusetheMP2methodwithanaug-cc-pVTZbasisset regime forthecomputationofthevibrationaltransitionfrequenciesand ToestimatethepossibilityofdetectingPAandOD-PAwith intensitiesofvariousnormalmodesofvibrationofPAandOD- present astronomical facilities, we use the CASSIS radiative PA. The prefix aug stands for diffuse functions and cc-pVTZ transfer model [http://cassis.irap.omp.eu] at LTE with the referstoPetersonandDunningscorrelationconsistentbasisset JPL molecular database [http://spec.jpl.nasa.gov]. Propenal (Peterson & Dunning, 2002) for performing our calculations. (CH CHCHO)hadalreadybeensuccessfullydetectedtowards Theinfluenceofthesolventinvibrationalspectroscopyistaken 2 the high mass star forming region Sagittarius B2(N) with the into account using the Polarizable Continuum Model (PCM) 100-meter Green Bank Telescope operating in the frequency withtheintegralequationformalismvariant(IEFPCM)asade- range18−26GHzbyHollisetal.(2004). Theyhaddetected faultSelf-consistentReactionField(SCRF)method(Tomasiet the2 −1 lineofCH CHCHOwith14mKintensity(≈7σ al.,2005;Miertusetal.,1981). TheIEFPCMmodelisconsid- 11 10 2 detection). From our chemical model, we found that the peak eredtobeaconvenientone,becausethesecondenergyderiva- abundances for PA and OD-PA are respectively 1.2 × 10−9 tiveisavailableforthismodelanditsanalyticformisalsoavail- and 2.14 × 10−12 with respect to H and for propenal, we able. Our calculation was performed in the presence of a sol- 2 use an abundance of 2.3×10−9 with respect to H (Requena vent(watermolecule)byplacingthePAorOD-PAsoluteinthe 2 et al., 2008). As the input parameters, we use the following cavitywithinthereactionfieldofthesolvent. Sincethedielec- parameters, which represent a typical high mass star forming tric constant of ice (85.5) is slightly higher than that of water regionanalogoustoSgrB2(N): (78.5),thevibrationalfrequenciesreportedherearenotexactly thosethatpertainiticebutareclosetothesevalues. ColumndensityofH =1024cm−2, InTable5,wepresentthevibrationalfrequenciesandinten- 2 Excitationtemperature(T )=100K, sities of PA and OD-PA in both the gas and ice phases. We ex FWHM=5km/s, show the band assignments and compare our results with the V =74km/s, existing low resolution experimental results (Sivaraman et al., LSR Sourcesizetotakeintoaccountthebeamdilution=3(cid:48)(cid:48) 2015; Nyquist, 1971). Vibrational detections of molecules in Fromourradiativetransfermodel,theintensityfor2 −1 theinterstellarmediummayincreaseinimportancewiththeim- 11 10 line (18.28065 GHz) of PA is found to be 0.011 mK, which pendinglaunchofthetheJamesWebbSpaceTelescope(JWST) is far below than the detection limit of GBT (3σ≈6mK). Be- althoughthespectroscopicresolutionmaynotbesufficientfor cause of the higher dipole moment components of propenal rotationalsubstructureinthegas. (µ = 3.052 D, µ = 0.630 D and µ = 0 D) compared with We find that the most intense mode of PA in the gas phase a b c PA (µ = 1.037 D, µ = 0.147 D and µ = 0.75 D), intensity appears at 1070.33 cm−1, which corresponds to CO stretch- a b c of PA is much lower compared with propenal. This prompted ing with an integral absorbance coefficient of 1.70×10−17 cm us to suggest the use of high-spatial and high-spectral reso- molecule−1. This peak is shifted downward in the ice water lution observations from the Atacama Large Millimeter/Sub- phase by nearly 17 cm−1 and appears at 1053.89 cm−1 with millimeterArray(ALMA).Inparticular,wecanuselowerfre- anintegralabsorbancecoefficient2.46×10−17 cmmolecule−1. quency bands of ALMA – Bands 1, 2 and 3 (31−116 GHz) Our frequencies can be compared with the experimental ice 9 Table3:PeakabundancesofPA,OD-PAandtheirrelatedspecieswithrespecttoH2. Species Isothermalphase Warm-upphase gasphase icephase gasphase icephase Time(year) Abundance Time(year) Abundance Time(year) T(K) Abundance Time(year) T(K) Abundance PA 3.29×105 1.07×10−12 7.40×105 5.03×10−10 1.10×106 200.0 1.20×10−09 1.01×106 32.2 7.09×10−11 OD-PA 2.89×105 1.32×10−13 5.66×105 1.33×10−11 1.10×100 200.0 2.14×10−12 1.02×106 47.3 7.23×10−12 C3H3 9.36×105 2.52×10−10 1.85×105 4.32×10−13 1.034×106 75.0 1.92×10−9 1.022×106 50.9 3.25×10−14 HC2CHO 8.22×104 6.71×10−10 2.17×105 2.60×10−10 1.028×100 63.5 1.66×10−9 1.01×106 29.2 2.05×10−9 HC2CH2O 5.14×105 5.26×10−15 1.85×105 9.43×10−18 1.054×106 114.1 2.52×10−13 1.01×106 29.2 7.23×10−17 Table4:Calculatedlineparametersformm-wavetransitionsofPropenal,PAandOD-PAwithALMA Species ALMA Frequency Jka(cid:48)K(cid:48)c−Jka(cid:48)(cid:48)K(cid:48)c(cid:48) Eup Aij Intensity Band (GHz) (K) (mK) Propenal Band1 44.4975 505−404 6.4 4.33×10−6 287 Propenal Band2/Band3 89.05935 1046−945 56.44 3.06×105 2320 1047−946 Propenal Band3 115.80341 13410−1249 71.83 7.34×10−5 3920 1349−1248 PA Band1 44.54410 505−404 6.42 5.03×10−7 8.18 PA Band2/Band3 89.39184 1065−964 72.01 2.74×10−6 102 1064−963 89.39212 1056−955 57.23 3.21×10−6 102 1055−954 PA Band3 116.23546 1359−1258 72.68 8.10×10−6 160 1358−1257 116.23586 1386−1285 125.04 5.97×10−6 160 1385−1284 OD-PA Band1 44.88953 514−413 7.68 4.93E-7 0.013 OD-PA Band2/Band3 87.65146 1055−954 53.43 3.43×10−6 0.112 OD-PA Band3 113.96386 1359−1258 68.58 8.59×10−6 0.206 113.96429 1358−1257 Band1=31-45GHz Band2=67-90GHz Band3=84-116GHz phase value of 1030.7 cm−1 (Sivaraman et al., 2015) and the 5. Conclusions vaporphasevalueof1051cm−1 (Nyquist,1971). Otherstrong modes of vibrations are the OH torsion, CCC bending, CH 2 wagging,CHstretching,andOHstretching. InthecaseofOD- Earlier results suggest that PA could be treated as a precur- PA,theCCOstretchingmodeisthemostintensemodeandap- sorofbenzeneformation(Wilsonetal.,2003). Theinterstellar pearsat1068.32cm−1inthegasphaseand1050.98cm−1inthe identification of propenal, which is an isomer of PA, encour- icephasewithintegralabsorbancecoefficientsof1.61×10−17 agedustocheckthepossibilityofdetectingPAandOD-PAin and2.49×10−17cmmolecule−1respectively. the ISM. Our quantum chemical calculations, combined with Table5clearlyshowsthedifferencesbetweenspectroscopic existingrotationaltransitionfrequenciesforPAandOD-PA,al- lowustoachievethefollowingmajorresultsforthispaper: parameters computed for propargyl alcohol between the two phases. Thesedifferencescanbeexplainedduetoelectrostatic • A complete reaction network has been prepared for the for- effects,whichareoftenmuchlessimportantforspeciesplaced mationanddestructionofPA.Variousformationroutesaredis- inasolventsuchaswaterwithhighdielectricconstantthanthey cussed and identified based on calculated exothermicities and areinthegasphasebutthesechangeswillbesignificantifboth endothermicities. thesolventandthemoleculearepolar. • The predicted abundances of PA and OD-PA in the gas and ice phases yield some information for both cold and warm in- The two sets of results for PA are in good agreement as terstellar sources on the possibility of observing this molecule regards frequencies with the existing experimental results, as intheISM,especiallyinhot-coreregionssuchasSgrB2(N). shown above for the single case of the strong CO stretching. Differencesbetweentheresultscouldbeattributedtomultiple • A simple radiative transfer model has been employed to reasons,oneofwhichconcernstheGaussian09program,with discuss the possibility of detecting PA in the ISM through whichweareunabletoconsiderthemixingofdifferentmodes millimeter-wave observations. The most likely environments under harmonic oscillator approximations. Another reason is for detection of PA are hot-core regions with a temperature thatinourquantumchemicalsimulationoficespectra,wecon- above100K. sidered a single propargyl alcohol molecule inside a spherical • The frequencies and intensities of vibrational transitions of cavityandimmerseditinacontinuousmediumwithadielec- PAandOD-PAbothinthegasandinamimeticforaniceenvi- tric constant, while in the experiment propargyl alcohol was ronmentarecalculatedandcomparedwithexistinglowresolu- deposited atop the ice, which could lead to the formation of tionexperimentalfrequencies.Thecalculatedintensitiesshould clusters. beparticularlyuseful. 10