High-velocity gas associated ultracompact HII regions

Vol. 45 No. 7SCIENCE IN CHINA(Series A)July 200ASTRONOMYHigh-velocity gas associated ultracompact Hl regionsⅫUYe(徐烨)2, JIANG Dongrong(蒋栋荣)2, YANG Chuanyi(杨传义)°,ZHENG Xingwu(郑兴武)4, GU Minfeng(顾敏峰)2& PEI Chunchuan(裴春传)1. Shanghai Astronomical Observatory, Chinese Academy of Sciences, Shanghai 200030, China;2. National Astronomical Observatories, Chinese Academy of Sciences, Beijing 100012, China3. Astronomy Department, Columbia University, New York, NY, 10027, US:4. Department of Astronomy, Nanjing University, Nanjing 210093, China;5. Purple Mountain Observatory, Nanjing 210008, ChinaCorrespondenceshouldbeaddressedtoXuYe(email:xy@center.shaoac.cn)Received August 6, 2001abstract We present the results of a survey for high-velocity 2co(1-0)emission associated H2Omasers and ultracompact(UC)Hll regions. The aim is to investigate the relationship between H2Omasers, Co high-velocity gas(HVG) and their associated infrared sources. Our sample satisfiesWood Churchwell criterion. Almost 70 % of the sources have full widths(FWs)greater than 15km.s at Ta=100 mk and 15% have FWs greater than 30 km.s. In most of our objects thereis excess high velocity emission in the beam. There is a clear correlation between CO line FWsand far-infrared luminosities: thecreases with the FIR luminosity. The relation suggests thatmore luminous sources are likely to be more energetic and able to inject more energy into theirsurroundings. As a result, larger FW of the Co line could be produced. In most of our sources, thevelocities of peak of the H2O emission are in agreement with those of the co cloud, but a numberof them have a large blueshift with respect to the Co peak. These masers might stem from theamplifications of a background source, which may amplify some unobservable weak masers to anobservable levelKeywords: HIl regions, ISM: jets and outflows, stars: formation, masersIt is now generally believed that during the earliest stages of evolution, most, if not all,starundergo a phase of very energetic mass ejection, frequently characterized by the occurrence ofmassive bipolar outflows of cold molecular gas. These outflows appear to be driven by strongstellar disk winds, whose other notable manifestations include rapidly moving Herbig-Haro(HH)objects, high-velocity water maser sources, shock-excited molecular hydrogen emission regionsand optically visible jets appearing to emanate directly from the immediate circumstellar environsof the young driving stars themselvesSome relationships between H2O masers, iraS obiects and outflow properties have beenfound!4,6,8,13-16 For instance. H,o luminosities correl中国煤化工 d thpeak velocity of the H2O are close to central velocityCNMHGSt of thples were associated with low and intermediate mass formation regions. This is partly becauselow-mass young stellar objects (YSOs) are far more abundant than high-mass ones and massive938SCIENCE IN CHINA (Series A)Vol 45star formation occurs in distant(> 500 pc), dense stellar clusters. So identification of the drivingsource is rather difficultShepherd Churchwell(hereafter SC) surveyed high velocity gas in 122 massive star for-mation regions and found that 90 of their samples were associated with high-velocity gas(fullwidth greater than 15 km.S), which are generally significantly higher than those toward lowmass outflow sources. At the same time, they also found that the full widths are also far less thanose seen inlow star formation regions" 7. Later, by mapping the co line toward a sample of 69luminous iras point sources, Zhang et al. found 90 %o of the sample to possess molecular outflows. This value is also far higher than those seen in low star formation regions 8Based on the far-infrared spectra of the known ultra-compact HIl regions, WoodChurchwell (hereafter wC criterion) evolved a two-color criterion that characterizes themg(F2m/F12m)>0.57 and lg(F6m/F12m)>1319In this paper, we produce a sample to further study the relationships betweeCO HVG and associated IRas objects. We choose a sample of H2O masers for high-velocitymolecular emission 20-22. The whole sample consists of 77 objects and all sources coincide withIRAS point sources. Our selection criteria are the following: (i) Colors satisfy WC criterion;(ii)8>-10ObservationThe observations of the J=1-0 transition of Co were made with the 13. 7 m telescope atthe Qinghai Station of Purple Mountain Observatory, United Radio Astronomy Laboratory of theChinese academy of Sciences, in March and December, 2000. This antenna is located in the gobidesert, a very dry and arid region 3200 m above sea level in western China. The beam size at thefrequency 115 GHz is 54. The pointing and tracing accuracy are better than 10". A cooled superonducting mixer and a FET IF-I were employed. Its double-side band system temperature islower than 250 K. The acoustic-optical spectrometer provides 1024 channels for a total bandwidth of 170 MHz. The frequency resolution is 255 kHz. The antenna efficiency is 45%050%depending on the elevation. The observations were made in position-Switching mode. The integration time is 4 minutes. The rms is about 0.20 K and the absolute calibration for flux density wasabout 202 Results and data reductionData reduction was performed using the Draspec package. In order to increase the signal-to-noise ratio, several spectra of the same source were added to lead the integration time to bemore than 8 minutes for every source. All data were spa中国煤化工way to a reso-lution of 108". After smoothing, the rms noise was reduCNMHGH,O maser sources and the observationaltabulated in table 1. The first three columns of table I present the corresponding IRAS sourceHIGH-VELOCITY GAS ASSOCIATED ULTRACOMPACT HII REGIONS939Table I Parameters of CO spectra and H2ODEC(1950)FWHMFWm/km·s00211+6542109.66549267.1112.12.7×1000260+5624002600.0563.6389-40.69.014×103-28.716.102232+6138022317.761385822148448.818.19.8×103259+310503255793105500.30.3×10203414+320003412803200000.3518.212.85,6×1004579+470304575664703032711-18316.91304.0×10305137+3919051345839190106666.614.11.2×1005168+3634051653.636342162.0x05274+345052727.63345371.84-494.322.144×10305345+3556053435935565716.5216×10405391-0217053907.7021729187112.815.11.2×104553+1631055520.31631461.214.51006006+30150600414301504-35,124.83×10406067+2138060647821384729.17.2×100617+1350061147.11350344019.458×10406291+042106290930421441612.209×10306306+0437063037.30437161.1618.1.1x06501+014306500980143586.043.952.08.0×100657904328265+0028182632.600283514.117.134×10+18315-090318312970903434463.70×10183160602183139.0060207340.145,242.43.4×10418319-0903183156.209030412.855.677.245.76.6×1030650183532.60650345.895,92.57.6×105184030417184019.50417016095.1184690132184658.60132256.187493.67230.81.8×1058507+0121185045.20121093.655.534.228×1018517+0437185145.30437423.1643.243.51501.8×10418530+0215185303.302151312.87692425.8×10518545+0134185434.601344912.746.447.73133.5×1019074+081419072560814445.313.127.8.1×104+0902192309022512.841.11661.6×1019207+1410192044.61410506.663.468.611.226.55.6×10519213+173219212291726.80×10319268+1754192649.51754540.522.91764.2×1029325+1925193233.819250339,1-11.017.39366+2301193640.123014231.210.719374+2352193726.423122.0×10419388+2357193852.52357364.36中国煤化工1811.1×101989330195857033204783 CNMHG253761019410+2336194104.222713.6×10420050+272020050252720092.7×104To be continued on the next page)940SCIENCE IN CHINA (Series A)Vol 45R.A.(1950)DEC(195)FWHMIRAS name/km·s/km/km·s1/km0056+335020053603350531.310.41.8×10320081+312220080993122421012583.8×10316.412.83.3×10320126+410420124104104205014.815082×1020144-3726201424.737262210.5121-5762.6×104160+3911201605.2391223.814.320215+3725202135637251527120.1-14.7×1044.014.31.6×10320227+4154202246.341542919.120231+34402010.73440165.21214914.152×10320286+4105202840.64105393.918.13.4×1020350+4126203504.84126024871-3.423.15.02579.5×10421007+49512100449495113074-1.93,2×1021008+4700210049147004266423-49614.2×10421078+5211210750.2521129191-7.810.11.5×10421368+5502213648.355025710.11-67.1-83.812.174×10321391+5802213110.35802290840.71762.6×10221413+544221212544230757-65.018×10521479+55102147591551047861303.2×10321553+5908215523459081910.689.91.8×1022506+5944225038.75944505.0651,7-47,94×10422517+6215225143.7621011305.8×105×105225637.05830525.311-51.719.19.3×1044.5×10423004+564223002705642455611-548562907.8×10317.11×10523116+6111231136.06157,2-60.0629×105name and their 1950 equatorial coordinates. Column 4 gives the distances D, of which some arequoted from the corresponding references and those that have no distances available are heliocentric kinematics distance, computed from the peak velocity of Co emission using the galactic rottion curve of Wouterloot Brand (1989). Columns 5 and 6 are the peak velocity of the H2O andthe Co emission, VH.o and Vco, respectively. Columns 7 to 8 list the measured half-powerwidths(FWHM) estimated from a gaussian fit to the line core, and the measured full width of theCo(=1 0) line at Ta=100 mK Column 9 is the iras luminosity L Fir which is calculated with the formula of Casoli et al. 24(Lo represents s中国煤化工3 DiscussionCNMHGCO spectra in most of our sources show broad line wing. Nearly 70 %o of the sources haveHIGH-VELOCITY GAS ASSOCIATED ULTRACOMPACT HII REGIONS941FWs greater than 15 km .s at Ta =100 mK, 15 have FWs greater than 30 km.s, and fourobjects even more than 40 km.s, as given in table 1. The FWs are generally substantially largerthan those seen toward low-mass stars. In low star formation regions, the FWs of most objects areless than 15 km. 23,25. In the results of SC, 90 of their sources have FWs greater than 15km.s almost 50 have FWs between 15 and 30 km.s. 30 have FWs between 30 and 45have FWs greater than 45 km.s. It is obvious that the FWs obtained by SCare significantly larger than those we observed. Such difference may be due to the FWs measuredat different noise levels. Their FWs measured at Ta =20 mK, whereas ours at Ta =100 mKSo, they can detect lower intensity of CO outflows, even extremely high velocity molecular gasThe measured FW versus the Fw predicted by a Gaussian fit to the line core at the samenoise level, GFW, is plotted in fig. 1. Excepting several sources in which measured line full widthsare less than or nearly equal to the expected values from Gauss fit, the vast majority have FWssubstantially greater than those predicted by a gaussian fit to the line core. It means that in mostof the objects there is excess high velocity in the beam, which should be originated from turbulence, rotation, collapse, or outflows. Seven of them have been mapped and the results show thatWe investigate the relation between the Fw and the luminosity of the corresponding IRASsources in fig. 2. A significant correlation is found at 99.99 %o confidence and correlation coeffiient 0.48 using Spearmans correlation coefficient. As a comparison, we also present the samerelation in SC sample, where 68 sources are available after the exclusion of those sources havingcomplex blend of lines and severe subtraction problems. There is also a good correlation in theirsample and the correlation coefficients and corresponding confidences are 0.47 and 99.99 %, respectively. This indicates that there is a possible connection between Co line FWs and far-infraredluminosities. A clear correlation is also found between the FWHM and the IRAs luminosity0510152025303540455055600210LAir(Lo)Fig.中国煤化工 C sample, open ciridths got with Gauss fit to the line core at the same levclesHlid circlesIt shows that most sources observed have Fw>GFwIRACNMHGIC v./ anu 99.99. 0.48 and99.99%, respectively, revealing that the correlations in both942SCIENCE IN CHINA (Series A)Vol 45shown in fig 3. The correlation coefficients and corresponding confidences are 0. 45 and 99.99 %,respectivelyMore luminous objects are often located at larger distances, and the pencil beam of the telecope covers a larger area for objects located at larger distances. Thus, the relative motions ofclouds within the beam can broaden the line. We examine the connection between the velocityrange of masers and the distances to the source and find no correlation between them. So, thishows that there is indeed a connection between co line fws and far -infrared luminosities TheFWs are proportional to the IrAs luminositiesAt the initial phases of a star formation, its increase in mass is mainly by accretion from acircumstellar accretion disk. The disk transports the materials to the central star and also producesa strong disk wind. This extremely fast wind, moving at velocities of several hundred km.smay very well drive the molecular outflows when it collides with the ambient cold interstellarmedium-. Zhang et al. found that IRAS sources are often located at/near the center of outflowThis indicates that the stars are likely to be the driving source of the outflow. Zhang et al. alsofound the flows associated with massive star formation have momenta generally substantially larger than those seen toward low-mass stars, which suggests that more luminous sources are likelyto be more energetic and would produce a stronger disk wind. As a result, more energy would benjected into their surroundings and thus produce a more energetic molecular outflow. This is ableto easily affect the Co emission and increase its FW, as shown in figs. 2 and 3: the larger thefar-infrared luminosity is, the larger CO line Fw would be producedThe H2O maser peak velocity versus the Co peak velocity is plotted in fig. 4. From the plotone can see that in most of these objects the peak velocities of H2O maser emission are close tothe CO emission peak velocities. In general, this indicates that the masers are well associated withthe co molecular cloud. It is notable that although H2o masers show a redshift and also a blue-shift relative to the corresponding cloud velocity, high velocity H2O masers only show a largeblueshift. These h,o masers have a velocity, >30 km .s relatively to co molecular clouds, two16l0010LEiR(Lo)T]中国煤化工Fig 3. Measured FWHM versus IRAS luminosity. TheCN MAGus thecorrelation coefficient and corresponding confidence arevelocity. The plot shows that a number of H2O m0.45 and 99.99%0, demonstrating that the correlation ishave a large blueshift relative to Co molecular clstrongHIGH-VELOCITY GAS ASSOCIATED ULTRACOMPACT HII REGIONS43objects even up to 120 km.sIf high-velocity H2O masers occur in outflow, they not only show a large redshift, but also alarge blueshift, but only large blueshifted one was detectedFirst, a large velocity difference suggests that the H2o masers may stem from different direc-tions because masers can in principle occur in either outflows or disks 27, whereas the velocity ofmasers from outflows or disks can be very different If a maser occurs in a disk its average veloc-ity could be close to that of the ambient gas while those occurring in outflows are able to get larger velocity relative to Co peak due to being accelerated. On the other hand, the masers may stemfrom outflow in different directions. Observations show that HO maser velocities can increasefrom about 20 to 200 km .s- in the outflow 28, 29 When stellar wind strikes denser material lowyelocity features may be produced and higher velocity ones may be produced when it strikes relatively less dense ambient material. Next, infalling protostellar envelopes or expanding compactHII regions may also shock dense gas and produce masers 50.31. Finally, masers may originatefrom clump-clump collisions. If collisions take place among clumps at sufficient relative velocities, the powerful energy can excite H20 maser emission532. For masers caused by differentmechanisms their velocities can be very different. Thus, although masers are associated with theame CO cloud well, a large velocity difference can take place between themThe large velocity difference between the H2O masers and the corresponding cloud can becaused by various mechanisms as already mentioned above. However, these H2O masers can notonly have a large blueshift, but also a large red-shift, but no large red-shifted one was detected inour samplcIf masers stem from expanding materials, the intensity of a redshifted maser may decay moresharply than that of a blueshifted one due to its much deeper optical depth along the line of sightwhen they pass though the Hll region which is likely to be optically thick at centimeter wavelengths. Therefore, the blueshifted masers are likely to stem from expanding materialsIf an unsaturated maser which may be unobservable amplifies a background source which iseither a radio continuum possibly in the center of the corresponding protostellar34-37) it can amplify the continuum flux to an observable level, and even becoming the most component Observations show that there is a large abundance of H20 in the outflows of massive stars 58.Mostprobably a lot of unsaturated H2O masers occur there and they could not be detected due to theirIn short, we suggest that the large blueshift masers may be caused by amplification of abackground continuum source by an unsaturated maser which was expanding far away associatedcentral star and the back ground continuum source is liTYHn of UC hil中国煤化工CNMHG4 ConclusionWe have performed a survey to search for high-velocity-Co(1-0)emission toward 7794SCIENCE IN CHINA (Series A)Vol 45sources associated with H2O masers and UC HIl regions. The results can be summarized as fol1)Most objects observed show Fw velocities greater than what might have been expected ifthe HVG were due solely to micro-turbulence and thermal broadening. This means that most ofthe objects have excess high velocity emission in the beam2) We find a significant correlation between Co line FWs and IrAs luminosities and it isalso true between FWHMs and IRAS luminosities: both widths are proportional to the iras luminosities. It means that high velocity outflow and extremely high velocity molecular gas are easier to occur in massive star formation region3)In most of our objects the peak velocities of the H2O masers are in agreement with theof the co hvg but a number of them have large blueshifts relative to the co molecular cloudsThey may be caused by amplification of a background source by an unsaturated maserAcknowledgements We want to thank all the staff at Qinghai Station, Purple Mountain Observatory for their assistanceduring the observation. We thank Y. Wu and J. Sun for helpful discussionsReferencesWouterloot, J. G. A, Brand, J, Fiegle, K, IRAS sources beyond the solar circle Ill. Observations of H2O. OH, CH3 OH andC,A&AS,1993Henning, T, Martin, K, Reimann, H et al., Multi-wavelength study of NGC 281A, a&A, 1994, 288: 282--292.tudies of luminous stars in nearby galaxies I. Supergiants and O stars in the Milky Way, ApJ Suppl1978,38:309350.4. Felli, M, Palagi, F. Tofani, G, Molecular outflows and HO masers: what type of connection?a& A, 1992, 255: 293Gueth, F, Guilloteau, s, The jet-driven molecular outflow of HH 211, a& A 1999. 343: 571--5846. Palla, F, Brand, J, Cesaroni, R, et al., What masers associated with dense molecular clouds and ultracompact HII regionsA&A,1991,246:249263.7. Baudry, A, Desmurs, J. F, wilson, T L. et al., A survey of star-forming regions in the 5 cm lines of oH, a& A, 1997, 325255-268.8. Henning, T, Cesaroni, R, Walmsley, M. et al., Maser search towards young stellar objects, a& A S, 1992. 93: 525-5389. Eiroa, C, Casali, M. M., Miranda, L. F et al., S269-IRS2: a massive young stellar object powering Herbig-Haro emission,A&A,1994.290:59960410. Williams, J P, Maddalena, R.J., A large photodissociation region around the cloud G216-2.5, Ap J, 1996, 464: 247--2511. Slysh, V, I, Val'tts, I. E, Kalenskii, S. V. et al., The Medicina survey of methanol masers at 6. 7 GHz, a &As, 1999, 134:115-1282. Palla, F, Cesaroni,R, Brand, J. et al., H]O masers associated with dense molecular clouds and ultracompact HIl regions IlThe extended sample, A&A, 1993, 288: 599--60813. Wouterloot, J G. A, Walmsley, C. M, H2O masers associated with IRAS sources in regions of star formation, a& a1986,168:237—2474. Codella. C. Felli, M. Natale, V et al.. The occurrence of H,o中国煤化工1:2612705. Xiang, D, Turner, B E, Newly discovered galactic H20HCNMHGpl,1995.99:12116. Wouterloot, J. G A, Fiegle, K, Brand, J. et al., RAS sources beyond the solar circle VI. Analysis of the FIR, H2O, and COemission,A&A,1995,301:236-260HIGH-VELOCITY GAS ASSOCIATED ULTRACOMPACT HII REGIONS14517. Shepherd, D.S., Churchwell, E, High velocity molecular gas from high mass star formation regions, Ap J, 1996, 45767-27618. Zhang, Q. et al., Search for CO outflows toward a sample of 69 high-mass protostellar candidates: frequency of occurrence,, ApJ Letter,2001,552:167-17019. Wood, D.O.S., Churchwell, E, Massive stars embedded in molecular clouds: their population and distribution in theGalaxy,ApJ,1989,340:265-27220. Han, F, Mao, R, Lu, J et al., New detections of H2O maser sources on the 13.7 m radio telescope of Purple MountaiObservatory, A&AS, 1998, 127: 181-18421. Comoretto, G, Palagi, F, Cesaroni, R et al., The Arcetri atlas of H2O maser sources, A&AS, 1990, 84: 179--22522. Brand, J, Cesaroni, R, Caselli, P. et al., The Arcetri catalogue of H2O maser sources update, a&A s, 1994, 10323, Wouterloot, J G.A. Brand, J, IRAS sources beyond the solar circle I CO observations, A&AS, 1989, 80: 149-1824. Casoli, F, Dupraz, C, Gerin, M. et al., CO and"CO observations of cold IRAS unidentified point sources in the galaxy.A&A,1986,169:281-29725. Wu, Y, Huang, M, He, J, A catalogue of high velocity molecular outflows, a&A S, 1996, 115: 28326. Shu, F. H, Najita, J, Ostriker, E. C. et al., Magnetocentrifugally driven flows from young stars and disk. V. Asymptoticollimation into jets, ApJ 1995, 455: L155--15827. Greenhill, L. J, Gwinn, C.R., Schwartz, C. et al., Coexisting conical bipolar and equatorial outflows from a high-massprotostar, Nature, 1998, 396: 650-65328. Gwinn, C.R., Moran, J. M, Reid, M. J, Distance and kinematics of the w49 HO maser outflow, ApJ, 1992, 393149—164.29, Gwinn, C.R., Hypersonic acceleration and turbulence of H2O masers in W49N, ApJ, 1994, 429: 241-25230. Leppanen, K, Liljestrom, T, Diamond, P. J, Submilliarcsecond linear polarization observations of water masers inWSM,ApJ,1998,507:90991931. Furuya, R.s., Kitamura, Y, Saito, M. et al., VLA observations of H2O masers in the class 0 protostar S106 FIR: evidencefor a 10 AU scale accelerating jetlike flow, ApJ, 1999, 525: 821-83132. Tarter, J. C, Welch, J. w, A cloud collision model for water maser excitation, Ap J, 1986, 305: 467-48333. Hollenbach, D, Johnstone, D, Lizano, S. et al., Photoevaporation of disk around massive stars and application to ultra-ompact HIl regions, Ap J, 1994, 428: 654--664. Deguchi, S, Watson, W. D, Interacting masers and the extreme brightness of astronomical masers, Part Ill: Filamentarymasers,ApJ,1989,340:L17-20.35. Tofani, G, Felli, M, Taylor, G B et al., Exploring the engines of molecular outflows. Radio continuum and HO maserobservations. a& A112. 299-3466. Torrelles, J M, Gomez, J. F, Rodriguez, L F, Systems with H2O maser and 1.3 centimeter continuum emission in Ce-pheu a,ApJ.1998.509:262-26937. Boboltz, D. A, Simonetti, J. H, Dennison, B. et al., A water maser flare in W49N: amplification by a rotating foregroundloud,ApJ,1998,509:25626138. Wright, C M, van Dishoeck, E. F, Black, J.H., ISO-sws observations of pure rotational H2O absorption lines towardOion-lrc2,A&A,2000,358:689700中国煤化工CNMHG