Search for the building block of deuterated ammonia

Coordinator: M. Gerin, D. Lis, E. Roueff, G. Guesten

We propose to use APEX for investigating deuterated ammonia isotopologues (ND2 H and ND3 , and the gas phase intermediate of multiply deuterated ammonia, ND, in the L1689N dense core. The ground state submillimeter ND2 H and ND3 lines have been successfully detected at CSO by our team in this source, showing that the lines are easily detected and the emission might be spatially extended. The spatial distribution of the ND2 H and ND3 emission bears important information on the process of D fractionation as it is extremely sensitive to the local physical and chemical conditions. Furthermore, ND is predicted to be abundant in the pre-stellar core conditions from a gas-phase model, which considers the detailed deuterium substitutes of ammonia and simple compounds species. The interest for searching ND has been renewed by the experimental finding that NH is an important product in the dissociative recombination of N2 H+ . A similar behaviour is expected for the dissociative recombination of N2 D+ leading to ND. ND has an extensive hyperfine structure pattern near 492, a range well suited for APEX. With the 1GHz bandpass, ND and C can be observed simultaneously, C providing a stable frequency reference.

Program is available and data products can be downloaded

Scientific Background:

Deuteration of nitrogen compounds, such as ammonia and N2 H+ , is spectacular in a number of environments including dark clouds such as L134N (Tin� et al. 2000, Roueff et al. 2000), e low mass star forming regions and prestellar cores such as L1689N and Barnard 1 (Loinard et al. 2001, Gerin et al. 2001, Lis et al. 2003). The discussion of the respective contributions of grain and gas-phase processes in the deuteration is active but no definite solution is available yet. The lack of detection of deuterated water in ices toward low mass young stellar objects (YSO) by Dartois et al. (2003) and Parise et al. (2003) suggests that 'another mechanism than pure solid state chemistry may be active to produce very high deuterium enrichment in the gas phase'.
We propose to add further constraints in the chemical studies by mapping ammonia deuterated isotopologue ND2 H and ND3 and by searching for deuterated imidogen, ND, a member of the ammonia family. Gas phase chemistry of imidogen has received recent interest as it has been found that it is produced at a level of 60% from the dissociative recombination of N2 H+ in a storage ring experiment, as reported by Geppert et al. (2004). The prediction that ND is also a major product of the dissociative recombination of N2 D+ is logical. We propose to search for ND in the L1689N dense core where N2 D+ and multiply deuterated ammonia have been found abundant.
In a gas phase model devoted to multideuterated ammonia (Roueff et al. 2005), ND is predicted to be formed at a level higher than 90% from the dissociative recombination of the N2 D+ molecular ion. We thus propose to study the correlation between ND and other N bearing species (NH2 D, ND2 H) as a test of the occurence of gas phase chemistry. We already have good quality N2 D+ data from the CSO for the L1689N core. Roberts et al 2004 discuss the formation of NH and ND with these new branching ratios and predict a maximum fractional abundance of ND of about 10-8 for a density of molecular hydrogen of 104 cm-3 . This high abundance is obtained when accretion is taken into account to lower the gas phase abundance of CO, when the favorable branching ratios are used, and and one assumes that N2 H+ and N2 D+ that land on grains are immediately returned to the gas as N2 and H or D.
The ND radical has a sub-mm spectrum studied by Saito and Goto (1993). The complete hyperfine structure pattern of ND corresponding to the 1�0 rotational transition occurs be- tween 491.9 and 546.2 GHz and is available from the Cologne data base (http://www.ph1.uni- Only the hyperfine components around 492 GHz (see Table 1) are observable from the ground at good submillimeter sites.
For a molecular hydrogen column density of N(H2 ) 1023 cm-2 , the maximum column density for ND we can expect is therefore about 1015 cm-2 which would produce saturated emission lines of TA 0.2 - 2 K depending on the excitation temperature. The line opacity will be easily obtained from the ratio of the most intense hyperfine components. From our previous observations at the CSO, the ND2 H and ND3 line intensities are about 0.6 K and 0.15 K respectively.

Time estimate
The best sources to search for ND are those where we have previously detected ND2 H and ND3 emission. Out of these, L1689N (RA(1950)=16:29:27.6,DEC(1950)=-24:22:33) can be easily observed from APEX in July. Under good weather conditions (225 0.05), we expect a system temperature of 1000 K at 492 GHz. Under these conditions, the expected rms in a 125 kHz channel (using 256 MHz and 2048 channels, and observing in Position-Switching mode), is 120 mK in 1 hour integration time (ON+OFF source). Such an integration will take approximately 2 hours of time, including overheads. Since velocity shifts will be required to confirm sideband assignment of any lines that are detected, we expect that 4 hours of observations are required for obtaining a secure detection of ND. Observations of the fine structure transition of neutral carbon at 492 GHz, and of the ground state transition of NH2 D can be obtained simultaneously as the line frequencies fit into the 4GHz IF. The carbon line is very close to ND, and should be observed as a frequency reference. In addition, we propose to use marginal weather (the sky is fairly transparent at 310 and 335 GHz) for mapping ND2 H (335.5 GHz) and ND3 (310 GHz) using the APEX-2a receiver. Our previous attemps using the CSO have shown that the ND3 line is marginally extended. These observations are aimed at determining the true extent of the high deuteration region, and at locating the position of the maximum deuteration. We wish to know whether the peak of D fractionation is associated with the maximum of the dust submillimeter emission of if a spatial offset can be measured. Indeed, by comparing the gas and dust distribution, we want to study the influence of molecular depletion due to freezing on the chemistry. Molecule freezing leads to differences in the spatial distribution of gas and dust, which can be readily identified on line maps. Assuming a system temperature of 400 K, and a frequency resolution of 125 kHz, ND2 H lines can be detected in 15 minutes, while 1 hour is requested for the ND3 ground state transition. We propose to spend 4 hours for mapping the 335 ND2 H line, and then 5 hours for sampling the same area in the weaker ND3 line. As a whole, we request 4hours with FLASH and 9hours with APEX-2a


Dartois, E. et al., 2003, A&A 399, 1009
Geppert, W. et al., 2004, ApJ 609, 459
Gerin, M. et al., 2001, ApJ 551, L193
Lis, D.C. et al., 2003, ApJ 571, L55
Lis, D.C., Gerin, M., & Roueff, E. 2005, in Dusty and Molecular Universe, ed. A. Wilson (ESA SP-577), in press
Loinard L. et al., 2001, ApJ 552 L163
Parise, B. et al., 2003, A&A 410, 897
Roberts H. et al., 2004, A&A 424, 905
Roueff, E. et al., 2000, A&A 354, L63
Roueff, E. et al. 2005, A&A, in press (astro-ph0504445)
Tine, S. et al., 2000, A&A 356, 1039