Simulations of binary galaxy mergers and the link with Fast Rotators, Slow Rotators, and Kinematically Distinct Cores

1 MB·English

by  Maxime Bois

**VolumeTitle**
ASPConferenceSeries,Vol.**VolumeNumber**
**Author**
(cid:13)c**CopyrightYear**AstronomicalSocietyofthePacific
Simulations ofbinary galaxymergers andthe linkwithFast
Rotators, SlowRotators, andKinematicallyDistinctCores
2
1
0 MaximeBois1,EricEmsellem2,Fre´de´ricBournaud3,KatherineAlatalo4,Leo
2 Blitz4,MartinBureau5,MicheleCappellari5,RogerL.Davies5,TimothyA.
Davis5,P.T.deZeeuw2,6,Pierre-AlainDuc3,SadeghKhochfar7,Davor
n
a Krajnovic´2,HaraldKuntschner8,Pierre-YvesLablanche2,RichardM.
J McDermid9,RaffaellaMorganti10,11,ThorstenNaab12,TomOosterloo10,11,

4 MarcSarzi13,NicholasScott5,PaoloSerra10,Anne-MarieWeijmans14 andLisa
M.Young15
]
O
1Observatoire deParis,France;2ESO,Garching,Germany;3CEA,
C Paris-Saclay, France;4University ofCalifornia, USA;5UniversityofOxford,
h. UK;6LeidenUniversity, theNetherlands; 7MPE,Garching,Germany;8ESO,
p Garching,Germany;9GeminiObservatory, Hilo,USA;10ASTRON,
- Dwingeloo, TheNetherlands; 11UniversityofGroningen, TheNetherlands;
o
r 12MPA,Garching,Germany;13UniversityofHertfordshire, UK;14University
st ofToronto,Canada;15NewMexicoInstitute ofMiningandTechnology,
a Socorro, USA
[

1 Abstract. We study the formation of early-type galaxies (ETGs) through mergers
v with a sample of 70 high-resolutionnumerical simulations of binary mergers of disc
5
galaxies. These simulationsencompassvariousmass ratios, initial conditionsand or-
8
bitalparameters. We findthatbinarymergersofdiscgalaxieswithmassratiosof3:1
8
and 6:1 are nearly always classified as Fast Rotators according to the ATLAS3D cri-
0
terion: they preserve the structure of the inputfast rotatingspiral progenitors. Major
.
1 discmergers(massratiosof2:1and1:1)leadtobothFastandSlowRotators. Mostof
0 theSlowRotatorsholdastellarKinematicallyDistinctCore(KDC)intheir1-3central
2
kilo-parsec: theseKDCsarebuiltfromthestellarcomponentsoftheprogenitors. The
1
massratiooftheprogenitorsisafundamentalparameterfortheformationofSlowRo-
:
v tatorsinbinarymergers,butitalsorequiresaretrogradespinfortheprogenitorgalaxies
i withrespecttotheorbitalangularmomentum.Theimportanceoftheinitialspinofthe
X
progenitorsisalsoinvestigatedinthelibraryofgalaxymergersoftheGalMerproject.
r
a
1. Introduction
Numerical simulations have clearly shown that the global characteristics of the rem-
nantsofbinarymergersbetweentwoequal-massspiralgalaxies,calledmajormergers,
resemble those of Early-Type Galaxies (i.e ellipticals & lenticulars), hereafter ETGs
(Barnes 1992). Modern work tends to quantify in detail the properties of major and
minor merger remnants, together withthorough comparisons withobserved properties
ofETGs(Bournaud etal.2005).
In this paper, we further examine the role of the initial conditions (mass ratios,
impact parameters, relative velocities, inclinations andspins oftheprogenitors) onthe
1
2 MaximeBoisetal.
All Mass Ratios Mass Ratio 1:1 Mass Ratio 2:1 (10% gas)
0.5 0.5 0.5
0.4 0.4 0.4
Re0.3 Re0.3 Re0.3
λ λ λ
0.2 0.2 0.2
0.1 0.1 0.1
0.0 0.0 0.0
0.0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8
ǫ ǫ ǫ
e e e
Mass Ratio 2:1 (33% gas) Mass Ratio 3:1 Mass Ratio 6:1
0.5 0.5 0.5
0.4 0.4 0.4
Re0.3 Re0.3 Re0.3
λ λ λ
0.2 0.2 0.2
0.1 0.1 0.1
0.0 0.0 0.0
0.0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8
ǫ ǫ ǫ
e e e
Figure 1. λ −ǫ diagramfor all simulationsof binarymergersof disc galaxies.
R
The different mass ratios are labelled on top of each panels. The red symbols are
for the projection which maximizes the ellipticity for a given remnant. The limit
definingtheslow/fastcategoriesfromATLAS3D isplottedasthesolidblackline.
morphology and the kinematics of the remnants of binary galaxy mergers. We build
two-dimensional momentum (intensity, velocity and velocity dispersion) maps of the
mergerremnantsandanalysetheirapparentproperties,directlylinkedwiththeirorbital
structures(Jesseit etal.2005). Usingtwo-dimensional mapsenablesustocompareour
mergerremnantsdirectlywiththecomplete,volume-limitedsampleof260localETGs
fromtheATLAS3D survey(Cappellari etal.2011).
2. Sampleofsimulations
Weusetheparticle-meshcodedescribedinBournaudetal.(2008)andreferencestherein.
The softening lenght is 58 pc and the number of particles is 2 × 106 for each com-
ponents (gas, stars and dark matter) for each galaxies. This resolution is required to
resolve properly the fluctuations of the gravitational potential during the merger, i.e
violent relaxation, which can significantly impact the morphology and kinematics of
mergerremnants(Boisetal.2010).
Wehave simulated binary mergers of spiral galaxies with mass ratios of 1:1, 2:1,
3:1and6:1. Thefirstprogenitor,whichisdefinedasthemostmassiveforunequal-mass
mergers, hasabaryonic massof1.3×1011 M . Thebulge fraction is B/T = 0.20and
⊙
the gas fraction in the disc is usually 10 per cent (33 per cent in some 2:1 mergers).
Thisinitial galaxy isrepresentative for aSb spiral galaxy. Theother progenitor has its
total mass determined by the mass ratio and a gas fraction of 10 per cent. The main
MaximeBoisetal. 3
0.6
Slow-rotators S+S dir
20 Fast-rotators S+S ret
0.5
15 0.4
λR0.3
10
0.2
5
0.1
080 60 40 20 i0 20 40 60 80 0.00.0 0.1 0.2 0.3 0ǫ.4 0.5 0.6 0.7 0.8
Sb
Figure2. Left:Fractionofslow/fastrotatorsasafunctionoftheinclinationofthe
mainprogenitorwithrespecttothemergerorbitalplaneformassratio1:1,2:1g10,
2:1g33. ThenegativeinclinationoccurswhenthespinoftheSbprogenitorandthe
spinofthe orbitalangularmomentumareanti-parallel. Right: λ −ǫ diagramfor
R
theGalMermergerremnants. Inwhitesymbolsthespiralprogenitorshaveadirect
spin,inblacksymbolsthespiralshavearetrogradespinwrttheorbitalspin.
difference with the first progenitor is the lower bulge fraction with B/T = 0.12. This
secondprogenitor galaxyisdenotedastheScspiralprogenitor.
We simulate 70 galaxy mergers with different initial conditions. We vary five
parameters to study their importance on the properties of the final remnant: the mass
ratio,theinclinationofthespiralwrttheorbitalplane,theimpactparameter,therelative
velocity of the spirals and the orientation of the spin of the spirals. The spin of the
spirals can be then either direct (parallel) or retrograde (anti-parallel) wrt the orbital
spin.
3. Populationsoffastandslowrotators
To quantify the global kinematics of each system, using the velocity and velocity dis-
persion maps, we measure the λ parameter, which is a robust proxy for the stellar
R
projected angular momentum defined in Emsellemetal. (2007). This parameter is
used to separate ETGs in two families: the fast rotators and the slow rotators (see
Krajnovic´ etal. 2011; Emsellemetal. 2011, for detailed properties on these families).
To quantify the morphology, we measure the ellipticity ǫ = 1 − b/a where a and b
are the semi major- and minor-axes, respectively. To compute these two parameters,
weprojectoursimulatedmergerremnantsover200differentsviewinganglestoobtain
statistically representative results.
3.1. High-resolution simulations
Figure 1 shows the λ − ǫ diagram for the 70 high-resolution simulations of binary
R
mergersfortheir200projections. Allprojectedgalaxieswhichareabove(resp. below)
the black line are classified as fast rotators (resp. slow rotators). We first note that
almost all the merger remnants formed with a 3:1 or 6:1 are classified as fast rotators.
They preserved the structure of the input fast rotating spiral progenitors. Major spiral
mergers (withmassratios of1:1and2:1)leadtobothfastandslowrotators. Alookat
4 MaximeBoisetal.
their velocity maps reveals that most of the slow rotators hold a stellar Kinematically
Distinct Core (KDC) in their 1-3 central kilo-parsec: these KDCs are built from the
stellar components oftheprogenitors. Themassratio isthen afundamental parameter
fortheformationofslowrotators inbinarymergers.
The left panel of Figure 2 shows the number of fast and slow rotators for the 1:1
and 2:1 mergers as a function of the inclination of the Sb spiral progenitor (a negative
inclination occurs when the spin of the spiral is retrograde wrt the orbital spin). The
formation of the slow rotators requires a retrograde spin for the Sb spiral with respect
totheorbital angular momentum. Duetoitsmassive bulge, the spinofthe Sbspiral is
preserved during the merger while the Sc spiral (with a lower bulge) always acquires
the direct spin of the orbit. It thus forms a merger remnant with two contributions in
counter-rotation and forms, in our simulations, an apparent KDC.These slow rotating
merger remnants arethen representative ofthe2-σ galaxies observed intheATLAS3D
survey(Krajnovic´ etal.2011;Emsellemetal.2011). Theproperties oftheseobserved
galaxiesindicatethattheyarecomposedoftwocounter-rotating componentsandcould
havebeenformedviaasinglebinarymerger(seealsoCrockeretal.2009).
3.2. TheGalMersimulations
TheGalMerprojectconsistsofoneofthelargestpubliclyavailablesampleofnumerical
simulations of interacting galaxies (Chilingarian etal. 2010): with a high number of
simulations butarelativelylowresolution (atotalnumberofparticlesof1.2×105 and
asoftening lengthof280pc).
To confirm the importance of the spins’ orientation of the progenitors, we have
selected in the GalMer database the simulations of 1:1 spiral – spiral mergers. The
right panel of Figure 2 shows the λ −ǫ diagram for the edge-on view of the merger
R
remnants. This plot confirms our previous findings: merging spirals with a retrograde
orientation wrt the orbital spin produces a remnant with a lower angular momentum
content. However, there is a large scatter in the λ values of these simulated galaxies.
R
This is probably due to the lower resolution of the simulations: a look at the velocity
mapsofthemergerremnantsrevealsahighlevelofnumericalnoise.
References
BarnesJ.E,1992,ApJ,393,484
BoisM.,etal.,2010,MNRAS,406,2405
BournaudF.,JogC.J.,CombesF.,2005,A&A,437,69
BournaudF.,DucP.-A.,EmsellemE.,2008,MNRAS,389,8
CappellariM.,etal.,2011,MNRAS,413,813
ChilingarianI.V.,etal.,2010,A&A,518,61
CrockerA.F.,etal.,2009,MNRAS,393,1255
EmsellemE.,etal.,2007,MNRAS,379,401
EmsellemE.,etal.,2011,MNRAS,414,888
JesseitR.,NaabT.,BurkertA.,2005,MNRAS,360,1185
Krajnovic´D.,etal.,2011,MNRAS,414,2923