Deuterated formaldehyde in rho Ophiuchi A
aa r X i v : . [ a s t r o - ph . GA ] N ov Astronomy&Astrophysicsmanuscript no. 15012 c (cid:13)
ESO 2018August 7, 2018
Deuterated formaldehyde in ρ Ophiuchi A ⋆ P. Bergman , B. Parise , R. Liseau , and B. Larsson Onsala Space Observatory, Chalmers University of Technology, SE-439 92 Onsala, Sweden e-mail: [email protected] Max Planck Institut f¨ur Radioastronomie, Auf dem H¨ugel 69, 53121 Bonn, Germany Department of Earth and Space Sciences, Chalmers University of Technology, SE-439 92 Onsala, Sweden Department of Astronomy, Stockholm University, AlbaNova, SE-10691 Stockholm, SwedenReceived ? ; accepted ?
ABSTRACT
Context.
Formaldehyde is an organic molecule that is abundant in the interstellar medium. High deuterium fractionation is a commonfeature in low-mass star-forming regions. Observing several isotopologues of molecules is an excellent tool for understanding theformation paths of the molecules.
Aims.
We seek an understanding of how the various deuterated isotopologues of formaldehyde are formed in the dense regions of low-mass star formation. More specifically, we adress the question of how the very high deuteration levels (several orders of magnitudeabove the cosmic D / H ratio) can occur using H CO data of the nearby ρ Oph A molecular cloud.
Methods.
From mapping observations of H CO, HDCO, and D CO, we have determined how the degree of deuterium fractionationchanges over the central 3 ′ × ′ region of ρ Oph A. The multi-transition data of the various H CO isotopologues, as well as fromother molecules (e.g., CH OH and N D + ) present in the observed bands, were analysed using both the standard type rotation diagramanalysis and, in selected cases, a more elaborate method of solving the radiative transfer for optically thick emission. In addition tomolecular column densities, the analysis also estimates the kinetic temperature and H density. Results.
Toward the SM1 core in ρ Oph A, the H CO deuterium fractionation is very high. In fact, the observed D CO / HDCO ratio is1 . ± .
19, while the HDCO / H CO ratio is 0 . ± . CO / HDCO abundanceratio is observed to be greater than 1. The kinetic temperature is in the range 20-30 K in the cores of ρ Oph A, and the H densityis (6 − × cm − . We estimate that the total H column density toward the deuterium peak is (1 − × cm − . As depletedgas-phase chemistry is not adequate, we suggest that grain chemistry, possibly due to abstraction and exchange reactions along thereaction chain H CO → HDCO → D CO, is at work to produce the very high deuterium levels observed.
Key words. astrochemistry – ISM: abundances – ISM: clouds – ISM: individual objects: ρ Oph A – ISM: molecules
1. Introduction
The study of deuterated molecules in the interstellar medium(ISM) has been intensified over the past decade ever sinceit was discovered that singly and multiply deuterated species(like the D-containing versions of H CO, CH OH, NH ) oc-curred at abundances that were orders of magnitude higher thanwould be expected from the local ISM D / H ratio of about 1 . × − (Linsky 2003). The regions that show these elevated D-abundances are mainly associated with low-mass protostars. Forinstance, around IRAS16293 − CO) which previously had only been seen inthe Orion KL ridge (Turner 1990). Later, Loinard et al. (2002)extended the study to a larger set of protostars arriving atD CO / H CO abundance ratios as high as 0.05-0.4. Likewise,high degrees of deuterium fractionation in CH OH (Parise et al.2002, 2004, 2006) and NH (Roue ff et al. 2000; Loinard et al.2001; Lis et al. 2002) were subsequently discovered.These studies are all related to single telescope pointings.To our knowledge, only one e ff ort to delineate the D CO / H COabundance variation within a single source (IRAS16293 − CO / H CO ra- ⋆ Based on observations with the Atacama Pathfinder EXperiment(APEX) telescope. APEX is a collaboration between the Max-Planck-Institut f¨ur Radioastronomie, the European Southern Observatory, andthe Onsala Space Observatory tio peaks some 20 ′′ away from the protostar. Further out, thedegree of deuteration seems to be lower in the quiscent gas(Ceccarelli et al. 2002). In the case of deuterated ammonia, themapping study by Roue ff et al. (2005) revealed that the deu-terium peak is typically o ff set from the positions of the embed-ded protostars. These authors argued that the observed scenariocould be explained by the formation of deuterium-enriched icesduring the cold pre-collapse phase. At a later stage, a newlyformed protostar evaporates the ices.In their extensive study of deuterated H CO and CH OH,Parise et al. (2006) concluded that formation of CH OH ongrain surfaces was a likely explanation for the high degrees ofdeuterium fractionation seen in the various isotopologues. ForH CO, the situation was less clear and a formation path in-volving gas-phase reactions could not be ruled out. In fact, forthe warmer Orion Bar PDR region, Parise et al. (2009) very re-cently advocated that gas-phase chemistry is entirely responsi-ble for the deuterium fractionation seen for some singly deuter-ated species (including HDCO). It should be noted that singly-deuterated species have been detected in dark and translucentclouds (Turner 2001) where the deuteration occurs only in thedense parts.In interstellar clouds, deuterium is mostly molecular in theform of HD. Deuterium can be transferred from this main molec-ular reservoir to other molecules via exothermic reactions withthe molecular ions H + , CH + , and C H + . These basic reactionsare followed by e ffi cient ion-neutral reactions and, in some
1. Bergman et al.: Deuterated formaldehyde in ρ Ophiuchi A cases, by reactions on grain surfaces. The exothermicity of thereactions involving these three molecular ions are in the range230-550 K (Gerlich et al. 2002; Asvany et al. 2004; Herbst et al.1987). As a result, deuterium fractionation is e ffi cient in coldenvironments. It is also well established that a high degree ofdepletion on the grains of CO, O, and other heavy species,which would otherwise destroy e ffi ciently H + (and its deuter-ated analogues), is another important condition for deuteriumfractionations taking place. Correlations between CO depletionand fractionation have been observed towards prestellar cores by,e.g., Bacmann et al. (2003) and Crapsi et al. (2005), and in theenvelope of Class 0 protostars by Emprechtinger et al. (2009).Because of these two conditions (low temperature and high COdepletion), deuterium fractionation is particularly e ffi cient dur-ing the early stages of star formation. The surface reactionson cold grains important for deuterium fractionation of H COhave recently been investigated through laboratory experiments(Hidaka et al. 2009).The aforementioned low-mass protostar IRAS16293 − ρ Ophiuchi cloud complex.To the west in the same complex, more than 1 degree away, liesthe ρ Oph A cloud (Loren et al. 1990) at a distance of about 120pc (Lombardi et al. 2008; Loinard et al. 2008; Snow et al. 2008).This cloud core is well-studied by infrared, submillimeter,and millimeter continuum observations (Ward-Thompson et al.1989; Andr´e et al. 1993; Motte et al. 1998). It hosts a well-collimated molecular outflow (Andr´e et al. 1990) and its driv-ing source VLA 1623 (Andr´e et al. 1993). Moreover, it was inthe direction of this cloud core that Larsson et al. (2007), us-ing the Odin satellite, detected the 119 GHz line from molecularoxygen as well as the ammonia ground state line at 572 GHz(Liseau et al. 2003). More recently, while searching for the 234GHz O O line Liseau et al. (2010) detected several lines due toD CO toward the millimeter continuum peaks. This study alsoincludes C O(3-2) mapping observations. Earlier, Loinard et al.(2002) reported a high D CO / H CO ratio toward the VLA 1623source. The existence of D CO in several positions of this 2 ′ × ′ cloud core formed the incentive of the present study as an ex-cellent source to delineate the distribution of H CO, HDCO, andD CO as we know the distribution of the dust (Motte et al. 1998)and the gas in terms of C O (Liseau et al. 2010). Here we re-port on mapping observations of the ρ Oph A cloud core in sev-eral frequency settings that cover most of the low-energy H CO,HDCO, and D CO lines in the 1.3 millimeter band using the 12m APEX telescope (G¨usten et al. 2006). In addition, lines frommany other species were observed simultaneously (eg. CH OH,SO, and N D + ). Here the CH OH results are of importance be-cause CH OH is directly involved in the H CO chemistry andthe N D + results are, of course, of interest for the deuteration.Although the sulphur chemistry is not of immediate interest herewe chose to include the SO and SO observational and analysisresults since they are of importance as a complementary tool fordetermining the physical conditions. This paper is organized asfollows. In Sect. (2) we describe the observations and then, inSect. (3), we present the observational results. In Sect. (4) weobtain the physical conditions. Before concluding, in Sect. (6),we discuss our results in Sect. (5).
2. Observations
The APEX telescope at Chajnantor (Chile) was used to map the3 ′ × ′ (with a spacing of 30 ′′ ) area centred on the coordinates α (J2000) = h m . s and δ (J2000) = − ◦ ′ ′′ whichis very close to the SM1N position in the ρ Oph A cloud core as designated by Andr´e et al. (1993) in their submillimeter con-tinuum maps. We used the single-sideband tuned APEX-1 re-ceiver which is part of the Swedish heterodyne facility instru-ment (Vassilev et al. 2008). It has a sideband rejection ratio morethan 10 dB. The image band is 12 GHz above or below the ob-serving frequency depending on whether the tuning is optimizedfor operation in the upper or lower sidebands, respectively. Fortwo frequency settings, the strong CO(2-1) and C O(2-1) linesentered via the image band and from the strength we could es-timate the sideband rejection ratio to be about 15 dB in bothcases. As backend we used two 1 GHz modules of the FFTS(Klein et al. 2006). Each FFTS 1 GHz module has 8192 e ff ectivechannels and the modules can be placed independently withinthe IF band width of 4-8 GHz. At 230 GHz this channel spacingcorresponds to 0 .
16 km s − and the telescope beamsize (HPBW)is 27 ′′ .The telescope control software APECS (Muders et al. 2006)was used to control the raster mapping. All observations wereperformed in position-switching mode using a reference positiono ff set by 300 ′′ east and 200 ′′ north of the map center. The point-ing of the telescope was maintained and checked regularly bymeans of small CO(2-1) cross maps of the relatively nearby car-bon stars IRAS15194 − ff sets were generally consistent from day to day andwe believe we have an absolute pointing uncertainty better than5 ′′ . To optimize the antenna focussing we used continuum ob-servations on Jupiter and Saturn.The observations took place in two blocks and one additionalday during 2009: April 24 - May 1, May 21, and July 4 - July 9.The column of precipitable water vapour was typically around0.7 mm (varied between 0.3-2.9 mm). Typical system tempera-tures for the frequencies in question (218-252 GHz) were 200-220 K. The telescope main beam e ffi ciency is η mb = .
73 at 345GHz (G¨usten et al. 2006). Hence, using the antenna surface ac-curacy of 18 µ m we estimate that (using the Ruze formula) thatthe APEX main beam e ffi ciency around 230 GHz is just slightlyhigher than at 345 GHz, about 0.75 which we adopt when con-verting the observed intensities from the T ∗ A scale to the T mb intensity scale. The heterodyne calibration procedure at APEXis more elaborate than the standard chopper wheel calibrationscheme and involves three measurements by apart from the nor-mal sky observation it also measures the receiver temperature byobserving a hot and cold loads. Moreover, the atmospheric con-tribution is based on the model (Pardo et al. 2001) that has beenadapted to the atmospheric characteristics at the Chajnantor site.The absolute calibration uncertainty is estimated to be 10% inthe 1 mm band.In Table 1 we summarize the targeted formaldehyde lines.The additional lines are summarized in Table 2. In the tableswe include the line parameters: the transition frequency (typi-cal uncertainty is 0.05 MHz or better), energy of lower level,and Einstein A -coe ffi cient. These have been compiled fromthe Cologne Database of Molecular Spectroscopy (M¨uller et al.2001, 2005). We here also indicate the symmetry due to the nu-clear spin direction of the H (or D) nuclei, which for H CO,H
CO and D CO can be ortho (parallel spins) or para (anti-parallel spins). The statistical weight ratio of the symmetries isalso noted if applicable. In the case of CH OH, the internal rota-tion of the methyl group results in the symmetry species A andE. Note that radiative or non-reactive collisional transitions areforbidden between levels of di ff erent symmetry.
2. Bergman et al.: Deuterated formaldehyde in ρ Ophiuchi A
Table 1.
Observed H CO, H
CO, HDCO, and D CO lines
Frequency Transition E l A ul Symmetry(MHz) (K) (s − )H CO ( o / p = , − , . × − para218475.63 3 , − , . × − para218760.07 3 , − , . × − para225697.78 3 , − , . × − orthoH CO ( o / p = , − , . × − orthoHDCO227668.45 1 , − , . × − , − , . × − D CO ( o / p = , − , . × − para231410.23 4 , − , . × − ortho233650.44 4 , − , . × − ortho234293.36 a , − , . × − para234331.06 a , − , . × − para245532.75 4 , − , . × − para a only observed at (0 , − ′′ ) Table 2.
Additional lines
Frequency Transition E l A ul Symmetry(MHz) (K) (s − )CH OH ( A / E = , − , . × − E241700.22 5 , − , . × − E241767.22 5 , − , . × − E241791.43 5 , − , . × − A241879.07 5 , − , . × − E241904.15 5 , − , . × − E241904.65 5 , − , . × − ESO219949.44 5 − . × − a − . × − SO246663.47 6 − . × − SO , − , . × − , − , . × − , − , . × − SO , − , . × − , − , . × − , − , . × − N D + − . × − a only observed at (0 , − ′′ )
3. Results
In this section we will first display the formaldehyde mapping re-sults of the ρ Oph A cloud core, both as spectra, integrated inten-sity maps or velocity position diagrams where appropriate. Afterthat, the mapping results for the other molecules will presented.Comparisons will be made with the existing APEX C O(3-2)data at 329 GHz of Liseau et al. (2010) and the IRAM 30 mcontinuum map at 1.3 mm of Motte et al. (1998). The angularresolution of the two data sets is similar; the C O data havean HPBW of 19 ′′ while the 1.3 mm continuum map has an an-gular resolution of 15 ′′ . Both these maps are shown in Fig. 1where the 1.3 mm continuum map is shown in contours on topof a grey-scale image of the C O(3-2) integrated intensity. The
Fig. 1.
Colour image: The C O(3-2) integrated intensity mapof ρ Oph A cloud core from Liseau et al. (2010). The inten-sity scale is shown to the right. Contours: The 1.3 mm contin-uum data by Motte et al. (1998). The contours start at 0.15 Jywith subsequent contours at every increment of 0.15 Jy. The fluxdensities are given in a 15 ′′ beam. We show the positions ofthe mm peaks from Motte et al. (1998) and the C O peaks ofLiseau et al. (2010). The map o ff sets are given relative the posi-tion α (J2000) = h m . s and δ (J2000) = − ◦ ′ ′′ .3 ′ × ′ region shown in Fig. 1 corresponds to the area that hasbeen mapped here in formaldehyde. All lines listed in Table 1 were detected except the 1 , − , tran-sition of HDCO at 227 GHz. This transition has a much lowerspontaneous rate coe ffi cient than the other transitions. Two ofthe D CO lines were not mapped and were only observed in the(0 ′′ , − ′′ ) position. The typical rms in a map spectrum is in therange 0.06-0.1 K.In Fig. 2 we plot the H CO(3 , − , ) and D CO(4 , − , )spectra. They consist of 49 spectra separated by 30 ′′ on a 7x7grid. The H CO lines posess a relatively complicated structureas compared to the D CO lines. The latter line shows only onevelocity component around 3 . − with full width at halfmaximum (FWHM) of 0 . − and it is peaking strongly atthe (0 ′′ , − ′′ ) position which, within the positional uncertain-ties, coincides with the SM1 or P3 position (cf. Fig. 1). We willhereafter refer to this position as the D-peak. The D-peak is alsoclearly seen in the integrated intensity maps of the three di ff erentH CO isotopologues (Fig. 3).Maret et al. (2004) present a H CO(3 , − , ) IRAM 30 mspectrum toward the position of VLA1623. The nearest posi-tion to VLA1623 in our map is at (0 ′′ , − ′′ ). The shape ofour spectrum at this position is similar to the one presented byMaret et al. (2004) and our T mb peak temperature of 3 . T ∗ a /η mb ≈ . / . ≈ . ′′ in the IRAM 30 m spectrum. Thiswould be expected for a relatively extended source which showslittle variation on the scale of 11 ′′ − ′′ . Also, from the distribu-
3. Bergman et al.: Deuterated formaldehyde in ρ Ophiuchi A
Fig. 2. H CO(3 , − , ) and D CO(4 , − , ) map spectra toward the ρ Oph A cloud. The (0 ′′ , ′′ ) position is the same as in Fig. 1.The vertical scale in each spectrum represents T mb scale in K and the horizontal scale is velocity v LSR with respect to local standardof rest in km s − as shown in the upper rightmost panel.tion of the H CO emission we do not see any strong componentthat can be attributed to VLA1623. The velocity as obtained from the N H + (1-0) observationsby Di Francesco et al. (2004, 2009) (see also Andr´e et al. 2007)toward SM1 is 3 . − with an FWHM of 0 . − , i.e. verysimilar to our values determined from the D CO lines at the D-peak. The peak intensity of the D CO line is about half that ofH CO. The size of D CO 4 , − , emission region (see Fig. 3),is found, by fitting a 2-dimensional gaussian, to be 33 ′′ × ′′ .This size corresponds to a deconvolved source size of about S = ′′ × ′′ for a beam size of B = ′′ . The filling factor whenpointing the 27 ′′ beam toward the center of the source, is then S / ( S + B ) = (19 ′′ × ′′ ) / (33 ′′ × ′′ ) ≈ .
5. This fillingfactor is merely an upper limit because any unresolved small-scale clumpiness may decrease the filling factor further.In order to investigate the H CO line distribution over the ρ Oph A cloud core we plot four velocity position diagrams(Fig. 4) between ∆ δ = ± ′′ at ∆ α = ′′ . Here we also seethat HDCO is peaking at the D-peak at ∆ δ = − ′′ . Lookingat the HDCO emission, there appears to be a velocity gradientover the D-peak source of about 0 . − when going from ∆ δ = ′′ to ∆ δ = − ′′ . It is also close to the D-peak where theC O(3-2) emission has its maximum. The secondary C O(3-2) Based on the radial intensity distributions of the sub-millimetercontinuum, Jayawardhana et al. (2001) suggested that the infall zonearound VLA1623 is surrounded by a constant-density region whichis the dominating contribution to the radial intensity profiles at scales > ∼ ′′ . peak (P1 in Fig. 1) is at ∆ δ = + ′′ and with a lower velocity of ≈ . − as compared to the velocity of + . − at theD-peak position and it is also here where the low-energy H COline intensities reach their maximum value at v LSR = . − .We denote this peak P1 from now on. There is also weak emis-sion emanating from both HDCO and D CO at the ∆ δ = + ′′ position.Interestingly, there is a third H CO peak at ∆ δ = ′′ and v LSR ≈ + . − which has no obvious counterpart in theC O velocity position map. This peak makes the H CO(3 , − , ) spectrum at (0 ′′ , ′′ ) to look doubly peaked (Fig. 2).However, these are two di ff erent cloud components and the dipis not an e ff ect of self-absorption. This is evident from the veloc-ity position diagrams in Fig. 4, but is also clear when comparingthe H CO(3 , − , ) and H CO(3 , − , ) spectra. In Fig. 5the central three spectra from the ortho lines H CO(3 , − , )and H CO(3 , − , ) are displayed. Toward the (0 ′′ , ′′ ) posi-tion we see the doubly peaked line profile also for this H COline, however there is no hint that the H
CO line is peaking atthe velocity of the dip for H CO. If anything, the emission fromthe rarer species seems to follow that of the main species. Wewill return to the low-velocity component seen in the (0 ′′ , ′′ )position below when we present the CH OH results. CH OH results In Fig. 6 we show a map of the integrated intensity for theCH OH 5 , − , and 4 , − , E-lines. The latter line comes
4. Bergman et al.: Deuterated formaldehyde in ρ Ophiuchi A
Fig. 3.
Maps of the H CO(3 , − , ), HDCO(4 , − , ), andD CO(4 , − , ) integrated line intensity over the range 3 . − . − .from levels of higher energy (Table 1). The distribution ofCH OH is quite di ff erent from that of H CO (see Fig. 3). Itshould be noted that the CH OH 4 , − , line and the H COlines at 218 GHz were observed simultaneously so the di ff er- Fig. 6.
Integrated intensity maps of the CH OH 5 , − , and4 , − , E-lines. First contour is at 0 . − and the incre-ment is 0 . − .ent distributions cannot be a result of large pointing o ff sets. TheCH OH emission has its maximum at ( − ′′ ,
0) and in Fig. 7 allobserved 5 − v LSR ≈ . − . The cloudcomponent at ( − ′′ , ′′ ) extends into adjacent positions. Thelow-velocity feature seen for H CO (Fig. 4) is very likely as-sociated with the CH OH-peak. The higher energy H CO lines3 , / − , / peak near this position. In addition, there is littleCH OH at the D-peak position (Fig. 6).
Two lines from SO(5 − ) and SO(6 − ) were coveredduring the observations. The map spectra are shown in Fig. 9.There is a single strong peak in the SO line intensity towardthe position ( − ′′ , + ′′ ) which we will call the S-peak. Herethe SO and SO lines are very narrow, only about 0 . − .This is about twice the width one would expect from thermalbroadening only (at T kin =
20 K). Further west, at ( − ′′ , + ′′ ),the SO line is broader, about 1 . − . The profile of SOshows line wings here which could be related to outflow ac-tivity. However, it is at the very border of our map and a clear
5. Bergman et al.: Deuterated formaldehyde in ρ Ophiuchi A
Fig. 7. CH OH 5-4 lines around 241 GHz toward the ( − ′′ , ′′ )position. Note that the 5 , − , and 5 , − , E-lines areblended.delineation into red and blue wings are impossible. It shouldbe pointed out that the S-peak has no structural counterpart inthe C O(3-2) map nor the 1.3 mm continuum map (cf. Fig. 1).Toward the D-peak, the SO lines exhibit a similar emission ve-locity ( v LSR ≈ . − ) and line width ( ∆ v ≈ .
83 km s − ) asthe other molecules. The covered SO and SO lines (Table 2)show a distribution similar to that of SO and SO. At the P1-position the SO and SO emission velocity of about ≈ . − is lower than the velocity of ≈ . − for the formaldehydeisotopologues and CH OH.The deuterated version of N H + (observed byDi Francesco et al. 2004, 2009; Andr´e et al. 2007), N D + (3-2)at 231 GHz was also covered during the formaldehyde obser-vations. The integrated intensity map of N D + (3-2) is shownin Fig. 10. The peak intensity of the N D + emission is clearlyassociated with the D-peak, with a secondary weaker sourceto the south-east. This secondary peak is not coincident withSM2 (see Fig. 1) but is located in the direction where the dustemission extends. Moreover, it shows up in the N H + map ofDi Francesco et al. (2004) and is designated N6 by them. Theemission velocity of N D + (3-2) at the D-peak is 3 . − andthe FWHM is 0 .
85 km s − in agreement with what is found forthe other species. After presenting the results of the distribution of the di ff erentmolecules toward the ρ Oph A cloud above we will now makea summary of the results which will be used in the analysis be-low. For H CO and CH OH we have identified three positionsof interest which seem to represent distinctive sources; the D-peak at (0 ′′ , − ′′ ), the P1 peak at (0 ′′ , + ′′ ), and the CH OHpeak at ( − ′′ , ′′ ). In addition to these three peaks, the sulphurcontaining molecules are localized to ( − ′′ , + ′′ ) (the S-peak).The emission velocities and FWHMs as obtained for the di ff er-ent species in these sources are summarized in Table 3 (fromgaussian fits of the lowest energy lines in those cases where wehave mapped multiple lines). Integrated line intensities for therelevant positions are tabulated in Table 4, Table 5, and Table 6.In these tables the 1 σ uncertainties of the integrated intensitiesdue to noise have been entered. Table 3.
Properties of the H CO, CH OH, SO, and SO emissionin selected cloud positions from gaussian fits O ff set Molecule v LSR ∆ v (FWHM)( km s − ) ( km s − )(0 ′′ , − ′′ ) H CO 3.8(0.1) 0.95(0.05)HDCO 3.6(0.1) 0.70(0.05)D CO 3.8(0.1) 0.68(0.02)CH OH 3.7(0.1) 0.73(0.05)SO 3.7(0.1) 0.83(0.05)SO ′′ , + ′′ ) H CO 3.4(0.1) 1.20(0.05)HDCO 3.2(0.1) 0.85(0.12)D CO 3.2(0.1) 0.60(0.15)CH OH 3.4(0.1) 1.18(0.05)SO 3.0(0.1) 1.06(0.05)SO − ′′ , ′′ ) a H CO 2.8(0.2) 0.95(0.10)3.6(0.2) 0.87(0.10)CH OH 2.6(0.2) 0.62(0.10)3.2(0.2) 1.27(0.10)( − ′′ , + ′′ ) SO 3.0(0.1) 0.78 (0.05) SO 2.9(0.1) 0.43(0.02)SO SO a Fitted by two components, full intensity used in models
Table 4. H CO, H
CO, HDCO, and D CO integrated line in-tensities in three positions of the ρ Oph A cloud
Molecule Line Frequency I mb = R T mb dv ∆ I mb a (MHz) ( K km s − ) ( K km s − )D-peak (0 ′′ , − ′′ )H CO 3 , − , , − , , − , , − , CO 3 , − , , − , CO 4 , − , , − , , − , , − , , − , , − , ′′ , + ′′ )H CO 3 , − , , − , , − , , − , CO 3 , − , , − , , − , CO 4 , − , , − , OH-peak ( − ′′ , ′′ )H CO 3 , − , , − , , − , , − , CO 3 , − , , − , CO 4 , − , , − , a σ error due to noise only6. Bergman et al.: Deuterated formaldehyde in ρ Ophiuchi A
Fig. 10.
Integrated intensity map of the N D + ρ Oph A. First contour is at 0 . − and the increment is0 . − . Table 5. CH OH integrated line intensities in three positions ofthe ρ Oph A cloud
Line Frequency I mb = R T mb dv ∆ I mb a (MHz) ( K km s − ) ( K km s − )D-peak (0 ′′ , − ′′ )4 , − , E1 218440 0.21 0.055 , − , E1 241700 0.13 0.055 , − , E2 241767 0.72 0.055 , − , A 241791 0.81 0.055 , − , E1 241879 0.09 0.055 , / − , / E1 241904 0.07 0.05P1 (0 ′′ , + ′′ )4 , − , E1 218440 0.61 0.055 , − , E1 241700 0.59 0.055 , − , E2 241767 2.39 0.055 , − , A 241791 2.88 0.055 , − , E1 241879 0.20 0.055 , / − , / E1 241904 0.31 0.05CH OH-peak ( − ′′ , ′′ )4 , − , E1 218440 1.43 0.055 , − , E1 241700 1.49 0.055 , − , E2 241767 4.53 0.055 , − , A 241791 5.25 0.055 , − , E1 241879 0.96 0.055 , / − , / E1 241904 1.43 0.05 a σ error due to noise only
4. Analysis
For several of the detected species (Table 4) we have mul-tiple transitions that have significantly di ff erent energy lev-els. In such cases the so-called rotation diagram analysis (eg.Goldsmith & Langer 1999) can be employed to determine therotation temperature, T rot , and the molecular column density, N mol . If all lines are optically thin, the rotation diagram methodoften is adequate to analyse multi-transition data. However, hereit is likely that we have a mixture of optically thin and thicklines and therefore we adopt the modified approach to the rota-tion diagram method described by Nummelin et al. (2000). The Table 6.
Integrated line intensities for other molecules in fourpositions of the ρ Oph A cloud
Transition Frequency I mb = R T mb dv ∆ I mb a (MHz) ( K km s − ) ( K km s − )D-peak (0 ′′ , − ′′ )SO(5 − ) 219949 7.22 0.06SO(6 − ) 251826 3.37 0.04 SO(6 − ) 246663 0.19 0.03SO (5 , − , ) 241616 0.69 0.02SO (10 , − , ) 245563 0.16 0.04N D + (3 −
2) 231322 3.38 0.06P1 (0 ′′ , + ′′ )SO(5 − ) 219949 13.1 0.06 SO(6 − ) 246663 0.55 0.04SO (5 , − , ) 241616 1.70 0.05SO (10 , − , ) 245563 0.57 0.04N D + (3 −
2) 231322 0.37 0.08CH OH-peak ( − ′′ , ′′ )SO(5 − ) 219949 8.25 0.06 SO(6 − ) 246663 0.08 0.04SO (5 , − , ) 241616 0.23 0.05SO (10 , − , ) 245563 0.05 0.04N D + (3 −
2) 231322 0.17 0.07S-peak ( − ′′ , + ′′ )SO(5 − ) 219949 14.5 0.06 SO(6 − ) 246663 1.31 0.04SO (5 , − , ) 241616 3.76 0.05SO (10 , − , ) 245563 1.14 0.04SO (14 , − , ) 226300 0.27 0.03 SO (4 , − , ) 246686 0.24 0.04 SO (8 , − , ) 241985 0.18 0.04 SO (11 , − , ) 219355 0.46 0.04N D + (3 −
2) 231322 0.07 a σ error due to noise only modification involves the inclusion of the peak optical depth andthus, in addition to T rot and N mol , also the beam filling factor, η bf ,can be determined by minimization of a χ -value. For a gaussiansource distribution, η bf is given by η bf = S / ( S + B ), where S is the FWHM source size and B is the FWHM beam size. If alltransitions included in the analysis are optically thin and η bf isset to 1 ( B << S ), the method is very similar to the rotationdiagram analysis with N mol then representing a beam averagedcolumn density.In selected cases we will also check the results and refine themodels obtained with the modified rotation diagram analysis byemploying a more accurate treatment of the line excitation andradiative transfer. We have here adopted the accelerated lambdaiteration (ALI) technique outlined by Rybicki & Hummer (1991,1992). The ALI model cloud consists of spherically concentricshells and allows only for radial gradients of the physical param-eters (kinetic temperature, molecular hydrogen density, molecu-lar abundance, and radial velocity field). Moreover, and in con-trast to the rotation diagram method, collisional excitation is in-cluded in the analysis, so the collision partner (here H ) densityis a physical input parameter. The code used in the present workhas been tested by Maercker et al. (2008) and it allows the in-clusion of dust as a source of continuum emission in the shells.When the radiative transfer has been iteratively solved using the The only di ff erence is that in the normal rotation diagram analy-sis a straight line is least-square fitted to quantities proportional to thelogarithm of the line intensities but here the fit is performed directly byminimizing the sum of the squared and error-weighted di ff erences ofobserved and modelled line intensities. 7. Bergman et al.: Deuterated formaldehyde in ρ Ophiuchi A
ALI approach, a model spectrum is produced by convolving thevelocity dependent intensity distribution of the model cloud witha gaussian beam.
As can be seen in Table 4 all four H CO lines are clearly detectedin the three sources (D-peak, P1, and CH OH-peak). The threepara lines at 218 GHz have all been observed simultaneously sotheir relative strengths are not a ff ected by any pointing or cal-ibration uncertainties. The only ortho transition, the 3 , − , line at 225 GHz, was observed in a separate frequency setting. Inour rotation diagram analysis we treat the ortho and para H COlines together by assuming that the population distribution is de-termined by T rot also between the states of di ff erent symmetry(through some exchange reaction or formation mechanism). ForH CO the lowest ortho level (1 , ) is about 15 K above the low-est para level (0 , ). For D CO the reverse situation is in e ff ectand it is the lowest level 0 , that is an ortho state while the low-est para level 1 , is 8 K higher up in energy. For a very low T rot (during molecule formation) most of the molecules will be inthe lowest energy state (para for H CO and ortho for D CO). Onthe other hand, if T rot is much greater than the energy di ff erenceof the symmetry states, the ortho to para population ratio willbe governed by the statistical weight ratio which is o / p = CO and o / p = CO.In Table 7 the results of the modified rotation diagram anal-ysis are shown. Here the number of lines used in the analy-sis for each molecule is tabulated together with best fit val-ues of T rot , N mol , and η bf . In the last column, we list the min-imum and maximum optical depths of the used transitions inthe analysis. If there are not enough of lines or all lines areoptically thin, one or two of the parameters have been set toa result obtained by another molecule in a previous fit. For in-stance, in the D-peak source, all three parameters could be de-termined in the analysis of H CO, while for the optically thinD CO lines, η bf had to be set to the value found for H CO.Using the same η bf for all formaldehyde isotopologues willalso make the determined column densities directly compara-ble with each other. Interestingly, the T rot obtained for D CO,17 . CO( T rot = . ff ect of subthermal exci-tation and di ff erence in optical depths, where the higher opticaldepths for H CO make the excitation more e ffi cient via photontrapping as compared to D CO. A lower rotation temperature ofthe less abundant formaldehyde isotopologues was also seen inIRAS16293 − OH peaks) the number of detected D CO transitions isnot su ffi cient to allow for a T rot determination and we assumethat the ratio of T rot (D CO) / T rot (H CO) is the same in these twosources as in the D-peak source. The deduced T rot is higher in theP1 and CH OH peaks than in the D-peak. This could be expectedbecause the ratio of the 3 , − , and 3 , − , H CO linesis a good measure of kinetic temperature (Mangum & Wootten1993) and this ratio is highest toward the CH OH-peak. Ofcourse, from the models we can also estimate the intensity oflines not included in the analysis. In particular, we find for theHDCO 1 , − , line that for the determined model parameterstoward the D-peak its expected intensity is about 5 times belowthe noise level, consistent with our non-detection.The derived H CO rotation temperatures for the three coresare quite similar to the ones obtained by Parise et al. (2006) forother low-mass protostar sources. Likewise, the derived H CO Table 7.
Analysis results for H CO, H
CO, HDCO, D CO,CH OH, SO, and SO from using the modified rotation diagramtechnique Molecule No. of T rot N mol η bf τ min , τ max lines (K) cm − D-peak (0 ′′ , − ′′ )H CO 4 22.5 2 . × CO 6 17.4 3 . × (0.446) 0.01,0.54HDCO 1 (17.4) 2 . × (0.446) 0.23H CO 1 (17.4) 4 . × (0.446) 0.06CH OH 6 6.8 7 . × . × (1) 0.50,1.21 SO 1 (18.3) 9 . × (1) 0.02SO . × (1) 0.01,0.07P1 (0 ′′ , + ′′ )H CO 4 25.9 1 . × CO 2 (20) 1 . × (0.469) 0.01,0.02HDCO 1 (20) 6 . × (0.469) 0.05H CO 1 (20) 1 . × (0.469) 0.02CH OH 6 7.4 1 . × . × (1) 0.04,0.12CH OH-peak ( − ′′ , ′′ )H CO 4 26.3 2 . × CO 2 (20) 1 . × (0.325) 0.01,0.03HDCO 1 (20) 2 . × (0.325) 0.02H CO 1 (20) 3 . × (0.325) 0.05CH OH 6 9.2 1 . × . × (1) 0.01S-peak ( − ′′ , + ′′ )SO . × SO . × (1) 0.04,0.05A value surrounded by (...) indicates a fixed parameter. column densities are just slightly higher than the correspondingrotation diagram values of Parise et al. (2006). The filling factor η bf = .
446 determined toward the D-peak is in agreement withthe filling of ∼ . CO 4 , − , sourcesize in Sect. (3.1). The CH OH analysis using the modified rotation diagram tech-nique is based on the integrated line intensities listed in Table 5.In all sources 6 CH OH lines with lower state energies rangingfrom 23 to 46 K, have been used. In the D-peak source a coupleof lines are just marginally detected but are included in the bestfit analysis since they are weighted with the uncertainty and donot a ff ect the fit significantly. The line feature at 241904 MHzis a blend of two CH OH lines of about the same energy and A -coe ffi cient and we have not been able to clearly resolve theminto individual components. All lines at 241 GHz were observedsimultaneously. The only CH OH line not belonging to the 241GHz 5 − , − , E-line which is located inthe 218 GHz H CO-band.Only one line from the methanol A-species has been ob-served and included in the analysis. Just like in the case of or-tho and para formaldehyde, the A- and E-species of methanolare treated together. The energy di ff erence between the lowestA and E-methanol states is 8 K with the A-species having thelowest energy. Furthermore, we only consider methanol to be inits lowest torsional state.The best fit results in the CH OH analysis have been enteredin Table 7. In all three sources we find a low T rot of 7 to 9 K. Alsothe η bf is found to be very close to 1. The much lower rotation
8. Bergman et al.: Deuterated formaldehyde in ρ Ophiuchi A temperatures found in the CH OH analysis as compared to theH CO and D CO results suggest that the excitation of CH OHis quite sub-thermal and a more elaborate treatment of the exci-tation and radiative transfer is needed (as was demonstrated byBachiller et al. (1998) in their Fig. 9a). This is also supported bythe fact that none of the CH OH fits are good since several of themodelled line intensities deviate substantially from the observedline intensities.
Our aim with the non-LTE modelling using the ALI code isto construct a homogeneous (in terms of physical parametersbut not in terms of excitation which may vary radially) spher-ical model cloud, for each of the source positions (D-, P1, andCH OH-peaks), that produces the observed H CO and CH OHspectra. The results from the modifed rotation diagram analy-sis above are used as a guide when adopting the model cloudparameters. The ALI setup for the statistical equilibrium equa-tions includes collisional excitation rates and for formaldehydewe use the He-H CO collisional coe ffi cients calculated by Green(1991). The coe ffi cients have been multiplied by 2.2 to approx-imate the H -H CO collision system. For the H -CH OH colli-sion coe ffi cients we adopt those of Pottage et al. (2004). For bothspecies only levels below 200 K are included and we divide themodel cloud into 29 shells. We do not include any radial velocityfield in the source and instead we simply use a microturbulentvelocity width, v turb , that reproduces the observed line widths.For an optically thin line, the FWHM line width is √ v turb .The results of the three ALI models have been summarizedin Table 8. The listed cloud radii R (in the first column) are basedon the corresponding H CO beam filling results (Table 7) and adistance to ρ Oph A of 120 pc. The tabulated column densitiesare the source-averaged values 2 N peak /
3, where N peak is the col-umn density through the center of the spherical model cloud. Itshould be noted that the derived n (H ) (and hence N (H )) is acompromise value since, for all three models, a 20% lower valueyields better fits for CH OH while a similarly higher value isoptimum for H CO. However, as the di ff erences are less thanabout 50% any real di ff erence in n (H ) is within the uncertain-ties of the adopted collisional coe ffi cients. The ortho-to-para ra-tio is about 2 for all three models but since the ortho results arebased on a single optically thick line ( τ ∼
4) only and not ob-served simultaneously with the three para lines, we cannot reallyexclude ortho-to-para ratios in the range 1-3. The derived H COcolumn densities are about 30% lower than those derived fromthe modified rotation diagram analysis. In contrast to N (H CO)and N (H ) which do not vary much among the three cores, theCH OH column densities vary from 5 × cm − in the D-peakto 4 × cm − at the CH OH-peak position. In addition, theE / A CH OH ratios seem to be close to 1 and since the A-lineis observed simultaneously with the E-lines the determined A / E-ratio of 1 is less uncertain than the estimated ortho / para ratios forformaldehyde. The optical depth for the strongest of the CH OHlines are about 0.5-0.6 toward the CH OH-peak. The total H mass of a core is about 0 . − . ⊙ .To our knowledge, there are no collision coe ffi cients avail-able for the collision system H -D CO, so we cannot use ourALI model for D CO. However, using the H -H CO collisionrates we check what happens to the excitation temperatureswhen adopting a factor of 7 lower column density (the ratio of N (H CO) and N (D CO) in Table 7 is close to 7). We then findthat the excitation temperatures of the 3 , − , transition drop between 20-30% at di ff erent radii. This drop in excitation tem-perature is in line with the 20% lower T rot = . CO as compared to H CO value of T rot = . CO as compared to H CO canbe explained by less e ffi cient photon trapping in the excitationof D CO. We also checked the influence on our results by con-tinuum emission of dust, and the inclusion of a dust componentwith standard dust parameters ( T d =
22 K, gas-to-dust mass ratioof 100) had negligible impact on our modelling results.Liseau et al. (2003) report results from a large velocity gra-dient (LVG) analysis of the ρ Oph A using CH OH 2 − − ′′ and 35 ′′ ) than for the presentstudy. The analysed data were averaged from spectra spaced by30 ′′ in the north-south direction around the P2 position (seeFig. 1) and their spectra is thus a partial blend of the contri-bution of the three cores. This is especially the case for their2 − T kin =
20 K, n (H ) = . × cm − , X [CH OH] = . × − ) are in goodagreement with those reported here. In the S-peak position we have detected three SO lines and amodified rotation diagram analysis gave best-fit results for η bf very close to 1. For SO we find a rotation temperature of about20 K and a column density of N (SO ) = . × cm − . Thesame analysis was made for two of the three detected SO linesand we get N ( SO ) = . × cm − when using the rotationtemperature of the more common variant. The 11 , − , SO line was excluded in the fit since its observed strengthis incompatible (too strong by about a factor of 2) with the ro-tation temperature found for SO , cf. Table 6. The reason forthis is unclear but could be due to an unknown blend or a non-LTE e ff ect involving K a = . × cm − (againassuming η bf = lines give T rot =
23 K and a beam averaged column density of9 . × cm − . Furthermore, assuming T rot =
20 K we arrive ata beam averaged SO column density of 8 . × cm − towardthe CH OH-peak. The SO and SO results have been entered inTable 7.The analysis results of the other molecules are summarizedin Table 9. The column densities listed here are all calculatedunder the assumption of optically thin emission. The C O(3-2)column densities are based on the data from Liseau et al. (2010)and have been derived using the kinetic temperatures derivedfrom the ALI analysis as excitation temperature as the lowerC O transitions are expected to be thermalized at the high densi-ties of the cores. For the S-peak, the SO rotation temperature of20 K was used as excitation temperature for all molecules. Alsotabulated is the N (H ) column density deduced from N (C O)and assuming a C O abundance of X [C O] = × − (stan-dard value of undepleted gas interstellar gas, Frerking et al.(1982), but see also Wouterloot et al. (2005) for a discussionapplicable to the ρ Oph cloud). In this context it is enlight-ning to estimate how much the cores contribute to the observedC O(3-2) integrated line intensity. Adopting the physical pa-rameters (Table 8) and assuming X [C O] = × − we find,using the ALI code for C O, that the cores make up 31%,37%, and 67% of the observed C O(3-2) emission in the D-, P1-, and CH OH-cores, respectively. Firstly, this tells us that
9. Bergman et al.: Deuterated formaldehyde in ρ Ophiuchi A
Table 8.
Model cloud properties and results for the ALI analysis
Source R / v turb T kin n (H ) N (H ) M (H ) X [H CO] × N (H CO) X [CH OH] × N (CH OH)(cm) (km s − ) (K) (cm − ) (cm − ) (M ⊙ ) para ortho (cm − ) A E (cm − )D-peak 3.3 0.4 24 6 × × . . . × . . . × P1 3.4 0.6 27 7 × × . . . × . . . × CH OH-peak 2.7 0.6 30 1 × × . . . × . . . × the C O abundance in the cores cannot be much higher than2 × − as they then would produce too much C O(3-2) emis-sion. Secondly, an appreciable part of the observed C O(3-2)emission is likely to arise in a lower H density ( < ∼ cm − )environment. This would explain the higher N (H ) column den-sities deduced from optically thin C O (Table 9) as comparedto the ALI results (Table 8). A similar finding was made byWouterloot et al. (2005) in their CO study of other regions in the ρ Oph cloud where the cold ( ∼
10 K) and dense ( > ∼ cm − )cores were surrounded by a warmer ( ∼
30 K) and less dense ( ∼ cm − ) envelope. Also, given the observed ratio of C O(3-2) and C O(3-2) of about 23 toward the D-peak (Liseau et al.2010), it is likely that the C O(3-2) emission is somewhat op-tically thick ( τ ≈
2) and hence the listed N (C O) and N (H )(Table 9) would be about a factor τ/ (1 − e − τ ) ≈ . column density toward the D-peak would then be N (H ) = . × cm − . Likewise, the SO column density inthe D-peak position of 1 . × cm − is lower than the cor-responding value of 2 . × cm − (Table 7) where the lattervalue includes compensation by optical depth. Using the ALImodel for the D-peak core (Table 8) we find that adopting anabundance of X [SO] = . × − results in integrated line in-tensites for the SO 5 − and 6 − transitions that are closeto the observed values (Table 6). Hence, using the previously de-rived physical properties of the D-peak we can also explain theobserved SO emission. We assume that the bulk of SO emissionoriginates in the D-peak core itself and not from the low-densityenvelope. With the same assumption for the optically thin SO lines toward the D-peak, we find an SO abundance of X [SO ] = N [SO ] / N [H ] ≈ (4 . × / . / × ≈ × − us-ing the beam averaged SO column density of 4 . × cm − ,D-peak filling factor of 0.446 (Table 7) and the D-peak coreH column density (Table 8). Similarly, for the P1 and CH OHpeak positions we obtain X [SO ] = × − and 6 × − , re-spectively. Here the value for the P1 position is uncertain sinceit is likely that the SO emission do not originate in the samesource as H CO since they exhibit di ff erent emission velocities(cf. Table 3). Using the P1 H column density from C O(3-2)(Table 9) instead will lower the abundance by about a factor offive. Taken together, the SO abundances in the D-peak, P1, andCH OH cores seem to fall in the range (0 . − × − . However,these SO abundances are all lower than the S-peak abundance,which can be estimated to X [SO ] = × − . To estimate the ratios of di ff erent formaldehyde isotopologueswe use the results obtained by the modified rotation diagramtechnique (Table 7). Even for the main isotopic species weuse the rotation diagram analysis results and not the ALI re-sults, since e ff ects like adopting a spherical source will a ff ectthe results and we rather be consequent in as many aspects aspossible when estimating column density ratios. The four col-umn density ratios; HDCO / H CO, D CO / HDCO, D CO / H CO,H CO / H CO have been calculated for each source position and
Table 9.
Analysis results for other molecules assuming opticallythin emission N (C O) N (H ) a N ( SO) N (SO) N (N D + )(cm − ) (cm − ) (cm − ) (cm − ) (cm − )D-peak (0 ′′ , − ′′ )1 . × × . × . × . × P1 (0 ′′ , + ′′ )1 . × × . × . × . × CH OH-peak ( − ′′ , ′′ )7 . × × < . × . × . × S-peak ( − ′′ , + ′′ ) b . × × . × . × < . × a Assuming X [C O] = × − b Adopting T ex =
20 K entered into Table 10. The listed ratio uncertainties have been es-timated by including an absolute calibration uncertainty of 10%together with the 1 σ uncertainty due to noise, where the latteris the most prominent source of error for the weak lines. TheHDCO / H CO ratio of 0 . ± .
015 toward the D-peak is simi-lar to the values found by Parise et al. (2006) for low-mass pro-tostars. The D CO / HDCO ratio of 1 . ± .
19 is higher than theParise et al. (2006) values which all fall in the range 0.3-0.9. Infact, it is the only reported D CO / HDCO ratio that is signifi-cantly greater than 1.The quantity F = (HDCO / H CO) D CO / H CO = HDCO / H COD CO / HDCO (1)is also tabulated in Table 10. This quantity was introduced byRodgers & Charnley (2002) to discern between gas phase andgrain surface formation mechanisms of deuterated formalde-hyde. Our D-peak F -value of 0.08 is similar to the values 0.03-0.2 found by Roberts & Millar (2007). In the last column ofTable 10 we also list the H CO / CH OH column density ratiosbased on the ALI model results.Bensch et al. (2001) used observations of rare CO isotopo-logues to determine the C / C ratio toward ρ Oph C to 65 ± ρ Oph C cloud core is quite far from the ρ Oph Aposition they belong to the same nearby cloud complex and ourH CO / H CO ratios (Table 10) are in good agreement with theirvalue. Consequently, we will adopt this C / C ratio of 65 alsofor ρ Oph A and the opacity for the C O(3-2) line toward the D-peak can be estimated from the optical depth of the C O(3-2)line of 0.03 (Liseau et al. 2010) as τ [C O(3 − ≈ × . ≈
2. This is then consistent with the discussion in the previous sec-tion on the C O(3-2) opacity corrections toward the D-peak.In Table 11 the SO / SO and SO / SO column densityratios have been listed. As noted in the table, some of theratios are a ff ected by opacity in the main isotopologue andshould be regarded as lower limits. The interstellar S / Sratio (Lucas & Liszt 1998) appears to be close to its solarvalue of 22 (Asplund et al. 2009). We can use this isotopo-logue ratio to estimate the SO abundances from the expres-
10. Bergman et al.: Deuterated formaldehyde in ρ Ophiuchi A
Table 10.
Formaldehyde column density ratios
Source HDCO / H CO D CO / HDCO D CO / H CO H CO / H CO F H CO / CH OHD-peak (0 ′′ , − ′′ ) 0 . ± .
015 1 . ± .
19 0 . ± .
020 54 ±
13 0 . ± .
016 3 . ± . ′′ , + ′′ ) 0 . ± .
006 0 . ± .
08 0 . ± . ±
32 0 . ± .
04 1 . ± . OH-peak ( − ′′ , ′′ ) 0 . ± .
004 0 . ± . . ± . ±
18 0 . ± . Table 11.
Other column density ratios
Source SO / SO SO / SO N D + / N H + ab D-peak 25 ± . ± . ± b . ± . OH-peak > c . ± . . ± . b ± a N (N H + ) estimated from Andr´e et al. (2007) b Not corrected for opacity c σ limit sion 22 N [ SO] / N [H ]. For the D-peak we already have anabundance estimate, made in the previous section, of X [SO] = . × − . In the P1 and CH OH peaks we then get 2 × − and 3 × − , respectively. The P1 value is quite uncertain, asthe SO emission velocities di ff er here from that of H CO (cf. thecase of SO in the previous section), will be lower by a factor5 when using beam averaged column densities. Using the val-ues in Table 9 for the S-peak we find that the SO abundance is X [SO] = × . × / × ≈ × − .The ratio N D + / N H + is also tabulated in Table 11 and herewe use the N H + (1-0) integrated intensity data from Andr´e et al.(2007) which have a beam size of 26 ′′ , i.e. almost identical to theresolution of our N D + (3-2) data. The listed N (N D + ) / N (N H + )ratios have not been compensated for optical depth of the N H + lines.
5. Discussion
As described in Sect. (3.3), the observed line profiles for SO andSO show broader wings in the north-western corner of our map(Fig. 9). Also the H CO spectra exhibit line wings in this region(Fig. 2). The CO red wing emission from the prominent outflowemanating from VLA 1623 extends in this direction while theblue wing emission is extended both in the NW and SE direc-tions (Andr´e et al. 1990). The SO and SO emission peaks veryclose to the IR reflection nebula GSS30 (Castelaz et al. 1985)which is at o ff sets ( − ′′ , + ′′ ) in our maps. The SO and SO abundances (3 × − and 2 × − , respectively) are here clearlyhigher than in other positions. The SO and SO line wings couldvery well be associated with GSS30 rather than the VLA 1623CO outflow. The CH OH-peak at ( − ′′ , ′′ ), on the other hand,could be a result of an interaction with the outflow and a denseclump (cf. Jørgensen et al. 2004). The methanol abundance isabout 5 times higher here compared to its abundance in theD-peak and P1 cores. It appears also to be somewhat warmer( ∼
30 K) than the other cores.
The D-peak position where the deuterium fractionation is high-est coincides with the emission peak (SM1) in the 1.3 mmcontinuum map of Motte et al. (1998). The 1.3 mm flux den-sity, S . , can be converted into an H column density using Fig. 11.
Gas-to-dust ratio map of ρ Oph A calculated as theH column density from C O relative to the H column den-sity from dust. The gas column data are based on the C O(3-2)map of Liseau et al. (2010) and the dust continuum data are fromMotte et al. (1998). The gas column data have not been correctedfor opacity. N (H ) = × cm − ( S . / mJy beam − ) assuming T dust =
20 K, a dust mass opacity of κ . = . g − , and a gas-to-dust mass ratio of 100 (e.g. Andre et al. 1996) over the entiremap (see Eq. (1 ′ ) in Motte et al. 1998). This can be comparedto the H column density derived from the C O(3-2) map ofLiseau et al. (2010). Following the discussion in Sect. (4.4) wehere assume X [C O] = × − and an excitation temperatureof 20 K in all positions. The ratio, N (H ) gas / N (H ) dust , of thesetwo estimates of N (H ) are displayed in Fig. 11. The lowest val-ues of this ratio are around 0.15 and they are found in the centralpart of the cloud. Opacity corrections for C O in this centralpart are of the order 2 (which will increase the ratio to 0.3), seeSect. (4.4) and probably less (closer to 1) further out. This wouldstill mean that the N (H ) gas estimate is a factor of 3 lower thanthat from dust in the central portions of ρ Oph A. Here the majoruncertainty is likely to be related to the adopted C O abundanceand dust mass opacity, κ . , while influences from temperaturechanges are smaller. The low ratios seen could be due to deple-tion of CO and would then correspond to a CO depletion factorof 2-3. Given the uncertainty of the adopted dust parameters anddegree of depletion (ie. C O abundance) we estimate that thefull H column density toward the D-peak, cf. Sect. (4.4), is inthe range (1 − × cm − .The freeze-out of CO on grains (together with low tem-peratures) is thought to increase the deuterium fractionation.
11. Bergman et al.: Deuterated formaldehyde in ρ Ophiuchi A
Although the D-peak at (0 ′′ , − ′′ ) is located in the part wherethe N (H ) gas / N (H ) dust opacity corrected ratio is low ∼ . − . ff ect more at(0 ′′ , − ′′ ) than in any of the other central positions. Hence, theamount of CO depletion (at least as measured by the C O emis-sion and the 1.3 mm continuum data) is not directly responsibleto the elevated deuterium fractionation seen in formaldehyde to-ward the D-peak. However, the D-peak is located within a largerregion that is likely to be depleted and this may have been im-portant in an earlier stage of the evolution of the D-peak source.Interestingly, there is a minimum in the N (H ) gas / N (H ) dust ratio in the SE part of the map (Fig. 11). This minimum co-incides with the secondary N D + peak (N6). This would be inline with the findings that the amount of CO depletion and theN D + / N H + ratio are correlated in starless cores (Crapsi et al.2005). Moreover, the Class 0 source VLA 1623 is also associ-ated with a minimum in the N (H ) gas / N (H ) dust ratio and possi-bly also with a higher N D + / N H + ratio, compare Fig. 10 andFig. 11. A correlation between CO depletion and N D + / N H + ratio has been found by Emprechtinger et al. (2009) toward sev-eral Class 0 objects. Pure gas-phase chemistry can lead to elevated deuterium frac-tionations for H CO. We have here adopted the low-metalicity(or depleted) model of Roue ff et al. (2007) as their warm-core(and un-depleted) model shows low deuterium fractionations.The steady-state model work presented here includes the updatesand modifications of Parise et al. (2009), and we have used it topredict relevant D / H ratios for H CO and N H + . The models areshown in Fig. 12. The di ff erent steady-state D / H model ratios areshown as a function of temperature and have been calculated fortwo di ff erent H densities; 5 × cm − and 1 × cm − . Themodelled ratios show little variation due to the used H density.The observed column density ratios (taken from Table 10 andTable 11) are represented as coloured boxes with size accord-ing to estimated uncertainty. The dashed rectangle indicates theN D + / N H + ratio toward the D-peak with an opacity correction.It is noteworthy that the gas-phase depleted models yield F -values, see Eq. (1), very close to 0.5 for T < ∼
30 K forformaldehyde which are significantly higher than the observedvalues F ≈ .
1, cf. Table 10. The absolute H CO abundances,(3 − × − , for these models are much lower than the ob-served H CO abundances ∼ × − . Our observations point to the fact that D CO is more abundantthan HDCO toward the D-peak. This is, to our knowledge, thefirst source where such case is observed. We note that HDCO andH CO showed a similar anomaly towards L1527 (Parise et al.2006), but even in that case, D CO was less abundant thanHDCO. This anomaly is di ffi cult to explain in terms of pure gas-phase chemistry, as discussed above.We discuss here the possibility that this anomaly stems fromgrain chemistry. Simple grain models which only account forH and D additions to CO to form formaldehyde and methanolcannot account for enrichment of D CO compared to the sta-tistically expected value. However, including the abstractionor exchange reactions of the type HDCO + H → DCO + H ,Rodgers & Charnley (2002) argue that the enrichment is any-way limited to HDCO / H CO > D CO / HDCO (or F >
1, see
Fig. 12.
Gas-phase steady-state models based on the reac-tion network by Roue ff et al. (2007). Shown are the D / H pre-dicted ratios; HDCO / H CO (top), D CO / HDCO (middle), andN D + / N H + (bottom) as function of temperature and for two dif-ferent H densities (5 × cm − and 1 × cm − ). The colouredboxes correspond to the observed ratios for the three cores as in-dicated in the top panel. The size of the rectangles represents theestimated uncertainties.Eq. (1)). Our observational results (with F ≈ . ff erent types of ex-periments have been done (see Watanabe & Kouchi 2008, foran overview). Nagaoka et al. (2005) have exposed CO iceswith H and D atoms (with D / H of 0.1) and have shown thatH CO and CH OH (as well as their deuterated counterparts)are formed. In this experiment, they can reproduce the frac-
12. Bergman et al.: Deuterated formaldehyde in ρ Ophiuchi A tionation of all deuterated methanol molecules observed to-wards IRAS16293 − CO higher than HDCO(Naoki Watanabe, priv. comm.). In a second type of ex-periment, Hidaka et al. (2009) aimed at clarifying the di ff er-ent formation routes by exposing an H CO ice sample withD atoms. Exchanges are observed, showing that deuteriumfractionation occurs very e ffi ciently along the reaction pathH CO → HDCO → D CO. In this case, the observed D COcan become more abundant than HDCO (see their Figure 3).Although a direct extrapolation from these experiments is dif-ficult, our observations may be the definitive evidence that ab-straction and exchange reactions are playing an important rolein grain chemistry. It is not clear if reproducing our observationswill require an increased atomic D / H ratio (from the 0.1 valueused by Nagaoka et al. 2005) incoming on the grains, or if thelonger timescales involved in the ISM chemistry compared tothe laboratory experiments also can play a role. Detailed mod-elling of the complex processes taking place on the grains wouldbe needed to definitely settle this question.There is no known central source towards the D-peak. So,unlike the case of IRAS16293 − ff ective in the SM1 core as compared to the other cores. Hence,the underlying question why the D-peak shows such a high deu-terium fractionation remains unanswered.
6. Conclusions
We have observed a very high degree of deuteration of formalde-hyde towards a core (SM1) in ρ Oph A. In this D-peak,the deuterium fractionation manifests itself by a very highD CO / HDCO ratio of 1 . ± .
19 while the ratio HDCO / H COis 0 . ± . / H ratio). For instance, at the P1position, about 1 ′ (or 0.035 pc) north of the D-peak, the corre-sponding ratios are 0 . ± .
08 and 0 . ± . D + / N H + . The H CO abundance rel-ative to H is estimated to be around 5 × − over the centralcore.The CH OH distribution is clearly di ff erent from that ofH CO and it has its maximum about 45 ′′ to the northwest ofthe deuterium peak. The elevated methanol abundance (by abouta factor of 5 relative to the D-peak) here could be due to an inter-action of the outflow with a dense clump. It would be interestingto study how the deuterated versions of CH OH are distributedin this source (cf. Parise et al. 2006).In order to understand the reason of the very high deutera-tion level observed toward the D-peak we have performed gas-phase chemistry modelling (for a depleted source). By look-ing at the ratio (HDCO / H CO) / (D CO / HDCO) (the F -value,see Eq. (1)) we find that the models result in values around0.5 while the observed values are always around 0.1. Also,the absolute H CO abundances obtained in these models aretoo low by a factor of 100. Hence, the gas-phase chemistry scheme cannot account for the observed H CO abundances anddeuterium ratios. Instead, we advocate that grain chemistry, interms of abstraction and exchange reactions in the reaction chainH CO → HDCO → D CO (Hidaka et al. 2009), can be respon-sible for the very high deuterium fractionations observed in ρ Oph A. However, before being too conclusive about, e.g., therequired atomic D / H ratio, these grain chemistry model resultsneed to be expressed in observable quantities like the H CO F -value. Again, observations of multiply deuterated methanol iso-topologues toward the D-peak in ρ Oph A will provide additionalinsight as they are key ingredients in the grain chemistry scheme.
Acknowledgements.
We are very grateful to Fr´ed´erique Motte for sending us hercontinuum data of ρ Oph and to Naoki Watanabe for sending us unpublished lab-oratory results. Excellent support from the APEX sta ff during the observations isalso greatly appreciated. BP is funded by the Deutsche Forschungsgemeinschaft(DFG) under the Emmy Noether project number PA1692 / References
Andr´e, P., Belloche, A., Motte, F., & Peretto, N. 2007, A&A, 472, 519Andr´e, P., Martin-Pintado, J., Despois, D., & Montmerle, T. 1990, A&A, 236,180Andr´e, P., Ward-Thompson, D., & Barsony, M. 1993, ApJ, 406, 122Andre, P., Ward-Thompson, D., & Motte, F. 1996, A&A, 314, 625Asplund, M., Grevesse, N., Sauval, A. J., & Scott, P. 2009, ARA&A, 47, 481Asvany, O., Schlemmer, S., & Gerlich, D. 2004, ApJ, 617, 685Bachiller, R., Codella, C., Colomer, F., Liechti, S., & Walmsley, C. M. 1998,A&A, 335, 266Bacmann, A., Lefloch, B., Ceccarelli, C., et al. 2003, ApJ, 585, L55Bensch, F., Pak, I., Wouterloot, J. G. A., Klapper, G., & Winnewisser, G. 2001,ApJ, 562, L185Castelaz, M. W., Hackwell, J. A., Grasdalen, G. L., Gehrz, R. D., & Gullixson,C. 1985, ApJ, 290, 261Ceccarelli, C., Castets, A., Loinard, L., Caux, E., & Tielens, A. G. G. M. 1998,A&A, 338, L43Ceccarelli, C., Loinard, L., Castets, A., et al. 2001, A&A, 372, 998Ceccarelli, C., Vastel, C., Tielens, A. G. G. M., et al. 2002, A&A, 381, L17Crapsi, A., Caselli, P., Walmsley, C. M., et al. 2005, ApJ, 619, 379Di Francesco, J., Andr´e, P., & Myers, P. C. 2004, ApJ, 617, 425Di Francesco, J., Andr´e, P., & Myers, P. C. 2009, ApJ, 700, 1994Emprechtinger, M., Caselli, P., Volgenau, N. H., Stutzki, J., & Wiedner, M. C.2009, A&A, 493, 89Frerking, M. A., Langer, W. D., & Wilson, R. W. 1982, ApJ, 262, 590Garrod, R., Park, I. H., Caselli, P., & Herbst, E. 2006, Chemical Evolution of theUniverse, Faraday Discussions, volume 133, 2006, p.51, 133, 51Gerlich, D., Herbst, E., & Roue ff , E. 2002, Planet. Space Sci., 50, 1275Goldsmith, P. F. & Langer, W. D. 1999, ApJ, 517, 209Green, S. 1991, ApJS, 76, 979G¨usten, R., Nyman, L. Å., Schilke, P., et al. 2006, A&A, 454, L13Herbst, E., Adams, N. G., Smith, D., & Defrees, D. J. 1987, ApJ, 312, 351Herbst, E. & Cuppen, H. M. 2006, Proceedings of the National Academy ofScience, 103, 12257Hidaka, H., Watanabe, M., Kouchi, A., & Watanabe, N. 2009, ApJ, 702, 291Jayawardhana, R., Hartmann, L., & Calvet, N. 2001, ApJ, 548, 310Jørgensen, J. K., Hogerheijde, M. R., Blake, G. A., et al. 2004, A&A, 415, 1021Klein, B., Philipp, S. D., Kr¨amer, I., et al. 2006, A&A, 454, L29Larsson, B., Liseau, R., Pagani, L., et al. 2007, A&A, 466, 999Linsky, J. L. 2003, Space Science Reviews, 106, 49Lis, D. C., Roue ff , E., Gerin, M., et al. 2002, ApJ, 571, L55Liseau, R., Larsson, B., Bergman, P., et al. 2010, A&A, 510, A98 + Liseau, R., Larsson, B., Brandeker, A., et al. 2003, A&A, 402, L73Loinard, L., Castets, A., Ceccarelli, C., Caux, E., & Tielens, A. G. G. M. 2001,ApJ, 552, L163Loinard, L., Castets, A., Ceccarelli, C., et al. 2002, Planet. Space Sci., 50, 1205Loinard, L., Castets, A., Ceccarelli, C., et al. 2000, A&A, 359, 1169Loinard, L., Torres, R. M., Mioduszewski, A. J., & Rodr´ıguez, L. F. 2008, ApJ,675, L29Lombardi, M., Lada, C. J., & Alves, J. 2008, A&A, 480, 785Loren, R. B., Wootten, A., & Wilking, B. A. 1990, ApJ, 365, 269Lucas, R. & Liszt, H. 1998, A&A, 337, 246Maercker, M., Sch¨oier, F. L., Olofsson, H., Bergman, P., & Ramstedt, S. 2008,A&A, 479, 779Mangum, J. G. & Wootten, A. 1993, ApJS, 89, 123Maret, S., Ceccarelli, C., Caux, E., et al. 2004, A&A, 416, 577
13. Bergman et al.: Deuterated formaldehyde in ρ Ophiuchi A
Motte, F., Andr´e, P., & Neri, R. 1998, A&A, 336, 150Muders, D., Hafok, H., Wyrowski, F., et al. 2006, A&A, 454, L25M¨uller, H. S. P., Schl¨oder, F., Stutzki, J., & Winnewisser, G. 2005, Journal ofMolecular Structure, 742, 215M¨uller, H. S. P., Thorwirth, S., Roth, D. A., & Winnewisser, G. 2001, A&A, 370,L49Nagaoka, A., Watanabe, N., & Kouchi, A. 2005, ApJ, 624, L29Nummelin, A., Bergman, P., Hjalmarson, Å., et al. 2000, ApJS, 128, 213Pardo, J. R., Cernicharo, J., & Serabyn, E. 2001, IEEE Transactions on Antennasand Propagation, 49, 1683Parise, B., Castets, A., Herbst, E., et al. 2004, A&A, 416, 159Parise, B., Ceccarelli, C., Tielens, A. G. G. M., et al. 2006, A&A, 453, 949Parise, B., Ceccarelli, C., Tielens, A. G. G. M., et al. 2002, A&A, 393, L49Parise, B., Leurini, S., Schilke, P., et al. 2009, A&A, 508, 737Pottage, J. T., Flower, D. R., & Davis, S. L. 2004, MNRAS, 352, 39Roberts, H. & Millar, T. J. 2007, A&A, 471, 849Rodgers, S. D. & Charnley, S. B. 2002, Planet. Space Sci., 50, 1125Roue ff , E., Lis, D. C., van der Tak, F. F. S., Gerin, M., & Goldsmith, P. F. 2005,A&A, 438, 585Roue ff , E., Parise, B., & Herbst, E. 2007, A&A, 464, 245Roue ff , E., Tin´e, S., Coudert, L. H., et al. 2000, A&A, 354, L63Rybicki, G. B. & Hummer, D. G. 1991, A&A, 245, 171Rybicki, G. B. & Hummer, D. G. 1992, A&A, 262, 209Snow, T. P., Destree, J. D., & Welty, D. E. 2008, ApJ, 679, 512Turner, B. E. 1990, ApJ, 362, L29Turner, B. E. 2001, ApJS, 136, 579Vassilev, V., Meledin, D., Lapkin, I., et al. 2008, A&A, 490, 1157Ward-Thompson, D., Robson, E. I., Whittet, D. C. B., et al. 1989, MNRAS, 241,119Watanabe, N. & Kouchi, A. 2008, Progress In Surface Science, 83, 439Wouterloot, J. G. A., Brand, J., & Henkel, C. 2005, A&A, 430, 549
14. Bergman et al.: Deuterated formaldehyde in ρ Ophiuchi A , Online Material p 1
Fig. 4.
Velocity position diagrams of the H CO(3 , − , ), HDCO(4 , − , ), D CO(4 , − , ), and C O(3-2) lines along thedeclination axis at ∆ α = ′′ . The molecule is indicated in each panel. The C O(3-2) data are from Liseau et al. (2010). The verticalscale in each panel indicates the declination o ff set and the horizontal scale is the v LSR velocity. The intensity scale in all four panelsis T mb . . Bergman et al.: Deuterated formaldehyde in ρ Ophiuchi A , Online Material p 2
Fig. 5.
Ortho H CO(3 , − , ) and H CO(3 , − , ) spectra at declination o ff sets − ′′ (bottom, D-peak), 0 ′′ (middle) and + ′′ (top, P1). The o ff sets are relative the center position in Fig. 1. The H CO spectra (in colour) have been scaled up with a factor of10 and appear much more noisy. . Bergman et al.: Deuterated formaldehyde in ρ Ophiuchi A , Online Material p 3
Fig. 9.
SO(5 − ) and SO(6 − ) map spectra toward the ρ Oph A cloud. The34