diffuse clouds along the lineâofâsight toward several stars in the solar neighborhood (i.e. within â¼ 90 pc). ... first such direct measurement in a dense molecular cloud without the complicating effects of chemical fractionation. 2. ..... Teams
Sep 14, 2005 - high mass star forming region Orion-KL and the first detection of ... acetone may be prevalent in other star forming regions such as Orion-KL.
Your thesaurus codes are: 08(02.19.1; 09.01.1; 09.09.1 Orion Peak 1; 09.13.2; 12.03.3; 13.09.4). ASTRONOMY. AND .... cessed Data (SPD) stage within the SWS Interactive Anal- ysis (IA) system. Between the SPD and Auto Analysis Re- sult (AAR) stages, a
Sep 14, 2005 - mass or low mass star forming region despite being structurally similar to dimethyl ether. ((CH3)2O). Conversely, (CH3)2O has been detected in both low mass star forming regions including IRAS 16293-2422 (Cazaux ... email: [email protected]
Dec 10, 2008 - phous silicate SiâO stretching absorption feature show considerable variations from the local diffuse interstellar medium (ISM) to Galactic ...
ABSTRACT. The Lyman and Werner band systems of deuterated molecular hydrogen (HD) occur in the far UV range below 1200 Ã . The high sensitivity of the FUSE mission can give access, at moderate resolution, to hot stars shining through translucent clou
Jun 10, 2015 - (3) The wavelength range corresponding to the given frequencies is 44 Âµmâ188 Âµm. Favre et al. 2011). The positions of IRc1, IRc2, IRc3, IRc4,. IRc5, IRc6, and IRc7 infrared emission peaks are taken from. Shuping et al. (2004). The
Sep 25, 2007 - [email protected], .... 2. Observations. We observed HD 209458b over a period of 8.1 hours on UT 2005 Nov. 27, spanning a single secondary eclipse, using the Infrared Array Camera (IRAC) ... images dominated by a single bright st
Nov 5, 2015 - found a trend that the ratio shows higher value at edges of the cloud. In particular the ratio at the north-eastern edge of ... The estimation shows that the gas associated with the edge of the cloud is warm (Tkin â¼ 60 K), dense (nH2
May 2, 2012 - (l-v) maps of [C ii], H i and CO emissions in the vicinity of the SâC tangency: (a) â (c) (l-v) maps of [C ii] intensity at b =0.0â¦,. Â±0.5â¦ and Â±1.0â¦ ... The color bars shown on top represent the [C ii] and H i map intensiti
Abstract. We present the spectral energy distribution. (SED) between 10 and 200 Âµm obtained for the prototype. Narrow-Line X-Ray Galaxy NGC 7582 with ISOPHOT, the photometer on board the Infrared Space Observatory. The emission is spatially extended
Mar 14, 2011 - HST to calculate a model of the interstellar cloud toward. HD 37903. ... (2009) to de- termine the TOP = 170 K/ln(9N0/N1)=68 Â± 7 K gas âki-.
Mar 14, 2011 - Î½â=0 vibrational level observed by the FUSE satellite. The T01 temperature should not be interpreted as a rota- tional temperature, but rather as a temperature of thermal equilibrium between the ortho and para H2. The ortho to para
Mar 14, 2011 - Î½â=0 vibrational level observed by the FUSE satellite. The T01 temperature should not be interpreted as a rota- tional temperature, but rather as a temperature of thermal equilibrium between the ortho and para H2. The ortho to para
Mar 14, 2011 - The presence of both FUSE (Far Ultraviolet Spec- troscopic Explorer) and HST STIS spectra for HD. 147888 (Ï Oph D) gives us a rare opportunity to mea- sure column density of H2 energy levels excited by fluorescence (Î½ > 0) and also b
Feb 26, 2001 - Pasadena, CA 91125; email: [email protected] Lynne A. Hillenbrand. California Institute of Technology, Department of Astronomy, MS ...
Sep 2, 2006 - A study of the metastable (J, K)=(18,18) line of ammonia. (NH3) is presented aiming at better defining the physical prop- erties of molecular gas in extreme environments. The spectra were collected with the Effelsberg 100-m telescope an
2MASS observing mode by scanning the telescope in declination to cover tiles of ...... to thank the 2MASS Observatory Staff and Data Management Team for acquiring .... http://www.ipac.caltech.edu/2mass/releases/second/doc/explsup.html.
To estimate the signal to noise ratio of the photometry, Figure 2 plots the observed photometric ... limitations. Historically, variable stars were identified through visual inspection of image data or light curves. This approach has the advantage th
Feb 25, 2012 - âO2-struggleâ was recently reviewed by Goldsmith et al. (2011). .... 2010) with continuous frequency coverage from 480 to. 1250 GHz (Î»Î» 624 ...
During the first flight, an area of 30â². Ã15. â² in Orion A was mapped. These observations extend over a larger area than previous observations, the map is.
Lockman Window is just below the detection limit of the observation. For HD 93521 there is also strong evidence for a high velocity HÎ± emission component at â90 km sâ1, but no corresponding feature in the H I spectrum. This sightline is very nea
Apr 21, 2015 - Based on the five detected C2H transitions, a single component rotational diagram analysis gives a rotation temperature of â¼64 K and ... Cloud and an Hii region illuminated by the Trapezium cluster. The FUV radiation ..... computed u
Apr 3, 2014 - stars using Kepler data. Both studies demonstrated that low-mass / small planets are much more frequent than massive / large planets. As the probability of observing a planet transiting in front of its host stars decreases with orbital
ISO–LWS detection of the 112 µm HD J=1→0 line toward the Orion Bar1
arXiv:astro-ph/9901391v1 28 Jan 1999
Christopher M. Wright2,6 , Ewine F. van Dishoeck2 , Pierre Cox3 , Sunil D. Sidher4 and Martin F. Kessler5
Based on observations with ISO, an ESA project with instruments funded by ESA
Member States (especially the PI countries: France, Germany, The Netherlands and the United Kingdom) and with participation of ISAS and NASA. 2
Leiden Observatory, P.O. Box 9513, 2300 RA Leiden, The Netherlands
Institut d’Astrophysique Spatiale, Universit´e de Paris, 91405 Orsay, France
Rutherford Appleton Laboratory, Chilton, Didcot, Oxon OX11 0QX, UK
ISO Science Operations Centre, Astrophysics Division of ESA, P.O. Box 50727, E–28080
Villafranca/Madrid, Spain 6
School of Physics, University College, ADFA, UNSW, Canberra ACT 2600, Australia
We report the first detection outside of the solar system of the lowest pure rotational J=1→0 transition of the HD molecule at 112 µm. The detection was made toward the Orion Bar using the Fabry-P´erot of the Long Wavelength Spectrometer (LWS) on board the Infrared Space Observatory (ISO). The line appears in emission with an integrated flux of (0.93±0.17)×10−19 W cm−2 in the LWS beam, implying a beam–averaged column density in the v=0, J=1 state of (1.2±0.2)×1017 cm−2 . Assuming LTE excitation, the total HD column density is (2.9±0.8)×1017 cm−2 for temperatures between 85 and 300 K. Combined with the total warm H2 column density of ∼ (1.5 − 3.0)×1022 cm−2 derived from either the H2 pure rotational lines, C18 O observations or dust continuum emission, the implied HD abundance, HD/H2 , ranges from 0.7 × 10−5 to 2.6 × 10−5 , with a preferred value of (2.0 ± 0.6) × 10−5 . The corresponding deuterium abundance of [D]/[H]=(1.0 ± 0.3) × 10−5 is compared with recent values derived from ultraviolet absorption line observations of atomic H i and D i in interstellar clouds in the solar neighborhood and in Orion.
Subject headings: ISM: abundances : molecules : individual : (Orion Bar) — infrared: ISM : lines and bands
The deuterium abundance is one of the most sensitive probes of the baryon density in the early universe (Wilson & Rood 1994), since it is thought that all the deuterium was produced in the Big Bang, with no subsequent production via nuclear reactions in stars. Conversely, it is destroyed in stellar interiors, so that current measurements provide only a lower limit on the primordial deuterium abundance. Thus, an observed [D]/[H] ratio may also be used as a measure of galactic chemical evolution. However, previous ground and airplane based attempts to measure the [D]/[H] ratio in a variety of interstellar sources have met with several problems, such as the very low intrinsic strength of the D i 92 cm line (Heiles et al. 1993) or chemical fractionation and line saturation effects in molecules (Penzias et al. 1977). The most successful measurements so far have been through satellite ultraviolet absorption line observations of the Lyα lines of atomic H i and D i through diffuse clouds along the line–of–sight toward several stars in the solar neighborhood (i.e. within ∼ 90 pc). These give [D]/[H]=1.6×10−5 with a typical uncertainty of 15%, a value so far independent of the line–of–sight (e.g., Dring et al. 1997, Piskunov et al. 1997). Hydrogen deuteride, HD, has also been detected in local diffuse molecular clouds through ultraviolet absorption line observations (e.g., Spitzer et al. 1973), but its abundance is low in these tenuous clouds because of rapid photodissociation of the molecule. In contrast, virtually all of the deuterium is expected to be contained within HD in dense, warm molecular clouds. Measurements of the v=0–0 J=1→0 R(0) 112.072 µm line can potentially provide a direct and accurate determination of the HD abundance, and thereby also the deuterium abundance, in such clouds. A previous attempt to observe this line was made by Watson et al. (1985) toward the Orion KL region, but resulted only in an upper limit of 1 × 10−18 W cm−2 . In this paper we report the first detection of the 112 µm HD J=1→0 line outside of the solar system.
–4– The detection was made toward the Orion Bar, a warm Photon Dominated Region (PDR), and the line is observed to be in emission. PDRs have the advantage over other source types, such as embedded young stellar objects, in that they have a large amount of warm molecular gas at sufficiently high temperatures to excite the HD J=1→0 112 µm line. They do not have cold surrounding material which could absorb and cancel part of the emission. The main drawback is that HD is rapidly photodissociated at the edge of the PDR, so that its abundance does not become large until AV ≈ 2 − 3 mag into the cloud (e.g., Jansen et al. 1995a). The physical and chemical structure of the Orion Bar is, however, well understood from a variety of molecular line observations (e.g., Hogerheijde et al. 1995, Jansen et al. 1995b, van der Werf et al. 1996). It has a particularly high total H2 column density due to its edge–on geometry, facilitating the detection of HD. Our successful observation provides a determination of the HD abundance, as well as a probe of the chemistry of molecular clouds. When compared with observations of molecular hydrogen, it also allows a constraint to be placed on the cosmologically important [D]/[H] ratio, the first such direct measurement in a dense molecular cloud without the complicating effects of chemical fractionation.
Observations and data reduction
The Orion Bar was observed during revolution 823 using the LWS04 Fabry–P´erot (FP) mode of the Long Wavelength Spectrometer (LWS, Clegg et al. 1996) on board the Infrared Space Observatory (ISO, Kessler et al. 1996). The rest frequency of the HD J=1→0 line is 2,674,986.66±0.15 MHz (Evenson et al. 1988) corresponding to a vacuum wavelength of 112.072506±0.000007 µm. The observed coordinates were RA = 05h 35m 20.3s and DEC = −05◦ 25′ 20′′ (J2000), a position which closely corresponds to the peak column density of molecular gas (e.g., Burton et al. 1990; Parmar, Lacy & Achtermann 1991; Hogerheijde et
–5– al. 1995). The observations consist of 120 separate LWS FP scans centered on the frequency of the HD J=1→0 line, with 7 spectral elements on either side of the line. The data were taken with the LW2 detector in fast scanning mode, with 4 spectral samples per resolution element and 45.5 s per sample. The total on–target–time was 3913 s. The FWHM beam size at 122 µm, the central wavelength of detector LW2, in the spacecraft Y–Z directions is 78′′ × 75′′ with a systematic uncertainty of up to 20% (Swinyard et al. 1998). The resolving power is of order 9500 or ∼ 30 km s−1 (Clegg, Heske & Trams 1994), and the wavelength calibration accuracy is good to a third of a resolution element, or 10 km s−1 (Trams et al. 1998). A full range 43–197 µm LWS01 grating spectrum, consisting of 5 scans with 0.7 s integration time per step and a spectral sampling interval of 2, was obtained during the same revolution for calibration purposes (fringes, continuum level). Initial data reduction was carried out using the ISO–LWS Off Line Processing (OLP) software, version 7.0, up to the Auto Analysis Result stage. Further data processing, such as removal of bad data points, flat–fielding, sigma clipping and co–adding, was performed using software in the ISO Spectral Analysis Package, and the LWS and Short Wavelength Spectrometer (SWS) Interactive Analysis (IA) packages. Of particular note is the dark current subtraction and grating position correction. The relatively high flux of the Bar results in some straylight leakage through the misaligned Fabry–P´erot plates during the dark current measurements, meaning that they are not true dark currents. Therefore, the dark current was iteratively modified such that the resultant continuum flux was equal to that obtained in the LWS01 observation. Using the LWS IA tool FP PROC (Swinyard et al. 1998, Sidher et al., in preparation), the subsequent product was corrected for a grating positioning problem, which introduces a spurious slope in the spectrum. By adopting these procedures, our photometric accuracy is estimated to be 30% or better (Burgdorf et al.
–6– 1997, Swinyard et al. 1998).
Figure 1 displays the resulting spectrum after co–addition of all FP scans. An emission line is clearly visible at vLSR ≈ +13 km s−1 , close to the expected vLSR ≈ +10 km s−1 known for the Bar (e.g., Hogerheijde et al. 1995). The feature is unresolved, with an observed FWHM of ∼ 30 km s−1 , i.e., near the resolving power of the LWS and implying an intrinsic FWHM less than this, again consistent with previous millimeter observations which show ∆V ≈ 2 km s−1 . The observed integrated line flux is (7.3±1.3)×10−20 W cm−2 , obtained by fitting a Gaussian to the line profile, and a first order polynomial plus sinusoid due to fringing (see below) to the baseline. The uncertainty is statistical and represents the range of values obtained using different fitting procedures. When divided by the LW2 beam size, 1.08 × 10−7 sr, the inferred surface brightness is I = (6.76 ± 1.20) × 10−6 erg s−1 cm−2 sr−1 . As noted in the LWS Data Users Manual (Trams et al. 1998), the LWS OLP software corrects the data of detector LW2 by a factor of 0.68 for the effective aperture of the instrument, assuming that the source is point–like and located at the aperture centre. This is not the case for the Orion Bar, which is extended in both its gas and dust emission. The Data Users Manual gives a correction factor of 0.87/0.68 at 100–120 µm which must be applied to observed fluxes for extended sources. The resulting best estimate of the flux is (9.3 ± 1.7) × 10−20 W cm−2 and of the surface brightness is (8.7 ± 1.5) × 10−6 erg s−1 cm−2 sr−1 . This corresponds to an antenna temperature of ∼0.22 K in the LWS beam for ∆V ≈2 km s−1 . The beam averaged column density in the HD v=0, J=1 state is obtained from N0,1 = (4πI1→0 )/(A1→0 hν1→0 ), where I1→0 is the observed surface brightness. Using A1→0 = 5.12 × 10−8 s−1 (Abgrall, Roueff & Viala 1982), N0,1 =(1.20±0.21)×1017 cm−2 is
–7– found. The small A–value assures that the optically thin relation is valid. Although we are confident that the observed feature is real and corresponds to the 112 µm J=1→0 line of HD, there are a few noteworthy points to be made. First, it is well known that the LWS grating spectral responsivity calibration file at 112 µm has a spurious “absorption” feature resulting from HD absorption in the calibration source Uranus. However, this does not affect the FP spectrum, where the resolving power is a factor of ∼ 50 greater than the grating. Across the width of our FP spectrum the calibration file is virtually flat. Second, a similarly sensitive HD 112 µm observation was made toward the Galactic Center, where no corresponding emission feature is seen (Wright et al., in preparation). Another possible source of spurious emission is from leakage into the FP of adjacent spectral orders. The FP free spectral range at 112 µm is 1.12 µm, and the LWS Observer’s Manual (Clegg, Heske & Trams 1994) discusses possible contamination. The grating acts as an order sorter for the FP, and its resolving power was specifically set such that contamination is avoided. In any case, at submillimeter wavelengths the Orion Bar is very poor in lines compared with spectroscopically rich sources such as Orion–KL (Hogerheijde et al. 1995), and there are no other line-rich sources in the LWS beam. The only strong lines expected in the LWS wavelength range are the high–J transitions from
There is no such transition, nor any other common low–lying (i.e. ≤ 300 K) molecular feature, known to be coincident with HD J=1→0 within 0.01 µm, or to be a possible contaminant from adjacent spectral orders. The LWS–FP data are also affected by fringing, in a similar way to LWS grating data of extended sources. However, the single high frequency fringing component apparent in the data has an essentially constant period of about 0.0095 cm−1 (Sidher et al. in preparation), which makes it easier to correct. The effect appears to occur as a result of interference
–8– within the instrument.
Analysis and Discussion
In the following, we will discuss the derivation of the column densities of HD and H2 and compare the HD and deuterium abundances in the Orion Bar with those found in other regions.
The HD column density
Because of the limited sensitivity of the LWS detector SW2, observations of the 56.23 µm 0–0 J=2→1 R(1) line of HD were not feasible, so that no direct information on the excitation of the molecule is available. However, the HD excitation can be readily computed assuming LTE: N(HD) = [N0,1 Q(T )/g0,1] exp(E0,1 /kT ), where g0,1 is the statistical weight of the v=0, J=1 level, and Q(T ) is the partition function. Hogerheijde et al. (1995) find gas kinetic temperatures in the Bar of 85±30 K from observations of a variety of molecules, whereas other analyses give values up to at least 300 K (Parmar et al. 1991, see §4.2). The corresponding total HD column densities range from (2.6±0.5)×1017 cm−2 at 115–200 K to (3.1±0.6)×1017 cm−2 at 85 and 300 K. Including the temperature range as an additional uncertainty, our best estimate of the total HD column density is (2.9±0.8)×1017 cm−2 . The estimated hydrogen densities of the Orion Bar, such as 104 to 105 cm−3 for the so–called interclump medium, or ≥ 106 cm−3 for the high density clumps (e.g., Hogerheijde et al. 1995), are all above the critical density of the J=1 level. For the higher HD levels, the departures from LTE are small for these conditions.
The HD abundance
To derive the HD abundance with respect to H2 , the total H2 column density, N(H2 ), is needed in addition to N(HD). The most direct method is to use observations of the pure rotational lines of H2 itself. Parmar, Lacy & Achtermann (1991) present data of the 17.0348 µm 0–0 J=3→1 S(1) and 12.2786 µm 0–0 J=4→2 S(2) lines of H2 in a 10′′ × 2′′ slit at positions covering a 10′′ × 16′′ region in the Bar, and within our ∼ 76′′ aperture. Our observed position is within a few arcseconds of their position 3, the point of peak emission, while their positions 1, 2, 4 and 5 fall well within our aperture. The averaged line fluxes for positions 1–5 (covering 10′′ × 10′′ ) give a mean excitation temperature Tex between levels J=3 and 4 of 482 K and a column density of 9.9×1020 cm−2 . However, this is unlikely to represent the total H2 column density, since there is a strong temperature gradient throughout the PDR with a significant amount of cooler gas present, as indeed is shown in other molecular tracers. Several methods may be used to determine the total amount of gas. First, the clumpy PDR models of Burton, Hollenbach & Tielens (1990, 1992) can be used to estimate the expected Tex between levels J=2 and 3. For the conditions appropriate to the interclump medium of the Orion Bar, where Burton et al. (1992) state that the bulk of the H2 emission originates, Tex =180–260 K. Using the observed column density in the J=3 level, we can then extrapolate to the J=2 level to determine its column density and thereby the total warm H2 column density, assuming that the ortho–para ratio is in LTE (Sternberg & Neufeld 1999). For Tex =180 K, N(H2 )=1.5×1022 cm−2 is found, while for Tex =260 K, N(H2 )=4.4×1021 cm−2 . We note that recent observations with the ISO–SWS at the same position in the Bar, including the 28.2188 µm 0–0 J=2→0 S(0) line, show that Tex between the J=2 and J=3 levels is 150–175 K, and that the total warm H2 column density is ∼1.5×1022 cm−2 , with a statistical uncertainty of ∼10% (e.g. Rosenthal et al., in preparation).
– 10 – Using N(H2 )=1.5×1022 cm−2 , the HD abundance, N(HD)/N(H2 ) (denoted as HD/H2 ), is (2.0±0.6)×10−5 . Our estimate of N(H2 ) may be an underestimate since the H2 S(0) line may still be weighted toward warmer gas than the HD J=1→0 line, given that their upper level energies are ∼ 510 and 128 K above ground, respectively. This would lead to an overestimate of HD/H2 . This may be counter–balanced by an intrinsic underestimate due to photodissociation of HD at the PDR edge and the fact that the LWS beam is significantly larger than the apertures used for the H2 observations (14′′ × 27′′ at 12 and 17 µm and 20′′ × 27′′ at 28 µm for ISO–SWS; see also below). N(H2 ) has also been estimated from observations of CO and its isotopic forms to be between 3 and 6.5×1022 cm−2 , varying with precise position, line and beam size (e.g., Graf et al. 1990, Hogerheijde et al. 1995, van der Werf et al. 1996). Using the molecular line maps and the C18 O 2→1 cut across the Bar by Hogerheijde et al. (1995), the beam dilution in the LWS aperture is estimated to be a factor of ∼ 0.4 − 0.5 compared with their peak emission. This results in a LWS beam–averaged H2 column density of ∼ 3 × 1022 cm−2 if an H2 /C18 O ratio of 5 × 106 is used. The corresponding HD/H2 is (1.0±0.3)×10−5 . This value refers to HD with respect to the total amount of warm and cold H2 . However, because the HD J=1 level lies at 128 K above ground it is only excited efficiently in gas with temperatures above ∼ 30 K. In addition, HD is photodissociated at the edge of the PDR, so that the region over which H2 and HD are coexistent is less. These effects can be taken into account by comparison with detailed physical and chemical models of the Orion Bar, such as developed by Jansen et al. (1995a,b). The models show that HD becomes abundant 2 visual magnitudes deeper into the cloud than H2 , and that the temperature stays above 30 K up to depths of 5–6 mag. These values need to be convolved with the edge–on geometry of the Bar as outlined by Jansen et al. (1995b). At the positions included in the LWS beam, the outer, warm layers are enhanced significantly, and it is estimated
– 11 – that at least 50% of the total column density contains HD at sufficiently high temperatures. Thus, the relevant N(H2 ) in the LWS beam for comparison with HD is ∼ 1.5 × 1022 cm−2 , consistent with that determined from the warm H2 above. Similar values are obtained from 13
CO 6→5 data from Lis (private communication) averaged over a 70′′ beam and from the
CO lines detected in our LWS01 spectrum. The corresponding HD abundance is
again (2.0±0.6)×10−5 . Another method for determining the HD abundance is to utilise the dust continuum emission as a tracer of molecular hydrogen. Our LWS01 spectrum gives a dust temperature of 60 K and a 112 µm optical depth of 0.05, obtained from an optically thin fit to the data. Inserting these values into Eq. (4) of Watson et al. (1985), HD/H2 =(1.5±0.3)×10−5 is obtained. Alternatively, the 400 µm observations of Keene, Hildebrand & Whitcomb (1982) can be used along with Eq. (9) in Table 1 of Hildebrand (1983) to find N(H2 )=2.6×1022 cm−2 and HD/H2 =(1.1±0.3)×10−5 , using a 400 µm optical depth of 4.4×10−3 . These methods assume thermal equilibrium between the gas and dust, however, and are uncertain by a factor of 2 due to uncertainty in the dust opacity. If the gas and dust are in thermal equilibrium at ∼60 K, N(HD) increases to (4.6 ± 0.8) × 1017 cm−2 and HD/H2 =(1.8 ± 0.3) × 10−5. Although each of the above methods to derive the HD abundance have their own intrinsic uncertainties, our observations constrain the range of HD/H2 in the Orion Bar to be between 0.7×10−5 and 2.6×10−5 , not including the uncertainty due to the LWS photometric calibration and beam size. Our preferred value is (2.0 ± 0.6) × 10−5 , the value obtained using the H2 pure rotational lines or corrected CO isotopic emission as tracers of N(H2 ).
– 12 – 4.2.1. Comparison with other determinations of the HD abundance It is interesting to compare our result with those previously obtained toward other source types. The HD/H2 ratios obtained from ultraviolet absorption line observations through diffuse clouds are typically ∼ 10−6 (e.g. Spitzer et al. 1973, Snow 1975, Wright & Morton 1979). Such low values are explained through preferential ultraviolet dissociation of HD, since HD does not self–shield. Although gas–phase HD formation is more rapid than that of H2 due to the H+ + D → D+ + H and D+ + H2 → HD + H+ reactions, the net effect is still a low HD/H2 ratio. Future observations with the Far Ultraviolet Space Explorer (FUSE) will be able to extend these ultraviolet observations to thicker, translucent clouds in the solar neighborhood, where more of the deuterium is in HD. Recently, Bertoldi et al. (1999) measured the 19.4 µm HD 0–0 J=6→5 R(5) line in the Orion shock using the ISO–SWS, and found an HD abundance of (9.0 ± 3.5) × 10−6, corrected for non–LTE population. In partially dissociative shocks, the HD abundance may be reduced relative to H2 by ∼40% due to the HD + H → H2 + D reaction. The HD abundance has also been determined for the giant planets using the SWS and LWS on board ISO to measure the 37.7 µm 0–0 J=3→2 R(2) and 56.2 µm J=2→1 R(1) lines, respectively. These yield (preliminary) values of (2.6–5.8)×10−5 for Jupiter and (3.0–8.0)×10−5 for Saturn (Lellouch 1999, and references therein). Our value is at the low end of these ranges, perhaps implying some Galactic evolution of the HD (and thereby D) abundance over the 5 billion year lifetime of our solar system.
The deuterium abundance
Our derived HD/H2 ratios refer specifically to the regions of the PDR where deuterium is in molecular form. Thus, the amount of atomic deuterium can be neglected and
– 13 – [D]/[H]≈ 0.5× HD/H2 . Our derived deuterium abundance ranges from 0.35 to 1.30 ×10−5 , with a preferred value of (1.0 ± 0.3) × 10−5 . This result is marginally outside the error estimate (∼ 15%) of the value for the solar neighborhood of 1.6×10−5 (Piskunov et al. 1997). However, it is close to the value of (0.76 ± 0.29) × 10−5 found for the Orion shock −5 −5 −5 by Bertoldi et al. (1999), and (0.74+0.19 and (1.4+0.5 −0.13 ) × 10 , (0.65 ± 0.3) × 10 −1.0 ) × 10
for the lines–of–sight to the stars δ, ǫ and ι Orionis by Jenkins et al. (1999) and Laurent et al. (1979), respectively. Additional ultraviolet and far-infrared observations are needed to determine whether true variations in the deuterium abundance exist within Orion and between Orion and the local interstellar medium. Our value is however definitely below that of (3.9±1.0)×10−5 found through radio observations of the 21 cm H i and 92 cm D i hyperfine transitions in a region of reduced star formation activity in the outer Galaxy by Chengalur, Braun & Burton (1997). This provides some evidence of a Galactic deuterium abundance gradient, and is consistent with its destruction in stars. Better determinations of [D]/[H] in the Galactic Center will be important to further constrain the gradient across the Galaxy. The successful detection of the HD 112 µm J=1→0 line by ISO in a reasonable integration time implies that the next generation of infrared airborne observatories, such as the Stratospheric Observatory for Far–Infrared Astronomy (SOFIA) and the Far–InfraRed and Submillimeter Space Telescope (FIRST), will be able to detect this line in a variety of sources, and possibly determine a [D]/[H] gradient across the Galaxy. At the high spectral resolution of the SOFIA and FIRST instruments, other sources such as dense molecular cloud cores may be more favorable for HD searches, because of their higher column densities compared with PDRs. The largest uncertainty in the [D]/[H] analysis remains the determination of the corresponding H2 column density.
This work was supported by NFRA/NWO grant 781-76-015. We thank Frank Bertoldi,
– 14 – John Black, Francois Boulanger, David Jansen, Ed Jenkins, Darek Lis, Tom Phillips, Dirk Rosenthal, Ralf Timmermann, Laurent Verstraete and Jonas Zmuidzinas for their help and stimulating discussions, and the referee John Lacy for instructive comments. CMW acknowledges support of an ARC Research Fellowship. The ISO Spectral Analysis Package (ISAP) is a joint development by the LWS and SWS Instrument Teams and Data Centers. Contributing institutes are CESR, IAS, IPAC, MPE, RAL and SRON.
– 15 – REFERENCES Abgrall, H., Roueff, E. & Viala, Y. 1982, A&A Supp. Ser., 50, 505 Bertoldi, F., Timmermann, R., Rosenthal, D., Drapatz, S. & Wright. C.M. 1999, A&A, submitted Burgdorf, M., Clegg, P., Ewart, D. 1997, in First ISO Workshop on Analytical Spectroscopy, ESA-SP 419 (Noordwijk: ESTEC), eds. A.M. Heras et al., p. 51. Burton, M.G., Geballe, T.R., Brand, P.W.J.L. & Moorhouse, A. 1990, ApJ, 352, 625 Burton, M.G., Hollenbach, D.J. & Tielens, A.G.G.M. 1990, ApJ, 365, 620 Burton, M.G., Hollenbach, D.J. & Tielens, A.G.G.M. 1992, ApJ, 399, 563 Chengalur, J.N., Braun, R. & Burton, W.B. 1997, A&A, 318, L35 Clegg, P.E., Heske, A. & Trams, N.R. 1994, LWS Observers Manual, LWS/PEC/2038.01, Issue 1.0, 6th March 1994, Clegg, P.E., Ade, P.A.R., Armand, C. et al. 1996, A&A, 315, L38 Dring, A.R., Linsky, J., Murthy, J. et al. 1997, ApJ, 488, 760 Evenson, K.M., Jennings, D.A., Brown, J.M., Zink, L.R., Leopold, K.R., Vanek, M.D. & Nolt, I.G. 1988, ApJ, 330, L135 Graf, U.U., Genzel, R., Harris, A.I. et al. 1990, ApJ, 358, L49 Heiles, C., McCullough, P.R. & Glassgold, A.E. 1993, ApJS, 89, 271 Hildebrand, R.H. 1983, Q. Jl. R. astr. Soc., 24, 267 Hogerheijde, M.R., Jansen, D.J. & van Dishoeck, E.F. 1995, A&A, 294, 792 Jansen, D.J., Spaans, M., Hogerheijde, M.R. & van Dishoeck, E.F. 1995b, A&A, 303, 541 Jansen, D.J., van Dishoeck, E.F., Black, J.H., Spaans, M., & Sosin, C. 1995a, A&A, 302, 223
– 16 – Jenkins, E.B., Tripp, T.M., Wozniak, P.R., Sofia, U.J. & Sonneborn, G. 1999, ApJ, submitted Keene, J., Hildebrand, R.H. & Whitcomb, S.E. 1982, ApJ, 252 L11 Kessler, M.F., Steinz, J.A., Anderegg, M.E. et al. 1996, A&A, 315 L27 Laurent, C., Vidal–Madjar, A. & York, D. 1979, ApJ, 229, 923 Lellouch, E. 1999, in The Universe as seen by ISO, ESA-SP (Noordwijk: ESTEC), eds. P. Cox et al., in press Parmar, P., Lacy, J.H. & Achtermann, J.M. 1991, ApJ, 372, L25 Penzias, A.A., Wannier, P.G., Wilson, R.W. & Linke, R.A. 1977, ApJ, 211, 108 Piskunov, N., Wood, B.E., Linsky, J.L., Dempsey, R.C. & Ayres, T.R. 1997, ApJ, 474, 315 Snow, T.P. 1975, ApJ, 201, L21 Spitzer, L., Drake, J.F., Jenkins, E.B. et al. 1973, ApJ, 181, L116 Sternberg, A. & Neufeld, D.A. 1999, ApJ, in press Swinyard, B.M., Burgdorf, M.J., Clegg, P.E., Davis, G.R., Griffin, M.J., Gry, C., Leeks, S.J., Lim, T.L., Pezzuto, S., & Tommassi, E. 1998, in Infrared Astronomical Instrumentation, SPIE 3354, in press Trams, N. et al. 1998, ISO LWS Data Users Manual, SAI/95–219/Dc, Issue 5.0, June 1 1998 van der Werf, P.J., Stutzki, A., Sternberg, A. & Krabbe, A. 1996, A&A, 313, 633 Watson, D.M., Genzel, R., Townes, C.H. & Storey, J.W.V. 1985, ApJ, 298, 316 Wilson, T.L. & Rood, R.T. 1994, ARAA, 32, 191
This manuscript was prepared with the AAS LATEX macros v4.0.
– 17 – Fig. 1.— The ISO–LWS Fabry–P´erot spectrum toward the Orion Bar of the HD 0–0 J=1→0 112.0725 µm line. The presented spectrum has not been corrected for the effective aperture of the instrument. The error bar represents the typical standard deviation in individual data points. The baseline has been corrected for fringing.