Mass functions and mass segregation in young starburst clusters [Elektronische Ressource] / presented by Andrea Stolte

English
173 Pages
Read an excerpt
Gain access to the library to view online
Learn more

Description

Dissertationsubmitted to theCombined Faculties for the Natural Sciences and for Mathematicsof the Ruperto-Carola University of Heidelberg, Germanyfor the degree ofDoctor of Natural Sciencespresented byDiplom-Physicist Andrea Stolteborn in DuisburgthOral examination 25 June 2003Mass Functions and Mass Segregationin Young Starburst ClustersReferees: Prof. Dr. Hans-Walter RixProf. Dr. Immo AppenzellerThesis AbstractThe Milky Way starburst clusters Arches in the Galactic Center region and NGC3603 in the Carinaspiralarmarestudiedwiththeaimtogaindeeperinsightintothestellarmassdistributioninstarburstclusters. The dense stellar population in both clusters is resolved in unprecedented detail with highangular resolution, near-infrared instruments. In the case of the Arches cluster, difiraction-limited,adaptive optics observations are analysed, and the achievements and limitations of ground-based vs.space-based difiraction-limited imaging are revealed by comparison with HST/NICMOS data. In thecase of NGC3603, seeing-limited JHKL photometry is used to derive colour-excess fractions, and iscomplemented by space-based Hfi data, both serving as tracers for circumstellar disks. Disk survivalin starburst clusters is discussed. The present-day mass function (MF) of both clusters is deducedfrom colour-magnitude diagrams. Radial variations in the MFs reveal a heavily mass-segregatedcore in both starburst clusters.

Subjects

Informations

Published by
Published 01 January 2003
Reads 17
Language English
Document size 18 MB
Report a problem

Dissertation
submitted to the
Combined Faculties for the Natural Sciences and for Mathematics
of the Ruperto-Carola University of Heidelberg, Germany
for the degree of
Doctor of Natural Sciences
presented by
Diplom-Physicist Andrea Stolte
born in Duisburg
thOral examination 25 June 2003Mass Functions and Mass Segregation
in Young Starburst Clusters
Referees: Prof. Dr. Hans-Walter Rix
Prof. Dr. Immo AppenzellerThesis Abstract
The Milky Way starburst clusters Arches in the Galactic Center region and NGC3603 in the Carina
spiralarmarestudiedwiththeaimtogaindeeperinsightintothestellarmassdistributioninstarburst
clusters. The dense stellar population in both clusters is resolved in unprecedented detail with high
angular resolution, near-infrared instruments. In the case of the Arches cluster, difiraction-limited,
adaptive optics observations are analysed, and the achievements and limitations of ground-based vs.
space-based difiraction-limited imaging are revealed by comparison with HST/NICMOS data. In the
case of NGC3603, seeing-limited JHKL photometry is used to derive colour-excess fractions, and is
complemented by space-based Hfi data, both serving as tracers for circumstellar disks. Disk survival
in starburst clusters is discussed. The present-day mass function (MF) of both clusters is deduced
from colour-magnitude diagrams. Radial variations in the MFs reveal a heavily mass-segregated
core in both starburst clusters. Dynamical timescales are estimated and interpreted with respect
to primordial and dynamical segregation. The implications for massive star and cluster formation
scenarios are discussed. Evidence for a low-mass cut-ofi is observed in the Arches MF, but not in
NGC3603, indicatingareducedformatione–ciencyforM• 10M starsintheGalacticCenter. Thisfl
environmental difierence has strong implications for the formation of stellar populations in galactic
nuclei and starburst galaxies.
˜Ubersicht der Dissertationsinhalte
Die Galaktischen Starburst Sternhaufen Arches im Galaktischen Zentrum und NGC3603 im Carina
Spiralarm wurden mit dem Ziel untersucht, ein tieferes Verst˜andnis fur˜ die stellar Massenverteilung
in Starburst Haufen zu gewinnen. Die dichte, stellare Population in beiden Sternhaufen konnte
mithilfe hochau ˜osender nah-infrarot Instrumente in einzigartigem Detail aufgel˜ost werden. Im Falle
des Arches Sternhaufens wurden beugungsbegrenzte, mit adaptiver Optik gewonnene Beobachtun-
gen im Hinblick auf die M˜oglichkeiten bodengebundener, beugungsbegrenzter Aufnahmen analysiert,
und technische Grenzen im Vergleich zu Weltraumbeobachtungen mit HST/NICMOS aufgezeigt. Im
Falle von NGC3603 wurden Infrarot-Farbexzesse aus JHKL Aufnahmen bestimmt, und gemein-
˜sam mit beugungsbegrenzten HST Hfi Daten verwendet um die Existenz und Uberlebensrate von
zirkumstellaren Scheiben in Starburst Haufen nachzuweisen. Die aktuelle Massenfunktion (MF) der
Starburst Haufen wurde aus Farb-Helligkeits-Diagrammen hergeleitet. Dynamische Zeitskalen wurden
abgesch˜atzt, die Efiekte primordialer oder dynamischer Massensegregation werden diskutiert, und die
Folgen fur˜ Entstehungsmodelle massereicher Sterne und Sternhaufen werden angesprochen. Evidenz
fur˜ einen Abbruch der MF unterhalb von M • 10M wird im Arches Sternhaufen beobachtet, je-fl
doch nicht in NGC3603, was auf eine reduzierte Entstehungsrate von Sternen niedriger Masse nahe
des Galaktischen Zentrums hindeutet. Dieser Ein u… der Sternentstehungsumgebung auf die MF hat
signiflkante Bedeutung fur˜ Sternentstehung in Galaxienzentren und Starburst Galaxien.Mass Functions and Mass Segregation in Young Starburst Clusters
NGC 3603
Arches
Sun
Orion
Portrait of the Milky Way - Painting by Jon Lomberg
Astronomical star and cluster data were used to obtain this view of the Milky Way’s spiral structure.
The position of the Sun and the Orion star-forming region are indicated. The Arches cluster close to
the Galactic Center and NGC 3603 in the tangent point of the Carina spiral arm are shown as
inserts (not to scale).dedicated to Rai Weiss
Astronomy is like leaf counting
Rai Weiss
Although the work was often as laborious and painstaking as leaf-counting,
I hope that even a hard-core physicist as Rai Weiss
can flnd his interesting bits and pieces in this thesis.Contents
1 Introduction 1
2 From Massive Star Forming Regions to Mass Functions 4
2.1e Star Forming Regions. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4
2.2 Massive Star Formation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6
2.3 The Stellar Mass Function - a short history . . . . . . . . . . . . . . . . . . . . . . . . 7
2.4 Milky Way Starburst Clusters . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8
2.4.1 Arches . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8
2.4.2 NGC 3603 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 10
3 Adaptive Optics Observations and Technical Analysis 15
3.1 Introduction to Adaptive Optics . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 15
3.1.1 Characterisation of atmospheric perturbations . . . . . . . . . . . . . . . . . . 15
3.1.2 of AO images . . . . . . . . . . . . . . . . . . . . . . . . . . . 17
3.1.3 Strehl ratio . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 17
3.1.4 Isoplanatic Angle . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 19
3.2 Gemini/Hokupa’a . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 21
3.2.1 Observations . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 21
3.2.2 HST/NICMOS data . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 31
3.3 Luminosity Functions and incompleteness efiects . . . . . . . . . . . . . . . . . . . . . 31
3.3.1 Integrated luminosity function . . . . . . . . . . . . . . . . . . . . . . . . . . . 31
3.3.2 Radial variation of the luminosity function . . . . . . . . . . . . . . . . . . . . 33
3.4 NAOS/CONICA . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 35
3.4.1 VLT/NAOS-CONICA observations . . . . . . . . . . . . . . . . . . . . . . . . . 35
3.4.2 Data Reduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 36
3.5 Image Quality . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 38
3.5.1 Strehl ratio vs. ux ratio inside the PSF kernel . . . . . . . . . . . . . . . . . . 38
3.5.2 Detectability of individual sources . . . . . . . . . . . . . . . . . . . . . . . . . 39
3.6 The Efiects of Oversampling . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 44
3.7 Deconvolution and higher Strehl images as coordinate input . . . . . . . . . . . . . . . 47
3.7.1 H-band low vs. high Strehl . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 47
3.7.2 Image deconvolution . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 47
3.8 Photometric Residuals . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 51
3.9 NAOS-CONICA luminosity functions . . . . . . . . . . . . . . . . . . . . . . . . . . . 52
3.10 Radial variations in the NACO LF . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 55
3.11 Colour-magnitude diagrams . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 58
3.12 Summary of the technical analysis . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 58
iii CONTENTS
4 The Arches Cluster 62
4.1 Photometric results from Gemini/Hokupa’a data . . . . . . . . . . . . . . . . . . . . . 62
4.1.1 Radial colour gradient . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 62
4.1.2 Extinction maps . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 65
4.1.3 Colour-magnitude diagrams . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 67
4.1.4 HST/NICMOS colour-colour diagram . . . . . . . . . . . . . . . . . . . . . . . 67
4.2 Implications for the mass function derivation . . . . . . . . . . . . . . . . . . . . . . . 69
5 The Mass Function of the Arches cluster 70
5.1 The mass function from Gemini/Hokupa’a data . . . . . . . . . . . . . . . . . . . . . . 70
5.1.1 Integrated mass function. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 70
5.1.2 Efiects of the chosen isochrone, bin size, and metallicity . . . . . . . . . . . . . 72
5.1.3 Radial variation in the mass function . . . . . . . . . . . . . . . . . . . . . . . . 72
5.1.4 Formation locus of massive stars . . . . . . . . . . . . . . . . . . . . . . . . . . 74
5.1.5 Comparison with cluster formation models . . . . . . . . . . . . . . . . . . . . 75
5.2 The Arches mass function derived from NACO data . . . . . . . . . . . . . . . . . . . 77
5.2.1 Integrated mass function. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 77
5.2.2 Radial variations in the NACO mass function . . . . . . . . . . . . . . . . . . . 80
5.2.3 Combining difierent binnings - a step towards binning independence . . . . . . 83
5.2.4 Away from binning - Cumulative Functions . . . . . . . . . . . . . . . . . . . . 88
5.2.5 Summary of Arches results from NACO . . . . . . . . . . . . . . . . . . . . . . 91
5.2.6 Dyamical parameters revised from NACO data . . . . . . . . . . . . . . . . . . 92
6 NGC 3603 96
6.1 NGC 3603 ISAAC data . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 96
6.1.1 Basic Data Reduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 97
6.1.2 Image combination . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 99
6.1.3 Photometry . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 99
6.1.4 L-band observations . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 100
6.2 Colour-Magnitude diagram of HD97950 . . . . . . . . . . . . . . . . . . . . . . . . . . 101
6.2.1 Colour-magnitude diagram and pre-main sequence turn-ofi . . . . . . . . . . . 101
6.2.2 Distance, extinction and age derived from the PMS . . . . . . . . . . . 101
6.2.3 Comparison with Yale evolutionary models . . . . . . . . . . . . . . . . . . . . 104
6.2.4 Binary candidate sequence. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 105
6.2.5 sequence and Hfi emission . . . . . . . . . . . . . . . . . . . . . . . . . 106
6.2.6 Stars with enhanced Hfi . . . . . . . . . . . . . . . . . . . . . . . . . . 106
6.2.7 L-band photometry . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 112
6.3 Colour-colour diagrams of HD97950 . . . . . . . . . . . . . . . . . . . . . . . . . . . . 112
6.3.1 Location of Hfi emission stars in the two-colour diagram . . . . . . . . . . . . . 117
6.4 Extinction map . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 118
6.5 Consequences for the mass function derivation. . . . . . . . . . . . . . . . . . . . . . . 120
7 The Mass Function in the central NGC3603 cluster 121
7.1 Physical parameters of NGC3603 entering the MF . . . . . . . . . . . . . . . . . . . . 121
7.1.1 Distance to NGC 3603 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 121
7.1.2 Metallicity of NGC 3603 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 121
7.1.3 Age of HD97950 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 122
7.2 General remarks on the mass function derivation in HD97950 . . . . . . . . . . . . . . 122
7.3 Mass function derivations . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 123
7.3.1 Incompleteness correction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 123CONTENTS iii
7.3.2 Field star correction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 124
7.3.3 Simple star counts . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 125
7.3.4 Individual dereddening . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 126
7.3.5 Binary rejection . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 127
7.3.6 correction - MS and transition region . . . . . . . . . . . . . . . . . . . 127
7.3.7 Binary - statistical PMS correction . . . . . . . . . . . . . . . . . . . 128
7.4 Discussion of the resultant mass functions . . . . . . . . . . . . . . . . . . . . . . . . . 128
7.5 Radial variation in the mass function of HD97950 . . . . . . . . . . . . . . . . . . . . 130
7.6 Mass segregation in HD97950 and dynamical timescales . . . . . . . . . . . . . . . . . 136
7.6.1 Cumulative Functions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 137
7.6.2 Fraction of high- to low-mass stars . . . . . . . . . . . . . . . . . . . . . . . . . 139
7.6.3 Dynamical Timescales . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 140
8 From Orion to R136 - a structural comparison 143
8.1 NGC3603 and 30 Doradus . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 143
8.1.1 A morphological comparison . . . . . . . . . . . . . . . . . . . . . . . . . . . . 143
8.1.2 The nature of the massive stars . . . . . . . . . . . . . . . . . . . . . . . . . . . 145
8.1.3 Age spread . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 146
8.2 Arches, NGC3603 and Orion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 146
8.3 Stellar mass distributions on four scales . . . . . . . . . . . . . . . . . . . . . . . . . . 149
9 Summary and Outlook 151
9.1 Technical analysis of AO data in a crowded stellar fleld . . . . . . . . . . . . . . . . . 152
9.2 Scientiflc Results . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 152
9.2.1 Colour-magnitude and Colour-colour diagrams . . . . . . . . . . . . . . . . . . 152
9.3 Mass Functions and Mass Segregation . . . . . . . . . . . . . . . . . . . . . . . . . . . 154
Abbreviations
AO Adaptive Optics
CF Cumulative Function
CMD Colour-Magnitude Diagram
GC Galactic Center
GMC Giant Molecular Cloud
IMF Initial Mass Function
MC Molecular Cloud
MF Mass Function
NACO NAOS/CONICA
TCD Two-Colour DiagramWe had the sky, up there, all speckled with stars,
and we used to lay on our backs and look up at them,
and discuss whether they was made,
or only just happened.
Jim allowed they was made,
but I allowed they happened;
I judged it would have took too long to MAKE so many.......
Huckleberry Finn by Mark TwainChapter 1
Introduction
Star-forming regions come in a wide variety of appearances - from sparse associations such as the
low density Taurus region to dispersed OB associations, from small clustered agglomerations with up
to a few hundred stars to massive, rich clusters such as the Orion nuclear cluster with more than
2000 members. While these are the dominant modes of star formation in our Galaxy, they are far
from giving us a complete notion of star formation in the Universe. Giant Hii regions such as the
30 Doradus region in the Large Magellanic Cloud (LMC) produce young star clusters with masses
comparable to the lower tail of the Milky Way globular cluster mass scale. At the high-mass end, very
compact massive clusters condense in the extreme conditions in the tidal tails of merging galaxies,
where tidal shear forces concentrate interstellar material. In the Antennae galaxies, young clusters
4 6span a mass range from 10 to 10 M (Whitmore et al. 1999), a similar scale as found in globularfl
clusters. Even more extreme star-forming environments are found in the nuclei of starburst galaxies
and distant quasars, representing star-formation during an early stage of the Universe.
Mostofthesemassivesourcesareatlargedistances,andcanconsequentlynotberesolvedintoindi-
vidual stars. Numerous assumptions have to be made to understand their observed integrated proper-
ties. The underlying stellar population has to be modelled using population synthesis techniques. The
basic model assumptions needed to reproduce a stellar population from integrated quantities are the
age or age range, the metallicity and the stellar mass distribution, the so-called mass function. While
the age and metallicity might be deduced from integrated spectra, there is no means to derive the
mass function from integrated quantities alone. At the same time, the mass is the crucial parameter
determining the evolution of a star, and thus the evolution of a stellar population. As this evolution
is dominated by the core and shell burning processes, with the duration and time of occurance of each
phase depending mainly on the star’s mass for a given metallicity, the inherent physical processes are
fairly well understood, allowing detailed stellar evolution modelling. If the age and metallicity are
known, the stellar mass distribution is the remaining uncertainty on the way to understand a stellar
population. Therefore, it is particularly important to understand the distribution of stellar masses at
the instance of birth, the initial mass function.
The observation that most nearby, resolved star-forming regions display a very similar power-law
mass spectrum with a small range in exponents led to the concept of a \standard" or \universal"
initial mass function. As stars form from fragmentation of molecular clouds, the flnal mass of a star
3=2 ¡1=2was long thought to be determined by the Jeans-mass, M »T ‰ , deflned as the mass where aJ
spherical cloud fragment becomes gravitationally unstable and collapses. As molecular clouds display
a diversity of densities, temperatures and turbulent structure proflles, this scenario suggests a strong
dependence of the emergent mass distribution on the molecular environment. Typical Jeans masses
range from 1¡ 100M , in the right mass range to form the observed population of stars. Thisfl
concept, however, faces two severe limitations. First of all, the collapse of a massive star occurs so
rapidlythatradiationpressureafterhydrogenignitioninvertstheinfallassoonasamassof15¡20Mfl
is accumulated. Secondly, if the environmental properties are the crucial parameters determining the
1