COMPLEX MOLECULES TOWARD LOW-MASS PROTOSTARS: THE SERPENS CORE^*

CH₃OH observations toward the Serpens core are presented in Kristensen et al. (2010) in the form of a James Clerk Maxwell Telescope (JCMT) HARP-B map. To compare CH₃OH and complex molecule abundances, the fully sampled CH₃OH map, obtained with a 15'' beam, in Figure 1 was convolved using MIRIAD (Sault et al. 1995) with the complex molecule beam sizes of ∼17'' and 28'' for the 2 and 3 mm observations, respectively. The resulting CH₃OH line intensities at the SMM1, SMM4, and SMM4-W positions are used to derive new column densities and rotational temperatures using the rotational diagram method (see the Appendix). The column densities and excitation temperatures are listed in Table 2, demonstrating that the applied beam size has only a marginal impact. CH₃OH thus seems equally abundant on 17'' and 28'' scales, suggesting that the emission is dominated by the protostellar envelope rather than a central hot core.

The derived column densities toward SMM1 and SMM4-W are up to 30% lower compared to the results presented in Kristensen et al. (2010) based on a 15'' beam. This is due to underweighting of some weak lines in Kristensen et al. (2010), which resulted in higher rotational temperatures and higher column densities.

Figure 2 shows the complete spectra for all settings. Figure 3 zooms in on the most diagnostic lines and shows that SMM1, SMM4, and SMM4-W all contain complex organic molecules with peak intensities of 20–100 mK. CH₃CHO and CH₃OCH₃ are detected in all lines of sight, while at least two HCOOCH₃ lines are detected toward SMM1 and SMM4-W and tentatively detected toward SMM4. C₂H₅OH, HCOCH₂OH, and C₂H₅CN are not detected.

Zoom In Zoom Out Reset image size

Zoom In Zoom Out Reset image size

Table 3 lists all detections and significant upper limits. The FWHM vary between 0.7 and 2.5 toward SMM1, 1.0 and 2.6 km s⁻¹ toward SMM4, and 1.6 and 4.4 km s⁻¹ toward SMM4-W. Only the HCOOCH₃ lines around 88.8 GHz and the CH₃CHO line around 115 GHz are not blended, however, and these lines have FWHM of 0.7–1.2, 1.0, and 1.6–2.3 km s⁻¹ toward SMM1, SMM4, and SMM4-W, respectively. These lines are narrow compared to the CH₃OH FWHM of 3.4–4.0 km s⁻¹ toward all three sources (Kristensen et al. 2010). There is thus no evidence that the complex organics originate in a hotter or more turbulent region than CH₃OH, justifying our presentation of beam-averaged abundances below.

The line positions of all complex molecules toward SMM1 and SMM4 are consistent with a constant source velocity, indicating that all molecular emission has the same origin. This is not the case toward SMM4-W. The observed HCOOCH₃ emission has the same source velocity as SMM4, while the CH₃CHO and the CH₃OCH₃ emission are shifted by ∼5 km s⁻¹, comparable to the shift in CH₃OH emission observed by Kristensen et al. (2010). CH₃CHO and CH₃OCH₃ emission then probably originate in the observed CH₃OH outflow, while HCOOCH₃ emission originates in a more quiescent region associated with the extended SMM4 envelope.

The lines are integrated numerically assuming central line velocities of 7, 7, and 7–12 km s⁻¹, for SMM1, SMM4, and SMM4-W, respectively. 3σ upper limits are derived assuming 1 km s⁻¹ line widths toward SMM1 and SMM4 and 2 km s⁻¹ line widths toward SMM4-W, and the standard equation $\sigma ={\rm rms}\times {\rm FWHM}/\sqrt{n_{\rm ch}}$, where n_ch is the number of channels across the FWHM. The rms and resolutions are listed in Table 1 and the resulting line intensities and upper limits in Table 3. The lowest upper limits are ∼20 mK km s⁻¹, while detections range between 30 and 560 mK km s⁻¹; most upper limits are significant compared to the detected line intensities.

Assuming that the molecules are thermalized at a rotational temperature, optically thin emission, and no beam dilution, we can use

$$\begin{equation} N = \frac{1.67\times 10^{14}}{\nu \mu ^2S} Q(T_{\rm rot})e^{{-E_u/T_{\rm rot}}}\int T_{\rm b}dv, \end{equation} \tag{ 1 }$$

from Thi et al. (2004), where N is the total number of molecules, Q(T_rot) is the temperature-dependent partition function, E_u is the energy of the upper level in K, T_rot is the rotational temperature, and ∫T_bdv is the integrated line intensity. μ²S is gathered from Cologne Database for Molecular Spectroscopy (CDMS) for each molecule or calculated from I(300 K) in JPL, Q(T_rot) is calculated at the CH₃OH rotational temperature by second-order interpolation between the three closest values, and E_u is calculated from the listed E_low values. Using the CH₃OH excitation temperature is a reasonable approximation since the thermal desorption temperatures and the non-thermal desorption efficiencies are expected to be similar for CH₃OH and more complex ices. Also, because of the fact that μ²S is similar for all considered molecules, they should be excited to a similar level.

The resulting column densities are listed in Table 4 and when constraints are available from multiple lines, the column densities are consistent within a factor of two. This supports the assumption that the CH₃OH gas and the complex molecules have a similar excitation temperature. The abundances of complex molecules are calculated based on the derived complex molecule column densities and the CH₃OH column densities listed in Table 2. SMM1 has the highest overall complex molecular abundances with 10% HCOOCH₃, 6% CH₃CHO, and 5% CH₃OCH₃, all with respect to CH₃OH. In the other two sources most complex abundances are ∼1% with respect to CH₃OH.

The three observed positions in the Serpens core are different both with respect to their total complex molecular content, as traced by HCOOCH₃, CH₃CHO, and CH₃OCH₃, and the relative abundances of the targeted molecules. The summed-detected complex molecular abundance with respect to CH₃OH is ∼22% toward SMM1, 3% toward SMM4, and 5% toward SMM4-W. The abundances thus vary over almost an order of magnitude, with the most complex molecules toward the most luminous protostar. This is in line with expectations that, averaged over the envelope, more complex molecules are produced in the CH₃OH ices when a larger portion of the envelope is heated and exposed to more intense UV radiation.

Toward SMM1 and SMM4-W, HCOOCH₃ is the most abundant complex molecule, while CH₃CHO dominates the observed chemistry toward SMM4. In a scenario where most complex molecules originate from UV chemistry in CH₃OH-rich ices, both HCOOCH₃ and CH₃CHO are expected to mainly form when cold, CO-rich ices are irradiated (Öberg et al. 2009a). In contrast, CH₃OCH₃ is only expected to form efficiently from CH₃OH ice photochemistry once CO has evaporated. The HCOOCH₃/CH₃OCH₃ ratio is therefore a potential tracer of the relative importance of cold versus lukewarm complex ice chemistry. The (HCOOCH₃+CH₃CHO)/CH₃OCH₃ abundance ratios are ∼3 toward all three sources, providing evidence for the cold complex chemistry dominating the beam-averaged emission—consistent with the derived low rotational temperatures. Gas observations of ice chemistry products in such cold regions are possible due to, e.g., UV photodesorption, desorption due to release of chemical energy, and sputtering in shocks.

Based on the observations, the main desorption mechanisms may differ for different complex molecules close to outflows. The low central velocities of the HCOOCH₃ lines toward the outflow SMM4-W suggests that it does not originate at the shock front where ices are sputtered by the shock, but rather in the surrounding colder material, where UV rays from the outflow cavity may still be able to penetrate. In contrast, the CH₃CHO and CH₃OCH₃ emission coincides with the outflow (traced by, e.g., CH₃OH). Similarly toward B1-b, HCOOCH₃ has no outflow signature, while CH₃CHO seems partially associated with a weak outflow (Öberg et al. 2010).

This difference between HCOOCH₃, CH₃CHO, and CH₃OCH₃ emission may be the result of different ice surface and ice mantle compositions together with the experimental results that UV photodesorption is only active in the ice surface layer (Öberg et al. 2009b, 2009c), while the shock front results in desorption of the entire ice mantle. If HCOOCH₃ mainly forms on ice surfaces, low levels of HCOOCH₃ gas should be present throughout the protostellar envelopes. Its relative importance to other complex molecules should be less in regions where the entire ice mantle has desorbed at, e.g., shock fronts and inside hot cores. Determining the mechanism behind the different ice surface and ice mantle compositions requires detailed modeling beyond the scope of this paper. Below we instead aim to provide more empirical constraints on the ice chemistry evolution by comparing our main results with previous observations toward low- and high-mass protostars.

Table 5 lists the low-mass sources toward which at least one hot-core-type molecule has been detected. The list comprises one illuminated cold core edge, two outflows, six protostellar envelopes, and three low-mass hot cores observed with 2'' beams. Most previous single-dish studies assumed some level of beam dilution based on the assumption that the complex molecules mainly originate from thermal desorption close to the protostar. Öberg et al. (2010) showed that abundant HCOOCH₃ may instead be due to non-thermal desorption at large scales, and a priori it is difficult to tell whether the complex organics observed with single-dish telescopes toward low-mass young stellar objects mainly reside in a hot core or are spread out in the envelope. We have therefore used abundances derived under the assumption that CH₃OH and complex organics have the same distributions (Herbst & van Dishoeck 2009).

Table 5. CH₃OH Data and Beam-averaged Complex Molecule Abundances Toward Low-mass Sources

Source	CH3OH			HCOOCH3	CH3CHO	CH3OCH3	C2H5OH	Refs.
	N/1014 cm−2	Beam/''	Trot/K	% CH3OH
B1-b core	4.7	19	10[5]	2.3	1.2	<0.8	<1.0	1
SMM4-W	22[7]	22.5	11[1]	3.5	0.6	1.1	<0.4
L1157 outflow	15	10.5	12[2]	1.8	...	...	0.7	2, 3
SMM1	2.5[0.3]	22.5	16[1]	10.0	6.6	5.3	<3.4
SMM4	10.5[1.0]	22.5	13[1]	<1.0	2.2	0.8	<0.6
NGC1333 IRAS 4Aa	5.1	10	24	56	...	<22	...	4, 5, 6
NGC1333 IRAS 4Ba	3.5	10	34	26	...	<19	...	4, 5, 6
IRAS 16293 env.a	8.8	20	85	30	4	20	...	6, 7, 8
IRAS 16293 Ab	1.1 × 104	13 × 27	100	0.6	<0.02	0.6	1.4	9, 10
IRAS 16293 Bb	5.0 × 103	13 × 27	100	0.8	0.6	1.6	...	9, 10
NGC1333 IRAS 2Ab	2 × 104	18 × 10	>75 K	...	...	2	...	11, 12

Notes. ^aThe complex abundances values were calculated assuming the same beam dilution for CH₃OH and the complex species, which should correspond to comparing beam-averaged abundances since the beams are of similar size (Herbst & van Dishoeck 2009). ^bThe complex abundances are also from interferometric observations. References. (1) Öberg et al. 2010; (2) Bachiller & Perez Gutierrez 1997; (3) Arce et al. 2008; (4) Maret et al. 2005; (5) Bottinelli et al. 2007; (6) Herbst & van Dishoeck 2009; (7) van Dishoeck et al. 1995; (8) Cazaux et al. 2003; (9) Kuan et al. 2004; (10) Bisschop et al. 2008; (11) Huang et al. 2005; (12) Jørgensen et al. 2005.

Download table as:  ASCIITypeset image

The summed-detected complex molecule abundances are 2.5%–6% toward the outflows (B1-b, SMM4-W, L1157), 9%–56% toward the protostellar envelopes (excluding SMM4), where emission from the cold and hot regions are blended, and 2%–3% toward the low-mass hot cores IRAS 16293A, IRAS 16293B, and IRAS2A for which interferometer data exist. The complex organics abundance is generally higher in protostellar envelopes compared to in outflows. This suggests a chemical evolution where more CH₃OH ice has been converted into complex organics in the protostellar envelope compared to the more pristine ices that are released by the outflow shocks and radiation (assuming the main function of the shocks is to desorb ices rather than initiate chemistry). The SMM4 abundances are more similar to the outflow sources than envelopes around the other, much more luminous protostars, which may be due to the fact that too small an envelope region is affected by the protostar in this case to affect the observable chemistry.

The low-mass hot cores should be further chemically evolved than the envelopes since the ices are expected to evolve during infall from the colder parts of the envelope. It is therefore unexpected that the complex abundances with respect to CH₃OH are smaller in the angularly resolved observations. As discussed in Section 4.1, different ice surface and ice mantle compositions may play a large role in the observed gas-phase abundance ratios. Assuming that complex molecules form mainly in the ice surface layer in cold regions, their abundances with respect to CH₃OH in the surface layer should increase with radiation dose, e.g., between the outflows and the protostar. As long as non-thermal desorption regulates the gas-phase abundances (which should be true in most of the protostellar envelope), it is only the complex-organics-enriched ice surface layer that is reflected in the gas-phase abundances. Once thermal desorption becomes the dominant mechanism to release CH₃OH and complex organics into the gas phase, the gas-phase abundances instead reflect the complex organics to CH₃OH ratio in the bulk of the ice mantle, which may be much lower. To confirm this scenario clearly requires a combination of laboratory experiments on surface versus bulk chemistry, astrochemical modeling, and more hot-core interferometric observations.

In the unresolved observations that probe both cold and warm material, HCOOCH₃ is generally the most abundant complex molecule (except for CH₃CHO toward SMM 4) while CH₃OCH₃ or C₂H₅OH is the most abundant molecule in the low-mass hot-core observations. This supports an ice chemistry scenario where complex molecules form sequentially starting with HCO-rich molecules (HCOOCH₃ and/or CH₃CHO) as long as CO ice is abundant, followed by CH_3/2-rich molecules at higher ice temperatures.

Figure 4 also shows how the low-mass sample of complex molecule abundances compare with average values from a single-dish survey toward high-mass hot cores (Bisschop et al. 2007). Generally, the abundances are quite similar within a factor of a few toward low- and high-mass sources, but the low-mass sources do generally have lower CH₃OCH₃ abundances compared to the hot-core sample. While the difference in absolute numbers may be due to the lack of applied beam dilution toward the low-mass sources, the average HCOOCH₃/CH₃OCH₃ ratio toward the high-mass hot cores of 0.2 is an order of magnitude lower than the ratio toward the low-mass sources observed with a single-dish telescope. The molecules in the high-mass hot core have a higher temperature chemical history. The difference between the low-mass and high-mass sample further supports that the formation of complex organics is a sequential process, starting with HCO-rich organics at low temperatures.

Zoom In Zoom Out Reset image size