A GLIMPSE-based search for 6.7-GHz methanol masers and the lifetime of their spectral features

The University of Tasmania Mt Pleasant 26-m and Ceduna 30-m radio telescopes have been used to search for 6.7-GHz class II methanol masers towards 200 GLIMPSE ( The Galactic Legacy Infrared Mid-Plane Survey Extraordinaire ) sources. The target regions were selected on the basis of their mid-infrared colours as being likely to be young high-mass star formation regions and are either bright at 8.0 μm, or have extreme [3.6]–[4.5] colour. Methanol masers were detected towards 38 sites, nine of these being new detections. The prediction was that approximately 20 new 6.7-GHz methanol masers would be detected within 3.5 arcmin of the target GLIMPSE sources, but this is the case for only six of the new detections. A number of possible reasons for the discrepancy between the predicted and actual number of new detections have been investigated. It was not possible to draw any firm conclusions as to the cause, but it may be because many of the target sources are at an evolutionary phase prior to that associated with 6.7-GHz methanol masers. Through comparison of the spectra collected as part of this search with those in the literature, the average lifetime of individual 6.7-GHz methanol maser spectral features is estimated to be around 150 yr, much longer than is observed for 22-GHz water masers.     

A comprehensive search for extragalactic 6.7-GHz methanol masers

We have used the Australia Telescope Compact Array (ATCA) to search for 6.7-GHz methanol maser emission towards 87 galaxies. We chose the target sources using several criteria, including far-IR luminosities and the presence of known OH megamasers. In addition, we searched for methanol masers in the nearby starburst galaxy NGC 253, making a full spectral-line synthesis image. No emission was detected in any of the galaxies, with detection limits ranging from 25 to 75 mJy. This is surprising, given the close association of OH and methanol masers in Galactic star formation regions, and significantly constrains models of OH megamaser emission. This absence of maser emission may be a result of low methanol abundances in molecular clouds in starburst galaxies.     

Class I methanol masers: masers with extended green objects

We have compared the results of a number of published class I methanol maser surveys with the catalogue of high-mass outflow candidates identified from the Galactic Legacy Infrared Mid-Plane Survey Extraordinaire survey (known as extended green objects or EGOs). We find class I methanol masers associated with approximately two-thirds of EGOs. Although the association between outflows and class I methanol masers has long been postulated on the basis of detailed studies of a small number of sources, this result demonstrates the relationship for the first time on a statistical basis. Despite the publication of a number of searches for class I methanol masers, a close physical association with another astrophysical object which could be targeted for the search is still lacking. The close association between class I methanol masers and EGOs therefore provides a large catalogue of candidate sources, most of which have not previously been searched for class I methanol masers. Interstellar masers and outflows have both been proposed to trace an evolutionary sequence for high-mass star formation, therefore a better understanding of the relationship between class I methanol masers and outflow offers the potential for comparison and amalgamation of these two evolutionary sequences.     

The relationship between class I and class II methanol masers

The Australia Telescope National Facility Mopra millimetre telescope has been used to search for 95.1-GHz class I methanol masers towards 62 6.6-GHz class II methanol masers. A total of 26 95.1-GHz masers were detected, 18 of these being new discoveries. Combining the results of this search with observations reported in the literature, a near complete sample of 66 6.6-GHz class II methanol masers has been searched in the 95.1-GHz transition, with detections towards 38 per cent (25 detections; not all of the sources studied in this paper qualify for the complete sample, and some of the sources in the sample were not observed in the present observations).
There is no evidence of an anticorrelation between either the velocity range, or peak flux density of the class I and II transitions, contrary to suggestions from previous studies. The majority of class I methanol maser sources have a velocity range that partly overlaps with the class II maser transitions. The presence of a class I methanol maser associated with a class II maser source is not correlated with the presence (or absence) of main-line OH or water masers. Investigations of the properties of the infrared emission associated with the maser sources shows no significant difference between those class II methanol masers with an associated class I maser and those without. This may be consistent with the hypothesis that the objects responsible for driving class I methanol masers are generally not those that produce main-line OH, water or class II methanol masers.     

Near-infrared imaging of the environment of 6.7-GHz methanol masers

Near-infrared (NIR) images at I , J , H and K bands were made of 12 southern 6.7-GHz methanol maser sources. Astrometry accurate to 0.5 arcsec was obtained. The positions of known H  ii regions, water masers, hydroxyl masers, and mid- and far-infrared objects in the region are examined in order to try to determine the nature of the methanol maser source. Deeply embedded NIR sources were found close to seven out of 14 maser sites. In three cases, no NIR source, H  ii region, water maser or hydroxyl maser could be found in likely association with the methanol masers.     

Methanol masers at 107.0 and 156.6 GHz

From a search of more than 80 southern class II methanol maser sites, we report measurements of 22 masers at 107.0 GHz and four at 156.6 GHz, mostly new discoveries. Class II sites, recognized by their strong emission at the 6.6-GHz methanol transition, are indirect indicators of new-born massive stars, and several hundred have been documented; only a handful of these had previously been found to exhibit maser emission at the 107.0- or 156.6-GHz transition. The present survey increases the number of known 107.0-GHz masers to 25, providing a sufficiently large sample to assess their general properties. For the stronger ones, our position measurements confirm that, to an accuracy of 5 arcsec, they coincide with the dominant maser emission at 6.6 GHz. Intensity variations exceeding 50 per cent have occurred in some 107.0-GHz maser features that we observed in both 1996 October and 1998 June.
We find that masers are rare at the 156.6-GHz transition. Two new detections increase the total now known to four. Each 156.6-GHz maser is substantially weaker than its corresponding 107.0-GHz maser. Despite the scarcity of masers, our 156.6-GHz spectra at most observed sites show emission, but apparently of a quasi-thermal variety; it is usually accompanied by somewhat weaker thermal emission at 107.0 GHz, and the intensity ratio of the transitions allows us to begin exploration of the physical characteristics of the small molecular clouds (diameter less than 60 mpc) at these sites. The thermal emission thus provides estimates of the environmental conditions that are needed to support strong masing from spots that are apparently embedded within these clouds.     

A search for 4750- and 4765-MHz OH masers in southern star-forming regions

We have used the Australia Telescope Compact Array (ATCA) to make a sensitive  (5 σ ≃100 mJy)  search for maser emission from the 4765-MHz ²Π_1/2   F =1→0  transition of OH. 55 star formation regions were searched and maser emission with a peak flux density in excess of 100 mJy was detected toward 14 sites, with 10 of these being new discoveries. In addition we observed the 4750-MHz ²Π_1/2   F =1→1  transition towards a sample of star formation regions known to contain 1720-MHz OH masers, detecting marginal maser emission from G348.550−0.979. If confirmed this would be only the second maser discovered from this transition.
The occurrence of 4765-MHz OH maser emission accompanying 1720-MHz OH masers in a small number of well-studied star formation regions has led to a general perception in the literature that the two transitions favour similar physical conditions. Our search has found that the presence of the excited-state 6035-MHz OH transition is a much better predictor of 4765-MHz OH maser emission from the same region than 1720-MHz OH maser emission is. Combining our results with those of previous high-resolution observations of other OH transitions we have examined the published theoretical models of OH masers and find that none of them predicts any conditions in which the 1665-, 6035- and 4765-MHz transitions are inverted simultaneously.     

The association of OH and methanol masers in W3(OH)

We present single-baseline Multi-Element Radio-Linked Interferometer Network (MERLIN) measurements of excited OH 6.0-GHz masers and methanol 6.7-GHz masers for the source W3(OH). These allow us to compare the positions of individual maser spots of these two species to ∼15 mas accuracy for the first time, and to compare these with previously published positions of ground-state OH masers near 1.7 GHz and excited-state OH masers near 4.7 GHz. There is a strong association between OH 6035-MHz and 1665-MHz masers. OH and methanol have very similar distributions, but associations of individual masers are relatively rare: most methanol 6.7-GHz masers are within 100 mas of OH 6.0-GHz masers, but only four methanol masers are within 15 mas of an OH 6.0-GHz maser. There are no correspondences of either species with excited OH 4.7-GHz masers. Zeeman splitting of the 6.0-GHz OH lines indicates an ordered magnetic field ranging from 3.2 to 14.4 mG. The magnetic fields estimated from co-propagating masers such as 6035 and 1665 MHz are generally in good agreement with each other.     

Methanol masers and their environment at high resolution

We have used the Australia Telescope Compact Array (ATCA) to make high-resolution images of the 6.7-GHz 5₁ → 6₀A⁺ maser transition of methanol towards 33 sources in the Galactic plane. Including the results from 12 methanol sources in the literature, we find that 17 out of 45 sources have curved or linear morphology. Most of the 17 have a velocity gradient along the line, which is consistent with masers lying in an edge-on circumstellar disc surrounding a massive star. We also made simultaneous continuum observations of the sources at 8.6 GHz, in order to image any associated H  ii region. 25 of the sources are associated with an ultracompact H  ii region, with a detection limit of ∼0.5 mJy beam⁻¹. We argue that the methanol sources without an associated H  ii region represent less massive embedded stars, not an earlier stage in the lifetime of the star, as previously suggested.     

Statistical Properties of 6.7 GHz Methanol Maser Sources

We present a statistical analysis of 482 6.7 GHz methanol maser sources from the available literature, on their maser emission and the characteristics of their associated infrared sources. On the color-color diagram, more than 70% of the objects fall within a very small region (0.57 ≤ [25 - 12] ≤ 1.30 and 1.30 ≤[60 - 12] ≤ 2.50). This suggests that 6.7 GHz methanol maser emission occurs only within a very short evolutionary phase during the earliest stage of star formation. The velocity ranges of the masers belong to two main groups: one from 1 to 10 km s^-1, and one from about 11 to 20 km s^-1. These velocity ranges indicate that the masers are probably associated with both disks and outflows. The correlations between the maser and infrared flux densities, and between the maser and infrared luminosities, suggest that far-infrared radiation is a possible pumping mechanism for the masers which most probably originate from some outer molecular envelopes or disks.     

Infrared emission from 6.7-GHz methanol maser sources

Near-infrared photometry was performed on 56 southern 6.7-GHz methanol maser sources. A simple spherically symmetric model of the radiative transfer through a dust shell was developed and used to study the conditions in the dust cloud in which the masers are produced. The parameters investigated were the size of the cloud, the spectral type of the embedded star, the optical depth of the dust cloud and the dust density distribution. It was found that the infrared colours of the models have a complex dependence on the parameters and that no unique combination of parameter values explains the spectral energy distribution of any particular source. The model effectively reproduces the far-infrared ( IRAS ) colours but cannot simultaneously explain the near-infrared colours for any of the observed sources.