The factors that influence protostellar multiplicity II. Gas temperature and mass in Perseus with APEX

TL;DR

Protostellar multiplicity is a fundamental outcome of star formation, and this paper investigates how gas temperature and mass in Perseus relate to multiplicity by using molecular tracers. The authors derive gas kinetic temperatures and gas masses from line ratios of $DCO^+$, $H_2CO$, and $c$-C$_3$H$_2$ observed with APEX and NRO toward 31 protostellar systems at envelope scales of about 5,000–8,000 AU. They find that gas temperatures are generally not correlated with multiplicity, while envelope gas and dust masses show a tendency for higher values in higher-order multiples, suggesting a continuum of formation pathways rather than a dichotomy between core and disk fragmentation. The results imply that large mass reservoirs and their spatial distribution govern multiplicity outcomes, highlighting the need for spatially resolved studies of mass delivery and magnetic fields to fully understand multiple system formation and evolution.

Abstract

Protostellar multiplicity is a common outcome of the star formation process. To fully understand the formation and evolution of these systems, the physical parameters of the molecular gas together with the dust must be systematically characterized. Using observations of molecular gas tracers, we characterize the physical properties of cloud cores in the Perseus molecular cloud (average distance of 295 pc) at envelope scales (5000-8000 AU). We used Atacama Pathfinder EXperiment (APEX) and Nobeyama 45m Radio Observatory (NRO) observations of DCO$^+$, H$_2$CO and c-C$_3$H$_2$ in several transitions to derive the physical parameters of the gas toward 31 protostellar systems in Perseus. Gas kinetic temperature was obtained from DCO$^+$, H$_2$CO and c-C$_3$H$_2$ line ratios. Column densities and gas masses were then calculated for each species and transition. Gas kinetic temperature and gas masses were compared with bolometric luminosity, envelope dust mass, and multiplicity to search for statistically significant correlations. Gas kinetic temperature derived from DCO$^+$, H$_2$CO and c-C$_3$H$_2$ line ratios have average values of 14 K, 26 and 16 K, respectively, with a range of 10-26 K for DCO$^+$ and c-C$_3$H$_2$. The gas kinetic temperature obtained from H$_2$CO line ratios have a range of 13-82 K. Column densities of all three molecular species are on the order of 10$^{11}$ to 10$^{14}$ cm$^{-2}$, resulting in gas masses of 10$^{-11}$ to 10$^{-9}$ M$_{\odot}$. Statistical analysis of the physical parameters finds: i) similar envelope gas and dust masses for single and binary protostellar systems; ii) multiple (>2 components) protostellar systems tend to have slightly higher gas and dust masses than binaries and single protostars; iii) a continuous distribution of gas and dust masses is observed regardless of separation between components in protostellar systems.

Figures (6)

• Figure 1: Physical parameters derived from APEX observations. Top row: Gas kinetic temperature and corresponding errors obtained from molecular line ratios. Cyan circles show the temperature obtained from DCO+ $J$=5--4 / $J$=3--2, orange squares show the average temperature obtained from the three H2CO ratios, and the black diamonds show the average temperature obtained from the two c-C3H2 ratios. Upper limits are indicated with a downward arrow. The horizontal red dash-dotted line indicates the average gas kinetic temperature ($T_{kin}~\sim$15 K) derived from the I(HCN)/I(HNC) $J$=1--0 (Paper I). The gray crosses show the gas kinetic temperature from NH3 observations ($T_{kin, \ce{NH3}}$) obtained from the core catalog of rosolowsky2008. Left and right panels plot gas kinetic temperature versus bolometric luminosity ($L_{\rm bol}$) and envelope dust mass ($M_{\rm env}$), respectively. Bottom row: Column density for DCO+ (cyan circles), H2CO (orange squares) and c-C3H2 (black diamonds) obtained from the respective lowest transition in the APEX data. Solid lines and shaded areas show the linear regression for the data with the corresponding color. Left and right panels plot column density versus $L_{\rm bol}$ and $M_{\rm env}$. Error bars are smaller than the symbols.
• Figure 2: Gas mass ($M_{\rm gas}$) versus dust mass ($M_{\rm env}$) for all transitions of DCO+ (left panel), H2CO (center panel), and c-C3H2 (right panel). This plot shows that the gas masses from DCO+ and c-C3H2 are consistent between transitions, while H2CO shows a broader spread. In the center panel, the gas mass from H2CO 3$_{2,2}$--2$_{2,1}$ is not shown to avoid a crowded plot, but follows a similar distribution as the other transitions shown in the panel.
• Figure 3: Relations of the DCO+ $J$=1--0 gas mass ($M_{\rm \ce{DCO+}}$, left column), H2CO $J$=3$_{0,3}$--2$_{0,2}$ gas mass ($M_{\rm \ce{H2CO}}$, center column), and c-C3H2 $J$=3$_{3,0}$--2$_{2,1}$ gas mass ($M_{\rm \ce{c-C3H2}}$, right column) versus envelope dust mass ($M_{\rm env}$) for the sample. In all panels the red dash-dotted line shows the N2H+ gas mass versus envelope dust mass from Paper I as reference. The full sample with no differentiation for multiplicity is shown in the top row. In the other three rows, orange stars indicate multiple protostellar systems, cyan diamonds show binary systems, and gray circles indicate single protostellar systems. Each row represents one of four ways of grouping the sample and their corresponding correlations (see Sect. \ref{['subsec:stats']}). Lines and shaded areas show the linear regression for the data with the corresponding color. Solid lines indicate statistically significant correlations in the subsamples (Censored Kendall rank correlation p-value $<$ 0.05), while the dashed line shows subsamples with p-values $>$ 0.05.
• Figure 4: Comparison of envelope dust masses calculated with SCUBA (x-axis, COMPLETE survey, ridge2006) and SCUBA-2 (y-axis, JCMT GBS, kirk2018) data. The top panel shows the dust masses per source in our sample (i.e., APEX pointing), while the bottom panel shows the dust masses per system (i.e., summed masses of all components in multiple protostellar systems).
• Figure 5: A sample of APEX spectra toward sources from the extended sample (see Sect. \ref{['sec:observations']}). The spectra for DCO+ $J$=5--4 and p-H2CO $J$=5$_{0,5}$--4$_{0,4}$ were obtained with APEX-2 with a beam of 18$\arcsec$, the other spectra were obtained with APEX-1 and a beam of 28.7$\arcsec$. From left to right, the first two panels show spectra for the components of the system B1 Per6+Per10. The third panel shows spectra for the close binary L1455 Per17. The fourth panel shows spectra for the single protostellar source NGC1333 RNO 15 FIR. The spectra are averaged to 0.4 km s$^{-1}$ in order to increase sensitivity. All emission lines shown are found to have S/N $\geq$5 based on GILDAS/CLASS gaussian fitting routines.