Deuterium and 15 N fractionation in N 2 H+ during the formation of a Sun-like star

Deuterium and 15 N fractionation in N 2 H+ during the formation of a Sun-like star Abstract Although chemical models predict that the deuterium fractionation in N2H+ is a good evolutionary tracer in the star formation process, the fractionation of nitrogen is still a poorly understood process. Recent models have questioned the similar evolutionary trend expected for the two fractionation mechanisms in N2H+, based on a classical scenario in which ion-neutral reactions occurring in cold gas should have caused an enhancement of the abundance of N2D+, 15NNH+, and N15NH+. In the framework of the ASAI IRAM-30m large program, we have investigated the fractionation of deuterium and 15N in N2H+ in the best known representatives of the different evolutionary stages of the Sun-like star formation process. The goal is to ultimately confirm (or deny) the classical ‘ion-neutral reactions’ scenario that predicts a similar trend for D and 15N fractionation. We do not find any evolutionary trend of the 14N/15N ratio from both the 15NNH+ and N15NH+ isotopologues. Therefore, our findings confirm that, during the formation of a Sun-like star, the core evolution is irrelevant in the fractionation of 15N. The independence of the 14N/15N ratio with time, found also in high-mass star-forming cores, indicates that the enrichment in 15N revealed in comets and protoplanetary discs is unlikely to happen at core scales. Nevertheless, we have firmly confirmed the evolutionary trend expected for the H/D ratio, with the N2H+/N2D+ ratio decreasing before the pre-stellar core phase, and increasing monotonically during the protostellar phase. We have also confirmed clearly that the two fractionation mechanisms are not related. stars: formation, ISM: clouds, ISM: molecules, radio lines: ISM 1 INTRODUCTION Observations and chemical models agree that the process of deuterium enrichment in N2H+ is a robust evolutionary indicator in the star formation process. The root ion-neutral reaction that forms N2D+ is (e.g. Millar, Bennett & Herbst 1989; Ceccarelli et al. 2014)   \begin{eqnarray} {\rm H_3^+ + HD \leftrightarrow H_2D^+ + H_2 + \Delta E} \,, \end{eqnarray} (1)followed by the reaction of H2D+ with N2 to give N2D+. Reaction (1) is exothermic by ΔE ∼ 230 K, and hence it is fast only from the left- to right-hand side below ∼20 K. At these low temperatures, and if CO is highly depleted, the abundance of H2D+ is thus boosted, and so are the species directly formed from it, like N2D+. This implies that the N2D+/N2H+ ratio is expected to decrease from the pre-stellar to the proto-stellar phase, when the nascent protostar begins to heat up its surrounding material (e.g. Caselli 2002). In fact, observations of pre-stellar cores and young protostars have shown D/H ratios in N2H+ of the order of 0.1–0.01 (e.g. Bacmann et al. 2003; Crapsi et al. 2005; Emprechtinger et al. 2009), i.e. several orders of magnitude higher than the cosmic D/H elemental abundance (∼10−5, e.g. Linsky et al. 2006), and a decrease with core evolution in both low- and high-mass star-forming cores (Emprechtinger et al. 2009; Fontani et al. 2011; Ceccarelli et al. 2014; Bianchi et al. 2017). This trend is followed by other molecules formed by H2D+, such as DNC (e.g. Fontani et al. 2014; Gerner et al. 2015) and DCO+ (e.g. Emprechtinger et al. 2009; Gerner et al. 2015). Despite the gradual decrease with time expected after the protostar birth, the huge D enrichment in the early cold phase is able to explain the enhanced D/H ratio measured in comets (∼10−4, e.g. Hartogh et al. 2011; Altwegg et al. 2015), with respect to the cosmic value (Cleeves et al. 2014). For this reason, the D/H ratio is believed to be an excellent chemical tool to link the different phases of the Solar system formation, from the earliest pre-stellar stage to the formation of the planets and other Solar system bodies (Ceccarelli et al. 2014). Similarly to D, 15N is enriched in pristine Solar system bodies such as comets (14N/15N ∼150, Manfroid et al. 2009; Shinnaka et al. 2016) and carbonaceous condrites (∼50, Bonal et al. 2009) by factors ∼2 to ∼10 with respect to the protosolar nebula value (∼441, Marty et al. 2010). As for N2D+, the 15NNH+ and N15NH+ abundances were thought to be enhanced through ion-neutral reactions occuring in cold gas (Terzieva & Herbst 2000):   \begin{eqnarray} {\rm ^{15}N + ^{14}\!N_2H^+ \leftrightarrow \ ^{14}N + ^{15}\!NNH^+ + 36 K\,,} \end{eqnarray} (2)  \begin{eqnarray} {\rm ^{15}N + ^{14}\!N_2H^+ \leftrightarrow \ ^{14}N + N^{15}NH^+ + 28 K\,.} \end{eqnarray} (3)However, recent chemical models (Roueff, Loison & Hickson 2015) have questioned the efficiencies of these reactions, because of the presence of energy barriers higher than previously predicted, which make significant fractionation unlikely to occur even in the earliest cold phases of the protostellar evolution. The few observational findings obtained to date make the puzzle even more intriguing: the 14N/15N measurements in N2H+ in high-mass star-forming cores obtained by Fontani et al. (2015) do not show an evolutionary dependence, in agreement with Roueff et al. (2015) predictions, but the same observations indicate a huge dispersion of the values (14N/15N ∼ 100–1000), which cannot be reproduced by the models. The models also do not predict the high 14N/15N ratio found in the low-mass pre-stellar core L1544 (∼1000, Bizzocchi et al. 2013). In this paper, we show the first evolutionary study of the combined 14N/15N and D/H ratio in N2H+ in the best representatives of the main evolutionary stages of the Sun-like star formation process. The primary goal of the study is to understand whether the two ratios are linked and, consequently, whether (also) the 14N/15N is an evolution-dependent parameter and can be considered a chemical link between the earliest and the most evolved stages during the formation of Sun-like stars. 2 SAMPLE AND OBSERVATIONS The targets are extracted from the Large Program ASAI1 (Lefloch et al., in preparation) and represent the best known precursors of a Sun-like star, in the main phases of its formation: the pre-stellar core L1544, the Class 0 protostars IRAS 4A, the Class 0/Class I protostar L1527, and the Class I protostar SVS13-A. We have analysed also the Class 0 source IRAS 16293 using the data of the TIMASSS survey (Caux et al. 2011). Together with this ‘evolutionary’ sample, we have studied other sources to provide additional relevant information about the chemical evolution in a Sun-like star-forming environment: the protocluster OMC–2 FIR4, the chemically rich protostellar shock L1157–B1, and the intermediate-mass Class 0 protostar CepE. In the following, we give a brief description of each target (for more details, please see López-Sepulcre et al. 2015): L1544: L1544 is a starless core in the Taurus molecular cloud complex (d ∼140 pc, Cernicharo & Guélin 1987). It is considered the prototypical pre-stellar core on the verge of the gravitational collapse (Caselli et al. 2012, and references therein). L1544 is characterized by a nucleus with a high H2 density peak (2 × 106 cm−3) and low temperature (∼7 K), surrounded by a lower density envelope undergoing extended infall (Caselli et al. 2012). For these reasons, in the central region, CO is depleted by a factor ∼10 (Caselli et al. 1999) and the deuterium fractionation is high, like in the interiors of dark clouds, although differentiated chemical processes can take place in the external layers (Caselli et al. 1999; Vastel et al. 2014). IRAS 16293: IRAS 16293 is a well known Class 0 protostar located in the small L1689N molecular cloud in the ρ Ophiuchus complex at a distance of 120 pc (Loinard et al. 2008). It is a system with two main components, IRAS16293 A and IRAS16293 B, separated by ∼5 arcsec, characterized by a strong chemical differentiation (Bisschop et al. 2008; Jørgensen et al. 2011, 2012). It is the first source where a hot corino has been discovered, (Ceccarelli et al. 2000; Cazaux et al. 2003; Bottinelli et al. 2004), and a well-studied astrochemical laboratory thanks to its richness in complex organic molecules, and its high deuterium fractionation, for example, in formaldehyde (Ceccarelli et al. 1998, 2001), methanol (Parise et al. 2004), methyl formate (Demyk et al. 2010), and water (Coutens et al. 2012). IRAS 4A: IRAS 4A is the second hot corino ever discovered (Bottinelli et al. 2007), and is a binary source in the Perseus molecular complex (d ∼ 235 pc, Hirota et al. 2007; Hirota et al. 2011); it is composed of two Class 0 objects separated by 1.8 srcsec: IRAS4 A1 and IRAS4 A2 (e.g. Looney, Mundy & Welch 2000). The nature of IRAS4 A1 and IRAS4 A2 has been discussed in several papers (e.g. Persson, Jørgensen & van Dishoeck 2012; Santangelo et al. 2015; Taquet et al. 2015; De Simone et al. 2017; López-Sepulcre et al. 2017). In particular, Santangelo et al. (2015) have concluded that A1 is brigther than A2 in the millimeter continuum, but only A2 is associated with a hot-corino. L1527: L1527 is a dark cloud in the Taurus molecular complex (d ∼ 140 pc), with a heavily obscured IRAS source (IRAS 04368+2557) located at the core centre, classified as a borderline Class 0/Class I object according to Andre, Ward-Thompson & Barsony (2000). This source is considered as a prototypical warm carbon-chain chemistry source (WCCC, Sakai et al. 2008, 2010), and high-angular resolution observations indicate the presence of a compact, infalling, and rotating envelope (Sakai et al. 2016, 2017) surrounding the protostar. SVS13-A: SVS13-A is a part of the multiple system NGC1333-SVS13 (distance 235 pc, Hirota et al. 2007) in the Perseus complex, dominated by three millimetre sources identified by interferometric observations (Bachiller et al. 1998; Looney et al. 2000; Tobin et al. 2016; Lefavre et al. 2017, and references therein): SVS13-A, SVS13-B separated by ∼15 arcsec and associated with two different evolutionary stages, and SVS13-C, ∼20 arcsec away from SVS13-A. Although SVS13-A is still deeply embedded in a large-scale envelope (Lefloch et al. 1998), it is considered a young Class I protostar because of its extended outflow (>0.07 pc) and its low ratio between submillimeter and bolometric luminosity (∼0.8 per cent), becoming an almost unique laboratory to investigate how deuteration and fractionation change from the Class 0 to the Class I phases (Chen, Launhardt & Henning 2009; Tobin et al. 2016; Bianchi et. al 2017; Lefloch et al., in preparation). In the analysis of SVS13-A, care needs to be taken in the fit of the lines because of the contamination of core IRAS 4A at velocity ∼7 km s−1 in the off position (Santangelo et al. 2015), while that of SVS13-A is at about 8–9 km s−1, which results in an artificial absorption near the emission lines of interest. This contamination has affected only the border of the blue side of the lines (see the figures in the on-line Appendix-A); in the case of the N2H+ (1–0) line, we used the observation carried out in Position Switching mode, and not in Wobbler Switching mode, to avoid this contamination. OMC–2 FIR4: OMC–2 FIR4 is a protocluster of low- and intermediate-mass protostars at a distance of 420 pc (e.g. Hirota et al. 2007). Its ∼30 M⊙ envelope extends to about 104 au (Crimier et al. 2009) and contains several clumps, probably a cluster of protostars (Shimajiri et al. 2008; López-Sepulcre et al. 2013), which makes it one of the best and closest environment analogues of the one in which the Sun was born (e.g. Ceccarelli et al. 2014; Fontani et al. 2017). In fact, growing evidence indicates that our Sun was born in a crowded environment, forming together with many other protostars, including very likely more massive objects (e.g. Adams 2010; Taquet et al. 2016). The fractionation of nitrogen has been studied by Kahane et al. (2018) in several molecular species (HCN, HNC, HC3N, CN), who derived 14N/15N ∼ 290, regardless of the molecule used, and in remarkable agreement with the present-day local interstellar 14N/15N. L1157–B1: L1157–B1 is a chemically rich bipolar outflow (Bachiller et al. 1997, 2001) driven by a low-mass (∼4 L⊙) Class 0 protostar (L1157-mm, at a distance of 250 pc, Looney, Tobin & Kwon 2007), and associated with several blue and red-shifted shocks at different ages (Gueth, Guilloteau & Bachiller 1996, 1998). It may be regarded as the ideal laboratory for observing the effects of shocks on the gas chemistry (Bachiller et al. 2001; Codella et al. 2010; Lefloch et al. 2010) and on isotopic fractionation (Fontani et al. 2014; Busquet et al. 2017). The analysis of the N2H+ emission performed by Codella et al. (2013) showed that the N2H+ emission detected towards the L1157-B1 shock can be considered a fossil record of the pre-shock phase, when the gas density was ∼104 cm−3. In other words, N2H+ is sampling the chemical–physical conditions of the quiescient gas evolved for more than 104 yr, before the gas was perturbed by the protostellar shock induced by the L1157 protostellar activity. CepE: CepE is an intermediate-mass Class 0 protostar (L = 100 L⊙, Lefloch, Eisloeffel & Lazareff 1996) in the Cepheus OB3 association at a distance of 730 pc (Sargent 1977), which drives an exceptionally powerful and luminous molecular outflow. Both the protostar and the outflow have been intensively studied at mm and IR wavelengths (Lefloch et al. 1996; Moro-Martin et al. 2001; Noriega-Crespo et al. 2004). The observations have been performed during several runs between 2011 and 2016 with the IRAM 30-m telescope near Pico Veleta (Spain) in the context of the Astrochemical Surveys at IRAM (ASAI) Large Program, using the broad-band EMIR receivers E090, E150, and E230. These were carried out in Wobbler Switching mode, with a throw of 3 arcmin, in order to ensure a flat baseline across the spectral bandwidth observed. The instrumental setup was decided according to the sources: L1544 was observed in the band 72 to 115 GHz using the Fast Fourier Transform Spectrometer (FTS) in its 50 kHz resolution mode; about Protostars and Outflow Shocks, the 3 mm (72–116,GHz) and 2 mm (126–170 GHz) bands were observed at the same time. The 1.3 m (200–272 GHz) band was covered observing LSB and USB simultaneously, using the FTS in its 200 kHz resolution mode (Lefloch et al., in preparation.). The data were reduced with the class software of the gildas2 package. Calibration uncertainties are ∼10 per cent at 3 mm and ∼20 per cent at shorter wavelengths. All the spectra are displayed in antenna temperature units. The main spectroscopic parameters of the transitions observed, and the main observational parameters, are summarized in Table 1. The spectroscopic parameters have been taken from the Cologne Database for Molecular Spectroscopy (CDMS).3 Table 1. Spectroscopic and observational parameters of the observed lines: line rest frequency, ν, energy of the upper level, Eu, line strength, $$S \mu _{\rm ij}^2$$, Einstein coefficient of spontaneous emission, Aij, main beam angular size, ΘMB, and main beam efficiency, Beff. The spectral parameters have been taken from the CDMS (Müller et al. 2001; Müller et al. 2005). Transition  ν  Eu  $$S \mu _{\rm ij}^2$$  Aij  ΘMB  Beff    GHz  K  D2  s−1  arcsec    N2H+ (1–0)  93.173  4.5  104  4 × 10−5  26  0.80  (3–2)  279.512  27  312  1 × 10−3  9  0.49  N2D+ (1–0)  77.109  4  104  2 × 10−5  32  0.83  (2–1)  154.217  11  208  2 × 10−4  16  0.71  (3–2)  231.322  22  312  7 × 10−4  11  0.59  15NNH+ (1–0)  90.264  4  35  3 × 10−5  27  0.80  (3–2)  270.784  26  104  1 × 10−3  9  0.51  N15NH+ (1–0)  91.206  4  35  3 × 10−5  27  0.80  (3–2)  273.609  26  104  1 × 10−3  9  0.51  Transition  ν  Eu  $$S \mu _{\rm ij}^2$$  Aij  ΘMB  Beff    GHz  K  D2  s−1  arcsec    N2H+ (1–0)  93.173  4.5  104  4 × 10−5  26  0.80  (3–2)  279.512  27  312  1 × 10−3  9  0.49  N2D+ (1–0)  77.109  4  104  2 × 10−5  32  0.83  (2–1)  154.217  11  208  2 × 10−4  16  0.71  (3–2)  231.322  22  312  7 × 10−4  11  0.59  15NNH+ (1–0)  90.264  4  35  3 × 10−5  27  0.80  (3–2)  270.784  26  104  1 × 10−3  9  0.51  N15NH+ (1–0)  91.206  4  35  3 × 10−5  27  0.80  (3–2)  273.609  26  104  1 × 10−3  9  0.51  View Large The distances in our sample vary from 120 pc (for the closest object, IRAS 16293) to ∼420 pc (for the farthest, OMC–2 FIR4). Therefore, the linear scales probed differ by a factor ∼3.5 at all frequencies, although in reality the difference is less than a factor 2 for all objects but OMC–2 FIR4. However, most of our calculations are based on the (1–0) transition (see Section 3.2), whose angular beam size corresponds to a linear scale that ranges from a minimum of ∼3000 au for IRAS 16293, to a maximum of ∼11 000 au for OMC–2 FIR4. According to the description of the sources presented above, these scales probe the lower-density envelope in all objects, including OMC–2 FIR4, which is a protocluster intrinsically larger than the other cores, and thus the isotopic fractions that we will derive in Section 3.2 are always associated with the envelope of the sources, and not with their inner nuclei. 3 RESULTS 3.1 Detected lines and fitting procedure We have detected at least one transition of N2H+ and N2D+ in all the sources of our sample. In Tables 2 and 3, we list the lines observed and detected in each source. The N2H+ (1–0) line is detected in all targets, while N2D+ was observed and detected either in the (1–0) and/or in the (2–1) and (3–2) lines. Transitions of N15NH+ have also been detected in all sources, except L1157–B1 and L1527, while 15NNH+ lines were detected only towards L1544, IRAS 4A, Cep E, and OMC–2 FIR4. All the transitions are split into multiple hyperfine components due to the coupling of the 14N nuclear spin with the angular momentum of the molecule. Therefore, we have attempted to fit all lines taking the hyperfine structure into account. The method assumes that the components are Gaussians with the same line width separated in frequency according to the laboratory value, and have all the same excitation temperature. A detailed description of the method and of the output parameters is given in the CLASS manual.4 For the N2H+ and N2D+ lines, the method has given generally good results, while it was not appropriate for several 15NNH+ and N15NH+ lines, mainly because of the faintness of the secondary components. In particular, for two lines, N15NH+ (1–0) in SVS13-A and N2D+ (1–0) in L1157–B1, the main component was clearly detected, but one or more satellites were undetected. Therefore, we have fitted the main component with a single Gaussian, and then calculated the expected contribution of the undetected hyperfine components assuming LTE and optically thin conditions. The final integrated area for these lines is the sum of these contributions (see Tables 2 and 3). In Columns 3–8 of Tables 2 and 3, we show the line parameters derived with the methods mentioned above. The Tables also give the 1σ rms level in the spectrum (Column 9), used for non-detected lines to compute the upper limits on the column densities. Transitions not shown in the Tables were not observed. The spectra of all lines detected are shown in the on-line Appendix A. Table 2. Line parameters obtained applying the hyperfine fit procedure (when possible). In these cases, the ouput parameters are: Tant · τ, the peak velocity of the main hyperfine component, Vpeak, the line width at half-maximum, FWHM, the opacity of the main component, τmain, and the integrated line area, Area. This latter was computed by integrating the spectrum under the channels with signal above the 3σ rms level. The associated uncertainty is computed from the propagation of errors: $$\sigma \times \Delta v_{\rm res} \times \sqrt{N}$$, where σ is the 1σ rms is the spectrum, Δvres is the spectral resolution, and N is the number of channels with signal. The lines without Tant · τ and τmain could not be fit with the hyperfine method, so that the associated Vpeak, FWHM, and Area have been derived from a Gaussian fit. In these cases, the associated uncertainty on the Area represents the 1σ rms given by the fit procedure. The upper limit on the Area for the undetected lines has been calculated assuming a Gaussian line with intensity peak equal to the 3 σ rms level in the spectrum, and FWHM as listed in Column 5 (for details, see end of Section 3.2). The last column lists the 1σ rms level in the spectrum. Molecule  Transition  Tant · τ  Vpeak  FWHM  τmain  Area  rms      K  km s−1  km s−1  –  (K km s−1)  (mK)  L1544  N2H+  1–0  0.9(0.1)  6.0(0.1)  0.3(0.1)  0.5(0.1)  5.11(0.04)  4  N2D+  1–0  4.9(0.1)  5.8(0.2)  0.4(0.6)  2.7(0.1)  1.89(0.02)  3  15NNH+  1–0  0.06(0.02)  6.9(0.1)  0.4(0.1)  1.9(1.8)  0.023(0.004)  1  N15NH+  1–0  0.10(0.04)  6.2(0.1)  0.3(0.1)  3.7(2.9)  0.03(0.01)  2  IRAS4 A  N2H+  1–0  0.3(0.1)  5.9(0.1)  1.2(0.1)  0.1(0.01)  9.10(0.03)  6  N2D+  1–0  1.0(0.1)  7.2(0.1)  1.2(0.1)  0.1(0.01)  1.58(0.02)  3    2–1  2.2(0.01)  7.0(0.1)  1.2(0.1)  1.2(0.1)  2.31(0.04)  6    3–2  1.5(0.1)  6.0(0.1)  1.0(0.1)  1.3(0.2)  1.17(0.05)  8  15NNH+  1–0  –  7.3(0.1)  1.33(0.2)  –  0.017(0.002)  1    3–2  –  6.8(0.1)  0.8(0.3)  –  0.02(0.02)  7  N15NH+  1–0  0.03(0.01)  6.2(0.6)  1.3(2.1)  0.4(0.1)  0.043(0.001)  0.3    3–2  –  6.6(0.1)  0.3(0.1)  –  0.009(0.002)  9  L1527  N2H+  1–0  0.2(0.1)  5.6(0.1)  1.2(0.1)  0.1(0.01)  3.47(0.02)  4  N2D+a  2–1  –  –  5.1b  –  ≤0.12  8.3  15NNH+a  1–0  –  –  3.3b  –  ≤0.04  4  N15NH+a  1–0  –  –  4.5b  –  ≤0.05  4  IRAS 16293  N2H+  1–0  0.3(0.1)  3.0(0.1)  1.7(0.1)  0.1(0.01)  11.78(0.07)  16  N2D+  2–1  2.7(0.1)  4.0(0.1)  1.0(0.1)  2.0(s0.2)  2.17(0.01)  3    3–2  –  4.5(0.2)  2.8(0.4)  0.7(0.1)  0.72(0.02)  10  15NNH+a  1–0  –  –  5.6b  –  ≤0.11  6.4  N15NH+  3–2  –  5.6(0.2)  2.8(0.5)  –  0.13(0.02)  5  SVS13-A  N2H+  1–0  0.6(0.1)  8.2(0.1)  0.8(0.1)  0.1(0.003)  11.40(0.01)  3  N2D+  2–1  0.4(0.1)  8.3(0.1)  0.6(0.1)  0.1(0.01)  0.21(0.02)  3    3–2  –  8.9(0.1)  0.8(0.1)  –  0.17(0.02)  7  15NNH+a  1–0  –  –  1.9b  –  ≤0.015  2.5  N15NH+  1–0  –  7.6(0.1)  1.4(0.3)  –  0.020(0.009)(c)  1.5  Molecule  Transition  Tant · τ  Vpeak  FWHM  τmain  Area  rms      K  km s−1  km s−1  –  (K km s−1)  (mK)  L1544  N2H+  1–0  0.9(0.1)  6.0(0.1)  0.3(0.1)  0.5(0.1)  5.11(0.04)  4  N2D+  1–0  4.9(0.1)  5.8(0.2)  0.4(0.6)  2.7(0.1)  1.89(0.02)  3  15NNH+  1–0  0.06(0.02)  6.9(0.1)  0.4(0.1)  1.9(1.8)  0.023(0.004)  1  N15NH+  1–0  0.10(0.04)  6.2(0.1)  0.3(0.1)  3.7(2.9)  0.03(0.01)  2  IRAS4 A  N2H+  1–0  0.3(0.1)  5.9(0.1)  1.2(0.1)  0.1(0.01)  9.10(0.03)  6  N2D+  1–0  1.0(0.1)  7.2(0.1)  1.2(0.1)  0.1(0.01)  1.58(0.02)  3    2–1  2.2(0.01)  7.0(0.1)  1.2(0.1)  1.2(0.1)  2.31(0.04)  6    3–2  1.5(0.1)  6.0(0.1)  1.0(0.1)  1.3(0.2)  1.17(0.05)  8  15NNH+  1–0  –  7.3(0.1)  1.33(0.2)  –  0.017(0.002)  1    3–2  –  6.8(0.1)  0.8(0.3)  –  0.02(0.02)  7  N15NH+  1–0  0.03(0.01)  6.2(0.6)  1.3(2.1)  0.4(0.1)  0.043(0.001)  0.3    3–2  –  6.6(0.1)  0.3(0.1)  –  0.009(0.002)  9  L1527  N2H+  1–0  0.2(0.1)  5.6(0.1)  1.2(0.1)  0.1(0.01)  3.47(0.02)  4  N2D+a  2–1  –  –  5.1b  –  ≤0.12  8.3  15NNH+a  1–0  –  –  3.3b  –  ≤0.04  4  N15NH+a  1–0  –  –  4.5b  –  ≤0.05  4  IRAS 16293  N2H+  1–0  0.3(0.1)  3.0(0.1)  1.7(0.1)  0.1(0.01)  11.78(0.07)  16  N2D+  2–1  2.7(0.1)  4.0(0.1)  1.0(0.1)  2.0(s0.2)  2.17(0.01)  3    3–2  –  4.5(0.2)  2.8(0.4)  0.7(0.1)  0.72(0.02)  10  15NNH+a  1–0  –  –  5.6b  –  ≤0.11  6.4  N15NH+  3–2  –  5.6(0.2)  2.8(0.5)  –  0.13(0.02)  5  SVS13-A  N2H+  1–0  0.6(0.1)  8.2(0.1)  0.8(0.1)  0.1(0.003)  11.40(0.01)  3  N2D+  2–1  0.4(0.1)  8.3(0.1)  0.6(0.1)  0.1(0.01)  0.21(0.02)  3    3–2  –  8.9(0.1)  0.8(0.1)  –  0.17(0.02)  7  15NNH+a  1–0  –  –  1.9b  –  ≤0.015  2.5  N15NH+  1–0  –  7.6(0.1)  1.4(0.3)  –  0.020(0.009)(c)  1.5  Notes. aTransition undetected: the column density given in Table 4 is an upper limit; bAssumed value for FWHM (see Section 3.2). View Large Table 3. Same as Table 2 for the ‘non-evolutionary’ sources. Molecule  Transition  Tant · τ  Vpeak  FWHM  τmain  Area  rms      K  km s−1  km s−1  -  (K km s−1)  (mK)  L1157–B1  N2H+  1–0  0.004(0.001)  1.4(0.1)  3.0(0.1)  0.1(0.01)  0.28(0.01)  2  N2D+  1–0  –  3.3(0.2)  1.8(0.5)  –  0.025(0.014)c  1  15NNH+a  1–0  –  –  7.5b  –  ≤0.02  1  N15NH+a  1–0  –  –  9.6b  –  ≤0.03  1  OMC–2 FIR4  N2H+  1–0  1.4(0.1)  10.2(0.1)  1.3(0.1)  0.1(0.01)  39.40(0.02)  6  N2D+  2–1  0.8(0.1)  11.0(0.1)  1.3(0.1)  0.3(0.6)  1.24(0.08)  14    3–2  –  10.7(0.02)  0.9(0.03)  –  0.80(0.08)  21  15NNH+  1–0  0.14(0.1)  11.1(0.1)  1.2(0.1)  1.2(0.9)  0.15(0.01)  2  N15NH+  1–0  0.14(0.1)  10.3(0.1)  1.3(0.1)  0.8(0.8)  0.19(0.01)  1  CepE  N2H+  1–0  0.2(0.1)  −13.6(0.1)  1.2(0.1)  0.1(0.01)  4.17(0.02)  4    3–2  –  −11.7(0.1)  4.5(0.1)  –  4.20(0.02)  4  N2D+  1–0  0.3(0.1)  −11.1(0.1)  1.3(0.1)  1.5(1.0)  0.29(0.02)  3    2–1  1.3(0.1)  −11.2(0.1)  0.8(0.1)  2.4(0.3)  0.79(0.02)  4    3–2  –  −12.1(0.3)  4.6(0.7)  –  0.28(0.08)  3  15NNH+  1–0  –  −11.9(0.4)  1.8(1.4)  –  0.010(0.004)  2  N15NH+  1–0  –  −13.2(0.6)  4.3(1.2)  –  0.070(0.001)  0.3  Molecule  Transition  Tant · τ  Vpeak  FWHM  τmain  Area  rms      K  km s−1  km s−1  -  (K km s−1)  (mK)  L1157–B1  N2H+  1–0  0.004(0.001)  1.4(0.1)  3.0(0.1)  0.1(0.01)  0.28(0.01)  2  N2D+  1–0  –  3.3(0.2)  1.8(0.5)  –  0.025(0.014)c  1  15NNH+a  1–0  –  –  7.5b  –  ≤0.02  1  N15NH+a  1–0  –  –  9.6b  –  ≤0.03  1  OMC–2 FIR4  N2H+  1–0  1.4(0.1)  10.2(0.1)  1.3(0.1)  0.1(0.01)  39.40(0.02)  6  N2D+  2–1  0.8(0.1)  11.0(0.1)  1.3(0.1)  0.3(0.6)  1.24(0.08)  14    3–2  –  10.7(0.02)  0.9(0.03)  –  0.80(0.08)  21  15NNH+  1–0  0.14(0.1)  11.1(0.1)  1.2(0.1)  1.2(0.9)  0.15(0.01)  2  N15NH+  1–0  0.14(0.1)  10.3(0.1)  1.3(0.1)  0.8(0.8)  0.19(0.01)  1  CepE  N2H+  1–0  0.2(0.1)  −13.6(0.1)  1.2(0.1)  0.1(0.01)  4.17(0.02)  4    3–2  –  −11.7(0.1)  4.5(0.1)  –  4.20(0.02)  4  N2D+  1–0  0.3(0.1)  −11.1(0.1)  1.3(0.1)  1.5(1.0)  0.29(0.02)  3    2–1  1.3(0.1)  −11.2(0.1)  0.8(0.1)  2.4(0.3)  0.79(0.02)  4    3–2  –  −12.1(0.3)  4.6(0.7)  –  0.28(0.08)  3  15NNH+  1–0  –  −11.9(0.4)  1.8(1.4)  –  0.010(0.004)  2  N15NH+  1–0  –  −13.2(0.6)  4.3(1.2)  –  0.070(0.001)  0.3  Notes. aTransition undetected: the column density given in Table 4 is an upper limit; bAssumed value for FWHM (see the text); cOne or more hyperfine components are under the 3σ rms level; the Area has been computed by summing the integrated intensity of the detected components and the expected area of the undetected components (assuming LTE and optically thin conditions). View Large 3.2 Derivation of the total column densities From the line parameters, we have calculated the total column densities N of N2H+, N2D+, 15NNH+, and N15NH+. Because different sources have been detected in different lines, and, above all, not all targets have been observed in the same lines, to be consistent, we have derived N from the (1–0) line for all species, when possible. This approach was suggested by the fact that the (1–0) line is detected in almost all species and targets, and because the isotopic ratio derived by dividing column densities obtained from the same transition are independent from the excitation temperature. The method has been used successfully for L1544, IRAS 4A, and L1157–B1. In the other targets, this method cannot be applied because the (1–0) line was either not observed or not detected in all molecules. For example, in IRAS 16293, the (1–0) transition is detected in N2H+ and 15NNH+ but neither in N2D+ nor in N15NH+. For these two molecules, we have detected the (2–1) and (3–2) lines (see Table 2). In this case, the column densities were derived adopting the transition with the best signal-to-noise ratio in each molecule, assuming a wide (but reasonable) range of excitation temperatures of 5–30 K. For L1544, we have assumed a more realistic temperature range of 5–15 K. The fit to the hyperfine structure gives an estimate of the optical depth of the main component, τmain. For lines with τmain ≤ 0.5, we have computed N from the total line integrated area according to equation (A4) in Caselli et al. (2002), valid for optically thin lines. The integrated areas have been computed from the spectra in antenna temperature units, and then converted into main beam brightness temperature units using the main beam efficiencies reported on the IRAM–30m website.5 The optically thin approach is good for almost all the detections (see Tables 2 and 3). Some N2D+ lines have τmain > 0.5, but poorly constrained (Δτmain/τmain ≥ 0.3), so that we have derived the column densities using the optically thin approach as well. For the few optically thick lines and with opacity well-constrained, the column density has been calculated by multiplying the value derived in optically thin approximation by the factor τ/(1 − e−τ) (where τ is the total opacity of the line). Finally, for the few 15NNH+ transitions in which the secondary components are not detected, we have used only the integrated intensity of the main one, assuming optically thin conditions, hence in these cases (see Tables 2 and 3) the derived N has to be regarded as lower limits. The column densities should be corrected for the filling factor to compare them in a consistent way, but in most cases we do not know the emission size of the transitions observed, therefore we do not have a direct estimate of the filling factor. However, as shown in Castets et al. (2001), in IRAS 16293 the source size in N2H+ (1–0) can be as extended as ∼30 arcsec, i.e. bigger than the beam size of all transitions observed. This is reasonably the extension of the emission also in the other sources, because it is well-known that the N2H+ (1–0) line traces the core envelope. For the (1–0) lines of the isotopologues, we have assumed the same emission size as for N2H+, and in the few cases in which only the (2–1) or (3–2) line was detected, we have assumed that the emission fills the telescope beam. The different transitions do not have the same critical density, which is ∼105 cm−3 for the (1–0) line, and ∼106 cm−3 for the (3–2) line (see Lique et al. 2015, for the collisional coefficients). Hence the angular size of the (3–2) line emitting region is likely smaller than that of the (1–0) line. However, because the telescope beam size of the (3–2) transition is also smaller (Table 1) than that of the (1–0) line, in the absence of high-angular resolution observations it is reasonable to assume that even the higher excitation transitions fill their (smaller) telescope beam sizes. Moreover, the isotopic ratio is derived comparing the (1–0) and (3–2) transitions only for IRAS 16293 and CepE (this latter does not even belong to the ‘evolutionary’ sample). Hence, the two isotopic ratios obtained from different excitation transitions, although suffering from higher uncertainties, are not crucial for the trends and the overall conclusions of the work. The total molecular column densities are given in Table 4, and the associated isotopic ratios are listed in Table 5. For each source, the estimates have been made assuming a reasonable range of excitation temperatures (5–30 K for all sources but L1544, for which a more realistic range 5–15 K is used). We stress that the temperature assumed is irrelevant when the 14N/15N is derived from the (1–0) line of both isotopologues, but even when derived from different excitation transitions, the difference is within the uncertainties (typical uncertainty in between 20 per cent and 50 per cent, see Table 4). The N2H+ average column densities are of the order of N(N2H+) ∼ 1014cm−2, while N(N2D+) range from ∼1011 up to ∼1013cm−2, and both N(15NNH+) and N(N15NH+) are in the range 1011–1012cm−2. Table 4. Total column densities of N2H+, N2D+, 15NNH+, and N15NH+ calculated as explained in Section 3. For each species, we have assumed a range in excitation temperature, Tex, based on the reference papers in the footnotes. In Column. 2, we indicate the transition from which the total column density of each species has been derived. Source  Transition  Tex  N(N2H+)  N(N2D+)  N(15NNH+)  N(N15NH+)      K  1013cm−2  1013cm−2  1010cm−2  1010cm−2  L1544  1–0  5–15  32.6–12.9  18.5–11.6  62.7–37.2  75.3–44.4  IRAS 4A  1–0  5–30  53.9–26.3  6.3–6.5  54.7–52.0  165–156  IRAS 16293  1–0  5–30  7.0–3.4  –  ≤35.4–33.7  –    2–1  5–30  –  2.0–0.7  –  –    3–2  5–30  –  –  –  393–12.9  L1527  1–0  5–30  20.6–10.0  –  ≤128.6–122.4  ≤142.7–135.1    2–1  5–30  –  ≤0.48- 0.49  –  –  SVS13-A  1–0  5–30  6.8–3.3  –  ≤4.8–4.6  6.3–6.0    2–1  5–30  –  0.08–0.03  –  –  L1157–B1  1–0  5–30  1.7–0.8  0.1–0.1  ≤61.1–58.2  ≤80.6–76.3  CepE  1–0  5–30  2.5–1.2  0.11–0.12  3.2–3.1  –    3–2  5–30  –  –  –  212–6.9  OMC–2 FIR4  1–0  5–30  19.8–9.6  –  40.9–38.9  51.1–48.3    2–1  5–30  –  0.4–0.1  –  –  Source  Transition  Tex  N(N2H+)  N(N2D+)  N(15NNH+)  N(N15NH+)      K  1013cm−2  1013cm−2  1010cm−2  1010cm−2  L1544  1–0  5–15  32.6–12.9  18.5–11.6  62.7–37.2  75.3–44.4  IRAS 4A  1–0  5–30  53.9–26.3  6.3–6.5  54.7–52.0  165–156  IRAS 16293  1–0  5–30  7.0–3.4  –  ≤35.4–33.7  –    2–1  5–30  –  2.0–0.7  –  –    3–2  5–30  –  –  –  393–12.9  L1527  1–0  5–30  20.6–10.0  –  ≤128.6–122.4  ≤142.7–135.1    2–1  5–30  –  ≤0.48- 0.49  –  –  SVS13-A  1–0  5–30  6.8–3.3  –  ≤4.8–4.6  6.3–6.0    2–1  5–30  –  0.08–0.03  –  –  L1157–B1  1–0  5–30  1.7–0.8  0.1–0.1  ≤61.1–58.2  ≤80.6–76.3  CepE  1–0  5–30  2.5–1.2  0.11–0.12  3.2–3.1  –    3–2  5–30  –  –  –  212–6.9  OMC–2 FIR4  1–0  5–30  19.8–9.6  –  40.9–38.9  51.1–48.3    2–1  5–30  –  0.4–0.1  –  –  View Large Table 5. D/H and 14N/15N isotopic ratios calculated as explained in Section 3.2. Source  Tex  $${\rm \frac{N_2H^+}{N_2D^+}}$$  $${\rm \frac{N_2H^+}{^{15}NNH^+}}$$  $${\rm \frac{N_2H^+}{N^{15}NH^+}}$$    K        L1157–B1  5–30  18.4–8.1  27.2–13.9l  20.6–10.6l  L1544  5–15  1.8–1.1  520–347  433–290  IRAS 4A  5–30  8.6–4.1  986–505  327–168  IRAS 16293  5–30  3.5–4.8  197–101l  18–264  L1527  5–30  42.9–20.4l  160–168l  70–74l  SVS13-A  5–30  84–110  1400–717l  1065–548  CepE  5–30  22–10  767–392  12–173  OMC–2 FIR4  5–30  52–74  484–247  388–199  Source  Tex  $${\rm \frac{N_2H^+}{N_2D^+}}$$  $${\rm \frac{N_2H^+}{^{15}NNH^+}}$$  $${\rm \frac{N_2H^+}{N^{15}NH^+}}$$    K        L1157–B1  5–30  18.4–8.1  27.2–13.9l  20.6–10.6l  L1544  5–15  1.8–1.1  520–347  433–290  IRAS 4A  5–30  8.6–4.1  986–505  327–168  IRAS 16293  5–30  3.5–4.8  197–101l  18–264  L1527  5–30  42.9–20.4l  160–168l  70–74l  SVS13-A  5–30  84–110  1400–717l  1065–548  CepE  5–30  22–10  767–392  12–173  OMC–2 FIR4  5–30  52–74  484–247  388–199  Note. llower limit. View Large Finally, for the sources undetected in N2D+, 15NNH+, and N15NH+, we have calculated the upper limits on the column density from the upper limit on the line integrated area. This was evaluated as the area of a single Gaussian having peak equal to the 3σ rms level of the spectrum. To compute the integrated area, however, one needs an estimate of the line width. This latter was evaluated as the width obtained from a Gaussian fit to the other, well-detected lines. In those cases in which the fit with a single Gaussian does not give reasonable results, we have computed the line width by applying to the line width obtained from the hyperfine structure a factor derived from the lines with good fit results with both methods. 4 DISCUSSION In Fig. 1, we show the trends of the three isotopic ratios. We indicate in different colours the low-mass sources with well established evolutionary stage, namely L1544, IRAS 4A, IRAS 16293, L1527, and SVS13-A, and the other objects. We have ordered the ‘evolutionary sample’ with increasing time. Although L1157–B1 does not belong to the evolutionary sample of cores, as explained in Section 2 the measurement of the fractionation in this object reflects that of the diffuse gas of the cloud before the condensation into a dense pre-stellar core, hence we have placed it in the evolutionary sequence before L1544. The two hot-corinos IRAS 4A and IRAS 16293 are placed closer than the other objects because there is no evidence that can indicate a different age of these two objects. Moreover, in Fig. 2, we plot the N2H+/N2D+ versus N2H+/15NNH+ and N2H+/N15NH+ ratios, in order to highlight a possible correlation between the D and 15N fractionation. Figure 1. View largeDownload slide Isotopic ratios H/D (top panel) and 14N/15N (bottom panel) obtained from the N2H+ isotopologues from the method described in Section 3. The points represent the values obtained for an average excitation temperature of 17 K, and the uncertainty is derived from the scatter calculated in the ‘reasonable’ temperature range of 5–30 K (see Table 4 and Section 3.2 for details). In the top panel, red diamonds represent the sources of the ‘evolutionary’ sample (see Section 1), while the other objects are represented by green hexagons. L1157–B1 has been placed before L1544 on the x-axis because the fractionation obtained in this object is associated with the diffuse gas before the passage of the shock (Codella et al. 2013). The points with an upward arrow represent the lower limit of the ratios derived from the upper limits of the column density. In the bottom panel, red and yellow diamonds indicate the isotopic fraction obtained from N15NH+ and 15NNH+, respectively, for the sources of the ‘evolutionary sample’, while for the others (CepE and OMC–2 FIR4), we use blue and green symbols, respectively. Figure 1. View largeDownload slide Isotopic ratios H/D (top panel) and 14N/15N (bottom panel) obtained from the N2H+ isotopologues from the method described in Section 3. The points represent the values obtained for an average excitation temperature of 17 K, and the uncertainty is derived from the scatter calculated in the ‘reasonable’ temperature range of 5–30 K (see Table 4 and Section 3.2 for details). In the top panel, red diamonds represent the sources of the ‘evolutionary’ sample (see Section 1), while the other objects are represented by green hexagons. L1157–B1 has been placed before L1544 on the x-axis because the fractionation obtained in this object is associated with the diffuse gas before the passage of the shock (Codella et al. 2013). The points with an upward arrow represent the lower limit of the ratios derived from the upper limits of the column density. In the bottom panel, red and yellow diamonds indicate the isotopic fraction obtained from N15NH+ and 15NNH+, respectively, for the sources of the ‘evolutionary sample’, while for the others (CepE and OMC–2 FIR4), we use blue and green symbols, respectively. Figure 2. View largeDownload slide D/H ratio derived from the column density ratio N2H+/N2D+ (see Section 3.2) against the 14N/15N ratio derived from both 15NNH+ (red and green symbols) and N15NH+ (blue and yellow symbols). Figure 2. View largeDownload slide D/H ratio derived from the column density ratio N2H+/N2D+ (see Section 3.2) against the 14N/15N ratio derived from both 15NNH+ (red and green symbols) and N15NH+ (blue and yellow symbols). The most direct results emerging from Figs 1 and 2 are: the N2H+/N2D+ decreases, as expected, from the diffuse cloud stage (represented by L1157–B1) to the pre–stellar core stage, represented by L1544, by a factor ∼10. Then, it increases monotonically in the more evolved stages, again by a factor ∼10 in the Class 0 objects, and by a factor ∼100 in the more evolved Class I object SVS13-A; the 14N/15N is quite uniform across the sample, which indicates that time is globally irrelevant in the fractionation of nitrogen in N2H+. Moreover, in the four objects in which both 15NNH+ and N15NH+ are detected, namely L1544, IRAS 4A, CepE, and OMC–2 FIR4, the 14N/15N ratio is the same, within the uncertainties, when derived from the two isomers. The only exception is IRAS 4A, in which they are different by a factor 2 (see Table 5 and Fig. 1). This indicates that the two isotopologues tend to follow the same chemical pathway, although the peculiar case of IRAS 4A deserves to be investigated further, possibly with the help of higher sensitivity and higher spectral resolution observations. Moreover, one can note that the 14N/15N ratio derived from 15NNH+ is systematically higher than that computed from N15NH+, which suggests that the N15NH+ is more abundant than 15NNH+. This difference could be due to the slightly different zero-point energy of the ground state level of the two molecules, which would favour the formation of N15NH+ (see e.g. Terzieva & Herbst 2000). However, all this needs to be supported and confirmed by a higher statistics; the two isotopic ratios D/H and 14N/15N are not related, as it can be seen from Fig. 2. This result is a direct consequence of the fact that the 14N/15N is independent of evolution, opposite to the D/H ratio. Therefore, we do not confirm the possible anticorrelation claimed by Fontani et al. (2015) in massive star-forming cores, although even in that study the anticorrelation was tentative due to the large dispersion of the points. Moreover, our data set and that of Fontani et al. (2015) are different in many aspects: The linear size of the observed region is much more extended in the high-mass objects of Fontani et al. 2015 (located at more than 1 kpc), and the evolutionary time-scales in massive star-forming cores are significantly shorter; inspection of Fig. 1 also shows that the D/H ratio in CepE is consistent with that measured in IRAS 16293 and IRAS 4A, which suggests that low- and intermediate-mass class 0 protostars have a comparable deuteration in N2H+. Of course, this finding needs to be supported by a more robust statistics. That the core evolution is basically irrelevant for the 14N/15N ratio was concluded also by Fontani et al. (2015) and Colzi et al. (2018) from observations of several molecular species (N2H+, CN, HNC, and HNC) towards high-mass star-forming cores belonging to different evolutionary stages. Both our study and those performed in high-mass star-forming regions indicate that the enrichment of 15N is unlikely to happen at core scales during the formation of stars of all masses. Therefore, we speculate that the 15N abundance enhancement measured in pristine Solar system material should be due to chemical processes occurred locally, perhaps at the scale of the protoplanetary disc, and not in the extended surrounding envelope. The same conclusion was provided also by Kahane et al. (2018) towards OMC–2 FIR4. We will discuss this point further in Section 4.1. Interestingly, cometary-like 14N/15N have indeed been measured by Guzmán et al. (2017) in a sample of protoplanetary disc, suggesting a chemical link. However, the study of Guzmán et al. (2017) is focused on HCN and HNC isotopologues, hence obviously what happens in N2H+ and its isotopologues could be completely different and still needs to be investigated in discs. Interestingly, our 14N/15N ratios are consistent, within the uncertainties, with those measured from ammonia and NH2D in a sample of dense and young star-forming cores (e.g. Gerin et al. 2009; Daniel et al. 2013). This would suggest that the 14N/15N ratio in the envelope of star-forming cores does not depend even on the molecule used. However, both our work and those on ammonia mentioned above are based on a small statistics, thus only observations on larger, carefully selected samples of star-forming cores, like that studied in this work, will confirm this possibility. 4.1 Comments on individual sources L1544: The 14N/15N ratio in N2H+ was already derived by Bizzocchi et al. (2013) through a non-LTE method using a Bonnor–Ebert sphere model for the source. This method provided 14N/15N =1000 ± 200, while with our LTE method we derive 14N/15N in the range 275–408 and 228–341 from 15NNH+ and N15NH+. However, we stress that within the uncertainties, the values are still marginally consistent, as we can note from Fig. 1. Also, even with the value given by Bizzocchi et al. (2013), the global evolutionary trend on the 14N/15N in Fig. 1 would not change significantly. On the other hand, the deuterated fraction of ∼0.24 previously calculated by Caselli et al. (2002) with the same approach adopted by us, is perfectly consistent with our estimates (∼0.25–0.33). SVS13-A: we find different values for the 14N/15N derived from 15NNH+ and N15NH+. In particular, the lower limit derived from 15NNH+ is larger (∼700–1400, see Table 5) than the upper limit calculated from N15NH+ (∼360–700), as well as any other 14N/15N measured in our survey. Please note that in Fig. 1, we have shown the intermediate values (1050 and 530, respectively). Because SVS13-A is the only class I object of the sample, if confirmed, this result would point to a different chemical behaviour between the two 15N isotopologues with time, which is not predicted by the most recent theoretical models (Roueff et al. 2015). Hence, it will be worth investigating this result with higher sensitivity observations, and possibly a larger statistics of similar objects. OMC–2 FIR4: as stated in Section 2, this protocluster is the closest analogue of the environment in which our Sun is thought to have been born. The nitrogen fractionation has been extensively studied by Kahane et al. (2018), who found a good agreement between the present-day local interstellar 14N/15N, and the 14N/15N measured from several molecules (HCN, HNC, HC3N, CN). Our results (Table 5) are perfectly consistent with this finding, indicating a remarkable uniformity of this ratio independently from the molecule used, and a further indication that the large-scale gas is not enriched in 15N, as concluded by Kahane et al. (2018). 5 CONCLUSIONS In the framework of the IRAM-30m Large Program ASAI (Lefloch et al., in preparation.), we have analysed the rotational transitions of N2H+, N2D+, 15NNH+, and N15NH+ in order to investigate if (and how) the isotopic fractions H/D and 14N/15N vary during the formation of a Sun-like star. We have confirmed in a robust way that the N2H+/N2D+ is a clear evolutionary indicator in low-mass star formation, because this ratio reaches the minimum value in L1544 (around ∼3–5), i.e. during the pre-stellar core phase at the onset of the gravitational collapse, and then increases with time monotonically by two orders of magnitude during the protostellar phase, as expected from chemical models. On the other hand, our data do not indicate an evolutionary trend for the 14N/15N ratio, and show clearly that the two isotopic ratios H/D and 14N/15N are basically independent. Therefore, our study confirms previous claims obtained in high-mass star-forming cores that the two ratios are not influenced by the same physical/chemical processes. Also, because our findings demonstrate that the 15N enrichment is a process unlikely to happen in the envelope of forming stars of all masses, the enrichment measured in pristine Solar system material should be caused by chemical processes occurred locally at the scale of the protoplanetary disc. ACKNOWLEDGEMENTS This study is based on observations carried out with the IRAM 30m Telescope. IRAM is supported by INSU/CNRS (France), MPG (Germany) and IGN (Spain).We are grateful to the anonymous referee for the useful and constructive comments that have improved the paper. We thank the IRAM staff for the precious help during the observations. F.F. and C. Codella akcnowledge financial support from the Italian Ministero dell'Istruzione, Università e Ricerca through the grant Progetti Premiali 2012 - iALMA (CUP C52I13000140001). C. Ceccarelli acknowledges the funding from the European Research Council (ERC) under the European Union's Horizon 2020 research and innovation programme, project DOC contract 741002. Footnotes 1 http://www.iram-institute.org/EN/content-page-344-7-158-240-344-0.html 2 http://www.iram.fr/IRAMFR/GILDAS 3 https://www.astro.uni-koeln.de/cdms 4 https://www.iram.fr/IRAMFR/GILDAS/doc/pdf/class.pdf 5 http://www.iram.es/IRAMES/mainWiki/Iram30mEfficiencies REFERENCES Adams F. C., 2010, ARA&A , 48, 47 CrossRef Search ADS   Altwegg K. et al.  , 2015, Science , 347, 387 CrossRef Search ADS PubMed  Andre P., Ward-Thompson D., Barsony M., 2000, Mannings V., Boss A. P., Russell S. S., , Protostars and Planets IV. Univ. Arizona Press, Tucson, p. 59 Bachiller R., Pérez Gutiérrez M., 1997, ApJ , 487, L93 CrossRef Search ADS   Bachiller R., Guilloteau S., Gueth F., Tafalla M., Dutrey A., Codella C., Castets A., 1998, A&A , 339, L49 Bachiller R., Pérez Gutiérrez M., Kumar M. S. N., Tafalla M., 2001, A&A , 372, 899 CrossRef Search ADS   Bacmann A., Lefloch B., Ceccarelli C., Steinacker J., Castets A., Loinard L., 2003, ApJ , 585, L55 CrossRef Search ADS   Bianchi E. et al.  , 2017, MNRAS , 467, 3011 CrossRef Search ADS   Bisschop S. E., Jørgensen J. K., Bourke T. L., Bottinelli S., van Dishoeck E. F., 2008, A&A , 488, 959 CrossRef Search ADS   Bizzocchi L., Caselli P., Leonardo E., Dore L., 2013, A&A , 555, 109 CrossRef Search ADS   Bonal L., Huss G. R., Nagashima K., Krot A. N., 2009, Meteorit. Planet. Sci. Suppl. , 72, 5178 Bottinelli S. et al.  , 2004, ApJ , 617, L69 CrossRef Search ADS   Bottinelli S., Ceccarelli C., Williams J. P., Lefloch B., 2007, A&A , 463, 601 CrossRef Search ADS   Busquet G., Fontani F., Viti S., Codella C., Lefloch B., Benedettini M., Ceccarelli C., 2017, A&A , 604, 20 CrossRef Search ADS   Caselli P., 2002, Planet. Space Sci. , 50, 1133 CrossRef Search ADS   Caselli P., Walmsley C. M., Tafalla M., Dore L., Myers P. C., 1999, ApJ , 523, L165 CrossRef Search ADS   Caselli P., Walmsley C. M., Zucconi A., Tafalla M., Dore L., Myers P. C., 2002, ApJ , 565, 344 CrossRef Search ADS   Caselli P. et al.  , 2012, ApJ , 759 L37 CrossRef Search ADS   Castets A., Ceccarelli C., Loinard L., Caux E., Lefloch B., 2001, A&A , 375, 40 CrossRef Search ADS   Caux E. et al.  , 2011, A&A , 532, 23 CrossRef Search ADS   Cazaux S. et al.  , 2003, ApJ , 593, L51 CrossRef Search ADS   Ceccarelli C., Castets A., Loinard L., Caux E., Tielens A. G. G. M., 1998, A&A , 338, L43 Ceccarelli C., Loinard L., Castets A., Tielens A. G. G. M., Caux E., 2000, A&A , 357, L9 Ceccarelli C. et al.  , 2001, Pilbratt G. L., Cernicharo J., Heras A. M., Prusti T., Harris R., , Proc. Symp. ESA-SP 460 , The Promise of the Herschel Space Observatory. Toledo, Spain, p. 219 Ceccarelli C., Caselli P., Bockelée-Morvan D., Mousis O., Pizzarello S., Robert F., Semenov D., 2014, Beuther H., Klessen R. S., Dullemond C. P., Henning T., , Protostars and Planets VI. , Univ. Arizona Press, Tucson, p. 859 Cernicharo J., Guelin M., 1987, A&A , 183, L10 Chen X., Launhardt R., Henning T., 2009, ApJ , 691, 1729 CrossRef Search ADS   Cleeves L. I., Bergin E. A., Alexander C. M. O. D., Du F., Graninger D., Öberg K. I., Harries T. J., 2014, Science , 345, 1590 CrossRef Search ADS PubMed  Codella C. et al.  , 2010, A&A , 518, L112 CrossRef Search ADS   Codella C. et al.  , 2013, ApJ , 776, 52 CrossRef Search ADS   Colzi L., Fontani F., Caselli P., Ceccarelli C., Hily-Blant P., Bizzocchi L., 2018, A&A , 609, A129 Coutens A. et al.  , 2012, A&A , 539, A132 CrossRef Search ADS   Crapsi A., Caselli P., Walmsley C. M., Myers P. C., Tafalla M. Lee C. W., Bourke T. L., 2005, ApJ , 619, 379 CrossRef Search ADS   Crimier N., Ceccarelli C., Lefloch B., Faure A., 2009, A&A , 506, 1229 CrossRef Search ADS   Daniel F. et al.  , 2013, A&A, 560 , 2013, A3 CrossRef Search ADS   De Simone M. et al.  , 2017, A&A , 599, A121 CrossRef Search ADS   Demyk K., Bottinelli S., Caux E. Vastel C., Ceccarelli C., Kahane C., Castets A., 2010, A&A , 517, A17 CrossRef Search ADS   Emprechtinger M., Caselli P., Volgenau N. H., Stutzki J., Wiedner M. C., 2009, A&A , 493, 89 CrossRef Search ADS   Fontani F. J. et al.  , 2011, A&A , 529, L7 CrossRef Search ADS   Fontani F., Sakai T., Furuya K., Sakai N., Aikawa Y., Yamamoto S., 2014, MNRAS , 440, 448 CrossRef Search ADS   Fontani F., Caselli P., Palau A., Bizzocchi L., Ceccarelli C., 2015, ApJ , 808, 46 CrossRef Search ADS   Fontani F. et al.  , 2018, A&A , 605, A57 Gérin M., Marcelino N., Biver N. E. Roueff, Coudert L. H., Elkeurti M., Lis D. C., Bockele-Morvan D., 2009, A&A , 498, L9 CrossRef Search ADS   Gerner T., Shirley Y. L., Beuther H., Semenov D., Linz H., Albertsson T., Henning Th., 2015, A&A , 579, 80 CrossRef Search ADS   Gueth F., Guilloteau S., Bachiller R., 1996, A&A , 307, 891 Gueth F., Guilloteau S., Bachiller R., 1998, A&A , 333, 287 Guzmán V. V., Öberg K. I., Huang J., Loomis R., Qi C., 2017, ApJ , 836, 30 CrossRef Search ADS   Hartogh P. et al.  , 2011, Nature , 478, 218 CrossRef Search ADS PubMed  Hirota T. et al.  , 2007, PASJ , 59, 897 CrossRef Search ADS   Hirota T. et al.  , 2008, PASJ , 60 37 CrossRef Search ADS   Hirota T., Honma M., Imai H., Sunada K., Ueno Y., Kobayashi H., Kawaguchi N., 2011, PASJ , 63, 1 CrossRef Search ADS   Jørgensen J. K., Bourke T. L., Nguyen Luong Q., Takakuwa S., 2011, A&A , 534, A100 CrossRef Search ADS   Jørgensen J. K., Favre C., Bisschop S. E., Bourke T. L., van Dishoeck E. F., Schmalzl M., 2012, ApJ , 757, L4 CrossRef Search ADS   Kahane C., Jaber Al-Edhari A., Ceccarelli C., López-Sepulcre A., Fontani F., Kama M., 2018, ApJ , 852, 130 Lefévre C. et al.  , 2017, A&A , 604, L1 CrossRef Search ADS   Lefloch B., Eisloeffel J., Lazareff B., 1996, A&A , 313, L17 Lefloch B., Castets A., Cernicharo J., Langer W. D., Zylka R., 1998, A&A , 334, 269 Lefloch B. et al.  , 2010, A&A , 518, 113 CrossRef Search ADS   Linsky J. L. et al.  , 2006, ApJ , 647, 1106 CrossRef Search ADS   Lique F., Daniel F., Pagani L., Feautrier N., 2015, MNRAS , 446, 1245 CrossRef Search ADS   Loinard L., Torres R. M., Mioduszewski A. J., Rodríguez L. F., 2008, ApJ , 675, L29 CrossRef Search ADS   Looney L. W., Mundy L. G., Welch W. J., 2000, ApJ , 529, 477 CrossRef Search ADS   Looney L. W., Tobin J. J., Kwon W., 2007, ApJ , 670, L131 CrossRef Search ADS   López-Sepulcre A. et al.  , 2013, A&A , 556, 62L CrossRef Search ADS   López-Sepulcre A. et al.  , 2015, MNRAS , 449, 2438 CrossRef Search ADS   López-Sepulcre A. et al.  , 2017, A&A , 606, A121 CrossRef Search ADS   Manfroid J. et al.  , 2009, A&A , 503, 613 CrossRef Search ADS   Marty B. et al.  , 2010, Geochim. Cosmochim. Acta , 74, 340 CrossRef Search ADS   Millar T. J., Bennett A., Herbst E., 1989, ApJ , 340, 906 CrossRef Search ADS   Moro-Martín A. et al.  , 2001, ApJ , 555, 146 CrossRef Search ADS   Müller H. S. P., Thorwirth S., Roth D. A., Winnewisser G., 2001, A&A , 370, L49 CrossRef Search ADS   Müller H. S. P., Schlüder F., Stutzki J., Winnewisser G., 2005, J. Mol. Struct. , 742, 215 CrossRef Search ADS   Noriega-Crespo A. et al.  , 2004, ApJS , 154, 402 CrossRef Search ADS   Parise B. et al.  , 2004, A&A , 416, 159 CrossRef Search ADS   Persson M. V., Jørgensen J. K., van Dishoeck E. F., 2012, A&A , 541, 39 CrossRef Search ADS   Roueff E., Loison J. C., Hickson K. M., 2015, A&A , 576, 99 CrossRef Search ADS   Sakai N., Sakai T., Hirota T., Yamamoto S., 2008, ApJ , 672, 371 CrossRef Search ADS   Sakai N., Sakai T., Hirota T., Yamamoto S., 2010, ApJ , 722, 1633 CrossRef Search ADS   Sakai N., Oya Y. et al.  , 2016, ApJ , 820, L34 CrossRef Search ADS   Sakai N., Oya Y. et al.  , 2017, MNRAS , 467, L76 Santangelo G. et al.  , 2015, A&A , 584, A126 CrossRef Search ADS   Sargent A. I., 1977, ApJ , 218, 736 CrossRef Search ADS   Shimajiri Y., Takahashi S., Takakuwa S., Saito M., Kawabe R., 2008, ApJ , 683, 255 CrossRef Search ADS   Shinnaka Y., Kawakita H., Jehin E., Decock A., Hutsemékers D., Manfroid J., Arai A., 2016, MNRAS , 462, 195 CrossRef Search ADS   Taquet V. et al.  , 2015, ApJ , 804, 81 CrossRef Search ADS   Taquet V., Furuya K., Walsh C., van Dishoeck E. F., 2016, MNRAS , 462, 99 CrossRef Search ADS   Terzieva R., Herbst E., 2000, MNRAS , 317, 563 CrossRef Search ADS   Tobin J. J. et al.  , 2016, ApJ , 818, 73 CrossRef Search ADS   Vastel C., Ceccarelli C., Lefloch B., Bachiller R., 2014, ApJ , 795, L2 CrossRef Search ADS   SUPPORTING INFORMATION Supplementary data are available at MNRAS online. Appendix A. Spectra. Please note: Oxford University Press is not responsible for the content or functionality of any supporting materials supplied by the authors. Any queries (other than missing material) should be directed to the corresponding author for the article. © 2018 The Author(s) Published by Oxford University Press on behalf of the Royal Astronomical Society http://www.deepdyve.com/assets/images/DeepDyve-Logo-lg.png Monthly Notices of the Royal Astronomical Society Oxford University Press

Deuterium and 15 N fractionation in N 2 H+ during the formation of a Sun-like star

Loading next page...
 
/lp/ou_press/deuterium-and-15-n-fractionation-in-n-2-h-during-the-formation-of-a-cOzMjgcURL
Publisher
Oxford University Press
Copyright
© 2018 The Author(s) Published by Oxford University Press on behalf of the Royal Astronomical Society
ISSN
0035-8711
eISSN
1365-2966
D.O.I.
10.1093/mnras/sty353
Publisher site
See Article on Publisher Site

Abstract

Abstract Although chemical models predict that the deuterium fractionation in N2H+ is a good evolutionary tracer in the star formation process, the fractionation of nitrogen is still a poorly understood process. Recent models have questioned the similar evolutionary trend expected for the two fractionation mechanisms in N2H+, based on a classical scenario in which ion-neutral reactions occurring in cold gas should have caused an enhancement of the abundance of N2D+, 15NNH+, and N15NH+. In the framework of the ASAI IRAM-30m large program, we have investigated the fractionation of deuterium and 15N in N2H+ in the best known representatives of the different evolutionary stages of the Sun-like star formation process. The goal is to ultimately confirm (or deny) the classical ‘ion-neutral reactions’ scenario that predicts a similar trend for D and 15N fractionation. We do not find any evolutionary trend of the 14N/15N ratio from both the 15NNH+ and N15NH+ isotopologues. Therefore, our findings confirm that, during the formation of a Sun-like star, the core evolution is irrelevant in the fractionation of 15N. The independence of the 14N/15N ratio with time, found also in high-mass star-forming cores, indicates that the enrichment in 15N revealed in comets and protoplanetary discs is unlikely to happen at core scales. Nevertheless, we have firmly confirmed the evolutionary trend expected for the H/D ratio, with the N2H+/N2D+ ratio decreasing before the pre-stellar core phase, and increasing monotonically during the protostellar phase. We have also confirmed clearly that the two fractionation mechanisms are not related. stars: formation, ISM: clouds, ISM: molecules, radio lines: ISM 1 INTRODUCTION Observations and chemical models agree that the process of deuterium enrichment in N2H+ is a robust evolutionary indicator in the star formation process. The root ion-neutral reaction that forms N2D+ is (e.g. Millar, Bennett & Herbst 1989; Ceccarelli et al. 2014)   \begin{eqnarray} {\rm H_3^+ + HD \leftrightarrow H_2D^+ + H_2 + \Delta E} \,, \end{eqnarray} (1)followed by the reaction of H2D+ with N2 to give N2D+. Reaction (1) is exothermic by ΔE ∼ 230 K, and hence it is fast only from the left- to right-hand side below ∼20 K. At these low temperatures, and if CO is highly depleted, the abundance of H2D+ is thus boosted, and so are the species directly formed from it, like N2D+. This implies that the N2D+/N2H+ ratio is expected to decrease from the pre-stellar to the proto-stellar phase, when the nascent protostar begins to heat up its surrounding material (e.g. Caselli 2002). In fact, observations of pre-stellar cores and young protostars have shown D/H ratios in N2H+ of the order of 0.1–0.01 (e.g. Bacmann et al. 2003; Crapsi et al. 2005; Emprechtinger et al. 2009), i.e. several orders of magnitude higher than the cosmic D/H elemental abundance (∼10−5, e.g. Linsky et al. 2006), and a decrease with core evolution in both low- and high-mass star-forming cores (Emprechtinger et al. 2009; Fontani et al. 2011; Ceccarelli et al. 2014; Bianchi et al. 2017). This trend is followed by other molecules formed by H2D+, such as DNC (e.g. Fontani et al. 2014; Gerner et al. 2015) and DCO+ (e.g. Emprechtinger et al. 2009; Gerner et al. 2015). Despite the gradual decrease with time expected after the protostar birth, the huge D enrichment in the early cold phase is able to explain the enhanced D/H ratio measured in comets (∼10−4, e.g. Hartogh et al. 2011; Altwegg et al. 2015), with respect to the cosmic value (Cleeves et al. 2014). For this reason, the D/H ratio is believed to be an excellent chemical tool to link the different phases of the Solar system formation, from the earliest pre-stellar stage to the formation of the planets and other Solar system bodies (Ceccarelli et al. 2014). Similarly to D, 15N is enriched in pristine Solar system bodies such as comets (14N/15N ∼150, Manfroid et al. 2009; Shinnaka et al. 2016) and carbonaceous condrites (∼50, Bonal et al. 2009) by factors ∼2 to ∼10 with respect to the protosolar nebula value (∼441, Marty et al. 2010). As for N2D+, the 15NNH+ and N15NH+ abundances were thought to be enhanced through ion-neutral reactions occuring in cold gas (Terzieva & Herbst 2000):   \begin{eqnarray} {\rm ^{15}N + ^{14}\!N_2H^+ \leftrightarrow \ ^{14}N + ^{15}\!NNH^+ + 36 K\,,} \end{eqnarray} (2)  \begin{eqnarray} {\rm ^{15}N + ^{14}\!N_2H^+ \leftrightarrow \ ^{14}N + N^{15}NH^+ + 28 K\,.} \end{eqnarray} (3)However, recent chemical models (Roueff, Loison & Hickson 2015) have questioned the efficiencies of these reactions, because of the presence of energy barriers higher than previously predicted, which make significant fractionation unlikely to occur even in the earliest cold phases of the protostellar evolution. The few observational findings obtained to date make the puzzle even more intriguing: the 14N/15N measurements in N2H+ in high-mass star-forming cores obtained by Fontani et al. (2015) do not show an evolutionary dependence, in agreement with Roueff et al. (2015) predictions, but the same observations indicate a huge dispersion of the values (14N/15N ∼ 100–1000), which cannot be reproduced by the models. The models also do not predict the high 14N/15N ratio found in the low-mass pre-stellar core L1544 (∼1000, Bizzocchi et al. 2013). In this paper, we show the first evolutionary study of the combined 14N/15N and D/H ratio in N2H+ in the best representatives of the main evolutionary stages of the Sun-like star formation process. The primary goal of the study is to understand whether the two ratios are linked and, consequently, whether (also) the 14N/15N is an evolution-dependent parameter and can be considered a chemical link between the earliest and the most evolved stages during the formation of Sun-like stars. 2 SAMPLE AND OBSERVATIONS The targets are extracted from the Large Program ASAI1 (Lefloch et al., in preparation) and represent the best known precursors of a Sun-like star, in the main phases of its formation: the pre-stellar core L1544, the Class 0 protostars IRAS 4A, the Class 0/Class I protostar L1527, and the Class I protostar SVS13-A. We have analysed also the Class 0 source IRAS 16293 using the data of the TIMASSS survey (Caux et al. 2011). Together with this ‘evolutionary’ sample, we have studied other sources to provide additional relevant information about the chemical evolution in a Sun-like star-forming environment: the protocluster OMC–2 FIR4, the chemically rich protostellar shock L1157–B1, and the intermediate-mass Class 0 protostar CepE. In the following, we give a brief description of each target (for more details, please see López-Sepulcre et al. 2015): L1544: L1544 is a starless core in the Taurus molecular cloud complex (d ∼140 pc, Cernicharo & Guélin 1987). It is considered the prototypical pre-stellar core on the verge of the gravitational collapse (Caselli et al. 2012, and references therein). L1544 is characterized by a nucleus with a high H2 density peak (2 × 106 cm−3) and low temperature (∼7 K), surrounded by a lower density envelope undergoing extended infall (Caselli et al. 2012). For these reasons, in the central region, CO is depleted by a factor ∼10 (Caselli et al. 1999) and the deuterium fractionation is high, like in the interiors of dark clouds, although differentiated chemical processes can take place in the external layers (Caselli et al. 1999; Vastel et al. 2014). IRAS 16293: IRAS 16293 is a well known Class 0 protostar located in the small L1689N molecular cloud in the ρ Ophiuchus complex at a distance of 120 pc (Loinard et al. 2008). It is a system with two main components, IRAS16293 A and IRAS16293 B, separated by ∼5 arcsec, characterized by a strong chemical differentiation (Bisschop et al. 2008; Jørgensen et al. 2011, 2012). It is the first source where a hot corino has been discovered, (Ceccarelli et al. 2000; Cazaux et al. 2003; Bottinelli et al. 2004), and a well-studied astrochemical laboratory thanks to its richness in complex organic molecules, and its high deuterium fractionation, for example, in formaldehyde (Ceccarelli et al. 1998, 2001), methanol (Parise et al. 2004), methyl formate (Demyk et al. 2010), and water (Coutens et al. 2012). IRAS 4A: IRAS 4A is the second hot corino ever discovered (Bottinelli et al. 2007), and is a binary source in the Perseus molecular complex (d ∼ 235 pc, Hirota et al. 2007; Hirota et al. 2011); it is composed of two Class 0 objects separated by 1.8 srcsec: IRAS4 A1 and IRAS4 A2 (e.g. Looney, Mundy & Welch 2000). The nature of IRAS4 A1 and IRAS4 A2 has been discussed in several papers (e.g. Persson, Jørgensen & van Dishoeck 2012; Santangelo et al. 2015; Taquet et al. 2015; De Simone et al. 2017; López-Sepulcre et al. 2017). In particular, Santangelo et al. (2015) have concluded that A1 is brigther than A2 in the millimeter continuum, but only A2 is associated with a hot-corino. L1527: L1527 is a dark cloud in the Taurus molecular complex (d ∼ 140 pc), with a heavily obscured IRAS source (IRAS 04368+2557) located at the core centre, classified as a borderline Class 0/Class I object according to Andre, Ward-Thompson & Barsony (2000). This source is considered as a prototypical warm carbon-chain chemistry source (WCCC, Sakai et al. 2008, 2010), and high-angular resolution observations indicate the presence of a compact, infalling, and rotating envelope (Sakai et al. 2016, 2017) surrounding the protostar. SVS13-A: SVS13-A is a part of the multiple system NGC1333-SVS13 (distance 235 pc, Hirota et al. 2007) in the Perseus complex, dominated by three millimetre sources identified by interferometric observations (Bachiller et al. 1998; Looney et al. 2000; Tobin et al. 2016; Lefavre et al. 2017, and references therein): SVS13-A, SVS13-B separated by ∼15 arcsec and associated with two different evolutionary stages, and SVS13-C, ∼20 arcsec away from SVS13-A. Although SVS13-A is still deeply embedded in a large-scale envelope (Lefloch et al. 1998), it is considered a young Class I protostar because of its extended outflow (>0.07 pc) and its low ratio between submillimeter and bolometric luminosity (∼0.8 per cent), becoming an almost unique laboratory to investigate how deuteration and fractionation change from the Class 0 to the Class I phases (Chen, Launhardt & Henning 2009; Tobin et al. 2016; Bianchi et. al 2017; Lefloch et al., in preparation). In the analysis of SVS13-A, care needs to be taken in the fit of the lines because of the contamination of core IRAS 4A at velocity ∼7 km s−1 in the off position (Santangelo et al. 2015), while that of SVS13-A is at about 8–9 km s−1, which results in an artificial absorption near the emission lines of interest. This contamination has affected only the border of the blue side of the lines (see the figures in the on-line Appendix-A); in the case of the N2H+ (1–0) line, we used the observation carried out in Position Switching mode, and not in Wobbler Switching mode, to avoid this contamination. OMC–2 FIR4: OMC–2 FIR4 is a protocluster of low- and intermediate-mass protostars at a distance of 420 pc (e.g. Hirota et al. 2007). Its ∼30 M⊙ envelope extends to about 104 au (Crimier et al. 2009) and contains several clumps, probably a cluster of protostars (Shimajiri et al. 2008; López-Sepulcre et al. 2013), which makes it one of the best and closest environment analogues of the one in which the Sun was born (e.g. Ceccarelli et al. 2014; Fontani et al. 2017). In fact, growing evidence indicates that our Sun was born in a crowded environment, forming together with many other protostars, including very likely more massive objects (e.g. Adams 2010; Taquet et al. 2016). The fractionation of nitrogen has been studied by Kahane et al. (2018) in several molecular species (HCN, HNC, HC3N, CN), who derived 14N/15N ∼ 290, regardless of the molecule used, and in remarkable agreement with the present-day local interstellar 14N/15N. L1157–B1: L1157–B1 is a chemically rich bipolar outflow (Bachiller et al. 1997, 2001) driven by a low-mass (∼4 L⊙) Class 0 protostar (L1157-mm, at a distance of 250 pc, Looney, Tobin & Kwon 2007), and associated with several blue and red-shifted shocks at different ages (Gueth, Guilloteau & Bachiller 1996, 1998). It may be regarded as the ideal laboratory for observing the effects of shocks on the gas chemistry (Bachiller et al. 2001; Codella et al. 2010; Lefloch et al. 2010) and on isotopic fractionation (Fontani et al. 2014; Busquet et al. 2017). The analysis of the N2H+ emission performed by Codella et al. (2013) showed that the N2H+ emission detected towards the L1157-B1 shock can be considered a fossil record of the pre-shock phase, when the gas density was ∼104 cm−3. In other words, N2H+ is sampling the chemical–physical conditions of the quiescient gas evolved for more than 104 yr, before the gas was perturbed by the protostellar shock induced by the L1157 protostellar activity. CepE: CepE is an intermediate-mass Class 0 protostar (L = 100 L⊙, Lefloch, Eisloeffel & Lazareff 1996) in the Cepheus OB3 association at a distance of 730 pc (Sargent 1977), which drives an exceptionally powerful and luminous molecular outflow. Both the protostar and the outflow have been intensively studied at mm and IR wavelengths (Lefloch et al. 1996; Moro-Martin et al. 2001; Noriega-Crespo et al. 2004). The observations have been performed during several runs between 2011 and 2016 with the IRAM 30-m telescope near Pico Veleta (Spain) in the context of the Astrochemical Surveys at IRAM (ASAI) Large Program, using the broad-band EMIR receivers E090, E150, and E230. These were carried out in Wobbler Switching mode, with a throw of 3 arcmin, in order to ensure a flat baseline across the spectral bandwidth observed. The instrumental setup was decided according to the sources: L1544 was observed in the band 72 to 115 GHz using the Fast Fourier Transform Spectrometer (FTS) in its 50 kHz resolution mode; about Protostars and Outflow Shocks, the 3 mm (72–116,GHz) and 2 mm (126–170 GHz) bands were observed at the same time. The 1.3 m (200–272 GHz) band was covered observing LSB and USB simultaneously, using the FTS in its 200 kHz resolution mode (Lefloch et al., in preparation.). The data were reduced with the class software of the gildas2 package. Calibration uncertainties are ∼10 per cent at 3 mm and ∼20 per cent at shorter wavelengths. All the spectra are displayed in antenna temperature units. The main spectroscopic parameters of the transitions observed, and the main observational parameters, are summarized in Table 1. The spectroscopic parameters have been taken from the Cologne Database for Molecular Spectroscopy (CDMS).3 Table 1. Spectroscopic and observational parameters of the observed lines: line rest frequency, ν, energy of the upper level, Eu, line strength, $$S \mu _{\rm ij}^2$$, Einstein coefficient of spontaneous emission, Aij, main beam angular size, ΘMB, and main beam efficiency, Beff. The spectral parameters have been taken from the CDMS (Müller et al. 2001; Müller et al. 2005). Transition  ν  Eu  $$S \mu _{\rm ij}^2$$  Aij  ΘMB  Beff    GHz  K  D2  s−1  arcsec    N2H+ (1–0)  93.173  4.5  104  4 × 10−5  26  0.80  (3–2)  279.512  27  312  1 × 10−3  9  0.49  N2D+ (1–0)  77.109  4  104  2 × 10−5  32  0.83  (2–1)  154.217  11  208  2 × 10−4  16  0.71  (3–2)  231.322  22  312  7 × 10−4  11  0.59  15NNH+ (1–0)  90.264  4  35  3 × 10−5  27  0.80  (3–2)  270.784  26  104  1 × 10−3  9  0.51  N15NH+ (1–0)  91.206  4  35  3 × 10−5  27  0.80  (3–2)  273.609  26  104  1 × 10−3  9  0.51  Transition  ν  Eu  $$S \mu _{\rm ij}^2$$  Aij  ΘMB  Beff    GHz  K  D2  s−1  arcsec    N2H+ (1–0)  93.173  4.5  104  4 × 10−5  26  0.80  (3–2)  279.512  27  312  1 × 10−3  9  0.49  N2D+ (1–0)  77.109  4  104  2 × 10−5  32  0.83  (2–1)  154.217  11  208  2 × 10−4  16  0.71  (3–2)  231.322  22  312  7 × 10−4  11  0.59  15NNH+ (1–0)  90.264  4  35  3 × 10−5  27  0.80  (3–2)  270.784  26  104  1 × 10−3  9  0.51  N15NH+ (1–0)  91.206  4  35  3 × 10−5  27  0.80  (3–2)  273.609  26  104  1 × 10−3  9  0.51  View Large The distances in our sample vary from 120 pc (for the closest object, IRAS 16293) to ∼420 pc (for the farthest, OMC–2 FIR4). Therefore, the linear scales probed differ by a factor ∼3.5 at all frequencies, although in reality the difference is less than a factor 2 for all objects but OMC–2 FIR4. However, most of our calculations are based on the (1–0) transition (see Section 3.2), whose angular beam size corresponds to a linear scale that ranges from a minimum of ∼3000 au for IRAS 16293, to a maximum of ∼11 000 au for OMC–2 FIR4. According to the description of the sources presented above, these scales probe the lower-density envelope in all objects, including OMC–2 FIR4, which is a protocluster intrinsically larger than the other cores, and thus the isotopic fractions that we will derive in Section 3.2 are always associated with the envelope of the sources, and not with their inner nuclei. 3 RESULTS 3.1 Detected lines and fitting procedure We have detected at least one transition of N2H+ and N2D+ in all the sources of our sample. In Tables 2 and 3, we list the lines observed and detected in each source. The N2H+ (1–0) line is detected in all targets, while N2D+ was observed and detected either in the (1–0) and/or in the (2–1) and (3–2) lines. Transitions of N15NH+ have also been detected in all sources, except L1157–B1 and L1527, while 15NNH+ lines were detected only towards L1544, IRAS 4A, Cep E, and OMC–2 FIR4. All the transitions are split into multiple hyperfine components due to the coupling of the 14N nuclear spin with the angular momentum of the molecule. Therefore, we have attempted to fit all lines taking the hyperfine structure into account. The method assumes that the components are Gaussians with the same line width separated in frequency according to the laboratory value, and have all the same excitation temperature. A detailed description of the method and of the output parameters is given in the CLASS manual.4 For the N2H+ and N2D+ lines, the method has given generally good results, while it was not appropriate for several 15NNH+ and N15NH+ lines, mainly because of the faintness of the secondary components. In particular, for two lines, N15NH+ (1–0) in SVS13-A and N2D+ (1–0) in L1157–B1, the main component was clearly detected, but one or more satellites were undetected. Therefore, we have fitted the main component with a single Gaussian, and then calculated the expected contribution of the undetected hyperfine components assuming LTE and optically thin conditions. The final integrated area for these lines is the sum of these contributions (see Tables 2 and 3). In Columns 3–8 of Tables 2 and 3, we show the line parameters derived with the methods mentioned above. The Tables also give the 1σ rms level in the spectrum (Column 9), used for non-detected lines to compute the upper limits on the column densities. Transitions not shown in the Tables were not observed. The spectra of all lines detected are shown in the on-line Appendix A. Table 2. Line parameters obtained applying the hyperfine fit procedure (when possible). In these cases, the ouput parameters are: Tant · τ, the peak velocity of the main hyperfine component, Vpeak, the line width at half-maximum, FWHM, the opacity of the main component, τmain, and the integrated line area, Area. This latter was computed by integrating the spectrum under the channels with signal above the 3σ rms level. The associated uncertainty is computed from the propagation of errors: $$\sigma \times \Delta v_{\rm res} \times \sqrt{N}$$, where σ is the 1σ rms is the spectrum, Δvres is the spectral resolution, and N is the number of channels with signal. The lines without Tant · τ and τmain could not be fit with the hyperfine method, so that the associated Vpeak, FWHM, and Area have been derived from a Gaussian fit. In these cases, the associated uncertainty on the Area represents the 1σ rms given by the fit procedure. The upper limit on the Area for the undetected lines has been calculated assuming a Gaussian line with intensity peak equal to the 3 σ rms level in the spectrum, and FWHM as listed in Column 5 (for details, see end of Section 3.2). The last column lists the 1σ rms level in the spectrum. Molecule  Transition  Tant · τ  Vpeak  FWHM  τmain  Area  rms      K  km s−1  km s−1  –  (K km s−1)  (mK)  L1544  N2H+  1–0  0.9(0.1)  6.0(0.1)  0.3(0.1)  0.5(0.1)  5.11(0.04)  4  N2D+  1–0  4.9(0.1)  5.8(0.2)  0.4(0.6)  2.7(0.1)  1.89(0.02)  3  15NNH+  1–0  0.06(0.02)  6.9(0.1)  0.4(0.1)  1.9(1.8)  0.023(0.004)  1  N15NH+  1–0  0.10(0.04)  6.2(0.1)  0.3(0.1)  3.7(2.9)  0.03(0.01)  2  IRAS4 A  N2H+  1–0  0.3(0.1)  5.9(0.1)  1.2(0.1)  0.1(0.01)  9.10(0.03)  6  N2D+  1–0  1.0(0.1)  7.2(0.1)  1.2(0.1)  0.1(0.01)  1.58(0.02)  3    2–1  2.2(0.01)  7.0(0.1)  1.2(0.1)  1.2(0.1)  2.31(0.04)  6    3–2  1.5(0.1)  6.0(0.1)  1.0(0.1)  1.3(0.2)  1.17(0.05)  8  15NNH+  1–0  –  7.3(0.1)  1.33(0.2)  –  0.017(0.002)  1    3–2  –  6.8(0.1)  0.8(0.3)  –  0.02(0.02)  7  N15NH+  1–0  0.03(0.01)  6.2(0.6)  1.3(2.1)  0.4(0.1)  0.043(0.001)  0.3    3–2  –  6.6(0.1)  0.3(0.1)  –  0.009(0.002)  9  L1527  N2H+  1–0  0.2(0.1)  5.6(0.1)  1.2(0.1)  0.1(0.01)  3.47(0.02)  4  N2D+a  2–1  –  –  5.1b  –  ≤0.12  8.3  15NNH+a  1–0  –  –  3.3b  –  ≤0.04  4  N15NH+a  1–0  –  –  4.5b  –  ≤0.05  4  IRAS 16293  N2H+  1–0  0.3(0.1)  3.0(0.1)  1.7(0.1)  0.1(0.01)  11.78(0.07)  16  N2D+  2–1  2.7(0.1)  4.0(0.1)  1.0(0.1)  2.0(s0.2)  2.17(0.01)  3    3–2  –  4.5(0.2)  2.8(0.4)  0.7(0.1)  0.72(0.02)  10  15NNH+a  1–0  –  –  5.6b  –  ≤0.11  6.4  N15NH+  3–2  –  5.6(0.2)  2.8(0.5)  –  0.13(0.02)  5  SVS13-A  N2H+  1–0  0.6(0.1)  8.2(0.1)  0.8(0.1)  0.1(0.003)  11.40(0.01)  3  N2D+  2–1  0.4(0.1)  8.3(0.1)  0.6(0.1)  0.1(0.01)  0.21(0.02)  3    3–2  –  8.9(0.1)  0.8(0.1)  –  0.17(0.02)  7  15NNH+a  1–0  –  –  1.9b  –  ≤0.015  2.5  N15NH+  1–0  –  7.6(0.1)  1.4(0.3)  –  0.020(0.009)(c)  1.5  Molecule  Transition  Tant · τ  Vpeak  FWHM  τmain  Area  rms      K  km s−1  km s−1  –  (K km s−1)  (mK)  L1544  N2H+  1–0  0.9(0.1)  6.0(0.1)  0.3(0.1)  0.5(0.1)  5.11(0.04)  4  N2D+  1–0  4.9(0.1)  5.8(0.2)  0.4(0.6)  2.7(0.1)  1.89(0.02)  3  15NNH+  1–0  0.06(0.02)  6.9(0.1)  0.4(0.1)  1.9(1.8)  0.023(0.004)  1  N15NH+  1–0  0.10(0.04)  6.2(0.1)  0.3(0.1)  3.7(2.9)  0.03(0.01)  2  IRAS4 A  N2H+  1–0  0.3(0.1)  5.9(0.1)  1.2(0.1)  0.1(0.01)  9.10(0.03)  6  N2D+  1–0  1.0(0.1)  7.2(0.1)  1.2(0.1)  0.1(0.01)  1.58(0.02)  3    2–1  2.2(0.01)  7.0(0.1)  1.2(0.1)  1.2(0.1)  2.31(0.04)  6    3–2  1.5(0.1)  6.0(0.1)  1.0(0.1)  1.3(0.2)  1.17(0.05)  8  15NNH+  1–0  –  7.3(0.1)  1.33(0.2)  –  0.017(0.002)  1    3–2  –  6.8(0.1)  0.8(0.3)  –  0.02(0.02)  7  N15NH+  1–0  0.03(0.01)  6.2(0.6)  1.3(2.1)  0.4(0.1)  0.043(0.001)  0.3    3–2  –  6.6(0.1)  0.3(0.1)  –  0.009(0.002)  9  L1527  N2H+  1–0  0.2(0.1)  5.6(0.1)  1.2(0.1)  0.1(0.01)  3.47(0.02)  4  N2D+a  2–1  –  –  5.1b  –  ≤0.12  8.3  15NNH+a  1–0  –  –  3.3b  –  ≤0.04  4  N15NH+a  1–0  –  –  4.5b  –  ≤0.05  4  IRAS 16293  N2H+  1–0  0.3(0.1)  3.0(0.1)  1.7(0.1)  0.1(0.01)  11.78(0.07)  16  N2D+  2–1  2.7(0.1)  4.0(0.1)  1.0(0.1)  2.0(s0.2)  2.17(0.01)  3    3–2  –  4.5(0.2)  2.8(0.4)  0.7(0.1)  0.72(0.02)  10  15NNH+a  1–0  –  –  5.6b  –  ≤0.11  6.4  N15NH+  3–2  –  5.6(0.2)  2.8(0.5)  –  0.13(0.02)  5  SVS13-A  N2H+  1–0  0.6(0.1)  8.2(0.1)  0.8(0.1)  0.1(0.003)  11.40(0.01)  3  N2D+  2–1  0.4(0.1)  8.3(0.1)  0.6(0.1)  0.1(0.01)  0.21(0.02)  3    3–2  –  8.9(0.1)  0.8(0.1)  –  0.17(0.02)  7  15NNH+a  1–0  –  –  1.9b  –  ≤0.015  2.5  N15NH+  1–0  –  7.6(0.1)  1.4(0.3)  –  0.020(0.009)(c)  1.5  Notes. aTransition undetected: the column density given in Table 4 is an upper limit; bAssumed value for FWHM (see Section 3.2). View Large Table 3. Same as Table 2 for the ‘non-evolutionary’ sources. Molecule  Transition  Tant · τ  Vpeak  FWHM  τmain  Area  rms      K  km s−1  km s−1  -  (K km s−1)  (mK)  L1157–B1  N2H+  1–0  0.004(0.001)  1.4(0.1)  3.0(0.1)  0.1(0.01)  0.28(0.01)  2  N2D+  1–0  –  3.3(0.2)  1.8(0.5)  –  0.025(0.014)c  1  15NNH+a  1–0  –  –  7.5b  –  ≤0.02  1  N15NH+a  1–0  –  –  9.6b  –  ≤0.03  1  OMC–2 FIR4  N2H+  1–0  1.4(0.1)  10.2(0.1)  1.3(0.1)  0.1(0.01)  39.40(0.02)  6  N2D+  2–1  0.8(0.1)  11.0(0.1)  1.3(0.1)  0.3(0.6)  1.24(0.08)  14    3–2  –  10.7(0.02)  0.9(0.03)  –  0.80(0.08)  21  15NNH+  1–0  0.14(0.1)  11.1(0.1)  1.2(0.1)  1.2(0.9)  0.15(0.01)  2  N15NH+  1–0  0.14(0.1)  10.3(0.1)  1.3(0.1)  0.8(0.8)  0.19(0.01)  1  CepE  N2H+  1–0  0.2(0.1)  −13.6(0.1)  1.2(0.1)  0.1(0.01)  4.17(0.02)  4    3–2  –  −11.7(0.1)  4.5(0.1)  –  4.20(0.02)  4  N2D+  1–0  0.3(0.1)  −11.1(0.1)  1.3(0.1)  1.5(1.0)  0.29(0.02)  3    2–1  1.3(0.1)  −11.2(0.1)  0.8(0.1)  2.4(0.3)  0.79(0.02)  4    3–2  –  −12.1(0.3)  4.6(0.7)  –  0.28(0.08)  3  15NNH+  1–0  –  −11.9(0.4)  1.8(1.4)  –  0.010(0.004)  2  N15NH+  1–0  –  −13.2(0.6)  4.3(1.2)  –  0.070(0.001)  0.3  Molecule  Transition  Tant · τ  Vpeak  FWHM  τmain  Area  rms      K  km s−1  km s−1  -  (K km s−1)  (mK)  L1157–B1  N2H+  1–0  0.004(0.001)  1.4(0.1)  3.0(0.1)  0.1(0.01)  0.28(0.01)  2  N2D+  1–0  –  3.3(0.2)  1.8(0.5)  –  0.025(0.014)c  1  15NNH+a  1–0  –  –  7.5b  –  ≤0.02  1  N15NH+a  1–0  –  –  9.6b  –  ≤0.03  1  OMC–2 FIR4  N2H+  1–0  1.4(0.1)  10.2(0.1)  1.3(0.1)  0.1(0.01)  39.40(0.02)  6  N2D+  2–1  0.8(0.1)  11.0(0.1)  1.3(0.1)  0.3(0.6)  1.24(0.08)  14    3–2  –  10.7(0.02)  0.9(0.03)  –  0.80(0.08)  21  15NNH+  1–0  0.14(0.1)  11.1(0.1)  1.2(0.1)  1.2(0.9)  0.15(0.01)  2  N15NH+  1–0  0.14(0.1)  10.3(0.1)  1.3(0.1)  0.8(0.8)  0.19(0.01)  1  CepE  N2H+  1–0  0.2(0.1)  −13.6(0.1)  1.2(0.1)  0.1(0.01)  4.17(0.02)  4    3–2  –  −11.7(0.1)  4.5(0.1)  –  4.20(0.02)  4  N2D+  1–0  0.3(0.1)  −11.1(0.1)  1.3(0.1)  1.5(1.0)  0.29(0.02)  3    2–1  1.3(0.1)  −11.2(0.1)  0.8(0.1)  2.4(0.3)  0.79(0.02)  4    3–2  –  −12.1(0.3)  4.6(0.7)  –  0.28(0.08)  3  15NNH+  1–0  –  −11.9(0.4)  1.8(1.4)  –  0.010(0.004)  2  N15NH+  1–0  –  −13.2(0.6)  4.3(1.2)  –  0.070(0.001)  0.3  Notes. aTransition undetected: the column density given in Table 4 is an upper limit; bAssumed value for FWHM (see the text); cOne or more hyperfine components are under the 3σ rms level; the Area has been computed by summing the integrated intensity of the detected components and the expected area of the undetected components (assuming LTE and optically thin conditions). View Large 3.2 Derivation of the total column densities From the line parameters, we have calculated the total column densities N of N2H+, N2D+, 15NNH+, and N15NH+. Because different sources have been detected in different lines, and, above all, not all targets have been observed in the same lines, to be consistent, we have derived N from the (1–0) line for all species, when possible. This approach was suggested by the fact that the (1–0) line is detected in almost all species and targets, and because the isotopic ratio derived by dividing column densities obtained from the same transition are independent from the excitation temperature. The method has been used successfully for L1544, IRAS 4A, and L1157–B1. In the other targets, this method cannot be applied because the (1–0) line was either not observed or not detected in all molecules. For example, in IRAS 16293, the (1–0) transition is detected in N2H+ and 15NNH+ but neither in N2D+ nor in N15NH+. For these two molecules, we have detected the (2–1) and (3–2) lines (see Table 2). In this case, the column densities were derived adopting the transition with the best signal-to-noise ratio in each molecule, assuming a wide (but reasonable) range of excitation temperatures of 5–30 K. For L1544, we have assumed a more realistic temperature range of 5–15 K. The fit to the hyperfine structure gives an estimate of the optical depth of the main component, τmain. For lines with τmain ≤ 0.5, we have computed N from the total line integrated area according to equation (A4) in Caselli et al. (2002), valid for optically thin lines. The integrated areas have been computed from the spectra in antenna temperature units, and then converted into main beam brightness temperature units using the main beam efficiencies reported on the IRAM–30m website.5 The optically thin approach is good for almost all the detections (see Tables 2 and 3). Some N2D+ lines have τmain > 0.5, but poorly constrained (Δτmain/τmain ≥ 0.3), so that we have derived the column densities using the optically thin approach as well. For the few optically thick lines and with opacity well-constrained, the column density has been calculated by multiplying the value derived in optically thin approximation by the factor τ/(1 − e−τ) (where τ is the total opacity of the line). Finally, for the few 15NNH+ transitions in which the secondary components are not detected, we have used only the integrated intensity of the main one, assuming optically thin conditions, hence in these cases (see Tables 2 and 3) the derived N has to be regarded as lower limits. The column densities should be corrected for the filling factor to compare them in a consistent way, but in most cases we do not know the emission size of the transitions observed, therefore we do not have a direct estimate of the filling factor. However, as shown in Castets et al. (2001), in IRAS 16293 the source size in N2H+ (1–0) can be as extended as ∼30 arcsec, i.e. bigger than the beam size of all transitions observed. This is reasonably the extension of the emission also in the other sources, because it is well-known that the N2H+ (1–0) line traces the core envelope. For the (1–0) lines of the isotopologues, we have assumed the same emission size as for N2H+, and in the few cases in which only the (2–1) or (3–2) line was detected, we have assumed that the emission fills the telescope beam. The different transitions do not have the same critical density, which is ∼105 cm−3 for the (1–0) line, and ∼106 cm−3 for the (3–2) line (see Lique et al. 2015, for the collisional coefficients). Hence the angular size of the (3–2) line emitting region is likely smaller than that of the (1–0) line. However, because the telescope beam size of the (3–2) transition is also smaller (Table 1) than that of the (1–0) line, in the absence of high-angular resolution observations it is reasonable to assume that even the higher excitation transitions fill their (smaller) telescope beam sizes. Moreover, the isotopic ratio is derived comparing the (1–0) and (3–2) transitions only for IRAS 16293 and CepE (this latter does not even belong to the ‘evolutionary’ sample). Hence, the two isotopic ratios obtained from different excitation transitions, although suffering from higher uncertainties, are not crucial for the trends and the overall conclusions of the work. The total molecular column densities are given in Table 4, and the associated isotopic ratios are listed in Table 5. For each source, the estimates have been made assuming a reasonable range of excitation temperatures (5–30 K for all sources but L1544, for which a more realistic range 5–15 K is used). We stress that the temperature assumed is irrelevant when the 14N/15N is derived from the (1–0) line of both isotopologues, but even when derived from different excitation transitions, the difference is within the uncertainties (typical uncertainty in between 20 per cent and 50 per cent, see Table 4). The N2H+ average column densities are of the order of N(N2H+) ∼ 1014cm−2, while N(N2D+) range from ∼1011 up to ∼1013cm−2, and both N(15NNH+) and N(N15NH+) are in the range 1011–1012cm−2. Table 4. Total column densities of N2H+, N2D+, 15NNH+, and N15NH+ calculated as explained in Section 3. For each species, we have assumed a range in excitation temperature, Tex, based on the reference papers in the footnotes. In Column. 2, we indicate the transition from which the total column density of each species has been derived. Source  Transition  Tex  N(N2H+)  N(N2D+)  N(15NNH+)  N(N15NH+)      K  1013cm−2  1013cm−2  1010cm−2  1010cm−2  L1544  1–0  5–15  32.6–12.9  18.5–11.6  62.7–37.2  75.3–44.4  IRAS 4A  1–0  5–30  53.9–26.3  6.3–6.5  54.7–52.0  165–156  IRAS 16293  1–0  5–30  7.0–3.4  –  ≤35.4–33.7  –    2–1  5–30  –  2.0–0.7  –  –    3–2  5–30  –  –  –  393–12.9  L1527  1–0  5–30  20.6–10.0  –  ≤128.6–122.4  ≤142.7–135.1    2–1  5–30  –  ≤0.48- 0.49  –  –  SVS13-A  1–0  5–30  6.8–3.3  –  ≤4.8–4.6  6.3–6.0    2–1  5–30  –  0.08–0.03  –  –  L1157–B1  1–0  5–30  1.7–0.8  0.1–0.1  ≤61.1–58.2  ≤80.6–76.3  CepE  1–0  5–30  2.5–1.2  0.11–0.12  3.2–3.1  –    3–2  5–30  –  –  –  212–6.9  OMC–2 FIR4  1–0  5–30  19.8–9.6  –  40.9–38.9  51.1–48.3    2–1  5–30  –  0.4–0.1  –  –  Source  Transition  Tex  N(N2H+)  N(N2D+)  N(15NNH+)  N(N15NH+)      K  1013cm−2  1013cm−2  1010cm−2  1010cm−2  L1544  1–0  5–15  32.6–12.9  18.5–11.6  62.7–37.2  75.3–44.4  IRAS 4A  1–0  5–30  53.9–26.3  6.3–6.5  54.7–52.0  165–156  IRAS 16293  1–0  5–30  7.0–3.4  –  ≤35.4–33.7  –    2–1  5–30  –  2.0–0.7  –  –    3–2  5–30  –  –  –  393–12.9  L1527  1–0  5–30  20.6–10.0  –  ≤128.6–122.4  ≤142.7–135.1    2–1  5–30  –  ≤0.48- 0.49  –  –  SVS13-A  1–0  5–30  6.8–3.3  –  ≤4.8–4.6  6.3–6.0    2–1  5–30  –  0.08–0.03  –  –  L1157–B1  1–0  5–30  1.7–0.8  0.1–0.1  ≤61.1–58.2  ≤80.6–76.3  CepE  1–0  5–30  2.5–1.2  0.11–0.12  3.2–3.1  –    3–2  5–30  –  –  –  212–6.9  OMC–2 FIR4  1–0  5–30  19.8–9.6  –  40.9–38.9  51.1–48.3    2–1  5–30  –  0.4–0.1  –  –  View Large Table 5. D/H and 14N/15N isotopic ratios calculated as explained in Section 3.2. Source  Tex  $${\rm \frac{N_2H^+}{N_2D^+}}$$  $${\rm \frac{N_2H^+}{^{15}NNH^+}}$$  $${\rm \frac{N_2H^+}{N^{15}NH^+}}$$    K        L1157–B1  5–30  18.4–8.1  27.2–13.9l  20.6–10.6l  L1544  5–15  1.8–1.1  520–347  433–290  IRAS 4A  5–30  8.6–4.1  986–505  327–168  IRAS 16293  5–30  3.5–4.8  197–101l  18–264  L1527  5–30  42.9–20.4l  160–168l  70–74l  SVS13-A  5–30  84–110  1400–717l  1065–548  CepE  5–30  22–10  767–392  12–173  OMC–2 FIR4  5–30  52–74  484–247  388–199  Source  Tex  $${\rm \frac{N_2H^+}{N_2D^+}}$$  $${\rm \frac{N_2H^+}{^{15}NNH^+}}$$  $${\rm \frac{N_2H^+}{N^{15}NH^+}}$$    K        L1157–B1  5–30  18.4–8.1  27.2–13.9l  20.6–10.6l  L1544  5–15  1.8–1.1  520–347  433–290  IRAS 4A  5–30  8.6–4.1  986–505  327–168  IRAS 16293  5–30  3.5–4.8  197–101l  18–264  L1527  5–30  42.9–20.4l  160–168l  70–74l  SVS13-A  5–30  84–110  1400–717l  1065–548  CepE  5–30  22–10  767–392  12–173  OMC–2 FIR4  5–30  52–74  484–247  388–199  Note. llower limit. View Large Finally, for the sources undetected in N2D+, 15NNH+, and N15NH+, we have calculated the upper limits on the column density from the upper limit on the line integrated area. This was evaluated as the area of a single Gaussian having peak equal to the 3σ rms level of the spectrum. To compute the integrated area, however, one needs an estimate of the line width. This latter was evaluated as the width obtained from a Gaussian fit to the other, well-detected lines. In those cases in which the fit with a single Gaussian does not give reasonable results, we have computed the line width by applying to the line width obtained from the hyperfine structure a factor derived from the lines with good fit results with both methods. 4 DISCUSSION In Fig. 1, we show the trends of the three isotopic ratios. We indicate in different colours the low-mass sources with well established evolutionary stage, namely L1544, IRAS 4A, IRAS 16293, L1527, and SVS13-A, and the other objects. We have ordered the ‘evolutionary sample’ with increasing time. Although L1157–B1 does not belong to the evolutionary sample of cores, as explained in Section 2 the measurement of the fractionation in this object reflects that of the diffuse gas of the cloud before the condensation into a dense pre-stellar core, hence we have placed it in the evolutionary sequence before L1544. The two hot-corinos IRAS 4A and IRAS 16293 are placed closer than the other objects because there is no evidence that can indicate a different age of these two objects. Moreover, in Fig. 2, we plot the N2H+/N2D+ versus N2H+/15NNH+ and N2H+/N15NH+ ratios, in order to highlight a possible correlation between the D and 15N fractionation. Figure 1. View largeDownload slide Isotopic ratios H/D (top panel) and 14N/15N (bottom panel) obtained from the N2H+ isotopologues from the method described in Section 3. The points represent the values obtained for an average excitation temperature of 17 K, and the uncertainty is derived from the scatter calculated in the ‘reasonable’ temperature range of 5–30 K (see Table 4 and Section 3.2 for details). In the top panel, red diamonds represent the sources of the ‘evolutionary’ sample (see Section 1), while the other objects are represented by green hexagons. L1157–B1 has been placed before L1544 on the x-axis because the fractionation obtained in this object is associated with the diffuse gas before the passage of the shock (Codella et al. 2013). The points with an upward arrow represent the lower limit of the ratios derived from the upper limits of the column density. In the bottom panel, red and yellow diamonds indicate the isotopic fraction obtained from N15NH+ and 15NNH+, respectively, for the sources of the ‘evolutionary sample’, while for the others (CepE and OMC–2 FIR4), we use blue and green symbols, respectively. Figure 1. View largeDownload slide Isotopic ratios H/D (top panel) and 14N/15N (bottom panel) obtained from the N2H+ isotopologues from the method described in Section 3. The points represent the values obtained for an average excitation temperature of 17 K, and the uncertainty is derived from the scatter calculated in the ‘reasonable’ temperature range of 5–30 K (see Table 4 and Section 3.2 for details). In the top panel, red diamonds represent the sources of the ‘evolutionary’ sample (see Section 1), while the other objects are represented by green hexagons. L1157–B1 has been placed before L1544 on the x-axis because the fractionation obtained in this object is associated with the diffuse gas before the passage of the shock (Codella et al. 2013). The points with an upward arrow represent the lower limit of the ratios derived from the upper limits of the column density. In the bottom panel, red and yellow diamonds indicate the isotopic fraction obtained from N15NH+ and 15NNH+, respectively, for the sources of the ‘evolutionary sample’, while for the others (CepE and OMC–2 FIR4), we use blue and green symbols, respectively. Figure 2. View largeDownload slide D/H ratio derived from the column density ratio N2H+/N2D+ (see Section 3.2) against the 14N/15N ratio derived from both 15NNH+ (red and green symbols) and N15NH+ (blue and yellow symbols). Figure 2. View largeDownload slide D/H ratio derived from the column density ratio N2H+/N2D+ (see Section 3.2) against the 14N/15N ratio derived from both 15NNH+ (red and green symbols) and N15NH+ (blue and yellow symbols). The most direct results emerging from Figs 1 and 2 are: the N2H+/N2D+ decreases, as expected, from the diffuse cloud stage (represented by L1157–B1) to the pre–stellar core stage, represented by L1544, by a factor ∼10. Then, it increases monotonically in the more evolved stages, again by a factor ∼10 in the Class 0 objects, and by a factor ∼100 in the more evolved Class I object SVS13-A; the 14N/15N is quite uniform across the sample, which indicates that time is globally irrelevant in the fractionation of nitrogen in N2H+. Moreover, in the four objects in which both 15NNH+ and N15NH+ are detected, namely L1544, IRAS 4A, CepE, and OMC–2 FIR4, the 14N/15N ratio is the same, within the uncertainties, when derived from the two isomers. The only exception is IRAS 4A, in which they are different by a factor 2 (see Table 5 and Fig. 1). This indicates that the two isotopologues tend to follow the same chemical pathway, although the peculiar case of IRAS 4A deserves to be investigated further, possibly with the help of higher sensitivity and higher spectral resolution observations. Moreover, one can note that the 14N/15N ratio derived from 15NNH+ is systematically higher than that computed from N15NH+, which suggests that the N15NH+ is more abundant than 15NNH+. This difference could be due to the slightly different zero-point energy of the ground state level of the two molecules, which would favour the formation of N15NH+ (see e.g. Terzieva & Herbst 2000). However, all this needs to be supported and confirmed by a higher statistics; the two isotopic ratios D/H and 14N/15N are not related, as it can be seen from Fig. 2. This result is a direct consequence of the fact that the 14N/15N is independent of evolution, opposite to the D/H ratio. Therefore, we do not confirm the possible anticorrelation claimed by Fontani et al. (2015) in massive star-forming cores, although even in that study the anticorrelation was tentative due to the large dispersion of the points. Moreover, our data set and that of Fontani et al. (2015) are different in many aspects: The linear size of the observed region is much more extended in the high-mass objects of Fontani et al. 2015 (located at more than 1 kpc), and the evolutionary time-scales in massive star-forming cores are significantly shorter; inspection of Fig. 1 also shows that the D/H ratio in CepE is consistent with that measured in IRAS 16293 and IRAS 4A, which suggests that low- and intermediate-mass class 0 protostars have a comparable deuteration in N2H+. Of course, this finding needs to be supported by a more robust statistics. That the core evolution is basically irrelevant for the 14N/15N ratio was concluded also by Fontani et al. (2015) and Colzi et al. (2018) from observations of several molecular species (N2H+, CN, HNC, and HNC) towards high-mass star-forming cores belonging to different evolutionary stages. Both our study and those performed in high-mass star-forming regions indicate that the enrichment of 15N is unlikely to happen at core scales during the formation of stars of all masses. Therefore, we speculate that the 15N abundance enhancement measured in pristine Solar system material should be due to chemical processes occurred locally, perhaps at the scale of the protoplanetary disc, and not in the extended surrounding envelope. The same conclusion was provided also by Kahane et al. (2018) towards OMC–2 FIR4. We will discuss this point further in Section 4.1. Interestingly, cometary-like 14N/15N have indeed been measured by Guzmán et al. (2017) in a sample of protoplanetary disc, suggesting a chemical link. However, the study of Guzmán et al. (2017) is focused on HCN and HNC isotopologues, hence obviously what happens in N2H+ and its isotopologues could be completely different and still needs to be investigated in discs. Interestingly, our 14N/15N ratios are consistent, within the uncertainties, with those measured from ammonia and NH2D in a sample of dense and young star-forming cores (e.g. Gerin et al. 2009; Daniel et al. 2013). This would suggest that the 14N/15N ratio in the envelope of star-forming cores does not depend even on the molecule used. However, both our work and those on ammonia mentioned above are based on a small statistics, thus only observations on larger, carefully selected samples of star-forming cores, like that studied in this work, will confirm this possibility. 4.1 Comments on individual sources L1544: The 14N/15N ratio in N2H+ was already derived by Bizzocchi et al. (2013) through a non-LTE method using a Bonnor–Ebert sphere model for the source. This method provided 14N/15N =1000 ± 200, while with our LTE method we derive 14N/15N in the range 275–408 and 228–341 from 15NNH+ and N15NH+. However, we stress that within the uncertainties, the values are still marginally consistent, as we can note from Fig. 1. Also, even with the value given by Bizzocchi et al. (2013), the global evolutionary trend on the 14N/15N in Fig. 1 would not change significantly. On the other hand, the deuterated fraction of ∼0.24 previously calculated by Caselli et al. (2002) with the same approach adopted by us, is perfectly consistent with our estimates (∼0.25–0.33). SVS13-A: we find different values for the 14N/15N derived from 15NNH+ and N15NH+. In particular, the lower limit derived from 15NNH+ is larger (∼700–1400, see Table 5) than the upper limit calculated from N15NH+ (∼360–700), as well as any other 14N/15N measured in our survey. Please note that in Fig. 1, we have shown the intermediate values (1050 and 530, respectively). Because SVS13-A is the only class I object of the sample, if confirmed, this result would point to a different chemical behaviour between the two 15N isotopologues with time, which is not predicted by the most recent theoretical models (Roueff et al. 2015). Hence, it will be worth investigating this result with higher sensitivity observations, and possibly a larger statistics of similar objects. OMC–2 FIR4: as stated in Section 2, this protocluster is the closest analogue of the environment in which our Sun is thought to have been born. The nitrogen fractionation has been extensively studied by Kahane et al. (2018), who found a good agreement between the present-day local interstellar 14N/15N, and the 14N/15N measured from several molecules (HCN, HNC, HC3N, CN). Our results (Table 5) are perfectly consistent with this finding, indicating a remarkable uniformity of this ratio independently from the molecule used, and a further indication that the large-scale gas is not enriched in 15N, as concluded by Kahane et al. (2018). 5 CONCLUSIONS In the framework of the IRAM-30m Large Program ASAI (Lefloch et al., in preparation.), we have analysed the rotational transitions of N2H+, N2D+, 15NNH+, and N15NH+ in order to investigate if (and how) the isotopic fractions H/D and 14N/15N vary during the formation of a Sun-like star. We have confirmed in a robust way that the N2H+/N2D+ is a clear evolutionary indicator in low-mass star formation, because this ratio reaches the minimum value in L1544 (around ∼3–5), i.e. during the pre-stellar core phase at the onset of the gravitational collapse, and then increases with time monotonically by two orders of magnitude during the protostellar phase, as expected from chemical models. On the other hand, our data do not indicate an evolutionary trend for the 14N/15N ratio, and show clearly that the two isotopic ratios H/D and 14N/15N are basically independent. Therefore, our study confirms previous claims obtained in high-mass star-forming cores that the two ratios are not influenced by the same physical/chemical processes. Also, because our findings demonstrate that the 15N enrichment is a process unlikely to happen in the envelope of forming stars of all masses, the enrichment measured in pristine Solar system material should be caused by chemical processes occurred locally at the scale of the protoplanetary disc. ACKNOWLEDGEMENTS This study is based on observations carried out with the IRAM 30m Telescope. IRAM is supported by INSU/CNRS (France), MPG (Germany) and IGN (Spain).We are grateful to the anonymous referee for the useful and constructive comments that have improved the paper. We thank the IRAM staff for the precious help during the observations. F.F. and C. Codella akcnowledge financial support from the Italian Ministero dell'Istruzione, Università e Ricerca through the grant Progetti Premiali 2012 - iALMA (CUP C52I13000140001). C. Ceccarelli acknowledges the funding from the European Research Council (ERC) under the European Union's Horizon 2020 research and innovation programme, project DOC contract 741002. Footnotes 1 http://www.iram-institute.org/EN/content-page-344-7-158-240-344-0.html 2 http://www.iram.fr/IRAMFR/GILDAS 3 https://www.astro.uni-koeln.de/cdms 4 https://www.iram.fr/IRAMFR/GILDAS/doc/pdf/class.pdf 5 http://www.iram.es/IRAMES/mainWiki/Iram30mEfficiencies REFERENCES Adams F. C., 2010, ARA&A , 48, 47 CrossRef Search ADS   Altwegg K. et al.  , 2015, Science , 347, 387 CrossRef Search ADS PubMed  Andre P., Ward-Thompson D., Barsony M., 2000, Mannings V., Boss A. P., Russell S. S., , Protostars and Planets IV. Univ. Arizona Press, Tucson, p. 59 Bachiller R., Pérez Gutiérrez M., 1997, ApJ , 487, L93 CrossRef Search ADS   Bachiller R., Guilloteau S., Gueth F., Tafalla M., Dutrey A., Codella C., Castets A., 1998, A&A , 339, L49 Bachiller R., Pérez Gutiérrez M., Kumar M. S. N., Tafalla M., 2001, A&A , 372, 899 CrossRef Search ADS   Bacmann A., Lefloch B., Ceccarelli C., Steinacker J., Castets A., Loinard L., 2003, ApJ , 585, L55 CrossRef Search ADS   Bianchi E. et al.  , 2017, MNRAS , 467, 3011 CrossRef Search ADS   Bisschop S. E., Jørgensen J. K., Bourke T. L., Bottinelli S., van Dishoeck E. F., 2008, A&A , 488, 959 CrossRef Search ADS   Bizzocchi L., Caselli P., Leonardo E., Dore L., 2013, A&A , 555, 109 CrossRef Search ADS   Bonal L., Huss G. R., Nagashima K., Krot A. N., 2009, Meteorit. Planet. Sci. Suppl. , 72, 5178 Bottinelli S. et al.  , 2004, ApJ , 617, L69 CrossRef Search ADS   Bottinelli S., Ceccarelli C., Williams J. P., Lefloch B., 2007, A&A , 463, 601 CrossRef Search ADS   Busquet G., Fontani F., Viti S., Codella C., Lefloch B., Benedettini M., Ceccarelli C., 2017, A&A , 604, 20 CrossRef Search ADS   Caselli P., 2002, Planet. Space Sci. , 50, 1133 CrossRef Search ADS   Caselli P., Walmsley C. M., Tafalla M., Dore L., Myers P. C., 1999, ApJ , 523, L165 CrossRef Search ADS   Caselli P., Walmsley C. M., Zucconi A., Tafalla M., Dore L., Myers P. C., 2002, ApJ , 565, 344 CrossRef Search ADS   Caselli P. et al.  , 2012, ApJ , 759 L37 CrossRef Search ADS   Castets A., Ceccarelli C., Loinard L., Caux E., Lefloch B., 2001, A&A , 375, 40 CrossRef Search ADS   Caux E. et al.  , 2011, A&A , 532, 23 CrossRef Search ADS   Cazaux S. et al.  , 2003, ApJ , 593, L51 CrossRef Search ADS   Ceccarelli C., Castets A., Loinard L., Caux E., Tielens A. G. G. M., 1998, A&A , 338, L43 Ceccarelli C., Loinard L., Castets A., Tielens A. G. G. M., Caux E., 2000, A&A , 357, L9 Ceccarelli C. et al.  , 2001, Pilbratt G. L., Cernicharo J., Heras A. M., Prusti T., Harris R., , Proc. Symp. ESA-SP 460 , The Promise of the Herschel Space Observatory. Toledo, Spain, p. 219 Ceccarelli C., Caselli P., Bockelée-Morvan D., Mousis O., Pizzarello S., Robert F., Semenov D., 2014, Beuther H., Klessen R. S., Dullemond C. P., Henning T., , Protostars and Planets VI. , Univ. Arizona Press, Tucson, p. 859 Cernicharo J., Guelin M., 1987, A&A , 183, L10 Chen X., Launhardt R., Henning T., 2009, ApJ , 691, 1729 CrossRef Search ADS   Cleeves L. I., Bergin E. A., Alexander C. M. O. D., Du F., Graninger D., Öberg K. I., Harries T. J., 2014, Science , 345, 1590 CrossRef Search ADS PubMed  Codella C. et al.  , 2010, A&A , 518, L112 CrossRef Search ADS   Codella C. et al.  , 2013, ApJ , 776, 52 CrossRef Search ADS   Colzi L., Fontani F., Caselli P., Ceccarelli C., Hily-Blant P., Bizzocchi L., 2018, A&A , 609, A129 Coutens A. et al.  , 2012, A&A , 539, A132 CrossRef Search ADS   Crapsi A., Caselli P., Walmsley C. M., Myers P. C., Tafalla M. Lee C. W., Bourke T. L., 2005, ApJ , 619, 379 CrossRef Search ADS   Crimier N., Ceccarelli C., Lefloch B., Faure A., 2009, A&A , 506, 1229 CrossRef Search ADS   Daniel F. et al.  , 2013, A&A, 560 , 2013, A3 CrossRef Search ADS   De Simone M. et al.  , 2017, A&A , 599, A121 CrossRef Search ADS   Demyk K., Bottinelli S., Caux E. Vastel C., Ceccarelli C., Kahane C., Castets A., 2010, A&A , 517, A17 CrossRef Search ADS   Emprechtinger M., Caselli P., Volgenau N. H., Stutzki J., Wiedner M. C., 2009, A&A , 493, 89 CrossRef Search ADS   Fontani F. J. et al.  , 2011, A&A , 529, L7 CrossRef Search ADS   Fontani F., Sakai T., Furuya K., Sakai N., Aikawa Y., Yamamoto S., 2014, MNRAS , 440, 448 CrossRef Search ADS   Fontani F., Caselli P., Palau A., Bizzocchi L., Ceccarelli C., 2015, ApJ , 808, 46 CrossRef Search ADS   Fontani F. et al.  , 2018, A&A , 605, A57 Gérin M., Marcelino N., Biver N. E. Roueff, Coudert L. H., Elkeurti M., Lis D. C., Bockele-Morvan D., 2009, A&A , 498, L9 CrossRef Search ADS   Gerner T., Shirley Y. L., Beuther H., Semenov D., Linz H., Albertsson T., Henning Th., 2015, A&A , 579, 80 CrossRef Search ADS   Gueth F., Guilloteau S., Bachiller R., 1996, A&A , 307, 891 Gueth F., Guilloteau S., Bachiller R., 1998, A&A , 333, 287 Guzmán V. V., Öberg K. I., Huang J., Loomis R., Qi C., 2017, ApJ , 836, 30 CrossRef Search ADS   Hartogh P. et al.  , 2011, Nature , 478, 218 CrossRef Search ADS PubMed  Hirota T. et al.  , 2007, PASJ , 59, 897 CrossRef Search ADS   Hirota T. et al.  , 2008, PASJ , 60 37 CrossRef Search ADS   Hirota T., Honma M., Imai H., Sunada K., Ueno Y., Kobayashi H., Kawaguchi N., 2011, PASJ , 63, 1 CrossRef Search ADS   Jørgensen J. K., Bourke T. L., Nguyen Luong Q., Takakuwa S., 2011, A&A , 534, A100 CrossRef Search ADS   Jørgensen J. K., Favre C., Bisschop S. E., Bourke T. L., van Dishoeck E. F., Schmalzl M., 2012, ApJ , 757, L4 CrossRef Search ADS   Kahane C., Jaber Al-Edhari A., Ceccarelli C., López-Sepulcre A., Fontani F., Kama M., 2018, ApJ , 852, 130 Lefévre C. et al.  , 2017, A&A , 604, L1 CrossRef Search ADS   Lefloch B., Eisloeffel J., Lazareff B., 1996, A&A , 313, L17 Lefloch B., Castets A., Cernicharo J., Langer W. D., Zylka R., 1998, A&A , 334, 269 Lefloch B. et al.  , 2010, A&A , 518, 113 CrossRef Search ADS   Linsky J. L. et al.  , 2006, ApJ , 647, 1106 CrossRef Search ADS   Lique F., Daniel F., Pagani L., Feautrier N., 2015, MNRAS , 446, 1245 CrossRef Search ADS   Loinard L., Torres R. M., Mioduszewski A. J., Rodríguez L. F., 2008, ApJ , 675, L29 CrossRef Search ADS   Looney L. W., Mundy L. G., Welch W. J., 2000, ApJ , 529, 477 CrossRef Search ADS   Looney L. W., Tobin J. J., Kwon W., 2007, ApJ , 670, L131 CrossRef Search ADS   López-Sepulcre A. et al.  , 2013, A&A , 556, 62L CrossRef Search ADS   López-Sepulcre A. et al.  , 2015, MNRAS , 449, 2438 CrossRef Search ADS   López-Sepulcre A. et al.  , 2017, A&A , 606, A121 CrossRef Search ADS   Manfroid J. et al.  , 2009, A&A , 503, 613 CrossRef Search ADS   Marty B. et al.  , 2010, Geochim. Cosmochim. Acta , 74, 340 CrossRef Search ADS   Millar T. J., Bennett A., Herbst E., 1989, ApJ , 340, 906 CrossRef Search ADS   Moro-Martín A. et al.  , 2001, ApJ , 555, 146 CrossRef Search ADS   Müller H. S. P., Thorwirth S., Roth D. A., Winnewisser G., 2001, A&A , 370, L49 CrossRef Search ADS   Müller H. S. P., Schlüder F., Stutzki J., Winnewisser G., 2005, J. Mol. Struct. , 742, 215 CrossRef Search ADS   Noriega-Crespo A. et al.  , 2004, ApJS , 154, 402 CrossRef Search ADS   Parise B. et al.  , 2004, A&A , 416, 159 CrossRef Search ADS   Persson M. V., Jørgensen J. K., van Dishoeck E. F., 2012, A&A , 541, 39 CrossRef Search ADS   Roueff E., Loison J. C., Hickson K. M., 2015, A&A , 576, 99 CrossRef Search ADS   Sakai N., Sakai T., Hirota T., Yamamoto S., 2008, ApJ , 672, 371 CrossRef Search ADS   Sakai N., Sakai T., Hirota T., Yamamoto S., 2010, ApJ , 722, 1633 CrossRef Search ADS   Sakai N., Oya Y. et al.  , 2016, ApJ , 820, L34 CrossRef Search ADS   Sakai N., Oya Y. et al.  , 2017, MNRAS , 467, L76 Santangelo G. et al.  , 2015, A&A , 584, A126 CrossRef Search ADS   Sargent A. I., 1977, ApJ , 218, 736 CrossRef Search ADS   Shimajiri Y., Takahashi S., Takakuwa S., Saito M., Kawabe R., 2008, ApJ , 683, 255 CrossRef Search ADS   Shinnaka Y., Kawakita H., Jehin E., Decock A., Hutsemékers D., Manfroid J., Arai A., 2016, MNRAS , 462, 195 CrossRef Search ADS   Taquet V. et al.  , 2015, ApJ , 804, 81 CrossRef Search ADS   Taquet V., Furuya K., Walsh C., van Dishoeck E. F., 2016, MNRAS , 462, 99 CrossRef Search ADS   Terzieva R., Herbst E., 2000, MNRAS , 317, 563 CrossRef Search ADS   Tobin J. J. et al.  , 2016, ApJ , 818, 73 CrossRef Search ADS   Vastel C., Ceccarelli C., Lefloch B., Bachiller R., 2014, ApJ , 795, L2 CrossRef Search ADS   SUPPORTING INFORMATION Supplementary data are available at MNRAS online. Appendix A. Spectra. Please note: Oxford University Press is not responsible for the content or functionality of any supporting materials supplied by the authors. Any queries (other than missing material) should be directed to the corresponding author for the article. © 2018 The Author(s) Published by Oxford University Press on behalf of the Royal Astronomical Society

Journal

Monthly Notices of the Royal Astronomical SocietyOxford University Press

Published: May 1, 2018

There are no references for this article.

You’re reading a free preview. Subscribe to read the entire article.


DeepDyve is your
personal research library

It’s your single place to instantly
discover and read the research
that matters to you.

Enjoy affordable access to
over 18 million articles from more than
15,000 peer-reviewed journals.

All for just $49/month

Explore the DeepDyve Library

Search

Query the DeepDyve database, plus search all of PubMed and Google Scholar seamlessly

Organize

Save any article or search result from DeepDyve, PubMed, and Google Scholar... all in one place.

Access

Get unlimited, online access to over 18 million full-text articles from more than 15,000 scientific journals.

Your journals are on DeepDyve

Read from thousands of the leading scholarly journals from SpringerNature, Elsevier, Wiley-Blackwell, Oxford University Press and more.

All the latest content is available, no embargo periods.

See the journals in your area

DeepDyve

Freelancer

DeepDyve

Pro

Price

FREE

$49/month
$360/year

Save searches from
Google Scholar,
PubMed

Create lists to
organize your research

Export lists, citations

Read DeepDyve articles

Abstract access only

Unlimited access to over
18 million full-text articles

Print

20 pages / month

PDF Discount

20% off