Far-Infrared detection of H2D+ toward Sgr B2

0.22 MB·English

by  J. Cernicharo

Recieved 2006December14;accepted 2007January18
PreprinttypesetusingLATEXstyleemulateapjv.10/09/06
FAR-INFRARED DETECTION OF H D+ TOWARD SGR B21
2
J. Cernicharo
DepartamentodeAstrof´ısicaMoleculareInfrarroja,InstitutodeEstructuradelaMateria,CSIC,C/Serrano121,28006,Madrid. Spain
and
E. Polehampton2
SpaceScience&Technology Department,RutherfordAppletonLaboratory,Chilton,Didcot,Oxfordshire,OX110QX,UK
and
J.R. Goicoechea
LERMA-LRA,UMR811,CNRS,ObservatoiredeParisetE´coleNormaleSup´erieure,24rueLhomond, 75231ParisCedex05,France
7 Recieved2006 December14; accepted 2007 January 18
0
0 Abstract
2
We report on the first far–IR detection of H D+, using the Infrared Space Observatory, in the line
2
n ofsighttowardSgrB2 in the galacticcenter. The transitionatλ=126.853µmconnecting the ground
a levelofo-H D+,1 ,withthethe2 levelat113K,isobservedinabsorptionagainstthecontinuum
2 1,1 1,2
J emissionofthecolddustofthesource. Thelineisbroad,withatotalabsorptioncovering350kms−1,

9 i.e., similar to that observed in the fundamental transitions of H O, OH and CH at ∼179, 119 and
2
1 149 µm respectively. For the physical conditions of the different absorbing clouds the H D+ column
densityrangesfrom2to5×1013cm−2,i.e.,nearanorderofmagnitudebelowtheupperlim2itsobtained
1 from ground based submillimeter telescopes. The derived H D+ abundance is of a few 10−10, which
v 2
agrees with chemical models predictions for a gas at a kinetic temperature of ≃20 K.
9
5 Subject headings: Astrochemistry — molecular processes — line : identification — ISM : molecules
5 — ISM : individual (Sgr B2) — infrared : ISM
1
0
7 1. INTRODUCTION spectrum. Thesespeciesplayacrucialroleinthedeuter-
0 The H+ molecular ion is a key molecule for the gas ation enhancement of many molecular species in clouds
/ 3 with kinetic temperatures below ≃30 K (see, e.g., Ger-
h phase chemistry in interstellar clouds (Herbst & Kem-
lich et al., 2002 and references therein). In particular,
p plerer1973;Watson1973). Producedbythefastreaction
o- of H2 with H+2, it reacts with almost all neutral atoms tchaellyHi2nDcr+e/aHse+3aabbouvendtahneceelermateinotaisleDx/pHecrteadtiotoasdrtahmeagtais-
and molecules, and thus it is the precursor of a large
r coolsdown. The firstcalculationofthe H D+ rotational
t number of complex neutral and ionic molecular species. 2
s spectrum was reported by Dalgarno et al. (1973), who
Initially suggested to be present in molecular clouds by
a
suggested that it could be detectable in dense and cold
: Martin et al. (1961), there is a long history behind the
v clouds. TheH D+ 1 -1 transitionhasbeenobserved
searchforthiscrucialmolecule. Duetoitslackofperma- 2 1,0 1,1
Xi nentdipolemoment,H+mustbeobservedinthenear–IR in the laboratory at 372.42134 GHz (Bogey et al. 1984;
3 Warner et al. 1984). This precise measurement opened
through its ro-vibrational transitions (Oka 1980). The
r the quest for the detection of H D+ in dense molecular
a first searches only provided upper limits to the H+ col- 2
3 clouds. Initialsearchesfortheo-H D+ 1 -1 line,and
umn density (Geballe & Oka 1989; Black et al., 1990). 2 1,0 1,1
alsothep-H D+ 1 -0 (at1370GHz),performedwith
However, H+ has been finally detected toward GL2136 2 0,1 0,0
3 the Kuiper Airborne Observatory (Phillips et al., 1985;
andtheW33Aclouds(Geballe&Oka1996). Inaddition,
Paganietal.,1992a;Boreiko&Betz,1993),andwiththe
H+ hasalsobeendetectedtowardthe diffuse interstellar
3 Caltech Submillimeter Observatory (van Dishoeck et al.,
medium (ISM) by McCall et al. (1998), and more re- 1992) were, however, unsuccessful. Fortunately, H D+
2
cently in the line of sight toward several galactic center
hasfinallybeendetectedtowardtheyounglow-masspro-
clouds (Oka et al. 2005). Unfortunately, these obser-
tostarNGC1333-IRAS4A(Starketal. 1999),andinsev-
vations only provide information on the H+3 abundance eral dark clouds (Caselli et al. 2003; Vastel et al., 2004,
in the most diffuse and warm external layers of molecu- 2006a,b; Hogerheijde et al., 2006; Harju et al., 2006),
lar clouds, and are limited to lines of sight with bright andtentativelyinprotoplanetarydisks(Ceccarellietal.,
near-IR background sources. 2004;see also Guilloteau et al., 2006).
A different approach to study H+3 in denser regions Observational evidence for the crucial role of H2D+,
obscured by dust extinction is to observe the asymmet- HD+ (Vasteletal.,2004),andevenofD+,inthedeuter-
ricalspeciesH2D+ andHD+2,whichdohavearotational atio2n processes of the ISM have been w3idely probed by
thedetectionofdoubleandtripledeuteratedspeciessuch
1Basedon observations withISO,an ESAproject withinstru-
as D CO (Turner, 1990; Ceccarelli et al., 1998), NHD
mentsfundedbyESAMemberStates(especiallythePIcountries: 2 2
(Roueff et al., 2000; Loinard et al., 2001; Gerin et al.
France, Germany, the Netherlands and the United Kingdom)and
withparticipationofISASandNASA. 2006), ND3 (van der Tak et al., 2002; Lis et al., 2002),
2DepartmentofPhysics,UniversityofLethbridge,4401Univer- CHD OH (Parise et al., 2002), CD OH (Parise et al.,
2 3
sityDrive,Lethbridge, Alberta,T1J1B1,Canada
2004), D CS (Marcelino et al., 2005).
2
2 Cernicharo, Polehampton & Goicoechea
Ground-basedobservationsofH D+ andHD+ haveal-
2 2
ways been performed toward very dense and cold cores
as the high Einstein coefficients of the involved transi-
tionsrequirehighvolumedensitiestoproducesignificant
emission. Therefore, it is difficult to estimate the H D+
2
abundance in lower density regions through submillime-
ter observations. In this Letter, we report on the first
detection of H D+ in the far-IR spectrum of Sgr B2 ob-
2
served with the Infrared Space Observatory (ISO). The
observedtransitionconnectsthe groundstate,1 , with
1,1
the2 levelat113.4K,andhasawavelengthof126.853
1,2
µm. ThisdetectionopensthepossibilitytodetectH D+
2
incloudswithmoderatevolumedensityinwhichthe1 -
0,1
1 transition at 372 GHz will be subthermally excited
1,1
and its intensity too weak to be detected with currently
available radio telescopes. It also opens the quest to
detect H D+ toward far-IR luminous galaxies and high
2
redshift objects.
2. OBSERVATIONS
The data presented in this paper are based on the fi-
nal calibration analysis of the high spectral resolution
line survey of Sgr B2 obtained with the Long Wave-
length Fabry-P´erot Spectrometer (LWS/FP) on board
ISO in its L03 mode (Polehampton 2002; Polehamp-
ton et al., 2007). The data for NH , which are neces-
2
sary to interpret the H D+ absorption reported in this
2
paper, were previously presented by Goicoechea et al.
(2004). The spectral resolution of the LWS/FP spec-
trometer is ≃ 35 km s−1. The telescope was pointed to- Fig. 1.—ObservedISOspectrumat126.853µmtowardSgrB2.
ward α(2000)=17h 47m 21.75s; δ(2000)=-28o 23’ 14.1”. Theupperpanel showsthedataandtheexpected contribution of
TheH D+ andNH dataarepresentedinFigs. 1and2. p-NH2 (see Figure 2). The middle panel shows the o-H2D+ line
2 2 profileaftersubtractingthecontributionfromp-NH2. Finally,the
The procedure to reduce the LWS/FP data have been
lowerpanelshowstheabsorptionofCHforcomparisonpurposes.
described in detail by Polehampton et al. (2007). This
Only a few lines remain unidentified in the ISO far-IR
provides a significant improvement over the previous re-
line survey of SgrB2. The feature discussed in this pa-
duction of Goicoechea et al. (2004), who did not apply
per must correspond to a line arising from a low energy
any interactive processing to the standard pipeline data
level or to a blend of lines. However, molecules heavier
products except for the removal of glitches, averaging
than H D+ could have other lines in the far-IR domain.
of individual scans and removal of baseline polynomials. 2
We have considered the possibility of b-type transitions
The improvements applied in the full spectralsurvey re-
fromslightlyasymmetricrotorslikeHNCOandHOCO+.
duction include determination of accurate detector dark
Both have one transition at similar wavelengths. How-
currents (including stray light), improved instrumental
ever, the line of HNCO involves two high energy levels
response calibration (including a correction to account
and can be easily ruled out. HOCO+ has several other
for adjacent FP orders) and interactive shifting of indi-
lines in the 120-127 µm range involving low energy lev-
vidual miniscans (to correct for uncertainty in the angle
els. None of which are detected in the far-IRline survey
of the LWS grating). The signal-to-noise ratio was also
of Sgr B2 (Polehampton et al., 2007). Hence, we are
enhancedby including everyavailableobservationin the
confident that the broadfeature detected at 126.853µm
survey (in this case 2 independent observations). The
corresponds to H D+.
broad absorption feature shown in Figure 1 was previ- 2
The intensity scale in figures 1 and 2 was obtained
ously assigned only to p-NH2 (Goicoechea et al., 2004;
by dividing the observed flux by a fitted continuum as
transition 2 -1 J=5/2-3/2). However, the final re-
2,1 1,0
described by Polehampton et al. (2007). Hence, it is
ductionoftheLWS/FPsurveytowardSgrB2showsthat
directly related with the line opacity convolvedwith the
the broadening is real. The feature appears in 2 inde-
instrumental spectral profile of the LWS/FP spectrome-
pendent observations : ISO TDT number 50601112,ob-
ter. The velocity scale in Figure 1 has been obtained as-
servedon1997April5withLWSdetectorLW2;andISO
sumingλ =126.853µm(2363.325GHz),i.e.,thewave-
TDT number 50800515, observed on 1997 April 7 with rest
lengthoftheo-H D+2 -1 transition(Amano&Hirao,
LWSdetectorLW3. Thefactthatthefeatureappearson 2 12 11
2005). The expected erroron the frequency of this tran-
2 separate detectors rules out any spurious instrumental
sitionis≃5MHz,i.e.,avelocityuncertainty<2kms−1
broadening of the p-NH2 line.
(3σ),whichismuchlowerthanthespectralresolutionof
Taking into account the latest data reduction, the
broad feature detected at ∼126 µm cannot be due to the LWS/FP spectrometer. The p-NH2 22,1-11,0 J=5/2-
p-NH alone. In fact, NH far-IR lines are narrow and 3/2 line is separated by ≃-123 km s−1 with respect to
2 2
appear only at the velocity of Sgr B2 (see, van Dishoeck the 212-111 line of o-H2D+.
et al., 1993, for the detection of NH in the submm).
2
Far-Infrared detection of H D+ 3
2
3. RESULTSANDDISCUSSION
The upper panel of Figure 2 shows the three spin–
rotational components of the o-NH 2 -1 transition
2 2,0 1,1
at ∼117 µm (Goicoechea et al. 2004). In order to com-
pute the expected contribution of the p-NH 2 -1
2 2,1 1,0
triplet at ≃126 µm we have modeled the NH absorp-
2
tion assuming a rotational temperature of 20 K, an in-
trinsic linewidth of 10 km s−1, a total NH column den-
2
sity of 1.2×1015 cm−2 (Goicoechea et al. 2004), and
an ortho/para ratio of 3. NH wavelengths and line
2
strengths are from the Cologne Database for Molecu-
lar Spectroscopy (Mu¨ller et al. 2001). The NH hyper-
2
fine structure is unresolved by the LWS/FP. The syn-
thetic spectrum, convolved with the LWS/FP spectral Fig. 2.—RestISO/LWS/FPspectrumofSgrB2around117and
resolution,is shownin Figure 2. The expected contribu- 1th2e6sµomur,cceo(r6re2c.7tedkmbys−th1e).avLeinraegsedduveetloocoit-yNoHf2fa(ru–pIpRerlinpeasnetlo)waanrdd
tion from p-NH2 to the absorption feature is shown as p-NH2 plus o-H2D+ (lower panel) are shown. Theexpected NH2
a continuous line in the upper panel of Figure 1. The absorptionisshown bycontinuous lines inboth panels (see text).
middle panel of this Figure shows the same data with An NH2 ortho/para ratio of 3 has been assumed. Note that the
the expected contributionfromp-NH removed. The re- NH2 linesshowabsorptiononlyatthevelocityofSgrB2.
2
maining absorption feature is broad and has several ab- at372GHzwouldbe0.024and0.006forT =3and30K
r
sorptionvelocitypeaks,whicharesimilar tothose found respectively.
in unsaturated ground–state transitions of light species The interpretation of the absorption produced by o-
such as the (2,3/2)-(1,1/2)line of CH (see bottom panel H D+ in Sgr B2 requires knowledge of the rotational
2
of Figure 1). Therefore, in the following, we assign this temperatureofthetransition. Thecloudhasalargedust
unidentified broad feature to the o-H2D+ 21,2-11,1 line. opacity in the far-IR, τd(100µm)=3.5 (Cernicharo et al.
H2D+ isanasymmetricmoleculewithµa ≃0.6D(Dal- 1997, 2006; Goicoechea et al. 2004). Hence, infrared
garno et al., 1973). It has two different species, ortho photons cloud play an important role in the pumping of
(Ka=odd), and para (Ka=even), with a spin weight ra- themolecularlevelsofo-H D+. Inaddition,the fraction
2
tio of 3:1. The three lowest energy levels, 11,1, 11,0 and of the cloud depth that is ”seen” at this wavelengthcor-
21,2, which are at 0, 17.9 and 113.4 K (see insert in the respondsonlytoitsexternallayers,whereτd(127µm)≃1,
middle panel of Figure 1), are enough to compute the i.e., N(H )≃3×1023 cm−2, rather than the ≥1024 cm−2
2
partition function (Fpar) at low temperatures. Hence, correspondingto the total column density along the line
F (T ) ≃ 3(1+e−17.8/Tr) is a good approximation for of sight (τ (127 µm)≃3). In order to estimate the ra-
par r d
T < 30 K, where T is the rotational temperature. For diative and collisional effects on the excitation of the
r r
example, F (3 K)= 3.004 (all molecules in the ground 2 -1 transition, we have assumed de-excitation col-
par 1,2 1,1
state), while Fpar(30 K)=4.8. Although the 11,0 level lisional rates σ(11,0-11,1)=σ(21,2-11,0)= 10−10 cm3 s−1,
startstobepopulatedinthiscase,the21,2oneisstillun- and σ(21,2-11,1)=5×10−11 cm3 s−1. For a gas at 30
populated. Consequently, if there is a strong continuum K purely excited by collisions with N(o-H D+)= 1013
2
source behind or inside the source, we could expect to cm−2, and ∆v=10 km s−1, the rotational temperature
seethe21,2-11,1 transitionofo-H2D+ inabsorption. The varies from 9 to 12 K for H densities of 105 and 106
2
same behavior is observed in CH toward Sgr B2, where cm−3 respectively. Reducing the collisional rates by a
only the ground-state line is observed with the line pro-
factor 10 produces rotationaltemperatures of 6 and 8 K
file shown in the bottom panel of Figure 1 (Goicoechea
for the same volume densities.
et al. 2004; Polehampton et al. 2007).
The role of radiative pumping by far–IR dust pho-
The Einstein coefficient of the o-H D+ 2 -1 tran-
2 1,2 1,1 tons can be estimated assuming that the gas we observe
sition is large, 1.7×10−2 s−1, which implies that the ro- in absorption, τ (127µm)=1, surrounds a region with
d
tational temperature of the line will be extremely low τ (127µm)=2.0 and T =30 K. If the radius of the ab-
d d
in most relevant cases (or close to the dust temper- sorbingshellistwicethatoftheregionemittingthebulk
ature if continuum emission is optically thick), even oftheinfraredemission,R /R =2,weobtainanex-
abs cont
when the 11,0-11,1 transition is already collisionally ex- citation temperature of 17 K, with a little dependency
cited (its Einstein coefficient is two orders of magni- on the assumed collisional rates. For R /R =3 and
abs cont
tude lower). Therefore, in order to derive a column 1.5, the rotational temperature is 15 and 20 K respec-
density from absorption measurements of the 21,2-11,1 tively. In the following we assume Tr=10-20 K, and
line, we can assume that all molecules are in the 11,1 we use these values to derive upper and lower limits
ground level, and that the transition has a low ro- to the opacity of the H D+ 2 -1 line. Assuming a
2 1,2 1,1
tational temperature in absence of infrared pumping. linewidth of 10 km s−1 we obtain τ(2 -1 )=0.18 and
1,2 1,1
The opacity of the 2 -1 line can be written as
1,2 1,1 0.36 for T =10 and 20 K respectively. The correspond-
r
τ(Tr) ≃ 0.067(N/1013)(10/∆v)/Fpar(Tr), where ∆v is ing column densities are ≃ 9×1013 and 2.3×1014 cm−2,
the linewidth at half intensity in km s−1, and N is the and the abundance of o-H D+, for a total gas column
2
columndensityofo-H2D+ in cm−2. IfN(o-H2D+)=1013 density of 3×1023 cm−2, is ≃ 3×10−10 for Tr=10 K
cm−2 and∆v=10kms−1,τ(21,2-11,1)wouldbe0.024for and 9×10−10 for Tr=20 K. The latter can be ruled out
Tr=3 K and 0.014 for Tr=30 K respectively. For com- from the upper limit obtained by Pagani et al (1992a)
parison purposes, the opacity of the 11,0-11,1 transition for the emission of the 11,0-11,1 line at 372 GHz, N(o-
4 Cernicharo, Polehampton & Goicoechea
H D+)<1.1×1014 cm−2. Hence, a low rotational tem- servedinthissource,nolineshavebeenassignedtoNH
2 2
perature, T ≃10 K, and an abundance of 3×10−10 are so far. Hence, the broad absorption feature assigned to
r
the best fit to the o-H D+ absorption at the velocity of NH by Gonz´alez-Alfonsoet al.,couldalsocontainsome
2 3
Sgr B2. contribution from o-H D+. The detection of H D+ pre-
2 2
For the other absorbing clouds which are not directly sented in this work,shows that far–IR ortho–and para–
associatedtoSgrB2,itisreasonabletoassumeradiative H D+ linescanbeusedtotracethedeuterationfraction-
2
thermalization with the cosmic background, T ≃3 K, ationofH+ ingalaxies(seealsoCeccarelliandDominik,
r 3
i.e.,nosignificantcollisionalexcitationorinfraredpump- 2006). Therefore, far–IR observations with the PACS
ing. Assuming an intrinsic line width of 10 km s−1, the instrument on board Herschel satellite (or in the future
observedabsorptionfeatures at-100,10 and120kms−1 with SOFIA), will be an important tool to understand
haveopacities of of0.1,0.15 and0.06and columndensi- thecoldISMchemistryindistantgalaxieswheretheD/H
ties of ≃ 4.5, 6.7, and 2.7 1013 cm−2 respectively. These ratio will be different and the gas temperature affected
clouds have low volume densities and total gas column by the local cosmic background temperature.
densities of a few 1022-1023 cm−2. The associated o- Finally,theopacityofthe2 -1 transitionforadark
1,2 1,1
H2D+ abundances are of a few 10−10–10−9. cloudwith∆v=0.5kms−1 (seeVasteletal.,2004),N(o-
The derived abundance of o-H2D+ in Sgr B2 corre- H2D+)=1012 cm−2, and Tr= 3 K is ≃0.05. While the
sponds to the predictions of chemical models for TK expected emission in the 11,0-11,1 transition at 372 GHz
<20 K (see Pagani et al., 1992b; Walmsley et al., 2004; will be negligible, the expected absorption at 126.853
Floweret al., 2006). These models predict that the para µm will be 5%, i.e., easy to detect with the expected
species could be more abundant than the ortho one for sensitivities of the high spectral resolution instruments
that temperature and the expected density of the gas (a planned for SOFIA. Dark clouds (see, e.g., Nisini et
few 104-105 cm−3). Hence, the fundamental line of p- al.,1999),prestellarcores,andprotoplanetarydisks(see
H2D+ could produce significant absorption at 218.8 µm Wyatt et al., 2003) have enough continuum emission in
in the direction of Sgr B2. The abundances observed the far-IR, hence, the 2 -1 transition can be used
1,2 1,1
toward the other clouds in the line of sight of Sgr B2 as a key observing tool to trace column densities of o-
indicate lowerkinetic temperatures,TK ≃10K.Inthese H2D+ larger than a few 1011 cm−2, and to understand
clouds the predicted ortho/para abundance ratio favors thedeuteriumfractionationatlargespatialscalesincold
the ortho species. Due the high dependency of the or- darkclouds. Observationsof the fundamental lines of p-
tho/para ratio with the gas temperature, simultaneous HD+at691.6605GHzando-HD+203.03µmwillprovide
2 2
observation of both the 126.853 µm (o-H2D+) and the additional constraints on these deuteration processes in
218.8 µm (p-H2D+) lines will provide a direct measure- molecular clouds and protoplanetary disks.
ment of the efficiency of the reactions of H D+ with H
2 2
and of the conversion of ortho–to–para H D+ and vice-
2
versa, i.e., to check chemical models.
DuetothepoorspectralandangularresolutionofISO, We wouldliketo thank Cecilia Ceccarelli,SerenaViti,
and due to the limited sensitivity of the LWS/FP in- and our referee for useful comments and suggestions.
strument, the o–H D+ line at 126.853µm could only be We also thank the spanish MEC for funding support
2
expected in sources with a large column density of cold through grant ESP2004-665, AYA2003-2785, and “Co-
gas with broad linewidths in the line of sight. As an munidaddeMadrid”GovernmentunderPRICITproject
example, a broad absorption feature has been also re- S-0505/ESP-0237 (ASTROCAM). This study is sup-
ported in the bright IR galaxy Arp220 at 127 µm by ported in part by the European Community’s human
Gonz´alez-Alfonso et al. (2004) using the LWS grating potential Programmeunder contractMCRTN-CT-2004-
spectrometer (spectral resolution more than 20 times 51230, Molecular Universe. JRG was supported by an
lower than the LWS/FP). These authors assigned it to individual Marie Curie fellowship under contractMEIF-
NH . While several lines of NH and NH have been ob- CT-2005-5153340.
3 3
REFERENCES
AmanoT.,HiraoT.,2005, J.Mol.Spectrosc.,233,7 Gerin,M.,Lis,D.C.,Philipp,S.,Gusten,R.,Roueff,E.&Reveret,
Black,J.H.,vanDishoeck,E.F.,Willner,S.P.,Woods,R.C.,1990, V.2006,A&A,454,L63.
ApJ,358,459 Gerlich, D., Herbsst, E., Roueff, E., 2002, Planetary and Space
BogeyM.,DemuynckC.,DenisM.,etal.,1984,A.&A.,137,L15 Science, 50,1275
Boreiko,R.T.,Betz, L.,1993, ApJ,405,L39 Goicoechea,J.R.,Rodriguez-Fernandez,N.J.,Cernicharo,J.,2004,
Caselli,P.,vanderTak,F.F.S.,Ceccarelli,C.,Bacmann,A.,2003, ApJ,600,214
A.&A.,403,L37 Gonza´lez-Alfonso,E.etal.,2004,ApJ,613,247
Ceccarelli, C., Castets, A., Loinard, L., et al., 1998, A.&A., 338, Guilloteau,S.,Pi´etu,V.,Dutrey,A.,Gu´elin,M.,2006,A.&A.,448,
L43 L5
Ceccarelli,C.,Dominik,C.,Lefloch,B.,etal.,2004,ApJ,607,L51 Harju,J.,Haikala,L.K.,Lehtinen,K.,etal.,2006,A.&A.,454,L55
Ceccarelli,C.,Dominik,C.,2006,ApJ,640,L131 Herbst,E.,Kemplerer,W.,1973, ApJ,185,505
Cernicharo,J.,Lim,T.,Cox,P.,etal.,1997,A.&A.,323,L25 Hogerheide, M.R., Caselli, P., Emprechtinger, M., et al. (2006),
Cernicharo,J.,Goicoechea, J.R.,Pardo,J.R.,Asensio-Ramos,A., A.&A.,454,L59
2006, ApJ,642,940 Lis,D.C.,Roueff,E.,Gerin,M.etal.,2002,ApJ,571,L55
Dalgarno,A.,Herbst,E.,Novick,S.,KemplererW.,1973ApJ,183, Loinard, L., Castets, A., Ceccarelli, C., Tielens, A.G.G.M., 2001,
L131 ApJ,552,L163
Flower, D., Pineau des Foˆrets, G., Walmsley, C.M.,2006, A.&A., Marcelino, N., Cernicharo, J., Roueff, E., et al., 2005, ApJ, 620,
449,621 308
Geballe,T.R.,Oka,T.,1989, ApJ,342,855 Martin,D.W.,McDaniel,E.W.,Meeks,M.L.,1961,ApJ,134,1012
Geballe,T.R.,Oka,T.,1996, Nature,384,344
Far-Infrared detection of H D+ 5
2
McCall B.J., Geballe, T.R.,Hinkle, K.H.,Oka, T., 1998, Science, Turner,B.,1990, ApJ,362,L29
279,1910 vanderTak,F.F.S.,Schilke,P.,Muller,H.S.P.,etal.,A.&A.,388,
Mu¨ller, H. S. P., Thorwirth, S., Roth, D. A. & Winnewisser, G. L53
2001, 370,L49. van Dishoeck, E.F., Phillips, T.G., Keene, J., Blake, G.A., 1992,
Nisini,B.,Benedettini, M.,Giannini,T.,etal.,1999,A.&A.,343, A.&A.,261,L13
266 van Dishoeck, E.F.,Jansen, D.J., Schilke P., Phillips,T.G., 1993,
Oka,T.,Phys.Rev.Lett.,1980, 45,531. ApJ,416,L83
Oka, T., Geballe, T.R., Goto, M., Usuda, T., McCall, BJ. 2005, Vastel,C.,Phillips,T.G.,Yoshida,H.,2004, ApJ,606,L127
ApJ,632,882. Vastel,C.,Caselli,P.,Ceccarelli,C.,etal.,2006,ApJ,645,1198
Pagani, L., Wannier, P.G., Frerking, M.A., et al., 1992a, A.&A., Vastel,C.,Phillips,T.,Caselli,P.,etal.,2006,Proceedingsofthe
258,472 Royal Society meeting Physics, Chemistry, and Astronomy of
Pagani,L.,Salez,M.,Wannier,P.G.,1992b, A.&A.,258,479 H+
3
Parise, B., Ceccarelli, C., Tielens, A.G.G.M., et al., A.&A., 393, Walmsley,C.M.,Flower,D.R.,PineaudesFoˆrets,G.,2004,A.&A.,
L49 418,1035
Parise,B.,Castets A.,Herbst,E.,etal.,A.&A.,416,159 Warner,H.E.,Conner,W.T.,Petrmichl,R.H.,Woods,J.,1984,J.
Polehampton, E.T.,2002,Ph.D.thesis,LinacreCollege,Oxford Chem.Phys.,81,2514
Polehampton,E.T.,Baluteau,J.P.,Swinyard,B.,Goicoechea,J.R. Watson,W.D.,1973,ApJ,183,L17
etal.2007,MNRAS,submitted. Wyatt,M.C.,Dent,W.R.F.,Greaves,J.S.,2003,MNRAS,342,876
Roueff,E.,Tin´e,S.,Coudert,L.H.,etal.,A.&A.,354,L63
Stark,R.,vandetak,F.F.S.,vanDishoeck,E.F.,1999,ApJ,521,
L67