Online material

Appendix A: Model predictions for the nitrogen chemistry

As discussed in Sect. 5.2, our model best-fit model predicts that N₂H⁺ contributes to about 20–30% of the total charge in both cores. Here we briefly discuss our model prediction for this species, and how it affects our results on the C¹⁸O (1–0) and H¹³CO⁺ (1–0) emission in the cores. Our best fit model for L1498 predicts a N₂H⁺ abundance of 3 × 10^-9, roughly constant throughout the core. This is more than an order of magnitude larger than the abundance derived by Tafalla et al. (2004) in the same object (9 × 10^-11 with respect to H nuclei), from N₂H⁺ (1−0) and (3–2) line observations. This is likely because our model overproduces the N₂ abundance (a precursor of N₂H⁺) which is the main nitrogen reservoir in our simulations. In order to reconcile our model predictions with the N₂H⁺ observations, we have updated the rate of several reactions that are important for nitrogen bearing species. For the reaction rate of N⁺ + H₂, we used the rate of Dislaire et al. (2012), assuming a H₂ ortho-to-para ratio of 10^-3 (the predicted steady-state value at 10 K; Faure et al. 2013). We have also updated the rates for the neutral-neutral reactions that lead to production of N₂ (see Table 3 from Le Gal et al. 2013, and references therein). Finally, we have updated the branching ratio of the N₂H⁺ dissociative recombination (Vigren et al. 2012).

The left panel of Fig. A.1 shows the predicted abundances of several nitrogen bearing species that we obtain for our best-fit model with the updated chemistry network. We find that the new rates have no influence on the N₂H⁺ abundance in the

Fig. A.1 Left panel: predicted abundances for several nitrogen bearing species in L1498 for our best-fit model with an updated chemistry network (see text). The solid lines correspond to gas-phase abundances, and the dashed lines to ice abundances. Right panel: same as in the left panel, but now assuming that most of nitrogen is initially in form of NH3 ices, with an abundance of 8.5% relative to H2O ices.

Open with DEXTER

core center: although the N abundance is increased as a result of the reduced conversion of N to N₂ through neutral reactions, this has little influence on the N₂ abundance, which remains the main nitrogen reservoir. To reduce the predicted N₂ abundance, we have run a model in which we assume that some of the nitrogen is initially in form of NH₃ ices, the rest of it being in atomic form. Assuming an initial NH₃ ice abundance of 5.5% relative to H₂O ices (the average value measured towards young stellar objects by Bottinelli et al. 2010), our model predicts a N₂H⁺ abundance at the core center of 1 × 10^-9, in better agreement with the observations, but still a factor 10 higher. Increasing the initial NH₃ ice abundance to 8.5% (still in agreement with the values measured by Bottinelli et al. that range between 2 and 15%), our model predicts a N₂H⁺ abundance of 2 × 10^-10, in reasonable agreement with the observations (see the right panel of Fig. A.1). In addition, the predicted NH₃ gas phase abundance is 1 × 10^-8. This is also consistent with Tafalla et al. (2004), who measured a para-NH₃ abundance of 7 × 10^-9 (with respect to H nuclei) at the core center, i.e. a total (ortho-NH₃ + para-NH₃) abundance of 1 × 10^-8, assuming an ortho-to-para ratio of 0.7 (Faure et al. 2013). Of course, the model presented here should be compared to observations of other nitrogen bearing species, in order to obtain better constraints of the nitrogen chemistry as a whole (which is out of the scope if the present paper). However, it is important to note that it predicts the same C¹⁸O (1–0) and H¹³CO (1–0) integrated intensities as our best-fit model (within 10%). Our conclusions on the CO and HCO⁺ depletion are therefore independent of the assumptions made for the nitrogen chemistry.

© ESO, 2013