Interaction of Massive Stars with Gas Clouds in the Milky Way: from shooting stars to breaking bubbles

Interaction of Massive Stars with Gas Clouds in the Milky Way Ümit Kavak

INVITATION

to attend the defence of my

doctoral thesis

Interaction of Massive Stars with

Gas Clouds in the Milky Way

From shooting stars to breaking bubbles

Paranymphs

Kristiina Verro

Pooja Bilimogga

on

Friday December 3, 2021

at 16:15, Rijksuniversiteit

Groningen, Aula

Followed by a reception

Interaction of Massive Stars with

Gas Clouds in the Milky Way

From shooting stars to breaking bubbles

PhD thesis

to obtain the degree of PhD at the

University of Groningen

on the authority of the

Rector Magnificus Prof. Cisca Wijmenga

and in accordance with

the decision by the College of Deans.

This thesis will be defended in public on

Friday 3rd December 2021 at 16:15 hours

by

Ümit Kavak

born on 16 February 1990

in Bakırköy, Turkey

Supervisors

Prof. F. F. S. van der Tak

Prof. A. G. G. M. Tielens

Assessment committee

Prof. J. Tan

Prof. G. J. Stacy

Prof. I. E. E. Kamp

To my family...

Cover design by: Yakup Kurt/Adgency Koeln. The background image on the

front cover is a star-forming region known as LH-95 in the Large Magellanic

Cloud taken from hubblesite.org. The interpretation of Starry Night on the

back-cover and figures at the end of each chapter were provided by Garip Ay.

The blue tulip created by Evren Kaan at page 222 was taken from kulturportali.gov.tr

with ID of #8942.

Printed by: Gildeprint

Contents

1 Introduction 1

1.1 Interstellar Medium . . . . . . . . . . . . . . . . . . . . . 2

1.2 Low and Massive Star Formation . . . . . . . . . . . . . . 3

1.3 Massive Stars . . . . . . . . . . . . . . . . . . . . . . . . . 7

1.3.1 Jets and Outflows . . . . . . . . . . . . . . . . . . 9

1.3.2 Photodissociation Regions . . . . . . . . . . . . . . 10

1.3.3 Interstellar Bubbles . . . . . . . . . . . . . . . . . . 13

1.4 Feedback from Massive Stars . . . . . . . . . . . . . . . . 15

1.4.1 Mechanical Feedback . . . . . . . . . . . . . . . . . 16

1.4.2 Radiative Feedback . . . . . . . . . . . . . . . . . . 16

1.5 Aims and Methods . . . . . . . . . . . . . . . . . . . . . . 18

1.5.1 Herschel Space Observatory . . . . . . . . . . . . . 18

1.5.2 Very Large Array (VLA) . . . . . . . . . . . . . . . 20

1.5.3 Stratospheric Observatory for Infrared Astronomy

(SOFIA) . . . . . . . . . . . . . . . . . . . . . . . . 21

1.5.4 Meudon PDR code . . . . . . . . . . . . . . . . . . 22

1.5.5 RADEX . . . . . . . . . . . . . . . . . . . . . . . . 23

1.5.6 GILDAS . . . . . . . . . . . . . . . . . . . . . . . . 23

1.5.7 CASA . . . . . . . . . . . . . . . . . . . . . . . . . 23

1.5.8 HIPE . . . . . . . . . . . . . . . . . . . . . . . . . 24

1.5.9 Python Libraries . . . . . . . . . . . . . . . . . . . 24

1.6 Goals of this thesis . . . . . . . . . . . . . . . . . . . . . . 24

1.7 Outline . . . . . . . . . . . . . . . . . . . . . . . . . . . . 25

2 Search for radio jets from massive young stellar objects.

Association of radio jets with H 2 O and CH 3 OH masers 29

2.1 Abstract . . . . . . . . . . . . . . . . . . . . . . . . . . . . 29

i

2.2 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . 31

2.3 Observations . . . . . . . . . . . . . . . . . . . . . . . . . 33

2.3.1 Selected sample . . . . . . . . . . . . . . . . . . . . 33

2.3.2 VLA observations . . . . . . . . . . . . . . . . . . . 35

2.4 Results . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 36

2.4.1 Continuum emission . . . . . . . . . . . . . . . . . 36

2.4.2 Spectral index analysis . . . . . . . . . . . . . . . . 37

2.4.3 Maser emission . . . . . . . . . . . . . . . . . . . . 41

2.5 Analysis and discussion . . . . . . . . . . . . . . . . . . . 41

2.5.1 Nature of the radio continuum emission . . . . . . 42

2.5.2 Association with molecular outflows . . . . . . . . 44

2.5.3 Association with EGOs . . . . . . . . . . . . . . . 45

2.5.4 Association with masers . . . . . . . . . . . . . . . 48

2.5.5 Radio-jet candidates . . . . . . . . . . . . . . . . . 48

2.6 Implications for high-mass star formation . . . . . . . . . 57

2.7 Summary . . . . . . . . . . . . . . . . . . . . . . . . . . . 58

2.8 Acknowledgements . . . . . . . . . . . . . . . . . . . . . . 60

2.9 Comments on individual sources . . . . . . . . . . . . . . 60

2.10 Catalog of the continuum sources . . . . . . . . . . . . . . 72

3 Origin of hydrogen fluoride emission in the Orion Bar.

An excellent tracer for CO-dark H 2 gas clouds 109

3.1 Abstract . . . . . . . . . . . . . . . . . . . . . . . . . . . . 109

3.2 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . 111

3.3 Observation and data reduction . . . . . . . . . . . . . . . 113

3.4 Results . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 116

3.5 Analysis . . . . . . . . . . . . . . . . . . . . . . . . . . . . 119

3.5.1 Column density . . . . . . . . . . . . . . . . . . . . 119

3.5.2 Spatial distribution of HF . . . . . . . . . . . . . . 122

3.6 Discussion . . . . . . . . . . . . . . . . . . . . . . . . . . . 124

3.6.1 Collisional excitation . . . . . . . . . . . . . . . . . 124

3.6.2 Infrared pumping . . . . . . . . . . . . . . . . . . . 129

3.6.3 Chemical Pumping . . . . . . . . . . . . . . . . . . 130

3.7 Summary . . . . . . . . . . . . . . . . . . . . . . . . . . . 130


3.9 Appendix . . . . . . . . . . . . . . . . . . . . . . . . . . . 133

3.9.1 SEDs of Three Positions in the HF map . . . . . . 133

ii

4 Breaking Orion’s Veil bubble with fossil outflows 139

4.1 Abstract . . . . . . . . . . . . . . . . . . . . . . . . . . . . 139

4.2 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . 141

4.3 Observations . . . . . . . . . . . . . . . . . . . . . . . . . 145

4.3.1 [C ii] Observations . . . . . . . . . . . . . . . . . . 145

4.3.2 Molecular Gas Observations . . . . . . . . . . . . . 147

4.3.3 Ionized Gas Observations . . . . . . . . . . . . . . 148

4.3.4 Far-IR photometric observations . . . . . . . . . . 148

4.3.5 Mid-IR Observations . . . . . . . . . . . . . . . . . 150

4.4 Results . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 150

4.5 Analysis . . . . . . . . . . . . . . . . . . . . . . . . . . . . 152

4.5.1 Expansion Velocity . . . . . . . . . . . . . . . . . . 152

4.5.2 Morphology of the protrusion . . . . . . . . . . . . 152

4.5.3 Expansion Timescale . . . . . . . . . . . . . . . . . 155

4.5.4 Line Profile Analysis . . . . . . . . . . . . . . . . . 155

4.5.5 Kinetic Energy and Momentum . . . . . . . . . . . 157

4.6 Discussion . . . . . . . . . . . . . . . . . . . . . . . . . . . 160

4.6.1 Persistence of fossil outflow cavity . . . . . . . . . 165

4.6.2 Ionizing source . . . . . . . . . . . . . . . . . . . . 166

4.6.3 Correlation of Intensities . . . . . . . . . . . . . . . 167

4.7 Conclusion . . . . . . . . . . . . . . . . . . . . . . . . . . . 169


4.9 Appendix . . . . . . . . . . . . . . . . . . . . . . . . . . . 170

4.9.1 Geometric correction Factor . . . . . . . . . . . . . 170

4.10 Additional Maps . . . . . . . . . . . . . . . . . . . . . . . 172

5 Unveiling the Veil: Protostellar feedback in Orion 183

5.1 Abstract . . . . . . . . . . . . . . . . . . . . . . . . . . . . 183

5.2 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . 185

5.3 Observations . . . . . . . . . . . . . . . . . . . . . . . . . 187

5.3.1 [C ii] observations . . . . . . . . . . . . . . . . . . 187

5.3.2 Molecular Gas observations . . . . . . . . . . . . . 189

5.3.3 Mid-IR observations . . . . . . . . . . . . . . . . . 189

5.3.4 Far-IR photometric observations . . . . . . . . . . 189

5.4 Identification of Dents . . . . . . . . . . . . . . . . . . . . 190

5.4.1 Position-velocity (PV) Diagrams . . . . . . . . . . 190

5.4.2 High-velocity [C ii] emission . . . . . . . . . . . . . 193

iii

5.4.3 Line profiles . . . . . . . . . . . . . . . . . . . . . . 193

5.5 Analysis . . . . . . . . . . . . . . . . . . . . . . . . . . . . 194

5.5.1 Characteristics of the dents . . . . . . . . . . . . . 194

5.5.2 Origin of the dents . . . . . . . . . . . . . . . . . . 196

5.5.3 Momentum of the dents . . . . . . . . . . . . . . . 196

5.5.4 Potential shock signature of the dents . . . . . . . 199

5.5.5 Collimation factor and opening angle . . . . . . . . 201

5.6 Summary . . . . . . . . . . . . . . . . . . . . . . . . . . . 202


5.8 Appendix . . . . . . . . . . . . . . . . . . . . . . . . . . . 204

5.8.1 Gaussian Fitting Results . . . . . . . . . . . . . . . 204

5.8.2 Massive Stars and Geometry . . . . . . . . . . . . 204

5.8.3 PV diagram of the dents . . . . . . . . . . . . . . . 205

6 Conclusions and Outlook 215

6.1 Summary and conclusions . . . . . . . . . . . . . . . . . . 215

6.2 Future Outlook . . . . . . . . . . . . . . . . . . . . . . . . 218

7 Additional Sections 223

7.1 Contributed Publications . . . . . . . . . . . . . . . . . . 223

7.2 Talks . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 224

7.3 Posters . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 224

7.4 Türkçe Özet . . . . . . . . . . . . . . . . . . . . . . . . . . 226

7.5 Nederlandse samenvatting . . . . . . . . . . . . . . . . . . 231


iv

Chapter 1

Introduction

When we look up to the sky at night, we see a fair where all the stars are

posing on the podium of the Milky Way. All the stars twinkle and make

the steps as if on a red carpet along the sky with the background music

of silence. We, as astronomers, climb mountains and send spectacular

cameras into space to explore the character of these stars. Not only are

astronomers interested in the fabled landscape of the Milky Way, but

many cultures throughout human history believe in the sanctity of the

mysterious sky. Based on their belief and imagination, they designate

the scintillations on their above with various names 1 .

Around 1600, brave people who had succumbed to their curiosity

built small-sized telescopes and pointed them at the sacred sky, to look

closely at the heavens. In the 19 th century, thanks to developing technology

and understanding of the nature of light, considerable improvement

was made by astronomers. William Herschel first showed the doors of

the heavens 2 because he found fewer stars than expected in some parts

of the holy sky. Medieval studies were unsatisfactory in exploring the

doors of the heavens. A more realistic proposal, first made by Edward

E. Barnard on the basis of photographs of the star-deficient regions, implied

dark markings on the sky were intervening opaque masses between

us and the stars (Barnard 1919). Barnard compiled the first catalogue

of dark markers on the sky, and over the next half century various on

these markers studies were made (Bok & McCarthy 1974). In the early

1

1 See list of names for the Milky Way on Wikipedia in different cultures

2 Medieval studies claimed that this area of the sky was the entrance of the heavens.

1

CHAPTER 1: Introduction

.......................................................................

20 th century, it was found that these dark markers also exist between the

stars and not just between the Earth and the stars. These results triggered

a revolutionary realisation about interstellar matter, or its current

terminology, the interstellar medium (ISM; Spitzer 1978).

1.1 Interstellar Medium

1

As a term, ISM refers to all the material found between stars in outer

space. The components of the ISM is mainly gas, dust, and other components

such as magnetic fields and cosmic rays. In the Milky Way, 1%

of the mass is in small solid particles as dust and 99% is in gas. The dust

grain sizes range from 0.35 nm to 1 µm in molecular clouds (Kennicutt &

Evans 2012). The baryonic components (i.e., besides dark matter) of the

Milky Way are compact objects (e.g., neutron stars and black holes), and

stars, in addition to the ISM. All these structures together, the Milky

Way galaxy, houses us on a rocky planet orbiting the Sun.

Today, we know that different types of galaxies exist in the Universe.

The Milky Way, as a spiral galaxy, is just one of these galaxies and is

made up of several components. Some galaxies have components similar

to those of the Milky Way. Therefore, the compilation of the properties

of dust and gas of the Milky Way is useful to understand the morphology

and dynamics of the Galaxy, and to compare to the properties of other

galaxies.

Like humans, stars have also a lifetime, and unlike humans, new

stars are born from the remains of earlier generations of stars. When

stars finish their lifetime (10 6 to 10 9 years on the main sequence) they

eject a large fraction of their material into the ISM either releasing their

outer parts slowly in a wind or via supernova explosions that have a

strong impact on the ISM. Different regimes of the ISM are considered

in order to understand its dynamics. The ISM is divided in the cold

neutral medium with hydrogen density n ∼ 20−50 cm −3 and temperature

T ∼ 50−100 K, warm neutral medium (WNM) with n ∼ 0.3 cm −3 and

T ∼ 8000 K, the warm ionized medium (WIM) with similar density

and temperature of WNM but hydrogen completely ionized, and the hot

ionized medium (HIM) with n ∼ 3 × 10 −3 cm −3 and T ∼ 10 6 K (Draine

2011, and references therein). There is a small fraction of the volume

and mass that has a higher thermal pressure than average and which

2

1.2 Low and Massive Star Formation

.......................................................................

Orion B

NGC 1981

NGC 1977

“orphan cluster”

Integral

shaped

filament

Ghost

filament

1 °

7.3 pc

NGC 1977

M43

M42

L 1641

cloud

Herschel N(H) map

WISE and N(H) map WISE 3.4 and 4.6 µm

Figure 1.1: Images of the Orion A star-forming cloud, showing the

integral-shaped filament, the two star clusters outside NGC 1981 and

1977, and the cloud L1641 to the South. Left: hydrogen atom column

density (N H ) map reconstructed from Herschel data, right: mid-infrared

image taken with the WISE space telescope (Lang 2014), center: combination

of the two images on the left and right panels. Image credit:

Amy M. Stutz/MPIA.

1

is gravitationally bound. The temperature of these regions is between

10−20 K. These regions are called molecular clouds and are the densest

parts (10 2 − 10 3 cm −3 on average) of the ISM (Kennicutt & Evans 2012).

As the densest and coolest part of the Galaxy, molecular clouds are the

formation sites of new stars.


Star formation is one of the most hotly debated questions in modern astrophysics.

It is difficult to establish a general theory for star formation

because a variety of physical processes occur on multiple scales in large

(∼10 pc), dense (n ≥ 10 3 cm −3 ), and cold (T ∼ 10 K) giant molecular

3


.......................................................................

1

Figure 1.2: Schematic for the formation of a low-mass star. Image credit:

Visser 2009.

clouds (GMCs) (Krumholz 2011; Rosen et al. 2020). GMCs have a hierarchical

structure ranging from dense cores (0.1 pc) to the large-scale

clumps (a few pc) within the Galaxy. In the last decade, several missions

have mapped (see also Fig. 1.1; Dame et al. 2001; André et al. 2010)

star-forming regions on pc and kpc scales. The densest condensations

in the GMCs are gravitationally confined cores, also known as prestellar

cores. The classification of stars depends on their mass: low-mass (< 8

M ⊙ ) and massive stars (> 8 M ⊙ ). The main difference is that massive

stars arrive on the main sequence while still accreting matter. At that

point, the luminosity of the star is provided by nuclear burning in the

core (Zinnecker & Yorke 2007).

The formation of low-mass stars has been better explained and modelled

than that of massive stars (see Fig. 1.2; Shu et al. 1987; Evans 1999;

André 2002; McKee & Ostriker 2007; André et al. 2008). In the low-mass

star formation scenario, prestellar cores collapse under their own gravity

(Fig. 1.2). The formation of cloud cores is thought to proceed via

filament instabilities (e.g., Hacar et al. 2018). This collapse results in

4


.......................................................................

the formation of a central object which is the nascent protostar because

of conservation of angular momentum. The internal temperature of the

central hydrostatic core increases when it becomes optically thick. In

this phase, the protostars are embedded in a gas and dust envelope.

The early stages of star formation are observed at millimetre and submillimeter

wavelengths, tracing dust continuum emission and molecular

emission from the outer layer of the envelope (André et al. 2008). During

formation, the newborn protostar, still embedded in its natal cloud,

energetically ejects some (∼10%) of its mass into the ISM in a bipolar

highly collimated supersonic jet and less collimated outflows (Arce et al.

2007). These ejections allow the initially deposited angular momentum

to dissipate, driving accretion of disk material onto the protostar. In

addition, the ejected material via shocks can accelerate surrounding gas

to velocities of 10−100 km s −1 in low-mass star-forming regions (Richer

et al. 2000). The typical size of the outflow can be 0.1 to 1 pc with a

collimation factor (ratio of length/width) of 3 or 20. After the envelope

mass is accreted or ejected through jet activity, the protostar reaches

the zero-age main sequence (ZAMS) and evolves into a mature star with

a remnant disk. Subsequently, the disk is cleaned up by gas and dust

condensation and planets and small bodies such as cometary bodies are

formed (McKee & Ostriker 2007).

The formation of massive stars is not as clear-cut as that of lowmass

stars, since massive stars are rare and tend to form within groups

of stars, in other words, higher multiplicity. Moreover, massive stars

are highly obscured by circumstellar dust (with extinction A V ≥ 100)

and are located at large distances (a few kpc) (Zinnecker & Yorke 2007;

Motte et al. 2018). Although their number is smaller than that of lowmass

stars, their strong radiation, powerful outflows, and winds play

a crucial role in shaping the structure and energetics of the ISM on

various scales, and affect galaxy formation and evolution (Walter et al.

2005), and even the re-ionization of the early universe (Kennicutt 2005).

Therefore, understanding the formation and impact of massive stars is

one of the main considerations of modern astronomy, and is the focus of

this thesis.

High-mass stars form in molecular cloud complexes along a filamentary

structure termed a ridge or hub on a parsec scale (1-10 pc) (see

step 1 in Fig. 1.3). Ridges and hubs collapse to create IR-quiet massive

1

5


.......................................................................

1

Figure 1.3: Schematic evolutionary diagram proposed for the formation

of high-mass stars as a function of time (Motte et al. 2018).

dense cores over a timescale of 10 4−5 years. Additionally, the collapse of

ridges/hubs leads a gas stream into the massive dense core, increasing

the mass of the core, which has an average size of 0.1 pc (step 2). At

this stage, massive dense cores host low-mass stars. In 3 × 10 5 years, an

IR-quiet massive dense core becomes a protostar housing a low-mass star

embryo (steps 3 to 4). If a gas stream can feed a low-mass star embryo

with a mass less than 8 M ⊙ , the latter develops into a high-mass star. At

6

1.3 Massive Stars

.......................................................................

this stage, high accretion rates and strong outflows are seen pulling angular

momentum out from the star-forming system. Massive protostars,

in contrast to their low-mass counterparts, reach the main sequence before

accretion is halted (step 5). UV photons with energies higher than

13.6 eV ionize the hydrogen in the surrounding material, resulting in

ionized gas regions (step 6; see also Kurtz 2005). Following that, expansion

of the ionized gas and radiation pressure on the surrounding

environment compete with accretion for the continued development of

massive stars. In the rest of this chapter, we summarize the phenomena

observed in massive star-forming regions, such as jets and outflows,

photodissociation regions (PDRs), and HII regions.

1.3 Massive Stars

Massive stars ionize and energize the ISM via UV radiation and fast

winds throughout their lifetime and via supernova ejecta by their death.

The interaction of massive stars with the ISM leads to a variety of remarkable

astronomical phenomena. Notable objects are jets and outflows

(Arce et al. 2007; Anglada et al. 2018; Kavak et al. 2021), PDRs (Tielens

& Hollenbach 1985a; Hollenbach & Tielens 1999), supernova remnants

(Thielemann et al. 2011), ultracompact HII regions (Churchwell 2002),

and hot molecular cores (Kurtz et al. 2000). On the other hand, massive

stars play a crucial role in the HIM of ISM. Moreover, massive stars are

believed to be critical in galaxy evolution because the escaping ionized

gas from the massive star-forming regions is thought to be the source of

ionized gas in the early universe at redshifts z = 6−20 (Abel et al. 2002).

Hence, understanding high mass star formation is at the core of many

key questions of modern astrophysics (Zinnecker & Yorke 2007).

One of the ways to build a global picture of the formation of massive

stars is to study the similarity between low- and high-mass star formation

(Tan 2016). To do this, phenomena in low-mass star-forming regions are

searched for in massive star-forming regions such as jets, outflows, and

disks (Beuther et al. 2002a; López-Sepulcre et al. 2011; Sánchez-Monge

et al. 2013a; Johnston et al. 2015; Cesaroni et al. 2016; Rosen et al. 2020;

Kavak et al. 2021).

1

7


.......................................................................

1

Figure 1.4: In the upper panel, a close-up view of HH 34 and its driving

jets in the upper panel (Credit: ESO) is oriented to facilitate the comparison

with the image in the lower panel, which is a cartoon depicting

the major components of a protostellar outflow lobe. The sizes of the

disk (purple), the poloidal component of the disk and stellar magnetic

fields (red), and the biconical molecular outflow are greatly exaggerated.

Thick, colored bands trace cavity shocks and UV-heated gas along the

cavity walls, while spot-shocks indicate supersonic velocity changes in

the jet. Forward shocks were highlighted in bright green whereas reverse

shocks were highlighted in magenta in both the terminal and internal

working surfaces. Low-J CO emission were depicted in blue, high-J CO

emission in green, and shock-heated H 2 emission in yellow. The atomic

or ionized cavity wall is shown by the dashed yellow line. The figure in

the lower panel is from Bally (2016).

8

1.3 Massive Stars

.......................................................................

1.3.1 Jets and Outflows

Jets and outflows from young stellar objects (YSOs) are two phenomena

(see Fig. 1.4) that help removing angular momentum observed in lowmass

star-forming regions (Anglada et al. 2018; Frank et al. 2014). Jets

are the first type of ejection observed during the early stages of star

formation. When a jet is expelled along the polar axis of the YSO, it

opens a channel on its velocity vector, allowing outflows to travel into the

surrounding cloud. As a result, on large scales, outflows emerge slightly

later than jet ejection. Both energetic mechanisms are launched from

the protostellar disk around newly born stars (Ray et al. 2007). They

carry sufficient energy and momentum to play a role in shaping their

parental molecular cloud and even regulating star formation by triggering

and perturbing the surrounding environment on cloud-scales (Arce et al.

2007; Federrath et al. 2014). Hence, feedback of jets and outflows are

a significant part of the evolution of molecular clouds. To understand

this, their physical and kinematic properties must be studied at some

100 AU scales to understand their interaction with the environment and

evolution in time.

Protostars are embedded in a thick envelope of gas and dust, so that

direct estimation of their position is non-trivial. Jets are found to be

most prominent near the protostar in all low-mass star-forming (Anglada

1996) regions. Since low-mass stars in nearby molecular clouds in Taurus

tend to form in relative isolation, the position of the protostar can be

estimated from jets and larger-scale outflows. However, massive stars

form in crowded regions (Bally 2016), so linking massive protostars with

large-scale outflows is not straightforward. Nevertheless, the presence of

outflows in massive star-forming regions has been part of the discussion

over the last two decades. A dozen surveys have investigated outflows

(Beuther et al. 2002a; Lopez et al. 2014; Sánchez-Monge et al. 2013d)

and jets (Sanna et al. 2018; Purser et al. 2018) in massive star-forming

regions. Rosero et al. (2019) found that almost half of their sample

lacks thermal jets 3 that can be detected with high-resolution observations

(∼1 ′′ ) in the radio wavelength range (i.e., a few centimetres). This raises

the question whether massive stars form via disk-mediated accretion as

1

3 The spectral index (α) is defined as change of flux as a function of frequency.

Given frequency (ν), S ν ∝ ν α , where S ν is the flux density and ν is the frequency.

For thermal sources α > −0.1 and non-thermal α < −0.1.

9


.......................................................................

1

Figure 1.5: Schematic diagram of a PDR illuminated by the strong interstellar

radiation field (ISRF) or nearby hot stars from the left. The

dissociation front for the Orion Bar is 15 ′′ (about 0.03 pc) from the ionization

front. The layer between black and red dashed lines indicate

CO-dark H 2 gas region. The snow line refers to the point where molecular

gases start to freeze and grains become coated by ices (Goicoechea

et al. 2016).

low-mass stars. In other words: Is massive star formation a scaled-up

version of low-mass star formation? A recent magneto-hydrodynamic

(MHD) simulation by Kölligan & Kuiper (2018) argues that a molecular

cloud with 100 M ⊙ core can produce powerful jets like their low-mass

counterparts, and the main difference between low-mass and massive

stars is how the protostar is embedded within the cloud. Further studies

(observations and simulations) are therefore needed to fully characterize

the jets and outflows from massive protostars.

1.3.2 Photodissociation Regions

Once a massive star emits UV radiation into the ISM, this radiation

travels away from the surface of the star into the ISM until it encounters

10

1.3 Massive Stars

.......................................................................

the surface of a nearby molecular cloud. The penetration of the UVphotons,

which are emitted by massive stars leads to bright emission at

the edges of molecular clouds. These regions are called ’photodissociation

regions’ (PDRs) (Hollenbach & Tielens 1999; Wolfire et al. 2003) 4 .

All of the atomic and most of the molecular gas in the Galaxy are in

PDRs that have a layered structure which is shown in Figure 1.5. In

these zones, the radiation field coming from high-mass stars regulates

the physical and chemical conditions of the gas. A PDR can also be defined

as an interface region between an HII region and a molecular cloud.

In these transition regions, hydrogen is mostly molecular but carbon is

mostly ionized (Kennicutt & Evans 2012). Therefore, these regions allow

us to understand the chemistry, thermal balance (between heating and

cooling), and evolution of the interstellar medium.

Among others, PDRs have been modeled by Tielens & Hollenbach

(1985a); Hollenbach et al. (1991); Kaufman et al. (1999); Le Petit et al.

(2006); Bisbas et al. (2012). The main parameters of these models are

the gas density (n), and the incident far-ultraviolet intensity (G 0 ). The

models aim to understand the physical conditions by considering thermal

balance, chemistry, and radiative transfer through a PDR layer. The

main heating mechanism in the layered structure is the photo-electric

effect on dust grains. The FUV photon flux gradually decreases into the

PDR which results in a layered structure where chemical transitions such

as H + → H → H 2 and C + → C → CO occur (van der Tak et al. 2012a).

The gas temperature is mainly governed by heating processes driven by

UV radiation (photoelectric heating, H 2 excitation, and dissociation).

Near the surface (A V < 4 mag), heating is mostly dominated by the

photoelectric effect on Polycyclic Aromatic Hydrocarbons (PAHs). The

dominant cooling processes are far-infrared (FIR) fine structure lines

such as [C ii] 158 µm, [O i] 63 µm, [S ii] 35 µm, [C i] 609 and 370 µm, and

H 2 pure rotational lines (van der Wiel et al. 2009; Koumpia et al. 2015).

Regarding the density distribution within PDRs, the presence of

small-scale density variations, called clumps have been observed by several

researchers using single dish (Stutzki et al. 1988; Hogerheijde et al.

1

4 Photodissociation regions have also been called ’photon-dominated regions’. In

this thesis, we prefer ’photo-dissociation regions’ because it implies the presence of

UV photons (6 eV < hν < 13.6 eV or 91.2 nm−200 nm wavelength region or Far

Ultraviolet (FUV) photons) as preferred by Hollenbach & Tielens (1999).

11


.......................................................................

175

150

J = 2

HF

J = 7

CO

GHz

∆E/kB [K]

125

100

75

50

25

0

J = 1

J = 0

2463.4 GHz

2 P 3/2

[CII]

1900 GHz

1232.4 GHz

2 P 1/2

J = 6

J = 5

J = 4

J = 3

J = 2

J = 1

J = 0

806.6

691.4

576.2

461.0

345.8

230.5

115.2

1

Figure 1.6: Energy level diagrams for HF J = 1-0 and 2-1, fine structure

lines of [C ii] and the first seven rotational levels of CO. The rotational

J levels and transition frequencies in GHz units are shown next the

proper place. See website of Leiden Atomic and Molecular Database

(LAMDA) for energy levels and transition frequencies: https://home.

strw.leidenuniv.nl/~moldata/.

1995; Wang et al. 1993) and interferometric observations (Young Owl

et al. 2000; Goicoechea et al. 2016). The density of the clumps varies

from 1.5 × 10 6 to 6 × 10 6 , while the interclump medium estimate yields

relatively lower densities of 10 4 − 10 5 (Lis & Schilke 2003). The density

parameter in other PDRs remains uncertain as high angular resolution

is needed.

H 2 is (almost) impossible to observe, because of the large level spacing

and because ∆J = 1 transitions are forbidden. The best molecule

that can be used is CO molecule as a proxy for H 2 because it is most

abundant molecule after H 2 and because it is easy to populate higher

energy levels via collisions in dense molecular clouds (see Fig. 1.6). In

12

1.3 Massive Stars

.......................................................................

addition, the CO molecule has numerous pure rotational transitions (see

Fig. 1.6 for the lowest seven transitions of CO), especially the ground

state transition (J = 1-0) which is associated with the cold regime (with

a typical temperature of 10 to 30 K) of the molecular cloud. However,

strong UV radiation of nearby massive star(s) dissociates CO molecules

near the surface of a molecular cloud and causes the ionization of carbon

atoms. In this layer of the molecular cloud, ionized carbon has a significant

abundance. These regions are called CO-dark H 2 gas (see also

Fig. 1.5; Madden et al. 1997; Leroy et al. 2011; Langer et al. 2014).

Alternative ways and tracers have been sought to understand the

density of CO-dark H 2 gas and to obtain a complete picture of it. The

presence of CO-dark H 2 gas in the Galaxy is inferred via various observations

such as dust emission (Reach et al. 1994), γ-rays (Grenier

et al. 2005), and [C ii] 158 µm (Langer et al. 2014). Guzmán et al.

(2012b) presented CF + J = 1-0 (∆E/k B = 4.92 K at 102.587 GHz)

as a useful tracer tracer of CO-dark H 2 using Horsehead PDR observations.

Among proposed tracers, the [C ii] 158 µm line stands out

as the best tool for several reasons. First, [C ii] 158 µm is observed

as emission in a significant fraction of the Galactic ISM (Pineda et al.

2014) and in other galaxies (Madden et al. 1997). The fine-structure

transition ( 2 P 3/2 → 2 P 1/2 at 158 µm or 1.9 THz, i.e., ∆E/k B = 91.2 K,

see also Fig. 1.6) is the most important and dominant cooling line of

the warm neutral medium (T ∼ 50−300 K) and intermediate density

(10 3 − 10 4 cm −3 ). Velocity-resolved [C ii] line observations provide us

invaluable information about the dynamics and kinematics in CO-dark

H 2 gas (Goicoechea et al. 2015; Pabst et al. 2019). Hence, [C ii] 158 µm

observations are the best tracer for the interaction of massive stars with

their environment. The disadvantage of [C ii] observations at 158 µm

is that it is not accessible from the ground because the Earth’s atmosphere

blocks FIR radiation at certain frequencies (Risacher et al. 2016).

Therefore, space-based (e.g., Herschel) or stratospheric (SOFIA, STO2,

or GUSTO) observatories are required for [C ii] emission at 158 µm.

1

1.3.3 Interstellar Bubbles

One of the most apparent manifestations of newly formed massive stars

in the Galaxy is HII regions. The formation of HII regions around massive

stars is discussed in Sect. 1.2. The notation HII (in spectroscopic

13


.......................................................................

1

notation H + ) means that hydrogen atoms are ionized by photons which

have energies (hν) more than 13.6 eV. HII regions are created by extreme

ultraviolet radiation from stars, e.g., O− and early B−type stars, L ∗ >

10 4 L ⊙ , with a effective temperature T ∗ > 20000 K, (Ward-Thompson &

Whitworth 2015). The ionization is balanced by recombination of thermal

electrons with protons. The balance between photo-ionization and

radiative recombination determines the degree of ionization. The excess

energy over the ionization potential is carried away as kinetic energy by

the photo-electron (Tielens 2010).

Many HII regions are ionized by the radiation from several stars, for

example an OB association, or a subgroup of an OB association (Blaauw

1964; de Zeeuw et al. 1999). Since massive stars have lifetimes shorter

than low-mass stars, HII regions are located near sites of recent and ongoing

star formation, in giant molecular clouds that are located in the

spiral arms of disk galaxies. Thus, much of the light which delineates the

spiral arms in other galaxies comes either from HII regions or recently

formed OB stars.

Classic HII regions are generally ascribed a spherical morphology like

a bubble, which is also known as Strömgen sphere. The size of the bubble

is also called Strömgen radius because Bengt Strömgen derived the theory

of HII region (Strömgren 1939). Ultracompact HII regions often have

different morphological classes such as spherical, cometary, core-halo,

shell, and multiply peaked (Wood & Churchwell 1989). Observationally,

the ionized gas around massive stars is classified based on its size (R s ),

the electron density (n e ), the number of ionizing stars within the ionized

region. In short, the classification of HII regions starts with hypercompact

HII regions (HC HII). For a homogeneous nebula, HC HII regions

are tiny bubbles with a diameter of ∼3 × 10 −3 pc and a density of

≥ 10 6 cm −3 (Lizano 2008). This phase is considered to be the phase

immediately after the hot molecular core phase.

Observations targeting HC HII regions show that the linewidth of

recombination lines such as H92α 5 at 8.3 GHz is extremely broad (∆v

> 40 km s −1 ; Sewilo et al. 2004; Keto et al. 2008), indicating the sum of

significant bulk gas motions, accretion, rotation, and/or expanding gas

(Lizano 2008). At this point, the high pressure of the ionized gas will

5 H92α, which trace ionized gas in star-forming regions, denotes hydrogen transition

line from n 93 to n 92.

14

1.4 Feedback from Massive Stars

.......................................................................

drive the expansion of the HC HII region, evolving it to UC HII region

(diameter of ∼5 × 10 −2 pc and density of ≥ 10 4 cm −3 , Churchwell

2002). In this way, the internal energy and momentum, which are also

called ‘stellar feedback’ drive the dynamics of the HII region and expand

it by sweeping out the gas outward in the ionization front, which is a

thin layer separating ionized gas in HII region from the surrounding HI

region (Tielens 2010).

Churchwell et al. (2006) showed that the Milky Way is full of bubbles

by uncovering parsec-sized bubbles throughout the Galactic plane

using the mid-IR Galactic Legacy Infrared Mid-Plane Survey Extraordinaire

(GLIMPSE) obtained with NASA’s Spitzer Space Telescope such

as compact HII regions (e.g., M43) with densities of ∼10 3 cm −3 and sizes

of ∼0.2 pc.

Many of the bubbles in the GLIMPSE Survey have been classified

as compact HII regions. Beaumont et al. (2014) developed a machinelearning

algorithm for identifying bubbles in the Galactic plane as part

of the Milky Way Project, a citizen science project based on GLIMPSE

maps. In data-release 2 (DR2) of this project, Jayasinghe et al. (2019)

reported the identification of 1394 bubbles located within the Galactic

plane. Despite their high numbers in the Galaxy, it is still a longsought

goal to discover ‘which feedback mechanisms drive the evolution

of HII regions.’

1


The enormous energy input of massive stars into their environment during

their lifetime plays an important role in shaping the morphology of

molecular clouds in which stars are formed in the Galaxy. In general,

a few sources of internal energy and momentum can drive the dynamics

(expansion, perturbation, and breaking) of HII regions: UV radiation

from stars (e.g., Jijina & Adams 1996; Krumholz & Matzner 2009;

Lopez et al. 2011), infrared radiation processed by dust inside an HII shell

(Thompson et al. 2005; Murray et al. 2011; Andrews & Thompson 2011),

the warm gas ionized by massive stars within HII region (e.g., Whitworth

1979; Dale et al. 2005; Pabst et al. 2020), the hot gas shock-heated by

stellar winds and SNe (e.g., Yorke et al. 1989; Harper-Clark & Murray

2009; Pabst et al. 2019), and protostellar outflows/jets (e.g., Quillen

15


.......................................................................

et al. 2005; Cunningham et al. 2006; Li & Nakamura 2006; Nakamura &

Li 2008; Wang et al. 2010). Each of these mechanisms has been employed

separately in the literature in models, simulations, and observations. Using

SOFIA velocity-resolved [C ii] observations, Pabst et al. (2019, 2020)

showed that the stellar winds of the Trapezium stars are responsible for

the expansion of the Orion Nebula and the over-pressured ionized gas in

M43 and NGC 1977 based on the energetic arguments.

1.4.1 Mechanical Feedback

1

Not only radiative feedback, but also mechanical feedback of high-mass

stars via jets/outflows affects the morphology and chemistry of the surrounding

gas. Interest in outflow feedbacks from massive protostars from

the core-scale (∼0.1 pc) to the cloud-scale (≥1 pc) has increased in recent

years because outflows carry kinetic energy comparable to the turbulent

energy and gravitational binding energy of their parental cloud

(Arce et al. 2007). Because of their kinetic energies, their effect can be

devastating or constructive in terms of star formation rates. Moreover,

large-scale outflows cause parsec-scale velocity gradients in star-forming

regions and produce dense massive shells/cavities far from the massive

protostar ejecting outflow(s) (Cunningham et al. 2006). These types of

giant outflows create dense, massive shells of swept up gas (Bally et al.

1999; Quillen et al. 2005). They can even cause breaches or protruding

structures on spherical HII regions as a result of energetic mass ejections

onto their parental cloud in massive star-forming regions (Benedettini

et al. 2004).

Massive star-forming regions tend to form in high multiplicity ejecting

multiple outflows from the molecular cloud (Bally 2016). Recent

simulations reveal that jets and outflows are crucial in determining the

mass of stars in molecular cloud (STARFORGE project; Guszejnov et al.

2021). Also, past outflow activities, which are also called ‘fossil outflows’

from a group of stars leave a signpost on their hosting cloud creating a

number of cavities or dents on the HII regions (e.g., Orion Nebula).

1.4.2 Radiative Feedback

Over their main sequence lifetime massive stars emit a large amount of

energy via powerful UV photons causing photoionization (Oey & Clarke

16


.......................................................................

Orion’s big Protrusion

Fossil outflows

M43 Dark Lane

M43

Northeast Dark Lane

Orion’s small

Protrusion

Dark Bay

Orion Bar

Trapezium

Stars

Orion-S

cloud

Extension of

Orion Bar

Bubbles

Veil’s Bird

X-ray North

West Rim

Kelvin-Helmholtz

Instabilities

East Rim

X-ray South

Rayleigh-Taylor

Instabilities

1

South Rim

3

Figure 1.7: Image of the Orion Nebula from the ESO/VLT Survey

with some apparent structures highlighted with different colors (see

also Robberto et al. 2013). The Veil shell lies in front of the Trapezium

stars and its border is along the rims. The background image

of the Orion Nebula can be retrieved from the ESO website: https:

//www.eso.org/public/images/eso1723a/.

2007; Schneider et al. 2020). This feedback has two effects creating

HII regions and generating diffuse ionized ISM. A small portion of the

energy is delivered via stellar winds driven by massive stars. For example,

a O5 star has a luminosity of ∼4 × 10 39 ergs s −1 and a mechanical

power of its stellar winds of ∼1.3 × 10 35 ergs s −1 for a mass loss rate

17


.......................................................................

of 10 −7 M ⊙ yr −1 and a wind terminal velocity of 2000 km s −1 (Chu &

Gruendl 2011).

The radiation of a massive YSO heats (up to 10 4 K) and repels the

surrounding material outward, photodissociating molecules and ionizing

atoms. Stellar winds clear out the circumstellar medium, creating

HII regions. From the formation of a massive star, it is unclear what

mechanism drives the expansion of the HII region. Castor et al. (1975)

and Weaver et al. (1977) derived an analytical model for the expansion

of the HII region via stellar winds. More recent simulations are able to

incorporate different mechanisms to determine the dominant mechanism

responsible for the expansion (Haid et al. 2018). Velocity-resolved [C ii]

observations from SOFIA allow us to quantify the different feedback

mechanisms. The Orion Nebula is one of the best candidates to study

stellar feedback because of its proximity and richness in terms of star

formation activity. The main structures in the Orion Nebula are shown

in Fig. 1.7. Pabst et al. (2019) find that the Orion Nebula is mainly

blown by stellar winds of the Trapezium stars, in particular θ 1 Ori C,

rather than over-pressurized ionized gas.

1

1.5 Aims and Methods

Recent developments in ground-based, space-based, and stratospheric

mission instruments allow us to characterize the effects of massive stars

on their environment. With this motivation, we aim to study the interaction

of massive stars with the surrounding gas components and their

formation. Therefore, various structures of the ISM, the formation of

massive stars, and feedback mechanisms are the focus of this thesis,

which relies on low- and high-resolution millimetre and submillimetre,

and far-IR observations. The observations, methods, and model software

used in this thesis are listed in the following sections.

1.5.1 Herschel Space Observatory

Herschel Space Observatory or Herschel is a L2-referenced 6 space-based

single-dish telescope (the size of primary mirror of 3.5 m with f/0.5 op-

6 As seen from the Sun, Lagrange-2 point (L2) is 1.5 × 10 6 km immediately behind

the Earth-Moon system.

18


.......................................................................

Figure 1.8: Herschel Space Observatory (HSO or Herschel). Image

credit: ESA−C. Carreau.

erated between 55 to 672 µm) which has made significant contribution

to our understanding of the ISM (see Fig. 1.8). Its contribution to astrophysics

is still ongoing even in the current years although it ceased operation

in 2013. There are three instruments onboard Herschel: Heterodyne

Instrument for the Far Infrared (HIFI), Photodetector Array Camera

and Spectrometer (PACS), and the Spectral and Photometric Imaging

REceiver (SPIRE). HIFI (de Graauw et al. 2010) is a high resolution

heterodyne spectrometer covering the 490−1250 GHz and 1410−1910

GHz ranges in seven bands. HIFI is specialized to observe the sky with

a single pixel but is not well suited for imaging.

Our focus in this thesis is on a hydride: hydrogen fluoride (HF) to

study the density of the Orion Bar. The HF molecule has been found

in absorption towards to Sgr B2 by Neufeld et al. (1997) who observed

the HF J = 2-1 transition using the Infrared Space Observatory (ISO).

However, ISO was unable to observe the ground-state transition of HF.

Also high densities or strong radiation fields are needed to populate

the HF J = 2 level and the derived HF column density contains large

1

19


.......................................................................

Figure 1.9: A panoramic view of VLA antennas located in New Mexico.

Image credit: NRAO.

1

uncertainties. With Herschel/HIFI, the observation of the ground-state

rotational transition of HF became possible for the first time. Since

Herschel can observe the J = 1−0 transition of HF (1232.4 GHz) at

high resolution (R > 10 6 ) its data is going to be used to clarify the

origin of the HF emission in the Orion Bar.

1.5.2 Very Large Array (VLA)

The Very Large Array (VLA) is a centimetre-wavelength interferometer

operating at radio frequencies (1 to 50 GHz) located at Socorro, New

Mexico at an altitude of 2120 metres. The VLA interferometer consists

of 28 steerable antennas (including one spare), of 25-metre arranged in

an equiangular Y-shaped rail configuration, 9 antennas per arm. The

farthest antenna is 21 km from the center of the Y-shape (Thompson

et al. 1980). Part of the interferometer is shown in Fig. 1.9. The VLA

observatory is a National Science Foundation (NSF) facility operated

under the cooperative agreement of Associated Universities, Inc.

The VLA interferometer covers the radio frequency range from 1 to

50 GHz over eight receiver bands (L, S, C, X, Ku, K, Ka, and Q bands)

utilizing state-of-the-art interferometry technology. The basic character-

20


.......................................................................

Figure 1.10: SOFIA flies over the Sierra Nevada mountains during a test

flight. Image credit: NASA/Jim Ross.

istics such as bandwidth, antenna sensitivity, and the sensitivity of the

continuum and line observations are given in Perley et al. (2011). It is

difficult to give a standard angular and velocity resolution for the VLA

interferometer. Instead, the resolution is determined by the array configuration

and the observing frequency. Information about the resolution

for different configurations can be found on the NRAO website 7 .

In Chapter 2, we use flux measurements in the C (6 cm) and K

(1.3 cm) bands to compute the spectral index for each continuum source

detected in the VLA maps of 18 massive star-forming regions. Chapter 2

consists of more information about the VLA and our observations.

1

1.5.3 Stratospheric Observatory for Infrared Astronomy

(SOFIA)

Stratospheric Observatory for Infrared Astronomy (SOFIA), is an airborne

observatory project of the US National Aeronautics and Space Administration

(NASA), and the German Aerospace Centre (DLR). SOFIA

is a modified aeroplane of the type Boeing 747-SP, which carries a telescope

with a diameter of 2.7 m in the rear fuselage (see also Fig. 1.10

7 https://science.nrao.edu/facilities/vla/docs/manuals/oss/performance/resolution

21


.......................................................................

1

Young et al. 2012). By flying up to 45000 ft, SOFIA makes it possible

to observe at frequencies blocked by the atmosphere from the ground. A

large part of the spectrum at far infrared (FIR) frequencies (1-10 THz)

becomes accessible. At the same time, a few molecular species (H 2 O, O 3 )

in the Earth’s atmosphere still block FIR radiation at certain frequencies

(Risacher et al. 2016).

SOFIA can access the frequency range from 0.3 to 1600 µm with

eight instruments with different resolving powers aimed for various scientific

purposes (Risacher et al. 2018). upGREAT is a heterodyne array

receiver with 21 pixels. At the time of the observations it contained

2 × 7-pixel sub-arrays with a hexagonal layout are designed for

low-frequency array receiver (LFA) with dual-band polarization. These

cover the 1.83−2.07 THz frequency range where the [C ii] 158 µm and

[O i] 145 µm lines can be found. The other hexagonal 7-pixel array is

located in the high-frequency array (HFA) that covers the [O i] 63 µm

line. The GREAT instrument uses local oscillators (LO) and heterodyne

techniques to achieve high spectral resolution (ν/∆ν = 10 7 ). In Chapter

4 and 5, we used velocity-resolved map of the Orion Nebula obtained

upGREAT instrument, a heterodyne spectrometer, including the sevenbeam

receiver array, within the framework of SOFIA C+ SQUAD Large

Program led by A. G. G. M. Tielens.

1.5.4 Meudon PDR code

The Meudon PDR code simulates a stationary plane-parallel slab of gas

and dust illuminated by an external radiation field coming from one or

both sides, where the two intensities can be different. It solves, at each

point in the cloud, the radiative transfer in the UV, taking into account

the absorption in the continuum by dust and in discrete transitions of

H and H 2 . The model also computes the thermal balance, taking into

account heating processes such as the photoelectric effect on dust, chemistry,

cosmic rays, etc., and cooling resulting from infrared and millimeter

emission of the abundant ions, atoms, and/or molecules (Le Petit et al.

2006). We use the code to compute the abundance changes of atoms

and/or molecules (such as H 2 , [C ii] , and HF) in Chapter 3 assuming

that abundances of the molecules or atoms are known.

22


.......................................................................

1.5.5 RADEX

RADEX, is a radiative transfer code, that has been developed to infer

physical and chemical parameters such as temperature, density, and

molecular abundances, based on statistical equilibrium calculations (van

der Tak et al. 2007). It is available as part of the Leiden Atomic and

Molecular Database (LAMDA) package (Schöier et al. 2005). RADEX

is a one-dimensional non-LTE radiative transfer code that uses the escape

probability formalism assuming an isothermal and homogeneous

medium without large-scale velocity fields. Various geometries are available

in RADEX, e.g. isothermal and homogeneous medium, slab parallel

model, and expanding sphere. One of these can be selected within the

off-line RADEX version but the main input parameters are the same.

The input parameters of RADEX are kinetic temperature (T kin ) gas density

(n H2 ), and the molecular column density (N). The FWHM of the

observed line, collisional partners and their collisional data, and the radiation

field (CMB and dust emission) have to be given as input as well.

The software can be used in two ways: (i) one can compare modelled line

intensities with observed ones or (ii) one can compare the observed intensity

ratios of lines of the same molecule. RADEX is used to construct

a column density map of the Orion Bar in Chapter 3.

1

1.5.6 GILDAS

GILDAS (Grenoble Image and Line Data Analysis Software) is a multipackage

software developed by IRAM-Grenoble to reduce and visualize

(sub-)millimeter observations 8 . GILDAS includes CLASS, GreG, AS-

TRO, GRAPHIC, and CLIC. Of these packages, CLASS, designed for

the reduction of spectroscopic data obtained with a single-dish telescope,

is the main sub-package for the reduction/display of HIFI and SOFIA

observations throughout this thesis.

1.5.7 CASA

CASA, the Common Astronomy Software Applications package, is the

primary data processing software used for data reduction and analysis

of radio observations. CASA is written with C++ application libraries

8 Official GILDAS homepage: https://www.iram.fr/IRAMFR/GILDAS.

23


.......................................................................

running in the background for data reduction and analysis of radio astronomical

data. All these applications are scriptable through the IPython

interface to Python. The software is used to reduce and analyse the radio

continuum observations of the VLA (McMullin et al. 2007).

1.5.8 HIPE

Herschel Interactive Processing Environment (HIPE) is the application

developed for data reduction and analysis of photometric and spectroscopic

Herschel observations taken between 55−672 µm (Herschel Science

Ground Segment Consortium 2011). The HF observations of Herschel/HIFI

are retrieved via HIPE from the ESA repository. The observations

are exported to FITS format for further processing with python

packages in Section 1.5.9.

1.5.9 Python Libraries

1

Python is a general-purpose programming language with many libraries

for a wide range of applications. Throughout this thesis, various packages

specialized to astronomical research and general aims are employed.

These packages are Numpy, matplotlib, spectral-cube (Ginsburg et al.

2019), APLpy (Robitaille & Bressert 2012), and Astropy (Astropy Collaboration

et al. 2013). Among these applications, matplotlib, APLpy,

and Astropy are mostly used for astronomical image display and plots.

Numpy is utilized for multidimensional arrays and operations to compute

various parameters in the thesis. Finally, spectral-cube is a recently

developed software that provides a simple way to read, modify, and analyze

data cubes with two spatial dimensions and one spectral dimension

collected from various observatories, such as integration over particular

velocity ranges and moment maps.

1.6 Goals of this thesis

This thesis is concerned with the formation of massive stars and the

effects they have on their environment throughout their lifetimes. The

major scientific aims on which we concentrated are as follows.

24

1.7 Outline

.......................................................................

• Goal 1: Investigate the similarities between massive and low-mass

star formation by searching for radio jets and establishing a probable

association between masers (H 2 O and CH 3 OH) and radio jets.

• Goal 2: Investigate the density structure of the Orion Bar, a wellknown

PDR in the Orion Nebula by using the ground state transition

of HF at 1.232 THz and establish a novel CO-dark H 2 gas

tracer in the ISM.

• Goal 3: Explore the distorted morphology of the Orion Nebula in

its northern part and determine the driving mechanism producing

the perturbation on the ionization front of the HII region.

• Goal 4: Quantify the protostellar feedback via outflows on the

whole Orion Veil by using a 1.2 degree [C ii] 158 µm map taken

with SOFIA and reveal impinging outflows onto the Veil.

1.7 Outline

This section gives an outline of the thesis and gives a short summary of

the content.

In Chapter 2, we utilize VLA continuum observation obtained in two

different bands (C and K bands) in B-configuration. We construct a

continuum source database of 146 continuum sources in 18 massive starforming

regions. We identify the radio jets based on spectral indices, and

association with outflows and masers. In total, we find 7 new radio jets

emanating from massive YSOs. Also, we find that CH 3 OH masers are

mostly associated with thermal radio jets while H 2 O masers are associated

with non-thermal radio jets. We conclude that low- and high-mass

star formation processes are similar, at least as far as jets are concerned.

Furthermore, finding radio jets in high-mass star-forming regions is more

likely in somewhat more developed star-forming regions.

In Chapter 3, we employ velocity-resolved HF J = 1−0 maps of the

Orion Bar PDR taken with the HIFI instrument onboard Herschel to

investigate the clumpy density structure of the bright Bar and the origin

of HF emission. To this end, we compare HF observations to the outputs

of radiative and chemical models created with physical conditions of the

Orion Bar. We find that the HF molecules are excited by collision with

1

25


.......................................................................

1

H 2 in the interclump medium at a density of 10 5 cm −3 . We also conclude

that HF emission traces CO-dark H 2 gas in the Orion Bar which may

help to understand the nature of galactic and extra-galactic HF emission.

In Chapter 4, we use velocity-resolved [C ii] observations with the

upGREAT instrument at the SOFIA observatory to understand the origin

of a protruding structure, which we call Orion’s protrusion, on the

north-west of the Veil shell. We estimate that the protrusion expands

at 12 km s −1 towards us similar to the Veil. Its size and thickness is

1.3 pc and 0.1 pc, respectively. The momentum budget of the protrusion

indicates that the north-western part of the pre-existing cloud was

perturbed by the relics of outflows, which are extinct by now, of the

Trapezium stars, most likely θ 1 Ori C. Moreover, we find that the protrusion

on the Veil will break the bubble and vent hot ionized gas into

the ISM before a possible supernova in the Trapezium cluster.

Chapter 5 makes use of [C ii] 158 µm observations of the whole

Veil obtained in the upGREAT survey of Orion to reveal interacting

jets/outflows with the Veil shell. To pinpoint these interaction spots

on the Veil, we use position-velocity diagrams and high-velocity [C ii]

emission moving relative to the background cloud OMC-1. In total, we

find six shock-accelerated [C ii] emitting gas spots on the Veil surface

through collimated jet/outflows from stars with luminosities higher than

10 3 L ⊙ which indicate B-type stars located in the Orion Nebula. These

ejections may cause local density and temperature gradients. We conclude

that not only the most massive stars affect the dynamics of the

expanding ionization fronts of HII regions.

26

1.7 Outline

.......................................................................

1

27


.......................................................................

1

28

Chapter 2

Search for radio jets from

massive young stellar objects.

Association of radio jets with

H 2 O and CH 3 OH masers

Ü. Kavak, Á. Sánchez-Monge, A. López-Sepulcre, R. Cesaroni, F. F. S. van

der Tak, L. Moscadelli, M. T. Beltrán, and P. Schilke 1

2

2.1 Abstract

Recent theoretical and observational studies debate the similarities of the

formation process of high- (> 8 M ⊙ ) and low-mass stars. The formation

of low-mass stars is directly associated with the presence of disks and

jets. Theoretical models predict that stars with masses up to 140 M ⊙ can

be formed through disk-mediated accretion in disk-jet systems. According

to this scenario, radio jets are expected to be common in high-mass

star-forming regions. We aim to increase the number of known radio

jets in high-mass star-forming regions by searching for radio-jet candidates

at radio continuum wavelengths. We used the Karl G. Jansky Very

Large Array (VLA) to observe 18 high-mass Galactic star-forming in the

C band (6 cm, ≈1 ′′ resolution) and K band (1.3 cm, ≈0.3 ′′ resolution).

1 Kavak et al., 2021, A&A, Volume 645, A29

29

CHAPTER 2: Search for radio jets from massive young stellar objects. Association

of radio jets with H 2O and CH 3OH masers

.......................................................................

2

We searched for radio-jet candidates by studying the association of radio

continuum sources with shock activity signs (e.g., molecular outflows, extended

green objects, and maser emission). Our VLA observations also

targeted the 22 GHz H 2 O and 6.7 GHz CH 3 OH maser lines. We have

identified 146 radio continuum sources, 40 of which are located within

the field of view of both images (C and K band maps). We derived

the spectral index, which is consistent with thermal emission (between

−0.1 and +2.0) for 73% of these sources. Based on the association with

shock-activity signs, we identified 28 radio-jet candidates. Out of these,

we identified 7 as the most probable radio jets. The radio luminosity of

the radio-jet candidates is correlated with the bolometric luminosity and

the outflow momentum rate. About 7–36% of the radio-jet candidates

are associated with nonthermal emission. The radio-jet candidates associated

with 6.7 GHz CH 3 OH maser emission are preferentially thermal

winds and jets, while a considerable fraction of radio-jet candidates associated

with H 2 O masers show nonthermal emission that is likely due

to strong shocks. About 60% of the radio continuum sources detected

within the field of view of our VLA images are potential radio jets. The

remaining sources could be compact HII regions in their early stages of

development, or radio jets for which we currently lack further evidence

of shock activity. Our sample of 18 regions is divided into 8 less evolved

infrared-dark regions and 10 more evolved infrared-bright regions. We

found that ≈71% of the identified radio-jet candidates are located in the

more evolved regions. Similarly, 25% of the less evolved regions harbor

one of the most probable radio jets, while up to 50% of the more evolved

regions contain one of these radio-jet candidates. This suggests that the

detection of radio jets in high-mass star-forming regions is more likely in

slightly more evolved regions.

30

2.2 Introduction

.......................................................................

2.2 Introduction

High-mass stars (O- and B-type stars with masses ≥ 8 M ⊙ ) play a crucial

role in the chemical and physical composition of their host galaxies

throughout their lifetimes by injecting energy and material on different

scales through energetic outflows, intense UV radiation, powerful stellar

winds, and supernova explosions. Despite its importance, the formation

process of massive stars is still only poorly understood because it is observationally

and theoretically challenging (e.g., massive stars form in

crowded environments and are located at far distances, see reviews by

Tan et al. 2014; Motte et al. 2018). On the other hand, the formation of

low-mass stars is better understood and is explained with a model based

on accretion through a circumstellar disk and a collimated jet or outflow

that removes angular momentum and enables accretion to proceed

(e.g., Larson 1969; Andre et al. 2000). Circumstellar disks have indeed

been observed around low-mass protostars (e.g., Williams & Cieza 2011;

Luhman 2012), while ejection of material has mainly been observed as

large-scale collimated jets and outflows (e.g., Bachiller 1996; Bally 2016).

For high-mass stars, the role that (accretion) disks and jets/outflows play

in their formation remains to be understood, also how their properties

vary with the mass of the forming star and the environment. For observations,

some studies have concentrated on disks and jets/outflows in

selected high-mass star-forming regions (see e.g., Beuther et al. 2002a;

Arce et al. 2007; López-Sepulcre et al. 2009; Bally 2016). The advent

of facilities such as the Atacama Large Millimeter/Submillimeter Array

(ALMA) or the upgraded Karl G. Jansky Very Large Array (VLA)

provides the required high spatial resolution and sensitivity to fully resolve

the structure of disks and jets/outflows in high-mass star-forming

regions. While disks are bright at millimeter wavelengths and constitute

perfect targets for ALMA observations (e.g., Sánchez-Monge et al.

2013b, 2014; Beltrán et al. 2014; Johnston et al. 2015; Cesaroni et al.

2017; Maud et al. 2019), jets are found to be bright at the centimeter

wavelengths that are observable with the VLA (e.g., Carrasco-González

et al. 2010, 2015; Moscadelli et al. 2013, 2016).

Surveys of low-mass star-forming regions with the VLA (e.g., Anglada

1996; Anglada et al. 1998; Beltrán et al. 2001) revealed radio-continuum

sources elongated in the direction of the large-scale molecular outflows.

2

31



.......................................................................

These sources are called thermal radio jets because their emission is

interpreted as thermal (free-free) emission of ionized, collimated jets at

the base of larger-scale optical jets and molecular outflows (e.g., Curiel

et al. 1987, 1989; Rodriguez 1995). Because of the high spatial resolution

that can be achieved at radio wavelengths with interferometers such as

the VLA, thermal radio jets constitute strong evidence of collimated

outflows on small scales (∼100 au; Torrelles et al. 1985; Anglada 1996)

and permit defining the location of the star that is forming and powering

the jet/outflow seen on larger scales. Although the emission of jets at

radio wavelengths is mainly thermal, some jets show a contribution from

a nonthermal component (e.g., Reid et al. 1995; Carrasco-González et al.

2010; Moscadelli et al. 2013, 2016).

2

Following the strategy used in the study of low-mass star-forming

regions, we aim to search for radio jets associated with high-mass starforming

regions in a large sample of sources. Until recently, only a limited

number of regions harboring high-mass stellar objects were known to be

associated with radio jets (e.g., HH80/81: Marti et al. 1993; Carrasco-

González et al. 2010, CepAHW2: Rodriguez et al. 1994, IRAS 16547-

4247: Rodríguez et al. 2008, IRAS 16562−1732: Guzmán et al. 2010,

G35.20-0.74 N: Beltrán et al. 2016). In the past years, progress has been

made to increase the number of known jets associated with high-mass

young stellar objects (e.g., Moscadelli et al. 2016; Rosero et al. 2016;

Sanna et al. 2018; Purser et al. 2018). We used the VLA in two different

frequency bands to search for radio jets in a sample of 18 high-mass

star-forming regions associated with molecular outflow emission.

This paper is structured as follows. In Section 2.3 we present the

sample and the details of the observations. The results of the observations

of the radio continuum (and maser) emission are presented in

Section 2.4. The analysis of the properties of the discovered sources is

presented in Section 2.5, while Appendix 2.9 describes the properties of

each region in more detail. In Section 2.6 we discuss the implications of

our results in the context of high-mass star formation, and in Section 2.7

we summarize the most important conclusions.

32

2.3 Observations

.......................................................................

2.3 Observations

2.3.1 Selected sample

We selected 18 high-mass star-forming regions from the samples of López-

Sepulcre et al. (2010, 2011) and Sánchez-Monge et al. (2013d) using the

following criteria: (i) clump mass > 100 M ⊙ , to exclude regions that

mainly form low-mass stars, (ii) distance < 4 kpc, to resolve spatial

scales < 4000 AU when observed with interferometers at a resolution of

1 arcsec, (iii) declination > −15 ◦ , to be observable from northern telescopes,

(iv) association with an HCO + bipolar outflow and SiO emission

with line widths broader than > 20 km s −1 (López-Sepulcre et al. 2011;

Sánchez-Monge et al. 2013d), and (v) absence of bright centimeter continuum

emission, to exclude developed HII regions.

We used the NVSS 2 (Condon et al. 1998), the MAGPIS 3 (Helfand

et al. 2006), CORNISH 4 (Hoare et al. 2012; Purcell et al. 2013), and

RMS 5 (Urquhart et al. 2008; Lumsden et al. 2013) surveys to eliminate

star-forming regions with developed HII regions that would hinder the

detection of faint radio jets. Our final sample of 18 high-mass starforming

regions is listed in Table 2.1.

2

2 NRAO VLA Sky Survey

3 The Multi-Array Galactic Plane Imaging Survey

4 Co-Ordinated Radio ‘N’ Infrared Survey for High-mass star formation

5 Red MSX Source survey

33

2

34

Table 2.1: High-mass star-forming regions observed with the VLA

R.A. (J2000) Dec. (J2000) d a M a C band b K band b

Region (h:m:s) ( ◦ : ′ : ′′ ) (kpc) (M ⊙) θ beam , PA rms θ beam , PA rms

IRAS 05358+3543 † 05:39:12.2 +35:45:52.0 1.8 127 1.27 × 1.23, +61 8.1 . . . c . . .

G189.78+0.34 † 06:08:34.5 +20:38:51.0 1.8 150 1.28 × 1.09, +23 16.0 . . . c . . .

G192.58−0.04 † 06:12:52.9 +18:00:34.9 2.6 500 1.40 × 1.19, +21 23.6 . . . c . . .

G192.60−0.05 † 06:12:54.0 +17:59:23.0 2.6 460 1.36 × 1.14, +20 26.5 . . . c . . .

G18.18−0.30 † 18:25:07.3 −13:14:22.9 2.6 110 1.74 × 1.05, −16 10.0 0.54 × 0.31, +25 16.7

IRAS 18223−1243 † 18:25:10.9 −12:42:27.0 3.7 980 1.89 × 1.14, −14 24.5 0.58 × 0.34, +37 15.0

IRAS 18228−1312 † 18:25:42.3 −13:10:18.0 3.0 740 1.88 × 1.16, −14 35.0 0.73 × 0.67, −07 59.1

G19.27+0.1M2 ‡ 18:25:52.6 −12:04:47.9 2.4 114 2.08 × 1.14, −14 9.8 0.50 × 0.32, −26 20.7

G19.27+0.1M1 ‡ 18:25:58.5 −12:03:58.9 2.4 113 1.95 × 1.19, −16 9.6 0.79 × 0.42, +49 20.0

IRAS 18236−1205 † 18:26:25.4 −12:03:50.9 2.7 780 1.99 × 1.11, −18 10.2 0.51 × 0.32, −26 16.9

G23.60+0.0M1 ‡ 18:34:11.6 −08:19:05.9 2.5 365 1.85 × 1.13, −22 8.7 0.54 × 0.30, +32 36.1

IRAS 18316−0602 † 18:34:20.5 −05:59:30.0 3.1 1000 1.71 × 1.09, −21 8.4 0.53 × 0.30, +35 40.0

G24.08+0.0M2 ‡ 18:34:51.1 −07:45:32.0 2.5 201 1.80 × 1.13, −21 17.0 0.53 × 0.30, +32 37.0

G24.33+0.1M1 ‡ 18:35:07.8 −07:35:04.0 3.8 1759 1.69 × 1.25, −13 21.0 0.49 × 0.31, −31 26.1

G24.60+0.1M2 ‡ 18:35:35.7 −07:18:08.9 3.7 483 1.66 × 1.03, −20 18.2 0.79 × 0.41, +51 21.8

G24.60+0.1M1 ‡ 18:35:40.2 −07:18:37.0 3.7 192 1.67 × 1.22, −10 12.7 0.75 × 0.38, +57 22.5

G34.43+0.2M3 ‡ 18:53:20.3 +01:28:23.0 2.5 301 1.57 × 1.46, −59 17.9 0.53 × 0.29, +40 19.0

IRAS 19095+0930 † 19:11:54.0 +09:35:52.0 3.0 500 1.61 × 1.37, −83 17.0 0.53 × 0.29, +44 63.4

(a)

Notes. Distances (d) and clump masses (M) from López-Sepulcre et al. (2011) and Sánchez-Monge et al. (2013d).

(b) Synthesized beam (θ beam ) major and minor axis in arcsecond, and position angle (PA) in degrees. The rms noise level

is given in units of µJy beam −1 . Regions marked with † and ‡ in the first column indicate IR-loud and IR-dark sources,

respectively, based on the classification of López-Sepulcre et al. (2010). (c) Region not observed in the K band.


of radio jets with H2O and CH3OH masers

.......................................................................

2.3 Observations

.......................................................................

2.3.2 VLA observations

We used the VLA of the NRAO 6 to observe the 18 selected regions (see

Table 2.1). The observations were conducted between June and August

2012 (project number 12A-099), when the array was in transition to

its current upgraded phase and was known as expanded VLA (EVLA).

During the observations, the array was in its B configuration, which provides

a maximum baseline of 11 km. We observed the frequency bands C

(4–8 GHz) and K (18–26.5 GHz) with 16 spectral windows of 128 MHz

each, covering a total bandwidth of 2048 MHz in each band. Each spectral

window has 128 channels, with a channel width of 1 MHz. The time

spent per source is ∼20 minutes and ∼30 minutes at 6 cm (C band)

and 1.3 cm (K band), respectively. Flux calibration was achieved by

observing the quasars 3C286 (F 1.3 cm =2.59 mJy, F 6 cm =7.47 mJy) and

3C48 (F 1.3 cm =1.13 mJy, F 6 cm =5.48 mJy). The amplitude and phase

were calibrated by monitoring the quasars J0555+3948, J0559+2353,

J1832−1035, and J1851+0035. We used the standard guidelines for the

calibration of VLA data. The data were processed using the Common

Astronomy Software Applications (CASA; McMullin et al. 2007).

Continuum images of each source were obtained after channels with

line emission were excluded, corresponding to H 2 O and CH 3 OH maser

lines. The images were obtained using the ‘clean’ task with the Briggs

weighting parameter set to 2, which results in a typical synthesized beam

of 1.5 arcsec and 0.4 arcsec for the C and K bands, respectively, and

typical rms noise levels of ∼22 µJy beam −1 at 6 cm and ∼30 µJy beam −1

at 1.3 cm (see Table 2.1).

The spectral resolution of the observations is limited (about 50 km s −1

and 13 km s −1 for the C and K bands, respectively) and insufficient to resolve

spectral features. Despite this limitation, we produced image cubes

of spectral windows that cover the frequencies of the H 2 O maser line at

22235.0798 MHz and the CH 3 OH maser line at 6668.519 MHz. This allowed

us to search for maser features that can be associated with the continuum

emission. The rms noise levels of these cubes are 0.5 mJy beam −1

and 0.3 mJy beam −1 per channel of 13 and 50 km s −1 for the H 2 O and

CH 3 OH images, respectively.

2

6 The Very Large Array (VLA) is operated by the National Radio Astronomy

Observatory (NRAO), a facility of the National Science Foundation operated under

cooperative agreement by Associated Universities, Inc.

35



.......................................................................

Sources detected in the …

K-band-only

C & K both

C-band-only

27 10 3

67 1 1 37

iC/iK iC/oK oC

2

Figure 2.1: Number of radio continuum sources detected in the K-band

images (marked in blue and corresponding to 15 sources) and in the C-

band images (marked in green and corresponding to 142 sources). Only 4

of the 146 detections are detected in the K-band images alone. The vast

majority (131) are detected only in the C-band images (see Sect. 2.4.1

for more details). The bottom labels mark the sources that are located

within the primary beams of the K-band and C-band images. Thirtyeight

sources are located outside the C-band primary beam (oC), 68

sources are located inside the C-band primary beam but outside the

K-band primary beam (iC/oK), and 40 sources are located within the

primary beam of both images (iC/iK). See Sects. 2.4.1 and 2.4.2 for more

details.

2.4 Results

2.4.1 Continuum emission

We detected compact continuum emissions in all 18 observed high-mass

star-forming regions. A total of 146 compact sources are identified with

intensities above 3σ level, where σ is the rms noise level of each map (see

Table 2.1). In Table 2.5 we list the coordinates, fluxes, and source sizes.

Most of the sources (a total of 131) are only detected in the C-band

image, while 4 of them are only detected in the K band (see Fig. 2.1).

Only 11 sources are detected at both frequencies. The higher detection

rate of sources in the C band is due to several factors. First, four regions

were only observed in the C band (see Table 2.1). This results in

26 radio continuum sources for which we have no access to K-band images.

Second, the field of view of the C-band images (primary beam ≈9

36

2.4 Results

.......................................................................

arcmin) is larger than that of the K-band primary beam (≈2 arcmin).

Only 40 sources are located within the K-band primary beam (identified

as iC/iK, see Fig. 2.1). This number is reduced to only 24 when we consider

only the sources that have been observed in both frequency bands.

A total of 68 sources are inside the C-band primary beam (identified as

iC) but outside the K band primary beam (identified as oK, see Fig. 2.1).

The remaining 38 sources are outside the primary beam of the C-band

observations (marked oC; see also Table 2.5). The sources that are outside

the primary beam are bright enough to be detected even when the

telescope sensitivity is highly reduced. Third, the spatial filtering of the

interferometer is different at the two frequencies. In the B configuration,

the VLA recovers scales up to 11 arcsec in the C band, and only up to 4

arcsec in the K band (see also Appendix A of Palau et al. 2010). Finally,

we cannot exclude the possibility that some of these sources are extragalactic

objects that can only be detected at low frequencies. We have

followed the approach of Anglada et al. (1998) to determine the possible

contamination of background sources in our catalogue. The expected

number of background sources N bg is given by

{ [ ( ) ]}

2 θF (

ν

) 2

N bg =1.4 1 − exp −0.0066

arcmin 5 GHz

( ) (2.1)

−0.75

S0

(

ν

) −2.52

×

,

mJy 5 GHz

2

where θ F is the area of the sky that has been observed (18 fields in

C band, and 14 fields in K band), ν is the frequency of the observations,

and S 0 is the detectable flux density threshold (3×rms, with an average

rms of 22 µJy beam −1 in the C band, and 30 µJy beam −1 in the K band).

This results in N bg = 11 and N bg = 0.2 for the C- and K-band images,

respectively. Less than 5% of the sources detected in the C band might

be background objects not related to the star-forming regions, while we

do not expect contamination in the K-band images.

2.4.2 Spectral index analysis

The spectral index (α) is defined as S ν ∝ ν α , where S ν is the flux density

and ν is the frequency. We calculated the spectral index for the contin-

37



.......................................................................

uum sources using the measured flux densities at 1.3 cm (K band) and

6 cm (C band). For the sources without detection in one of the bands,

we assumed an upper limit of the flux density equal to 5σ. The flux

densities of the sources were corrected for the primary beam response

of the antennas. The sources far away from the phase centers (listed

in Table 2.1) have larger uncertainties in the correction factors of the

primary beam and therefore in the final (corrected) flux. The sources

located within the primary beams in both frequency bands (i.e., sources

listed as ‘iC/iK’ in Table 2.5) accordingly have more accurate flux estimates.

For the sources outside one of the primary beams (i.e., oK or

oC), we did not determine the spectral index because of the high uncertainty

involved in the fluxes. In the last column of Table 2.5 we list the

calculated spectral indices. For the sources detected at both frequencies,

we improved the determination of the spectral index by creating new

images with the same uv (visibility) coverage (see Table 2.7). This ensured

that the interferometer is sensitive to similar spatial scales at both

frequencies.

2

In Fig. 2.2 we present the spectral index against the ratio of flux density

to intensity peak for the 24 continuum sources that were observed at

both bands and located within the primary beams. For the sources detected

at 6 cm only, we derive an upper limit to the spectral index, while

we derive a lower limit for the spectral index for the sources detected only

at 1.3 cm. We note that the real spectral index may not always be an

upper limit if the source emission is completely filtered out in our K-band

images. Further observations at different wavelengths, with a similar uv

sampling and angular resolution are necessary to constrain the spectral

index of the sources detected only in the C-band images. The sources

detected at both wavelengths (black dots) have a more precise determination

of the spectral index. For most sources, we derive spectral indices

consistent with thermal emission (i.e., in the range of −0.1 to +2), and

in agreement with observations of other radio-jets (e.g., Anglada et al.

2018). Only a few sources show very negative spectral indices (sources

48, 96, and 144). These sources are likely to be partially filtered out

in the K-band images, which may result in lower limits for the actual

value of the spectral index. In particular, source 48 appears as three

distinguishable peaks, which we refer to as a, b, and c, surrounded by a

more diffuse and extended structure that is mainly visible in the C-band

38

Sp

−2

−3

48b

Panel (a)

144

10 −1 10 0 10 1 10 2

Flux (mJy)

2.4 Results

.......................................................................

3

2

Spectral Index

1

0

−1

−2

96

Thermal free-free emission spectral index

48c

48a

48b

−3

144

10 0 10 1

Flux/Intensity

Figure 2.2: Spectral index (α, see Sect. 2.4.2) against the flux-to-intensity

ratio for the radio continuum sources detected in both frequency bands

and inside the primary beam of both images (sources listed as ‘iC/iK’

in Table 2.5). The gray shaded region depicts the spectral index regime

associated with thermal free-free emission (i.e., in the range from −0.1

to +2). Black dots correspond to sources detected in both bands (see

spectral indices in Table 2.3), blue upward-pointing triangles correspond

to sources detected only in the K band (i.e., lower limits), and red

downward-pointing triangles correspond to sources detected only in the

C band (i.e., upper limits).

2

image. High flux-to-intensity ratios indicate that the source is likely extended

and most likely partially filtered out in the K-band images, which

may result in negative spectral indices.

39

2

40

Table 2.2: H 2 O and CH 3 OH maser features

R.A. (J2000) Dec. (J2000) Vmaser a V H13 CO +

LSR Intensity Continuum

Region Maser (h:m:s) ( ◦ : ′ : ′′ ) (km s −1 ) (km s −1 ) (Jy beam −1 ) source ID b

IRAS 05358+3543 CH 3 OH 05:39:13.071 +35:45:50.938 −304 −15.8 0.028 2

G189.78+0.34 CH 3 OH 06:08:35.304 +20:39:06.405 −13 +9.2 0.014 14

G192.58−0.04 CH 3 OH 06:12:54.026 +17:59:23.060 −14 +9.1 0.72 22

G18.18−0.30 H 2 O 18:25:07.575 −13:14:31.487 −3 +50.0 0.57 –

IRAS 18223−1243 H 2 O 18:25:10.804 −12:42:26.234 +24 +45.2 0.006 –

IRAS 18228−1312 H 2 O 18:25:41.935 −13:10:19.591 +24 +33.1 0.022 48

IRAS 18236−1205 H 2 O 18:26:25.677 −12:03:48.402 +28 +26.5 0.010 63

H 2 O 18:26:25.575 −12:03:48.502 +28 +26.5 0.006 63

H 2 O 18:26:25.782 −12:03:53.263 +15 +26.5 0.010 64

H 2 O 18:26:27.149 −12:03:54.888 +15 +26.5 0.014 –

CH 3 OH 18:26:25.788 −12:03:53.456 +5 +26.5 0.26 64

G19.27+0.1M1 H 2 O 18:25:58.546 −12:03:58.516 +28 +26.5 0.022 –

G23.60+0.0M1 H 2 O 18:34:11.237 −08:19:07.680 +108 +106.5 0.44 –

H 2 O 18:34:11.452 −08:19:07.138 +108 +106.5 0.10 –

IRAS 18316−0602 H 2 O 18:34:20.918 −05:59:41.638 +41 +42.5 11.1 83

CH 3 OH 18:34:20.913 −05:59:42.087 −233 +42.5 0.014 83

G24.33+0.1M1 H 2 O 18:35:08.123 −07:35:04.216 +108 +113.6 4.03 110

CH 3 OH 18:35:08.147 −07:35:04.260 −182 +113.6 0.010 110

G24.60+0.1M2 H 2 O 18:35:35.728 −07:18:08.796 +122 +115.3 0.031 –

G24.60+0.1M2 H 2 O 18:35:40.120 −07:18:37.417 +54 +53.2 1.02 136

IRAS 19095+0930 H 2 O 19:11:53.975 +09:35:50.559 +37 +43.9 26.8 143

H 2 O 19:11:53.990 +09:35:49.848 +37 +43.9 10.3 143

CH 3 OH 19:11:53.993 +09:35:50.641 +39 +43.9 0.043 143

Notes.

(a) Uncertainties in the reported maser velocities (V maser) are expected to be ∼50 km s −1 for the CH 3OH masers

and ∼13 km s −1 for the H 2O masers (see Sect. 2.3). The systemic velocities (V H13 CO +

LSR ) are reported in López-Sepulcre et al.

(2011). (b) Radio continuum source spatially associated with the maser feature and listed as identified in Table 2.5.



.......................................................................

2.5 Analysis and discussion

.......................................................................

2.4.3 Maser emission

H 2 O and CH 3 OH masers are excellent indicators of star formation activity

(e.g., Beuther et al. 2002b; Moscadelli et al. 2005; de Villiers et al.

2015). We created image cubes of the H 2 O and CH 3 OH spectral lines and

searched for maser features by scanning the entire velocity range. Despite

the limited spectral resolution of our observation setup (see Sect. 2.3), we

found maser emission in 14 of the 18 regions. In Table 2.2 we list the coordinates

of the maser features detected in each region together with the

velocity at which the feature is detected and its intensity. We also compare

the velocities of the maser features with the systemic velocities determined

from H 13 CO + (1–0) observations (López-Sepulcre et al. 2011).

We find that the velocities of H 2 O masers match the H 13 CO + (1–0) velocities,

while the CH 3 OH masers have a larger discrepancy, probably

due to the lower spectral resolution.

The low spectral resolution in our observations compared to the typical

maser line widths (a few km s −1 ; Elitzur 1982; Kalenskii & Kurtz

2016) leads to smearing of the maser intensities. The intensities given

in Table 2.2 should be considered as lower limits. Despite this limitation,

the high angular resolution of our observations can be used to

spatially associate the H 2 O and CH 3 OH masers with the detected continuum

sources. When the angular separation between the continuum

source and the maser is smaller than the synthesized beam size (listed

in Table 2.1), we assume that the maser is associated with the continuum

source. In the last column of Table 2.2 we specify the identifier

of the continuum source (see Table 2.5) with which the maser is associated.

We find 10 continuum sources associated with maser features (see

Section 2.5.4 for more details).

2


In this section, we determine how many sources in our sample are potential

radio-jets. For this purpose, we study the nature of the detected

radio continuum emission, and investigate the association with molecular

outflows, masers, and EGOs 7 /IRAC 4.5 µm band and are usually

7 The so-called extended green objects (EGOs) are sources with bright emission in

the Spitzer maps.

41



.......................................................................

found to be associated with strong shocks and jets (e.g., Cyganowski

et al. 2008, 2009). Based on these criteria, we identify the best radio-jet

candidates in our sample and characterize their properties.

2.5.1 Nature of the radio continuum emission

2

Usually, two mechanisms are invoked to explain the origin of thermal

free-free radiation from ionized gas in star-forming regions: photoionization,

and ionization through shocks (e.g., Gordon & Sorochenko 2002;

Kurtz 2005; Sánchez-Monge et al. 2008, 2013c; Anglada et al. 2018).

In the case of photoionization, ultraviolet (UV) photons with energies

above 13.6 eV are emitted by massive stars and ionize the surrounding

atomic hydrogen. In the second scenario, the ionization is produced

when ejected material associated with outflows and jets interacts in a

shock with neutral and dense material surrounding the forming star (e.g.,

Curiel et al. 1987, 1989; Anglada et al. 1992).

Anglada (1995, 1996) showed that the relation between the radio

luminosity and the bolometric luminosity of young stellar objects (YSOs)

depends on the origin of the ionization: stellar UV radiation, or shocks

(see also Anglada et al. 2018). We used this relation to investigate the

nature of our continuum sources. The solid line in Fig. 2.3 shows the

maximum radio luminosity that a high-mass object of a given luminosity

may have according to its UV radiation, the so-called Lyman continuum

limit that is usually associated with HII regions. The radio luminosity

decreases fast with decreasing bolometric luminosity. In contrast, the

radio luminosity originated in shocks (i.e., radio jets) has a flatter curve.

The dotted line in Fig. 2.3 shows the least-squares fit to the sample of

radio jets studied in Anglada et al. (2018) that are shown as gray squares

in the figure.

We calculated the radio luminosity of our continuum sources as L radio

= S ν d 2 , where S ν is the observed flux density in the C band (listed in

Table 2.5) and d is the distance to the source (listed in Table 2.1). The

bolometric luminosity (L bol ) of each source is uncertain because we lack

high-resolution data at far-infrared wavelengths. The bolometric luminosity

of each region is given in Table A.1 of López-Sepulcre et al. (2011)

and provides an upper limit to the actual luminosity. As a simple approach,

we divided the bolometric luminosity by the number of radio

sources detected within the primary beam to have an estimate of the

42


.......................................................................

Lradio [mJy kpc 2 ]

10 6

10 5

10 4

10 3

10 2

10 1

10 0

Lyman Continuum

Radio jet fit (Anglada et al. 2018)

Radio jet data (Anglada et al. 2018)

This work (see Table B.1)

Spectral index (α) > +0.0

Spectral index (α) < +0.0

96

102

53

63

48

143

144

10 −1

10 −2

10 −3

2

10 −4

10 −3 10 −2 10 −1 10 0 10 1 10 2 10 3 10 4 10 5 10 6

L bol [L ⊙ ]

Figure 2.3: Scatter plot of bolometric luminosity (L bol ) and observed

radio continuum luminosity (L radio ) at 6 cm (C band). Open black circles

correspond to the continuum sources detected in our work that are

located within the primary beam of the K-band (1.3 cm) images (i.e.,

‘iC/iK’ in Table 2.5). Blue and red symbols mark the sources with positive

and negative spectral indices, respectively, as listed in Table 2.5 and

shown in Fig. 2.2. The solid line represents the values expected from

Lyman continuum radiation for a zero-age main-sequence star of a given

luminosity (Thompson 1984). The dashed line is the least-squares fit to

the radio jets reported by Anglada et al. (2018, shown as gray squares),

corresponding to [L radio /mJy kpc 2 ] = 10 −1.90 [L bol /L ⊙ ] +0.59 (see their

Eq. 28).

43



.......................................................................

expected average luminosity for the continuum sources in the region.

Circle symbols in Fig. 2.3 show the continuum sources detected in our

work and located inside the K-band primary beam (i.e., with reliable

flux measurements and listed as iC/iK in Table 2.5). Colored symbols

correspond to those sources for which we could derive the spectral index

(see Fig. 2.2), with blue symbols corresponding to positive spectral

indices (i.e., α > +0.0, mainly thermal emission) and red symbols corresponding

to negative spectral indices. In general, our sources lie in

between the two lines defining the radio jet and HII region regimes 8 . Interestingly,

sources with positive spectral indices (blue symbols) seem to

preferentially follow, although with some dispersion, the relation found

for radio jets, while sources with negative spectral indices (red symbols)

are located closer to the Lyman continuum regime. This favors our previous

interpretation that sources with negative spectral indices may be

slightly extended HII regions that are partially filtered out in the K-band

images.

2

2.5.2 Association with molecular outflows

We investigated the association of radio continuum sources with molecular

outflows by comparing the location of radio sources with respect to

the molecular outflow emission reported mainly by López-Sepulcre et al.

(2010) and Sánchez-Monge et al. (2013d). It is expected that the most

promising radio-jet candidates is in the center of the molecular outflow

emission.

We find a total of 24 radio continuum sources that are spatially associated

with molecular outflow emission (see Table 2.3 and Fig. 2.4 for

more details). Out of these sources, 18 (sources 2, 4, 13, 14, 15, 16, 22,

23, 25, 48a, 48b, 48c, 74, 83, 95, 110, 137 and 143) are located at or

near the geometric center of the molecular outflow emission, while the

remaining 6 (sources 12, 63, 64, 65, 73, and 144) are located within the

outflow lobes. Although we cannot confirm that these 6 sources are at

the base of the outflows detected with single-dish telescopes (with angular

resolutions of 11–29 arcsec), we cannot exclude that they might drive

8 Some sources lie above the solid curve depicting the Lyman continuum limit.

This is in agreement with other studies that report the existence of a population

of HII regions with radio fluxes higher than the Lyman continuum limit (see, e.g.,

Sánchez-Monge et al. 2013a; Cesaroni et al. 2016).

44


.......................................................................

molecular outflows. Further observations of outflow tracers at higher

angular resolution are necessary to confirm and better associated the

molecular outflows with the radio continuum sources. In Table 2.3 we

list the outflow momentum rates reported in the literature (see López-

Sepulcre et al. 2010; Sánchez-Monge et al. 2013d). For source 137, no

outflow momentum rate has been reported (Hatchell et al. 2001; Liu

et al. 2013).

2.5.3 Association with EGOs

In this section, we investigate the association of radio continuum sources

with Spitzer/IRAC 4.5 µm emission tracing EGOs, which are considered

related to the shocked gas. For the association with EGOs we used the

catalogs of Cyganowski et al. (2008, 2009). In total, we found six sources

(sources 42, 63, 64, 119, 137 and 139) with an EGO counterpart (see

Table 2.3).

We also inspected the Spitzer/IRAC images of the different regions

to search for other possible EGOs not included in previous catalogues.

We identified nine radio continuum sources in this category (see sources

48, 65, 73, 74, 83, 110, and 143, marked with a questionmark in Table

2.3). The association of these sources with bright 4.5 µm emission

suggests their association with strong shocks and favors the hypothesis

of a radio-jet origin for the radio continuum emission of these objects.

However, a more detailed characterization of the infrared properties of

the nine additional sources is necessary to confirm whether these objects

are EGOs.

2

45

2

46

Table 2.3: Properties of the radio-jet candidates.

Flux properties a Source size properties b Outflow/shock activity c

ID S C band S K band α θ C band θ K band β log( ˙ P out) EGOs Masers

Radio-jet candidates with signposts of outflow activity

2 d 0.53±0.01 — — 0.75 — — −3.9 n CH 3 OH

4 d 0.33±0.01 — — 1.23 — — −3.9 n . . .

12 d 0.72±0.03 — — 0.97 — — −3.1 † n . . .

13 d 1.24±0.05 — — 0.75 — — −3.1 n . . .

14 d 0.69±0.05 — — 1.08 — — −3.1 n CH 3 OH

15 d 0.94±0.04 — — 1.60 — — −3.1 n . . .

16 d 1.04±0.04 — — 1.16 — — −3.1 n . . .

22 d 10.27±0.22 — — 1.55 — — −3.3 n CH 3 OH

23 d 1.56 ±0.05 — — . . . — — −3.3 n . . .

25 d 0.49 ±0.02 — — . . . — — −3.3 n . . .

42 4.16±0.08 . . . . . . 0.81 . . . . . . . . . Y . . .

48a e 55.32±3.50 15.19±1.51 −0.99 ± 0.09 2.70 2.07 −0.20 −2.9 ? H 2 O

48b e 75.41±9.50 10.53±1.30 −1.50 ± 0.14 3.84 1.64 −0.65 −2.9 ? . . .

48c e 129.52±8.30 54.41±3.51 −0.66 ± 0.07 2.28 1.89 −0.14 −2.9 ? . . .

63 e 0.99±0.06 0.75±0.13 −0.22 ± 0.07 1.19 1.11 −0.05 −2.6 † Y H 2 O

64 e 0.28±0.02 0.35±0.08 +0.18 ± 0.04 <1.38 0.66 > −0.56 −2.6 † Y H 2 O, CH 3 OH

65 e 0.57±0.14 2.31±0.09 +1.08 ± 0.19 <1.46 <0.75 . . . −2.6 † ? . . .

73 e 0.24±0.15 0.43±0.09 +0.45 ± 0.50 0.38 <0.74 < +0.51 −2.4 † ? . . .

74 e 0.35±0.03 0.49±0.13 +0.26 ± 0.21 0.53 0.42 −0.18 −2.4 ? . . .

83 e 3.35±0.21 3.73±0.29 +0.08 ± 0.08 0.87 0.85 −0.02 −1.8 ? H 2 O, CH 3 OH

95 <0.042 0.25±0.07 > +1.36 . . . <0.42 . . . −1.8 n . . .

110 e 0.46±0.21 1.20±0.06 +0.73 ± 0.35 <1.39 0.37 > −1.01 −2.9 ? H 2 O, CH 3 OH

119 1.11±0.06 < 0.022 . . . 2.28 . . . . . . . . . Y . . .

136 <0.10 0.85±0.12 > +1.67 . . . 1.50 . . . . . . n H 2 O

137 e 14.40±1.20 2.21±0.20 . . . 2.53 1.60 −0.35 . . . ‡ Y . . .

139 0.73±0.03 < 0.019 . . . 0.87 . . . . . . . . . Y . . .

143 e 39.50±1.60 130.87±2.60 +1.12 ± 0.04 0.55 0.28 −0.52 −3.4 ? H 2 O, CH 3 OH

144 15.57±0.89 <0.32 < −2.97 2.44 . . . −3.4 † n . . .



.......................................................................

47

Radio continuum sources consistent with positive spectral index, but with no signposts of outflow activity

61 0.14±0.03 <2.10 < +2.07 1.79 . . . . . . . . . n . . .

62 <0.05 0.24±0.09 > +1.23 . . . <0.44 . . . . . . n . . .

86 0.35±0.01 <2.32 < +1.43 <1.45 . . . . . . . . . n . . .

109 0.08±0.03 <1.75 < +2.33 <0.94 . . . . . . . . . n . . .

113 0.28±0.01 <0.31 < +0.07 0.85 . . . . . . . . . n . . .

126 0.11±0.01 <0.37 < +0.96 <1.37 . . . . . . . . . n . . .

129 0.16±0.01 <0.20 < +0.18 1.23 . . . . . . . . . n . . .

145 2.84±0.02 <4.34 < +0.32 <1.48 . . . . . . . . . n . . .

Notes.

(a) Primary beam corrected fluxes in mJy as listed in Table 2.5. For sources 42, 119, 137 and 139 it was not possible

primary beam correct the fluxes at both bands (see Sect 2.4), resulting in not usable spectral indices. The spectral index α is

defined in Eq. 2.2. (b) Source sizes in arcsec determined as √ θ major × θ minor, with θ major and θ minor listed in Table 2.6. Upper

limits corresponds to sources for which we could not determine a deconvolved source size. The source size index β is defined

in Eq. 2.2.

(c) Association of the radio continuum source with outflow and shock activity. The associations correspond to

(i) molecular outflows, with the outflow momentum rate P ˙ out given in units of M ⊙ yr −1 km s −1 (from López-Sepulcre et al.

2010; Sánchez-Monge et al. 2013d), with the dagger indicating those radio continuum sources located within the outflow lobes

and not at the center of the outflow, (ii) EGOs (or extended green objects), based on the catalog of Cyganowski et al. (2008,

questionmarks indicate the presence of bright Spitzer/IRAC 4.5 µm emission although without confirmation of the object being

an EGO), and (iii) H 2O and CH 3OH masers, as listed in Table 2.2. Source 137, marked with a double cross, is associated with

molecular outflow emission (Hatchell et al. 2001; Liu et al. 2013), but no outflow momentum rate has been reported. (d) Sources

not observed in the K band. For these sources we do not have information on the K-band flux and presence of H 2O masers.

(e) Sources detected at both frequency bands and for which we have created new images using a common uv-range that allows

us to sample similar spatial scales. Fluxes and source sizes for these sources are taken from Table 2.7. Fluxes for source 137

can not be primary beam corrected and cannot be used to determine a spectral index.


.......................................................................

2



.......................................................................

2.5.4 Association with masers

2

Our VLA observations (see Sect. 2.3.2) allow us to search for H 2 O and

CH 3 OH maser spots associated with radio continuum sources. As shown

in Table 2.2, we have found 16 H 2 O and 7 CH 3 OH maser spots.

We find ten radio continuum sources associated with maser features,

of which three (sources 2, 14, and 22) are associated with CH 3 OH masers

only, three sources (sources 48, 63, and 136) are associated only with H 2 O

masers, and four sources (sources 64, 83, 110, and 143) are associated

with both types of masers (see Table 2.3 and Fig. 2.4). It is worth noting

that the three sources associated only with CH 3 OH masers correspond

to regions observed only in the C band. Future observations of these

sources in the K band together with observations of the H 2 O maser line

could confirm that all sources associated with CH 3 OH maser are also

associated with H 2 O maser features.

The observed Class II 6.7 GHz CH 3 OH masers are ideal indicators

for embedded YSOs and mark the location of deeply embedded massive

protostars (e.g., Breen et al. 2013). On the other hand, 22 GHz H 2 O

masers have been found associated with outflow activity (e.g., Torrelles

et al. 2011) as well as tracing disk-like structures around young stellar

objects (e.g., Moscadelli et al. 2019). Our maser observations have

therefore enabled us to identify at least seven potential candidates for a

radio-jet (i.e., sources associated with outflow activity).

Table 2.4: Number of the sources with thermal and nonthermal radio

continuum emission associated with different outflow activity signatures.

Nonthermal Sources Thermal Sources

Outflows 5/5 (100%) 7/8 (88%)

EGOs 1/5 (20%) 1/8 (13%)

Masers (all) 2/5 (40%) 5/8 (63%)

H 2 O 2/5 (40%) 5/8 (63%)

CH 3 OH 0/5 (0%) 4/8 (50%)

2.5.5 Radio-jet candidates

Out of the 146 radio continuum sources detected in our study, we identified

28 sources (see list at the beginning of Table 2.3) as possible radio-jet

48


.......................................................................

Radio continuum sources

associated with…

outflows

14

1

7

2

0

1

masers

3

EGOs

Association with outflows: Association with EGOs: Association with masers:

8

Non-thermal

Thermal

5

+11 with no

spectral index

1

Non-thermal

Thermal

1

+4 with no

spectral index

5

Non-thermal

Thermal

2

+3 with no

spectral index

Figure 2.4: Diagrams summarizing the outflow-activity associations of

the radio-jet candidates studied in this work. The integer numbers indicate

the number radio-jet candidates in a specific group (see Table 2.3).

The top diagram summarizes the association of the radio-jet candidates

with molecular outflows, masers, and EGOs (see Sect. 2.5, for EGOs we

only consider an association if the source is labeled ‘Y’ in Table 2.3).

The bottom row diagrams summarize the results regarding the thermal

(spectral index > −0.1) and nonthermal (spectral index < −0.1) properties

of the radio continuum emission. The number of sources for which

we could not derive the spectral index is also indicated.

2

candidates, based on their association with outflow and shock activity.

We find 24 of these sources associated with molecular outflow emission,

6 of them with EGOs, and 10 with masers. In the sketch presented in

Fig. 2.4, we summarize these findings.

In addition to these 28 sources, we also identified 8 radio-continuum

sources with spectral indices consistent with thermal emission (see bottom

list in Table 2.3). Based on the results shown in Fig. 2.3, these

sources could also be radio-jet candidates, despite their lack of association

with tracers of outflow and shock activity. In the following, we build

49



.......................................................................

on the properties of the identified radio-jet candidates.

Radio continuum properties

Reynolds (1986) describe radio-jets with a model that assumes a jet of

varying temperature, velocity, and ionization fraction. In case of constant

temperature, the relations of the flux density (S ν ), and source size

(θ ν ) with frequency are given by

S ν ∝ ν α = ν 1.3−0.7/ɛ and θ ν ∝ ν β = ν −0.7/ɛ , (2.2)

2

where ɛ depends only on the geometry of the jet and is the power-law

index that describes how the width of the jet varies with the distance

from the central object. In this model, the spectral index α is always

smaller than 1.3 and drops to values < 0.6 for confined jets (ɛ < 1;

Anglada et al. 1998).

In Table 2.3 we list the spectral index (α) and the source size index

(β) for our radio-jet candidates. The latter only for the sources

detected at both frequencies. Nine of the radio-jet candidates associated

with outflow/shock activity have spectral indices consistent with thermal

emission (> −0.1), with six showing clear positive (> +0.4) spectral

indices. These values are consistent with the model of Reynolds (1986)

for values of ɛ > 0.6. For such geometries of the jet, the source size

index (β) is expected to be about −1. The source size indices reported

in Table 2.3 are mainly in the range −0.1 to −1.0, in agreement with the

model of Reynolds (1986) for radio-jets.

Although most of our radio-jet candidates have spectral indices consistent

with thermal emission (64% of the sample, see Table 2.3), we find

some sources (accounting for 36% of the sample, five sources 9 ) that show

negative spectral indices. This finding is in agreement with some recent

works. For example, Moscadelli et al. (2016) find about 20% of their

sample of 15 radio continuum sources to be associated with nonthermal

emission. The presence of nonthermal emission is explained in terms

of synchrotron emission from relativistic electrons accelerated in strong

shocks within the jets, and a number of cases have been studied in more

9 As discussed in Sect. 2.5.5, four of these five sources are most likely HII regions.

This would reduce the number of nonthermal radio-jets to only one out of 14 (7% of

our sample).

50


.......................................................................

detail (e.g., Carrasco-González et al. 2010; Sanna et al. 2019). Further

detailed observations toward these new four nonthermal radio-jet candidates,

can provide further constraints to understand the characteristics

of these types of objects.

We searched for possible relations between the presence of thermal

and non-thermal radio-jets and different outflow/shock activity signs

(i.e., outflows, masers, and EGOs). We summarize our findings in the

bottom panels of Fig. 2.4. We do not find a preferred relation between

thermal and nonthermal radio-jets with the outflow activity signs, since

we find similar percentages (see Table 2.4) for the association with outflows

(88% and 100%, respectively), EGOs (13% and 20%), and masers

(55% and 40%). The low number of objects included in our analysis prevents

us from deriving further conclusions, and we indicate that a larger

sample of radio-jets needs to be studied to better understand the properties

and differences between thermal and nonthermal radio-jets. It is also

worth noting that all the four objects associated with both CH 3 OH and

H 2 O masers are thermal radio-jet candidates (see Table 2.3), while only

one of the three objects associated with only H 2 O masers shows thermal

emission. This might suggest that radio-jets associated with CH 3 OH

masers tend to have positive spectral indices (i.e., thermal emission),

while radio-jets associated with only H 2 O masers might preferentially

have negative spectral indices (i.e., nonthermal emission). However, the

low number of sources studied in our sample prevents from deriving

further conclusions. The different levels of association of the radio continuum

sources with maser emission may be affected by the variability

of masers (Felli et al. 2007; Sugiyama et al. 2017; Ashimbaeva et al.

2017). Moreover, we cannot discard that the poor spectral resolution of

our observations, which may smear out the intensity of the maser lines

making some of them undetectable with our sensitivity limits, may also

affect our detectability limits. Despite these limitations, our results are

in agreement with the 6.7 GHz CH 3 OH masers tracing the actual location

of the newly born YSOs usually associated with thermal winds/jets,

while H 2 O masers may be originated in strong shocks where nonthermal

synchrotron emission can be relevant (see, e.g., Moscadelli et al. 2013,

2016).

2

51



.......................................................................

log Lradio [mJy kpc 2 ]

5

4

3

2

1

0

−1

−2

−3

Radio jet fit (Anglada et al. 2018)

Anglada et al. (2018)

This work (at the outflow center)

This work (in the outflow lobe)

143

144

22

48

−4

−6 −4 −2 0

log Ṗout [M ⊙ yr −1 km s −1 ]

2

Figure 2.5: Relation between radio luminosity (L radio ) and outflow

momentum rate ( P ˙ out ). Open boxes show data from Anglada

et al. (1992, 2018). The dashed line is a least-squares fit to

the radio-jets reported by Anglada et al. (2018), corresponding to

[L radio /mJy kpc 2 ] = 10 +2.97 [ P ˙ /M ⊙ yr −1 km s −1 ] +1.02 (see their Eq. 31).

Jet-outflow connection

It has been found that the radio luminosity (L rad = S ν d 2 ) of thermal

radio-jets is correlated with the energetics of the associated molecular

outflows. The relation between radio luminosity and momentum rate in

the outflow ( P ˙ out ) was empirically derived by Anglada et al. (1992, see

also Anglada et al. 2018). In Fig. 2.5 we compare our radio-jet candidates

(see Table 2.3) with the radio-jets studied by Anglada et al. (2018).

As reported by Anglada et al. (2018), there is a tight correlation between

the radio luminosity of the jet and the outflow momentum rate. This

relation is interpreted as proof that shocks are the ionization mechanism

of radio-jets (see, e.g., Rodríguez et al. 2008; Anglada et al. 2018). Most

of the radio-jet candidates investigated in this work, with the exception

of only a few sources, follow this relation, suggesting a radio-jet origin

52


.......................................................................

for the detected radio continuum emission. The exceptions are mainly

sources 48a, 48b, 48c, 143, and 144, which have a much higher radio

luminosity than the associated outflow momentum rate. This excess

suggests that another mechanism could be responsible for a large fraction

of the observed radio continuum emission. Based on the location

of these sources in the diagram shown in Fig. 2.3, these sources may

correspond to more evolved and extended HII regions instead of radiojets,

which would explain the discrepancy between the observed radio

luminosity and outflow momentum rate. In this case, we could be facing

two possible scenarios. The first is that the sources are indeed radio-jets

that transition into a more evolved HII region phase (similar to what

has been proposed for G35.20−0.74N, Beltrán et al. 2016). The second

scenario is that the radio continuum sources that we are detecting

are only associated with an HII region, and the spatial coincidence with

the molecular outflow emission is due to the presence of another (lower

mass) object that powers the outflow, but with nondetectable radio continuum

emission in our maps. Higher angular resolution observations of

the molecular outflow can better establish the location of the powering

source and its association with the detected radio continuum sources.

Following Eq. 8 of Anglada et al. (2018, see also Reynolds 1986), we

estimated the ionized mass-loss rate (Ṁion) of our radio-jet candidates

as

(

)

Ṁ ion

10 −6 M ⊙ yr −1 =0.108

( d

) 1.5 [ ] (2 − α) (0.1 + α) 0.75

kpc 1.3 − α

( ) T −0.075 [( )

Sν

(

ν

) ] −α 0.75

×

10 4 K mJy 10 GHz

(

)

V ( jet νm

) 0.75α−0.45

×

200 km s −1 10 GHz

( ) 0.75 θ0

× (sin i) −0.25 ,

rad

(2.3)

2

where α is the spectral index and S ν is the radio continuum flux, both

listed in Table 2.3, and d is the distance to the source. The opening angle

of the jet θ 0 can be approximated as 2 arctan(θ min /θ maj ) (Beltrán et al.

2001; Anglada et al. 2018). We assumed a value of 0.5 for the ratio of

53



.......................................................................

2

the minor and major axis of the jet. We also assumed that the velocity

of the jet (V jet ) is 500 km s −1 and that it lies in the plane of the sky (i.e.,

sin i = 1). For a random orientation of the jet in the celestial plane, the

value of sin i is on average π/4 (e.g., Beltrán et al. 2001). Usually, a

value of T = 10 4 K is adopted for ionized gas. For the turnover frequency

ν m , we assumed a value of 40 GHz (see the discussion in Anglada et al.

2018). In Fig. 2.6 we show the relation between the mass-loss rates of the

ionized and the molecular outflow for the thermal radio-jet candidates

listed in Table 2.3 and associated with the molecular outflows. Major

uncertainties in the determination of Ṁ ion may arise from parameters

such as the jet velocity, the turnover frequency, or the aspect ratio of

the jet because they cannot be determined from our observational data.

However, their effects are almost negligible, and variations within reasonable

ranges result in variations of the ionized mass-loss rate of less

than a factor of 10. Our derived Ṁion are mainly in the range of 10 −7 to

10 −5 M ⊙ yr −1 , consistent with the values reported for low-mass radiojets

(≈10 −10 M ⊙ yr −1 ) and high-mass radio-jets (≈10 −5 M ⊙ yr −1 , see

Rodriguez et al. 1994; Beltrán et al. 2001; Guzmán et al. 2010, 2012a).

The dashed lines in Fig. 2.6 indicate different degrees of ionization for

the mass-loss rate. Most of our radio-jet candidates, with the exception

of source 143, which is probably associated with an already developed

HII region (see Fig. 2.5 and discussion above), have ionization levels of

Ṁ ion = 10 −3 × Ṁout. These values are about one to two orders of magnitude

lower than those reported in previous studies (see, e.g., Rodriguez

et al. 1990; Hartigan et al. 1994; Bacciotti et al. 1995; Anglada et al.

2018). This difference may be due to uncertainties in the assumed parameters

of Eq. 2.3, as well as to the fact that our molecular outflow

emission is studied with a single dish (sensitive to all scale structures),

while the radio-jet observations were carried out with a large interferometric

configuration that probably resolved out part of the radio-jet

emission.

Best radio-jet candidates

In previous sections, we have analyzed the properties of the 146 detected

radio continuum sources and built a sample of possible radio jet candidates

based on their association with outflow activity: molecular outflows,

EGOs, and masers (see Table 2.3). In Sect. 2.5.5 and 2.5.5 we have

54


.......................................................................

10 −3

10

1

10 −4

143

10 −1

Ṁion [M⊙ yr −1 ]

10 −5

10 −6

10 −7

65

64

110

73

74

10 −2

10 −3

10 −4

83

10 −4 10 −3

Ṁ out [M ⊙ yr −1 ]

Figure 2.6: Relation between ionized (Ṁion) and molecular outflow

(Ṁout) mass-loss rates for the radio-jet candidates listed in Table 2.3.

The ionized mass-loss rate is derived for the thermal radio jets using

Eq. 2.3, while the molecular outflow mass-loss rate is provided in López-

Sepulcre et al. (2010) and Sánchez-Monge et al. (2013d). The dashed

lines indicate different ionization levels given by the ratio Ṁion to Ṁout.

2

investigated in more detail the possible nature of the radio continuum

emission and its relation to the outflow activity tracers, in particular,

the outflow momentum rate. The results presented in Figs. 2.3 and 2.5

allow us to identify sources with properties that differ from those expected

from radio jets, which therefore suggests that these sources are

not radio jets after all. From this analysis and the individual description

of selected sources (see Appendix 2.9), we discuss in this section which

objects are most likely to be radio jets.

Out of the 28 sources listed in Table 2.3, 5 have radio luminosities

similar to those expected for HII regions: sources 48a, 48b, 48c, 143, and

144 (see Fig. 2.3). In addition, all of these sources, with the exception of

source 143, have negative spectral indices. These negative values could

be due to a slight extension of the sources (as expected for HII regions)

55



.......................................................................

2

and partial filtering-out in the K-band images. Moreover, these 5 sources

also exhibit higher radio luminosities than their associated outflow momentum

rates (see Fig. 2.5), which supports the interpretation that there

may be an excess of radio continuum emission that is not necessarily associated

with a radio jet, but with an HII region. In the absence of further

evidence, we are not in a position to interpret further, and we cannot

consider these sources to be among the best radio jet candidates. Further

observations, sensitive to extended emission, can provide the necessary

information to better characterize these sources in terms of their spatial

extent and the nature of the emission. It is also worth noting that

sources with negative spectral indices could correspond to background

sources with synchrotron radiation because we expect about 11 objects

in our sample to have this possible origin (see Sect. 2.4.1). Source 22

also shows an excess of radio continuum emission compared to its outflow

momentum rate, which suggests that this is also a dubious radio-jet

candidate. From the individual source descriptions presented in Appendix

2.9, sources 42 and 137 seem to be radio continuum sources with

most of their emission dominated by cometary/ultracompact HII regions,

which makes it difficult to identify a radio jet in our data.

Out of the remaining sources listed in Table 2.3, we can identify seven

that have a high probability to be radio jets. These are sources 2, 14, 22,

64, 74, 83, and 110, which are associated with additional outflow/shock

activity such as masers and EGOs. Source 74 is adjacent to two H 2 O

maser features, which are only 2 ′′ away and coincident with extended

4.5 µm emission (see Fig. 2.13). The remaining sources do still classify

as radio-jet candidates because we do not have clear evidence against

it. Some of them are located at the center of molecular outflows (e.g.,

source 95) but are not associated with additional outflow/shock signs.

This could be related to the variability of H 2 O masers (see Sect. 2.5.5).

Others are located within molecular outflow lobes (e.g., sources 12, 63,

64, 65, 73, and 144), and for which higher angular resolution observations

of outflow tracers are necessary to confirm if they are powering some of

the molecular outflows. Although they are not associated with molecular

outflows, other sources show other shock activity signs such as the presence

of EGOs (e.g., sources 119, 136, and 139). Further observational

constraints are therefore needed to fully confirm or discard these objects

as radio jets. The results acquired so far allow us to classify them as

56

2.6 Implications for high-mass star formation

.......................................................................

radio-jet candidates.

2.6 Implications for high-mass star formation

Recently, Rosero et al. (2019) studied the properties of 70 radio continuum

sources associated with the earliest stages of high-mass star formation.

They find that ≈30–50% of their sample are ionized jets. This

fraction is in agreement with our findings. Out of the 146 radio continuum

sources detected in our study, we identify 28 possible radio jets

(i.e., about 19% of our sample). However, when we focus on the sources

for which we have more accurate information (i.e., sources classified as

iC/iK in Table 2.5, see also Sect. 2.4.1), 24 out of 40 sources are potential

radio jets. Therefore the percentage of radio continuum sources that are

radio jets increases to 60%. This suggests that about half of the radio

continuum sources found in star-forming regions at early evolutionary

stages may indeed be radio jets powered by young stars. The remaining

≈50% of objects could still be radio jets for which we have not yet identified

shock activity signs, or they could represent extremely compact

HII regions in early stages of their development. These objects could be

powered by early B-type stars and not necessarily by the most massive

stars, and could be an intermediate stage between radio jets and already

developed HII regions (see, e.g., Beltrán et al. 2016; Rivera-Soto et al.

2020).

López-Sepulcre et al. (2010) classified the regions studied in our

work as infrared-dark (IRDC, infrared-dark cloud) and infrared-bright

(HMSFR, high-mass star-forming region) based on their detectable infrared

emission. Our sample therefore consists of two subclasses: IRDC

(eight regions) and HMSFR (ten regions; see Table 2.1). We assume that

these two types belong to different evolutionary phases of massive star

formation, with the IRDC regions being less evolved than the HMSRF

regions. Considering the 40 sources located within the primary beam

of our images (i.e., sources classified as iC/iK), we find ≈2.8 radio continuum

sources per HMSFR region, and ≈1.8 radio continuum sources

per IRDC region. This suggests that it is more probable to detect compact

radio continuum emission in more evolved regions. Regarding the

presence of radio jets in these two evolutionary stages, we find 21 out of

the 28 radio jet candidates listed in Table 2.3 in HMSFR regions (cor-

2

57



.......................................................................

responding to 75%), while we only find 7 (corresponding to 25%) in the

less evolved IRDC regions. When we consider only the best radio-jet

candidates (see Sect. 2.5.5), we find five radio jets (sources 2, 14, 22,

64, and 83; corresponding to 71%) in HMSRF regions and two radio jets

(sources 74 and 110; corresponding to 29%) in IRDC regions. This shows

a preference of radio jets to be found in more evolved clouds. Complementary

to this, we can determine the fraction of IRDC and HMSFR

regions that harbor radio jets. Out of the eight IRDC regions studied in

this work, only two (corresponding to 25%) harbor one of the best radiojet

candidates. This increases to 50% (five out of ten) for the HMSFR

regions. This means that the frequency of radio jets in IRDC regions

is lower than in HMSFR regions. One possible explanation is that the

jets may become larger and brighter with time. Our limited data do not

show that IRDC jets are smaller or fainter than HMSFR jets, but future

work on larger source samples may provide further insight.

2

2.7 Summary

We have used of the VLA in two different bands (C and K band, corresponding

to 6 and 1.3 cm wavelengths) to search for radio jets powered

by high-mass YSOs. We studied a sample of 18 high-mass star-forming

regions with signs of SiO and HCO + outflow activity. In the following

we summarize our main results.

We have identified 146 radio continuum sources in the 18 high-mass

star-forming regions, with 40 of the radio continuum sources located

within the primary beams of our images (i.e., labeled iC/iK and with

reliable flux measurements). Out of these sources, 131 (27 iC/iKs) are

only detected in the C band, 4 (3 iC/iKs) are only detected in the K

band, and 11 (10 iC/iKs) are detected in both bands. This different

detection level is likely due to different factors: (i) four regions were not

observed in the K band, (ii) the C-band images have a larger field of

view, and (iii) the K-band images are affected by a larger interferometric

spatial filtering. In addition to the continuum emission, we detected 23

maser features in the 6.7 GHz CH 3 OH and 22 GHz H 2 O lines.

Out of the 146 continuum sources, only 40 sources are located within

the field of view of both images, allowing for an accurate characterization

of their radio properties. For these sources we derived the spectral index,

58

2.7 Summary

.......................................................................

which we find to be consistent with thermal emission (i.e., in the range

−0.1 to +2.0) for most of the objects (73%).

We investigated the nature of the radio continuum emission by comparing

the radio luminosity to the bolometric luminosity. We find that

most sources with positive spectral indices (i.e., thermal emission) follow

the trend expected for radio jets, while sources with large negative spectral

indices seem to follow the relation expected for HII regions. These

large negative spectral indices probably arise because the emission in the

K-band images is partially filtered out.

Based on the association of the radio continuum sources with shock

activity signs (i.e., association with molecular outflows, EGOs, or masers),

we compiled a list of 28 radio-jet candidates. This corresponds to ≈60%

of the radio continuum sources located within the field of view of both

VLA images. In this sample of radio-jet candidates, we identified 7 objects

(sources 2, 14, 22, 64, 74, 83, and 110) as the most probable radio

jets. The remaining 21 require additional observations, either at different

radio frequency bands or of molecular outflow tracers at higher

resolution, to confirm or discard them as radio jets.

We find about 7–36% of the possible radio jet candidates to show

nonthermal radio continuum emission. This is consistent with previous

studies reporting ≈20% nonthermal radio jets. We do not find a clear

association of the nonthermal emission with the presence of outflows,

EGOs, or masers. However, despite the low statistics, we find that radio

jet candidates associated with CH 3 OH masers have thermal emission,

while the radio-jet candidates associated with only H 2 O masers tend

to have nonthermal emission. This is in agreement with the 6.7 GHz

CH 3 OH masers tracing the actual location of newly born YSOs powering

thermal winds and jets, while the H 2 O masers may originate in strong

shocks where nonthermal emission becomes relevant.

As previously found in other works, we find a correlation between the

radio luminosity of our radio-jet candidates and their associated outflow

momentum rates. We derive an ionized mass-loss rate in the range 10 −7

to 10 −5 M ⊙ yr −1 , which results in ionization levels of Ṁ ion = 10 −3 Ṁ out

(i.e., ≈0.1% of the outflow mass that is ionized).

The 18 high-mass star-forming regions studied in this work are classified

into two different evolutionary stages: eight less evolved IRDC,

and ten more evolved HMSFR. We find more radio continuum sources

2

59



.......................................................................

(≈2.8 sources per region) in the more evolved HMSFR than in the IRDC

(≈1.8). Regarding radio jets, we find about 71% of the radio jet candidates

to be located in HMSFR regions, and only 29% in IRDC regions.

Complemenary to this, 25% of the IRDC regions harbor one of the most

probable radio-jet candidates, while this percentage increases to 50% for

the HMSFR regions. This suggests that the frequency of radio jets in

the less evolved IRDC regions is lower than in the more evolved HMSFR

regions.

2.8 Acknowledgements

2

The authors thank the referee for his/her review and greatly appreciate

the comments and suggestions that have contributed significantly to improve

the quality of the publication. ÜK also thanks Jonathan Tan for

useful discussions. This work has been partially supported by the Scientific

and Technological Research Council of Turkey (TÜBİTAK), project

number: 116F003. Part of this work was supported by the Research

Fund of Istanbul University, project number: 44659. ASM research is

carried out within the Collaborative Research Centre 956, sub-project

A6, funded by the Deutsche Forschungsgemeinschaft (DFG; project ID

184018867). ÜK would like to thank William Pearson for checking the

language of the paper and Kyle Oman for helping with Python issues.

2.9 Comments on individual sources

In the following, we comment on various aspects of selected sources for

which additional literature information is available. We presented the

region(s) where the radio-jet maps with outflow emission contours are

given.

IRAS 05358+3543 (sources 2 and 4)

In region IRAS 05358+3543, we identified two radio continuum sources

that can be potential radio jets (see Fig. 2.7). Sources 2 and 4 were observed

only in the C band, and therefore we cannot determine a spectral

index for these sources. Despite this limitation, both sources are located

at the center of the molecular outflow reported by López-Sepulcre et al.

60


.......................................................................

(2010). Furthermore, source 2 is associated with a 6.7 GHz CH 3 OH

maser emission, suggesting that this marks the position of a massive

YSO. Of the two radio continuum sources, source 2 is likely the main

object powering the molecular outflow for which its centimeter emission

traces a radio jet. Further observations at different frequency bands are

necessary to better constrain its properties.

G189.78+0.34 (sources 12, 13, 14, 15, and 16)

In region G189.78+0.34, we identified five radio continuum sources (from

12 to 16) associated with the molecular outflow reported by López-

Sepulcre et al. (2010). All of them are located at the center of the

outflow, with source 12 slightly offset from the rest (see Fig. 2.8). All

these sources were observed only in the C band, and no spectral index can

be derived. Out of the five sources, source 14 is associated with CH 3 OH

maser emission (see also Caswell et al. 2010), suggesting that this marks

the location of a massive YSO. The radio continuum sources are found

in an elongated chain extending from the southeast to the northwest.

This direction is consistent with the orientation of the molecular outflow

(López-Sepulcre et al. 2010). Overall, we consider that source 14 is the

powering source and most likely the main component of the radio jet.

The other sources could correspond to different radio continuum knots

located along the jet, as seen in other sources (e.g., HH 80-81: Carrasco-

González et al. 2010, and G35.20−0.74 N: Beltrán et al. 2016), where the

radio continuum knots usually show nonthermal spectral indices. Observations

at different frequency bands are necessary to gather information

on the spectral index and thermal/nonthermal nature of the different

radio continuum sources in the region.

2

G192.58−0.04 (sources 22, 23, and 25)

In region G192.58−0.04, we identified sources 22, 23, and 25 as potential

radio jets (see Fig. 2.9). These sources were observed only in the C band

and no spectral index can be derived. Out of the three sources, source

22 is associated with CH 3 OH maser emission, suggesting that this source

may mark the location of a massive YSO. The sources are located at the

center of the molecular outflow reported by López-Sepulcre et al. (2010).

Out of the three sources, we consider that source 22 is the most probable

61



.......................................................................

Dec (J2000)

56 ′′

54 ′′

52 ′′

50 ′′

+35 ◦ 45 ′ 48 ′′

IRAS 05358+3543

0.05 pc

2

5 h 39 m 13 s

RA (J2000)

4

0.40

0.35

0.30

0.25

0.20

0.15

0.10

0.05

0.00

mJy/Beam

2

Figure 2.7: VLA C-band (6 cm) continuum emission map of radio-jet

candidates 2 and 4 located in the region IRAS 05358+3543. The pink

ellipse is the beam size of the C band. The orange star marks the location

of the CH 3 OH maser (see Table 2.2). The gray double-headed arrow

indicates the direction of the outflows.

radio jet. The comparison between the radio continuum luminosity with

the outflow momentum rate (see Fig. 2.5) also confirms this possibility,

although there appears to be a slight excess of radio continuum emission,

suggesting that there can be an additional contribution to the radio

continuum source (e.g., from an HII region in an early stage).

IRAS 18223−1243 (source 42)

In the region IRAS 18223−1243, we identified the radio continuum source

42 (see Fig. 2.10) as adjacent to the one reported in Cyganowski et al.

(2011) EGO F G18.67+0.03−CM1. This is the only sign of shock activity

because neither molecular outflow nor maser emission are found

for this object. In addition, Cyganowski et al. (2012) reported a massive

protocluster consisting of a hot molecular core and an ultracompact

HII region. Our source appears to be located at the same position as the

ultracompact HII region, which makes us to doubt whether this can be

62


.......................................................................

Dec (J2000)

12 ′′

09 ′′

06 ′′

03 ′′

+20 ◦ 39 ′ 00 ′′

0.05 pc

G189.78+0.34

16

14

15

13

12

6 h 08 m 35 s

RA (J2000)

0.40

0.35

0.30

0.25

0.20

0.15

0.10

0.05

0.00

mJy/Beam

Figure 2.8: VLA C-band (6 cm) continuum emission map of radio jet

candidates 12, 13, 14, 15, and 16 located in the region G189.78+0.34.

The pink ellipse is the beam size of the C band. The orange star marks

the location of the CH 3 OH maser (see Table 2.2). The gray doubleheaded

arrow indicates the direction of the outflows.

2

a radio-jet candidate.

IRAS 18228−1312 (source 48)

Radio continuum source 48 is observed as a group of three compact

sources (sources 48a, 48b, and 48c) surrounded by an extended and more

diffuse structure. One of these compact sources (source 48a) is clearly

associated with H 2 O maser emission (see Fig. 2.11). The Spitzer/IRAC

4.5 µm emission is extended and spatially coincident with the radio continuum

extended emission. The spectral indices for these three compact

sources are in the range −0.6 to −1.5, most likely due to additional filtering

of the emission in the K-band image. Previous studies have classified

this extended source as a region containing hypercompact (HC) and

ultracompact (UC) HII regions (e.g., Chini et al. 1987; Lockman 1989;

Kurtz et al. 1994; Kuchar & Clark 1997; Leto et al. 2009), which is consistent

with our derived radio continuum luminosity (see Figs. 2.3 and

63



.......................................................................

Dec (J2000)

30 ′′

27 ′′

24 ′′

21 ′′

+17 ◦ 59 ′ 18 ′′

0.05 pc

25

G192.58-0.04

22

23

6 h 12 m 54 s

RA (J2000)

0.8

0.7

0.6

0.5

0.4

0.3

0.2

0.1

0.0

mJy/Beam

2


candidates 22, 23, and 25 located in the region G192.58−0.04. The

pink ellipse is the beam size of the C band. The orange star marks the

location of the CH 3 OH maser (see Table 2.2). The gray double-headed

arrow indicates the direction of the outflows.

2.5). The three sources are spatially located at the center of a molecular

outflow (e.g., López-Sepulcre et al. 2010). This may suggest that

one of the compact sources may be powering the molecular outflow. In

this case, this object would be in a evolutionary stage where the radio

jet still exists but a young HII region has already developed, similar to

the high-mass young stellar object G35.20−0.74 N (e.g., Sánchez-Monge

et al. 2013b, 2014; Beltrán et al. 2016).

IRAS 18236−1205 (sources 63, 64, and 65)

We identified nine radio continuum sources in the IRAS 18236−1205 region

(also referred to in the literature as G19.36−0.03), three of which

have been classified as radio jet candidates: sources 63, 64, and 65 with

spectral indices of −0.22 ± 0.07, +0.18 ± 0.04, and +1.08 ± 0.19. We

identified four H 2 O maser features near these sources (see Fig. 2.12).

Two maser features are associated with source 63, one maser feature is

64


.......................................................................

Dec (J2000)

−12 ◦ 39 ′ 06 ′′

12 ′′

18 ′′

24 ′′

30 ′′

36 ′′

0.05 pc

IRAS 18223-1243

42

53 s

54 s RA (J2000)

18 h 24 m 52 s

0.8

0.7

0.6

0.5

0.4

0.3

0.2

0.1

0.0

mJy/Beam


candidate 42 located in the region IRAS 18223−1243. The pink ellipse

is the beam size of the C band. The green circle with a radius of ∼5

arcsec marks the EGO F G18.67+0.03-CM1 reported by Cyganowski

et al. (2011).

2

associated with source 64 (which also spatially coincides with a CH 3 OH

maser), and the last maser feature is located in the center of the redshifted

outflow lobe where no radio continuum emission is detected. The

molecular outflow in this region has been mapped in the lines SiO (2–1)

and HCO + (1–0) by Sánchez-Monge et al. (2013d).

Sources 63 and 64 are associated with EGOs (see Cyganowski et al.

2009), indicating strong shock activity in these two sources. Their location

near the center of the molecular outflow together with their association

with H 2 O maser emission and EGOs suggests that these two sources

could be candidates for radio jets. Source 64 is spatially more coincident

with the geometric center of the outflow, and its association with 6.7 GHz

CH 3 OH maser emission suggests that a massive YSO is located at this

position. This massive YSO could be the driving source of the molecular

outflow seen on larger scales. The third candidate (source 65) is located

∼28 arcsec away from source 64 and the center of the outflow, and

65



.......................................................................

Dec (J2000)

−13 ◦ 10 ′ 00.0 ′′

10.0 ′′

20.0 ′′

30.0 ′′

IRAS 18228-1312

0.1 pc

48b

48a

48c

25

20

40.0 ′′

44 s 43 s 42 s 18 h 25 m 0

41 s

RA (J2000)

15

10

5

mJy/Beam

2

Figure 2.11: VLA C-band (6 cm) continuum emission map of source 48

located in the region IRAS 18228−1312. Three bright peaks are visible

and labeled 48a, 48b, and 48c. The green contour levels of the K-band

(1.3 cm) continuum emission are 3, 5 and 9 times 0.7 mJy beam −1 .

The magenta contours show the Spitzer/GLIMPSE 4.5 µm emission.

The pink and green ellipses are the beam sizes of the C and K bands,

respectively. The white star marks the location of the H 2 O maser (see

Table 2.2).

has been studied by Cyganowski et al. (2011, G19.36-0.03-CM2). This

source is associated with an emission of 4.5 µm, although it is unclear

whether it can be convincingly classified as an EGO (Cyganowski et al.

2009). The positive spectral index indicates thermal emission, which

could come from a radio jet. However, there is no clear evidence of outflow

or shock activity. Source 65 is also located in the vicinity of a dense

core (18236−1205 P8) identified by Lu et al. (2014) in the VLA NH 3

maps, which supports the interpretation of this source as an embedded

YSO. Overall, source 65 could be a YSO-powered radio jet in the vicinity

of the more massive object (sources 63 and 64) in IRAS 18236−1205.

66


.......................................................................

Dec (J2000)

−12 ◦ 03 ′ 30 ′′

45 ′′

04 ′ 00 ′′

15 ′′

30 ′′

28 s IRAS 18236-1205

0.40

0.05 pc

0.35

0.30

63 0.25

64

0.20

0.15

65

0.10

0.05

27 s 26 s 2518 s h 26 m 0.00

24 s

RA (J2000)

mJy/Beam

63

64

Dec (J2000)

−12 ◦ 03 ′ 46 ′′

63 and 64 - zoom

0.05 pc

48 ′′ 63

50 ′′

52 ′′

64

54 ′′

56 ′′

18 h 26 m 26 s RA (J2000)

0.40

0.35

0.30

0.25

0.20

0.15

0.10

0.05

0.00

mJy/Beam

Dec (J2000)

−12 ◦ 04 ′ 14 ′′

16 ′′

18 ′′

20 ′′

65

22 ′′

24 ′′

0.05 pc

65 - zoom

RA (J2000)

0.40

0.35

0.30

0.25

0.20

0.15

0.10

0.05

18 h 26 m 0.00

26 s

mJy/Beam

2


candidates 63, 64, and 65 located in the region IRAS 18236−1205. A

close-up view of the three radio sources is shown in the bottom and right

panels. The green contour levels of the K-band (1.3 cm) continuum emission

are 3, 5, and 7 times 30 µJy beam −1 . The magenta contours show

the Spitzer/GLIMPSE 4.5 µm emission. The blue- and redshifted outflow

lobes of SiO (2−1) are shown as solid blue and dashed red contours,

respectively (see Sánchez-Monge et al. 2013d). The pink and green ellipses

are the beam sizes of the C and K bands, respectively. The white

stars mark the location of the H 2 O masers (see Table 2.2).

65

G23.60+0.0M1 (sources 73 and 74)

The G23.60+0.01M1 star-forming region has been studied in the literature

by various authors (e.g., Rathborne et al. 2006; Battersby et al.

67



.......................................................................

2010; Ginsburg et al. 2013), who reported the presence of two massive

dense clumps with masses of 100 M ⊙ and 120 M ⊙ . The two candidate

radio jets (73 and 74) have positive spectral indices consistent with thermal

emission of radio jets. In particular, source 74 is located at the center

of the molecular outflow reported by Sánchez-Monge et al. (2013d) and

is associated with a strong 4.5 µm emission (see Fig. 2.13). The two

H 2 O masers detected in the region are slightly displaced from the radio

continuum source, but coincide with the extended 4.5 µm emission (see

Fig. 2.13-bottom). The association of molecular outflow emission, bright

and extended 4.5 µm emission, and close H 2 O maser features favor the

interpretation of this source as a good radio-jet candidate.

The second radio continuum source (source 73) is located relatively

close to the center of the outflow. However, no maser or EGOs are found

in connection with the source. Although we cannot reject this source as

a radio jet, we prefer source 74 as the main object driving the outflow

observed in the region.

2

IRAS 18316−0602 (sources 83 and 95)

We identified 13 radio continuum sources in the IRAS 18316−0602 region,

2 of which were classified as radio-jet candidates: sources 83 and

95. Source 83 has an almost flat spectral index (0.08 ± 0.08) and shows

a weak extension to the south, which is better resolved in the K-band

image. Source 95 is fainter, appears to be located about 3 arcsec to the

southeast of source 83, and is visible only in the K band (some faint

emission slightly above the noise level is visible in the C-band image,

see Fig. 2.14). As for the maser emission, we identified both a H 2 O and

a CH 3 OH maser feature associated with the brightest source 83. This

source has been studied in previous works (e.g., Roueff et al. 2006; Cutri

et al. 2012; Azatyan et al. 2016; Stecklum et al. 2017), in some of them,

it is called RAFGL 7009S. The source is detected in the near-infrared

together with two other objects separated by about 10 arcsec, and it is

surrounded by a diffuse and extended structure (see, e.g., Stecklum et al.

2017).

López-Sepulcre et al. (2010) and Sánchez-Monge et al. (2013d) reported

on the molecular outflow emission in the region. The blue and

red contours in Fig. 2.14 show the SiO (2–1) blueshifted and redshifted

outflow emission. The two radio continuum sources (83 and 95) are lo-

68


.......................................................................

cated close to the center of the outflow. Although this region was not

included in the Cyganowski et al. (2008, 2009, 2011) surveys, we identified

a bright 4.5 µm source in the Spitzer/GLIMPSE data. However, the

source is located at the edge of the area surveyed by Spitzer, which prevents

a detailed characterization of its infrared emission. The association

of source 83 with a 6.7 GHz CH 3 OH maser emission suggests that this

source harbors a massive YSO. Together with its central location within

the outflow and its association with a H 2 O maser, this indicates that this

source is the radio jet that drives the outflow. The almost flat spectral

index may indicate that this is a fully ionized radio jet. However, further

observations in different frequency bands may help to better determine

the spectral index and the status of radio continuum emission.

G24.08+0.0 M2 (source 96)

We identified 14 radio continuum sources in the region G24.08+0.0 M2,

one of which, source 96, is detected in both frequency bands and has a

negative spectral index (−0.84 ± +0.08, see Fig. 2.15). The outflow activity

in the region has been studied by López-Sepulcre et al. (2011) and

Sánchez-Monge et al. (2013d), who found molecular outflow in different

tracers. However, this outflow is not spatially related to any of the radio

continuum sources identified in this work. Moreover, this source could

be a background source.

2

G24.33+0.1 M1 (source 110)

In the region G24.33+0.1 M1 we identified a radio continuum source

(source 110) located in the center of our field of view and detected in

both frequency bands (see Fig. 2.16). This source has a positive spectral

index of +0.73±0.35, which is consistent with thermal emission. We also

detected maser features of both H 2 O and CH 3 OH associated with the

continuum source. In addition, other authors have reported OH masers

towards this object (see, e.g., Caswell & Green 2011). Rathborne et al.

(2007) studied this source in the millimeter regime and identified a singular

compact source with a rich chemistry characteristic of hot molecular

cores. Sánchez-Monge et al. (2013d) reported molecular outflow activity,

with source 110 at its geometric center. Overall, this source is one of the

best candidates for a thermal radio jet.

69



.......................................................................

G24.60+0.1M1 (source 119)

The only information we have for source 119 is that it is associated with

an EGO. The nearest studied object is a extended H 2 emission that

is 19 arcsec away from source 119 (Froebrich et al. 2015). We cannot

determine the spectral index because the source is outside the primary

beam of the K-band images, but we propose that source 119 is a radio

jet.

G24.60+0.1M2 (136)

2

Source 136 is detected only in the K band, resulting in a spectral index

limit (> +1.67) that is consistent with thermal emission. The source

is associated with H 2 O maser emission. This source, although not directly

associated with one EGO, is located in the vicinity of G24.63+0.15

reported by Cyganowski et al. (2008, see green circle in Fig. 2.17).

Rathborne et al. (2007) suggested that the main dense condensation,

hosting source 136, may contain several condensations, referred to as

G024.60+00.08 MM1 (A, B, and C). Our radio continuum source appears

to be related to component C which is an IRDC condensation.

Further observations of this object are necessary to confirm its possible

nature as a radio jet.

G34.43+0.2M3 (sources 137 and 139)

Our radio continuum observations toward the region G34.43+0.2 M3

have led to the discovery of six radio continuum sources, although most of

them are located far from the central region studied in Sánchez-Monge

et al. (2013d). The brightest source is source 137, which is about 13

arcmin from the phase center of our observations, that is, the source is

outside the primary beam responses of the VLA antennas on both bands.

This object is so bright that it is detected in both the C and K bands. At

6 cm, the source appears as a comet-like structure (see Fig. 2.18), which

resembles cometary HII regions. The 1.3 cm continuum emission also

shows an arc-shaped structure shifted to the east, probably tracing the

head of the cometary object. This source, referred to in the literature as

G34.26+0.15, has been studied by other authors who reported two hypercompact

HII regions (A and B) and one ultracompact HII region (C),

70


.......................................................................

all marked in the figure (see also Reid & Ho 1985; Gaume et al. 1994;

Sewiło et al. 2011). Various studies (e.g., Hatchell et al. 2001; Liu et al.

2013) have reported outflow activity in this region, but no information

on the outflow energetics such as the outflow momentum rate is reported.

Although this source is associated with an EGO (Cyganowski et al. 2008)

and with a molecular outflow, the bright emission together with previous

studies suggests that a major fraction of the radio continuum emission we

detected originates in a HII region rather than in a radio jet. Source 139

in the region is also classified as a radio-jet candidate in Table 2.3. However,

the only information we have for this object is its association with

an EGO (i.e., G34.41+0.24, Cyganowski et al. 2008). Shepherd et al.

(2004) suggested that the embedded object (G34.4 MM) appears to be a

massive B2 protostar at an early stage of evolution. This region is also

associated with the H 2 O maser activity (Cyganowski et al. 2013), which

may favor a radio-jet origin for the detected radio continuum emission.

Further observations are needed to better constrain its properties.

IRAS 19095+0930 (sources 143 and 144)

We identified four radio continuum sources in the region IRAS 19095+0930,

also known in the literature as G43.80−0.13. Two of these radio sources,

sources 143 and 144, are located close to each other and in the center of

a molecular outflow (Sánchez-Monge et al. 2013d, see also Figure 2.19).

Source 143 has a brighter flux, is also clearly visible in the K-band image,

and is associated with H 2 O maser emission features.

This region has been studied in the past by different authors at different

wavelengths (Kurtz et al. 1994; Lekht 2000; De Buizer et al. 2005).

De Buizer et al. (2005) reported a kidney-bean shape structure at midinfrared

wavelengths that matches the radio continuum sources reported

by Kurtz et al. (1994) that were referred to as a HII region. The object

is also associated with OH masers. We did not find this source in the

EGO catalogs (Cyganowski et al. 2008, 2009), but we identified a 4.5 µm

source associated with source 143. No 4.5 µm infrared source appears to

be associated with the eastern source 144.

We derive a spectral index of +1.12 ± 0.04 for source 143, which

is consistent with thermal emission. Together with its location at the

geometrical center of the outflow and its association with masers, this

may suggest that source 143 is a good radio-jet candidate. However, the

2

71



.......................................................................

high radio continuum flux of this source does not appear to be consistent

with the typical properties of other radio jets (see Figs. 2.3 and 2.5). This

might mean that this source is in a transition phase from a radio jet to

a HII region. However, this requires further investigation.

2.10 Catalog of the continuum sources

2

In the following tables and figures, we provide information on the properties

of the radio continuum sources detected in the VLA observations

presented in this work. In Table 2.5 we list the coordinates of the 146

radio continuum sources together with their flux density, intensity peak,

and deconvolved size at 6 cm (C band) and 1.3 cm (K band). The fluxes

and intensities are corrected for the primary beam response of the VLA

antennas. For sources outside the C-band primary beam (listed as ‘oC’

in Column (11) of Table 2.5) the primary beam correction is not reliable

and the flux has to be taken with caution. Similarly, for sources located

outside the K-band primary beam (labeled ‘oK’ in the table), the K-

band flux has to be taken with caution. The last column of the table

lists the spectral index, α. For sources with no reliable flux estimate at

one of the bands, we did not determine the spectral index. For sources

detected at both frequency bands (C and K bands), the spectral index

was determined using the fluxes determined after creating images with

a common uv range (see Sect. 2.4.2 for more details). In Table 2.6 we

list the observed and deconvolved source sizes of all the detected sources.

The source sizes are determined as √ θ major × θ minor , where θ major and

θ minor are listed in Table 2.6. We transformed the angular size of each

source into astronomical unit (au) using the distances listed in Table 2.1.

We give the source size in Table 2.5. Finally, in Table 2.7, we list the

intensities, flux densities, and sizes determined from the images generated

using a common uv range at the C and K bands. In Figs. 2.20 to

2.35, we present close-up views of the C- and K-band emission for the

146 detected continuum sources.

72


.......................................................................

Dec (J2000)

−8 ◦ 18 ′ 30.0 ′′

45.0 ′′

19 ′ 00.0 ′′

15.0 ′′

30.0 ′′

14 s G23.60+0.0M1

0.05 pc

0.14

0.12

0.10

73 74 0.08

0.06

0.04

0.02

13 s 12 s 118 s h 34 m 10 s 0.00

RA (J2000)

mJy/Beam

Dec (J2000)

−8 ◦ 18 ′ 55.0 ′′

19 ′ 00.0 ′′

05.0 ′′

10.0 ′′

15.0 ′′

0.05 pc

G23.60+0.0M1

73 74

18 h 34 m 12 s

RA (J2000)

0.14

0.12

0.10

0.08

0.06

0.04

0.02

0.00

mJy/Beam

2


candidates 73 and 74 located in the region G23.60+0.0M1. A closeup

view of the two radio sources is shown in the bottom panel. The

green contour levels of the K-band (1.3 cm) continuum emission are 3,

5, 7, 9, and 11 times 20 µJy beam −1 . The magenta contours show the

Spitzer/GLIMPSE 4.5 µm emission. The blue- and redshifted outflow

lobes of SiO (2−1) are shown as solid blue and dashed red contours, respectively

(see Sánchez-Monge et al. 2013d). The pink and green ellipses

are the beam sizes of the C and K bands, respectively. The white stars

mark the location of the H 2 O masers (see Table 2.2).

73



.......................................................................

Dec (J2000)

−5 ◦ 59 ′ 15.0 ′′

30.0 ′′

45.0 ′′

−6 ◦ 00 ′ 00.0 ′′

IRAS 18316-0602

2.5

2.0

83 95

1.5

1.0

0.5

15.0 ′′

23 s 22 s 21 s 2018 s h 34 m 0.0

19 s

RA (J2000)

mJy/Beam

83 - zoom

2.5

2

Dec (J2000)

−5 ◦ 59 ′ 40.0 ′′

42.0 ′′

44.0 ′′

83

95

2.0

1.5

1.0

0.5

mJy/Beam

46.0 ′′

18 h 34 m 21 s

RA (J2000)

0.0


candidates 83 and 95 located in the region IRAS 18316−1602. A closeup

view of the two radio sources is shown in the bottom panel. The

green contour levels of the K-band (1.3 cm) continuum emission are 3,

5, 9, and 11 times 27 µJy beam −1 . The magenta contours show the

Spitzer/GLIMPSE 4.5 µm emission (note that half of the region was not

covered in the mapped area). The blue- and redshifted outflow lobes of

SiO (2−1) are shown as solid blue and dashed red contours, respectively

(see Sánchez-Monge et al. 2013d). The pink and green ellipses are the

beam sizes of the C and K bands, respectively. The white and orange

stars mark the location of the H 2 O and CH 3 OH masers, respectively (see

Table 2.2).

74


.......................................................................

Dec (J2000)

−7 ◦ 46 ′ 36.0 ′′

38.0 ′′

40.0 ′′

G24.08+0.0M2

96

42.0 ′′

44.0 ′′

46.0 ′′

18 h 34 m 49 s RA (J2000)

0.005

0.004

0.003

0.002

0.001

0.000

Jy/beam

2

Figure 2.15: VLA C-band (6 cm) continuum emission map of radio

source 96 located in the region G24.08+0.0M2. The green contour levels

of the K-band (1.3 cm) continuum emission are 3, 5, and 9 times

20 µJy beam −1 . The pink and green ellipses are the beam sizes of the C

and K bands, respectively.

75



.......................................................................

Dec (J2000)

−7 ◦ 34 ′ 50.0 ′′

35 ′ 00.0 ′′

10.0 ′′

20.0 ′′

0.005 pc

G24.33+0.1M1

08 s

09 s RA (J2000)

0.40

0.35

0.30

0.25

0.20

0.15

0.10

0.05

18 h 35 m 0.00

07 s

mJy/Beam

2

Dec (J2000)

−7 ◦ 35 ′ 00.0 ′′

02.0 ′′

04.0 ′′

06.0 ′′

08.0 ′′

10.0 ′′

0.005 pc

110 - zoom

18 h 35 m 08 s

RA (J2000)

0.40

0.35

0.30

0.25

0.20

0.15

0.10

0.05

0.00

mJy/Beam


candidate 110 located in the region G24.33+0.1 M1. A close-up view

of the radio source is shown in the bottom panel. The green contour

levels of the K-band (1.3 cm) continuum emission are 3, 5, and 9 times

7 µJy beam −1 . The blue- and redshifted outflow lobes of SiO (2−1) are

shown as solid blue and dashed red contours, respectively (see Sánchez-

Monge et al. 2013d). The pink and green ellipses are the beam sizes of

the C and K bands, respectively. The white star marks the location of

the H 2 O (see Table 2.2).

76


.......................................................................

G24.60+0.1M2

−7 ◦ 18 ′ 34 ′′

0.25

Dec (J2000)

36 ′′

38 ′′

136

0.20

0.15

0.10

mJy/Beam

40 ′′

18 h 35 m 40 s

RA (J2000)

0.05

0.00

2

Figure 2.17: VLA K-band (1.3 cm) continuum emission map of radiojet

candidate 136 located in the region G24.60+0.1 M2. The green circle

with a radius of ∼4 arcsec marks EGO G24.63+0.15, reported by

Cyganowski et al. (2008). The pink ellipse is the beam size of the K band.

The white star marks the location of the H 2 O (see Table 2.2).

77



.......................................................................

A

B

C

2


candidate 137 located in the region G34.43+0.2 M3. The green contour

levels of the K-band (1.3 cm) continuum emission are 3, 5, and 7 times

20 µJy beam −1 . The magenta contours show the Spitzer/GLIMPSE

4.5 µm emission. The pink and green ellipses are the beam sizes of the

C and K bands, respectively. The white circles (A, B, and C) mark the

position of the 2 mm continuum sources reported by Gasiprong et al.

(2002).

78


.......................................................................

Dec (J2000)

15.0 ′′

36 ′ 00.0 ′′

45.0 ′′

30.0 ′′

+9 ◦ 35 ′ 15.0 ′′

0.05 pc

IRAS 19095+0930

30

25

20

15

10

56 s 5

55 s 54 s 5319 s h 11 m 0

52 s

RA (J2000)

mJy/Beam

Dec (J2000)

54.0 ′′

52.0 ′′

50.0 ′′

48.0 ′′

46.0 ′′

+9 ◦ 35 ′ 44.0 ′′

143 and 144 - zoom

0.05 pc

144

143

19 h 11 m 54 s

RA (J2000)

30

25

20

15

10

5

0

mJy/Beam

2

Figure 2.19: VLA C-band (6 cm) continuum emission map of radio

sources 143 and 144 located in the region IRAS 19095+0930. The green

contour levels of the K-band (1.3 cm) continuum emission are 3, 5,

and 9 times 2 mJy beam −1 . The blue- and redshifted outflow lobes

of SiO (2−1) are shown as solid blue and dashed red contours, respectively

(see Sánchez-Monge et al. 2013d). The magenta contours show the

Spitzer/GLIMPSE 4.5 µm emission. The pink and green ellipses are the

beam sizes of the C and K bands, respectively. The white and orange

stars mark the of the H 2 O and CH 3 OH masers (see Table 2.2).

79

2

80

Table 2.5: Catalogue of radio continuum sources detected with the VLA. Coordinates (RA and

Dec) are given in J2000 epoch.

Coordinates C band a K band b

RA Dec Intensity Flux Size Intensity Flux Size Spectral

ID Region (h:m:s) ( ◦ : ′ : ′′ ) (mJy/beam) (mJy) (au) (mJy/beam) (mJy) (au) index c

1 05358+3543 05:39:25.83 +35:45:53.86 37.7±0.3 38.1±0.3 <2200 — — — iC/oK —

2 05:39:13.06 +35:45:51.12 0.387±0.010 0.534±0.014 1300 — — — iC/iK —

3 05:39:10.05 +35:46:07.73 0.105±0.003 0.075±0.002 <1800 — — — iC/iK —

4 05:39:12.83 +35:45:50.85 0.147±0.006 0.332±0.015 2200 — — — iC/iK —

5 05:39:09.93 +35:45:17.19 0.267±0.008 0.403±0.013 1500 — — — iC/iK —

6 05:39:33.53 +35:48:03.74 0.282±0.006 0.345±0.007 1000 — — — iC/oK —

7 05:39:14.36 +35:44:29.63 0.191±0.006 0.266±0.009 1300 — — — iC/iK —

8 05:39:12.12 +35:44:02.15 0.238±0.006 0.262±0.007 500 — — — iC/iK —

9 05:38:54.93 +35:44:41.72 0.521±0.004 0.571±0.005 600 — — — iC/oK —

10 05:38:39.15 +35:49:43.53 0.886±0.060 1.640±0.110 2000 — — — iC/oK —

11 05:39:37.24 +35:46:45.15 0.265±0.014 0.197±0.010 <1900 — — — iC/oK —

12 G189.78+0.34 06:08:35.11 +20:39:12.57 0.429±0.016 0.722±0.027 1700 — — — iC/iK —

13 06:08:35.27 +20:39:09.01 0.874±0.032 1.241±0.046 1300 — — — iC/iK —

14 06:08:35.30 +20:39:06.77 0.372±0.024 0.687±0.045 1900 — — — iC/iK —

15 06:08:35.38 +20:39:04.42 0.313±0.013 0.938±0.040 2800 — — — iC/iK —

16 06:08:35.44 +20:39:03.11 0.503±0.021 1.040±0.044 2000 — — — iC/iK —

17 06:08:43.60 +20:41:08.59 0.183±0.008 0.119±0.005 <1700 — — — iC/oK —

18 06:08:45.19 +20:38:16.64 0.114±0.008 0.164±0.011 1300 — — — iC/oK —

19 06:08:36.59 +20:43:14.36 0.185±0.015 0.300±0.024 800 — — — iC/oK —

20 06:08:44.25 +20:33:06.46 0.595±0.038 0.569±0.036 <2000 — — — iC/oK —

21 G192.58−0.04 06:12:53.60 +18:00:26.36 12.03±0.46 18.21±0.70 2400 — — — iC/iK —

22 06:12:54.01 +17:59:23.20 3.998±0.087 10.27±0.22 4000 — — — iC/iK —

23 06:12:53.84 +17:59:22.16 1.540±0.047 1.558±0.047 <3300 — — — iC/iK —

24 06:12:54.29 +17:59:33.80 0.637±0.033 0.494±0.025 <2900 — — — iC/iK —

25 06:12:54.33 +17:59:24.27 0.303±0.011 0.470±0.049 5800 — — — iC/iK —

26 06:12:49.91 +18:03:05.36 0.313±0.009 0.408±0.012 1500 — — — iC/oK —

27 G18.18−0.30 18:25:01.06 −13:15:39.515 4.240±0.262 21.2±1.37 6600 . . . . . . . . . iC/oK . . .

28 18:25:07.46 −13:17:58.83 2.412±0.089 3.863±0.144 2200 . . . . . . . . . iC/oK . . .

29 18:25:06.97 −13:18:10.63 1.745±0.034 3.390±0.069 3000 . . . . . . . . . iC/oK . . .

30 18:25:01.18 −13:15:45.88 1.303±0.020 1.989±0.032 2400 . . . . . . . . . iC/oK . . .

31 18:25:00.95 −13:15:35.72 1.298±0.039 6.719±0.204 7000 . . . . . . . . . iC/oK . . .

32 18:24:44.85 −13:14:45.80 2.647±0.036 4.447±0.064 2400 . . . . . . . . . iC/oK . . .

33 18:24:53.82 −13:12:52.49 0.713±0.007 0.670±0.007 <3400 . . . . . . . . . iC/oK . . .

34 18:25:06.70 −13:12:28.02 0.307±0.008 0.331±0.009 <3600 . . . . . . . . . iC/oK . . .

35 18:25:42.35 −13:10:21.21 0.361±0.023 1.177±0.075 4500 . . . . . . . . . oC . . .

36 18:25:18.54 −13:15:33.06 0.182±0.006 0.259±0.009 2100 . . . . . . . . . iC/oK . . .

37 18:24:55.78 −13:10:16.39 0.404±0.023 1.782±0.101 6400 . . . . . . . . . iC/oK . . .

38 18:24:55.94 −13:10:17.44 0.291±0.012 1.226±0.053 6100 . . . . . . . . . iC/oK . . .

39 18223−1243 18:25:31.47 −12:41:24.03 54.2±1.1 105.6±2.1 4500 . . . . . . . . . iC/oK . . .

40 18:25:04.14 −12:37:44.91 48.6±1.2 75.3±1.9 3800 . . . . . . . . . iC/oK . . .

41 18:25:26.73 −12:40:33.23 1.694±0.063 2.477±0.094 3600 . . . . . . . . . iC/oK . . .

42 18:24:52.60 −12:39:19.80 2.989±0.054 4.158±0.075 2900 . . . . . . . . . iC/oK . . .

43 18:24:36.30 −12:51:03.31 1.762±0.091 6.550±0.340 8800 . . . . . . . . . oC . . .

44 18:24:34.61 −12:52:03.91 0.374±0.016 1.583±0.069 9600 . . . . . . . . . oC . . .

45 18:26:04.96 −12:34:35.17 0.250±0.009 0.577±0.023 6000 . . . . . . . . . oC . . .

46 18:26:05.24 −12:34:35.62 0.289±0.039 0.579±0.039 <8200 . . . . . . . . . oC . . .

47 18:24:34.69 −12:50:59.45 0.771±0.083 1.542±0.083 <16800 . . . . . . . . . oC . . .

48a 18228−1312 18:25:41.92 −13:10:17.63 21.20±0.45 219.9±5.1 13500 6.9±0.8 13.7±2.3 <7500 iC/iK −2.11±0.13 †

Table 2.5: continued.


81



48b 18:25:42.05 −13:10:16.54 19.37±0.72 383.0±9.9 19100 6.2±1.2 14.3±3.7 5400 iC/iK −2.51±0.20 †

48c 18:25:42.40 −13:10:22.14 38.50±2.70 297.0±9.9 11200 25.0±1.8 64.5±6.4 6500 iC/iK −1.17±0.10 †

49 18:26:26.27 −13:05:48.53 0.468±0.010 0.445±0.009 <4300 . . . . . . . . . oC . . .

50 18:25:00.73 −13:15:35.80 0.312±0.019 13.17±0.80 28300 . . . . . . . . . oC . . .

51 G19.27+0.1M1 18:27:37.96 −11:56:33.89 0.348±0.023 2.140±0.140 8200 . . . . . . . . . oC . . .

52 18:26:20.97 −12:05:33.12 0.797±0.033 0.878±0.037 1100 . . . . . . . . . iC/oK . . .

53 18:26:03.67 −12:04:37.08 1.130±0.019 1.198±0.020 800 . . . . . . . . . iC/iK < −0.57

54 18:25:45.52 −12:05:18.59 0.265±0.005 0.326±0.007 1400 . . . . . . . . . iC/oK . . .

55 18:25:54.27 −11:52:19.02 0.186±0.012 1.138±0.071 3600 . . . . . . . . . oC . . .

56 18:26:04.22 −11:52:32.05 0.299±0.041 7.300±1.000 17600 . . . . . . . . . oC . . .

57 18:26:48.63 −12:26:17.05 0.199±0.009 0.599±0.030 5000 . . . . . . . . . oC . . .

58 18:26:34.30 −11:57:59.91 0.284±0.008 0.518±0.016 3100 . . . . . . . . . oC . . .

59 18:26:13.78 −12:01:19.59 0.258±0.007 0.249±0.006 <3500 . . . . . . . . . iC/oK . . .

60 18:27:37.16 −11:56:26.27 0.168±0.008 0.409±0.021 2200 . . . . . . . . . oC . . .

61 18:26:05.57 −12:04:33.53 0.123±0.003 0.139±0.003 4300 . . . . . . . . . iC/iK < +2.07

62 18:25:51.96 −12:05:13.44 . . . . . . . . . 0.2±0.01 0.2±0.1 <1100 iC/iK > +1.23

63 18236−1205 18:26:25.06 −12:03:48.92 0.070±0.004 0.099±0.006 <1100 0.28±0.01 0.14±0.01 2800 iC/iK +0.26±0.21 †

64 18:26:25.78 −12:03:53.20 0.331±0.013 0.383±0.015 <4300 0.2±0.01 0.28±0.01 930 iC/iK −0.24±0.12 †

65 18:26:26.38 −12:04:19.78 0.558±0.004 0.570±0.004 <4000 1.7±0.01 1.9±0.01 310 iC/iK +0.94±0.02 †

66 18:26:35.47 −12:01:13.80 0.125±0.002 0.113±0.002 <3800 . . . . . . . . . iC/oK . . .

67 18:26:22.55 −12:05:58.88 0.224±0.035 0.708±0.060 <7400 . . . . . . . . . iC/oK . . .

68 18:26:22.30 −12:05:58.06 0.135±0.005 0.270±0.020 <6000 . . . . . . . . . iC/oK . . .

69 18:26:22.05 −12:07:28.99 0.092±0.003 0.449±0.034 <9700 . . . . . . . . . iC/oK . . .

70 18:26:21.84 −12:07:31.84 0.070±0.001 0.143±0.019 <4300 . . . . . . . . . iC/oK . . .

71 18:26:21:65 −12:07:35.40 0.198±0.001 0.394±0.031 <6500 . . . . . . . . . iC/oK . . .

72 G23.60+0.0M1 18:33:53.47 −08:07:12.19 0.817±0.110 9.900±1.300 11900 . . . . . . . . . oC . . .

73 18:34:12.33 −08:19:01.19 0.191±0.003 0.190±0.003 <3600 0.35±0.04 0.48±0.05 <1200 iC/iK +0.70±0.26 †

74 18:34:11.57 −08:19:06.42 0.310±0.004 0.328±0.004 <3700 0.11±0.04 1.04±0.43 1100 iC/iK +0.87±0.94 †

75 18:34:21.06 −08:18:12.35 0.420±0.006 0.452±0.006 800 . . . . . . . . . iC/oK . . .

76 18:34:33.02 −08:15:26.75 11.3±0.2 12.813±0.219 1200 . . . . . . . . . iC/oK . . .

77 18:34:44.83 −08:31:05.33 0.210±0.003 0.272±0.003 1700 . . . . . . . . . oC . . .

78 18:33:44.01 −08:21:22.95 1.234±0.038 2.500±0.077 3500 . . . . . . . . . iC/oK . . .

79 18:33:47.80 −08:23:34.27 0.763±0.049 1.718±0.109 3900 . . . . . . . . . iC/oK . . .

80 18:34:06.12 −08:24:38.77 0.276±0.012 0.336±0.015 <3900 . . . . . . . . . iC/oK . . .

81 18:34:17.74 −08:21:08.23 0.096±0.004 0.119±0.005 <4000 . . . . . . . . . iC/oK . . .

82 18:34:14.29 −08:24:10.30 0.351±0.010 0.390±0.011 <3800 . . . . . . . . . iC/oK . . .

83 18316−0602 18:34:20.90 −05:59:41.96 2.275±0.081 3.345±0.120 2500 0.61±0.06 2.62±0.26 2000 iC/iK −0.19±0.24 †

84 18:34:33.93 −06:02:21.99 1.618±0.025 1.863±0.029 1200 . . . . . . . . . iC/oK . . .

85 18:34:27.63 −06:05:09.13 2.103±0.026 2.493±0.030 1300 . . . . . . . . . iC/oK . . .

86 18:34:14.92 −06:00:23.58 0.313±0.005 0.354±0.006 <4400 . . . . . . . . . iC/iK < +1.43

87 18:34:26.74 −05:57:21.18 0.907±0.016 0.998±0.019 <4400 . . . . . . . . . iC/oK . . .

88 18:34:16.53 −05:45:48.50 0.109±0.006 0.742±0.045 9900 . . . . . . . . . oC . . .

89 18:32:42.03 −06:10:19.83 0.146±0.011 0.301±0.022 3700 . . . . . . . . . oC . . .

90 18:34:08.90 −05:52:55.20 1.454±0.064 1.658±0.073 <4500 . . . . . . . . . iC/oK . . .

91 18:34:32.32 −06:00:15.07 0.130±0.007 0.147±0.008 <4400 . . . . . . . . . iC/oK . . .

92 18:34:13.82 −05:53:01.18 0.405±0.013 0.971±0.031 3400 . . . . . . . . . iC/oK . . .

93 18:35:10.90 −06:02:32.53 0.067±0.001 0.136±0.008 4000 . . . . . . . . . oC . . .

94 18:34:17.69 −06:05:05.39 . . . . . . . . . 0.22±0.01 0.31±0.01 2100 iC/oK . . .

2

2

82





95 18:34:20.81 −05:59:42.99 . . . . . . . . . 0.23±0.04 0.25±0.07 <1300 iC/iK > +1.36

96 G24.08+0.0M2 18:34:48.71 −07:46:41.51 16.9±0.3 21.7±0.4 1600 4.10±0.17 5.3±0.2 <1100 iC/iK −1.08±0.10 †

97 18:34:41.39 −07:43:55.45 7.986±0.172 10.67±0.24 1600 . . . . . . . . . iC/oK . . .

98 18:34:41.45 −07:43:47.69 1.411±0.017 2.558±0.032 3000 . . . . . . . . . iC/oK . . .

99 18:34:57.18 −07:43:26.18 3.122±0.052 4.364±0.073 2000 . . . . . . . . . iC/oK . . .

100 18:34:59.59 −07:43:00.52 3.762±0.104 7.750±0.217 3500 . . . . . . . . . iC/oK . . .

101 18:34:59.60 −07:42:57.06 1.208±0.021 2.517±0.089 2800 . . . . . . . . . iC/oK . . .

102 18:34:57.13 −07:45:22.48 1.141±0.014 1.454±0.017 1700 . . . . . . . . . iC/iK < −0.06

103 18:34:12.13 −07:52:54.06 0.136±0.004 0.202±0.005 2400 . . . . . . . . . oC . . .

104 18:34:11.27 −07:53:07.87 0.088±0.004 0.146±0.006 2800 . . . . . . . . . oC . . .

105 18:33:59.50 −07:52:36.63 0.304±0.005 0.608±0.056 <7600 . . . . . . . . . oC . . .

106 18:34:25.40 −07:54:46.13 0.239±0.074 59.0±18.00 55700 . . . . . . . . . oC . . .

107 18:34:51.28 −07:42:14.42 0.094±0.005 0.195±0.033 <2900 . . . . . . . . . iC/oK . . .

108 18:35:23.92 −07:37:38.20 1.510±0.150 2.080±0.020 <3500 . . . . . . . . . oC . . .

109 18:34:52.96 −07:47:03.05 0.041±0.003 0.082±0.032 <2300 . . . . . . . . . iC/iK < +2.33

110 G24.33+0.1M1 18:35:08.13 −07:35:04.17 0.438±0.007 0.434±0.007 <5400 0.77±0.03 0.90±0.03 <1600 iC/iK +0.55±0.09 †

111 18:35:13.56 −07:38:20.37 0.382±0.014 0.659±0.024 2700 . . . . . . . . . iC/oK . . .

112 18:35:33.97 −07:37:34.49 1.630±0.078 4.685±0.228 7400 . . . . . . . . . iC/oK . . .

113 18:35:10.88 −07:34:22.08 0.198±0.007 0.278±0.011 3200 . . . . . . . . . iC/iK < +0.07

114 18:35:02.82 −07:31:20.72 0.290±0.005 0.328±0.005 2000 . . . . . . . . . iC/oK . . .

115 18:34:48.65 −07:46:40.36 0.151±0.006 0.225±0.009 3700 . . . . . . . . . oC . . .

116 18:35:56.00 −07:27:23.27 0.196±0.009 0.289±0.015 2400 . . . . . . . . . oC . . .

117 G24.60+0.1M1 18:36:12.51 −07:12:09.92 0.437±0.015 0.526±0.018 1300 . . . . . . . . . oC . . .

118 18:36:12.60 −07:12:14.14 0.420±0.022 0.983±0.052 5700 . . . . . . . . . oC . . .

119 18:35:40.67 −07:22:05.79 0.305±0.016 1.107±0.059 8400 . . . . . . . . . iC/oK . . .

120 18:35:40.73 −07:22:00.86 0.284±0.009 0.457±0.014 3000 . . . . . . . . . iC/oK . . .

121 18:35:40.86 −07:21:57.88 0.286±0.020 1.705±0.126 10900 . . . . . . . . . iC/oK . . .

122 18:36:05.58 −07:31:21.64 0.349±0.036 3.790±0.390 16200 . . . . . . . . . oC . . .

123 18:35:53.12 −07:14:20.49 1.058±0.011 1.164±0.012 <5500 . . . . . . . . . iC/oK . . .

124 18:35:16.80 −07:05:08.35 0.239±0.011 0.400±0.019 3800 . . . . . . . . . oC . . .

125 18:35:41.07 −07:16:41.06 0.140±0.004 0.123±0.004 <4900 . . . . . . . . . iC/iK . . .

126 18:35:43.76 −07:19:26.18 0.114±0.003 0.105±0.002 <5000 . . . . . . . . . iC/iK < +0.96

127 18:35:03.75 −07:26:00.91 0.146±0.012 0.529±0.044 8400 . . . . . . . . . oC . . .

128 18:36:18.11 −07:08:50.10 0.128±0.003 0.705±0.018 11000 . . . . . . . . . oC . . .

129 18:35:40.40 −07:19:28.74 0.088±0.005 0.157±0.009 4500 . . . . . . . . . iC/iK < +0.18

130 18:34:59.85 −07:26:39.36 0.078±0.002 0.110±0.003 3000 . . . . . . . . . oC . . .

131 18:34:39.43 −07:02:39.15 0.068±0.002 0.081±0.003 <5700 . . . . . . . . . oC . . .

132 G24.60+0.1M2 18:35:47.19 −07:12:59.44 0.300±0.008 0.543±0.015 3800 . . . . . . . . . iC/oK . . .

133 18:35:40.95 −07:21:56.17 0.275±0.010 0.992±0.036 7600 . . . . . . . . . iC/oK . . .

134 18:35:56.07 −07:27:23.87 0.133±0.015 0.243±0.028 <6500 . . . . . . . . . oC . . .

135 18:35:29.95 −07:27:46.84 0.064±0.003 0.077±0.003 1900 . . . . . . . . . oC . . .

136 18:35:40.12 −07:18:37.39 . . . . . . . . . 0.82±0.11 0.85±0.12 <2100 iC/iK > +1.67

137 G34.43+0.2M3 18:53:18.84 +01:14:59.32 4.070±0.280 18.40±1.30 7000 0.24±0.02 1.27±0.12 2000 oC . . .

138 18:53:18.67 +01:24:47.73 6.046±0.080 7.393±0.097 1600 . . . . . . . . . iC/oK . . .

139 18:53:18.02 +01:25:25.60 0.439±0.017 0.729±0.029 2100 . . . . . . . . . iC/oK . . .

140 18:53:08.32 +01:29:33.99 0.858±0.013 0.820±0.012 3000 . . . . . . . . . iC/oK . . .

141 18:53:35.99 +01:35:18.77 2.290±0.147 10.43±0.672 <5600 . . . . . . . . . iC/oK . . .

142 18:54:14.01 +01:19:18.42 0.089±0.003 0.148±0.005 7100 . . . . . . . . . oC . . .

143 19095+0930 19:11:53.99 +09:35:50.40 30.6±0.9 40.4±1.1 1900 90.5±2.1 126.3±3.0 660 iC/iK +0.87±0.08 †

144 19:11:54.36 +09:35:49.25 4.211±0.240 15.56±0.89 7300 . . . . . . . . . iC/iK < −2.97





145 19:12:00.54 +09:36:24.23 2.847±0.023 2.838±0.023 <4400 . . . . . . . . . iC/iK < +0.32

146 19:11:46.43 +09:37:03.67 1.927±0.025 2.048±0.027 <4500 . . . . . . . . . iC/oK . . .

Notes.

(a) Flux density, intensity peak and deconvolved source size for the sources detected at 6 cm in the C band images.

The fluxes are corrected by the primary beam response of the antennas, except for sources located outside the primary beam

(listed as ‘oC’). Source sizes are calculated as indicated in Appendix 2.10 and based on the values reported in Table 2.6. Upper

limits in the source size indicate that the source could not be deconvolved (see more details in Table 2.6).

(b) Flux density,

intensity peak and deconvolved source size for the sources detected at 1.3 cm in the K band images. The fluxes are corrected

by the primary beam response of the antennas, except for sources located outside the primary beam (listed either as ‘iC/oK’ or

‘oC’, see Sect. 2.4.1 for more details about this classification). (c) Spectral index determined from the fluxes at 6 cm (C band)

and 1.3 cm (K band). For the sources detected in one band, we use a 5σ upper limit for the non-detected band flux. Only for

sources located with the primary beam of both images (sources listed as ‘iC/iK’) we can derive reliable fluxes and therefore

spectral indices. Sources marked with † have been re-imaged using the common uv-range (see Table 2.7). More accurate

spectral indices, derived using these new images, are listed in Table 2.3.

83

2

2

84

Table 2.6: Observed and deconvolved angular sizes for the radio continuum

sources (see Table 2.5).

Observed source size

Deconvolved diameter

Major axis Minor axis PA Major axis Minor axis PA

ID Region ( ′′ ) ( ′′ ) ( ◦ ) ( ′′ ) ( ′′ ) ( ◦ )

Observed and deconvolved source sizes for C band detections

1 IRAS 05358+3543 1.269±0.009 1.239±0.009 53.0±3.8 . . . . . . . . .

2 1.532±0.034 1.402±0.034 106.7±3.4 0.894±0.063 0.633±0.085 114.9±9.9

3 1.113±0.044 1.001±0.043 155.4±5.2 . . . . . . . . .

4 2.411±0.058 1.455±0.057 120.4±1.1 2.069±0.069 0.730±0.124 121.1±1.2

5 1.667±0.041 1.409±0.041 104.5±2.0 1.105±0.065 0.654±0.098 108.5±3.8

6 1.412±0.028 1.349±0.027 0.1±7.5 0.690±0.066 0.472±0.094 169.0±9.0

7 1.520±0.046 1.424±0.044 145.4±6.4 0.899±0.082 0.644±0.106 147.1±6.4

8 1.330±0.035 1.286±0.034 168.0±9.0 0.512±0.758 0.215±1.011 160.0±9.0

9 1.357±0.012 1.255±0.012 76.4±1.4 0.490±0.035 0.249±0.076 85.0±9.0

10 1.918±0.083 1.502±0.086 65.3±3.1 1.440±0.110 0.870±0.160 65.6±3.8

11 1.089±0.070 1.065±0.068 135.0±9.0 . . . . . . . . .

12 G189.78+0.34 1.673±0.041 1.397±0.048 12.0±2.2 1.095±0.070 0.860±0.089 0.5±9.6

13 1.578±0.041 1.249±0.048 14.6±1.8 0.934±0.074 0.602±0.115 6.7±9.5

14 1.781±0.072 1.441±0.083 10.1±3.9 1.250±0.110 0.930±0.150 1.7±9.4

15 2.461±0.073 1.688±0.050 158.3±2.0 2.161±0.088 1.194±0.079 154.2±2.7

16 1.920±0.050 1.493±0.050 156.9±1.8 1.525±0.068 0.880±0.096 148.3±3.9

17 1.014±0.052 0.892±0.061 18.6±6.4 . . . . . . . . .

18 1.598±0.078 1.254±0.092 21.7±4.9 0.958±0.136 0.629±0.226 20.0±9.0

19 1.959±0.091 1.148±0.100 174.7±2.0 1.524±0.091 0.136±0.100 169.9±2.0

20 1.382±0.071 0.961±0.083 16.5±2.4 . . . . . . . . .

21 G192.58−0.04 1.741±0.046 1.451±0.054 25.8±2.2 1.033±0.082 0.827±0.106 32.0±9.0

22 2.524±0.064 1.700±0.029 59.1±1.5 2.152±0.078 1.116±0.052 63.7±2.1



.......................................................................

85





ID Region ( ′′ ) ( ′′ ) ( ◦ ) ( ′′ ) ( ′′ ) ( ◦ )

23 1.454±0.036 1.162±0.043 30.3±1.8 . . . . . . . . .

24 1.206±0.066 1.073±0.072 170.4±6.4 . . . . . . . . .

25 2.903±0.050 2.375±0.038 148.1±2.4 2.614±0.060 1.961±0.052 143.0±3.1

26 1.745±0.050 1.247±0.042 28.1±1.9 1.043±0.064 0.360±0.221 32.0±4.4

27 G18.18−0.30 3.480±0.102 2.620±0.077 47.0±1.4 3.270±0.110 2.030±0.110 53.9±2.4

28 1.884±0.040 1.550±0.064 174.8±2.4 1.180±0.098 0.657±0.172 61.0±9.0

29 2.210±0.025 1.604±0.033 11.5±0.8 1.629±0.042 0.820±0.085 37.9±3.1

30 1.962±0.017 1.419±0.028 165.8±0.8 0.971±0.050 0.890±0.053 50.0±9.0

31 3.244±0.057 2.909±0.052 151.8±4.5 2.803±0.090 2.645±0.091 117.0±9.0

32 1.926±0.016 1.591±0.024 179.5±0.7 1.267±0.034 0.711±0.058 57.4±3.7

33 1.637±0.011 1.047±0.019 166.6±0.3 . . . . . . . . .

34 1.714±0.030 1.148±0.049 161.5±1.0 . . . . . . . . .

35 2.876±0.108 2.066±0.072 56.7±1.5 2.660±0.120 1.160±0.140 62.3±2.0

36 2.114±0.037 1.225±0.061 157.9±1.0 1.224±0.073 0.584±0.164 147.5±6.1

37 3.407±0.061 2.362±0.099 156.2±1.4 2.938±0.074 2.105±0.117 152.7±3.1

38 3.376±0.103 2.275±0.068 132.8±1.9 3.000±0.130 1.850±0.110 123.8±3.9

39 IRAS 18223−1243 2.267±0.027 1.865±0.034 18.1±0.8 1.715±0.043 0.882±0.088 54.0±3.4

40 2.167±0.030 1.550±0.048 172.3±0.7 1.163±0.073 0.920±0.096 33.0±9.0

41 2.230±0.044 1.423±0.072 166.4±1.0 1.178±0.085 0.844±0.132 168.0±9.0

42 2.064±0.021 1.461±0.034 175.2±0.5 1.053±0.055 0.622±0.099 41.7±7.7

43 3.002±0.060 2.684±0.098 161.3±2.7 2.440±0.120 2.320±0.100 91.0±9.0

44 3.271±0.056 2.803±0.078 142.1±1.7 2.810±0.081 2.401±0.104 116.0±9.0

45 2.538±0.046 1.970±0.075 158.5±1.5 1.748±0.088 1.538±0.113 129.0±9.0

46 2.548±1.973 1.973±0.091 157.0±7.0 . . . . . . . . .

47 5.050±0.250 4.120±0.200 143.0±9.0 . . . . . . . . .


.......................................................................

2

2

86





ID Region ( ′′ ) ( ′′ ) ( ◦ ) ( ′′ ) ( ′′ ) ( ◦ )

48a IRAS 18228−1312 6.450±0.144 3.501±0.070 4.3±1.2 6.189±0.153 3.268±0.080 5.7±1.5

48b 7.250±0.270 5.940±0.220 46.1±7.4 7.120±0.290 5.670±0.250 49.0±8.0

48c 4.610±0.340 3.640±0.250 65.0±9.0 4.450±0.370 3.130±0.320 67.0±9.0

49 1.784±0.025 1.160±0.040 163.8±0.7 . . . . . . . . .

50 11.100±0.080 8.271±0.107 140.4±0.5 10.960±0.082 8.164±0.109 139.5±0.6

51 G19.27+0.1M1 4.790±0.110 2.980±0.130 167.3±1.7 4.380±0.120 2.730±0.150 167.9±2.2

52 2.071±0.050 1.231±0.082 163.8±1.7 0.694±1.257 0.317±0.872 161.0±9.0

53 2.018±0.021 1.218±0.034 164.9±0.7 0.518±1.433 0.258±0.931 173.0±9.0

54 2.060±0.028 1.387±0.042 171.9±1.3 0.889±1.063 0.391±0.797 36.0±9.0

55 4.480±0.140 3.170±0.110 95.3±2.9 4.290±0.150 0.550±0.150 91.0±9.0

56 8.700±0.260 6.540±0.260 100.9±4.2 8.590±0.280 6.280±0.280 99.3±4.4

57 3.057±0.094 2.282±0.092 141.4±3.5 2.480±0.140 1.780±0.170 124.0±9.0

58 2.447±0.037 1.724±0.059 175.3±1.3 1.571±0.077 1.125±0.116 19.6±9.9

59 1.951±0.033 1.147±0.053 159.8±1.2 . . . . . . . . .

60 2.300±0.140 1.430±0.120 170.1±3.6 1.250±0.310 0.730±0.480 4.2±9.8

61 2.950±0.100 1.900±0.100 152.8±2.5 2.250±0.150 1.430±0.170 143.8±8.4

63 IRAS 18236−1205 0.485±0.019 0.393±0.021 134.6±7.2 . . . . . . . . .

64 2.229±0.061 1.148±0.079 170.1±1.5 . . . . . . . . .

65 2.033±0.012 1.112±0.017 163.6±0.3 . . . . . . . . .

66 1.785±0.044 1.121±0.042 156.1±1.6 . . . . . . . . .

67 4.068±1.886 1.886±0.093 116.0±2.0 . . . . . . . . .

68 2.633±0.156 1.893±0.092 138.0±6.0 . . . . . . . . .

69 6.040±0.820 2.180±0.210 3.0±3.0 . . . . . . . . .

70 1.840±0.170 1.400±0.100 126.0±9.0 . . . . . . . . .

71 3.193±0.223 1.823±0.093 159.0±4.0 . . . . . . . . .



.......................................................................

87





ID Region ( ′′ ) ( ′′ ) ( ◦ ) ( ′′ ) ( ′′ ) ( ◦ )

72 G23.60+0.0M1 5.110±0.210 4.960±0.210 114.0±9.0 4.930±0.290 4.660±0.300 86.0±9.0

73 1.770±0.029 1.175±0.035 160.1±1.4 . . . . . . . . .

74 1.793±0.021 1.236±0.025 160.3±1.0 . . . . . . . . .

75 1.884±0.020 1.198±0.026 156.3±0.8 0.444±1.410 0.290±0.842 105.0±9.0

76 1.926±0.019 1.230±0.032 157.3±0.8 0.541±0.083 0.475±0.117 141.0±9.0

77 1.975±0.017 1.372±0.027 162.3±0.9 0.847±0.061 0.602±0.087 35.0±9.0

78 2.285±0.041 1.856±0.057 152.2±3.1 1.497±0.095 1.312±0.111 88.0±9.0

79 2.672±0.090 1.760±0.116 174.9±2.6 2.010±0.140 1.220±0.220 8.9±9.8

80 1.817±0.065 1.405±0.084 173.7±4.9 . . . . . . . . .

81 1.828±0.051 1.418±0.080 169.0±4.0 . . . . . . . . .

82 1.762±0.033 1.324±0.054 152.6±2.3 . . . . . . . . .

83 IRAS 18316−0602 1.987±0.041 1.376±0.060 174.1±1.9 1.181±0.096 0.592±0.228 22.5±9.8

84 1.741±0.018 1.230±0.028 163.2±1.1 0.620±1.085 0.251±0.838 46.0±9.0

85 1.759±0.014 1.253±0.021 165.1±0.9 0.696±0.053 0.283±0.184 41.5±7.6

86 1.683±0.021 1.250±0.031 163.2±1.7 . . . . . . . . .

87 1.685±0.021 1.215±0.032 159.8±1.4 . . . . . . . . .

88 4.688±0.090 2.687±0.094 13.0±1.1 4.434±0.099 2.333±0.117 16.2±1.5

89 2.250±0.100 1.690±0.110 118.4±4.7 1.790±0.160 0.800±0.400 98.0±9.0

90 1.709±0.050 1.240±0.074 145.2±3.2 . . . . . . . . .

91 1.521±0.081 1.376±0.099 172.0±9.0 . . . . . . . . .

92 2.140±0.110 1.500±0.110 145.9±4.7 1.380±0.230 0.910±0.330 124.0±9.0

93 2.474±0.119 1.509±0.075 173.9±2.4 1.840±0.180 0.950±0.170 4.0±7.9

96 G24.08+0.0M2 1.977±0.023 1.317±0.036 150.6±1.0 0.942±0.068 0.487±0.151 118.6±8.7

97 1.935±0.026 1.402±0.039 171.8±1.5 1.008±0.078 0.418±0.265 37.1±7.8

98 2.469±0.022 1.492±0.022 174.4±0.6 1.759±0.035 0.842±0.052 6.8±1.8


.......................................................................

2

2

88





ID Region ( ′′ ) ( ′′ ) ( ◦ ) ( ′′ ) ( ′′ ) ( ◦ )

99 1.940±0.019 1.464±0.030 161.5±1.2 0.942±0.054 0.708±0.068 59.0±9.0

100 2.967±0.039 1.410±0.050 159.1±0.7 2.357±0.050 0.846±0.086 159.1±1.0

101 2.471±0.101 1.714±0.054 22.8±3.0 2.020±0.140 0.660±0.280 39.1±5.2

102 1.907±0.016 1.358±0.022 161.6±0.9 0.776±0.049 0.599±0.070 50.0±9.0

103 2.146±0.049 1.399±0.053 153.8±1.7 1.190±0.100 0.790±0.130 140.0±9.0

104 2.180±0.150 1.540±0.110 159.7±5.7 1.230±0.270 1.050±0.270 162.0±9.0

105 3.350±0.270 2.830±0.210 162.0±9.0 . . . . . . . . .

106 26.310±1.400 19.000±0.530 69.7±4.3 26.280±1.400 18.910±0.540 69.7±4.3

107 1.369±0.138 1.020±0.072 160.0±9.6 . . . . . . . . .

108 1.540±1.289 1.289±0.069 10.0±9.1 . . . . . . . . .

109 0.940±0.130 0.940±0.130 168.0±7.7 . . . . . . . . .

110 G24.33+0.1M1 1.661±0.021 1.259±0.029 166.9±0.7 . . . . . . . . .

111 2.114±0.064 1.722±0.047 78.2±1.8 1.708±0.081 0.315±0.931 78.0±9.0

112 2.571±0.064 2.361±0.081 147.2±4.0 2.070±0.110 1.860±0.120 103.0±9.0

113 1.957±0.051 1.511±0.066 150.4±1.8 1.120±0.110 0.650±0.220 121.0±9.0

114 1.821±0.022 1.314±0.030 167.3±0.7 0.670±0.064 0.414±0.114 170.0±9.0

115 1.976±0.051 1.584±0.068 155.8±2.0 1.120±0.120 0.860±0.160 121.0±9.0

116 1.861±0.079 1.673±0.072 112.4±4.5 1.321±0.079 0.307±0.072 87.9±4.5

117 G24.60+0.1M1 1.681±0.041 1.456±0.056 168.5±3.9 0.800±0.873 0.170±1.500 82.0±9.0

118 2.304±0.086 2.068±0.067 60.9±6.6 1.940±0.110 1.240±0.130 71.6±6.6

119 3.087±0.074 2.396±0.081 132.6±2.5 2.708±0.096 1.914±0.118 122.5±5.1

120 2.232±0.048 1.467±0.052 20.8±1.5 1.636±0.076 0.426±0.791 35.0±9.0

121 5.480±0.200 2.210±0.110 34.4±1.5 5.280±0.210 1.650±0.160 35.9±1.7

122 6.720±0.150 3.290±0.150 38.5±1.3 6.570±0.160 2.930±0.170 39.6±1.4

123 1.856±0.013 1.205±0.018 163.1±0.5 . . . . . . . . .



.......................................................................

89





ID Region ( ′′ ) ( ′′ ) ( ◦ ) ( ′′ ) ( ′′ ) ( ◦ )

124 1.902±0.066 1.788±0.073 25.0±9.0 1.380±0.130 0.800±0.250 71.0±9.0

125 1.594±0.041 1.122±0.056 168.9±2.3 . . . . . . . . .

126 1.708±0.042 1.097±0.044 162.8±1.8 . . . . . . . . .

127 2.910±0.120 2.530±0.130 132.0±7.6 2.530±0.170 2.050±0.200 114.0±9.0

128 4.754±0.072 2.358±0.042 5.9±0.6 4.462±0.078 1.992±0.052 7.2±0.8

129 2.374±0.071 1.524±0.096 176.8±2.1 1.690±0.110 0.900±0.190 1.1±5.6

130 1.908±0.037 1.501±0.047 151.2±3.0 1.115±0.097 0.612±0.201 118.0±9.0

131 2.076±0.077 1.161±0.061 165.9±2.1 . . . . . . . . .

132 G24.60+0.01M2 2.003±0.081 1.537±0.046 137.2±3.8 1.400±0.130 0.790±0.220 103.0±9.0

133 2.778±0.097 2.206±0.101 0.1±5.5 2.310±0.150 1.860±0.170 17.0±9.0

134 2.080±0.180 1.490±0.140 98.7±6.2 . . . . . . . . .

135 1.826±0.074 1.119±0.078 166.0±2.7 0.813±0.843 0.354±0.673 4.7±9.8

137 G34.43+0.2M3 3.890±0.110 2.660±0.110 165.1±1.4 3.580±0.120 2.190±0.130 166.2±1.7

138 1.764±0.020 1.587±0.020 156.0±1.8 0.898±0.045 0.483±0.084 172.0±5.7

139 2.024±0.065 1.596±0.054 143.8±2.5 1.287±0.087 0.594±0.181 149.0±5.0

140 1.553±0.023 1.880±0.061 15.4±8.2 1.390±0.100 1.040±0.120 21.0±8.5

141 3.575±0.098 1.410±0.024 150.5±2.5 . . . . . . . . .

142 2.217±0.056 2.918±0.098 74.9±4.1 3.240±0.110 2.490±0.120 72.7±4.9

143 IRAS 19095+0930 2.014±0.039 1.445±0.044 115.5±0.8 1.250±0.350 0.340±1.030 124.0±9.0

144 3.158±0.079 2.577±0.091 87.0±1.6 2.724±0.090 2.177±0.111 84.3±3.7

145 1.601±0.011 1.370±0.013 94.7±0.6 . . . . . . . . .

146 1.715±0.018 1.365±0.021 105.1±0.6 . . . . . . . . .

Observed and deconvolved source sizes for K band detections


.......................................................................

2

2

90





ID Region ( ′′ ) ( ′′ ) ( ◦ ) ( ′′ ) ( ′′ ) ( ◦ )

48a IRAS 18228−1312 2.950±0.390 2.090±0.220 166.0±9.0 . . . . . . . . .

48b 3.070±0.640 2.330±0.410 79.0±9.0 3.060±0.210 2.310±0.160 169.0±9.0

48c 7.619±0.716 1.042±0.085 81.9±0.9 2.300±1.000 1.400±1.000 119.0±9.0

62 G19.27+0.1M1 0.485±0.019 0.393±0.021 134.6±7.2 . . . . . . . . .

63 IRAS 18236−1205 1.670±0.024 0.970±0.011 133.5±7.9 1.510±0.028 0.690±0.017 132.5±9.4

64 0.868±0.036 0.738±0.026 121.0±8.9 0.499±0.079 0.240±0.168 109.0±9.0

65 0.799±0.016 0.713±0.013 109.9±6.5 0.394±0.040 0.035±0.136 93.2±9.2

73 G23.60+0.0M1 0.598±0.087 0.365±0.035 54.1±7.5 . . . . . . . . .

74 1.022±0.279 0.397±0.061 18.9±5.7 0.878±0.356 0.226±0.111 15.0±9.0

83 IRAS 18316−0602 1.069±0.105 0.611±0.048 178.4±5.4 0.969±0.126 0.451±0.096 171.2±9.2

94 0.946±0.284 0.682±0.180 57.0±9.0 0.809±0.374 0.576±0.511 70.0±9.0

95 0.455±0.081 0.388±0.059 16.0±9.0 . . . . . . . . .

96 G24.08+0.0M2 0.503±0.021 0.407±0.023 27.4±7.8 . . . . . . . . .

110 G24.33+0.1M1 0.528±0.018 0.350±0.020 33.4±3.3 . . . . . . . . .

136 G24.60+0.01M2 0.685±0.068 0.492±0.068 78.0±9.0 . . . . . . . . .

137 G34.43+0.2M3 1.025±0.060 0.797±0.070 61.0±9.0 0.899±0.085 0.715±0.108 123.8±3.9

143 IRAS 19095+0930 0.572±0.012 0.371±0.012 50.0±1.9 0.269±0.034 0.181±0.058 94.0±9.0



.......................................................................

Table 2.7: Intensities, fluxes and source sizes for the sources detected at both frequency bands and imaged

using the common uv range (see Sect. 2.4.2). (a) The intensities and fluxes are corrected for the primary

beam response of the antennas, except for source 137, which is located outside the primary beam of the

antennas and no correction factor can be determined.

Convolved Image Size

Deconvolved Image Size

Intensity a Flux a Major Axis Minor Axis PA Major Axis Minor Axis PA θ beam , PA rms

ID Region (mJy beam −1 ) (mJy) ( ′′ ) ( ′′ ) ( ◦ ) ( ′′ ) ( ′′ ) ( ◦ ) ( ′′ × ′′ , ◦ ) (mJy)

Intensities, fluxes and sizes from the C band images with common uv range

48a 18228−1312 11.50±0.61 55.3±3.5 4.60±0.28 2.08±0.10 0±2 4.24±0.31 1.72±0.13 2±3 1.88±1.66, −14 0.400

48b 8.76±1.0 75.4±9.5 4.35±0.50 3.93±0.44 40±44 4.15±0.72 3.56±0.65 52±54 1.88±1.66, −14 0.400

48c 32.3±1.7 129.5±8.3 3.21±0.18 2.48±0.12 61±8 3.00±0.21 1.73±0.21 66±8 1.88±1.66, −14 0.400

63 18236−1205 0.56±0.02 0.99±0.06 2.24±0.11 1.66±0.07 162±5 1.26±0.17 1.13±0.28 71±63 1.99±1.11, −18 0.021

64 0.30±0.01 0.28±0.02 1.97±0.09 0.97±0.02 168±1 . . . . . . . . . 1.94±1.08, −18 0.005

65 0.56±0.07 0.57±0.14 1.99±0.04 1.07±0.01 163±1 . . . . . . . . . 1.94±1.08, −18 0.004

73 G23.60+0.0M1 0.22±0.08 0.24±0.15 1.82±0.09 1.14±0.03 159±2 0.48±0.28 0.30±0.16 13±41 1.77±1.08, −23 0.008

74 0.29±0.01 0.35±0.03 1.94±0.11 1.27±0.05 163±4 0.73±0.28 0.38±0.29 25±40 1.85±1.13, −22 0.008

83 18316−0602 2.21±0.09 3.35±0.21 1.94±0.09 1.36±0.05 173±4 1.16±0.18 0.65±0.29 21±16 1.64±1.05, −22 0.011

96 G24.08+0.0M2 18.70±0.30 24.97±0.65 1.95±0.04 1.29±0.02 151±1 0.98±0.09 0.57±0.10 125±10 1.73±1.08, −21 0.624

110 G24.33+0.1M1 0.46±0.12 0.46±0.21 1.55±0.05 1.24±0.03 168±4 . . . . . . . . . 1.61±1.19, −13 0.009

137 G34.43+0.2M3 3.41±0.23 14.40±1.20 3.36±0.24 2.54±0.17 170±9 3.04±0.28 2.10±0.21 171±10 1.46±1.38, −59 0.242

143 19095+0930 30.03±0.74 39.50±1.60 1.88±0.05 1.34±0.03 113±3 1.18±0.01 0.26±0.17 121±5 1.49±1.28, −85 0.107

Intensities, fluxes and sizes from the K band images with the common uv range

48a 18228−1312 1.51±0.13 15.19±1.51 3.02±0.28 1.62±0.13 1±5 2.93±0.29 1.47±0.15 1±5 2.50±1.25, +57 0.200

48b 1.59±0.17 10.53±1.30 2.09±0.23 1.54±0.16 67±13 1.98±0.25 1.36±0.18 67±13 2.50±1.25, +57 0.200

48c 6.28±0.36 54.41±3.51 2.81±0.17 1.50±0.08 67±3 2.73±0.18 1.31±0.09 68±3 2.50±1.25, +57 0.200

63 18236−1205 0.21±0.03 0.75±0.13 1.58±0.23 1.13±0.14 115±15 1.41±0.28 0.88±0.21 113±23 0.51±0.32, −26 0.002

64 0.17±0.03 0.35±0.08 1.24±0.24 0.81±0.12 127±13 1.01±0.32 0.43±0.28 125±27 0.73±0.67, −25 0.004

65 2.03±0.05 2.31±0.09 0.81±0.02 0.70±0.02 110±6 . . . . . . . . . 0.73±0.67, −25 0.002

73 G23.60+0.0M1 0.44±0.05 0.43±0.09 0.89±0.12 0.61±0.06 74±10 . . . . . . . . . 0.76±0.71, −1 0.001

74 0.20±0.04 0.49±0.13 1.01±0.27 0.39±0.06 18±5 0.86±0.34 0.21±0.11 15±17 0.54±0.30, +32 0.002

83 18316−0602 1.55±0.09 3.73±0.29 1.34±0.08 0.96±0.05 3±6 1.11±0.10 0.65±0.08 3±7 0.91±0.84, −47 1.924

96 G24.08+0.0M2 4.75±0.32 8.26±0.83 1.18±0.09 0.78±0.05 128±6 0.94±0.12 0.24±0.13 125±7 0.73±0.70, +6 0.008

110 G24.33+0.1M1 0.97±0.31 1.20±0.06 0.93±0.03 0.85±0.03 146±13 0.45±0.11 0.30±0.18 1±37 0.85±0.75, −56 0.002

137 G34.43+0.2M3 0.53±0.04 2.21±0.20 2.33±0.19 1.50±0.10 179±6 1.19±0.14 2.14±0.21 180±7 0.95±0.88, −41 0.043

143 19095+0930 117.6±1.4 130.8±2.6 0.94±0.01 0.90±0.01 117±10 0.34±0.05 0.23±0.09 66±29 0.91±0.84, −47 1.924

91

2



.......................................................................

2

Dec (J2000)

Dec (J2000)

Dec (J2000)

Dec (J2000)

58"

56"

54"

52"

+35°45'50"

54"

52"

50"

48"

+35°45'46"

34"

32"

30"

28"

+35°44'26"

48"

46"

44"

42"

+35°49'40"

05358+3543 (1)

200 AU

5h39m26s

05358+3543 (4)

5h39m13s

05358+3543 (7)

25

20

mJy/Beam

15

10

5

0

0.25

0.20

0.15

mJy/Beam

0.10

0.05

0.00

0.000

5h39m14s

RA (J2000)

05358+3543 (10) 0.05

5h38m39s

RA (J2000)

0.175

0.150

0.125

0.100

mJy/Beam

56"

54"

52"

50"

+35°45'48"

22"

20"

18"

16"

+35°45'14"

02"

44'00"

58"

0.075

56"

0.050

0.025 +35°43'54"

0.04

0.03

mJy/Beam

0.02

0.01

0.00

50"

48"

46"

44"

+35°46'42"

05358+3543 (2)

5h39m13s

05358+3543 (5)

5h39m10s

05358+3543 (8)

5h39m12s

05358+3543 (11)

5h39m37s

RA (J2000)

0.5

12"

0.4

10"

0.3

08"

0.2

06"

0.1

+35°46'04"

0.0

mJy/Beam

0.30

0.25

0.20

mJy/Beam

0.15

0.10

08"

06"

04"

02"

0.05

+35°48'00"

0.00

0.175

0.150

0.125

mJy/Beam

46"

44"

0.100 42"

0.075

0.050

40"

0.025 +35°44'38"

0.000

0.08

0.06

mJy/Beam

0.04

0.02

0.00

05358+3543 (3)

5h39m10s

05358+3543 (6)

05358+3543 (9)

5h38m55s

RA (J2000)

0.08

0.06

mJy/Beam

0.04

0.02

0.00

0.14

0.12

0.10

0.08

0.06

0.04

0.02

0.00

mJy/Beam

0.35

0.30

0.25

mJy/Beam

0.20

0.15

0.10

0.05

0.00

Figure 2.20: Close-up views of the C- (color-scale image) and K-band

(contours) continuum images for the sources listed in Table 2.5. Maps

for the sources detected in region IRAS 05358+3543.

92


.......................................................................

Dec (J2000)

Dec (J2000)

Dec (J2000)

16"

14"

12"

10"

+20°39'08"

08"

06"

04"

02"

+20°39'00"

20"

18"

16"

14"

+20°38'12"

G189.78+0.34 (12)

200 AU

6h08m35s

G189.78+0.34 (15)

G189.78+0.34 (18)

6h08m45s

RA (J2000)

0.45

0.40

0.35

0.30

0.25

0.20

0.15

0.10

0.05

0.00

mJy/Beam

0.30

0.25

0.20

mJy/Beam

0.15

0.10

0.05

0.00

0.07

0.06

0.05

mJy/Beam

0.04

0.03

0.02

0.01

0.00

14"

12"

10"

08"

+20°39'06"

06"

04"

02"

39'00"

+20°38'58"

18"

16"

14"

12"

+20°43'10"

G189.78+0.34 (13)

0.8

0.7

0.6

0.5

0.4

0.3

0.2

0.1

0.0

6h08m35s

G189.78+0.34 (16)

0.5

G189.78+0.34 (19)

RA (J2000)

mJy/Beam

0.4

mJy/Beam

0.3

0.2

0.1

0.0

0.07

0.06

0.05

mJy/Beam

0.04

0.03

0.02

0.01

0.00

10"

08"

06"

04"

+20°39'02"

12"

10"

08"

06"

+20°41'04"

10"

08"

06"

04"

+20°33'02"

G189.78+0.34 (14)

0.35

0.30

0.25

mJy/Beam

0.20

0.15

0.10

0.05

0.00

6h08m35s

G189.78+0.34 (17) 0.10

G189.78+0.34 (20)

6h08m44s

RA (J2000)

0.08

0.06

mJy/Beam

0.04

0.02

0.00

0.10

0.08

0.06

mJy/Beam

0.04

0.02

0.00

2



for the sources detected in region G189.78+0.34.

93



.......................................................................

2

Dec (J2000)

Dec (J2000)

30"

28"

26"

24"

+18°00'22"

38"

36"

34"

32"

+17°59'30"

G192.58-0.04 (21)

200 AU

G192.58-0.04 (24)

12 28"

10

26"

8

24"

6

4 22"

2 +17°59'20"

0

mJy/Beam

0.5

0.4

mJy/Beam

0.3

0.2

0.1

0.0

6h12m54s

RA (J2000)

24"

22"

20"

18"

+17°59'16"

G192.58-0.04 (22)

6h12m54s

G192.58-0.04 (25)

6h12m55s

RA (J2000)

1.4

1.2 26"

1.0 24"

0.8

0.6

22"

0.4 20"

0.2

0.0

+17°59'18"

mJy/Beam

0.30

10"

0.25 08"

0.20

06"

0.15

04"

0.10

0.05 +18°03'02"

0.00

mJy/Beam

G192.58-0.04 (23)

6h12m54s

G192.58-0.04 (26)

6h12m50s

RA (J2000)

0.5

0.4

mJy/Beam

0.3

0.2

0.1

0.0

0.25

0.20

0.15

mJy/Beam

0.10

0.05

0.00



for the sources detected in region G192.58−0.04.

94


.......................................................................

Dec (J2000)

Dec (J2000)

Dec (J2000)

Dec (J2000)

-13°15'36"

38"

40"

42"

44"

-13°15'42"

44"

46"

48"

50"

-13°12'48"

50"

52"

54"

56"

-13°15'30"

32"

34"

36"

38"

G18.18-0.30 (27)

200 AU

18h25m01s

G18.18-0.30 (30)

18h25m01s

G18.18-0.30 (33)

18h24m54s

G18.18-0.30 (36)

RA (J2000)

3.0

2.5

2.0

mJy/Beam

1.5

1.0

0.5

0.0

1.0

0.8

0.6

mJy/Beam

0.4

0.2

0.0

0.30

0.25

0.20

mJy/Beam

0.15

0.10

0.05

0.00

0.05

0.04

0.03

mJy/Beam

0.02

0.01

0.00

-13°17'54"

56"

58"

18'00"

02"

-13°15'32"

34"

36"

38"

40"

-13°12'24"

26"

28"

30"

32"

G18.18-0.30 (28)

G18.18-0.30 (31)

18h25m01s

G18.18-0.30 (34)

18h25m07s

G18.18-0.30 (37)

-13°10'12"

14"

16"

18"

20"

22"

18h24m56s

RA (J2000)

1.4

1.2

1.0

0.8

0.6

0.4

0.2

0.0

mJy/Beam

1.0

0.8

0.6

mJy/Beam

0.4

0.2

0.0

0.200

0.175

0.150

0.125

0.100

0.075

0.050

0.025

0.000

mJy/Beam

0.14

0.12

0.10

0.08

0.06

0.04

0.02

0.00

mJy/Beam

-13°18'08"

10"

12"

14"

16"

-13°14'42"

44"

46"

48"

50"

-13°10'18"

20"

22"

24"

26"

-13°10'12"

14"

16"

18"

20"

22"

G18.18-0.30 (29)

18h25m07s

G18.18-0.30 (32)

18h24m45s

G18.18-0.30 (35)

G18.18-0.30 (38)

18h24m56s

RA (J2000)

0.9

0.8

0.7

0.6

0.5

0.4

0.3

0.2

0.1

0.0

mJy/Beam

0.7

0.6

0.5

0.4

0.3

0.2

0.1

0.0

mJy/Beam

0.35

0.30

0.25

mJy/Beam

0.20

0.15

0.10

0.05

0.00

0.09

0.08

0.07

0.06

0.05

0.04

0.03

0.02

0.01

0.00

mJy/Beam

2



for the sources detected in region G18.18−0.30.

95



.......................................................................

2

Dec (J2000)

Dec (J2000)

Dec (J2000)

-12°41'20"

22"

24"

26"

28"

-12°39'16"

18"

20"

22"

24"

-12°34'32"

34"

36"

38"

40"

IRAS 18223-1243 (39)

200 AU

IRAS 18223-1243 (42)

RA (J2000)

IRAS 18223-1243 (45)

18h26m05s

RA (J2000)

20.0

17.5

15.0

12.5

10.0

7.5

5.0

2.5

0.0

mJy/Beam

0.9

0.8

0.7

0.6

0.5

0.4

0.3

0.2

0.1

0.0

mJy/Beam

0.200

0.175

0.150

0.125

0.100

0.075

0.050

0.025

0.000

mJy/Beam

-12°37'40"

42"

44"

46"

48"

-12°51'00"

02"

04"

06"

08"

-12°34'32"

34"

36"

38"

40"

IRAS 18223-1243 (40)

18h25m04s

IRAS 18223-1243 (43)

17.5

15.0

12.5

10.0

7.5

5.0

2.5

0.0

mJy/Beam

1.8

1.6

1.4

1.2

1.0

0.8

0.6

0.4

0.2

0.0

18h24m36s

IRAS 18223-1243 (46)

0.12

18h26m05s

RA (J2000)

mJy/Beam

0.10

0.08

mJy/Beam

0.06

0.04

0.02

0.00

-12°40'30"

32"

34"

36"

IRAS 18223-1243 (41)

38"

18h25m27s

IRAS 18223-1243 (44)

-12°52'00"

02"

04"

06"

08"

-12°50'56"

58"

51'00"

02"

IRAS 18223-1243 (47)

04"

18h24m35s

RA (J2000)

0.8

0.7

0.6

0.5

0.4

0.3

0.2

0.1

0.0

mJy/Beam

0.30

0.25

0.20

mJy/Beam

0.15

0.10

0.05

0.00

0.14

0.12

0.10

0.08

0.06

0.04

0.02

0.00

mJy/Beam



for the sources detected in region IRAS 18223−1243.

96


.......................................................................

2




97



.......................................................................

2

Dec (J2000)

Dec (J2000)

Dec (J2000)

Dec (J2000)

-11°56'24"

28"

32"

36"

40"

-12°05'12"

16"

20"

24"

28"

-12°26'08"

12"

16"

20"

24"

-11°56'20"

24"

28"

32"

36"

G19.27+0.1M1 (51)

0.30

0.25

0.20

mJy/Beam

0.15

0.10

0.05

200 AU

0.00

18h27m38s

G19.27+0.1M1 (54) 0.16

0.14

0.12

0.10

0.08

0.06

0.04

0.02

0.00

46s 18h25m45s

G19.27+0.1M1 (57) 0.18

0.16

0.14

0.12

0.10

0.08

0.06

0.04

0.02

0.00

49s 18h26m48s

G19.27+0.1M1 (60) 0.10

18h27m37s

RA (J2000)

mJy/Beam

mJy/Beam

0.08

0.06

mJy/Beam

0.04

0.02

0.00

-12°05'24"

28"

32"

36"

40"

-11°52'12"

16"

20"

24"

28"

-11°57'52"

56"

58'00"

04"

08"

-12°04'24"

28"

32"

36"

40"

G19.27+0.1M1 (52)

18h26m21s

G19.27+0.1M1 (55)

18h25m54s

G19.27+0.1M1 (58)

18h26m34s

G19.27+0.1M1 (61)

0.200

0.175

0.150

0.125

0.100

0.075

0.050

0.025

0.000

mJy/Beam

0.14

0.12

0.10

0.08

0.06

0.04

0.02

0.00

mJy/Beam

0.25

0.20

0.15

mJy/Beam

0.10

0.05

0.00

0.10

0.08

0.06

mJy/Beam

0.04

0.02

0.00

06s 18h26m05s

RA (J2000)

-12°04'32"

36"

40"

44"

48"

-11°52'24"

28"

32"

36"

40"

05s

-12°01'12"

16"

20"

24"

28"

-12°05'10"

12"

14"

16"

18"

G19.27+0.1M1 (53)

1.0

0.8

0.6

mJy/Beam

0.4

0.2

0.0

04s 18h26m03s

G19.27+0.1M1 (56) 0.200

0.175

0.150

0.125

0.100

0.075

0.050

0.025

0.000

18h26m04s

G19.27+0.1M1 (59) 0.10

18h26m14s

RA (J2000)

G19.27+0.1M1 (62)

18h25m52s

RA (J2000)

0.08

0.06

0.04

0.02

0.00

mJy/Beam

mJy/Beam

0.000200

0.000175

0.000150

0.000125

0.000100

0.000075

0.000050

0.000025

0.000000

mJy/Beam



for the sources detected in region G19.27+0.1 M1.

98


.......................................................................

2




99



.......................................................................

2

Dec (J2000)

Dec (J2000)

Dec (J2000)

Dec (J2000)

-8°07'04"

08"

12"

16"

20"

-8°18'04"

08"

12"

16"

20"

-8°21'16"

20"

24"

28"

32"

-8°21'00"

04"

08"

12"

16"

G23.60+0.0M1 (72) 0.8

0.7

0.6

0.5

0.4

0.3

0.2

200 AU 0.1

0.0

54s 18h33m53s

G23.60+0.0M1 (75) 0.30

18h34m21s

G23.60+0.0M1 (78)

18h33m44s

G23.60+0.0M1 (81)

18h34m18s

RA (J2000)

mJy/Beam

0.25

0.20

mJy/Beam

0.15

0.10

0.05

0.00

0.12

0.10

0.08

mJy/Beam

0.06

0.04

0.02

0.00

0.07

0.06

0.05

mJy/Beam

0.04

0.03

0.02

0.01

0.00

-8°18'52"

56"

19'00"

04"

08"

-8°15'20"

24"

28"

32"

36"

-8°23'28"

32"

36"

40"

44"

-8°24'08"

12"

16"

20"

24"

13s

G23.60+0.0M1 (73)

18h34m12s

G23.60+0.0M1 (76)

18h34m33s

G23.60+0.0M1 (79)

18h33m48s

G23.60+0.0M1 (82)

18h34m14s

RA (J2000)

0.14

0.12

0.10

0.08

0.06

0.04

0.02

0.00

mJy/Beam

2.00

1.75

1.50

1.25

1.00

0.75

0.50

0.25

0.00

mJy/Beam

0.07

0.06

0.05

mJy/Beam

0.04

0.03

0.02

0.01

0.00

0.12

0.10

0.08

mJy/Beam

0.06

0.04

0.02

0.00

-8°19'00"

04"

08"

12"

16"

-8°30'56"

31'00"

04"

08"

12"

-8°24'32"

36"

40"

44"

48"

G23.60+0.0M1 (74)

0.30

0.25

0.20

mJy/Beam

0.15

0.10

0.05

0.00

12s 18h34m11s

G23.60+0.0M1 (77) 0.18

0.16

0.14

0.12

0.10

0.08

0.06

0.04

0.02

0.00

18h34m45s

G23.60+0.0M1 (80) 0.07

18h34m06s

RA (J2000)

mJy/Beam

0.06

0.05

mJy/Beam

0.04

0.03

0.02

0.01

0.00




100


.......................................................................

2

Dec (J2000)

18316-0602 (94)

-6°05'02"

04"

06"

08"

200 AU

10"

18h34m18s

RA (J2000)

0.200

0.175

0.150

0.125

0.100

0.075

0.050

0.025

0.000

mJy/Beam

-5°59'40"

42"

44"

46"

48"

18316-0602 (95)

18h34m21s

RA (J2000)

0.14

0.12

0.10

0.08

0.06

0.04

0.02

0.00

mJy/Beam


(contours) continuum images for the sources listed in Table 2.5. The

images of sources 94 and 95 correspond to the K-band maps. Maps for

the sources detected in region IRAS 18316−0602.

101



.......................................................................

2

Dec (J2000)

Dec (J2000)

Dec (J2000)

Dec (J2000)

-7°46'32"

36"

40"

44"

48"

-7°43'20"

24"

28"

32"

36"

-7°45'16"

20"

24"

28"

32"

-7°52'28"

32"

G24.08+0.0M2 (96)

200 AU

18h34m49s

G24.08+0.0M2 (99)

18h34m57s

G24.08+0.0M2 (102)

18h34m57s

G24.08+0.0M2 (105)

12

10

8

mJy/Beam

6

4

2

0

2.0

1.5

mJy/Beam

1.0

0.5

0.0

1.0

0.8

0.6

mJy/Beam

0.4

0.2

0.0

36"

0.04

0.03

40"

0.02

44"

0.01

34m00s

0.00

18h33m59s

-7°37'28"

G24.08+0.0M2 (108) 0.10

32"

36"

40"

44"

18h35m24s

RA (J2000)

0.07

0.06

0.05

mJy/Beam

0.08

0.06

mJy/Beam

0.04

0.02

0.00

G24.08+0.0M2 (97)

-7°43'48"

5

52"

4

56"

3

2

44'00"

1

04"

42s 18h34m41s

0

-7°42'52"

G24.08+0.0M2 (100)

2.5

56"

2.0

43'00"

1.5

04"

1.0

08"

0.5

35m00s

0.0

18h34m59s

G24.08+0.0M2 (103)

0.12

-7°52'48"

0.10

52"

56"

53'00"

04"

-7°54'30"

36"

42"

48"

54"

55'00"

-7°37'24"

28"

32"

36"

40"

18h34m12s

G24.08+0.0M2 (106)

26s 18h34m25s

G24.08+0.0M2 (109)

18h35m34s

RA (J2000)

mJy/Beam

mJy/Beam

0.08

mJy/Beam

0.06

0.04

0.02

0.00

0.40

0.35

0.30

0.25

0.20

0.15

0.10

0.05

0.00

mJy/Beam

0.10

0.08

0.06

mJy/Beam

0.04

0.02

0.00

-7°43'40"

44"

48"

52"

G24.08+0.0M2 (98)

56"

42s 18h34m41s

G24.08+0.0M2 (101)

-7°42'48"

0.9

0.8

0.7

0.6

0.5

0.4

mJy/Beam

0.3

0.2

0.1

0.0

0.7

52"

0.6

56"

0.5

0.4

43'00"

0.3

0.2

04"

0.1

35m00s

0.0

18h34m59s

G24.08+0.0M2 (104) 0.07

-7°53'00"

0.06

04"

08"

12"

16"

-7°42'08"

12"

16"

20"

24"

18h34m11s

G24.08+0.0M2 (107)

18h34m51s

RA (J2000)

mJy/Beam

0.05

mJy/Beam

0.04

0.03

0.02

0.01

0.00

0.07

0.06

0.05

mJy/Beam

0.04

0.03

0.02

0.01

0.00




102


.......................................................................

Dec (J2000)

Dec (J2000)

Dec (J2000)

-7°35'00"

02"

04"

06"

08"

-7°34'18"

20"

22"

24"

26"

-7°27'20"

22"

24"

26"

28"

G24.33+0.1M1 (110)

200 AU

18h35m08s

G24.33+0.1M1 (113)

18h35m11s

G24.33+0.1M1 (116)

18h35m56s

RA (J2000)

0.35

0.30

0.25

0.20

0.15

0.10

0.05

0.00

mJy/Beam

0.16

0.14

0.12

0.10

0.08

0.06

0.04

0.02

0.00

mJy/Beam

0.14

0.12

0.10

0.08

0.06

0.04

0.02

0.00

mJy/Beam

-7°38'16"

18"

20"

22"

24"

-7°31'16"

18"

20"

22"

24"

G24.33+0.1M1 (111)

G24.33+0.1M1 (114)

18h35m03s

RA (J2000)

0.175

0.150

0.125

0.100

0.075

0.050

0.025

0.000

mJy/Beam

0.14

0.12

0.10

0.08

0.06

0.04

0.02

0.00

mJy/Beam

-7°37'30"

32"

34"

36"

38"

-7°46'36"

38"

40"

42"

44"

G24.33+0.1M1 (112)

18h35m34s

G24.33+0.1M1 (115)

RA (J2000)

0.200

0.175

0.150

0.125

0.100

0.075

0.050

0.025

0.000

mJy/Beam

0.14

0.12

0.10

0.08

0.06

0.04

0.02

0.00

mJy/Beam

2




103



.......................................................................

2

Dec (J2000)

Dec (J2000)

Dec (J2000)

Dec (J2000)

-7°12'03"

06"

09"

12"

G24.60+0.1M1 (117)

15"

200 AU

18h36m13s

G24.60+0.1M1 (120)

-7°21'54"

57"

22'00"

03"

06"

-7°14'15"

18"

21"

24"

27"

-7°19'21"

24"

27"

30"

-7°19'24"

18h35m41s

G24.60+0.1M1 (123)

18h35m53s

G24.60+0.1M1 (126)

33"

18h35m44s

G24.60+0.1M1 (129)

27"

30"

33"

36"

0.45

0.40

0.35

0.30

0.25

0.20

0.15

0.10

0.05

0.00

mJy/Beam

0.10

0.08

0.06

mJy/Beam

0.04

0.02

0.00

0.35

0.30

0.25

mJy/Beam

0.20

0.15

0.10

0.05

0.00

0.10

0.08

0.06

mJy/Beam

0.04

0.02

0.00

0.07

0.06

0.05

mJy/Beam

0.04

0.03

0.02

0.01

0.00

18h35m40s

RA (J2000)

-7°12'09"

12"

15"

18"

G24.60+0.1M1 (118)

21"

18h36m13s

G24.60+0.1M1 (121)

-7°21'51"

54"

57"

22'00"

03"

-7°05'03"

06"

09"

12"

15"

-7°25'50"

55"

26'00"

05"

10"

-7°26'33"

36"

39"

42"

45"

18h35m41s

G24.60+0.1M1 (124)

18h35m17s

G24.60+0.1M1 (127)

0.40

0.35

0.30

0.25

0.20

0.15

0.10

0.05

0.00

mJy/Beam

0.10

0.08

0.06

mJy/Beam

0.04

0.02

0.00

0.20

0.15

mJy/Beam

0.10

0.05

0.00

0.00

04s 18h35m03s

G24.60+0.1M1 (130) 0.07

18h35m00s

RA (J2000)

0.10

0.08

0.06

mJy/Beam

0.04

0.02

0.06

0.05

mJy/Beam

0.04

0.03

0.02

0.01

0.00

-7°22'00"

03"

06"

09"

G24.60+0.1M1 (119)

12"

18h35m41s

G24.60+0.1M1 (122)

-7°31'15"

18"

21"

24"

27"

-7°16'36"

18h36m06s

G24.60+0.1M1 (125)

39"

42"

45"

48"

-7°08'45"

48"

51"

54"

57"

-7°02'33"

36"

39"

42"

45"

18h35m41s

G24.60+0.1M1 (128)

18h36m18s

G24.60+0.1M1 (131)

0.16

0.14

0.12

0.10

0.08

0.06

0.04

0.02

0.00

mJy/Beam

0.35

0.30

0.25

mJy/Beam

0.20

0.15

0.10

0.05

0.00

0.10

0.08

0.06

mJy/Beam

0.04

0.02

0.00

0.12

0.10

0.08

mJy/Beam

0.06

0.04

0.02

0.00

0.06

0.05

0.04

mJy/Beam

0.03

0.02

0.01

0.00

18h34m39s

RA (J2000)




104


.......................................................................

-7°18'30"

G24.60+0.1M2 (136)

0.10

2

33"

0.08

Dec (J2000)

36"

39"

0.06

mJy/Beam

0.04

42"

200 AU

18h35m40s

RA (J2000)

0.02

0.00


(contours) continuum images for the sources listed in Table 2.5. The

image of source 136 corresponds to the K-band map. Maps for the

sources detected in region G24.60+0.1 M2.

105



.......................................................................

2




Dec (J2000)

Dec (J2000)

54"

52"

50"

48"

+9°35'46"

28"

26"

24"

22"

+9°36'20"

19095+0930 (143)

200 AU

19h11m54s

19095+0930 (146)

RA (J2000)

30

25

20

mJy/Beam

15

10

5

0

2.25

2.00

1.75

1.50

1.25

1.00

0.75

0.50

0.25

0.00

mJy/Beam

54"

51"

48"

+9°35'45"

19095+0930 (144)

3.5

3.0

2.5

mJy/Beam

2.0

1.5

1.0

0.5

0.0

19h11m54s

RA (J2000)

58"

56"

54"

52"

+9°36'50"

19095+0930 (145)

19h11m47s

RA (J2000)

1.4

1.2

1.0

mJy/Beam

0.8

0.6

0.4

0.2

0.0



for the sources detected in region IRAS 19095+0930.

106


.......................................................................

2

107



.......................................................................

2

108

Chapter 3

Origin of hydrogen fluoride

emission in the Orion Bar.

An excellent tracer for

CO-dark H 2 gas clouds

Ü. Kavak, F. F. S. van der Tak, A. G. G. M. Tielens, and R. F. Shipman

1

3

3.1 Abstract

The hydrogen fluoride (HF) molecule is seen in absorption in the interstellar

medium (ISM) along many lines of sight. Surprisingly, it is

observed in emission toward the Orion Bar, which is an interface between

the ionized region around the Orion Trapezium stars and the

Orion molecular cloud. We aim to understand the origin of HF emission

in the Orion Bar by comparing its spatial distribution with other tracers.

We examine three mechanisms to explain the HF emission: thermal

excitation, radiative dust pumping, and chemical pumping. We used a

Herschel/HIFI strip map of the HF J = 1 → 0 line, covering 0.5 ′ by 1.5 ′

that is oriented perpendicular to the Orion Bar. We used the RADEX

non-local thermodynamic equilibrium (non-LTE) code to construct the

1 Kavak Ü., et al., 2019, Volume 631, A117

109

CHAPTER 3: Origin of hydrogen fluoride emission in the Orion Bar. An excellent

tracer for CO-dark H 2 gas clouds

.......................................................................

HF column density map. We use the Meudon PDR code to explain the

morphology of HF. The bulk of the HF emission at 10km s −1 emerges

from the CO-dark molecular gas that separates the ionization front from

the molecular gas that is deeper in the Orion Bar. The excitation of

HF is caused mainly by collisions with H 2 at a density of 10 5 cm −3 together

with a small contribution of electrons in the interclump gas of

the Orion Bar. Infrared pumping and chemical pumping are not important.

We conclude that the HF J = 1 → 0 line traces CO-dark molecular

gas. Similarly, bright photodissociation regions associated with massive

star formation may be responsible for the HF emission observed toward

active galactic nuclei.

3

110

3.2 Introduction

.......................................................................

3.2 Introduction

The penetration of UV-photons (hν < 13.6 eV), emitted by massive

stars, leads to bright regions at the edges of molecular clouds that are

called photo-dissociation regions (PDRs) 2 (Hollenbach & Tielens 1999;

Wolfire et al. 2003). PDRs can also be seen in high-mass star-forming

regions, protoplanetary disks, and the nuclei of active galaxies. The

penetration of FUV photons regulates the thermal and chemical balance

of the gas in a PDR. The gradual decrease of the FUV flux in a PDR

results in a layered structure (Tielens et al. 1993) where a chemical phase

transition, such as H + → H → H 2 and C + → C → CO, occurs (Kaufman

et al. 1999; Wolfire et al. 2003).

The Orion Bar is a prototypical PDR at a distance of 414 pc (Tauber

et al. 1994; Menten et al. 2007), located between the Orion molecular

cloud and the Orion Nebula, the HII region surrounding the Trapezium

stars. Observations at infrared and sub-millimeter wavelengths first indicate

a geometry for the bar where the PDR is wrapped around the

Orion Nebula and second, changes from a face-on to an edge-on view in

the Orion Bar where the molecular emission peaks (Hogerheijde et al.

1995; Walmsley et al. 2000). The mean temperature of the molecular

gas in the bar is 85 K, while the temperature rises to several 100 K

toward the ionization front (Ossenkopf et al. 2013), where the emission

from polycyclic aromatic hydrocarbon (PAH) particles and vibrationally

excited H 2 are observed (Walmsley et al. 2000).

While the temperature structure of the Orion Bar is reasonably well

understood (Tielens & Hollenbach 1985b; Ossenkopf et al. 2013; Nagy

et al. 2017), the same cannot be said about the density structure. The

mean density of the molecular gas is 10 5 cm −3 , but single-dish observations

already indicate the presence of random small-scale density variations,

usually called ‘clumps’ (Hogerheijde et al. 1995), which are also

seen toward other PDRs (Stutzki et al. 1988; Wang et al. 1993). While interferometric

observations have confirmed the presence of clumps (Young

Owl et al. 2000), the densities of both the clumps and the interclump

medium are somewhat uncertain. The interclump medium probably has

3

2 We prefer this term over photon-dominated region, because HII regions and AGN

nuclei are also dominated by photons; however, we use the term photo-ionization of

atoms rather than photodissociation of molecules).

111



.......................................................................

3

a density between a few 10 4 and 2 × 10 5 cm −3 (Simon et al. 1997), while

estimates of the clump density range from 1.5×10 6 cm −3 to 6×10 6 cm −3

(Lis & Schilke 2003). Goicoechea et al. (2016) show the presence of even

denser and small gas clumps that are close to the edge of the cloud using

high-resolution Atacama Large Millimeter Array (ALMA) observations.

In addition to gas clumps, dust condensations in the Orion Bar were

found by Qiu et al. (2018). These condensations have temperatures

between 50 − 73 K and masses of between 0.03 − 0.3 M ⊙ , and are very

compact, that is, r < 0.01 pc. They are located right behind the PAH

ridge of the Orion Bar.

We study the origin of the HF emission in the Orion Bar by using a

map of the HF J = 1 → 0 line. We also investigate whether we can use HF

as a tracer of CO-dark molecular gas or not. HF is an F-bearing hydride

molecule which has been established as a surrogate tracer of molecular

hydrogen in diffuse clouds (Emprechtinger et al. 2012). Halogencontaining

molecules like HF have a unique thermochemistry (Neufeld

& Wolfire 2009). In particular, only fluorine has a higher affinity to

hydrogen than hydrogen itself so that the reaction,

H 2 + F HF + H,

is exothermic. Models by Neufeld & Wolfire (2009) predict that, in the

presence of H 2 , all of the gas phase fluorine is rapidly converted into HF,

resulting in an abundance of ∼2 ×10 −8 in diffuse clouds, that is, they

are close to the Solar fluorine abundance (Neufeld et al. 2010). Herschel

observations of the HF J = 0 → 1 line confirm this prediction: the line is

seen in absorption toward several background sources, with abundances

of ∼2–3 ×10 −8 (Neufeld et al. 2010). Toward dense clouds, the HF

abundance is measured to be ∼100 times lower (Phillips et al. 2010),

suggesting significant depletion of F on grain surfaces. In PDRs, the

destruction of HF occurs by photo-dissociation (Neufeld et al. 1997) at

a rate of 1.17 × 10 −10 s −1 χ UV , where χ UV is the mean intensity of the

radiation field that is normalized with respect to the standard interstellar

UV-radiation field of Draine (1978). In addition, reactions with C + can

be an important destruction channel (Neufeld & Wolfire 2009).

HF has been detected in extragalactic sources; such as in emission toward

Mrk 231 (van der Werf et al. 2010), as a P Cygni profile toward Arp

220 (Rangwala et al. 2011), and in absorption toward nearby luminous

112

3.3 Observation and data reduction

.......................................................................

galaxies (Monje et al. 2014) as well as the Cloverleaf quasar at z = 2.56

(Monje et al. 2011b). The ground state transition of HF, that is, J =

0 → 1 appears in absorption in many Galactic lines of sight (Neufeld et al.

1997, 2010; Sonnentrucker et al. 2010; Monje et al. 2011a; Emprechtinger

et al. 2012; van der Wiel et al. 2016). In contrast, IRC+10216, a wellknown

Galactic asymptotic giant branch star, shows HF in emission

(Agúndez et al. 2011). The large dipole moment of HF and the high

frequency of its ground state transition indicate that radiative decay to

the ground state is swift. At the low densities of the diffuse ISM, most

of the HF is in the rotational ground state and emission would be very

weak. This explains why HF can then be readily detected in absorption

toward strong background sources. As an exception, the HF J = 1 → 0

line is observed in emission in the Orion Bar (van der Tak et al. 2012a),

which is illuminated by the Trapezium stars. Three hypotheses are suggested

to explain the HF emission: thermal excitation by collisions with

H 2 or other species; radiative pumping by warm dust continuum or H 2

line emission at ∼2.5 µm; or chemical pumping where most HF is formed

in excited rotational states. To address this issue, we analyzed a spatial

map of the HF emission in the Orion Bar.

We organize the paper as follows. In Section 3.3, we describe the

observations, observing modes, and data reduction. In Section 3.4, we

present direct observational results, while Section 3.5 consists of the analysis

of the data and a comparison of tracers. In Section 3.6, we discuss

the hypotheses and the most efficient excitation mechanism for the HF

emission. Finally, in Section 3.7, we summarize our main conclusions.

3


The observations were made with HIFI (de Graauw et al. 2010) onboard

Herschel (Pilbratt et al. 2010) on 2012 August 28 with observation id

(obsid) 1342250409. The area mapped in HF is outlined on emission

maps of various molecular tracers in Figure 3.1 assembled on the Spitzer

8 µm map. Receiver 5a was used as the front end for mapping of the

Orion Bar in OTF mode, where data are taken continuously while the

telescope scans back and forth across the source. In total, one thousand

spectra have been obtained. The acousto-optical Wide-Band Spectrometer

(WBS) was used as the back-end with full frequency coverage of in-

113



.......................................................................

HF BEAM

CO + PEAK

Theta1

Orionis C

3

Figure 3.1: Spitzer 8 µm map of the Orion Bar. Blue contours show

H 13 CN J = 1 → 0 (Lis & Schilke 2003), which traces dense gas clumps,

white contours are 12 CO J = 1 → 0 (Tauber et al. 1994), which traces

molecular gas, and black contours are [O i] 6300 (Weilbacher et al. 2015),

which traces the ionization front. The red squares show the HF strip map

perpendicular to the Orion Bar.

termediate frequency (IF) 4 GHz bandwidth in four 1140 MHz sub-bands

which have a spectral resolution of 1.1 MHz and a velocity resolution of

1 km s −1 that is smoothed from the native resolution of 0.2676 km s −1 .

The HF map of the Orion Bar was centered on the CO + peak, that

is, α = 05 h 35 m 20.8 s , δ = -05 ◦ 25 ′ 17.10 ′′ (J2000). Reference spectra

have been taken ∼5.5 ′ away at α = 05 h 35 m 45.0 m , δ = -05 ◦ 26 ′ 16.9 ′′

(J2000). The total integration time (OTF + Reference observation) is

105 minutes. The double-sideband system temperature (T sys ) is 920 K.

The full width at half maximum (FWHM) beam size at 1232.476 GHz

is 18.1 ′′ which corresponds to 7500 AU or 0.036 pc at the distance of the

Orion Bar.

114

We inspected the data in the Herschel Interactive Processing Envi-


.......................................................................

1

2

3

Figure 3.2: Map of integrated (between 5 − 13 km s −1 ) HF J = 1 → 0

intensity overlaid with [O i] 6300 , which traces the ionization front of the

Orion Bar and is shown with black contours, and the H 13 CN dense gas

tracer, shown in blue contours. The positions where the three spectra

in Figure 3.3 were extracted are indicated by numbers 1 through 3. The

black circle shows the (18.1 ′′ ) FWHM HIFI beam and the pixel size in

this map is 4.5 ′′ . SMA8 denotes a dust condensation (Qiu et al. 2018).

The light green star denotes the HF peak. The black star shows the

CO + peak.

3

ronment (Herschel Science Ground Segment Consortium 2011, HIPE)

version of 15.0.0 for both polarizations. The level 2 data, produced by

HIFI-pipeline (Shipman et al. 2017), were exported as a FITS file for

further processing in CLASS, which is a sub-package of GILDAS (Gildas

Team 2013). We have estimated the baseline by using a second degree

polynomial fit over the entire channel range. After that, we have converted

the intensity scale to T mb using the mean beam efficiency of 64%

provided by Roelfsema et al. (2012) to obtain the line parameters. Finally,

we have created an integrated intensity map over the 5−13 km s −1

range. The data cube is the combination of individual spectra at each

position.

115



.......................................................................

3.4 Results

Fig. 3.2 shows our HF integrated intensity map of the Orion Bar 3 . The

HF emission appears as a bright ridge separating the ionization front –

traced by [O i] 6300 (Weilbacher et al. 2015) – and the dense molecular

clumps – traced by H 13 CN J = 1 → 0 (Lis & Schilke 2003) – deeper

in the PDR (Fig. 3.2). Faint HF emission is also observed toward the

HII region and the molecular cloud, where we note that the former is

brighter than the latter.

Table 3.1: Parameters of Gaussian fits in Figure 3.3.

∫

Position V LSR Tmb ∆V ∆V T mb

No (km s −1 ) (K km s −1 ) (km s −1 ) (K)

1 8.5 (0.1) 3.7 (0.1) 3.6 (0.1) 1.05

2 10.2 (0.1) 8.5 (0.2) 4.4 (0.1) 1.85

3 10.1 (0.1) 2.4 (0.2) 3.8 (0.3) 0.59

3

We inspected all the lines in the data cube and find 3 distinct regions

(position 1, 2, and 3 in Figure 3.3) that are representative of the emission

in the regions (see Table 3.1 for the line parameters). We do not

see evidence for the weak absorption feature detected by van der Tak

et al. (2012a) at 5.5 km s −1 – and ascribed by them to absorption by

foreground atomic gas – presumably because of the more limited signal

to noise ratio (S/N) in our data, that has been revealed by van der Werf

et al. (2013) in the HI counterpart. The strongest absorption feature

peaks at 5 km s −1 that is a few km s −1 broad. Position 1, toward the

HII region (top panel in Fig. 3.3) reveals an HF emission line peaking

at 8.5 km s −1 and a width of 3.5 km s −1 . The HF profile toward the

molecular cloud, position 3, peaks at 10 km s −1 (Fig. 3.1), similar to the

main component at the peak of the HF emission, that is, position 2.

The velocity at position 1 corresponds to the velocity of the [C ii]

158 µm line (9 km s −1 ) rather than the CO background gas (10 km s −1 ;

Pabst et al. (2019)). Hence, the HF emission originates in the PDR evaporative

flow from the background molecular cloud as traced by the [C ii]

3 Figure 3.2 is only available in electronic form at the CDS via anonymous

ftp to cdsarc.u-strasbg.fr (130.79.128.5) or via http://cdsweb.u-strasbg.fr/cgibin/qcat?J/A+A/

116

3.4 Results

.......................................................................

3

Figure 3.3: Upper panel shows HF spectrum toward HII region at position

1 and Gaussian fit, which is in red. The middle panel, position 2,

shows the spectrum at HF peak, which has also been studied by van der

Tak et al. (2012a). The components of HF lines is given in Figure 3.7. Finally,

the bottom panel, position 3, shows the spectrum observed toward

the molecular cloud.

emission. The typical width of the HF emission is ∼4 km s −1 and does

not vary systematically with position across our map (see Figure 3.4).

Hence, the HF emission is likely associated with interclump gas, which

typically has ∼4–5 km s −1 wide emission lines (Nagy et al. 2013). In

contrast, the width of emission lines originating in the dense clumps is

typically ∼2–3 km s −1 .

117



.......................................................................

10.5

Dec (J2000)

-5°24'30.0"

25'00.0"

30.0"

26'00.0"

10.0

9.5

9.0

8.5

8.0

5 h 35 m 24.0 s 22.0 s 20.0 s 18.0 s 16.0 s

RA (J2000)

5.0

V [km s 1 ]

3

Dec (J2000)

-5°24'30.0"

25'00.0"

30.0"

26'00.0"

4.5

4.0

V [km s 1 ]

3.5

3.0

2.5

5 h 35 m 24.0 s 22.0 s 20.0 s 18.0 s 16.0 s 2.0

RA (J2000)

Figure 3.4: Upper panel: The map of the central velocity of the HF.

There are two velocity component of the HF in the strip map. The V

= 10.7 km s −1 component is moving with the Orion Bar itself since it

has same velocity distribution. Bottom panel: FWHM map of HF J =

1 → 0 which represents a distribution of the width of 4 km s −1 .

118

3.5 Analysis

.......................................................................

3.5 Analysis

The HF J = 1 → 0 transition has a critical density (10 9 cm −3 ) much

higher than the gas density (10 5 cm −3 ) in the Orion Bar. Thus the HF

line is sub-critically excited, and hence the derived column density and

abundance are sensitive to physical conditions, that are, density (n) and

temperature (T ). Therefore, we have modeled the HF lines to determine

the column density.

3.5.1 Column density

We used the RADEX non-LTE radiative transfer code that has been

developed to infer physical parameters such as temperature and density,

based on statical equilibrium calculations (van der Tak et al. 2007).

RADEX is available for public use as part of the Leiden Atomic and

Molecular Database (LAMDA; Schöier et al. 2005). The input parameters

are kinetic temperature (T kin ), gas density (n H2 ), and molecular

column density (N col ). In addition, the FWHM of the line, collisional

partners and their collisional data, and radiation field (CMB with or

without dust emission) have to be specified as input parameters.

We consider three collision partners for the RADEX models, namely

atomic H, H 2 , and electrons. We use the new rate coefficients for the HF-

H system by Desrousseaux & Lique (2018) which are provided between

10 and 500 K. Yang et al. (2015) published rate coefficients for p-H 2 with

HF for temperatures up to 3000 K. The previous coefficients for the HF-

H 2 system provided by Guillon & Stoecklin (2012) are consistent with the

more recent Yang et al. (2015) results, and hence we use the coefficients of

Guillon & Stoecklin (2012). Based on quantum mechanical calculations

of collisional cross sections for the e-HF system by (Thummel et al. 1992)

for T > 500 K, van der Tak et al. (2012a) estimated the excitation rate

by electrons for HF ∆J = 1 at T < 500 K.

For the Orion Bar, we adopt the mean gas temperature as 120 K

(Tauber et al. 1994), and the density as 10 5 cm −3 based on previous

observations (van der Tak et al. 2012a; Nagy et al. 2013). We calculated

the column density at each position in the HF integrated intensity

map iteratively to fit the observation for the construction of the column

density map in Figure 3.5 where only CMB emission, T = 2.73 K, is

considered as background emission.

3

119



.......................................................................

Dec (J2000)

-5°24'30.0"

25'00.0"

30.0"

26'00.0"

1e15

1.4

1.2

1.0

0.8

0.6

0.4

0.2

5 h 35 m 24.0 s 22.0 s 20.0 s 18.0 s 16.0 s 0.0

RA (J2000)

Ncol [cm 2 ]

3

Figure 3.5: The map of the HF column density in the J = 1 level.

Only cosmic microwave background (CMB) emission is considered as

background emission where T bg = 2.73 K.

We have also run models which include a contribution from dust,

which has a temperature between 35 − 70 K in the Orion Bar (Arab

et al. 2012). To that end, we have fitted the observed far-IR dust Spectral

Energy Distribution (SED) at different locations (see Appendix 3.12 for

the chosen positions and SEDs) and fitted those with a modified black

body (cf., Arab et al. (2012)) and used those parameters to describe

the IR radiation field in our RADEX analysis. We have investigated

the (excitation) effects of the IR radiation field. To that end we have

assembled the IR spectral energy distribution from Herschel observations

and included this in the RADEX models. The results are insensitive to

the IR radiation field because dust is highly optically thin (τ ∼ 0.02) at

three positions. Hereby, we report in Fig. 3.5 the results of our models

using only the CMB as a background radiation field (see Appendix 3.9.1

for details). RADEX calculates the optical depth for HF J = 1-0 is 9.6

at N(HF) = 10 14 cm −2 . Our models take line trapping into account as

RADEX allow us to quantify this.

120

Figure 3.6 shows how variations in the gas temperature and density

3.5 Analysis

.......................................................................

1.8

1.7

1.6

Ncol [cm −2 ]

1.5

1.4

1.3

1.2

1.1

×10 15 10 4 10 5 10 6

70 80 90 100 110 120

Temperature [K]

3

Ncol [cm −2 ]

10 15

10 14

Density [cm −3 ]

Figure 3.6: Effect of the assumed gas temperature from 70 to 120 K and

H 2 density from 10 4 to 10 5 cm −3 on the estimated column density of HF

based on the RADEX models.

121



.......................................................................

3

Figure 3.7: Sketch of the Orion Bar. HF emission is observed toward

the HII region background molecular cloud originated due to inclination

of the Orion Bar. The three example of HF spectra from 3 positions are

given in Fig. 3.3. The figure is not to scale.

affect the derived HF column density focusing on the HF peak. The

derived column density is inversely proportional to the temperature over

the range 70−120 K (see Figure 3.6). However, as the temperature of the

gas is much better constrained than the density, the main (systematic)

uncertainty in the column density is due to the uncertainty in the density.

Given the high critical density of the J = 0 → 1 line of HF, the derived

column density is inversely proportional to the density of the gas over

the relevant density range (10 4 − 5 × 10 6 cm −3 ; Figure 3.6).

3.5.2 Spatial distribution of HF

In Figure 3.8, we compare the spatial distribution of HF with other

species: [O i] 6300 (Weilbacher et al. 2015) traces the ionization front,

H 13 CN J = 1 → 0 traces dense clumps in the PDR from Lis & Schilke

(2003), and 13 CO J = 3 → 2 traces molecular gas in the PDR (Tauber

et al. 1994). For this, we use a crosscut starting from θ 1 Ori C through

the HF integrated intensity strip map in Figure 3.2. We find that the

122

3.5 Analysis

.......................................................................

HF emission peaks between the ionization front and the dense molecular

gas in the PDR (Fig. 3.8). HF has a flat intensity distribution at offsets

between 75 ′′ and 100 ′′ toward the HII region while its intensity is decreasing

toward the inner part of the molecular cloud. As evidenced by

its shifted peak velocity, the emission toward the north west of the strip

scan is likely due to the background PDR behind the HII region (Salgado

et al. 2016; Goicoechea et al. 2016). We describe the components of the

HF lines with a sketch of the Orion Bar (see Figure 3.7). The cross cut

in Fig. 3.8 clearly illustrates that the HF emission straddles the region

separating the [C ii] 158 µm and the 13 CO J = 1 → 0 emitting zones.

Distance [pc]

0.00 0.06 0.12 0.18 0.24 0.30 0.35 0.40

1.2

1.0

HF J = 1-0

[OI] 6300 A

H 13 CN J = 1-0

13 CO J = 3-2

[CII] 158 µm

H 2 v=1-0 S(1) 2.1 µm

Normalized Value

0.8

0.6

0.4

3

0.2

0.0

−0.2

0 20 40 60 80 100 120 140 160 180 200

Distance [arcsec]

Figure 3.8: The spatial distribution of different tracers along a crosscut

which was chosen over the Orion Bar where the layered structure of the

Orion Bar can be seen. The plot starts from θ 1 Ori C which is the

main ionizing member of the Trapezium stars. The spatial resolution of

HF, [O i], H 13 CN, 13 CO, [C ii] , and H 2 is 18.1 ′′ , 0.2 ′′ , 9.2 ′′ , 22 ′′ , 11.4 ′′ ,

respectively.

123



.......................................................................

3.6 Discussion

In this section, we address the observed morphology of the HF emission

in the Orion Bar. For this, we created chemical and excitation models

along the strip map.

3.6.1 Collisional excitation

3

The observed morphology of the HF map reveals a ridge of emission

that separates the peak of the H 2 and the C + emission near the front

of the PDR from the molecular emission deeper in. Moreover, the peak

of the HF emission is well displaced from the dense clumps traced in

H 13 CN. Hence, we attribute the HF emission to the interclump gas with

a typical density of 10 5 cm −3 and a temperature of 120 K (Tauber et al.

1994; Hogerheijde et al. 1995). This is supported by the rather broad

(4 km s −1 ) HF line which is characteristic for interclump gas (Nagy et al.

2013, see Section 3.4). To test this hypothesis, we now compare our

observations to the results of a PDR model.

We have run the Meudon PDR code (Le Petit et al. 2006) for a

one-dimensional, plane parallel, constant pressure model illuminated on

one side by a strong radiation field to determine the spatial distribution

of fluorine-bearing species in the PDR. The Meudon code provides the

abundances of the major species as a function of depth in the PDR. We

have used these results to determine abundances of atomic F, HF, and

CF + , using a chemical model (Neufeld & Wolfire 2009). Specifically, HF

is mostly formed in the exothermic reaction of F with H 2 and destroyed

by C + and UV photons (Fig. 3.9). The dominant reactions playing a

role in the HF abundance are:

H 2 + F

HF + hν

HF + C +

CF + + e

HF + H

H + F

CF + + H

C + F

The Meudon PDR code calculates self-consistently the temperature for

an isobaric model. The results show that the HF abundance increases at

the PDR surface between 0 < A v < 1 when atomic H is converted into

H 2 . HF becomes the major fluorine bearing species at a depth A v > 0.5

124

3.6 Discussion

.......................................................................

where it contains ∼90% of the gas phase F; that is, X(HF) = 1.8 × 10 −8

relative to H-nuclei (Fig. 3.9).

Using the calculated H, H 2 , and e abundances from the PDR model,

we have calculated the excitation of the J = 1 level of HF with RADEX

as a function of depth in the PDR (see Fig. 3.10). We focus on the range

of A v of 1.2 and 5.8 as we were only able to extract the gas temperature

from 12 CO observation of the Orion Bar (Tauber et al. 1994). We find

that the J = 1 level population is typically 0.07 within this range. This

low level population reflects the high critical density of the J = 1 → 0

transition. The level population is not very sensitive to the H-to-H 2

conversion near A v = 0.5 as both species can readily excite HF J = 1.

This is a result of a coincidental balancing of the availability of collision

partners with their collisional rate coefficients (Guillon & Stoecklin 2012;

Thummel et al. 1992; Desrousseaux & Lique 2018; Reese et al. 2005).

Deeper in the PDR, the J = 1 level population drops. Essentially, this

reflects the steep drop in temperature in the model, T ≪ E 10 /k as the

J = 1 level cannot be easily collisionally excited anymore. Anticipating

the discussion below, we note that over most of the bright HF emission

region of the PDR, excitation is mainly due to collisions with H 2 with a

small (15%) contribution by electrons. Atomic H is not important as a

collision partner as H is not abundant in regions where HF is abundant.

Using the PDR model abundance for HF and the excitation results

from RADEX, we can calculate the intensity of the HF J = 1 → 0 line.

For this calculation, we have to specify the column density of HF along

the line of sight. We adopt a line-of-sight length scale of 0.26 pc, derived

by Salgado et al. (2016) from their analysis of the IR emission from the

Orion Bar. With this length scale and our adopted density of H-nuclei,

the total column density is 8 × 10 22 cm −2 . Over much of the PDR,

the total column density of HF is thus 8 × 10 14 cm −2 . The model with

N(HF) = 8 × 10 14 cm −2 near the peak predicts a line intensity of 1.89 K

at 120 K. We have compared the integrated intensity from RADEX with

the observations at the peak of HF. Now, we only need to discuss the

drop in intensity deeper in the cloud.

The calculated model intensity distribution is compared to the observations

in Figure 3.11. With this choice for the HF column density,

we reproduce the observed intensity at the peak well. The drop in intensity

toward the surface – caused by the drop in HF abundance – is also

3

125



.......................................................................

X([CII])

X(e − )

X(H 2)

X(HF)

UV density (erg/cm 3 )

10 −2

10 −5

10 3

X

10 −8

10 −11

10 2

Tgas [K]

10 −14

10 −17

10 1

0 1 2 3 4 5 6 7

A V

10 6

X(HF)

X(F)

X(CF + )

HF/F

CF + /F

3

X

10 3

10 0

10 −3

10 −6

10 3

10 2

Tgas [K]

10 −9

10 −12

10 1

0 1 2 3 4 5 6 7

A V

Figure 3.9: Upper panel: Abundances of HF, C + , H 2 and electron with

UV density corresponding to a Meudon PDR model with a pressure of

P = 10 8 cm −3 K. The one illuminated PDR model is considered. The

radiation field of χ = 2.6 × 10 4 . Lower panel: The abundance of F

and the ratio of HF with F and CF + are given to figure how much of

F and CF + is pushed in to HF. It must be noted that X(F) denotes

the abundance of atomic fluorine while in the ratios for the total gas

phase fluorine (F + CF + + HF) abundance. The dashed magenta line

shows the gas temperature (T gas ) shown on the right-hand y-axis in both

panels.

126

3.6 Discussion

.......................................................................

well reproduced by the model. Fig. 3.11 shows the comparison of two

RADEX models with our observation. However, while the observations

show a drop in intensity deep in the cloud, the model underestimates the

observed HF intensity. In the model, this drop in intensity is a direct

consequence of the steep drop in temperature since the PDR model underestimates

the temperature at the surface (Shaw et al. 2009; Pellegrini

et al. 2009). The calculated temperature, 20 K, is much less than the

temperature derived from 12 CO observations, 40 K (Tauber et al. 1994).

We have calculated a model where we never let the temperature drop

below 40 K (Fig. 3.10) and this model reproduces the HF observations

well even in the deeper cloud.

HF J = 1 population

Distance [pc]

0.00 0.05 0.10 0.15 0.20 0.25 0.30 0.35 0.40

2 × 10 −1

10 −1

6 × 10 −2

4 × 10 −2

RADEX (H 2 + H + e)

RADEX (H 2 + e)

RADEX (H 2 )

140

120

100

80

60

Tgas [K] (Tauber et al. (1994))

3

3 × 10 −2

0.0 0.5 1.0 1.5 2.0 2.5 3.0 3.5 4.0 4.5 5.0 5.5 6.0 6.5 7.0

A V

Figure 3.10: HF J = 1 level population as a function the depth between

A v = 1.2–6, that is, gray-shaded area. The rest does not reflect proper

calculation. The J = 1 population is calculated based on the three

RADEX models. The blue line shows the model includes only H 2 as

collisional partner. The red curve shows the model consisting of H 2 and

electrons as collisional partners. The model consisting of H 2 , electron,

and atomic H does not effect the level population that indicate atomic

H is not important for HF excitation at this range. The temperature

values shown on right-hand y-axis are taken from Tauber et al. (1994).

See the text for the detailed discussion.

127



.......................................................................

Integrated Intensity [K km s −1 ]

Distance [pc]

0.00

10

0.05 0.10 0.15 0.20 0.25 0.30 0.35 0.40

9

Model with T gas from 12 CO

8

Model with Meudon T gas

HF J = 1-0 (Observation)

7

6

5

4

3

2

1

0

0.0 0.5 1.0 1.5 2.0 2.5 3.0 3.5 4.0 4.5 5.0 5.5 6.0 6.5 7.0

A V

3

Figure 3.11: Comparison of RADEX models with the HF observation.

While the green curve shows the HF observation, the orange curve show

the RADEX model we created with the temperature taken from Tauber

et al. (1994). Red curve shows a second RADEX model where we use

the temperature calculated by Meudon code. We run these models with

the same input parameters except for the temperature to figure out the

relative importance of the temperature. The temperature is warmer

than the model predict in the deep cloud. Since we are unable extract

the temperature profile near the surface from 12 CO observations because

CO is not formed, we have only focused on the decreasing profile of HF

between A v = 1.2 − 5.8, that is, gray-shaded region, for this comparison.

The rest does not reflect a proper calculation. See the text for detailed

discussion.

Our model reproduces well the observed spatial distribution of the

HF emission in the Orion Bar. The ridge of HF emission is an interplay

of two factors: the steep rise in the HF abundance when H is converted

into HF and the drop in temperature deeper in the PDR when the CO

abundance rises and gas cooling is more efficient. Namely, cooling is

dominated by CO the deep in the cloud as C + is not important anymore

because C is converted into CO. [OI] cooling is not important as the

gas temperature is too low. We conclude therefore that, qualitatively,

the HF J = 1 → 0 line measures the presence of warm dense, CO-dark

128

3.6 Discussion

.......................................................................

molecular gas. Quantitatively, the observed intensity is a strong function

of the H 2 density and the column density of HF. We emphasize that the

observations measure the HF J = 1 column density well. The total

HF column density scales then inversely with the adopted density (cf.,

Fig 3.6). Conversely, if we were to fix the total HF column density, then

we could adjust the density to reproduce the observed intensity. Our

observations cannot break this degeneracy.

3.6.2 Infrared pumping

It has been suggested that the HF line may be excited by infrared photons

through the v = 1 → 0 fundamental vibrational band at 2.55 µm given

the brightness of the Orion Bar at this wavelength (van der Tak 2012b).

We compare the vibrational pumping with the collisional excitation of

the HF J = 1 level. This mechanism is effective if

(n l B lu − n u B ul )J near−IR = n l nγ lu (3.1)

where the Bs are the Einstein coefficients for absorption and stimulated

emission, J ul the mean intensity of the near-IR radiation field, and γ lu is

the collision probability for pure rotational transitions, which depends on

the velocity of molecules in the gas and hence the kinetic temperature.

n l and n u are the number densities of HF in the lower and upper energy

state respectively, and n is the number density of collision partners in

the gas. The left-hand side of the equation gives the near-infrared net

pumping rate and the right side is the collisional excitation rate. When

the left-hand side is greater than the right-hand side, infrared pumping is

important. If we ignore stimulated emission as at this low critical density,

most of the HF molecule will be in ground state, Eq. 3.1 simplifies to,

3

J near−IR = ( A rot

)( 2hν3

A vib c 2 )( n )exp[−hν/kT k ]. (3.2)

n cr

We have used the Infrared Space Observatory (ISO) Short Wavelength

Spectrometer (SWS) spectrum of the Orion Bar (Bertoldi et al.

2000), which is labeled as D8 in the archive 4 . From the spectrum, we

estimate the surface brightness of the Bar at 2.55 µm where the HF

vibrational ground state transition lies. The aperture size of SWS is

4 https://irsa.ipac.caltech.edu/data/SWS/spectra/sws/69501409_sws.tbl

129



.......................................................................

14 ′′ × 20 ′′ , and the flux density at the D8 position is 6.16 Jy which corresponds

to a surface brightness of 9.24×10 −14 erg s −1 cm −2 Hz −1 sr −1 .

At 120 K, pumping rate equals 7.38 × 10 −11 s −1 from the left side of

Eq. 3.1. γ 01 which corresponds to γ 10 (g 0 /g 1 )exp(-hν/kT ) that is equal

to 4.43 × 10 −11 cm 3 s −1 of the HF molecule where g 0 and g 1 are the statistical

weights of the lower and upper level, respectively. The collisional

excitation (4.43 × 10 −6 s −1 ) is much bigger than the excitation by infrared

photons (7.38 × 10 −11 s −1 ). Therefore, infrared photons do not

play a role in the excitation of HF in the Orion Bar.

3.6.3 Chemical Pumping

3

The third possibility is chemical pumping, where HF is primarily formed

in the J = 1 or higher states at a reaction rate similar to its radiative

decay (van der Tak 2012b). To produce HF emission by chemical

pumping, the HF formation rate (R = k chem n(H 2 ) n(F)) must equal or

exceed the collisional excitation rate of the 1 → 0 line. The reaction rate

coefficient (k chem ) is equal to 7.78 × 10 −12 cm 3 s −1 at 120 K based on

Neufeld & Wolfire (2009). The density of F is constrained by the total

amount of fluorine, 1.8 × 10 −8 relative to H (Simón-Díaz & Stasińska

2011), that is, n(F) = 1.8 × 10 −8 × n(H 2 ) = 1.8 × 10 −3 cm −3 where we

assumed n(H 2 ) is equal to 1 × 10 5 cm −3 in the Orion Bar. Comparison

of the chemical pumping rate (7.78 × 10 −7 s −1 ) with the collisional rate

(nγ 01 = 4.43 × 10 −6 s −1 ) for HF J = 1 → 0 demonstrates that collisional

excitation is more important. Chemical pumping does not play a major

role in the excitation of the HF J = 1 level.

3.7 Summary

We have determined the most efficient excitation mechanism for HF emission

and compared its spatial distribution with other tracers in the Orion

Bar. We find that:

130

1. HF emission peaks between the ionization region and the dense

gas in the Orion Bar. The line width of HF indicates that HF

emission emerges from the interclump medium which has a density

of 1 × 10 5 cm −3 .


.......................................................................

2. Our model studies shows that the observed peak intensity and the

morphology of the emission is well reproduced by collisional excitation

by H 2 molecules with a minor contribution by electrons

(∼15%) while IR pumping or chemical pumping plays no role in

its excitation.

3. The observations reveal a bright ridge of emission that straddles

the boundary between the [C ii] 158 µm and the CO emission. This

morphology reflects the steep rise of the HF abundance near the

surface and the drop in temperature deeper into the PDR.

4. The HF J = 1 level population peaks in the region where the CO

molecule, the common tracer of H 2 , has a low abundance. Such

regions are called CO-dark H 2 gas (Madden et al. 1997; Grenier

et al. 2005). We conclude that HF emission traces CO-dark molecular

gas, especially from PDR surfaces, as H 2 has to be abundant

for the formation of HF. In other words, HF J = 1 → 0 can be

used to trace CO-dark H 2 gas between A v = 1.0–3.5 in the Orion

Bar. Studies of a wider sample of PDRs will help develop HF as

a tracer of CO-dark molecular gas and assist in the interpretation

of HF observations of luminous nearby galaxies and high redshift

galaxies.

3


Ü. Kavak wants to dedicate this paper to the memory of Kadir Kangel,

one of the biggest supporters of his academic career, who passed

away suddenly on 11 May 2019 at the age of 49. We want to thank

William Pearson for checking the language of the present paper and

Meudon PDR team, especially to Frank Le Petit and Jacques Le Bourlot,

for their help with the Meudon code. We also thank Benhui Yang

and Benjamin Desrousseaux for sharing their recent collisional data for

the HF-H 2 and HF-H systems. This paper uses Herschel-HIFI archival

data. HIFI was designed and built by a consortium of institutes and

university departments from across Europe, Canada, and the US under

the leadership of SRON Netherlands Institute for Space Research,

Groningen, The Netherlands, with significant contributions from Germany,

France, and the US. Consortium members are Canada: CSA,

131



.......................................................................

U.Waterloo; France: IRAP, LAB, LERMA, IRAM; Germany: KOSMA,

MPIfR, MPS; Ireland: NUI Maynooth; Italy: ASI, IFSI-INAF, Arcetri-

INAF; The Netherlands: SRON, TUD; Poland: CAMK, CBK; Spain:

Observatorio Astronomico Nacional (IGN), Centro de Astrobiología (CSIC

INTA); Sweden: Chalmers University of Technology – MC2, RSS &

GARD, Onsala Space Observatory, Swedish National Space Board, Stockholm

University – Stockholm Observatory; Switzerland: ETH Zürich,

FHNW; USA: Caltech, JPL, NHSC. HIPE is a joint development by the

Herschel Science Ground Segment Consortium, consisting of ESA, the

NASA Herschel Science Center, and the HIFI, PACS, and SPIRE consortia.

PACS was developed by a consortium of institutes led by MPE

(Germany) and including UVIE (Austria); KU Leuven, CSL, IMEC (Belgium);

CEA, LAM (France); MPIA (Germany); INAF/OAA/OAP/OAT,

LENS, SISSA (Italy); IAC (Spain).

3

132

3.9 Appendix

.......................................................................

3.9 Appendix

3.9.1 SEDs of Three Positions in the HF map

To determine the spatial distribution of dust temperature and column

density in the Orion Bar, we use Herschel PACS (70 µm and 160 µm) and

SPIRE (250 µm, 350 µm, and 500 µm) maps. All maps are convolved

to the SPIRE 500 µm beam size of 39 arcsec FWHM. To construct the

SED of the Orion Bar, we choose 3 positions within the HF integrated

intensity map (see Figure 3.2). The flux densities are modeled as a

modified blackbody,

I(λ) = B(λ, T d ) τ 0

(

λ 0

λ

) β

Here, T d denotes the effective dust temperature, τ 0 the dust optical depth

at the reference wavelength λ 0 , and β the dust grain opacity index. The

reference wavelength (λ 0 ) is the position of the HF 1232.476 GHz. T d

and τ 0 are free parameters. Here, we assume that the dust emission is

optically thin. The dust emissivity index (β) is fixed at 1.7 in all models

(Arab et al. 2012). We fit the fluxes with a modified blackbody at three

different positions. In front of the Bar, position 1, the fitted temperature

is 49 K and it decreases slightly to 43 K in the Orion Bar, position 2. The

temperature in the deeper cloud, position 3, is similar to the temperature

in the Bar.

We run two RADEX models at the HF peak, position 2. In the

first model, we run RADEX considering only CMB emission. For a gas

kinetic temperature of 120 K, this model predicts an intensity for the HF

J = 1 → 0 line of 1.97 K. The second model where we only added the

IR radiation field coming from dust at 50 K to CMB also predicts same

intensity for the HF J = 1 → 0 line, i.e., 1.97 K. The RADEX models

show that FIR pumping by 50 K warm dust is not important. More

detailed models have been developed by Shaw et al. (2009) involving

detailed temperature profile, but we feel that this is outside the scope

of this paper. We elected a more straightforward approach by Salgado

et al. (2016). Following Salgado et al. (2016), dust IR emission optically

thin at all positions. Subsequently, CMB emission is only used in the

models.

3

133



.......................................................................

Position 1: 05 h 35 m 18.731 s ;

5 d 24 m 41.015 s

10 12

B [ergs/cm 2 /s]

10 13

10 14

1 = 0.006

2 = 0.008

3 = 0.01

= 1.7

T = 49.0 K

10 13 (Hz)

10 12

Position 2: 05 h 35 m 21.055 s

5 d 25 m 17.511 s

10 12

3

B [ergs/cm 2 /s]

10 13

10 14

1 = 0.015

2 = 0.02

3 = 0.03

= 1.7

T = 43.0 K

10 13 (Hz)

10 12

Position 3: 05 h 35 m 22.597 s

5 d 25 m 49.631 s

B [ergs/cm 2 /s]

10 13

10 14

1 = 0.004

2 = 0.006

3 = 0.009

= 1.7

T = 43.0 K

10 13 (Hz)

10 12

Figure 3.12: SED of three positions within the HF map as labeled in the

Figure 3.2.

134

3.9 Appendix

.......................................................................

3

135

3

136

RA (J2000)

Dec (J2000)

Table 3.2: Line parameters and column densities of the spectrum at found each pixel.

∫

Tmb ∆V V LSR ∆V T mb N col N col (50 K) N col (35 K)

(h:m:s) ( ◦ : ′ : ′′ ) [K km s −1 ] [km s −1 ] [km s −1 ] [K] 10 15 [cm −2 ] 10 15 [cm −2 ] 10 15 [cm −2 ]

5:35:19.6 -5:24:24.6 3.62 ± 0.24 8.97 ± 0.09 2.63 ± 0.19 1.29 0.50 1.10 1.11

5:35:19.1 -5:24:24.6 3.62 ± 0.24 8.95 ± 0.10 2.95 ± 0.21 1.15 0.48 1.09 1.10

5:35:18.5 -5:24:24.6 4.04 ± 0.27 8.27 ± 0.13 4.22 ± 0.33 0.90 0.53 1.20 1.21

5:35:19.6 -5:24:32.8 3.44 ± 0.18 8.88 ± 0.08 2.89 ± 0.17 0.12 0.46 1.04 1.04

5:35:19.1 -5:24:32.8 3.55 ± 0.19 8.88 ± 0.08 3.16 ± 0.18 1.05 0.47 1.07 1.07

5:35:18.5 -5:24:32.8 4.04 ± 0.26 8.35 ± 0.14 4.26 ± 0.30 0.89 0.56 1.20 1.21

5:35:18.0 -5:24:32.8 4.02 ± 0.28 8.31 ± 0.16 4.32 ± 0.34 0.87 0.52 1.20 1.21

5:35:20.2 -5:24:41.5 3.95 ± 0.22 9.41 ± 0.11 4.05 ± 0.25 0.91 0.52 1.18 1.19

5:35:19.6 -5:24:41.5 3.59 ± 0.16 8.87 ± 0.08 3.44 ± 0.16 0.98 0.47 1.07 1.08

5:35:19.1 -5:24:41.5 3.73 ± 0.13 8.54 ± 0.06 3.57 ± 0.13 0.98 0.49 1.11 1.12

5:35:18.5 -5:24:41.5 3.94 ± 0.15 8.37 ± 0.07 3.79 ± 0.15 0.97 0.52 1.18 1.19

5:35:18.0 -5:24:41.5 4.09 ± 0.20 8.33 ± 0.11 4.16 ± 0.22 0.92 0.54 1.22 1.23

5:35:20.8 -5:24:50.4 2.76 ± 0.17 9.87 ± 0.07 2.60 ± 0.19 1.00 0.36 0.83 0.83

5:35:20.2 -5:24:50.4 3.15 ± 0.17 9.54 ± 0.07 2.89 ± 0.21 1.02 0.41 0.94 0.95

5:35:19.6 -5:24:50.4 3.73 ± 0.17 9.33 ± 0.08 3.73 ± 0.19 0.93 0.49 1.11 1.12

5:35:19.1 -5:24:50.4 4.14 ± 0.15 8.87 ± 0.08 4.35 ± 0.18 0.89 0.54 1.23 1.24

5:35:18.5 -5:24:50.4 4.62 ± 0.15 8.26 ± 0.07 4.14 ± 0.15 1.05 0.61 1.38 1.40

5:35:18.0 -5:24:50.4 4.98 ± 1.22 7.71 ± 0.62 3.68 ± 1.30 1.27 0.66 1.51 1.52

5:35:21.4 -5:24:59.0 7.12 ± 0.35 9.67 ± 0.11 4.73 ± 0.28 1.41 0.97 2.17 2.19

5:35:20.8 -5:24:59.0 4.55 ± 0.17 9.87 ± 0.07 3.62 ± 0.15 1.18 0.61 1.37 1.39

5:35:20.2 -5:24:59.0 3.17 ± 0.14 9.85 ± 0.07 3.40 ± 0.17 0.87 0.41 0.94 0.95

5:35:19.6 -5:24:59.0 3.52 ± 0.11 9.50 ± 0.06 3.92 ± 0.13 0.84 0.46 1.05 1.05

5:35:19.1 -5:24:59.0 3.74 ± 0.20 9.10 ± 0.10 3.87 ± 0.25 0.90 0.49 1.11 1.12

5:35:18.5 -5:24:59.0 4.42 ± 0.23 9.02 ± 0.12 4.69 ± 0.29 0.88 0.58 1.31 1.33

5:35:21.4 -5:25:07.3 8.61 ± 0.17 10.50 ± 0.04 3.86 ± 0.08 2.09 1.23 2.73 2.75

5:35:20.8 -5:25:07.3 7.09 ± 0.16 10.27 ± 0.05 4.17 ± 0.10 1.60 0.97 2.19 2.21

5:35:20.8 -5:25:07.3 5.65 ± 0.14 9.98 ± 0.05 4.28 ± 0.12 1.24 0.76 1.71 1.73

5:35:19.6 -5:25:07.3 4.08 ± 0.19 9.86 ± 0.09 4.05 ± 0.20 0.94 0.53 1.22 1.23

5:35:19.1 -5:25:07.3 3.76 ± 0.13 9.34 ± 0.08 4.27 ± 0.16 0.82 0.49 1.11 1.12

5:35:22.0 -5:25:16.0 8.66 ± 0.20 10.57 ± 0.05 3.88 ± 0.09 2.09 1.24 2.74 2.76

5:35:21.4 -5:25:16.0 9.35 ± 0.19 10.53 ± 0.04 3.87 ± 0.07 2.27 1.35 2.99 3.01

5:35:20.8 -5:25:16.0 8.93 ± 0.14 10.52 ± 0.03 3.87 ± 0.07 2.17 1.28 2.84 2.86

5:35:20.2 -5:25:16.0 8.51 ± 0.20 10.20 ± 0.05 4.41 ± 0.11 1.80 1.19 2.66 2.68

5:35:19.6 -5:25:16.0 6.51 ± 0.22 10.01 ± 0.08 4.77 ± 0.18 1.28 0.87 1.98 1.99

5:35:22.5 -5:25:25.0 5.34 ± 0.20 10.50 ± 0.08 4.00 ± 0.15 1.25 0.71 1.62 1.63

5:35:22.0 -5:25:25.0 6.06 ± 0.18 10.62 ± 0.05 3.73 ± 0.11 1.52 0.83 1.87 1.88

5:35:21.4 -5:25:25.0 7.79 ± 0.15 10.58 ± 0.04 3.88 ± 0.08 1.89 1.09 2.44 2.46

5:35:20.8 -5:25:25.0 8.72 ± 0.16 10.57 ± 0.04 3.94 ± 0.08 2.08 1.24 2.76 2.78

5:35:20.2 -5:25:25.0 9.17 ± 0.21 10.52 ± 0.05 4.19 ± 0.10 2.05 1.30 2.90 2.92

5:35:19.6 -5:25:25.0 9.14 ± 0.43 10.20 ± 0.11 4.27 ± 0.21 2.01 1.29 2.88 2.90

5:35:22.5 -5:25:33.7 4.43 ± 0.20 10.40 ± 0.09 4.13 ± 0.22 1.00 0.58 1.33 1.34

5:35:22.0 -5:25:33.7 4.99 ± 0.16 10.51 ± 0.06 3.99 ± 0.13 1.17 0.66 1.51 1.52

5:35:21.4 -5:25:33.7 5.79 ± 0.12 10.48 ± 0.04 3.94 ± 0.09 1.38 0.78 1.77 1.78


tracer for CO-dark H2 gas clouds

.......................................................................

137

RA (J2000)

Dec (J2000)


∫

Tmb ∆V V LSR ∆V T mb N col N col (50 K) N col (35 K)

(h:m:s) ( ◦ : ′ : ′′ ) [K km s −1 ] [km s −1 ] [km s −1 ] [K] 10 15 [cm −2 ] 10 15 [cm −2 ] 10 15 [cm −2 ]

5:35:20.8 -5:25:33.7 7.14 ± 0.17 10.49 ± 0.05 3.95 ± 0.10 1.70 0.99 2.22 2.23

5:35:20.2 -5:25:33.7 8.17 ± 0.21 10.55 ± 0.05 3.95 ± 0.11 1.94 1.15 2.57 2.58

5:35:23.1 -5:25:42.0 2.85 ± 0.20 10.15 ± 0.12 3.39 ± 0.26 0.79 0.37 0.84 1.85

5:35:22.5 -5:25:42.0 3.36 ± 0.17 10.29 ± 0.10 3.77 ± 0.21 0.84 0.44 0.99 1.00

5:35:22.0 -5:25:42.0 3.35 ± 0.13 10.10 ± 0.08 4.01 ± 0.17 0.78 0.43 0.99 1.00

5:35:21.4 -5:25:42.0 4.17 ± 0.16 10.39 ± 0.09 4.39 ± 0.18 0.89 0.54 1.24 1.25

5:35:20.8 -5:25:42.0 5.40 ± 0.16 10.30 ± 0.06 4.26 ± 0.14 1.19 0.72 1.63 1.64

5:35:23.7 -5:25:50.7 2.17 ± 0.33 9.63 ± 0.20 2.71 ± 0.57 0.75 0.28 0.64 0.64

5:35:23.1 -5:25:50.7 2.75 ± 0.22 9.73 ± 0.14 3.87 ± 0.38 0.68 0.35 0.81 0.81

5:35:22.5 -5:25:50.7 2.24 ± 0.13 9.83 ± 0.10 3.51 ± 0.22 0.60 0.29 0.66 0.66

5:35:22.0 -5:25:50.7 2.37 ± 0.16 10.07 ± 0.13 3.77 ± 0.29 0.59 0.30 0.69 0.70

5:35:21.4 -5:25:50.7 2.35 ± 0.17 9.91 ± 0.15 3.92 ± 0.29 0.56 0.30 0.69 0.69

5:35:23.7 -5:25:59.7 1.82 ± 0.35 9.52 ± 0.24 2.24 ± 0.64 0.76 0.23 0.54 0.54

5:35:23.1 -5:25:59.7 1.75 ± 0.31 9.52 ± 0.21 2.26 ± 0.65 0.73 0.22 0.52 0.52

5:35:22.5 -5:25:59.7 1.91 ± 0.17 9.35 ± 0.15 3.40 ± 0.35 0.53 0.24 0.56 0.56

5:35:22.0 -5:25:59.7 1.35 ± 0.15 9.71 ± 0.17 2.87 ± 0.38 0.44 0.17 0.39 0.39

5:35:21.4 -5:25:59.7 2.33 ± 0.28 9.82 ± 0.20 3.82 ± 0.42 0.57 0.30 0.68 0.69

5:35:23.1 -5:26:07.8 2.06 ± 0.27 9.97 ± 0.26 3.48 ± 0.51 0.56 0.26 0.60 0.60

5:35:22.5 -5:26:07.8 2.08 ± 0.28 9.35 ± 0.20 3.40 ± 0.49 0.58 0.27 0.61 0.61

5:35:22.0 -5:26:07.8 1.78 ± 0.25 9.18 ± 0.26 3.55 ± 0.52 0.47 0.22 0.52 0.52

3.9 Appendix

.......................................................................

3



.......................................................................

3

138

Chapter 4

Breaking Orion’s Veil bubble

with fossil outflows

Ü. Kavak, J. Goicoechea, C. H. M. Pabst, J. Bally, F. F. S. van der Tak,

and A. G. G. M. Tielens (A&A, submitted) 1

4.1 Abstract

4

The role of feedback in the self-regulation of star formation is a fundamental

question in astrophysics. The Orion Nebula is the nearest site of

ongoing and recent massive star formation. It is a unique laboratory for

the study of stellar feedback. Recent SOFIA [C ii] 158 µm observations

revealed an expanding bubble being powered by stellar winds and ionization

feedback. We have identified a protrusion-like substructure in the

Northwest portion of the Orion Veil Shell that may indicate additional

feedback mechanisms that are highly directional. Our goal is to investigate

the origin of the protrusion by quantifying its driving mechanisms.

We use the [C ii] 158 µm map of the Orion Nebula obtained with the up-

GREAT instrument onboard SOFIA. The spectral and spatial resolution

of the observations are 0.3 km s −1 and 16 ′′ , respectively. The velocityresolved

[C ii] observations allow us to construct position-velocity (pv)

diagrams to measure the morphology and the expansion velocity. For

the morphology, we also use new observations of 12 CO and 13 CO J =

1 Submitted in Astronomy & Astrophysics

139

CHAPTER 4: Breaking Orion’s Veil bubble with fossil outflows

.......................................................................

4

2-1 (to trace the molecular gas), Spitzer 8 µm (to trace the far-UV illuminated

surface of photodissociation regions), and Hα (to trace the

ionized gas). For the kinematics, we performed line-profile analysis of

[C ii] , 13 CO, and 12 CO at twelve positions covering the entire protrusion.

To quantify the stellar feedback, we estimate the mass of the protrusion

by fitting the dust thermal emission. We compare the kinetic energy

with the stellar wind of θ 1 Ori C and the momentum of the outflows of

massive protostars to investigate the driving mechanism of the protrusion.

The pv diagrams reveal two half-shells expanding at velocities of

+6 km s −1 and +12 km s −1 . We find that the protrusion has a diameter

of ∼1.3 pc with a ∼45 M ⊙ shell expanding at +12 km s −1 at the northwestern

rim of the Veil. The thickness of the expanding shell is ∼0.1 pc.

Using the mass in the limb-brightened shell and the maximum expansion

velocity, we calculate the kinetic energy and the momentum of the

protrusion as ∼7 × 10 46 erg and 540 M ⊙ km s −1 , respectively. Based on

the energetics and the morphology, we conclude that the northwestern

part of the pre-existing cloud was locally perturbed by outflows ejected

from massive stars in the Trapezium cluster. This suggests that the

protrusion of the Veil is the result of mechanical rather than radiative

feedback. Furthermore, we argue that the location of the protrusion is a

suitable place to break the Orion Veil. We conclude that the outflows of

massive protostars can influence the morphology of the future HII region

and even cause breakages in the ionization front. Specifically, the interaction

of stellar winds of main-sequence stars with the molecular core

pre-processed by the protostellar jet is important.

140

4.2 Introduction

.......................................................................

4.2 Introduction

Massive stars have luminosities larger than 10 3 L ⊙ , corresponding to

a spectral type of B3 or earlier, and have stellar masses higher than

8 M ⊙ . The formation of massive stars is far less understood than that

of low-mass stars (< 8 M ⊙ ; see reviews by Tan et al. 2014; Motte et al.

2018). Forming massive stars differ from forming low-mass stars in several

ways. Their Kelvin-Helmholtz times are much shorter owing to their

much higher luminosities. They tend to form in dense clusters and exhibit

a higher multiplicity fraction (Motte et al. 2018). While accreting

at high rates, massive stars growing through 10 to 15 M ⊙ develop extended

photo-spheres resembling red giants (Hosokawa & Omukai 2009).

Recent studies examine the formation of massive stars and its similarity

to low-mass star formation by searching ubiquitous phenomena found

in low-mass star-forming regions (such as disks, jets, and outflows in

the scenario of disk-mediated accretion; see Beuther et al. 2002a; López-

Sepulcre et al. 2010; Sánchez-Monge et al. 2013d; Cesaroni et al. 2017;

Purser et al. 2018; Sanna et al. 2018; Kavak et al. 2021). Massive stars,

in contrast to low-mass stars, reach their main-sequence luminosity while

still embedded in accreting a natal cloud of gas and dust (Hosokawa &

Omukai 2009; Kuiper et al. 2011). A massive protostellar embryo heats

and ionizes the gas of its surrounding envelope with Extreme Ultraviolet

photons (EUV; E>13.6 eV), creating an HII region (Spitzer 1978).

Young massive stars are surrounded by ultracompact (UC) HII regions

with size < 0.1 pc and density > 10 4 cm −3 (Churchwell 2002).

The gas in the UCHII region is photoionized and heated by EUV photons

leading to an increase in gas pressure. This highly pressurized gas

causes the HII region to expand until it reaches an equilibrium Strömgen

sphere with a much lower gas density (Newman & Axford 1968). In the

standard model of HII region evolution (Spitzer 1978), the thermal pressure

of the plasma drives a D-type shock into the surrounding neutral

medium that sweeps-up a dense, expanding shell which traps the ionization

front or photodissociation region (or PDR; see review by Tielens

& Hollenbach 1985a; Hollenbach & Tielens 1997; Wolfire et al. 2003).

HII regions are mainly classified on the basis of their size and internal

density (Kurtz 2005), which span orders of magnitude in size (from 0.02

to 100 pc) and density (from 10 to 10 6 cm −3 ). In addition, HII regions

4

141


.......................................................................

4

are associated with interstellar bubbles due to their spherical morphology.

The mid-IR Galactic Legacy Infrared Mid-Plane Survey Extraordinaire

(GLIMPSE), obtained with NASA’s Spitzer Space Telescope, revealed

parsec-sized bubbles throughout the Galactic plane (Churchwell

et al. 2006) 2 . Krumholz & Matzner (2009) showed that bubble expansion

driven only by ionized gas is insufficient and that other mechanisms

than the pressure of the photoionized gas are needed to reproduce giant

molecular clouds (GMCs).

Stellar feedback implies the injection of energy, momentum, and mass

into the interstellar medium (ISM) by massive stars. This feedback is a

combination of ionizing radiation, radiation pressure, stellar winds, and

supernovae on various spatial scales (from ∼1 to ∼100 pc) and dynamical

timescales (from 10 4 to 10 6 years). Without stellar feedback, the

temperature of interstellar matter drops rapidly, and as a consequence

of this cooling, new stars form rapidly by consuming the available gas

content in the Galaxy (Kereš et al. 2009; Naab & Ostriker 2017; Lopez

et al. 2014). By heating up the gas and removing angular momentum in

star-forming regions, stellar feedback plays a key role in preventing this

‘cooling catastrophe’ in the evolution of galaxies in which star formation

occurs and dispersal of cold gas in molecular clouds (Ceverino & Klypin

2009; Walch et al. 2012; Genzel et al. 2015).

Feedback processes are divided into momentum- and energy-driven

mechanisms which have different efficiencies in terms of energy input

and time ranges (Fierlinger et al. 2016). For example, although feedback

from supernovae could provide enormous energy input that can shape

the content of galaxies on large scales (10−100 pc), pre-SN feedback

is also crucial to reproduce the properties of GMCs (Fujimoto et al.

2019; Olivier et al. 2021). From a theoretical point of view, quantifying

the relative influence of stellar feedback in detail is individually possible

and still hotly debated (Naab & Ostriker 2017; Gatto et al. 2017; Haid

2 While many HII regions are seen as bubbles, there are many bubbles that do not

contain HII regions. These can be driven by soft, non-ionizing UV, stellar winds, or

radiation pressure. HII regions may simply be the result of thermal instabilities or

fossil cavities created by now extinct energy and momentum sources such as protostellar

outflows, long-gone supernovae, or faded HII regions whose ionizing sources

have evolved off the main sequence. Throughout the paper, we presume that the

Orion Nebula is mainly blown-up by stellar winds from the Trapezium stars (Pabst

et al. 2019).

142

4.2 Introduction

.......................................................................

250

−4 ◦ 40 ′

NGC 1977

200

−5 ◦ 00 ′

200

175

Dec (J2000)

−5 ◦ 00 ′

20 ′

M43

Trapezium Stars

Orion Bar

∫

Tmb dv [K km s −1 ]

150

100

Dec (J2000)

10 ′

20 ′

150

125

75

∫ Tmbdv [K km s −1 ]

100

50

M42

50

30 ′ RA (J2000)

5 h 35 m 00 s 34 m 30 s 00 s 33 m 30 s 00 s

25

0

40 ′ RA (J2000)

5 h 37 m 36 m 35 m 34 m 33 m

0

Figure 4.1: Left: The integrated (between −5 and +14 km s −1 ) intensity

[C ii] 158 µm map of the Orion Molecular Cloud observed by upGREAT

receiver on board SOFIA. The position of NGC 1977, θ 1 Ori C, M42,

M43, and the Orion Bar PDR are labelled. The green box shows the

extracted region from the map including the area of interest for this

study, the protrusion. Right: Close-up view of the protrusion extracted

from the map on the left.

4

et al. 2018). In the last two decades, observational studies have also

demonstrated that feedback mechanisms have an important role in the

dynamics of star-forming regions (Lopez et al. 2011; Naab & Ostriker

2017).

Wind bubbles produced by stars of spectral-type earlier than B2 are

described by Castor et al. (1975) and subsequently studied analytically

by Weaver et al. (1977). However, the expansion of the bubbles, in other

words, their main driving feedback mechanism and the underlying physical

process, are poorly understood, but are studied by simulations, which

are capable of incorporating several types of feedback mechanisms individually

(Walch et al. 2012; Haid et al. 2018). From observations, it has

been difficult to assess the relative contribution of feedback mechanisms

to bubble expansion.

Most commonly, the neutral gas in the shells that confine these bubbles

is translucent to far-UV (FUV) dissociating radiation, thus they

host little CO to be detected (e.g., Goicoechea et al. 2020) because CO

143


.......................................................................

4

is readily dissociated at low A V . In addition, most stars lie in the atomic

of ionized phases of the ISM and not in molecular clouds. Thus their

feedback mostly impacts atomic or ionized gas not traced by molecules

such as CO, as well as CO-dark H 2 gas (Grenier et al. 2005). To date, a

few alternative tracers have been reported to probe the CO-dark H 2 gas

(e.g., CF + J = 1-0 by Guzmán et al. (2012b), HF J = 1-0 by Kavak et al.

(2019)). However, both species produce faint emission lines and require

long integration times in the various regimes of the ISM. In addition to

these tracers, [C ii] has been proposed as a more suitable tracer because

its fine-structure transition ( 2 P 3/2 → 2 P 1/2 at 158 µm or 1.9 THz, i.e.,

∆E/k B = 91.2 K) is the main cooling agent of the neutral interstellar

gas (Hollenbach et al. 1991; Bennett et al. 1994). Also, [C ii] is an excellent

tracer of both neutral and weakly ionized phases of the ISM, the

[C ii] 158 µm line is an ideal tracer of many types of feedback mechanisms

powered by stars. The [C ii] line is also one of the brightest

lines in PDRs and 30% of total [C ii] emission in the Galaxy comes from

dense FUV-illuminated gas (Bennett et al. 1994; Pineda et al. 2014).

Moreover, velocity-resolved observations of the [C ii] line are an excellent

probe of the kinematic and physical conditions of extended PDR

gas (Goicoechea et al. 2015), in our case, bubble shells. Unfortunately,

its rest-frame emission is not accessible from ground-based observatories.

With the upGREAT instrument onboard SOFIA, it is possible to observe

this transition from the stratosphere (Risacher et al. 2018). Therefore,

[C ii] observation of regions with a range of massive star formation activity

with stars of different spectral types will provide invaluable input for

simulation of the Galaxy evolution (see SOFIA/FEEDBACK Survey 3 ;

Schneider et al. 2020).

Orion’s Veil (Veil for short) is a series of foreground layers of gas

and dust lying in front of the Trapezium stars along the line of sight

towards the Orion Nebula (O’Dell 2018; Abel et al. 2019). The Veil is a

unique laboratory to study the relative effects of feedback mechanisms,

as its proximity allows us to resolve the bubbles in the Orion Molecular

Cloud (OMC) spatially and spectrally. Recent SOFIA [C ii] 158 µm

3 FEEDBACK is a SOFIA (Stratospheric Observatory for Infrared Astronomy)

legacy program dedicated to study the interaction of massive stars with their environment.

It performs a survey of 11 galactic high mass star-forming regions in the

158 µm (1.9 THz) line of [C ii] and the 63 µm (4.7 THz) line of [O i].

144

4.3 Observations

.......................................................................

observations of the Veil focusing on the large scale emission and dynamics

have shown that stellar winds have swept up the surrounding material

and created the Veil shell, a half-shell of neutral gas and a mass of

∼1500 M ⊙ that expands at ∼15 km s −1 (Pabst et al. 2019, 2020). They

also find that stellar winds are more effective in disrupting OMC−1 than

photo-ionization, evaporation, or even a future supernova explosion. The

stellar wind is shocked, creating a hot plasma observed in X−rays with

Chandra (Güdel et al. 2008). The high pressure of this hot plasma has

driven a shock into the environment that has swept up a dense, expanding

shell of gas. In this paper, we zoom into a specific expanding structure

at the north-west of the Veil using [C ii] observations. This protrusion

is clearly seen in Herschel PACS (70 and 160 µm) and SPIRE (250,

350, and 500 µm), and in Spitzer 8 µm emission images. Moreover,

there is bright emission in the Hα map following a similar morphology

of the limb-brightened shell as seen in the mid- and far-IR, PAH, dust

emission, and [C ii] maps. In this study, we investigate the origin of the

protrusion using velocity-resolved SOFIA [C ii] maps and compare them

to the dust, CO and PAH emission. Finally, we use the energetics of the

protrusion to assess the driving mechanism.

We organize the paper as follows. In Section 4.3 we describe the

observations of [C ii] , 12 CO, and 13 CO as well as dust emission. In

Section 4.4 we derive observational results on the general morphology,

emission features, stars (YSO and early O−, B−, and A−stars) in the

Veil. Section 4.5 contains a detailed analysis of the morphology, the

expanding shell and its velocity, and calculations of the kinetic energy of

the protrusion. Finally, we discuss whether or not the Veil is breached

at the location of the protrusion in Section 4.7.

4

4.3 Observations

4.3.1 [C ii] Observations

The observations were conducted with the Stratospheric Observatory for

Infrared Astronomy (SOFIA), which is an airborne observatory project of

the US National Aeronautics and Space Administration (NASA), and the

German Aerospace Centre (DLR). SOFIA is a modified aeroplane of the

type Boeing 747-SP, which carries a telescope with a diameter of 2.7 m

145


.......................................................................

4

in the rear fuselage (Young et al. 2012). By flying up to 45000 ft, SOFIA

makes it possible to observe at frequencies blocked by the atmosphere

from the ground. A large part of the spectrum at far infrared (FIR)

frequencies (1-10 THz) becomes accessible. At the same time, a few

molecular species (H 2 O, O 3 ) in the Earth’s atmosphere still block FIR

radiation at certain frequencies (Risacher et al. 2016).

The data were collected with the German REceiver for Astronomy at

Terahertz Frequencies (upGREAT) Instrument onboard SOFIA (Risacher

et al. 2018) for the Large program of the C + SQUAD led by A. G. G. M.

Tielens. GREAT is a heterodyne array receiver with 21 pixels. At the

time of the observations it was 2 × 7 LFA plus 1 × 7 pixel HFA. 2 × 7-

pixel sub-arrays with a hexagonal layout are designed for low-frequency

array receiver (LFA) with dual-band polarization. These cover the 1.83-

2.07 THz frequency range where the [C ii] 158 µm and [O i] 145 µm lines

can be found. The other hexagonal 7-pixel array is located in the highfrequency

array (HFA) that covers the [O i] 63 µm line. The GREAT

instrument uses local oscillators (LO) to achieve high spectral resolution

(ν/∆ν = 10 7 ). An area of about 1 square degree in Orion was surveyed

in the [C ii] 1.9 THz line (cf. Fig. 4.1; Pabst et al. 2019). The native

spectral resolution of the map is about 0.04 km s −1 . The final data is

resampled to 0.3 km s −1 to achieve a better signal-to-noise ratio. The

final rms noise (in T mb ) is 1.14 K in 0.3 km s −1 velocity channels. The

spatial resolution of the map is 16 ′′ , which corresponds to 0.03 parsecs

at the distance of Orion, 414 pc 4 (Menten et al. 2007). The data cube

is made at LSR velocities between −50 and +50 km s −1 . The [C ii]

emission mostly appears between −10 and +15 km s −1 in the entire

cube. More detailed information about the observations has been given

in Pabst et al. (2019).

We extract the [C ii] observations within the green box from the map

presented in Fig. 4.1. The map is centered on an arbitrary point, that

is, α = 05 h 34 m 17.77 s , δ = -05 ◦ 20 ′ 03.89 ′′ (J2000) and covers the entire

protrusion at the north-east of Veil (Fig. 4.1).

4 We use 414 pc provided by Menten et al. (2007) as the distance. The Orion

Molecular cloud does show a substantial distance gradient (Großschedl et al. 2018)

but that is on a much larger scale and not relevant for our paper.

146

4.3 Observations

.......................................................................

Figure 4.2: Images of Orion’s protrusion at different wavelengths and angular

resolutions. The observed transition or frequency is given for each

panel. [C ii] , 12 CO (2-1), and 13 CO (2-1) observations are integrated

between −5 and +14 km s −1 .

4.3.2 Molecular Gas Observations

4

We use new 12 CO J = 2-1 (230.5 GHz) and 13 CO J = 2-1 (220.4 GHz)

line maps taken with the IRAM 30m telescope. These data are part of

the Large Program ‘Dynamic and Radiative Feedback of Massive Stars’

(PI: J. R. Goicoechea). This project uses the old CO HERA and the

new EMIR observations of the Orion Nebula. Goicoechea et al. (2020)

describes how the old HERA and the new EMIR CO maps were merged.

The last data relevant to this study were acquired during 2020. We

extract the same region indicated in Fig. 4.2 from the original CO-cubes.

The line intensities are presented in main-beam temperature (T mb ) for

both CO observations. In order to compare with the velocity-resolved

[C ii] map, we smoothed the 12 CO (2-1) and 13 CO (2-1) data to the

angular resolution of the SOFIA [C ii] maps of 16 ′′ . The average rms

noise level in these maps is 0.20 K in 0.41 km s −1 velocity channels.

A more detailed description of the CO observations can be found in

Goicoechea et al. (2020).

147


.......................................................................

4.3.3 Ionized Gas Observations

We use the Hα images of the calibrated ESO/Digitized Sky Survey 2

(DSS-2) image obtained at the ESO/MPI 2.2-m telescope at La Silla (Da

Rio et al. 2009). The Orion Nebula has been observed on two different

nights with the same observing strategy. After combining the dithered

exposures, the final map has been created after trimming to the overlapping

area. In the final map, the surroundings of the Trapezium stars are

saturated but no saturation is seen in our region of interest. We extract

the same region as indicated in Fig. 4.1 to trace ionized gas with the

Hα map within the protrusion. The trimmed Hα map we use is given in

Fig. 4.13.

4.3.4 Far-IR photometric observations

4

We use the archival Herschel images of the dust thermal emission for

comparison to the [C ii] data, and in particular use this to estimate the

mass of dust (and gas) associated with the protrusion. The Orion molecular

clouds have been observed as part of the Gould Belt Survey (André

et al. 2010) in parallel mode using the Photoconductor Array Camera

and Spectrometer ((PACS), Griffin et al. 2010) and Spectral and Photometric

Imaging Receiver ((SPIRE), Poglitsch et al. 2010) instruments

on-board Herschel. We use the photometric images of PACS at 70 µm,

100 µm, and 160 µm, and of SPIRE at 250 µm and 350 µm. Because

of the limited spatial resolution, we refrain from using the longest wavelength

SPIRE band at 500 µm in the comparison of the dust emission

with the SOFIA [C ii] emission. Inspection of the 350 µm map reveals

that omission of the 500 µm data does not compromise our analysis. We

give more details about the model for fitting the Herschel fluxes and the

results of the spectral energy distribution (SED) fitting in Section 4.5.

A comparison between the [C ii] and Herschel maps shows that the

shorter wavelengths have almost the same morphology, which clearly represents

FUV-heated warm dust in the protrusion (see Fig. 4.2). However,

faint emission, which could be physically connected to the protrusion itself,

appears to the NW of protrusion (see Fig. 4.3). This component

is also visible in most of the maps in Fig. 4.2. Unfortunately, our [C ii]

observations do not cover this component.

148

4.3 Observations

.......................................................................

Outflow-1

Protrusion

(Second shell)

Outflow-2

Outflow-3

First shell

Trapezium Stars

Weak emission

Orion Bar

Veil’s Wall

KH-instabilities

4

Orion Bar’s

extension

Figure 4.3: Schematic picture (almost to scale) of the protrusion with

apparent structures as seen in our data. Outflows 1, 2, and 3 can be seen

in the WISE image shown in Fig. 4.15. Red and blue lines show redand

blue-shifted structures in the [C ii] data, respectively. The lightgreen

area indicates weak emission in Herschel maps. KH-instabilities

indicates Kelvin-Helmholtz instabilities reported by Berné et al. (2010).

149


.......................................................................

4.3.5 Mid-IR Observations

We also make use of the Wide-field Infrared Survey Explorer (WISE)

map of the Extended Orion Nebula 5 (EON; see also Fig. 4.15). Blue

represents emission at 3.4 µm and cyan (blue-green) represents 4.6 µm,

both of which come mainly from hot stars. Relatively cooler objects,

such as the dust in the nebulae, appear green and red. Green represents

12 µm emission and red represents 22 µm emission. The field of view of

the image is 3 ◦ × 3 ◦ which covers the Veil and the extended emission

coming from the dust. We trimmed the map to show a few striking jetlike

structures that are present near the protrusion to the northeast of

the Trapezium cluster.

To trace the FUV-illuminated surface of PDRs, we use the Spitzer

8 µm image (see Fig. 4.2). As in all observations, we extract the same

region from the 8 µm image for further analysis.

4.4 Results

4

Figure 4.2 shows the integrated intensity map of the protrusion. The

protrusion is clearly seen in Herschel PACS 70 and 160 µm and SPIRE

500 µm images. We show three representative dust emission maps in

Fig. 4.2 that trace the emission of dust heated by the Trapezium stars to

∼40 K. We also use the 12 CO and 13 CO J = 2-1 observations to identify

CO molecular gas exposed to intense FUV radiation. To confirm the

location of PDRs, we overlay the Spitzer 8 µm emission produced by

PAHs on the [C ii] map in the right panel in Figure 4.12. We see that

the [C ii] emission has a similar distribution as the 8 µm emission map

at the bottom and along the arm-like structure of the protrusion. We

identify all structures in a schematic in Fig. 4.3.

We compare the Hα emission with [C ii] to trace the ionized gas emission

within the protrusion. The outlines of the protrusion are also quite

apparent in Hα. To find the possible driving star/source in the protrusion,

we show young stars and protostars detected with IRAC/Spitzer

(green circles in Fig. 4.13; Megeath et al. 2005, 2012). In addition, we

searched for O−, B−, and A−stars within a 0.5 ′ circle around the Veil

5 The WISE map of EON can be retrieved via: http://wise.ssl.berkeley.edu/

gallery_OrionNebula.html

150

4.4 Results

.......................................................................

and listed 54 stars in Table 4.7. This table consists of the ID and object

name of the stars, coordinates in RA and Dec (in degree units), spectral

type, and object type 6 . The closest star to the protrusion is an A3

star (star 39 in Table 4.7) which has a luminosity of 14 L ⊙ and a mass

of about 2.0 M ⊙ . We think that this star is insufficient to ionize the

surrounding gas and cause a protrusion because these type stars have

low effective temperature (T eff ) and ionizing luminosity (Q i ). Thus, we

find no nearby powerful star that could ionize the gas or locally affect

the shell or Veil in the north-west (see Section 4.5.5 for detailed analysis)

and hence, the ionizing photons from the Trapezium cluster must be able

to reach this surface almost unimpeded.

Perusal of the individual channel maps (see Fig. 4.17) reveals that

the protrusion is particularly noticeable in the velocity range of −3 to

+14 km s −1 in the [C ii] observations. It is clearly offset from the main

[C ii] emission associated with the OMC−1 core at +9 km s −1 . In the

12 CO J=2-1 velocity channel maps, the protrusion does not appear as in

the [C ii] map (see Fig. 4.17). We find that the protrusion seen in [C ii]

map does not consist of CO. Unlike the [C ii] map, the protrusion does

not appear in the 12 CO J = 2-1 velocity channel maps (−0.8 km s −1 in

Fig. 4.17) associated with the boundary of the Veil. On the other hand,

12 CO J = 2-1 shows a protrusion-like structure at higher velocities (12-

13 km s −1 ) than OMC−1 (see Fig. 4.17). However, it is not associated

with our protrusion and has been identified with an expanding shell

identified in CARMA CO J = 1-0 observations (Feddersen et al. 2018).

They argued that Bruno 193 − an F9IV star at the geometric center −

is driving this CO bubble. This bubble is thought to be embedded in the

OMC−1 cloud behind the Veil. Based upon the kinematic information,

we consider that this CO bubble is not related to the protrusion and this

star is insufficient to ionize the gas; the more as this star is 7 ′ (0.85 pc)

displaced from the center of our protrusion. The general morphology,

the sub-components, and expanding shells are discussed in more detail

in Section 4.5.1, 4.5.2, and 4.5.4.

4

6 For more information on object type, see http://simbad.u-strasbg.fr/simbad/

sim-display?data=otypes

151


.......................................................................

4.5 Analysis

4.5.1 Expansion Velocity

4

Guided by the velocity channel maps, we quantify the characteristics of

the protrusion in [C ii] position-velocity (pv) diagrams. We have created

pv diagrams along thirty diagonal crosscuts, which are the 30 ′′ wide white

and magenta arrows in Fig. 4.4. We illustrate the results with two pv

diagrams (cross cuts 8 and 23 in Fig. 4.4). The other pv diagrams are

presented in Figs. 4.19 and 4.20 and support the analysis presented here.

Both pv diagrams in Fig. 4.4 reveal two arc-like structures that are the

tell tale signs of two half bubbles, both expanding only towards us (see

also Fig. 4.3).

Inspection of all pv diagrams reveals two expanding shells. We fit

these two arc-like structures in the pv diagram with a least-square fit

over the chosen positions. The expansion velocity (V exp ) of the first shell

(yellow dashed line in Fig. 4.4) is V exp = 6 ± 0.2 km s −1 and the second

(white dashed line in Fig. 4.4) V exp = 12 ± 0.2 km s −1 , which indicates

the maximum expansion velocity of the outer shell. We fitted two pv

diagrams (number 8 and 23) representing the maximum expansion of

the protrusion using a simple bubble model (see Fig. 4.4). The emission

at V LSR = +9 km s −1 (i.e. the green-dashed line) seen horizontally in

both diagrams arises from the Orion cloud itself.

When we take a closer look at the [C ii] channel maps in Fig. 4.17,

we find two spatial components between −5 and +14 km s −1 . The first

component appears from −3 to +5 km s −1 . The second component is

identified between +6 and +14 km s −1 (see Fig. 4.4). We did not see

the expanding shells in the CO channel maps and pv diagrams (see also

Fig. 4.18) and only detected a CO globule (Globule #10 of Goicoechea

et al. 2020).

4.5.2 Morphology of the protrusion

Our observations (see Fig. 4.2) reveal two expanding bow-shaped cavities

in the northwest part of the Veil. The inside wall of these cavities is

ionized as shown by the Hα emission and the [C ii] , 8 µm, and 70 µm

emission trace the surrounding PDR. First, we explore the protrusion

itself, and later the ionizing star(s) and the origin of the protrusion. We

152

4.5 Analysis

.......................................................................

−5 ◦ 00 ′

200

Dec (J2000)

10 ′

20 ′

101112131415

9

8

567

4

3

2

1

30

29

28

27

26

25

24

23

22

21

20

19

18

30 ′ 17

16

5 h 35 m 00 s 34 m 30 s 00 s 33 m 30 s 00 s

RA (J2000)

175

150

125

100

75

50

25

0

∫ T mbdv [K km s −1 ]

Vlsr [km/s]

Vlsr [km/s]

20.0

10.0

0.0

Cut 8

-10.0

0.00 200.00 400.00 600.00 800.00 1000.00

Offset [arcsec]

20.0

10.0

0.0

Cut 23

-10.0

0.00 200.00 400.00 600.00 800.00 1000.00

Offset [arcsec]

8

20.0

6

10.0

4

0.0

2

Cut 8

-10.0

0

0.00 200.00 400.00 600.00 800.00 1000.00

Offset [arcsec]

8

20.0

6

10.0

4

0.0

2

Cut 23

-10.0

0

0.00 200.00 400.00 600.00 800.00 1000.00

Offset [arcsec]

Tmb [K]

Tmb [K]

Vlsr [km/s]

Vlsr [km/s]

8

6

4

2

0

8

6

4

2

0

Tmb [K]

Tmb [K]

4

Figure 4.4: Top: Selected crosscuts along the green arrows are overlaid

on the integrated [C ii] intensity map. The number of the crosscuts is indicated

at the starting point of the cut. Bottom: The middle and bottom

panels show the pv diagram generated along the magenta crosscuts (cuts

8 and 23, respectively). The pv diagram with horizontal green lines in

both panels show the [C ii] emission produced by the FUV-illuminated

surface of OMC and the arcuate white and yellow lines trace the shell

expanding at 12 km s −1 and 6 km s −1 , respectively. The remaining pv

diagrams in Fig. 4.19 and 4.20 have the same scale in both axes. A 12 CO-

PV diagram along the crosscut 23 is shown in Fig. 4.18 for comparison

with [C ii] .

153


.......................................................................

−5 ◦ 05 ′

10000

10 ′

8000

Dec (J2000)

15 ′

20 ′

25 ′

1

11

2

3

10

4

5

9

6

12

7

8

6000

4000

2000

MJy/sr

30 ′ RA (J2000)

5 h 35 m 00 s 34 m 30 s 00 s 33 m 30 s

0

20

1 2 3 4 5 6

15

10

5

4

Tmb [K]

0

20

15

10

5

7

[CII] 158 µm

13 CO (2-1)

12 CO (2-1)

8

9

10

11

12

0

−5 0 5 10 15

vLSR [km s −1 ]

−5 0 5 10 15

−5 0 5 10 15

−5 0 5 10 15

−5 0 5 10 15

−5 0 5 10 15

Figure 4.5: Upper panel: Spitzer 8 µm image of the protrusion. Red circles

indicate twelve positions that we use to extract line profiles with an

aperture of 16 ′′ . Lower panel: Velocity-resolved spectra of [C ii] (colored

in gray), 12 CO J = 2-1 (blue), and 13 CO J = 2-1 (cyan) in the direction

of protrusion for selected twelve positions in the upper panel. The

vertical, red dotted line at 9 km s −1 marks the approximate velocity

of the emission generated by the OMC and the associated star-forming

molecular cloud behind the Veil.

fit the elliptical structure of the limb-brightened shell in the channel map

at 12 km s −1 with a least-square fit to estimate the size and expansion

timescale (t exp ) of the protrusion. We find that the size of the protrusion

154

4.5 Analysis

.......................................................................

is 1.3 ± 0.1 pc from the Veil boundary to the NW direction. The minor

and major axes of the model are 0.5 ± 0.1 and 1.3 ± 0.1 pc, respectively.

The thickness of the shell we have derived is 0.1 ± 0.05 pc. We assume

an elliptical geometry to calculate the energetics of the protrusion in

Sect. 4.5.5 because the channel maps suggest an elliptical morphology.

In summary, we have examined the channel maps and determined the

size of the expanding structure to be 1.3 pc in the southeast-northwest

and 0.5 pc in the northeast-southwest direction. This ellipsoidal morphology

is already quite apparent from the 8 µm and 70 µm dust emission

maps. While morphologically, the structure resembles a half-cap in

the plane of the sky, perusal of the pv diagrams shows that in all cross

cuts, the structure starts and ends at the cloud velocity (+9 km s −1 )

even in the southeast-northwest direction (cf., cross cut 23 in Fig. 4.4).

The observed PV diagrams are reasonably well fitted by a coherent half

ellipsoidal shell with the dimensions discussed above and expanding at

+12 km s −1 .

4.5.3 Expansion Timescale

The classical way to calculate the expansion timescale (t exp ) for structures

moving perpendicular to the line-of-sight is to use the ratio between

the size of the outer shell and the maximum expansion velocity

(size/v exp ) (see also Beuther et al. 2002a; Maud et al. 2015). In Section

4.5.1, we estimate the expansion velocity as 12 km s −1 using pvdiagram

fit results. Using this expansion velocity and size (1.3 pc), t exp

we derived is ∼1.06 × 10 5 yr, which is ∼50% of the expansion timescale

of the entire Veil shell (Pabst et al. 2019, 2020).

4

4.5.4 Line Profile Analysis

Figure 4.5 shows the comparison of [C ii] 158 µm, 12 CO J=2-1, and 13 CO

J=2-1 spectra at twelve positions covering the protrusion.

12 CO and

13 CO always have the similar profile, but at different brightness. The CO

lines typically show two emission components (at +7 and +13 km s −1 )

at several positions (3, 4, 5, and 12) corresponding to the bottom of the

protrusion. The velocity separation between the two CO peaks varies

between 1−3 km s −1 . These peaks in both CO isotopologues show small

shifts (2−3 km s −1 ) to higher or lower velocities. The absence of these

155


.......................................................................

−5 ◦ 00 ′ 00 ′′

1 parsec

10 ′ 00 ′′

Dec (J2000)

20 ′ 00 ′′

1

11

2

3

10

4

5

9

8

7

6

12

30 ′ 00 ′′ RA (J2000)

5 h 35 m 00 s 34 m 30 s 00 s 33 m 30 s 00 s

−5 ◦ 00 ′ 00 ′′

1 parsec

10 ′ 00 ′′

4

Dec (J2000)

20 ′ 00 ′′

1

11

2

3

10

4

5

9

8

7

6

12

30 ′ 00 ′′ RA (J2000)

5 h 35 m 00 s 34 m 30 s 00 s 33 m 30 s 00 s

Figure 4.6: Three-color image of the protrusion. Blue emission is the

integrated emission between −5 and +3 km s −1 , green between +3 and

+12 km s −1 , and red between +12 and +15 km s −1 of the SOFIA [C ii]

158 µm (upper panel) and IRAM 12 CO (lower panel) emission maps.

White circles show the selected twelve positions in Fig. 4.5.

velocity peaks in the [C ii] line profiles indicates that the CO emission

is associated with structures deeper in OMC−1 that are not exposed to

156

4.5 Analysis

.......................................................................

FUV radiation.

In contrast, the [C ii] line shows a different behavior than CO, with

the exception of position 2. In addition to the OMC, which emits predominantly

at V LSR = +9 km s −1 (i.e., red-dotted line in Fig. 4.5), we

identify two other components on the [C ii] emission. To investigate the

origin of these components, we have integrated [C ii] emission between

−5 and +3 km s −1 (blue is first component), +3 and +12 km s −1 (green

is second component or OMC itself), and +12 and +15 km s −1 (red is

third component). Note that the first component shifts to somewhat

higher and lower velocities and that part of the profile of the first emission

structure may be confused by emission of the OMC−1 core surface

that dominates the total emission. We are therefore not able to use

a fixed integration range for this component. The integration range is

assumed based on positions 1 and 12 in Fig. 4.5. Using integrated intensity

maps, we create a three-color map of our protrusion using [C ii]

and 12 CO cubes and display them in Fig. 4.6. In the [C ii] RGB map,

relative to the background OMC−1 core, the protrusion and the other

structure are moving towards us at 9 km s −1 . Together with the OMC,

the blue component moving towards us is associated with the smaller (in

size) expanding shell that we identified in the pv diagrams in Fig. 4.19.

The presence of a red component at higher velocities (at 13 km s −1 )

than the OMC−1 core, suggests that there is a backward extension of

the Veil shell. It is possible that the Veil shell on the rear side is tilted

with respect to the background of the OMC−1 core, and sticking out of

it, allowing for extension away from us. We do note though that the extension

of the Orion Bar in the M42 HII region is also quite prominent in

this red channel and in that case, this velocity behavior could be related

to complex morphology/velocity structures within the HII region or at

the PDR/HII edges. In the 12 CO RGB map, we track several components

with velocities different from those of [C ii] . We conclude that the

limb-brightened shell of the protrusion observed [C ii] does not contain

CO and that the CO emission is associated with the molecular cloud in

the background.

4

4.5.5 Kinetic Energy and Momentum

To identify the driving mechanism of the protrusion, we calculate its

momentum and kinetic energy. For this, we follow the same methods

157


.......................................................................

Veil Shell Protrusion

size (pc) 2.7 1.3

thickness (pc) 0.5 0.1

density [× 10 3 cm −3 ] 1−10 0.1−1

E kin [10 46 erg] 250 7

Momentum (M ⊙ km s −1 ) 20000 360−540

expansion velocity [km s −1 ] 13 12

mass of neutral gas [M ⊙ ] 1500 30−45

Table 4.1: Comparison of the masses and energetics of the protrusion

with the Veil reported by Pabst et al. (2020). The protrusion size is

measured from the wall of Veil shell to the outer shell in the NW direction.

4

as in Pabst et al. (2020). This also allows us to directly compare our

results with the Veil shell (Pabst et al. 2019). To calculate the mass

in the limb-brightened shell of the protrusion we use Herschel PACS

(70 µm, 100 µm, and 160 µm) and SPIRE (250 µm and 350 µm) maps.

All maps are convolved to the SPIRE 350 µm beam size of 20 ′′ FWHM,

as this resolution is comparable to the spatial resolution of SOFIA [C ii]

. We convert the units of SPIRE maps from Jy beam −1 to Jy px −1 using

the beam areas given in the HIPE 7 manual. The flux densities at each

pixel are modeled as a modified blackbody,

I(λ) = B(λ, T d ) τ 0

(

λ 0

λ

) β

.

Here, T d denotes the effective dust temperature, τ 0 the dust optical depth

at the reference wavelength λ 0 , and β the dust grain opacity index. The

reference wavelength (λ 0 ) is 160 µm. T d and τ 160 are free parameters.

The dust emissivity index (β) is fixed at 2 in all models (Goicoechea et al.

2015; Kavak et al. 2019; Pabst et al. 2019). Maps of the fitted optical

depth and dust temperature are shown in Fig. 4.16. The statistical values

of the dust temperature which are maximum, minimum, and median are

7 The software package for Herschel Interactive Processing Environment (HIPE)

is designed to work with the Herschel data, including finding the data products,

interactive analysis, plotting of data, and data manipulation.

158

4.5 Analysis

.......................................................................

50 K, 20 K, and 26 K, respectively. The same statistics for the optical

depth at 160 µm are 2 × 10 −1 , 8 × 10 −4 , and 2 × 10 −3 , respectively.

Using an average value of the dust optical depth over the protrusion, we

calculate the hydrogen column density:

N H = τ 160

κ 160 m H

≃ 6 × 10 24 cm −2 τ 160 (4.1)

where κ 160 is the 160 µm dust opacity per H-atom 8 which is 2.3 × 10 −25

cm 2 /H-atom for R V = 5.5 (Weingartner & Draine 2001). Using these

values and the median optical depth, which is 2 × 10 −3 , we calculated

the column density N H ∼ 1.20 × 10 22 cm −2 (or a visual extinction of

A v = 8 mag) which includes contribution from the background molecular

cloud. However, we also note that the limb-brightened shell of the protrusion

seen in the [C ii] map does not appear in the 12 CO J = 2-1 map,

indicating a low column density (A v < 3 mag), in other words, a thin

expanding shell. The high column density we derived reflects a difference

in geometry. The dust emission estimate refers to the column density

along the line of sight of a limb-brightened shell. Assuming a spherical

homogeneous shell with a relative thickness of 0.1 pc, the column density

estimates will decrease by a factor five.

Assuming elliptical geometry the mass of the limb brightened shell is

given by the surface area, S, times the surface density along the line of

sight; M = S N H µ m H . With the dimensions of the ellipse and a thickness

of 0.1 pc, the surface area is calculated to be 0.13 pc 2 , corresponding

to a mass in the limb brightened shell of 18 M ⊙ . A geometric correction

factor (see Appendix 4.9.1) of 2.5 converts this then into the mass of

the [C ii] emitting shell, ∼45 M ⊙ . which is ∼3% of the mass estimate of

Veil shell (1500 M ⊙ ; Pabst et al. 2020). Using the mass estimate and the

expansion velocity (12 km s −1 ), we calculate the kinetic energy (E kin )

of [C ii] gas tracing the neutral shell to be ∼7 × 10 46 erg. Our energy

estimate is ∼3% of the kinetic energy of the entire expanding Veil shell

(Pabst et al. 2020). Also, the momentum of the protrusion would be

∼540 M ⊙ km s −1 .

4

8 https://www.astro.princeton.edu/~draine/dust/dustmix.html

159


.......................................................................

4.6 Discussion

If the protrusion is driven by stellar winds of the Trapezium stars, in

particular θ 1 Ori C, as found for the Veil, the protrusion itself should

expand like the Veil shell. However, despite that the velocity is (slightly)

less than that of the Veil, the protrusion goes far beyond the Veil wall.

Alternatively, the stellar winds could originate from another massive star

within or near the protrusion. For this, we superimpose the O−, B−, and

A-stars B-stars O-stars

500

−5 ◦ 15 ′

400

4

Dec (J2000)

30 ′

Trapezium Stars

∫ T mb dv [K km s −1 ]

300

200

45 ′ RA (J2000)

100

5 h 36 m 35 m 34 m

0

Figure 4.7: SOFIA [C ii] map of Orion with O−, B−, and A−stars found

in SIMBAD. The list of stars retrieved from the archive is given in Table

4.3. The blue, orange, and red circles are O−, B−, and A−stars,

respectively. The light-green arrow indicate the positions of the Trapezium

stars.

160

4.6 Discussion

.......................................................................

Outflow Momentum [M⊙ km s −1 ]

10 3

10 2

10 1

P protrusion

10 0

10 3 10 4 10 5 10 6

Source Luminosity [L ⊙ ]

4

Figure 4.8: Momentum of outflows from massive young stellar objects

as a proportion of the source luminosity of the cores (Maud et al. 2015).

The blue and red symbols indicate the blue- and red-shifted outflow lobe

values, respectively which are joined by a dotted line for each source. The

horizontal blue-shaded range indicates the momentum of the protrusion,

which is between 360−540 M ⊙ km s −1 . The cross at the bottom-right

shows the uncertainty for both axes.

A−stars on the [C ii] map (Fig. 4.7). There is no massive star within the

protrusion. Only two A−stars are found near the protrusion. However,

the nearest A−star does not follow the elongated morphology of the pro-

161


.......................................................................

Veil

Protrusion

10 −2

Veil

Protrusion

(a)

(d)

I [CII] [erg s −1 cm −2 sr −1 ]

10 −2

10 −3

10 −4

I [CII] [erg s −1 cm −2 sr −1 ]

10 −3

10 −4

10 2 10 3 10 4 10 5

I70 µm [erg s −1 cm −2 sr −1 ]

10 −3 10 −2 10 −1 10 0

I8 µm [erg s −1 cm −2 sr −1 ]

4

Figure 4.9: Correlation plots between the surface brightness of [C ii] with

70 µm and 8 µm, respectively. Gray dots show the Veil and colored dots

show the protrusion in both panels. We convolve and re-grid all images

to a resolution of 36 ′′ and a pixel size of 14 ′′ which allow us a proper

comparison with the Veil (see Pabst et al. submitted). In the y-axis of

each plot, all points above 3σ corresponding to 8 K km s −1 , which is

equal to 5 × 10 −5 erg s −1 cm −2 sr −1 , are shown. The lines in the graphs

show the least-square fits of various correlations. For example, the black

line is a power-law fit (using the form y = a x b ; see Table 4.2 for the

fit results of x, y, a, and b) for the Veil, and the different coloured lines

show power-law fits for the protrusion on a logarithmic scale.

trusion. The second A−star is located at a comparable distance to the

Trapezium stars. These findings force us to think of a pre-existing structure

that is now being overtaken by the expanding Veil shell. Another

way to estimate the role of the winds is to compare them with X−ray

observations, in which the hot X−ray emitting gas is traced inside the

Veil. Using X−ray observations of the Veil, Güdel et al. (2008) showed

that the X−ray emission from the ionized region indicates a hot plasma

heated to a few 10 6 K by the shocks created by the stellar winds. In

other words, the presence of X−ray emitting hot gas can be taken as

an indication of stellar winds. However, there is no X−ray observation

covering the protrusion. It should also be noted that X−ray emission is

very susceptible to extinction by foreground material (Güdel et al. 2008).

Therefore, X−ray observations may not be the best tool to investigate

the effect of stellar winds, at least in our case. Imaging of optical line

emission with the Apache Point Observatory (APO) will help us to de-

162

4.6 Discussion

.......................................................................

tect the hot plasma (T >30,000 Kelvin) inside the cavity (Bally et al.,

in prep).

The slightly lower expansion velocity of the protrusion than the Veil

and its extension beyond the boundary of the Veil argues that the protrusion

is a pre-existing structure in the OMC−1 core that is now being

overtaken by the Veil bubble. Following Bally et al. (in prep), we suggest

that this pre-existing structure is the result of fossil outflow activity

in the OMC−1 core created during the accretion phase of the massive

protostars in the Trapezium cluster. Once the protostellar jet switches

off, the cavity blown by this jet will enter the momentum conserving

phase and expand while slowing down. As θ 1 Ori C entered its main sequence

phase, its stellar wind started to blow the Veil bubble. The large

amount of momentum involved in this kinematic structure could indicate

outflow activity associated with the formation of the most massive

star. To identify the possible protostellar source(s), we use the bolometric

source luminosity and momentum of the outflows from Maud et al.

(2015). The momentum of M outflow red- and blue-shifted lobes are given

individually with red- and blue-shifted squares in Fig. 4.8, respectively.

The interpretation of the relation in Fig. 4.8 is that the jet or wind from

the most luminous protostar drives the strongest and most powerful outflows.

For the scatter of momentum values, Maud et al. (2015) argued

that it is caused either by outflow inclination angles or by multiple outflows

driven by sources within dense cores. We also emphasize that this

type of outflow activity is generally found in systems with ages less than

a few times 10 4 yr (Arce et al. 2007) and hence is a clear signature of

protostellar activity.

Using the relation in Fig. 4.8, we estimate that a massive dense shell

with a momentum of ∼540 M ⊙ km s −1 would require a luminosity of

3 × 10 4 to 3 × 10 5 L ⊙ . This corresponds to B0 to O7 type stars (cf. for

stellar parameters of O and B stars Vacca et al. 1996) and several stars

in the Trapezium region could be responsible. Likely, θ 1 Ori C, the most

massive star, is the culprit. We do notice that there are several other jetlike

morphological structures present in the 8 µm and WISE maps in the

area of the protrusion (Fig. 4.14 and 4.15). Our [C ii] observations do not

cover these structures and therefore we have no kinematic information

on their expansion. Further (deeper) studies are warranted to determine

their kinematics. Here, we recognize that these structures may indicate

4

163


.......................................................................

the presence of multiple protostellar outflows for example associated with

the several of the Trapezium star cluster. Alternatively, these jet-like

structures may reflect intermittent activity of a single, precessing object

in a binary of the Trapezium cluster. We do note that the trajectories

of these jet-like structures trace back to the Trapezium stars (Fig. 4.14).

4

The wind of θ 1 Ori C would produce a spherical bubble only if there

were no obstacles blocking the propagation of the wind and post-shock

hot plasma. However, we know that there is a dense cloud in the region

of K-H instabilities (see Fig. 4.2) containing CO whose surface is affected

by radiative feedback (maybe a wind) from the Trapezium cluster. The

H-alpha ionization front of the nebula wraps around this structure. This

also blocks the plasma flow to the west. A possible model for the NW

protrusions is that the plasma driving the Veil shell has found a path of

least resistance towards the NW. However, the slower expansion of the

protrusion forces us to think that the protrusions are fossil protostellar

outflow cavities that were powered by Orion’s massive stars prior to their

reaching the zero-age main sequence (ZAMS).

At this point, it is worth noting that the protrusion was likely created

by outflow activity when accretion in a protostar-disk structure

was accompanied by a jet/wind in the polar directions. On this basis,

it can be argued that the Trapezium stars (specifically θ 1 Ori C)

should have formed via disk-mediated accretion. This model of massive

star formation is supported by recent studies have found disks (Cesaroni

et al. 2017), outflows (López-Sepulcre et al. 2010; Sánchez-Monge et al.

2013d), and jets (Sanna et al. 2018; Kavak et al. 2019). If the protrusion

is made of fossil outflow cavities, there have to be the counter flows corresponding

to the red-shifted lobe of the northwest protrusions from the

Trapezium cluster. WISE and 8 µm images show a vague protrusion in

the opposite direction of the northwestern protrusion. Given the blueshifts

of the NW protrusions, this component should be the red-shifted

lobe (i.e., the red arrow shows the red-shifted lobe in Fig. 4.14). However,

the [C ii] emission is weak preventing us to study this red-shifted

lobe in this work. Note also that the fossil outflow activity is not related

to the explosive outflow and the H 2 fingers seen in near-IR lines (Bally

et al. 2017), as these fingers are still far (∼1.5 pc) from the boundary of

the Veil shell.

164

4.6 Discussion

.......................................................................

4.6.1 Persistence of fossil outflow cavity

The protrusion has a limited lifetime due to the photo-ablation of its

walls. Once the massive stars reach the ZAMS and begin to ionize their

surroundings, photo-ablation of the inner walls of these cavities will start

to fill their interiors with plasma. To first order, the plasma will expand

at the speed of sound in ionized gas at V [CII] = 10 km s −1 . Using our mass

estimations in Table 4.1, the surface area of the protrusion (0.385 pc 2 ),

and the incident flux of Lyman continuum photons (2 × 10 49 s −1 for

θ 1 Ori C), we can estimate the mass-loss rate of the protrusion walls and

how long the walls would survive (t sur ). The mass-loss rate is given by,

dM

dt

= f µ m H n e V [CII] R 2 (4.2)

where f is a factor of order unity depending on geometry which is taken

to be √ 3 to recover the Strömgren condition for a spherical HII region.

The plasma density (n e ) can be calculated assuming that the incident Lyman

continuum flux (L(LyC)/(4 π D 2 )) equals the recombination along

a path length (R),

4

n e = f D

[ L(LyC)

] 0.5

(4.3)

4 π α B R

where α B is the Case B recombination coefficient of H; 2.6 × 10 −13 cm 3

s −1 . The number of electron-proton recombinations per unit volume and

unit time is equal to n e n p α B . Using Eq. 4.3, we derive a plasma density

(n e ) of ∼2 × 10 3 cm −3 . The mass-loss rate from the protrusion walls is

1.8 × 10 −4 M ⊙ yr −1 . Therefore, the lifetime of the protrusion (i.e., t sur

= M/(dM/dt), where M is the mass of [C ii] emitting protrusion walls)

is ∼1.6 × 10 5 years, which is consistent with the age of the Trapezium

stars and the expansion timescale derived in Sect. 4.5.3, but not with the

age of the O9 to early B-stars below the bright Orion Bar (the θ 2 Ori A

stars) whose age is older than 10 6 years. We argue that the location of

the protrusion is an ideal place to break Orion’s Veil and ventilate its

hot plasma before a possible supernova occurs (∼5 × 10 6 years; see also

Williams & McKee 1997).

165


.......................................................................

4.6.2 Ionizing source

4

In Section 4.4, we show that the Hα emission follows a similar morphology

as the [C ii] emission. To understand the origin of ionized gas along

the limb-brightened shell, we use the Hα flux to estimate the source

of the ionizing photons. We can make an estimate for the extinction

associated with the protrusion from the thickness of the shell and the

estimated column density of the limb brightened shell. Adopting a spherical

half shell with a relative thickness of 0.1 pc, we estimate that the

column density along the line of sight is 0.2 times the column density

derived from the dust emission of the limb brightened shell, 2 × 10 21

H nuclei per cm 2 . Using the extinction curve of Weingartner & Draine

(2001), this corresponds to an extinction at Hα of 1.1 mag. Correcting

the observed surface brightness for extinction results in an intrinsic Hα

surface brightness of 525 MJy sr −1 or 2.7 × 10 −7 erg s −1 cm −2 arcsec −2 .

We converted the surface brightness into emission measure 9 (EM) using

Equation 4.4. Given a constant temperature of 8500 K obtained from

radio recombination line observations (Wilson et al. 1997),

[ ] EM

pc cm −6 = 4.197 × 10 17 × I Hα (4.4)

with Hα in units of erg s −1 cm −2 arcsec −2 . The EM we have derived as

1.40 × 10 7 pc cm −6 . We then calculated the total number of ionizing

photons (N Lyc ) emitted by the star (see Sect. 7.4.1 of Tielens 2010).

N Lyc = A × EM × 2.6 × 10 −13 (4.5)

where A is surface area in pc 2 and EM in pc cm −6 . We find 1.8 × 10 50 photons

s −1 . We measure the number of ionizing photons over a hole on the

wall of the Veil of 1 pc 2 , which is 1/16 of the total inner surface area of the

Veil. In this case, the final number of ionizing photons is 1.1 × 10 49 photons

s −1 . This indicates that the source of the ionizing photons should

be an O-type star. The only O-star in the Trapezium cluster is θ 1 Ori C,

the main ionizing star in Orion Nebula (O’Dell et al. 2017). Therefore,

we conclude that the source of the ionized gas in the protrusion should

be θ 1 Ori C.

9 The emission measure is defined as n 2 eL, where n e is the electron density and L

is the total path length in the ionized gas.

166

4.6 Discussion

.......................................................................

−5 ◦ 00 ′

−5 ◦ 00 ′

10 5

G 0

10 2

05 ′

05 ′

Dec (J2000)

10 ′

15 ′

20 ′

10 3 5 h 35 m 00 s 34 m 30 s 00 s 33 m 30 s 00 s

Dec (J2000)

10 ′

15 ′

20 ′

n 0

10 4

10 3

10 2

10 0

[cm −3 ]

25 ′

25 ′

10 1

30 ′ RA (J2000)

5 h 35 m 00 s 34 m 30 s 00 s 33 m 30 s 00 s

30 ′ RA (J2000)

Figure 4.10: The map of the incident radiation field G 0 (left) and the

density (right) of the protrusion for a face-on PDR model adopted from

Tielens (2010). See Section 4.6.3 for a more detailed discussion.

4.6.3 Correlation of Intensities

Fig. 4.9 show the pixel-by-pixel correlation of [C ii] with 70 µm and

8 µm, which we use to examine the behavioral coherence between the

Veil and the protrusion. For a proper comparison with the Veil, we

converted the units of [C ii] 158 µm 10 and Spitzer 8 µm observations

into surface brightness (I [CII] ; erg s −1 cm −2 sr −1 ). We note that there

is no correlation of the Hα and 13 CO J = 2-1 lines with the [C ii] line

intensities. The gray points in both panels correspond to the Veil and

colorful points indicate the distribution of the protrusion. The plots in

Fig. 4.9 imply the same results as the similarity in morphology in Fig. 4.2.

Fig. 4.9 shows that the protrusion behaves coherently with the Veil and

all three emission components trace the PDR.

We estimate the density of the protrusion by using the relation between

G 0 and 70 µm reported by Goicoechea et al. (2020) for Orion. G 0

is given by,

4

log 10 (G 0 ) = (0.975 ± 0.02) log 10 (I 70 ) − (0.668 ± 0.007) (4.6)

10 For [C ii] and 13 CO J = 2-1 lines, we use the following formula to calculate

the conversion factor from velocity-integrated line intensities (K km s −1 ) into surface

brightness (I [CII] ; erg s −1 cm −2 sr −1 ): I = 2k W ν 3 /c 3 . The conversion is

I(erg s −1 cm −2 sr −1 ) = 7.0 × 10 −6 W(K km s −1 ) and I(erg s −1 cm −2 sr −1 ) =

1.3 × 10 −8 W(K km s −1 ), respectively (see also Goicoechea et al. 2015).

167


.......................................................................

where I 70 is the 70 µm dust surface brightness in MJy sr −1 . The median

value of G 0 is ∼600 towards the protrusion, although a gradient can be

seen in the G 0 map (see Fig. 4.10). Using the estimate of G 0 , we calculate

the density of a face-on PDR using equation 9.4 of Tielens (2010), which

is given in Eq. 4.7. We isolate the density and express it in terms of G 0 .

G 0 ≃ 10 2 (

n 0

10 3 cm −3 ) 4/3

(4.7)

4

The resulting density map is also shown in Fig. 4.10. The density decreases

in the northwest direction from the boundary of the Veil to the

outer shell of the protrusion. We can check our gas density from the

observed [C ii] intensity using PDR models. For this purpose, we use

the intensity of the [C ii] 158 µm line emitted from the surface of an

edge-on PDR as a function of the density and radiation field based on

the PDR models 11 of Kaufman et al. (1999) adopting an average G 0

of 600 Habings. This results in an average density of 10 3 cm −3 , in

agreement with the estimates in Figure 4.10. The density along the

limb-brightened shell of the protrusion is comparable with the Veil shell

(Pabst et al. 2020) and two or three orders of magnitude lower than the

Orion Bar (Kavak et al. 2019; Pabst et al. 2020).

x y a b

The protrusion

70 µm [C ii] 1.0 (0.1) × 10 −6 83 (0.1) × 10 −1

8 µm [C ii] 2.6 (0.1) × 10 −2 80 (0.9) × 10 −2

The Veil

70 µm [C ii] 2.1 (0.02) × 10 −5 40 (0.01) × 10 −2

8 µm [C ii] 9.4 (0.02) × 10 −3 50 (0.09) × 10 −2

Table 4.2: The resulting fit coefficients for the correlations in Fig. 4.9

using a power-law function of the form y = a x b . The numbers in parentheses

are the standard deviations of the parameters.

At this point, it might be worth to note that we can calculate the

mass within the limb from the density and volume assuming an elliptical

geometry for the protrusion. We calculate the mass of the shell

11 http://dustem.astro.umd.edu/models/wk2006/cpweb.html

168

4.7 Conclusion

.......................................................................

of ∼30 M ⊙ .

section 4.5.5.

This is in good agreement with the values calculated in

4.7 Conclusion

We investigate the origin of the protrusion in the northwestern part of

the Orion Veil shell using velocity-resolved [C ii] 158 µm observations.

We find that the formation of the protrusion is caused by extinct or

previously active outflows from the Trapezium stars. This suggests that

mechanical feedback is the responsible mechanism for the formation of

the protrusion rather than radiative feedback. This is an important intermediate

stage in which fossil outflow activity influences the dynamics

of HII shells before Trapezium stars reach the supernova phase.

Moreover, SOFIA [C ii] observation of the Veil revealed that it will

break open as its expansion velocity exceeds the escape velocity of the

Orion Nebula (Pabst et al. 2019). Afterwards, the hot ionized gas inside

will escape into the surrounding medium (Pabst et al. 2019) and the

expansion will slow down. In Section 4.4, we also see that the lack of

CO detections in the protrusion indicate a low N H , or in other words, a

thin shell in the northwestern Veil. In Sect. 4.6.1, we also show that the

fossil outflow activity could cause breaks in the ionization front of the

Orion Veil because of photo-ablation from the protrusion walls. Also, the

expansion velocity of the protrusion exceeds the escape velocity (about

1−2 km s −1 ), making the protrusion a suitable place for the Veil to break

up.

The diagonal pv diagrams parallel to the direction of expansion,

in particular cuts 18, 19, and 20, show [C ii] emission that extends

somewhat beyond the protrusion. Moreover, the densities of the limbbrightened

shell are lower (a factor of up to two) at the head of the

protrusion. Outflows, particularly Outflow 3 in Fig. 4.14, appear to

be associated with the chimney-like top of the protrusion, suggesting

that the Veil shell has already been pierced here. This location could

be a suitable place for the bubble to break. Furthermore, beyond the

area mapped in [C ii] , the outflows and their extended morphology are

also seen in the Spitzer 8 µm image, the dust emission maps of Herschel

PACS 70 µm and WISE observations. Thus, future more sensitive

[C ii] observations could clarify whether or not the Veil is already broken

4

169


.......................................................................

at the location of the protrusion.


4

We want to thank Martin Vogelaar (Groningen) for his help for solving

Python programming problems and Anthony G.A. Brown (Leiden) for

retrieving the list of O−, B−, and A− stars from the SIMBAD database.

We also thank Marc William Pound and Mark Wolfire for their help on

the PDR Toolbox. Studies of interstellar dust and gas at Leiden Observatory

are supported by a Spinoza award from the Dutch Science agency,

NWO. JRG thanks the Spanish MICIU for funding support under grant

PID2019-106110GB-I00. This study was based on observations made

with the NASA/DLR Stratospheric Observatory for Infrared Astronomy

(SOFIA). SOFIA is jointly operated by the Universities Space Research

Association Inc. (USRA), under NASA contract NAS2-97001, and the

Deutsches SOFIA Institut (DSI), under DLR contract 50 OK 0901 to

the University of Stuttgart. upGREAT is a development by the MPI für

Radioastronomie and the KOSMA/Universität zu Köln, in cooperation

with the DLR Institut für Optische Sensorsysteme. We acknowledge

the work, during the C+ upGREAT square degree survey of Orion, of

the USRA and NASA staff of the Armstrong Flight Research Center

in Palmdale, the Ames Research Center in Mountain View (California),

and the Deutsches SOFIA Institut.

4.9 Appendix

4.9.1 Geometric correction Factor

The limb-brightened shell observed in different tracers is seen as an arc

of emission. If we assume that the emission seen in the dust tracer, the

[C ii] line or the CO line is proportional to the total volume, then we

need some geometry to figure out what the enhancement factor, f v , is

that scales the volume of the limb brightened part to that of the full

shell. We consider two concentric nested ellipsoids with major diameter

2C o and 2C i and minor diameter 2B o and 2B i . The protrusion is half

of this ellipsoid (see Fig. 4.11). If the cap height is h, the cap volume is

given by;

170

4.9 Appendix

.......................................................................

V cap = π 3 C2 o

( h

B o

) 2

(3Bo − h) (4.8)

The base surface area of the cap is,

( h

)

A cap = π (C o B o ) (2 − h ) (4.9)

B o B o

The volume of the cylinder is,

V cyl = 2 (B o − h) A cap (4.10)

The total volume of the outer ellipsoid is,

( 4π

)

V ell = Bo 2 C o (4.11)

3

The volume of the rim is then,

V rim = V ell − 2 V cap − V cyl (4.12)

We compare this to the volume in between the two nested ellipsoids,

V = 4π )

(B o 2 C o − Bi 2 C i (4.13)

3

4

C o

C i

Observer

h

B o

B i

h

Figure 4.11: Shell geometry

171


.......................................................................

Actually, as the ellipsoid only protrudes half out of the Veil, we should

divide all of these volumes by two. As we are really interested in V rim /V

these factors two drop. Now we have to express h in the sizes of the inner

and outer ellipsoid. The base area of the cap is equal to the surface area

of the inner spheroid.

A cap = π B i C i (4.14)

Thus, h can be found from,

A cap = π (C o B o ) ( h B o

) (2 − h B o

) = π B i C i (4.15)

For B o = 0.5 pc, C o = 1.3 pc, B i = 0.4 pc, and C i = 1.2 pc, we find

that the height of the cap would be 0.244 pc. Using this, we estimate

that the volume of the [C ii] emitting limb-brightened rim is 2.5 of the

total volume of half ellipse in Fig 4.11. In this case, the mass in the limbbrightened

shell would be between 45 M ⊙ which is in good agreement

with the mass estimation of 30 M ⊙ based on the PDR models. Finally,

the mass of the limb-brightened shell is between 30−45 M ⊙ .

4

4.10 Additional Maps

Figs. 4.12 and 4.13 show Spitzer 8 µm and Hα maps, respectively. In

both panels, blue contours are the integrated (between −5 and 14 km s −1 )

[C ii] observations. In Figs. 4.12, [C ii] traces 8 µm closely.

172


.......................................................................

−5 ◦ 00 ′ 00.0 ′′

5000

06 ′ 00.0 ′′

4000

Dec (J2000)

12 ′ 00.0 ′′

18 ′ 00.0 ′′

24 ′ 00.0 ′′

3000

2000

MJy/sr

4

30 ′ 00.0 ′′

1000

34 m

35 m RA (J2000)

5 h 33 m

Figure 4.12: Spitzer 8 µm image, which outlines the PDR surfaces. The

blue contours show the SOFIA [C ii] line integrated emission.

173


.......................................................................

Dec (J2000)

−5 ◦ 00 ′

10 ′

20 ′

20000

18000

16000

14000

12000

10000

8000

MJy/sr

4

6000

30 ′ RA (J2000)

5 h 35 m 34 m 33 m

Figure 4.13: Hα image, which traces the ionized gas in the protrusion.

The blue contours show the SOFIA [C ii] line integrated emission. Green

circles show the young stars and protostars surveyed by Megeath et al.

(2005, 2012).

174


.......................................................................

−5 ◦ 00 ′

6000

outflow 1

outflow 2

5000

Dec (J2000)

15 ′

30 ′

red-shifted lobe

Trapezium Stars

outflow 3

4000

3000

2000

MJy/sr

4

45 ′ RA (J2000)

1000

5 h 37 m 36 m 35 m 34 m 33 m

0

Figure 4.14: Spitzer 8 µm image and three potential outflows identified

toward protrusion.

175


.......................................................................

4

Figure 4.15: WISE image of the Orion Nebula provided by University

of Berkeley. Blue represents emission at 3.4 µm and cyan (blue-green)

represents 4.6 µm, both of which come mainly from hot stars. Relatively

cooler objects, such as PAHs, the dust of the nebulae, appear green

and red. Green represents 12 µm emission and red represents 22 µm

emission tracing very small grains (VSGs). The field of view (FOV) of

the original image is 3 ◦ × 3 ◦ , but we trimmed the image to show the

outflow beyond the protrusion. The original file can be retrieved via:

http://wise.ssl.berkeley.edu/gallery_OrionNebula.html

176


.......................................................................

200

Dust Temperature

34

200

Optical Depth (τ 160)

0.010

180

32

180

0.008

Dec Offset [pixel]

160

140

120

30

28

26

Kelvin [K]

Dec Offset [pixel]

160

140

120

0.006

0.004

100

80

24

22

100

80

0.002

50 75 100 125 150 175

RA Offset [pixel]

20

SED (1, 1); T = 28.4 K; τ = 3.38e-03

50 75 100 125 150 175

RA Offset [pixel]

Bν (erg/cm 2 /s/Hz/sr)

10 −12

4

10 13 Frequency (Hz)

10 12

Figure 4.16: The temperature (upper-left) and the optical depth at

160 µm (τ 160 ) (upper-right) map of dust emission, which traces the mass

of the shell. The bottom panel shows an example of SEDs from the

bottom-left of the dust temperature map.

177


.......................................................................

-3.2 km s °1 -2.0 km s °1 -0.8 km s °1 0.4 km s °1 1.6 km s °1

Globule #10

2.8 km s °1 4.0 km s °1 5.2 km s °1 6.4 km s °1 7.6 km s °1

8.8 km s °1 10.0 km s °1 11.2 km s °1 12.4 km s °1 13.6 km s °1

4

Figure 4.17: Channel map of [C ii] emission from V LSR from −3.2 to

+13.6 km s −1 overlaid with 12 CO J = 2-1 observations with white contours.

The contour levels are [3, 6, 10, 15, 20] K km s −1 . The velocity

resolution of both maps is smoothed to 0.5 km s −1 . Globule #10 which

is a bright CO emission at the velocity of −0.8 km s −1 indicates the

CO globule reported in Orion Veil (see also Fig. 4.18; Goicoechea et al.

2020).

Vlsr [km/s]

20.0

15.0

10.0

5.0

0.0

-5.0

0.00 200.00

Globule #10

400.00 600.00 800.00

Cut 23

1000.00

Offset [arcsec]

20

10

0

Tmb [K]

Figure 4.18: PV diagram of 12 CO J = 2-1 along the crosscut 23 in

Fig.4.4. Globule #10 which is a bright CO emission at 200 ′′ indicates

the CO globule reported by Goicoechea et al. (2020).

178


.......................................................................

Cut 1

Cut 2

Cut 3

Cut 4

Cut 5

Cut 6

Cut 7

Cut 8

Cut 9

4

Cut 10

Cut 11

Cut 12

Cut 13

Cut 14

Cut 15

Figure 4.19: [C ii] pv diagrams from the protrusion sliced along expansion

direction (i.e., cuts from 1 to 15 in Fig. 4.4). All diagram have same

scales as in Fig. 4.4.

179


.......................................................................

Cut 16

Cut 17

Cut 18

Cut 19

Cut 20

Cut 21

Cut 22

Cut 23

4

Cut 24

Cut 25

Cut 26

Cut 27

Cut 28

Cut 29

Cut 30

Figure 4.20: 2nd [C ii] pv diagrams from the protrusion sliced along

expansion direction (i.e., cuts from 16 to 30 in Fig. 4.4). All diagram

have same scales as in Fig. 4.4.

180


.......................................................................

Table 4.3: List of O, B, and A stars within 0.5 ′ circle which is centered

at Veil (RA: 83.6952553, Dec: -5.5075778) retrieved from SIMBAD. For

object type, see http://simbad.u-strasbg.fr/simbad/sim-display?

data=otypes. Star 39 is an A3 type star which has an luminosity of

14 L ⊙ and has mass from 1.4 to 2.1 M ⊙ on average. See Section 4.4 for

more detail.

ID Main ID RA Dec Spectral Type Object Type

(J2000) (J2000)

1 811408 HD 37000 83.795 -5.926 B3/5 Y*O

2 813903 HD 37115 83.975 -5.628 B7Ve Be*

3 812745 * iot Ori 83.858 -5.909 O9IIIvar SB*

4 813805 HD 36960 83.761 -6.002 B1/2Ib/II *

5 810070 * tet02 Ori B 83.860 -5.416 B2-B5 Y*O

6 800633 HD 37174 84.113 -5.408 B9V *

7 810062 V* V1230 Ori 83.836 -5.362 B1 Or*

8 800723 * tet02 Ori C 83.880 -5.421 B4V Or*

9 808906 Brun 328 83.666 -5.168 A0 *

10 800621 * tet01 Ori A 83.815 -5.387 B0V Ae*

11 809750 J05355545-0513556 83.981 -5.232 A0-A5 *

12 800755 V* KO Ori 83.735 -5.526 A7 Or*

13 811404 HD 37150 84.062 -5.647 B3III/IV *

14 800610 HD 294265 83.643 -5.051 A5 *

15 810606 V* V2254 Ori 83.808 -5.372 B Or*

16 804756 HD 37061 83.880 -5.267 O9V Or*

17 814861 * iot Ori B 83.860 -5.912 B8III Or*

18 811401 HD 36999 83.808 -5.826 B8(III) Y*O

19 800619 * tet01 Ori F 83.819 -5.390 B8 Em*

20 5385015 [AD95] 266 84.007 -5.553 A2-A7 *

21 803601 V* KS Ori 83.750 -5.421 A0V Or*

22 811506 HD 36918 83.704 -6.006 B8.3 *

23 800625 V* MR Ori 83.820 -5.362 A2:Vv Or*

24 805895 Brun 818 83.917 -5.291 B6 *

25 804750 * tet02 Ori A 83.845 -5.416 O9.5IVp SB*

26 11673670 V* V566 Ori 83.899 -5.205 A0V Or*

27 800617 * tet01 Ori D 83.821 -5.387 B1.5Vp Y*O

28 801426 Brun 633 83.829 -5.344 A4-A7 *

29 810775 * tet01 Ori C 83.818 -5.389 O7Vp **

30 811405 HD 37188 84.121 -5.770 A7II/III V*

31 804755 HD 36939 83.730 -5.506 B7/8II V*

32 800613 BD-05 1309 83.752 -5.085 A0 *

33 810618 V* T Ori 83.960 -5.476 A3IVeb Ae*

34 800325 HD 36917 83.695 -5.570 B9III/IV Or*

35 811399 HD 36866 83.639 -5.714 A0III/IV Y*O

36 813706 Brun 508 83.765 -5.983 B9V *

37 811400 HD 36983 83.781 -5.868 B5(II/III) Y*O

38 801423 V* V2338 Ori 83.828 -5.291 A8-F0 Or*

39 803639 V* V2056 Ori 83.708 -5.312 A3 Or*

40 800750 HD 36982 83.790 -5.464 B1.5Vp Or*

41 808571 HD 36981 83.775 -5.204 B7III/IV *

42 806045 V* V1073 Ori 83.868 -5.438 B9.5V Or*

43 800631 HD 37114 83.993 -5.375 B9V *

44 800605 HD 36655 83.281 -5.340 B9V *

45 802024 Brun 1018 84.161 -5.472 B6V *

46 800623 HD 37019 83.825 -5.065 A0 V*

47 810053 * tet01 Ori B 83.817 -5.385 B1V Al*

48 800614 BD-05 1310 83.765 -5.095 B9 *

49 812778 BD-05 1300 83.635 -5.762 A3 *

50 800611 HD 36899 83.676 -5.120 A0V Y*O

51 11682603 BD-05 1322 83.859 -5.804 A6Vn Y*O

52 11680584 HD 36919 83.702 -5.998 B9V *

53 811409 Brun 731 83.871 -5.913 A0 Y*O

4

181


.......................................................................

4

182

Chapter 5

Unveiling the Veil:

Protostellar feedback in

Orion

Ü. Kavak, J. Bally, J. Goicoechea, C. H. M Pabst, F. F. S. van der Tak,

A. G. G. M. Tielens (To be submitted)

5

5.1 Abstract

Interest in stellar feedback has recently increased because new studies

suggest that radiative and mechanical feedback from massive stars regulate

the physical and chemical composition of the interstellar medium

(ISM) significantly. Recent SOFIA [C ii] 158 µm observations of the

Orion Veil revealed that the expanding bubble is powered by stellar

winds and influenced by previously active molecular outflows of ionizing

massive stars. We aim to investigate the mechanical feedback on

the whole Veil shell by searching for jets/outflows interacting with the

Veil shell and determining the origin/driving mechanisms of these collisions.

We make use of the [C ii] 158 µm map of the Orion Nebula taken

with the upGREAT instrument onboard SOFIA. The spectral and spatial

resolutions of [C ii] observations are smoothed to 0.3 km s −1 and

16 ′′ , respectively. The velocity-resolved [C ii] observations are utilized

to produce position-velocity (PV) diagrams, which are used to detect

183

CHAPTER 5: Unveiling the Veil: Protostellar feedback in Orion

.......................................................................

5

locations of shock-accelerated [C ii] emitting gas, so called dents, and

to extract [C ii] line profiles to identify velocity components. We image

the [C ii] emission between −3 and −20 km s −1 to pinpoint the highvelocity

structures. Finally, we compare the intensity distribution of the

[C ii] emission with that of 8 µm, and 70 µm. We identify six dents on

the Veil shell with sizes between 0.3 and 1.35 pc and the expansion velocities

range from 4 to 14 km s −1 . The momenta of the dents and their

dynamical timescale suggest that the dents are created by the interaction

of collimated jets/outflows from protostars with the Orion Veil shell.

The measured outflow momenta suggest that the driving protostars have

luminosities with luminosities between 10 3 and 10 4 L ⊙ indicating B-type

stars. However, it is challenging to pinpoint the driving stars as they may

be displaced from the ejection points of the jets/outflows. The intensity

distribution of the [C ii] emission of the dents has a tight correlation

with that of 8 µm, and 70 µm as long as the OMC or the Veil do not

dominate its emission. We conclude that the dynamics of the expanding

Veil shell is influenced not just by the Trapezium stars, but also by other

massive stars in the Orion Nebula. Protostellar feedback appears to play

an important role in deciding the fate of HII regions.

184

5.2 Introduction

.......................................................................

5.2 Introduction

Interest in massive stars (with luminosities larger than 10 3 L ⊙ , corresponding

to a spectral type of B3 or earlier, and stellar masses higher

than 8 M ⊙ (Tan et al. 2014)) has increased in the last three decades as

they inject considerable energy and momentum to unbind and disperse

their natal clouds via stellar winds, powerful outflows, ionising radiation,

and supernova explosions (Krumholz et al. 2014; Bally 2016; Motte

et al. 2018). The injection of mass, momentum, and energy which is

called ‘stellar feedback’ can be seen on various spatial scales (from ∼1

to ∼100 pc) and dynamical timescales (from 10 4 to 10 6 years). At first

glance, supernova explosions are the most energetic feedback process delivering

immense energy (on the order of 10 51 erg seen in observations)

that can reshape the morphology and composition of star-forming galaxies

on large scales (10−100 pc) (Thielemann et al. 2011). However, recent

studies reveal that feedback via protostellar outflows is also vital in setting

the observed properties such as masses of stars (Olivier et al. 2021;

Guszejnov et al. 2021).

Massive stars, in contrast to their low-mass companions, reach their

main-sequence luminosity while still embedded and accreting in a natal

cloud of gas and dust due to their shorter Kelvin-Helmholtz timescales

(Zinnecker & Yorke 2007). Massive protostars eject energetic jets or outflows

to remove the angular momentum excess from the accretion process

till they reach the main-sequence (Beuther et al. 2002a; Sánchez-Monge

et al. 2013d; Kavak et al. 2021). This will result in the entrainment

of a significant amount of ambient molecular material. Even after the

jets/outflows switch off when the star reaches the zero-age main sequence

(ZAMS), relics of previously active molecular outflows, in other words,

fossil outflows will continue to expand on their velocity vector and interact

with the surrounding environment (Quillen et al. 2005). Furthermore,

massive stars tend to form in dense clusters and exhibit a high

multiplicity fraction (Motte et al. 2018). Therefore, it is possible to

find newly forming massive stars and their outflows in the same cluster

(O’Dell et al. 2015) while other massive stars radiate strong UV radiation

(as in the Orion Nebula; Bally 2016). From the point of observations,

quantifying the relative contribution of stellar feedback before and after

reaching the ZAMS has been challenging for years (Lopez et al. 2011),

5

185


.......................................................................

5

despite the fact that state-of-the-art simulations are capable of employing

stellar feedback modes individually (Walch et al. 2012; Haid et al.

2018; Guszejnov et al. 2021).

Orion’s Veil (Veil for short), which is a series of foreground layers

(e.g., 9 layers identified by Abel et al. 2019) of gas and dust lying in front

of the Trapezium stars (O’Dell 2018), is a unique laboratory for studying

the relative effects of feedback mechanisms because its proximity allows

us to resolve the bubbles in the Orion Molecular Cloud (OMC) spatially

and spectrally (O’Dell et al. 2011). The [C ii] emitting Veil layer is a

thin (0.5 pc) H i shell expanding at a velocity of 13 km s −1 toward us

on the OMC-1 core by the kinetic energy converted from stellar winds

of θ 1 Ori C, the most ionizing star in the Orion Nebula (O’Dell 2001;

Pabst et al. 2019). Some studies suggest a multi-layered structure model

for the Veil based on the velocity components characterized through

the emission and absorption lines (Abel et al. 2016; O’Dell 2018; Abel

et al. 2019). The main ionization component of the Veil is traced by

[C ii] fine-structure transition ( 2 P 3/2 → 2 P 1/2 at 158 µm or 1.9 THz, i.e.,

∆E/k B = 91.2 K), which is the main cooling agent of neutral interstellar

gas. While there are other tracers of CO-dark H2 gas (e.g., HF J =1-0;

Kavak et al. 2019), [C ii] is by far the best as C + is the dominant carbon

bearing species and the line is readily excited. Velocity-resolved [C ii]

line observations are the state-of-the-art technique in determining the

driving mechanisms of feedback in massive star-forming regions (Pabst

et al. 2019, 2020; Schneider et al. 2020; Tiwari et al. 2021; Luisi et al.

2021).

Not only photoionization of θ 1 Ori C, but high-velocity structures

such as jets/outflows from YSOs and Herbig-Haro objects play a role in

the dynamics of the Veil on various scales (Henney et al. 2007; Bally et al.

2006; O’Dell et al. 1997). Recently, Kavak et al. (submitted) showed that

even relics of previously active molecular outflows (i.e., fossil outflows)

from θ 1 Ori C affect the morphology of the Veil and could even shatter

the well-known Veil. Blue-shifted ejections, which have relatively weak

[O iii] emission, are impinging on the neutral foreground Veil shell (HH

202, HH 269, HH 203+204; O’Dell 2001) as Veil itself expands while being

and confined by OMC. The collision of such objects with the Veil shell

are a plausible explanation for the large temperature gradients (Peimbert

et al. 1991). In this work, we investigate the shock-accelerated atomic gas

186

5.3 Observations

.......................................................................

over the entire Veil in the PV diagrams generated in cuts along the Veil

and high velocity structures seen in [C ii] observations. Furthermore,

we attempt to investigate the origin of the shock-accelerated gas 1 by

estimating its momentum and dynamical timescale.

We organize the paper as follows. In Section 5.3 we describe the

observations of [C ii] , 12 CO and 13 CO, and mid- and far-IR observations

of the Veil. In Section 5.4, we define our methods to identify the dents

on the Veil and to decompose the observed [C ii] line profiles over the

dent position. Section 5.5 contains an analysis of the momentum and

origin of the dents. Finally, we discuss the origin of the dents and suggest

possible further studies in Section 5.6.

5.3 Observations

5.3.1 [C ii] observations

The observations were carried out with the Stratospheric Observatory for

Infrared Astronomy (SOFIA), an airborne observatory project funded by

the US National Aeronautics and Space Administration (NASA) and the

German Aerospace Centre (DLR). SOFIA is a Boeing 747-SP jetliner

that has been adapted to carry a 2.7-meter-diameter telescope in the

back fuselage (Young et al. 2012).

The data were collected with the German REceiver for Astronomy at

Terahertz Frequencies (upGREAT) Instrument onboard SOFIA (Risacher

et al. 2018) for the Large program of the C + SQUAD led by A. G. G. M.

Tielens. The spectral and spatial resolution during the observation is

about 0.04 km s −1 , and 14.1 ′′ . The final data is resampled to 0.3 km s −1

to achieve a better signal-to-noise ratio. The spatial resolution of the

map is smoothed 16 ′′ , which corresponds to ≃0.03 parsecs at the distance

of Orion, 414 pc (Menten et al. 2007). The final rms noise (in

T mb ) is 1.14 K in 0.3 km s −1 velocity channels. More information on observation

and data reduction can be found in Pabst et al. (2020); Higgins

et al. (2021).

5

1 We use the term of dent for the shock accelerated gas because the shocks collide

with the inner surface of the Veil, resulting in hollow-like structures on the Veil’s

surface.

187


.......................................................................

5

Figure 5.1: Integrated (between −50 and +50 km s −1 ) intensity [C ii]

158 µm map of the Orion Molecular Cloud observed with the upGREAT

receiver onboard SOFIA. The positions of NGC 1977, the Trapezium

stars, M42, M43, and the Orion Bar are labelled. The line at top-right

corresponds 1.5 pc.

188

5.3 Observations

.......................................................................

Vlsr [km/s]

20.0

10.0

0.0

-10.0

-20.0

10.0 20.0 30.0 40.0

Offset [arcmin]

10 1

10 0

10 −1

Tmb [K]

Figure 5.2: PV diagram of Dent 4 identified in this work. [C ii] emission

at +9 km s −1 is the OMC, the weak [C ii] emission of Veil is around

+13 km s −1 . The dent is indicated with a green arrow.

5.3.2 Molecular Gas observations

We make use of 12 CO J = 2-1 (230.5 GHz) and 13 CO J = 2-1 (220.4 GHz)

line maps taken with the IRAM 30m telescope in the framework of the

Large Program ‘Dynamic and Radiative Feedback of Massive Stars’ (PI:

J. R. Goicoechea). In order to facilitate comparison with the velocityresolved

[C ii] map, we smoothed the 12 CO (2-1) and 13 CO (2-1) data

to the angular resolution of the SOFIA [C ii] maps of 16 ′′ . The average

rms noise level in these maps is 0.20 K in 0.41 km s −1 velocity channels.

A more detailed description of the CO observations can be found in

Goicoechea et al. (2020).

5

5.3.3 Mid-IR observations

Mid−infrared observations were taken with the space-borne Spitzer telescope

(Werner et al. 2004) that conducted scientific observations between

2003 and 2020 with three focal plane instruments, one of which being

the Infrared Array Camera (IRAC; Fazio et al. 2004). IRAC is a fourchannel

camera that produces 5.2 × 5.2 arcminute images at 3.6, 4.5,

5.8, and 8 µm. We utilize Spitzer 8 µm observations of the Orion Nebula

to trace the UV-illuminated surface of the Veil.

5.3.4 Far-IR photometric observations

The OMC has been observed as part of the Gould Belt Survey (André

et al. 2010) in parallel mode using the Photoconductor Array Camera and

189


.......................................................................

Spectrometer ((PACS), Poglitsch et al. 2010) and Spectral and Photometric

Imaging Receiver ((SPIRE), Griffin et al. 2010) instruments onboard

Herschel. We use only the archival photometric images of PACS

instrument at 70 µm tracing the dust thermal emission for comparison

with the [C ii] map over the dents.

5.4 Identification of Dents

The dents can be detected with velocity-resolved channel maps and

position-velocity diagrams (Quillen et al. 2005). We first identify notable

dents in [C ii] PV diagrams along the Orion Veil (Section 5.4.1).

We find that the dents move at V LSR between −20 and −3 km s −1 (Section

5.4.3). We then integrate the [C ii] emission (red hue in Fig. 5.3)

between these velocities to identify further dents in high-velocity [C ii]

channels (Section 5.4.2).

5

5.4.1 Position-velocity (PV) Diagrams

Because structures in the Veil are hard to find in the integrated map

of [C ii] , the unbiased way of identifying shock-accelerated material or

dents is the PV diagram. We examine [C ii] PV diagrams of the Orion

Veil produced with 30 ′′ broad horizontal and vertical slices. The horizontal

cuts (east-to-west) are 60 ′′ long, while the vertical cuts (southto-north)

are 45 ′′ long because of the non-spherical morphology of the

Orion Nebula.

PV diagrams uncover the complicated structure of the Veil exposed

to ionizing radiation from Trapezium stars (O’Dell et al. 2017). In all

PV diagrams, the [C ii] emission at +9 km s −1 indicates the background

cloud OMC-1 (see PV diagram of Dent 4 in Fig. 5.2 and of all dents

in Appendix 5.8.3). The main blue-shifted expanding structure, moving

towards us, is the Veil shell at +12 km s −1 (Pabst et al. 2019). In addition

to these structures, we find ‘V-shaped’ substructures that expand faster

than the Veil. Using [C ii] PV diagrams, we identify four dents in the

Veil, which are listed with their properties in Table 5.1. The expansion

velocities are measured relative to the blue-shifted Veil shell. To this end,

we extract the peak velocity, which is determined via Gaussian fitting of

the dent spectrum, of the dent from that of Veil component. The average

190


.......................................................................

Dec (J2000)

40:00.0 -5:30:00.0 20:00.0

10:00.0

5

4

2

1

3

6

50:00.0

37:00.0 30.0 36:00.0 30.0 5:35:00.0 30.0 34:00.0 30.0 33:00.0

RA (J2000)

Figure 5.3: Three-color map of [C ii] emission in the Orion Nebula.

Green hue represents the integrated [C ii] emission from the OMC between

+20 and +3 km s −1 . Blue represents the blue-shifted [C ii]

emission generated by the Veil shell moving between +3 and −3 km s −1 .

Red represents high-velocity [C ii] emitting gas at velocities ranging from

−3 to −20 km s −1 . Gaussian smoothing of radius 35 ′′ is performed to

all three colors to reduce the image noise. The white circles indicate

the position and size of the dents measured via PV diagrams that is

also the aperture size for extracting [C ii] line profiles of the dent. More

information is given in Section 5.5.

4.9 -4.8 -4.6 -4.2 -3.4 -1.8 1.4 7.8 21 46

5

size of the dents is about 0.3 pc which is equal to 2.5 ′ . The size of the

dent is calculated along the RA axis in the PV diagrams. Since we know

the width of each crosscut, the size in Dec is estimated by the number

of PV diagrams in which the dent appears.

191

5

192

Table 5.1: Dents identified in this work. The location and sizes of the dents are shown in Fig. 5.3.

∫

RA (J2000) Dec (J2000) Tmb ∆V Size a I[CII] b (f, α) c Mass V exp,Veil P dent

ID (h:m:s) ( ◦ : ′ : ′′ ) [K km s −1 ] (pc × pc) (× Veil) (× OMC) d 2.0 pc [M ⊙ ] [km s −1 ] [M ⊙ km s −1 ]

1 +05:35:01.5 −05:29:03.5 5.27 ± 1.51 0.36 × 0.24 0.65 0.04 5, 10 3.3 ± 0.6 6.40 ± 1.1 21.5 ± 1.3

2 +05:34:53.4 −05:24:56.7 9.91 ± 4.23 0.16 × 0.18 0.28 0.33 12, 5 0.6 ± 0.1 9.60 ± 3.4 6.00 ± 0.9

3 +05:34:37.4 −05:20:19.9 11.6 ± 3.18 0.43 × 0.30 0.62 0.11 5, 12 4.8 ± 0.9 8.00 ± 1.7 38.4 ± 3.2

4 +05:34:54.2 −05:23:30.5 6.27 ± 1.50 0.39 × 0.30 0.20 0.12 5, 11 3.9 ± 0.8 13.5 ± 1.7 52.5 ± 3.1

5 +05:35:39.2 −05:37:05.2 22.8 ± 1.25 0.12 × 0.12 1.35 1.00 16, 3 1.4 ± 0.3 9.20 ± 0.6 >13.6 ± 0.7

6 +05:34:22.0 −05:40:36.7 6.23 ± 3.61 0.24 × 0.24 0.92 0.46 16, 3 1.5 ± 0.3 4.30 ± 2.3 >6.40 ± 2.0

Notes.

(a) One arcminute corresponds to the physical size of 0.12 pc at the distance of the Orion Nebula (414 pc; Menten

et al. 2007). The error in size is around 10%. (b) I [CII] denotes the integrated intensities of the dent. The values indicate which

component dominates [C ii] emission at the dent position. If the value in the Veil and the OMC columns are > 1, that dent

has brighter [C ii] emission. For fit results, see Table 5.2. (c) d, f, and α denote the distance between star and the Veil surface,

the collimation factor and opening angle (in degree) of possible outflows, respectively. f and α are given for distances of 2.0 pc.

See Section 5.5 for more detail.


.......................................................................


.......................................................................

5.4.2 High-velocity [C ii] emission

Inspection of the [C ii] channel maps reveals [C ii] emission between −20

and −3 km s −1 . We examine the [C ii] channel maps, which show blueshifted

gas with a rather high velocity associated with the Veil. This is

accomplished by superimposing the [C ii] emission within this velocity

range as a red hue across the [C ii] of emission of the Veil (blue hue) and

the OMC-1 background cloud (green hue) in Figure 5.3.

Besides Dents 1−4, we choose two more locations of high-velocity

[C ii] emission, which are Dents 5 and 6 in Fig. 5.3. These spots are

the brightest and farthest high-velocity gas from the Trapezium cluster,

respectively. The behaviour of these high-velocity structures does not

appear as a dent in the PV diagrams (see Section 5.4.1). Examination of

the consecutive PV diagrams covering dents 5 and 6 shows that there is

blue-shifted emission slightly faster than the Veil and that the velocity

of the [C ii] emitting gas increases towards the peak of these structures

(see Figs. 5.12 and 5.13).

Note also that there is high-velocity emission we have not considered

in this work, especially in the Huygens region. This region has been

the subject of many publications (van der Werf et al. 2013; O’Dell 2001;

O’Dell et al. 2009; Abel et al. 2019). The origin of this emission could

be a combination of the radiative and mechanical feedback from the

Trapezium cluster and perhaps also from the stars in the Orion-S cloud

(O’Dell et al. 2009). We also note that we limit this work focusing on

Dent 5 and 6 detected in high-velocity [C ii] emission.

5

5.4.3 Line profiles

The [C ii] emission towards the dents shows a complex structure in the

PV diagrams. To explore the origin of each component, we extract the

[C ii] spectral line profiles across the six dents from the data cube and

present them in Fig. 5.4 between −30 and +30 km s −1 . We utilize the size

of the dent in arcminutes estimated from the PV diagrams to determine

the size of the extraction region. In the direction of the dents, the line

profile suggests a multi-component structure.

We used a multi-Gaussian model to fit the [C ii] spectra and estimate

the line parameters. The fit results are listed in Table 5.2. In the local

standard of rest, all [C ii] spectra exhibit three major components: (i)

193


.......................................................................

the OMC at +9 km s −1 , (ii) Orion’s Veil at about −2 km s −1 , and

(iii) the dents at −10 km s −1 . Dent-1 is an outlier, since it exhibits a

double-peak at the Veil’s velocity (i.e., black and orange fits) as well as

a component at the extreme velocity of −19 km s −1 (see black Gaussian

fit in Fig. 5.4).

Dent 5 and 6 behave differently than the other dents (Figs. 5.12 and

5.13). The PV diagrams of these dents appear to have a bright head of

emission that is not linked with the Veil at first glance (see PV diagram

at the top in Figures 5.12 and 5.13). In addition, their brightness is

similar to the Veil’s. We check the spectra of the adjacent places to

see if accelerated gas is present at the dent positions. These two dents

are expanding somewhat faster than the Veil, according to the spectra

in Fig. 5.14. In Section 5.5, we provide the information regarding the

origin of the [C ii] emission at the head of the dents.

5.5 Analysis

5

In Section 5.5.1, we summarize the properties of the dents. To comprehend

the driving process, we first compare the [C ii] emission with

two crucial PDR tracers in Section 5.5.2. The momentum, which is the

key parameter used to analyze the driving mechanism, of each dent is

then estimated (Section 5.5.3). Finally, we discuss a possible tracer of

dent-like features on the ionization front of HII regions in Section 5.5.4.

5.5.1 Characteristics of the dents

In the previous section, we identify six dents that have diameters ranging

from 0.16 to 0.43 pc. The first dents in Table 5.1 are clearly visible in PV

diagrams, but the last two dents require confirmation by high-velocity

[C ii] emission maps. Four out of the six dents are detected near the

Huygens region hosting the Trapezium cluster. The other two are located

toward the extended extended Orion Nebula (EON; Güdel et al. 2008).

None of the dents appear in 12 CO-PV diagrams. Morphologically, all

dents have different structures, but only one property is the same that

their velocity is higher than the Veil shell. The expansion velocity of the

dents relative to the Veil ranges from 4 to 14 km s −1 .

194

5.5 Analysis

.......................................................................

6

Dent 1

6

Dent 1

4

4

2

2

0

0

6

Dent 2

6

Dent 2

4

4

2

2

0

0

6

Dent 3

6

Dent 3

4

4

2

2

0

0

6

Dent 4

6

Dent 4

4

4

2

2

0

6 Dent 5

4

2

0

Dent 6

0

6 Dent 5

4

2

0

Dent 6

5

Tmb [K]

2

0

Tmb [K]

2

0

−20 −10 0 10 20

V LSR [km s −1 ]

−20 −10 0 10 20

V LSR [km s −1 ]

Figure 5.4: [C ii] 158 µm line profiles using a circular extraction aperture

of dent-size width towards Veil dents listed in Table 5.1. The vertical

green-dashed line indicate the system velocity (9 km s −1 ) of Orion. The

light-green Gaussian fitting line is the OMC, the orange fitting line is

the Veil, and the magenta fitting line is the dent.

195


.......................................................................

5.5.2 Origin of the dents

5

We investigate the contribution of the dents to the total [C ii] emission

and the association of [C ii] emission with PDR tracers such as Spitzer 8

µm tracing the surface of the UV-illuminated molecular cloud and PACS

70 µm tracing warm dust (see Fig. 5.5). Dent 1 and 3 are located in front

of the extension of the Orion Bar and the PDR in the western wall of the

Veil, respectively. Therefore, it is difficult to find an association between

tracers following the intensity changes of these two examples. Dents 2

and 4 show an increase in [C ii] and 8 µm emission, while the decrease in

70 µm emission seems to be continuous. In contrast, Dents 5 and 6 show

an increase at all three tracers at the dent position. Thus it is difficult

to establish a consistent behaviour for the dents among 8 µm, 70 µm,

and [C ii] emissions.

Dent 1 is located just below the bright Orion Bar PDR. The spectrum

in the direction of the dent has four components (see Fig. 5.4). The

fitted lines in magenta, orange, and green represent the dent, the Veil

shell, and background cloud OMC, respectively. The black is component

is observed at the V LSR of −5 km s −1 . This components has been interpreted

as a broad arc in the shape of an incomplete semicircle near

the border of the Huygens region, but the dent at −18 km s −1 is not

observed in H i 21 cm observations (van der Werf et al. 2013). The semicircle

structure host a group of stars including a B-star (see Fig. 5.7).

We suggest that Dent 1 is accelerated by the jet/outflow of these stars

to high velocities such as 20 km s −1 . Based on this, we suggest that the

dents are made up of CO-dark H 2 gas similar to what seen in Orion Bar

(Goicoechea et al. 2015; Kavak et al. 2019). It is possible that all dents

are formed in this way, but this hypothesis require further observations

as discussed in Section 5.6.

5.5.3 Momentum of the dents

The momentum of the dent is estimated to determine the driving mechanism

on the assumption that the momentum of outflows from protostars

is conserved. In this regard, the mass and velocity of the dent must be

estimated. We measure the expansion velocity relative to the Veil shell

using the fit results of [C ii] line profiles. To this end, we extract the

velocity of dent from that of the Veil (see Section 5.2). The mass pa-

196

5.5 Analysis

.......................................................................

Normalized Intensity



1.0

0.8

0.6

0.4

1.0

0.8

0.6

1.0

0.8

0.6

0.4

Dent 1

[CII]

PAH

PACS70

0.0 0.1 0.2 0.3 0.4 0.5 0.6

Distance [pc]

Dent 3

0.0 0.1 0.2 0.3 0.4 0.5 0.6

Distance [pc]

Dent 5

[CII]

PAH

PACS70

[CII]

PAH

PACS70

0.0 0.1 0.2 0.3 0.4 0.5 0.6

Distance [pc]




1.0

0.8

0.6

0.4

0.2

1.0

0.8

0.6

0.4

0.2

1.0

0.8

0.6

Dent 2

[CII]

PAH

PACS70

0.0 0.1 0.2 0.3 0.4 0.5 0.6

Distance [pc]

Dent 4

[CII]

PAH

PACS70

0.0 0.1 0.2 0.3 0.4 0.5 0.6

Distance [pc]

Dent 6

[CII]

PAH

PACS70

0.0 0.1 0.2 0.3 0.4 0.5 0.6

Distance [pc]

5

Figure 5.5: Comparison of normalized [C ii] intensity with to that of

8 µm and PACS 70 µm along 0.6 pc long cuts spanning the dents. The

position of the dent is shown by the blue-dashed line. All observations

are convolved to 20 ′′ for a proper comparison.

197


.......................................................................

5

rameter is, however, rather uncertain because the Veil shell has density

variations of up to a factor of ten and very low N H of ∼10 21 cm −2 (Pabst

et al. 2020) toward the line-of-sight. However, it is possible to make an

estimation based on the mass calculation reported by Pabst et al. (2019).

The mass accelerated by shocks from the Veil shell outward equals

at least the to mass entrained in each dent. As the size (2.7 pc) and

gas mass (1500 M ⊙ ) of the Veil are known (Pabst et al. 2020), we can

roughly calculate the surface mass density of the Veil to calculate the

mass parameter assuming a half-sphere geometry with radius of 2.7 pc

for the Veil. The surface density of the Veil is 32.74 M ⊙ pc −2 . We

multiply the area of the dent by the surface density to estimate the

shock-accelerated mass, in other words, the mass in the dents. The mass

estimation and momentum of the dents are given in Table 5.1. We also

note that Dent 5 and 6 are oblique to the surface of the Veil as shown by

a series of PV-diagrams (see Figs. 5.12 and 5.13) in the Appendix 5.8.3.

We, therefore, give an lower limit for the momentum of these two dents

in Table 5.1.

The total momentum of the Veil is 18,000 M ⊙ km s −1 and the dents

carry thus between 0.5 and 1% of the total momentum injected by the

Trapezium stars (Pabst et al. 2020). For comparison, the momentum

contained in the protrusion to the northwest is 540 M ⊙ km s −1 that is

3% of the momentum of the Veil shell (Kavak et al., submitted).

There is a well-established correlation between the jet/outflow momentum

and the luminosity of the protostars ejecting the material (Bontemps

et al. 1996; Wu et al. 2004; López-Sepulcre et al. 2010; Duarte-

Cabral et al. 2013; Sánchez-Monge et al. 2013d; San José-García et al.

2013; Maud et al. 2015; Kavak et al. 2021). These results indicate a

relationship across the low- and high-mass regimes between these two

quantities (see Fig. 5.6). In this plot, each dent is individually marked

by its momentum. The momentum of the dent implies massive stars of

B-type with luminosities ranging between 10 3 and 10 4 L ⊙ .

Taking the size and the excess velocity as a guide, we estimate that

the formation of the dents would take between (1.3−5.5) × 10 4 years.

This represents 1/4 of the expansion timescale of the Veil shell (∼2 × 10 5

years; Pabst et al. 2019), suggesting that dents were produced during the

expansion of the Veil by newly generated massive B- and A-type stars

which is consistent with accreting massive stars reported by Duarte-

198

5.5 Analysis

.......................................................................

Outflow Momentum [M⊙ km s −1 ]

10 4

10 2

10 0

10 −2

10 −4

Dent 1

Dent 2

Dent 3

Dent 4

10 −1 10 0 10 1 10 2 10 3 10 4 10 5 10 6

Source Luminosity [L ⊙ ]

Figure 5.6: Plot of outflow momentum in M ⊙ km s −1 against the luminosity

of its driving source in L ⊙ . Black circles are from Maud et al.

(2015), and open triangles from Duarte-Cabral et al. (2013) for high-mass

Class−0 objects, and black squares from Dunham et al. (2014) showing

the outflows from low-mass stars.

5

Cabral et al. (2013). As a result, the jets/outflows from accreting massive

protostars are most likely the driving mechanisms.

5.5.4 Potential shock signature of the dents

If the dents were due to a jet interacting with the Veil indeed, then

we would expect to see this region light up in typical shock tracers.

Low velocity interstellar shocks can be J-type or C-shocks, depending

on the strength of the magnetic field and the shock velocity (Draine

et al. 1983). The line-of-sight magnetic field is measured to be ∼100

µG (Troland et al. 1989). For atomic gas, the critical velocity at which

a C-shock becomes J-type is ∼20 km s −1 (Lesaffre et al. 2013). The

observed velocities are consistent with a C-type shock in the range of 5

to 15 km s −1 . Such a shock would heat a column density of ∼10 20 cm −2

199


.......................................................................

to ∼1000 K (Lesaffre et al. 2013). For a velocity in excess of 20 km s −1 ,

the shock would be J-type. The gas is then heated to ∼10 5 K in the

shock front and, in the frame of the shock, would flow at 1/4 of the

shock velocity. As the gas cools down, its velocity would decrease. In a

J-type shock, cooling through atomic lines becomes more important.

5

Comparing PDR models (Kaufman et al. 2006; Pound & Wolfire

2008) and shock models, for the atomic cooling lines ([C ii] , [O i], [C i],

and etc.), the shock signature would be overwhelmed by the emission

generated by the UV irradiation. The best tracers are low-J H 2 lines

(Lesaffre et al. 2013) but there too, the UV−heated gas would have to

be accounted for. As an example, the H 2 0-0 S(1) intensity from a PDR

with G 0 = 10 2 is predicted to be 10 −5 erg cm −2 s −1 sr −1 ; very comparable

to the emission from a 10 km s −1 C-type shock and about 10 times the

emission from a J-type shock (Lesaffre et al. 2013). High velocity J-type

shocks will lead to emission in optical transitions such as [S ii] λ6731.

High velocity resolution will be required to separate this shock emission

from photo-ionized gas in the extended Orion Nebula. Finally, we note

that the substantial column density of warm gas in both C- and J-type

shocks would enable reactions with substantial energy barriers to proceed

and this could lead to detectable amounts of, for example, CH + , OH,

and SH (Lesaffre et al. 2013). As shocks heat the gas to much higher

temperatures than PDRs, these species could be used as the signature of

the presence of a shock. Similarly, near-IR [Fe ii] lines could be used as

shock tracers as these lines originate from levels that cannot be excited

in low density, low UV field PDRs.

The extended Orion region has been surveyed in the 1-0 S(1) line

at 2.12 µm (Stanke et al. 2002). We find that 12 out of 78 H 2 features

reported in Stanke et al. (2002) are situated in the direction of

the Orion Nebula but are unconnected to the dents. We conclude that

shocks revealing the interaction of protostellar jets with the Veil nebula

are difficult to trace directly. The best signature seems to be the velocity

shift induced by this shock interaction in the atomic fine-structure cooling

lines but, as argued above, the emission is dominated by UV heated

gas.

200

5.5 Analysis

.......................................................................

5.5.5 Collimation factor and opening angle

Assuming that the dents are driven by the jets/outflows of protostars,

the collimation factor (f) may also be indication of the type of star given

that outflow collimation decrease from low to massive stars (Bachiller &

Tafalla 1999). However, outflows from B- or O-type stars can be wellcollimated

with factor higher than five in a dynamical scale shorter than

10 4 yr (Arce et al. 2007). Moreover, Wu et al. (2004) report that the

collimation factor of outflow from a protostar with bolometric luminosity

higher than 10 3 L ⊙ is about two. In our case, the degree of the

collimation can be estimated depending on the distance (d) between the

star and the surface of the Veil shell (see also Fig. 5.7 for the assumed

geometry). For this purpose, we assume that the star, which is powering

the outflow is located in the core of the Orion Nebula cluster and

adopt a distance of 2 pc. The collimation factor varies between 5 and

12 while opening angle (α) is between 3 and 12 ◦ (see Table 5.1), indicating

collimated ejections such as molecular jets from massive stars (Arce

et al. 2007). If the star-dent distance were substantially smaller than the

adopted value, the collimation factor would decrease and the opening

angle would increase. For a distance of 0.5 pc, the typical collimation

factor and opening angle would be 2 and 40 ◦ , respectively. We estimate

a timescale of 5.5 × 10 4 years for Dent-1 involved in its formation, which

is 1/4 of the expansion timescale of the Veil shell (Pabst et al. 2019).

With this kind of a timescale, the star does not need to be directly behind

the dent because a massive star with the proper motion of 2 km s −1

can travel about 0.1 pc (∼0.8 ′ ) away in 5.5 × 10 4 years from where outflows

are ejected. Therefore, estimating the driving stars of the dent is

challenging.

In addition, the [C ii] spectra around Dents 5 and 6 suggest [C ii]

emission from accelerated gas (see Fig. 5.14). Only these two dents show

an increase in [C ii] brightness at the head of the dents. By comparing

the intensities in Fig. 5.5 and previous findings in Section 5.4, we argue

that these two dents are also formed by the same mechanisms on the

surface of the Veil shell. We, however, are unable to find the same

association for all dents in Fig. 5.5 (see also Table 5.1) because the [C ii]

emission of the Veil and the OMC dominate the [C ii] emission of Dent

1–4.

5

201


.......................................................................

5.6 Summary

5

Using SOFIA [C ii] observations, we trace the influence of protostellar

feedback by protostars on the Orion Veil. To that aim, we employ PV

diagrams and maps of blue-shifted [C ii] emission ranging from −3 to

−20 km s −1 . A dent is defined as a shock-accelerated structure that

expands outward on the ionization front of the HII region, in our case,

on the Veil shell. We identify six dents in the Veil shell that is expanding

towards us. Their sizes vary between 0.16 and 0.43 pc and they expand

at velocities from 4 to 14 km s −1 . Kavak et al. (submitted) found that

fossil outflows, generated by the Trapezium stars during their protostellar

phase, influence the shape of the Veil shell as well. The momentum

of the dents indicates newly forming stars with luminosities between 10 3

and 10 4 L ⊙ , i.e., B-type stars. The dents are, therefore, a consequence of

the collision of active, energetic jets/outflows expelled by massive protostars

with the ionization front of the HII regions. The Veil shell is

being driven mainly by the stellar wind of θ 1 Ori C, the most ionizing

star in Trapezium cluster (Abel et al. 2019; Pabst et al. 2019). The

Trapezium stars are on the main sequence and this wind is the result of

radiation pressure acting on gas in the stellar photosphere. In contrast,

the jets and outflows considered for the dents are driven by accretion

onto a protostar. We conclude that, in addition to radiative feedback,

both active and fossil outflow processes have a significant impact on the

morphology of the Veil shell. In particular, these protostellar jets and

outflows may create channels and holes in the Veil that will allow the

million degree plasma to escape the Veil confinement. Any escaping hot

plasma will entrain further Veil material and widen the dent’s aperture.

Eventually, the escape of the hot plasma will relieve the pressure of the

wind-blown bubble. At that point, the expansion of the Veil will enter

a momentum conserving phase and eventually merge with the material

in the Orion-Eridanus superbubble. Supernova explosions in the Orion

Ia/Ib associations will sweep up this loose material and transport it to

the walls of this superbubble (Ochsendorf et al. 2015).

According to van der Werf et al. (2013); Abel et al. (2019), the Veil

shell is ionized by the Trapezium stars and has a multilayered structure

along the line-of-sight. Because shocks in jets/outflows with high velocities

in low-density slabs (see also Lehmann et al. 2020) interact with the

202


.......................................................................

Veil, they may cause extra [C ii] emission on the Veil surface. We are,

however, unable to identify this emission in PV diagrams. This might

be attributed to a variety of factors. To begin with, the Veil shell has a

low, varying N H towards line-of-sight as it shows also a density gradient,

which might suggest that there is insufficient material on the Veil surface

to excite. In addition, the dents are positioned in front of the background

cloud OMC-1 core, which is likewise exposed to intense UV-radiation of

Trapezium stars. This radiation may dominate additional [C ii] emission

produced as a result of shock-cloud interaction.

Without velocity-resolved [C ii] observations, it is challenging to unveil

the dent-like structure on the Veil shell. Moreover, estimating the

driving stars of the dents is also difficult as that star could have moved

from the ejection point of its jets/outflows. The dents, unlike the COglobules

(Goicoechea et al. 2020), do not appear in 12 CO-PV diagrams,

indicating that, as for the Veil itself, their NH is low because the dents

are accelerated from the Veil.

In the future, we plan to search for alternative tracers to follow the

dents of the Veil and validate the presence of jets/outflows at their location.

H i 21 cm data may be useful for this purpose. Alternatively,

long-slit spectra observations of 1.644 µm [Fe ii] line, and [S ii] as can for

example be observed with the ARCTIC instrument employed at Apache

Point Observatory with high-resolution (>10,000) might aid in determining

the dynamics of the dents. Finally, we conclude that velocity-resolved

[C ii] observations of SOFIA observatory continue to be state-of-the-art

for discovering feedback mechanisms in massive star-forming regions.

5


We thank to Antoine Gusdorf for useful discussions on the shock models.

Studies of interstellar dust and gas at Leiden Observatory are supported

by a Spinoza award from the Dutch Science agency, NWO. JRG thanks

the Spanish MICIU for funding support under grant PID2019-106110GB-

I00. This study was based on observations made with the NASA/DLR

Stratospheric Observatory for Infrared Astronomy (SOFIA). SOFIA is

jointly operated by the Universities Space Research Association Inc.

(USRA), under NASA contract NAS2-97001, and the Deutsches SOFIA

Institut (DSI), under DLR contract 50 OK 0901 to the University of

203


.......................................................................

Stuttgart. upGREAT is a development by the MPI für Radioastronomie

and the KOSMA/Universität zu Köln, in cooperation with the DLR Institut

für Optische Sensorsysteme. We acknowledge the work, during the

C+ upGREAT square degree survey of Orion, of the USRA and NASA

staff of the Armstrong Flight Research Center in Palmdale, the Ames Research

Center in Mountain View (California), and the Deutsches SOFIA

Institut.

5.8 Appendix

5.8.1 Gaussian Fitting Results

5

Table 5.2: Fit results of multi-Gaussian fitting to [C ii] line profiles in

Fig. 5.4.

V LSR

∫

Tmb ∆V ∆V T mb

Component Position [km s −1 ] [K km s −1 ] [km s −1 ] [K]

Dent 1 −17.5 ± 0.81 5.27 ± 1.51 5.77 ± 1.92 0.85 ± 0.25

2 −9.59 ± 3.34 9.91 ± 4.23 15.8 ± 6.46 0.58 ± 0.09

3 −7.09 ± 1.55 11.6 ± 3.18 11.6 ± 2.54 0.94 ± 0.09

4 −12.8 ± 1.63 6.27 ± 1.50 14.3 ± 4.01 0.41 ± 0.06

5 −7.62 ± 0.19 22.8 ± 1.25 7.12 ± 0.47 3.00 ± 0.16

6 −9.23 ± 0.29 6.23 ± 3.61 3.86 ± 0.84 1.51 ± 0.58

Veil 1 +1.02 ± 0.27 8.02 ± 1.36 3.50 ± 0.68 2.15 ± 0.31

2 +0.01 ± 0.10 34.7 ± 3.52 6.16 ± 0.33 5.29 ± 0.32

3 +0.90 ± 0.21 18.6 ± 3.07 6.67 ± 0.49 2.62 ± 0.28

4 +0.63 ± 0.12 30.9 ± 1.37 7.76 ± 0.31 3.74 ± 0.09

5 +0.92 ± 0.48 16.8 ± 2.06 9.96 ± 1.46 1.59 ± 0.10

6 −4.91 ± 1.98 6.74 ± 3.82 7.23 ± 2.91 0.87 ± 0.18

OMC 1 +9.21 ± 0.02 146.7 ± 1.3 4.18 ± 0.04 32.9 ± 0.29

2 +10.2 ± 0.06 29.43 ± 0.8 4.42 ± 0.01 6.24 ± 0.16

3 +9.76 ± 0.01 106.5± 0.44 3.70 ± 0.02 27.0 ± 0.10

4 +9.54 ± 0.02 51.0 ± 0.55 3.62 ± 0.04 13.2 ± 0.12

5 +9.33 ± 0.02 23.0 ± 0.58 1.87 ± 0.04 11.5 ± 0.22

6 +8.81 ± 0.61 13.6 ± 0.46 3.71 ± 0.14 3.45 ± 0.11

5.8.2 Massive Stars and Geometry

This section contains [C ii] map of the Orion Nebula showing the location

of the dents and massive stars and geometry to estimate the collimation

204

5.8 Appendix

.......................................................................

A-stars B-stars O-stars

500

Dec (J2000)

−5 ◦ 15 ′

30 ′

5

1

4

2

3

6

400

∫ T mb dv [K km s −1 ]

300

200

45 ′ RA (J2000)

100

5 h 36 m 35 m 34 m

0

Star

α

d

Size of dent

H2

5

Veil

Figure 5.7: Left: SOFIA [C ii] map of Orion with O−, B−, and A−stars

found in SIMBAD. The blue−, orange−, and red−filled circles are O−,

B−, and A−stars, respectively. White open circles indicate the dents

identified in this work. Right: Geometry we used to calculate the collimation

factor and opening angle (α). d is the distance between driving

star and the Veil shell.

1

factor (f) and opening angle (α).

5.8.3 PV diagram of the dents

This section consists of a set of PV diagrams that cover the dents investigated

in this work. All PV diagrams are drawn using 60 ′ . Dents are

205


.......................................................................

5

Vlsr [km/s]

20

10 1

10

0

-10

-20 10 −1

0.0 10.0 20.0 30.0 40.0 50.0

20

10 1

10

0

-10

-20 10 −1

0.0 10.0 20.0 30.0 40.0 50.0

20

10 1

10

0

-10

-20 10 −1

0.0 10.0 20.0 30.0 40.0 50.0

20

10 1

10

0

-10

-20 10 −1

0.0 10.0 20.0 30.0 40.0 50.0

20.0

10 1

10.0

0.0

-10.0

-20.0

10 −1

0.0 10.0 20.0 30.0 40.0 50.0

Offset [arcmin]

Tmb [K]

Figure 5.8: Five consecutive PV diagrams showing the changes of the

Dent-1. The dent at 25 ′ is indicated with a green arrow.

denoted with a colored arrow.

206

5.8 Appendix

.......................................................................

Vlsr [km/s]

20

10 1

10

0

-10

-20 10 −1

0.0 10.0 20.0 30.0 40.0 50.0

20

10 1

10

0

-10

-20 10 −1

0.0 10.0 20.0 30.0 40.0 50.0

20

10 1

10

0

-10

-20 10 −1

0.0 10.0 20.0 30.0 40.0 50.0

20

10 1

10

0

-10

-20 10 −1

0.0 10.0 20.0 30.0 40.0 50.0

20.0

10 1

10.0

0.0

-10.0

-20.0

10 −1

0.0 10.0 20.0 30.0 40.0 50.0

Offset [arcmin]

Tmb [K]

5

Figure 5.9: Five consecutive PV diagrams showing the changes of the


207


.......................................................................

5

Vlsr [km/s]

20

10 1

10

0

-10

-20 10 −1

0.0 10.0 20.0 30.0 40.0 50.0

20

10 1

10

0

-10

-20 10 −1

0.0 10.0 20.0 30.0 40.0 50.0

20

10 1

10

0

-10

-20 10 −1

0.0 10.0 20.0 30.0 40.0 50.0

20.0

10 1

10.0

0.0

-10.0

-20.0

10 −1

0.0 10.0 20.0 30.0 40.0 50.0

Offset [arcmin]

Tmb [K]

Figure 5.10: Four consecutive PV diagrams showing the changes of the


208

5.8 Appendix

.......................................................................

Vlsr [km/s]

20

10 1

10

0

-10

-20 10 −1

0.0 10.0 20.0 30.0 40.0 50.0

20

10 1

10

0

-10

-20 10 −1

0.0 10.0 20.0 30.0 40.0 50.0

20.0

10 1

10.0

0.0

-10.0

-20.0

10 −1

0.0 10.0 20.0 30.0 40.0 50.0

Offset [arcmin]

Tmb [K]

5

Figure 5.11: Three consecutive PV diagrams showing the changes of the


209


.......................................................................

5

Vlsr [km/s]

20

10 1

10

0

-10

-20 10 −1

0.0 10.0 20.0 30.0 40.0 50.0

20

10 1

10

0

-10

-20 10 −1

0.0 10.0 20.0 30.0 40.0 50.0

20

10 1

10

0

-10

-20 10 −1

0.0 10.0 20.0 30.0 40.0 50.0

20

10 1

10

0

-10

-20 10 −1

0.0 10.0 20.0 30.0 40.0 50.0

20

10 1

10

0

-10

-20 10 −1

0.0 10.0 20.0 30.0 40.0 50.0

20

10 1

10

0

-10

-20 10 −1

0.0 10.0 20.0 30.0 40.0 50.0

20.0

10 1

10.0

0.0

-10.0

-20.0

10 −1

0.0 10.0 20.0 30.0 40.0 50.0

Offset [arcmin]

Tmb [K]

Figure 5.12: Seven consecutive PV diagrams showing the changes of the

Dent-5. The dent at 18 ′ is indicated with a red arrow.

210

5.8 Appendix

.......................................................................

Vlsr [km/s]

20

10 1

10

0

-10

-20 10 −1

0.0 10.0 20.0 30.0 40.0 50.0

20

10 1

10

0

-10

-20 10 −1

0.0 10.0 20.0 30.0 40.0 50.0

20

10 1

10

0

-10

-20 10 −1

0.0 10.0 20.0 30.0 40.0 50.0

20.0

10 1

10.0

0.0

-10.0

-20.0

10 −1

0.0 10.0 20.0 30.0 40.0 50.0

Offset [arcmin]

Tmb [K]

5

Figure 5.13: Four consecutive PV diagrams showing the changes of the

Dent-6. The dent at 33 ′ is indicated with a red arrow.

211


.......................................................................

10

1

2 1

0

10

2

0

2 2

0

10

3

0

2 3

0

10

Dent 5

0

2 Dent 6

5

5

0

10 5

5

0

10 6

5

0

0

2 5

0

2 6

0

Tmb [K]

10

0

[CII] 158 µm

−20 0 20

V LSR [km s −1 ]

7

Tmb [K]

2

0

[CII] 158 µm

−20 0 20

V LSR [km s −1 ]

7

Figure 5.14: Horizontally consecutive [C ii] spectra extracted over dents

5 and 6, demonstrating the change of the line profile.

212

5.8 Appendix

.......................................................................

5

213


.......................................................................

5

214

Chapter 6

Conclusions and Outlook

6.1 Summary and conclusions

This thesis focuses on the different stages of massive star formation and

their impacts and imprints on the surrounding gas environment. To

answer questions regarding excitation conditions and abundance profiles,

and dynamics of expanding bubbles, we employ a variety of data ranging

from radio to infrared wavelengths as well as radiative and chemical

models. Although we are making progress in understanding the origins

of many phenomena in massive star-forming regions, a number of new

issues have arisen that might be addressed in follow-up initiatives to this

thesis (see Section 6.2).

The early stages of the formation of massive stars is partially revealed

by searching for jets from massive YSOs in Chapter 2. We also study

the PDRs around massive stars in Chapter 3 and how outflows (both

active and fossil) impact and appear in massive star-forming regions in

Chapter 4 and 5.

In Chapter 2, we examine the similarity of the formation of massive

stars to their low-mass companions in relatively nearby star-forming regions

(d < 4 pc). We make use of VLA continuum images at 6 cm and

1.3 cm wavelength of the 18 massive star-forming regions containing outflow

activity as reported by Sánchez-Monge et al. (2013a). We identify a

total of 146 radio continuum sources (40 of them within the field of view

of both the C- and K-band images) we detect twenty-three maser spots,

namely sixteen H 2 O and seven CH 3 OH masers. The spectral index,

215

CHAPTER 6: Conclusions and Outlook

.......................................................................

which is our key criterion, is derived using the flux measurements from

the C- and K-band images for the continuum sources identified at both

frequencies. Of our radio continuum sources, 73% show thermal emission

with a spectral index between from −0.1 and +2.0. Using spectral indices

and associations with outflow and masers, we identify 28 radio jet

candidates among the 146 continuum sources. Following Anglada et al.

(2018), we make use of relationship between jet momentum rate and the

luminosity of its driving stars which suggest 7 probable radio jets and

infer that radio jets are more likely detected in somewhat more evolved

massive star-forming regions. The similar occurrence rate of radio jets in

massive star-forming regions suggest that massive stars form in the same

way as low-mass stars do, namely via disk-mediated accretion (Luhman

2012).

In Chapter 3, we investigate the physical and chemical structure of

the Orion Bar, which is the interface between the HII region and the

molecular cloud in the Orion Nebula. To this end, velocity-resolved

HF J = 1-0 maps from Herschel/HIFI are used to understand the origin

of the HF emission seen in the Orion Bar by employing radiative and

chemical models using the RADEX and Meudon PDR codes, respectively.

With RADEX, we construct the HF column density (N HF ) map.

Afterwards, we employ the Meudon PDR code to explain the abundance

changes of HF from the HII region toward the molecular cloud because

the intensity of HF peaks near the surface of the molecular cloud where

the depth A v > 0.5 and X(HF) = 1.8 × 10 −8 relative to H nuclei.

Combining the results from the RADEX and Meudon PDR models, we

compute the anticipated intensity of the HF J = 1-0 line and conclude

that the HF J = 1 level is excited mainly by collisions of HF molecules

with H 2 at a density of 10 5 cm −3 and a electron density of 10 cm −3 in

the interclump gas of the bright PDR. As a result, we propose HF as a

novel tracer for CO-dark H 2 to study a certain density regime of PDRs

and the properties of extragalactic HF emission.

In Chapter 4, we attempt to understand the non-spherical geometry

of the Orion Veil bubble using velocity-resolved [C ii] observations

(0.3 km s −1 and 16 ′′ ) obtained with the SOFIA/upGREAT instrument.

We identify a protruding-like substructure 1 that extends in a northwesterly

direction from the Orion’s Veil shell that indicates a feedback mech-

1 We use the term of Orion’s Protrusion or protrusion for short for this structure.

216

6.1 Summary and conclusions

.......................................................................

anism in addition to the kinetic energy delivered by the stellar wind of

θ 1 Ori C, which is the most luminous star in the Orion Nebula. Positionvelocity

(PV) diagrams derived from the [C ii] and CO J = 2-1 observations

indicate that this protrusion consists of two half-shells expanding

at different velocities. However, neither shell appears in the CO observations

implying that the protrusion is a very thin shell. The primary

protrusion consists of a 45 M ⊙ [C ii] emitting shell with a thickness of

0.1 pc extending 1.3 pc further from the shores of the Veil shell; in other

words, the Veil’s western rims. We calculate the expansion timescale

of the protrusion and its momentum which are 1.06 × 10 5 years, which

is the half of the expansion timescale of the Veil shell driven by stellar

winds (Pabst et al. 2019) and 540 M ⊙ km s −1 , respectively. Based on

these findings, we infer that the protrusion is a pre-existing cavity carved

by outflows from the Trapezium cluster during their protostellar phase,

which are extinct by now, so-called fossil outflows. Moreover, we estimate

that the Orion Veil shell will break up at the protrusion and the

entire hot gas will be vented into Orion-Eridanus superbubble, which is

about 100 times larger than the Veil shell. We conclude that besides mechanical

and radiative energy from the most massive stars, fossil outflows

are important for the dynamical evolution of HII regions.

In Chapter 5, we look for further signs of protostellar feedback in the

Orion Nebula by searching for high-velocity [C ii] emitting gas. To this

end, we use the same velocity-resolved [C ii] observations of the Orion

Nebula as in Chapter 4. The velocity-resolved [C ii] observations allows

us to construct PV diagrams which reveal high-velocity [C ii] emission.

In the PV diagrams, we identify four spots with shock-accelerated gas,

so called dent 2 . We find that these dents are blue-shifted relative to the

background cloud, OMC (at V LSR = +9 km s −1 ) at V LSR ranging from 4

to 14 km s −1 . Furthermore, we image the [C ii] emission at V LSR between

−3 and −20 km s −1 to trace high-velocity structures within the Veil shell.

We find four bright spots in the Extended Orion Nebula (EON; Güdel

et al. 2008). We choose two representative spots (the brightest spot in

the Nebula and the furthest spot from the Trapezium cluster) among

[C ii] emission seen as high-velocity [C ii] spots. In total, we study these

six dents and estimate their properties such as size, expansion velocity,

2 The term of dent refers to a hollow-like structure generated in the Veil surface as

a result of jets/outflows interacting with the inner surface of the Veil.

217


.......................................................................

mass, dynamical timescale, and momentum. Their sizes range between

0.3 and 1.35 pc and their momentum varies from 6 to 50 M ⊙ km s −1 .

We also estimate that the time required to form the dents ranges from

1 to 5.5 × 10 4 years which is ∼1/4 of the expansion timescale of the

Veil shell (Pabst et al. 2019). The momentum values and timescales

indicate jets/outflows of B-type stars as driving mechanisms. Within this

timescale, the driving star(s) may have moved away from the location,

where jets/outflows were ejected, up to 0.1 pc depending on the proper

motion of the driving star. Therefore, it is very difficult to pinpoint that

star. We conclude that the jets/outflows of the most massive stars in

the Orion Nebula play a major role in the dynamics of the Veil shell and

supply ∼7% of the Veil’s momentum.

6.2 Future Outlook

In Chapter 2, we identify 7 radio jets using the spectral index calculated

via two flux measurements (C and K bands) with the VLA. All of the

continuum sources require at least two more flux measurements for more

accurate spectral index estimation. As a follow-up project, our sample

can be observed in X (3 cm or 10 GHz) and Q bands (0.7 cm or 45 GHz) 3

to calculate more accurate spectral indices of all continuum sources. Because,

spectral index may change with more flux measurements. After

compiling the final list of radio jets, another effort may be to seek for

disks in the massive star-forming regions where we discovered radio jets

using the high-angular resolution of ALMA because the driving source

should contain disks as they host a jet and outflow system. Furthermore,

we discover no radio jets in the half of our sample. It would be

fascinating to know why no radio jets were identified in these objects.

Hence, we plan to compare regions with and without radio jets in order

to investigate why radio jets cannot be detected in some star-forming

regions. Finally, one source from our sample, G189.78+0.34 consists of 5

aligned radio continuum sources along the elongation of the outflow identified

in Sánchez-Monge et al. (2013a). This region might be interesting

to understand the genesis of these continuum sources as it could be an

3 Frequencies are center of each band. For all bands and their range,

please see https://science.nrao.edu/facilities/vla/docs/manuals/oss2013B/

performance/bands.

218

6.2 Future Outlook

.......................................................................

outflow lobe containing several shock knots. For this purpose, velocityresolved

high-angular resolution observations may reveal the nature of

these structures.

In Chapter 3, we find that the HF emission originates from the interclump

medium with a density of 10 5 cm −3 based on the broad linewidth

of the HF line in the Orion Bar PDR. According to Gorti & Hollenbach

(2002), photo-evaporation of clumps in a PDR causes mass flows off the

clump surface and induces a broadening in the linewidth of the observed

line profiles. In addition, the electron density we utilized appears to

be low according a recent study by (Cuadrado et al. 2019). Therefore,

observation of HF emission in different Galactic PDRs is needed to investigate

whether low density is a general characteristic of PDRs bright

in HF. However, the ground state transition of HF requires space-based

observatories. On the other hand, HF has been observed in emission in

extragalactic sources as well (Monje et al. 2014). Timmes et al. (1997)

argued that any positive detection of any fluorine at large redshift (z >

1.5) would suggest a positive screening of the activity of massive stars

strongly. We plan to submit an ALMA proposal to observe HF J =

1-0 at z between 3−7 to figure out the cosmic evolution of the fluorine

abundance via HF emission. Hence, HF emission can be very useful to

investigate properties of early massive stars after solving the origin HF

emission thoroughly.

In Chapter 4, we find that the protrusion in the Orion Veil Shell is a

result of the activity of fossil outflows perturbing the north-west of the

Veil shell. We find that these outflows lobes are blue-shifted and there

is some indications of the corresponding redshifted fossil outflow lobes

in the opposite direction toward the Eastern part of the Veil in the [C ii]

data, which also appears as highly red-shifted 12 CO J = 2-1 emission.

Therefore, we plan to focus on these red-shifted lobes in an upcoming

study. Moreover, we detect 8 µm and Hα emission beyond the protrusion

which seems to be the bow shocks at the tips of a fossil outflow. Our

[C ii] observations do not cover this part of the fossil outflow. We plan to

submit a SOFIA/upGREAT proposal to this region in the [C ii] 158 µm

line with longer integration times.

In Chapter 5, we study some high-velocity structures and dents identified

within the Veil shell. The high-velocity structures in the Huygens

region can be the focus of a follow-up project to unveil the protostellar

219


.......................................................................

feedback. Furthermore, velocity-resolved [C ii] 158 µm observations with

SOFIA are uniquely suited to identify shock accelerated gas and in other

shells around other HII regions. Another follow-up project will look into

shock tracers to determine if they can shed further light on the origin

and evolution of the dents. The tracers to be examined are CH + (J =1-0

at 835.079 GHz and J=2-1 at 1669.16 GHz), [O i] (at 6300λ) [S ii] (at

673.1 nm), S(1) H 2 (2.12 µm), H i 21 cm (VLA/NRAO), [Fe ii] (1.644 µm

with ARCTIC/APO) and [S ii] (6717/7631 with ARCTIC/APO). In addition,

numerical simulations of fossil/protostellar outflows with stellar

wind bubbles may validate the velocity and morphological signatures

that we identified as characteristics, help understand the future evolution

of stellar wind bubbles.

Recent studies reveal that velocity-resolved observation of the [C ii]

line at 158 µm is an excellent tool to probe kinematic and physical conditions

of massive star-forming regions. In this sense, 1.2 square-degree

[C ii] map of the Orion A provides critical insight into the interaction of

massive stars with their surroundings. However, Orion Nebula (or M42

bubble) consists of the Trapezium cluster (i.e., a single 07V star and its

companions) and a variety of stars. In this thesis, I study the protostellar

feedback processes that could change in different regimes of the

Galactic ISM. To investigate the protostellar feedback, I will utilize [C ii]

observation of regions with a range of massive star formation activity.

This will provide invaluable input for simulation of the Galaxy evolution

(see SOFIA/FEEDBACK Survey 4 ; Schneider et al. 2020). Having

characterized the signatures of the interaction of (fossil and active) protostellar

outflow activity on stellar wind bubbles in Orion, this survey will

allow us to address the general importance of protostellar activity on the

evolution of stellar bubbles. Furthermore, upcoming GUSTO (Galactic/Extragalactic

ULDB Spectroscopic Terahertz Observatory) observations

will map the Galactic plane at 158 µm with a beam size of ∼50 ′′

and a velocity resolution of 0.1 km s −1 . This survey may enable us to

explore the protostellar feedback on larger scales up to ∼100 pc, which

is ∼10 times bigger than SOFIA maps.

4 FEEDBACK is a SOFIA (Stratospheric Observatory for Infrared Astronomy)

legacy program dedicated to study the interaction of massive stars with their environment.

It performs a survey of 11 galactic high-mass star-forming regions in the

158 µm (1.9 THz) line of [C ii] and the 63 µm (4.7 THz) line of [O i]

220

6.2 Future Outlook

.......................................................................

221


.......................................................................

222

Chapter 7

Additional Sections

7.1 Contributed Publications

1. Seo, Young Min; Goldsmith, Paul F.; Walker, Christopher K.;

Hollenbach, David J.; Wolfire, Mark G.; Kulesa, Craig A.; Tolls,

Volker; Bernasconi, Pietro N.; Kavak, Ümit; van der Tak, Floris

F. S.; Shipman, Russ; Gao, Jian Rong; Tielens, Alexander; Burton,

Michael G.; Yorke, Harold; Young, Erick; Peters, William L.;

Young, Abram; Groppi, Christopher; Davis, Kristina Pineda, Jorge

L.; Langer, William D.; Kawamura, Jonathan H.; Stark, Antony;

Melnick, Gary; Rebolledo, David; Wong, Graeme F.; Horiuchi,

Shinji; Kuiper, Thomas B., Probing ISM Structure in Trumpler

14 and Carina I: Using the Stratospheric Terahertz Observatory 2;

ApJ, 2019, 878, 120. 1

2. Shipman, R.; Seo, Y.; Tolls, V.; Peters, W.; Kavak, Ü.; Kulesa,

C.; Walker, C., Data Processing of the Stratospheric Terahertz

Observatory-2 [C ii] Survey, Astronomical Data Analysis Software

and Systems XXVIII, 2019, 878, 120. 2

1 Youngmin Seo et al., 2019, ApJ, 878, 120

2 Shipman, R. F. et al., 2019, ASPC, 878, 120

223

CHAPTER 7: Additional Sections

.......................................................................

7.2 Talks

1. Breaking Orion’s Veil bubble via fossil outflows at Virtual NOVA

Network Meeting, the Netherlands, March 3 rd , 2021.

2. Watch out: Exoplanets are in the Netherlands at EAS 2020 Leiden

Virtual (SS23: NAC: Outreach Session), Leiden, the Netherlands,

July 1 st , 2020.

3. Origin of the Hydrogen Fluoride (HF) Emission in the Orion Bar

at JPL Lunch Talk, NASA/JPL-Caltech, Pasadena, California,

USA, July 9 th , 2018.

4. Origin of the Hydrogen Fluoride (HF) Emission in the Orion Bar

at the Olympian Symposium 2018, Paralia Katerini, Mount Olympus,

Greece, May 29 th , 2018.

5. Origin of the Hydrogen Fluoride Emission in the Orion Bar, 73rd

Netherlands Astronomers Conference (NAC2018), May 15 th , 2018.

6. The Interaction of Stars with Gas Clouds: HF in the Orion Bar,

Wednesday Launch Talk (WLT) of Interstellar Medium, A.G.M.M.

(Xander) Tielens, Leiden Observatory, Leiden, the Netherlands,

Oct 18 th , 2017.

7. Understanding the Chemical Complexity in Protostellar Outflows:

L1157-B1, Wednesday Lunch Talk (WLT) of Kapteyn Institute,

Groningen, The Netherlands, Oct 4 th , 2017.

7.3 Posters

224

1. Kavak, Ü., F.F.S. van der Tak, A.G.G.M. Tielens, and R.F. Shipman,

Origin of hydrogen fluoride emission in the Orion Bar: An

excellent tracer for CO-dark H 2 gas clouds, Virtual European Astromical

Society 2020, Leiden, June 29 - July 3, 2020.

2. Kavak, Ü., A.G.G.M. Tielens, F.F.S. van der Tak, J. Goicoechea,

Breaking the bubble: [C ii] observations of the Orion Veil, Virtual

European Astronomical Society 2020, Leiden, June 29 - July 3,

2020.

7.3 Posters

.......................................................................

3. Kavak, Ü., F.F.S. van der Tak, A.G.G.M. Tielens, R.F. Shipman,

CO-dark H 2 gas clouds are not dark anymore!, Astrochemistry

2018, Pasadena, California, USA, 10–14 July Sep 2018

4. Kavak, Ü. & Taquet V., Understanding of Chemical Complexity

in Protostellar Outflows: L1157-B1 Star Forming Region, Fundamentals

of the Life In Universe, Groningen, The Netherlands, 31

August - 1 September 2017.

225


.......................................................................

7.4 Türkçe Özet

Başımızı yukarı doğru kaldırıp gökyüzüne baktığımızda parıldayan yıldı

zları görürüz. Yalnız astronomlar değil, farklı kültürlere sahip toplumlar

da gökyüzüne benzer şekilde büyük bir ilgi göstermiş ve kutsal olduğunu

düşünmüştür. Hatta bu toplumlar, kutsal gökyüzünün parlak yıldızlarını

birleştirerek farklı şekiller veya hayvanlara benzetmişlerdir. Bu

sadece yıldızlarla kalmamış, Samanyolu Galaksi’sine de farklı isimler verilmiştir.

3 Bilimsel anlamda ise 1600’lü yıllara kadar, henüz teknolojinin

gelişmemiş olması nedeniyle, yapılan bir çok çalışma matematiksel

hesaplarla veya basit gözlemlerle sınırlı kalmıştır.

Orta Çağ’da, meraklarına yenik düşen cesur araştırmacılar küçük

ölçeklerde teleskoplar inşa edip, kutsal gökyüzünü olabildiğince yakından

gözlemek istemiştir. 19. yüzyıla gelindiğinde, teknolojinin gelişmesi

ve ışığın doğasının daha iyi anlaşılmasıyla, gökyüzünün bazı bölgelerinde

olması gereken daha az yıldız olduğu fark edilmiştir. William Herschel

bu bölgelerin cennete açılan kapılar olduğunu öne sürdü. Ancak bu açıklamalar

bilimsel olarak yetersizdi ve daha çok ispat gerekiyordu. Daha

gerçekçi bir yaklaşımla Edward E. Barnard, bu ilginç bölgelerin yıldızlarla

Dünya’mız arasındaki maddeler olduğunu iddia etmiş ve böylece

kutsala olan bakış açısı değişmeye başlamıştır (Barnard 1919). Sonra

ki bir kaç yüzyıl içinde bilim camiasının bu bölgelere olan ilgisi artmış

ve koyu bölgelerin kataloğu yayınlanmıştır (Bok & McCarthy 1974).

Yirminci yüzyılın başlarında ise bu karanlık bölgelerin sadece Dünyamız

ile yıldızlar arasında değil, yıldızların da arasında olduğu gösterilmiştir.

Bu sonuçlar yıldızlararası ortamla ilgili devrimsel keşiflere neden olmuştur

(Spitzer 1978).

Yıldızlararası Madde (kısaca YAM) terim olarak yıldızlar arası ortamda

bulunan herşeyi ifade etmektedir. Bugün biliyoruz ki YAM’ın

boyutları 0.35 nm ile 1 µm arasında değişen küçük toz taneciklerden

(%1), hidrojen atom ve moleküllerinden (%99) oluşmaktadır. YAM,

farklı sıcaklık ve yoğunlukta bölgelerde bulunmaktadır. YAM’ın bulunduğu

en soğuk bölgeler yaklaşık olarak 10 Kelvin veya −263 C ◦ ’dir.

3 Örnek olarak, kültürümüzde Samanyolu Galaksi’sine Kehkeşan da denilmektedir.

Kehkeşan kelime anlamı olarak saman taşıyan demektir. Dilimizde ise bir saman

balyasından yere dökülen saman taneciklerinin yerde oluşturduğu parıltılı yola denmektedir.

226

7.4 Türkçe Özet

.......................................................................

Orion B

NGC 1981

NGC 1977


Integral

shaped

filament

Ghost

filament

1 °

7.3 pc

NGC 1977

M43

M42

L 1641

cloud

Herschel N(H) map


Figure 7.1: Orion A yıldız oluşum bölgesi ve NGC 1981 ve 1977 yıldız

kümeleri. Telif hakkı: Amy M. Stutz / MPIA.

Bu soğuk bölgeler molekül bulutu olarak adlandırılmıştır ve biliyoruz ki

molekül bulutları yeni yıldızların oluştuğu bölgelerdir (Şekil 7.1).

İnsanlar gibi yıldızların da bir ömrü vardır; ancak insanların tersine

yeni yıldızlar önceki nesilden kalan yıldızların kalıntılarından doğmaktadır.

Yıldızlar ömürlerini tamamladıktan sonra (yani 10 6 ile 10 9 yıl)

etrafındaki maddeyi yavaşça yıldızlararası ortama salarak veya şiddetli

bir şekilde patlayarak ömürlerini bitirir. Bu iki farklı ölüm şekli ise

yıldızın kütlesine bağlı olarak değişir. Güneş kütlesinin (M ⊙ ) iki katı

veya daha aşağı kütlesindeki yıldızlara düşük kütleli yıldızlar olarak denilmektedir.

İki ve daha yüksek (> 2 M ⊙ ) kütleli yıldızlar ise yüksek

kütleli yıldızlar olarak sınıflandırılmaktadır.

Son yarım asırda, düşük kütleli yıldızların oluşum senaryosu çokça

çalışılmıştır (bkz. derleme yayın; Luhman 2012) ve detaylı bir senaryo

önerilmiştir (Şekil 7.2). Bu senaryoda molekül bulutunun en yoğun bölgelerindeki

yığışmalar kütle çekimsel etkinin altında çöker. Sonrasında

yoğunluğu ve sıcaklığı artan merkezcil çekirdek yapı yıldızımsı nesneyi

oluşturur. Bu sırada sistemde biriken açısal momentum, jet ve fışkırmalarla

yıldızlararası ortama aktarılır. Bu aktarımlar bulunduğu or-

227


.......................................................................

Figure 7.2: Düşük kütleli bir yıldızın oluşum senaryosu. 1 AU

Güneş−Dünya arasında uzaklığı göstermektedir. Telif hakkı: Visser

2009.

tamın yapısal ve kimyasal yapısını önemli ölçüde etkiler. Sistemde biriken

açısal momentum aktarımı esnasında gezegenleri oluşturacak disk yapısı

oluşur. Bu disk içindeki yoğun bölgeler yörüngelerini temizleyerek gezegenleri

oluşturur. Sonrasında evrimini tamamlayıp optik dalgaboylarında

ışımaya başlayan bir gezegenli bir yıldız sistemi oluşur. İnsanlar olarak

içinde bulunduğumuz yıldız-gezegen sistemine ise Güneş sistemi demekteyiz.

Yüksek kütleli yıldızlar ağır elementlerin sentezlenmesinde ve evrenin

erken fazlarında iyonize gazın oluşmasında rol oynar (bkz., Tan et al.

2014). Ancak, yüksek kütleli yıldızların oluşumu ise düşük kütleleri yıldızlara

göre daha az anlaşılmıştır. Çünkü bu konunun irdelenmesine yetecek

veri elde edilememiştir. İlk olarak büyük kütleli yıldızlar çok fazla

yıldızın bulunduğu molekül bulutlarında oluşmaktadır. Ayrıca yüksek

kütleli yıldızlar düşük kütleli yıldızlara göre daha uzakta gözlenmiştir.

228

7.4 Türkçe Özet

.......................................................................

Orion Çıkıntısı

Fosil Fışkırmalar

M43 Karanlık Şerit

Karanlık Körfezi

M43 Nebulası

Kuzeydoğu karanlık şeridi

Orion Çubuk

Trapezium

Yıldızları

Orion-S

cloud

Orion Çubuk

Uzantısı

Orion küçük

çıkıntısı

Baloncuklar

Peçe Kuşu

X-ray Kuzey

Batı Sınırı

Kelvin-Helmholtz

kararsızlıkları

Doğu Sınırı

X-ray Güney

Rayleigh-Taylor

kararsızlıkları

Güney Sınır

1

Figure 7.3: ESO/VLT taraması kapsamında alınan Orion Nebula’sının

görüntüsü. Bölgedeki önemli yapılar renkli çizgilerle gösterilmiştir.

Arka plandaki Orion Nebula görüntüsü ESO’nun sitesinden getirilebilir:

https://www.eso.org/public/images/eso1723a/

Buna ek olarak, yüksek kütleli yıldızlar etrafını saran gazı 100 pc 4 ölçeklerine

kadar iyonlaştırarak ederek galaksilerin iyonize fazı oluşturur (Abel

et al. 2002).

Yüksek kütleli yıldızların oluşumu ise son yıllarda artan bir ilgiyle

çokça çalışılan bir konu haline gelmiştir. ALMA ve VLA 5 gibi inter-

4 100 pc yaklaşık olarak üç peta kilometreye (3 × 10 15 km) karşılık gelmektedir.

5 Atacama Large Millimeter Array veya ALMA interfeometresi Şili’de konuşlandırılmış

66 tane çanaklı teleskopun birleştirilmesiyle milimetre ve milimetre-altı

229


.......................................................................

ferometrik gözlemevlerinin yüksek çözünürlüklü gözlemleri yıldız oluşum

bölgelerinin gerek dinamiğini gerekse de kimyasal yapısını anlamakta çok

değerli sonuçlar getirmiştir. Bu tez çalışmasında ilk olarak yüksek kütleli

yıldızların oluşumundaki jet atımlar aranmıştır (Bölüm 2). Daha sonra,

Orion Nebulası’nda 6 bulunan Orion Çubuk’un yoğunluğu çalışılmıştır

(bkz. Şekil 7.3 ve Bölüm 3).

Yüksek kütleli yıldızların etrafındaki materyali dışa doğru süpürmesiyle

yıldızlararası ortamda çokça görülen baloncuk yapılar (veya iyonize

hidrojen bölgeleri) oluşur. Orion Nebulası’nın SOFIA gözlemeviyle

yapılan son gözlemleri, bu nebulanın balon yapısı içinde bulunan çok yüksek

kütleli bir yıldızın yüzeyinden atılan yıldız rüzgarlarıyla şişirildiğini

göstermiştir (Pabst et al. 2019). Ancak basit iyonize hidrojen bölgesi

modellerinin aksine Orion Nebula’sı büyük çıkıntı benzeri bir yapı göstermiştir.

Bölüm 4’te bu çıkıntıya neden olan mekanizma tartıştık ve sonuç

olarak, bu çıkıntıya Trapezium yıldızlarından atılan ve şu anda sönmüş

olan fosil fışkırmaların neden olduğu tespit ettik. Bir sonraki aşamada ise

nebulanın içinde bulunan diğer yüksek kütleli yıldızların etkisi araştırdık.

Bunun sonucunda balon yapının iç yüzeyine çarpan altı tane fışkırmanın

izini bulduk (bkz. Bölüm 5).

Sonuç olarak, bu tez çalışmasında yüksek kütleli yıldızların da düşük

kütleli yıldızlar gibi oluşabileceği gösterilmiştir. Dahası yüksek kütleleri

yıldızların atımlarının iyonize hidrojen bölgelerinin yapısını önemli ölçüde

etkileyebileceği ve hatta bu atımların Orion Nebula’sını patlatabileceğini

gözlemsel olarak ispat ettik. Çünkü bu balon yapıların patlamasıyla

yıldızlararası ortama sıcak ve iyonize olmuş gaz atılır. Bu sayede, iyonize

bölgelerin dinamiği hakkında bilgi edinmek için önemli bir adım

atılmıştır.

dalgaboylarında astronomik gözlemler yapan bir gözlemevidir. Benzer şekilde Very

Large Array veya VLA New Mexico’da konuşlandırılmış 27 tane çanakla radyo dalgaboylarında

astronomik gözlemler yapan bir gözlemevidir.

6 Orion Nebula Dünyamıza en yakın (414 pc) yüksek kütleli yıldız oluşum bölgesini

içeren nebuladır. Bu balon benzeri yapıda, bize doğru saniyede 13 km hızla (yaklaşık

olarak bir saatte 46,800 km) genişleyen ve direkt olarak görünmeyen bir kabuk vardır.

Bu yapıya Orion Peçesi veya Kabuğu denilmektedir. Bu yapının bu denli hızlı bir şekilde

büyümesine yapının kuzeyinde bulunan Trapezium yıldızlarının en parlak üyesi

θ 1 Ori C neden olmuştur.

230

7.5 Nederlandse samenvatting

.......................................................................


Op het moment dat we naar de hemel kijken, zien we de sterren schijnen.

Niet alleen astronomen, maar ook samenlevingen van verschillende culturen

hebben veel aandacht geschonken aan de hemel. Daarnaast werd

gedacht dat de hemel heilig was. Tot ongeveer de 17e eeuw, wanneer

de technologie nog niet geëvolueerd was, waren de berekeningen beperkt

tot de wiskunde. Daarnaast werden heldere sterren van de heilige hemel

gecombineerd, vergeleken met dieren van het dierenrijk waarnaar ze genoemd

werden. Dit gold niet enkel voor de sterren, ook voor de Melkweg 7

werden verschillende benamingen gegeven.

In de Middeleeuwen waren er dappere onderzoekers, en gedreven door

hun nieuwsgierigheid, bouwden ze een kleine telescoop waarmee ze de

heilige hemel zo dichtbij mogelijk konden bekijken. In de 19e eeuw,

samenhangend met de evoluerende technologie en gestegen kennis over

het licht, werd in de hemel op bepaalde plaatsen gebrek aan sterren

geconstateerd. William Herschel noemde deze gebieden een opening tot

het paradijs. Doch deze uitleg van William Herschel was wetenschappelijk

gezien onvoldoende onderbouwd en veel meer onderbouwing was

nodig. Edward E. Barnard kwam met een realistischere benadering en

gaf aan dat deze gebieden, tussen de sterren en onze wereld, uit materie

bestonden waardoor de gedachte van heiligheid begon te veranderen

(Barnard 1919). Ondertussen werd in de wetenschapswereld de interesse

voor deze gebieden aangewakkerd en resulteerde dit in het publiceren

van een catalogus over de donkere gebieden (Bok & McCarthy 1974). In

het begin van de 20e eeuw werd vastgesteld dat deze donkere gebieden

niet enkel tussen de sterren en onze wereld lagen, maar ze werden ook

vastgesteld tussen de sterren. Dit resulteerde in grote ontdekkingen in

de omgeving tussen de sterren (Spitzer 1978).

Interstellaire Materie (kortweg ISM) als term drukt alles uit dat zich

tussen de sterren bevindt. Tegenwoordig weten we dat ISM uit kleine

stofdeeltjes variërend tussen de 0.35 nm en 1 µm (1%), waterstof atomen

7 Volgens de Griekse mythe werd de Melkweg gevormd toen Hera Heracles

borstvoeding gaf. Heracles was de zoon van Zeus uit een sterfelijke vrouw. Zeus’

vrouw Hera verachtte het kind, maar toen zijn halfzus Athena Heracles naar Hera

bracht, had ze geen idee wie hij was en gaf ze hem uit medelijden borstvoeding. Heracles

zoog zo hard dat hij Hera verwondde, en ze trok hem weg, waardoor melk het

universum in schoot en de Melkweg vormde.

231


.......................................................................

Orion B

NGC 1981

NGC 1977


Integral

shaped

filament

Ghost

filament

1 °

7.3 pc

NGC 1977

M43

M42

L 1641

cloud

Herschel N(H) map


Figure 7.4: Het stervormingsgebied Orion A en de sterrenhopen

NGC 1981 en 1977. Auteursrecht: Amy M. Stutz/MPIA.

en moleculen (99%) bestaat. De omgeving tussen de sterren blijkt een

verschillende temperatuur en dichtheid te hebben. De koudste gebieden

in het ISM zijn ongeveer 10 Kelvin oftewel -263 ◦ C. Deze koude gebieden

worden benoemd als zijnde moleculaire wolk of stervormingsgebied

(Figuur 7.4). Vandaag de dag weten we dat moleculaire wolken

gebieden zijn waar nieuwe sterren gevormd worden.

Net zoals mensen hebben sterren ook een leven; echter in tegenstelling

tot de mens ontstaan sterren uit resten van de vorige generatie sterren.

Wanneer sterren hun leven beëindigen (d.w.z. tussen de 10 6 en 10 9 jaar)

verspreiden zij hun materie langzaam in het ISM of ontploffen in een zogenaamde

supernova. De verschillen in beide manieren van het beëindigen

van het leven wordt bepaald door de stellaire massa. Sterren die twee

keer of minder massa dan de zonsmassa (M 0 ) hebben, worden lage-massa

sterren genoemd. In tegenstelling tot de lage-massa sterren, worden sterren

die twee keer of meer massa dan de zonsmassa (>2 M ⊙ ) hebben hoge

massa sterren genoemd. De laatste halve eeuw, zijn het ontstaansscenario

van lage-massa sterren vaak onderzocht (Luhman 2012). Om dit

232


.......................................................................

Figure 7.5: Scenario van lage massa stervorming. 1 AU is de afstand

tussen de zon en de aarde. Auteursrecht: Visser 2009.

scenario te kunnen verklaren is er een gedetailleerd scenario ontwikkeld

(Figuur 7.5). In dit scenario stort de moleculaire wolk ineen in de meest

dichte gebieden onder invloed van de zwaartekracht. Daarna, na het

toenemen van de dichtheid en temperatuur ontstaat er een kern die een

basis vormt van het ontstaan van een protostellaire fase. In de tussentijd

worden de geaccumuleerde impulsmomenten uitgeworpen door de protoster

door middel van jets en uitstroom in ISM. Deze uitstromen beïnvloeden

de morfologische en chemische structuur van de omgeving waarin

ze zich bevinden in aanzienlijke mate. Wanneer de geaccumuleerde impulsmomenten

worden uitgeworpen ontstaan er schijven die de basis vormen

voor het ontstaan van planeten. In deze schijven bevinden zich

dichte structuren die hun banen leegvegen zodat planeten gevormd worden.

De ster wordt optisch zichtbaar er ontstaat er een planetenstelsel.

Als mens zijnde bevinden we ons in het ster-planeet systeem dat we het

zonnestelsel noemen.

Er wordt gedacht dat sterren met hoge massa een rol spelen in het

233


.......................................................................

M43 Donkere laan

M43 Nevel

Noordoostelijke donkere baan

Trapeziumsterren

Orion-uitsteeksel

Bubbels

Kleine Orionuitsteeksel

Fossiele Uitstroom

Donkere Baai

Orion Bar

Orion-S

wolk

uitbreiding

van Orion Bar

Orion vogel

X-ray Noord

Kelvin-Helmholtz

instabiliteiten

Westelijke-rand

Oost-rand

X-ray Zuid

Rayleigh-Taylor

instabiliteiten

Zuid-rand

1

Figure 7.6: Afbeelding van de Orionnevel uit de ESO/VLT Survey met

duidelijke structuren gemarkeerd met verschillende kleuren. De Sluierschil

ligt voor de Trapezium-sterren en de rand is langs de randen. De

achtergrondafbeelding van de Orionnevel is te vinden op de ESO-website:

https://www.eso.org/public/images/eso1723a/

ontstaan van geïoniseerde gassen in het vroege universum (Tan et al.

2014). De vorming van sterren met hoge massa is relatief gezien minder

begrepen dan die van sterren met lage massa. Dit komt omdat er niet

voldoende data verkregen zijn om dit te begrijpen. Ten eerste, ontstaan

sterren met hoge massa in het algemeen in moleculaire wolken waar veel

sterren zich bevinden. Daarnaast, bevinden sterren met hoge massa zich

verder weg dan sterren met lage massa. Bovendien beïnvloeden sterren

234


.......................................................................

met een hoge massa het omringende gas tot 100 pc 8 het begin van de

ioniserende fase van het heelal maakt.

Het ontstaan van sterren met hoge massa verkreeg de laatste jaren

meer aandacht. Interferometrische observatoria zoals ALMA, en de VLA 9

maakten het mede door hun hoge resolutie mogelijk zowel de dynamiek

als de chemische structuur van de protostellaire gebieden te begrijpen.

Tijdens het werken aan dit proefschrift werd eerst het ontstaan van sterren

met hoge massa de jet structuren gedetecteerd (Hoofdstuk 2). Nadien

werd de dichtheid van de Orion Bar die zich in de Orionnevel bevindt,

bestudeerd (zie Figuur 7.4 en Hoofdstuk 3).

Het materiaal rondom sterren met hoge massa wordth naar buiten

geblazen waardoor er in het ISM vaak geziene ballonstructuren (oftewel

geïoni seerde waterstof gebieden) ontstaan. Waarnemingen aan de Orionnevel

door het SOFIA observatorium laten zien dat een structuur binnen

in de ballon opgeblazen is door de wind afkomstig van de oppervlakte

van een ster met hoge massa (Pabst et al. 2019). Echter, de Orionnevel,

in tegenstelling tot eenvoudig geïoniseerde waterstof gebieden, laat

een groot uitsteeksel zien. In hoofdstuk 4 wordt aandacht geschonken

aan het mogelijke mechanisme dat de oorzaak zou kunnen zijn van deze

uitsteeksels. De Trapezium sterren en met name hun op dit moment

uitgedoofde‘fossiele’ uitstromingen worden aangewezen als de oorzaak

van dit uitsteeksel. Als volgende stap werd het effect van de andere sterren

met hoge massa in de Orion-nevel onderzocht. Als gevolg van dit

onderzoek werd aan de binnenkant 6 sporen van uitsteeksels gevonden

(zie Hoofdstuk 5). Als resultaat van dit proefschrift werd aangetoond dat

sterren met hoge massa zoals sterren met lage massa kunnen ontstaan.

Bovendien, hebben wij gevonden dat de uitstroom van sterren met hoge

massa de structuur van de geïoniseerde waterstof gebieden in belangrijke

mate kan beïnvloeden en het is zelfs geobserveerd dat de nevel kan

ontploffen.

8 100 pc komt overeen met ongeveer drie peta-kilometers (3 × 10 15 km).

9 De Atacama Large Millimeter Array of ALMA-interferometer is een observatorium

dat astronomische waarnemingen doet op millimeter- en submillimetergolflengten,

waarbij 66 schoteltelescopen worden gecombineerd die in Chili zijn

geplaatst. Evenzo is de Very Large Array of VLA een observatorium dat astronomische

waarnemingen doet op radiogolflengten met 27 schotels gestationeerd in New

Mexico.

235


.......................................................................


Hoera!!! The PhD journey has flown by so quickly, and now it’s over!

Now it is the time to read the funniest part of the thesis. I have to thank

and send my greetings to a vast number of individuals, so let me sum up

in just a few pages and begin with my supervisors.

Floris, I owe you a big thank you for giving me this great opportunity

to become a researcher and take part in this challenging PhD path. You

were a helpful advisor and taught about science and on how to be an

independent researcher. To improve my skills, the most essential thing I

learnt from your experiences was to use an agenda to carefully organize

my days, weeks, and months. I look forward for further collaborations

with you!

Alexander Godfried Gerardus Maria or widely known as Xander or

even grandPAH, there are very few individuals that know all of your

middle names, and I am one of them. When I informed you that I

could say all your middle names by heart, you said that it was time

for me to graduate. For me, there are no words to thank you for what

you’ve done for me throughout my PhD. I have learnt a lot as I have

grown in my scientific thinking, and of course I have studied your ISM

book together with you, what an eye-opener. I would love to keep in

touch with you to gain better understanding of science, and marriage.

That’s right, marriage too, since you gave me a lot of golden points for a

happy relationship with my wife. Please accept my sincere gratitude and

appreciation for the post-doctoral position in Leiden and the opportunity

to fulfil my dream of working for SOFIA/NASA. Lastly, I completely

agree with you and believe wholeheartedly in what you think about the

real definition of happiness: "Do fun things with fun people!".

Russ, you were the researcher who was most exposed to, or perhaps

suffered, about my questions related to data reduction and the HIFI

instrument. You were a lifesaver for me when I first started working

on my PhD. Thank you very much for your patience and understanding!

Each and every time I hear the terms of HIFI/Herschel or data reduction,

your name will come up to my mind. I wish the best of the best for you!

Garip, I saw your artwork, which is ‘van Gogh on water’ a couple of

years ago on social media and it was the trending topic of that day. I

did not think of using it for my thesis at that time, but here we go! Your

236


.......................................................................

painting is on the back-cover of my thesis, and I feel that it was one of

the greatest contributions to the thesis. I am grateful for the permission

to use all the artworks, not only for the cover, but also for the end of

each chapter. I wish cheerfully that the Ebru or Turkish paper marbling

would take its rightful place with your invaluable contributions!

Yakup, we first met in Cologne about seven years ago. The Cologne

moments were easy to endure with your helps and I felt very welcome

to your friendly environment. In time, I have learnt about your talents

and your design firm, Adgency Koeln, and realized that you are much

more than a frequent Red Bull drinker night-owl. When we discussed

the cover design for this thesis, I had an idea of what I wanted, but

your creative talents and suggestions resulted in the final design, which

was precisely what I had envisioned. Thank you for your unconditional

friendship as well as the thesis cover design.

Kapteyn and SRON Secretaries, Administrators, and computer groups

(Christa, Eite, Leon, Lucia, Martine, Martin, Romana, Alie, Bert, Chantal,

Engelien, and Frank) you are all, in my opinion, VIPs of the two

institutions. Without your help, it would be difficult to move in the

institutes and complete our PhDs, since you handle the majority of the

paperwork.

Pooja, it was fantastic to spend the time together in the office with

you in the last several years. I’m grateful for your continuous support

and encouragement in each and every case. The only thing I wanted to

point up is that we were flat mates in the student dormitory when we

started our jobs, but the first two weeks, we didn’t see each other. One

day, we were complaining about our rooms. We then realized that we

reside in the same dormitory flat, but to put it bluntly, the first time in

Groningen was hectic for all of us. I’ll never forget your friendship or

your low-pitched singing in the office while I was working behind you. I

hope that we will be in touch forever.

Nelvy, without any doubt you are the most silent person I have ever

met. From my desk, it is not possible to see you directly, so I sometimes

prefer to talk to you without turning. Sometimes, I was ‘talking’ to you

without you because you left the office extremely silent. It was always a

great pleasure to speak with or without you, no matter how many times

it occurred. Nevertheless, I will never forget the moments that I realized

that I was talking alone and waited for your answer back in the office. I

237


.......................................................................

am very happy to meet you and have you in my life!

Kristiina, it’s wonderful to have you and Karlis in my contact list. I

am sure we have amassed a plethora of wonderful memories in Turkey and

the Netherlands over the past few years. I will never forget how joyful the

wedding was with your presence and that of other Kapteyners. I also

recall the dinner at your house with home-made pasta and delectable

dishes. I owe thanks not only for the invitation, but also for providing

me yummy foods with the largest platter. I wish you a wonderful life,

increased culinary abilities, and a continued relationship between our

families.

Nick, due to pandemic, we were in the same office only for one year,

but it was nice talking to you about anything that passed by that day. I

wished to have more time with you because I think describing anything

in any language is a very difficult task, but you are one of the most

clearest person I have ever met. I wish you a great life and hopefully we

will meet again soon, maybe in Belgium or somewhere else.

Avanti, I always have a lot of good memories to write about you and

your friendship, but when I think about you, I remember what you most

likely remember. You would say the same thing, which is the phrases

of nantar bhetu, which is roughly translated to mean ‘come later’ in

Marathi. I am certain that I frequently heard similar version of this

word in my home country as a kid as well. Things change everywhere,

but I want to maintain our friendship regardless of the length of the time

between now and the future.

Mr. William Pearson and Jorrit you are both gentlemen from the

British and Dutch cultures, as well as very amusing. I believe it is difficult

to have these two characteristics in the same individual. This was one of

the reasons for my visits to your office and our discussions on different

topics. For me, it was a pleasant moment when I corrected your English

grammar, William. And Jorrit, eierbal is a kind of Dutch crochet that is

widely available in Groningen. I also learnt a lot from you, particularly

regarding the he/she distinction :)) I wish you both the best of luck in

your future endeavours!

Pavel and Andre, I just could not stand the thought of writing your

names separately :)) For me, you were the color of the institute and my

wedding. I personally enjoyed my wedding, but your funny dancing style

was unique that night. I guess most of my cousins are willing to see you

238


.......................................................................

both again and again or even invite you both to their weddings. I believe

if one wants to produce anything humorous, both of you should be asked

to participate in as Meltem did for my birthday gift. I wish to have you

both for the rest of my life :))

Hyoyin, it was great to have you in the last two years of my time at

Kapteyn. You were at the wedding and played piano on my birthday-gift

video as well. I noticed that no one ever played the piano for me, so you

are the first person who played it. Please accept my sincere thanks for

your friendship, piano performance, and support!

Veronica, You were like an elder sister to me at some point! Namely,

I told you once that I had a Vitamin-D issue and needed to supply some

vitamin pills as soon as possible. You angrily said, ‘How don’t you take

Vitamin-D in this sunless country!’. I felt that I made a huge mistake

and bought vitamins from the pharmacy immediately. You were also

both a great friend and a group-mate to me. I guess you want to be

around in Europe for the rest of your life. If this is still the case, we will

have time to get together. I wish you a great life in the Netherlands for

you and your family.

Kostas, maybe we never spoke about our projects as astronomers, but

conversations on gaining muscles and losing weight were highly discussed

together. Nevertheless, it was a pleasure to hear about your advanced

body-building experiences and suggestions, and they really work. I’d

want to meet up with you in Greece or in Turkey to talk further.

Lisa and Annelis, after Meltem and I were married, I had two houses,

one in Belgium and one in the Netherlands. Thank you for your friendship,

which made it much easier for me to adjust to the Flemish region

of Belgium. In addition, it was apparent that you both had a warm,

sympathetic personality at first glance. Unquestionably, both of your

performance in my birthday video was a proof of this. Thank you for

being a great friend. May you have a wonderful life and have lovely

memories with your loved ones including us!

Umut, Fatime, and Duru, you have helped me so much in my academic

career to achieve the point where I am right now. During my

travels to the United States, you made me feel like I was part of your

family. After I will be settled in the USA in January 2022, we should

plan another road trip in the USA with you as soon as we can. It was an

excellent decision to go to Las Vegas from Pasadena for a few days and

239


.......................................................................

it should not be our the last stop. Our next stop should be a place like

Vegas where we can leave all memories behind. I am looking forward to

seeing you all again and we can discuss where we will be going next.

Şeyda, I believed I’d lost my Turkish language skills after nearly seven

years in Europe. But it didn’t happen thanks to you, so thanks for the

Turkish chats with wonderful cheese breads and cappuccinos, and your

friendship three thousand kilometers away from homeland. Soon, it will

be my turn to defence my thesis, and I would like to add that it was

really beneficial to be one of the paranymphs during your defence at the

Aula building before defending mine.

Mustafa, true, we were not in Kapteyn at the same time, but I was

one of the PhD students who worked in the office that we took over from

you. You also helped me to adapt to the PhD life in the Netherlands

during the last several years and paid two visits to my home. It was great

to have you in Groningen. To be honest, you are the most fascinating

person I have ever encountered! Please take this as a compliment, since

I agree with you on a number of things. I also welcome your help in

polishing the Turkish summary of the thesis. I’d want to pay you a visit

at your home in Kayseri (TR), and I’d like to see you again very soon.

Oya, I realized that our time together in Groningen was extremely

limited after getting to know you. You still hold the record of the longest

Skype call. I believe it is difficult to break this record with someone

else, except Meltem. Currently, you are in Istanbul with Meltem and

asking about how you contributed on my thesis while I am penning these

sentences. I believe that your friendship was particularly important on

this journey. You will also submit your thesis soon and I wish you defend

it successfully. Maybe this is the best place to thank all the GUTSA

(Turkish student association in Groningen) members. It would be great

if we could get together soon and explore the cities we both call home.

Kadir Kangel, I believe I was lucky to meet you in Cologne thanks

to a cup of traditional Turkish tea. Since that time, it has always been a

pleasure to speak with you. You and your family were also very helpful

during my Cologne journey. Your untimely demise really saddened me.

I wish we had more time to chat on anything with a tea. During your

funeral prayer, I learned how much you are loved throughout the NRW

region. I wish your soul serenity!

Zekiye Aydın, despite the fact that you were one of my closest cousins,

240


.......................................................................

I accepted you like an elder sister because of your interest, support, and

love. Your unexpected passing away at the age of 36 is tremendously

difficult to put into words. Everyone who had the pleasure of spending

even a little time with you is devastated and will greatly miss you. For

the rest of our lives, you’ll be at the center of our thoughts and hearts,

and every moment will be incomplete without your bright grin.

Even though I’ve been officially out of my former astronomical institution

in Turkey for nearly four years, I’m still there. I always feel

welcomed when I study and work in that department. Your polite relationship

at long distance has been wonderful. From A to Z, Damla, Ersin,

Emre, Fulin, Hikmet, İlhan, Kenan, Mehmet Hanifi, Nezahat, Nurdan,

Özlem, Pınar and your families, thank you very much for your support

and friendship.

There are many people to thank for their support in various nations

from west to east. Please accept these sentences for each of you: Thank

you very much for your support! One of the greatest thanks must go

to my family, family-in-law and relatives and the rest of the world. My

father, mother, and brothers, I have to also thank for your support and

love. Being your kid and brother makes me feel lucky!

And Meltem, it is very difficult to express my feeling and love to you

in any of the languages I speak. I am overjoyed to have you as a friend a

few years ago and to decide to marry you as my eternal friend and wife.

A word says: "An unmarried individual is half, but would be completed

when he/she is married". I believe I found my other half a few years

ago on July 21. Thank you very much for your unconditional love and

support.

Now, I am really done!

241


.......................................................................

242

Bibliography

Abel, N. P., Ferland, G. J., & O’Dell, C. R. 2019, ApJ, 881, 130

Abel, N. P., Ferland, G. J., O’Dell, C. R., & Troland, T. H. 2016, ApJ,

819, 136

Abel, T., Bryan, G. L., & Norman, M. L. 2002, Science, 295, 93

Agúndez, M., Cernicharo, J., Waters, L. B. F. M., et al. 2011, A&A, 533,

6

André, P. 2002, in EAS Publications Series, Vol. 3, EAS Publications

Series, ed. J. Bouvier & J.-P. Zahn, 1–38

André, P., Hennebelle, P., & Peretto, N. 2008, Ap&SS, 313, 29

André, P., Men’shchikov, A., Bontemps, S., et al. 2010, A&A, 518, L102

Andre, P., Ward-Thompson, D., & Barsony, M. 2000, in Protostars and

Planets IV, ed. V. Mannings, A. P. Boss, & S. S. Russell, 59

Andrews, B. H. & Thompson, T. A. 2011, ApJ, 727, 97

Anglada, G. 1995, in Revista Mexicana de Astronomia y Astrofisica Conference

Series, Vol. 1, Revista Mexicana de Astronomia y Astrofisica

Conference Series, ed. S. Lizano & J. M. Torrelles, 67

Anglada, G. 1996, in Astronomical Society of the Pacific Conference

Series, Vol. 93, Radio Emission from the Stars and the Sun, ed. A. R.

Taylor & J. M. Paredes, 3–14

Anglada, G., Rodriguez, L. F., Canto, J., Estalella, R., & Torrelles, J. M.

1992, ApJ, 395, 494

243

BIBLIOGRAPHY

.......................................................................

Anglada, G., Rodríguez, L. F., & Carrasco-González, C. 2018, A&ARv,

26, 3

Anglada, G., Villuendas, E., Estalella, R., et al. 1998, AJ, 116, 2953

Arab, H., Abergel, A., Habart, E., et al. 2012, A&A, 541, A19

Arce, H. G., Shepherd, D., Gueth, F., et al. 2007, in Protostars and

Planets V, ed. B. Reipurth, D. Jewitt, & K. Keil, 245

Ashimbaeva, N. T., Colom, P., Krasnov, V. V., et al. 2017, Astronomy

Reports, 61, 16

Astropy Collaboration, Robitaille, T. P., Tollerud, E. J., et al. 2013,

A&A, 558, A33

Azatyan, N. M., Nikoghosyan, E. H., & Khachatryan, K. G. 2016, Astrophysics,

59, 339

Bacciotti, F., Chiuderi, C., & Oliva, E. 1995, A&A, 296, 185

Bachiller, R. 1996, ARA&A, 34, 111

Bachiller, R. & Tafalla, M. 1999, in NATO Advanced Study Institute

(ASI) Series C, Vol. 540, The Origin of Stars and Planetary Systems,

ed. C. J. Lada & N. D. Kylafis, 227

Bally, J. 2016, ARA&A, 54, 491

Bally, J., Ginsburg, A., Arce, H., et al. 2017, ApJ, 837, 60

Bally, J., Licht, D., Smith, N., & Walawender, J. 2006, AJ, 131, 473

Bally, J., Reipurth, B., Lada, C. J., & Billawala, Y. 1999, AJ, 117, 410

Barnard, E. E. 1919, ApJ, 49, 1

Battersby, C., Bally, J., Jackson, J. M., et al. 2010, ApJ, 721, 222

Beaumont, C. N., Goodman, A. A., Kendrew, S., Williams, J. P., &

Simpson, R. 2014, ApJS, 214, 3

Beltrán, M. T., Cesaroni, R., Moscadelli, L., et al. 2016, A&A, 593, A49

244

BIBLIOGRAPHY

.......................................................................

Beltrán, M. T., Estalella, R., Anglada, G., Rodríguez, L. F., & Torrelles,

J. M. 2001, AJ, 121, 1556

Beltrán, M. T., Sánchez-Monge, Á., Cesaroni, R., et al. 2014, A&A, 571,

A52

Benedettini, M., Molinari, S., Testi, L., & Noriega-Crespo, A. 2004,

MNRAS, 347, 295

Bennett, C. L., Kogut, A., Hinshaw, G., et al. 1994, ApJ, 436, 423

Berné, O., Marcelino, N., & Cernicharo, J. 2010, Nature, 466, 947

Bertoldi, F., Draine, B. T., Rosenthal, D., et al. 2000, in IAU Symposium,

Vol. 197, From Molecular Clouds to Planetary, ed. Y. C. Minh

& E. F. van Dishoeck, 191

Beuther, H., Schilke, P., Sridharan, T. K., et al. 2002a, A&A, 383, 892

Beuther, H., Walsh, A., Schilke, P., et al. 2002b, A&A, 390, 289

Bisbas, T. G., Bell, T. A., Viti, S., Yates, J., & Barlow, M. J. 2012,

MNRAS, 427, 2100

Blaauw, A. 1964, ARA&A, 2, 213

Bok, B. J. & McCarthy, C. C. 1974, AJ, 79, 42

Bontemps, S., Andre, P., Terebey, S., & Cabrit, S. 1996, A&A, 311, 858

Breen, S. L., Ellingsen, S. P., Contreras, Y., et al. 2013, MNRAS, 435,

524

Carrasco-González, C., Rodríguez, L. F., Anglada, G., et al. 2010, Science,

330, 1209

Carrasco-González, C., Torrelles, J. M., Cantó, J., et al. 2015, Science,

348, 114

Castor, J., McCray, R., & Weaver, R. 1975, ApJL, 200, L107

Caswell, J. L., Fuller, G. A., Green, J. A., et al. 2010, MNRAS, 404,

1029

245

BIBLIOGRAPHY

.......................................................................

Caswell, J. L. & Green, J. A. 2011, MNRAS, 411, 2059

Cesaroni, R., Sánchez-Monge, Á., Beltrán, M. T., et al. 2017, A&A, 602,

A59

Cesaroni, R., Sánchez-Monge, Á., Beltrán, M. T., et al. 2016, A&A, 588,

L5

Ceverino, D. & Klypin, A. 2009, ApJ, 695, 292

Chini, R., Kruegel, E., & Wargau, W. 1987, A&A, 181, 378

Chu, Y.-H. & Gruendl, R. A. 2011, Bulletin de la Societe Royale des

Sciences de Liege, 80, 297

Churchwell, E. 2002, ARA&A, 40, 27

Churchwell, E., Povich, M. S., Allen, D., et al. 2006, ApJ, 649, 759

Condon, J. J., Cotton, W. D., Greisen, E. W., et al. 1998, AJ, 115, 1693

Cuadrado, S., Salas, P., Goicoechea, J. R., et al. 2019, A&A, 625, L3

Cunningham, A. J., Frank, A., Quillen, A. C., & Blackman, E. G. 2006,

ApJ, 653, 416

Curiel, S., Canto, J., & Rodriguez, L. F. 1987, Revista Mexicana de

Astronomía y AstrofÃŋsica, 14, 595

Curiel, S., Rodríguez, L. F., Cantó, J., et al. 1989, Astrophysical Letters

and Communications, 27, 299

Cutri, R. M., Wright, E. L., Conrow, T., et al. 2012, Explanatory Supplement

to the WISE All-Sky Data Release Products, Explanatory

Supplement to the WISE All-Sky Data Release Products

Cyganowski, C. J., Brogan, C. L., Hunter, T. R., & Churchwell, E. 2009,

ApJ, 702, 1615

Cyganowski, C. J., Brogan, C. L., Hunter, T. R., & Churchwell, E. 2011,

ApJ, 743, 56

Cyganowski, C. J., Brogan, C. L., Hunter, T. R., et al. 2012, ApJL, 760,

L20

246

BIBLIOGRAPHY

.......................................................................

Cyganowski, C. J., Koda, J., Rosolowsky, E., et al. 2013, ApJ, 764, 61

Cyganowski, C. J., Whitney, B. A., Holden, E., et al. 2008, AJ, 136,

2391

Da Rio, N., Robberto, M., Soderblom, D. R., et al. 2009, ApJS, 183, 261

Dale, J. E., Bonnell, I. A., Clarke, C. J., & Bate, M. R. 2005, MNRAS,

358, 291

Dame, T. M., Hartmann, D., & Thaddeus, P. 2001, ApJ, 547, 792

De Buizer, J. M., Radomski, J. T., Telesco, C. M., & Piña, R. K. 2005,

ApJS, 156, 179

de Graauw, T., Helmich, F. P., Phillips, T. G., et al. 2010, A&A, 518,

L6

de Villiers, H. M., Chrysostomou, A., Thompson, M. A., et al. 2015,

MNRAS, 449, 119

de Zeeuw, P. T., Hoogerwerf, R., de Bruijne, J. H. J., Brown, A. G. A.,

& Blaauw, A. 1999, AJ, 117, 354

Desrousseaux, B. & Lique, F. 2018, MNRAS, 476, 4719

Draine, B. T. 1978, ApJS, 36, 595

Draine, B. T. 2011, Physics of the Interstellar and Intergalactic Medium

(Princeton University Press)

Draine, B. T., Roberge, W. G., & Dalgarno, A. 1983, ApJ, 264, 485

Duarte-Cabral, A., Bontemps, S., Motte, F., et al. 2013, A&A, 558, A125

Dunham, M. M., Arce, H. G., Mardones, D., et al. 2014, ApJ, 783, 29

Elitzur, M. 1982, Reviews of Modern Physics, 54, 1225

Emprechtinger, M., Monje, R. R., van der Tak, F. F. S., et al. 2012, ApJ,

756, 136

Evans, II, N. J. 1999, ARA&A, 37, 311

247

BIBLIOGRAPHY

.......................................................................

Fazio, G. G., Hora, J. L., Allen, L. E., et al. 2004, ApJS, 154, 10

Feddersen, J. R., Arce, H. G., Kong, S., et al. 2018, ApJ, 862, 121

Federrath, C., Schrön, M., Banerjee, R., & Klessen, R. S. 2014, ApJ,

790, 128

Felli, M., Brand, J., Cesaroni, R., et al. 2007, A&A, 476, 373

Fierlinger, K. M., Burkert, A., Ntormousi, E., et al. 2016, MNRAS, 456,

710

Frank, A., Ray, T. P., Cabrit, S., et al. 2014, in Protostars and Planets

VI, ed. H. Beuther, R. S. Klessen, C. P. Dullemond, & T. Henning,

451

Froebrich, D., Makin, S. V., Davis, C. J., et al. 2015, MNRAS, 454, 2586

Fujimoto, Y., Chevance, M., Haydon, D. T., Krumholz, M. R., & Kruijssen,

J. M. D. 2019, MNRAS, 487, 1717

Gasiprong, N., Cohen, R. J., & Hutawarakorn, B. 2002, MNRAS, 336,

47

Gatto, A., Walch, S., Naab, T., et al. 2017, MNRAS, 466, 1903

Gaume, R. A., Fey, A. L., & Claussen, M. J. 1994, ApJ, 432, 648

Genzel, R., Tacconi, L. J., Lutz, D., et al. 2015, ApJ, 800, 20

Gildas Team. 2013, GILDAS: Grenoble Image and Line Data Analysis

Software

Ginsburg, A., Glenn, J., Rosolowsky, E., et al. 2013, ApJS, 208, 14

Ginsburg, A., Koch, E., Robitaille, T., et al. 2019, radio-astrotools/spectral-cube:

Release v0.4.5

Goicoechea, J. R., Pabst, C. H. M., Kabanovic, S., et al. 2020, A&A,

639, A1

Goicoechea, J. R., Pety, J., Cuadrado, S., et al. 2016, Nature, 537, 207

Goicoechea, J. R., Teyssier, D., Etxaluze, M., et al. 2015, ApJ, 812, 75

248

BIBLIOGRAPHY

.......................................................................

Gordon, M. A. & Sorochenko, R. L. 2002, Radio Recombination Lines.

Their Physics and Astronomical Applications, Vol. 282 (Russian

Academy of Sciences)

Gorti, U. & Hollenbach, D. 2002, ApJ, 573, 215

Grenier, I. A., Casandjian, J.-M., & Terrier, R. 2005, Science, 307, 1292

Griffin, M. J., Abergel, A., Abreu, A., et al. 2010, A&A, 518, L3

Großschedl, J. E., Alves, J., Meingast, S., et al. 2018, A&A, 619, A106

Güdel, M., Briggs, K. R., Montmerle, T., et al. 2008, Science, 319, 309

Guillon, G. & Stoecklin, T. 2012, MNRAS, 420, 579

Guszejnov, D., Grudić, M. Y., Hopkins, P. F., Offner, S. S. R., &

Faucher-Giguère, C.-A. 2021, MNRAS, 502, 3646

Guzmán, A. E., Garay, G., & Brooks, K. J. 2010, ApJ, 725, 734

Guzmán, A. E., Garay, G., Brooks, K. J., & Voronkov, M. A. 2012a,

ApJ, 753, 51

Guzmán, V., Pety, J., Gratier, P., et al. 2012b, A&A, 543, L1

Hacar, A., Tafalla, M., Forbrich, J., et al. 2018, A&A, 610, A77

Haid, S., Walch, S., Seifried, D., et al. 2018, MNRAS, 478, 4799

Harper-Clark, E. & Murray, N. 2009, ApJ, 693, 1696

Hartigan, P., Morse, J. A., & Raymond, J. 1994, ApJ, 436, 125

Hatchell, J., Fuller, G. A., & Millar, T. J. 2001, A&A, 372, 281

Helfand, D. J., Becker, R. H., White, R. L., Fallon, A., & Tuttle, S. 2006,

AJ, 131, 2525

Henney, W. J., O’Dell, C. R., Zapata, L. A., et al. 2007, AJ, 133, 2192

Herschel Science Ground Segment Consortium. 2011, HIPE: Herschel Interactive

Processing Environment, Astrophysics Source Code Library

249

BIBLIOGRAPHY

.......................................................................

Higgins, R., Kabanovic, S., Pabst, C., et al. 2021, arXiv e-prints,

arXiv:2106.15620

Hoare, M. G., Purcell, C. R., Churchwell, E. B., et al. 2012, PASP, 124,

939

Hogerheijde, M. R., Jansen, D. J., & van Dishoeck, E. F. 1995, A&A,

294, 792

Hollenbach, D. J., Takahashi, T., & Tielens, A. G. G. M. 1991, ApJ,

377, 192

Hollenbach, D. J. & Tielens, A. G. G. M. 1997, ARA&A, 35, 179

Hollenbach, D. J. & Tielens, A. G. G. M. 1999, Reviews of Modern

Physics, 71, 173

Hosokawa, T. & Omukai, K. 2009, ApJ, 691, 823

Jayasinghe, T., Dixon, D., Povich, M. S., et al. 2019, MNRAS, 488, 1141

Jijina, J. & Adams, F. C. 1996, ApJ, 462, 874

Johnston, K. G., Robitaille, T. P., Beuther, H., et al. 2015, ApJL, 813,

L19

Kalenskii, S. V. & Kurtz, S. 2016, Astronomy Reports, 60, 702

Kaufman, M. J., Wolfire, M. G., & Hollenbach, D. J. 2006, ApJ, 644,

283

Kaufman, M. J., Wolfire, M. G., Hollenbach, D. J., & Luhman, M. L.

1999, ApJ, 527, 795

Kavak, Ü., Sánchez-Monge, Á., López-Sepulcre, A., et al. 2021, A&A,

645, A29

Kavak, Ü., van der Tak, F. F. S., Tielens, A. G. G. M., & Shipman, R. F.

2019, A&A, 631, A117

Kennicutt, R. C. 2005, in Massive Star Birth: A Crossroads of Astrophysics,

ed. R. Cesaroni, M. Felli, E. Churchwell, & M. Walmsley, Vol.

227, 3–11

250

BIBLIOGRAPHY

.......................................................................

Kennicutt, R. C. & Evans, N. J. 2012, ARA&A, 50, 531

Kereš, D., Katz, N., Davé, R., Fardal, M., & Weinberg, D. H. 2009,

MNRAS, 396, 2332

Keto, E., Zhang, Q., & Kurtz, S. 2008, ApJ, 672, 423

Kölligan, A. & Kuiper, R. 2018, A&A, 620, A182

Koumpia, E., Harvey, P. M., Ossenkopf, V., et al. 2015, A&A, 580, A68

Krumholz, M. R. 2011, in American Institute of Physics Conference Series,

Vol. 1386, XV Special Courses at the National Observatory of

Rio de Janeiro, ed. E. Telles, R. Dupke, & D. Lazzaro, 9–57

Krumholz, M. R., Bate, M. R., Arce, H. G., et al. 2014, in Protostars

and Planets VI, ed. H. Beuther, R. S. Klessen, C. P. Dullemond, &

T. Henning, 243

Krumholz, M. R. & Matzner, C. D. 2009, ApJ, 703, 1352

Kuchar, T. A. & Clark, F. O. 1997, ApJ, 488, 224

Kuiper, R., Klahr, H., Beuther, H., & Henning, T. 2011, ApJ, 732, 20

Kurtz, S. 2005, in IAU Symposium, Vol. 227, Massive Star Birth: A

Crossroads of Astrophysics, ed. R. Cesaroni, M. Felli, E. Churchwell,

& M. Walmsley, 111–119

Kurtz, S., Cesaroni, R., Churchwell, E., Hofner, P., & Walmsley, C. M.

2000, in Protostars and Planets IV, ed. V. Mannings, A. P. Boss, &

S. S. Russell, 299–326

Kurtz, S., Churchwell, E., & Wood, D. O. S. 1994, ApJS, 91, 659

Lang, D. 2014, AJ, 147, 108

Langer, W. D., Velusamy, T., Pineda, J. L., Willacy, K., & Goldsmith,

P. F. 2014, A&A, 561, A122

Larson, R. B. 1969, MNRAS, 145, 271

Le Petit, F., Nehmé, C., Le Bourlot, J., & Roueff, E. 2006, ApJS, 164,

506

251

BIBLIOGRAPHY

.......................................................................

Lehmann, A., Godard, B., Pineau des Forêts, G., & Falgarone, E. 2020,

A&A, 643, A101

Lekht, E. E. 2000, A&AS, 141, 185

Leroy, A. K., Bolatto, A., Gordon, K., et al. 2011, ApJ, 737, 12

Lesaffre, P., Pineau des Forêts, G., Godard, B., et al. 2013, A&A, 550,

A106

Leto, P., Umana, G., Trigilio, C., et al. 2009, A&A, 507, 1467

Li, Z.-Y. & Nakamura, F. 2006, ApJL, 640, L187

Lis, D. C. & Schilke, P. 2003, ApJL, 597, L145

Liu, T., Wu, Y., & Zhang, H. 2013, ApJ, 776, 29

Lizano, S. 2008, in Astronomical Society of the Pacific Conference Series,

Vol. 387, Massive Star Formation: Observations Confront Theory, ed.

H. Beuther, H. Linz, & T. Henning, 232

Lockman, F. J. 1989, ApJS, 71, 469

Lopez, L. A., Krumholz, M. R., Bolatto, A. D., Prochaska, J. X., &

Ramirez-Ruiz, E. 2011, ApJ, 731, 91

Lopez, L. A., Krumholz, M. R., Bolatto, A. D., et al. 2014, ApJ, 795,

121

López-Sepulcre, A., Cesaroni, R., & Walmsley, C. M. 2010, A&A, 517,

A66

López-Sepulcre, A., Codella, C., Cesaroni, R., Marcelino, N., & Walmsley,

C. M. 2009, A&A, 499, 811

López-Sepulcre, A., Walmsley, C. M., Cesaroni, R., et al. 2011, A&A,

526, L2

Lu, X., Zhang, Q., Liu, H. B., Wang, J., & Gu, Q. 2014, ApJ, 790, 84

Luhman, K. L. 2012, ARA&A, 50, 65

252

BIBLIOGRAPHY

.......................................................................

Luisi, M., Anderson, L. D., Schneider, N., et al. 2021, Science Advances,

7, eabe9511

Lumsden, S. L., Hoare, M. G., Urquhart, J. S., et al. 2013, ApJS, 208,

11

Madden, S., Geis, N., Genzel, R., et al. 1997, in ESA Special Publication,

Vol. 401, The Far Infrared and Submillimetre Universe., ed. A. Wilson,

111

Marti, J., Rodriguez, L. F., & Reipurth, B. 1993, ApJ, 416, 208

Maud, L. T., Cesaroni, R., Kumar, M. S. N., et al. 2019, A&A, 627, L6

Maud, L. T., Moore, T. J. T., Lumsden, S. L., et al. 2015, MNRAS, 453,

645

McKee, C. F. & Ostriker, E. C. 2007, ARA&A, 45, 565

McMullin, J. P., Waters, B., Schiebel, D., Young, W., & Golap, K. 2007,

in Astronomical Society of the Pacific Conference Series, Vol. 376,

Astronomical Data Analysis Software and Systems XVI, ed. R. A.

Shaw, F. Hill, & D. J. Bell, 127

Megeath, S. T., Flaherty, K. M., Hora, J., et al. 2005, in IAU Symposium,

Vol. 227, Massive Star Birth: A Crossroads of Astrophysics, ed.

R. Cesaroni, M. Felli, E. Churchwell, & M. Walmsley, 383–388

Megeath, S. T., Gutermuth, R., Muzerolle, J., et al. 2012, AJ, 144, 192

Menten, K. M., Reid, M. J., Forbrich, J., & Brunthaler, A. 2007, A&A,

474, 515

Monje, R. R., Emprechtinger, M., Phillips, T. G., et al. 2011a, ApJL,

734, L23

Monje, R. R., Lord, S., Falgarone, E., et al. 2014, ApJ, 785, 22

Monje, R. R., Phillips, T. G., Peng, R., et al. 2011b, ApJL, 742, L21

Moscadelli, L., Cesaroni, R., & Rioja, M. J. 2005, A&A, 438, 889

253

BIBLIOGRAPHY

.......................................................................

Moscadelli, L., Cesaroni, R., Sánchez-Monge, Á., et al. 2013, A&A, 558,

A145

Moscadelli, L., Sánchez-Monge, Á., Goddi, C., et al. 2016, A&A, 585,

A71

Moscadelli, L., Sanna, A., Goddi, C., et al. 2019, A&A, 631, A74

Motte, F., Bontemps, S., & Louvet, F. 2018, ARA&A, 56, 41

Murray, N., Ménard, B., & Thompson, T. A. 2011, ApJ, 735, 66

Naab, T. & Ostriker, J. P. 2017, ARA&A, 55, 59

Nagy, Z., Choi, Y., Ossenkopf-Okada, V., et al. 2017, A&A, 599, A22

Nagy, Z., Van der Tak, F. F. S., Ossenkopf, V., et al. 2013, A&A, 550,

A96

Nakamura, F. & Li, Z.-Y. 2008, ApJ, 687, 354

Neufeld, D. A., Sonnentrucker, P., Phillips, T. G., et al. 2010, A&A, 518,

L108

Neufeld, D. A. & Wolfire, M. G. 2009, ApJ, 706, 1594

Neufeld, D. A., Zmuidzinas, J., Schilke, P., & Phillips, T. G. 1997, ApJL,

488, L141

Newman, R. C. & Axford, W. I. 1968, ApJ, 153, 595

Ochsendorf, B. B., Brown, A. G. A., Bally, J., & Tielens, A. G. G. M.

2015, ApJ, 808, 111

O’Dell, C. R. 2001, ARA&A, 39, 99

O’Dell, C. R. 2018, MNRAS, 478, 1017

O’Dell, C. R., Ferland, G. J., Henney, W. J., et al. 2015, AJ, 150, 108

O’Dell, C. R., Ferland, G. J., Porter, R. L., & van Hoof, P. A. M. 2011,

ApJ, 733, 9

O’Dell, C. R., Hartigan, P., Bally, J., & Morse, J. A. 1997, AJ, 114, 2016

254

BIBLIOGRAPHY

.......................................................................

O’Dell, C. R., Henney, W. J., Abel, N. P., Ferland, G. J., & Arthur,

S. J. 2009, AJ, 137, 367

O’Dell, C. R., Kollatschny, W., & Ferland, G. J. 2017, ApJ, 837, 151

Oey, M. S. & Clarke, C. J. 2007, arXiv e-prints, astro

Olivier, G. M., Lopez, L. A., Rosen, A. L., et al. 2021, ApJ, 908, 68

Ossenkopf, V., Röllig, M., Neufeld, D. A., et al. 2013, A&A, 550, A57

Pabst, C., Higgins, R., Goicoechea, J. R., et al. 2019, Nature, 565, 618

Pabst, C. H. M., Goicoechea, J. R., Teyssier, D., et al. 2020, A&A, 639,

A2

Palau, A., Sánchez-Monge, Á., Busquet, G., et al. 2010, A&A, 510, A5

Peimbert, M., Sarmiento, A., & Fierro, J. 1991, PASP, 103, 815

Pellegrini, E. W., Baldwin, J. A., Ferland, G. J., Shaw, G., & Heathcote,

S. 2009, ApJ, 693, 285

Perley, R. A., Chandler, C. J., Butler, B. J., & Wrobel, J. M. 2011,

ApJL, 739, L1

Phillips, T. G., Bergin, E. A., Lis, D. C., et al. 2010, A&A, 518, L109

Pilbratt, G. L., Riedinger, J. R., Passvogel, T., et al. 2010, A&A, 518,

L1

Pineda, J. L., Langer, W. D., & Goldsmith, P. F. 2014, A&A, 570, A121

Poglitsch, A., Waelkens, C., Geis, N., et al. 2010, A&A, 518, L2

Pound, M. W. & Wolfire, M. G. 2008, in Astronomical Society of the

Pacific Conference Series, Vol. 394, Astronomical Data Analysis Software

and Systems XVII, ed. R. W. Argyle, P. S. Bunclark, & J. R.

Lewis, 654

Purcell, C. R., Hoare, M. G., Cotton, W. D., et al. 2013, ApJS, 205, 1

Purser, S. J. D., Lumsden, S. L., Hoare, M. G., & Cunningham, N. 2018,

MNRAS, 475, 2

255

BIBLIOGRAPHY

.......................................................................

Qiu, K., Xie, Z., & Zhang, Q. 2018, ApJ, 855, 48

Quillen, A. C., Thorndike, S. L., Cunningham, A., et al. 2005, ApJ, 632,

941

Rangwala, N., Maloney, P. R., Glenn, J., et al. 2011, ApJ, 743, 94

Rathborne, J. M., Jackson, J. M., & Simon, R. 2006, ApJ, 641, 389

Rathborne, J. M., Simon, R., & Jackson, J. M. 2007, ApJ, 662, 1082

Ray, T., Dougados, C., Bacciotti, F., Eislöffel, J., & Chrysostomou,

A. 2007, in Protostars and Planets V, ed. B. Reipurth, D. Jewitt, &

K. Keil, 231

Reach, W. T., Koo, B.-C., & Heiles, C. 1994, ApJ, 429, 672

Reese, C., Stoecklin, T., Voronin, A., & Rayez, J. C. 2005, A&A, 430,

1139

Reid, M. J., Argon, A. L., Masson, C. R., Menten, K. M., & Moran,

J. M. 1995, ApJ, 443, 238

Reid, M. J. & Ho, P. T. P. 1985, ApJL, 288, L17

Reynolds, S. P. 1986, ApJ, 304, 713

Richer, J. S., Shepherd, D. S., Cabrit, S., Bachiller, R., & Churchwell,

E. 2000, in Protostars and Planets IV, ed. V. Mannings, A. P. Boss,

& S. S. Russell, 867

Risacher, C., Güsten, R., Stutzki, J., et al. 2018, Journal of Astronomical

Instrumentation, 7, 1840014

Risacher, C., Güsten, R., Stutzki, J., et al. 2016, A&A, 595, A34

Rivera-Soto, R., Galván-Madrid, R., Ginsburg, A., & Kurtz, S. 2020,

ApJ, 899, 94

Robberto, M., Soderblom, D. R., Bergeron, E., et al. 2013, ApJS, 207,

10

Robitaille, T. & Bressert, E. 2012, APLpy: Astronomical Plotting Library

in Python

256

BIBLIOGRAPHY

.......................................................................

Rodriguez, L. F. 1995, in Revista Mexicana de Astronomia y Astrofisica

Conference Series, Vol. 1, Revista Mexicana de Astronomia y Astrofisica

Conference Series, ed. S. Lizano & J. M. Torrelles, 1

Rodriguez, L. F., Garay, G., Curiel, S., et al. 1994, ApJL, 430, L65

Rodriguez, L. F., Ho, P. T. P., Torrelles, J. M., Curiel, S., & Canto, J.

1990, ApJ, 352, 645

Rodríguez, L. F., Moran, J. M., Franco-Hernández, R., et al. 2008, AJ,

135, 2370

Roelfsema, P. R., Helmich, F. P., Teyssier, D., et al. 2012, A&A, 537,

A17

Rosen, A. L., Offner, S. S. R., Sadavoy, S. I., et al. 2020, Space Science

Reviews, 216, 62

Rosero, V., Hofner, P., Claussen, M., et al. 2016, ApJS, 227, 25

Rosero, V., Hofner, P., Kurtz, S., et al. 2019, ApJ, 880, 99

Roueff, E., Dartois, E., Geballe, T. R., & Gerin, M. 2006, A&A, 447,

963

Salgado, F., Berné, O., Adams, J. D., et al. 2016, ApJ, 830, 118

San José-García, I., Mottram, J. C., Kristensen, L. E., et al. 2013, A&A,

553, A125

Sánchez-Monge, Á., Beltrán, M. T., Cesaroni, R., et al. 2014, A&A, 569,

A11

Sánchez-Monge, Á., Beltrán, M. T., Cesaroni, R., et al. 2013a, A&A,

550, A21

Sánchez-Monge, Á., Cesaroni, R., Beltrán, M. T., et al. 2013b, A&A,

552, L10

Sánchez-Monge, Á., Kurtz, S., Palau, A., et al. 2013c, ApJ, 766, 114

Sánchez-Monge, Á., López-Sepulcre, A., Cesaroni, R., et al. 2013d, A&A,

557, A94

257

BIBLIOGRAPHY

.......................................................................

Sánchez-Monge, Á., Palau, A., Estalella, R., Beltrán, M. T., & Girart,

J. M. 2008, A&A, 485, 497

Sanna, A., Moscadelli, L., Goddi, C., et al. 2019, A&A, 623, L3

Sanna, A., Moscadelli, L., Goddi, C., Krishnan, V., & Massi, F. 2018,

A&A, 619, A107

Schneider, N., Simon, R., Guevara, C., et al. 2020, PASP, 132, 104301

Schöier, F. L., van der Tak, F. F. S., van Dishoeck, E. F., & Black, J. H.

2005, A&A, 432, 369

Sewilo, M., Churchwell, E., Kurtz, S., Goss, W. M., & Hofner, P. 2004,

ApJ, 605, 285

Sewiło, M., Churchwell, E., Kurtz, S., Goss, W. M., & Hofner, P. 2011,

ApJS, 194, 44

Shaw, G., Ferland, G. J., Henney, W. J., et al. 2009, ApJ, 701, 677

Shepherd, D. S., Nürnberger, D. E. A., & Bronfman, L. 2004, ApJ, 602,

850

Shipman, R. F., Beaulieu, S. F., Teyssier, D., et al. 2017, A&A, 608,

A49

Shu, F. H., Adams, F. C., & Lizano, S. 1987, ARA&A, 25, 23

Simon, R., Stutzki, J., Sternberg, A., & Winnewisser, G. 1997, A&A,

327, L9

Simón-Díaz, S. & Stasińska, G. 2011, A&A, 526, A48

Sonnentrucker, P., Neufeld, D. A., Phillips, T. G., et al. 2010, A&A, 521,

L12

Spitzer, L. 1978, Physical processes in the interstellar medium (A Wiley-

Interscience Publication, New York: Wiley)

Stanke, T., McCaughrean, M. J., & Zinnecker, H. 2002, A&A, 392, 239

Stecklum, B., Caratti o Garatti, A., Klose, S., & Wiseman, P. 2017, The

Astronomer’s Telegram, 10842, 1

258

BIBLIOGRAPHY

.......................................................................

Strömgren, B. 1939, ApJ, 89, 526

Stutzki, J., Stacey, G. J., Genzel, R., et al. 1988, Physical conditions

in molecular clouds derived from sub-mm and far-IR spectroscopic

observations, Tech. rep., KONA SYMPOSIUM

Sugiyama, K., Nagase, K., Yonekura, Y., et al. 2017, PASJ, 69, 59

Tan, J. C. 2016, in From Interstellar Clouds to Star-Forming Galaxies:

Universal Processes, ed. P. Jablonka, P. André, & F. van der Tak, Vol.

315, 154–162

Tan, J. C., Beltrán, M. T., Caselli, P., et al. 2014, in Protostars and Planets

VI, ed. H. Beuther, R. S. Klessen, C. P. Dullemond, & T. Henning,

149

Tauber, J. A., Tielens, A. G. G. M., Meixner, M., & Goldsmith, P. F.

1994, ApJ, 422, 136

Thielemann, F. K., Hirschi, R., Liebendörfer, M., & Diehl, R. 2011,

Massive Stars and Their Supernovae, Vol. 812 (Springer), 153–232

Thompson, A. R., Clark, B. G., Wade, C. M., & Napier, P. J. 1980,

ApJS, 44, 151

Thompson, R. I. 1984, ApJ, 283, 165

Thompson, T. A., Quataert, E., & Murray, N. 2005, ApJ, 630, 167

Thummel, H. T., Nesbet, R. K., & Peyerimhoff, S. D. 1992, Journal of

Physics B: Atomic, Molecular and Optical Physics, 25, 4553

Tielens, A. G. G. M. 2010, The Physics and Chemistry of the Interstellar

Medium (Cambridge University Press)

Tielens, A. G. G. M. & Hollenbach, D. 1985a, ApJ, 291, 722

Tielens, A. G. G. M. & Hollenbach, D. 1985b, ApJ, 291, 747

Tielens, A. G. G. M., Meixner, M. M., van der Werf, P. P., et al. 1993,

Science, 262, 86

259

BIBLIOGRAPHY

.......................................................................

Timmes, F. X., Truran, J. W., Lauroesch, J. T., & York, D. G. 1997,

ApJ, 476, 464

Tiwari, M., Karim, R., Pound, M. W., et al. 2021, ApJ, 914, 117

Torrelles, J. M., Ho, P. T. P., Rodriguez, L. F., & Canto, J. 1985, ApJ,

288, 595

Torrelles, J. M., Patel, N. A., Curiel, S., et al. 2011, MNRAS, 410, 627

Troland, T. H., Heiles, C., & Goss, W. M. 1989, ApJ, 337, 342

Urquhart, J. S., Hoare, M. G., Lumsden, S. L., Oudmaijer, R. D., &

Moore, T. J. T. 2008, in Astronomical Society of the Pacific Conference

Series, Vol. 387, Massive Star Formation: Observations Confront

Theory, ed. H. Beuther, H. Linz, & T. Henning, 381

Vacca, W. D., Garmany, C. D., & Shull, J. M. 1996, ApJ, 460, 914

van der Tak, F. F. S. 2012b, Philosophical Transactions of the Royal

Society of London Series A, 370, 5186

van der Tak, F. F. S., Black, J. H., Schöier, F. L., Jansen, D. J., & van

Dishoeck, E. F. 2007, A&A, 468, 627

van der Tak, F. F. S., Ossenkopf, V., Nagy, Z., et al. 2012a, A&A, 537,

L10

van der Werf, P. P., Goss, W. M., & O’Dell, C. R. 2013, ApJ, 762, 101

van der Werf, P. P., Isaak, K. G., Meijerink, R., et al. 2010, A&A, 518,

L42

van der Wiel, M. H. D., Naylor, D. A., Makiwa, G., Satta, M., & Abergel,

A. 2016, A&A, 593, A37

van der Wiel, M. H. D., van der Tak, F. F. S., Ossenkopf, V., et al. 2009,

A&A, 498, 161

Walch, S. K., Whitworth, A. P., Bisbas, T., Wünsch, R., & Hubber, D.

2012, MNRAS, 427, 625

Walmsley, C. M., Natta, A., Oliva, E., & Testi, L. 2000, A&A, 364, 301

260

BIBLIOGRAPHY

.......................................................................

Walter, F., Brinks, E., de Blok, W. J. G., Thornley, M. D., & Kennicutt,

R. C. 2005, in Astronomical Society of the Pacific Conference Series,

Vol. 331, Extra-Planar Gas, ed. R. Braun, 269

Wang, P., Li, Z.-Y., Abel, T., & Nakamura, F. 2010, ApJ, 709, 27

Wang, Y., Jaffe, D. T., Evans, Neal J., I., et al. 1993, ApJ, 419, 707

Ward-Thompson, D. & Whitworth, A. P. 2015, An Introduction to Star

Formation (Cambridge University Press)

Weaver, R., McCray, R., Castor, J., Shapiro, P., & Moore, R. 1977, ApJ,

218, 377

Weilbacher, P. M., Monreal-Ibero, A., Kollatschny, W., et al. 2015, A&A,

582, A114

Weingartner, J. C. & Draine, B. T. 2001, ApJ, 548, 296

Werner, M. W., Roellig, T. L., Low, F. J., et al. 2004, ApJS, 154, 1

Whitworth, A. 1979, MNRAS, 186, 59

Williams, J. P. & Cieza, L. A. 2011, ARA&A, 49, 67

Williams, J. P. & McKee, C. F. 1997, ApJ, 476, 166

Wilson, T. L., Filges, L., Codella, C., Reich, W., & Reich, P. 1997, A&A,

327, 1177

Wolfire, M. G., McKee, C. F., Hollenbach, D., & Tielens, A. G. G. M.

2003, ApJ, 587, 278

Wood, D. O. S. & Churchwell, E. 1989, ApJS, 69, 831

Wu, Y., Wei, Y., Zhao, M., et al. 2004, A&A, 426, 503

Yang, B., Walker, K. M., Forrey, R. C., Stancil, P. C., & Balakrishnan,

N. 2015, A&A, 578, A65

Yorke, H. W., Tenorio-Tagle, G., Bodenheimer, P., & Rozyczka, M. 1989,

A&A, 216, 207

261

BIBLIOGRAPHY

.......................................................................

Young, E. T., Becklin, E. E., Marcum, P. M., et al. 2012, ApJL, 749,

L17

Young Owl, R. C., Meixner, M. M., Wolfire, M., Tielens, A. G. G. M.,

& Tauber, J. 2000, ApJ, 540, 886

Zinnecker, H. & Yorke, H. W. 2007, ARA&A, 45, 481

262

Interaction of Massive Stars with Gas Clouds in the Milky Way: from shooting stars to breaking bubbles

Create successful ePaper yourself

Delete template?

Save as template?