Extrasolar gas giants; eccentric and close-in planets models for orbital evolution Formation of giant planets; core accretion model M-a distribution dependence on stellar metallicity/mass Origin of diversity of giant planets Planetary Formation (architectures, theory) Shigeru Ida (Tokyo Institute of Technology) OUTLINE disk surface density model is very important
32
Embed
Planetary Formation (architectures, theory) Shigeru Ida ...nexsci.caltech.edu/conferences/2005/disks05/Invited_Talks/ida.pdf · Mizuno, Bodenheimer coagulation of planetesimals gas
This document is posted to help you gain knowledge. Please leave a comment to let me know what you think about it! Share it to your friends and learn new things together.
Transcript
Extrasolar gas giants; eccentric and close-in planets
models for orbital evolutionFormation of giant planets; core accretion model
M-a distributiondependence on stellar metallicity/mass
Origin of diversity of giant planets
Planetary Formation (architectures, theory)Shigeru Ida (Tokyo Institute of Technology)
OUTLINE
disk surface density model is very important
Orbits of discovered extrasolar planets
a(1-e)
asmall a
close-in planetshigh e
eccentric planets
mas
s x
sini
[M⊕
]
semimajor axis a [AU] a [AU]ec
cent
ricity
e
a(1+e)
detection limit
Origin of eccentric planets: jumping jupiterRasio & Ford(1996), Weidenschilling & Marzari(1996), Lin & Ida(1997),….
Δa [rHill]
Marzari & Weidenschilling(2002)
tcrosst cr
oss[y
r]
108
106
102
104
real
istic
rang
emore than 3 giant planets in circular orbitsOrbit crossing starts at ~tcrossOne is ejected. The others remain in stable eccentricorbits.
other model:tidal interaction with a gas disk. limited to e < 0.3?e.g., Sari & Goldreich (2004)
Origin of close-in planets: orbital migrationLin et al. (1996)
a giant planet at > a few AUThe planet opens up a gap
planet’s perturbation > viscous diffusion
The planet migrates inwardwith disk accretion
[If disk gas dissipates soon after the formation, no migration.]
The migration may stop near disk inner edge
⊕−≈> MM 21010p
planet’s perturbation
viscous diffusion
diffmig ττ ≈
Kley’s talk
Origin of close-in planets: slingshot
If pericenter distance a(1-e) < 0.05AU during Jumping Jupiter processe is dampedrare? (0.95 < e < 1.0)
Ford et al. (2001): 2 planets case, << 1%Marzari & Weidenschilling (2002): 3 planets case
- stable stage after ejection, ~10%Bessho, Tanaka, Ida (in prep.): 3 planets case,
tidal damping included in orbit integration- during orbit crossing, ~30% Not rare
Migration may be primary.But slingshot could also contribute.
Formation of gas giant planets(that will become close-in, eccentric, or solar-system type
giant planets)
Ida & Lin (2004a, ApJ, 604, 388-413)
formation sites of different kinds of gas giantsimplication from the results in Ida & Lin (2004a)
semimajor axis a [AU]0.1
0.11 10 100
1
eccentric
10solar-system type
close-insurface density of planetsimals
in disks
MMSNΣΣ /
formation sites of different kinds of gas giantsimplication from the results in Ida & Lin (2004a)
0.10.1
1 10 100
1
10the most massive disks
disks similar to that formed our solar system
the least massive disks
surface density of planetsimals
in disks
semimajor axis a [AU]
MMSNΣΣ /
formation sites of different kinds of gas giantsimplication from the results in Ida & Lin (2004a)
0.10.1
1 10 100
1
eccentric
10solar-system type
close-insurface density of planetsimals
in disks
semimajor axis a [AU]
MMSNΣΣ /
protoplanatery disk
gas giants
Models for formation of gas giants
terrestrialplanets
planetesimals
cores107 y
106 y
104 y
core accretion scenarioSafronov, Hayashi
Mizuno, Bodenheimer
coagulation of planetesimals
gas accretion onto cores
grav. instability scenario
Cameron, Boss
dust (micron) to planetsimals (km)
disk self-grav. instability
102-3 y
108y
107 y
106y
104 y
( ) y pp
p
310710 -MMMM
⊕≈&
deterministic planet formation modelbased on core accretion scenario
Ida & Lin (2004a, ApJ, 604, 388-413)
(atmosphere collapses)gas accretion
(clear gap;global depletion)gas inflow stops
Lin&Papaloizou (1985,PPII)
(partial gap)orbital migration
Lin&Papaloizou (1985,PPII)
⊕−>≈ M M 10core 5
⊕−> MM 2p 1010
depp τ>> ⊕− tMM ;1010 32
jumping jupiter:neglected
Ikoma et al. (2000,ApJ)
, aini =(integration on 109y)⇒ Mp, afinal
dust to planetesimals: not discussed
iniΣ
protoplanatery disk
gas giants
terrestrialplanets
planetesimals
cores107 y
106 y
108y
107 y
106y
104 yplanetesimal coagulation:
Kokubo & Ida (2002, ApJ)
type-II migrationincluded
planet’s perturbation
viscous diffusion
type-I migrationnot included
disk torque imbalance
yr AU p
,Imig,
MMSNg
g2/1
1
11510
≈
⊕
a-
MM
-
ΣΣ
τ
⊕−> MM )10010(
very rapid: neglected in our simulationdiscussed later
⊕−> MM )11.0(
yr AU J
p
g,
gIImig,
1-
MMSN
2/1
1
1
10610 4
≈ −
a-
MM α
ΣΣ
τ
NelsonKley
disk surface densitygasdust
Disk surface density model
ice ice
AUice
>−
≈<=
)(43)3(1
aaaa
η
( ) 2g/cmAUdepdiskgas 5.1
12400/ −− ×= aef t τΣ
( ) 2g/cmAUicediskdust 5.1110 −×= af ηΣ
Min. Mass Solar Nebula(disk that formed solar system)
orbital radius a1AU 10AU
diskf
diskf
(planetesimals)
dustΣ
gasΣ
icea
Σlog
aice= 2.7(M*/M )2 AU
gas giants(yellow region)• fdisk ~1 (MMSN): intermediate a (> aice)
-- small a: core isolation mass is too small;
-- large a: core growth istoo slow;
• fdisk > 5 (massive disks) :broad a, even inside aice
Formation sites of gas giantsIda & Lin (2004a, ApJ, 604, 388)
initial orbital radius aini [AU]
fdisk
icea
0.01M⊕
final planet massM*=1M τdep=107y
M⊕
M⊕M⊕
M⊕M⊕
isolationslow growth
100
2/34/3 )( icediskiso ηfaM ∝
110/27 )( −∝ icediskgrow ηfat
prediction of M-a distribution: Monte Carlo simulation
surface density distribution
gasdust
fdisk =0.1-10 Gaussian distribution in log scale
a - distribution : (0.1-100AU)disk lifetime:disk viscosity: α=1x10-4
type-II migration is included; artificially stopped at 0.04AU
0.001 0.1
Taurusρ Oph
N
0.1 1 10
fdisk
Beckwith & Sargent (1996)MMSNgas,dep
diskgas Σ×=Σ− τ/t
ef
MMSNdust,diskdust Σ=Σ f(planetesimals)
aa ∝∆yrsdep
76 1010 −=τ)10( AUfewaat depdiff −≈≈ aττ
final a [AU]
Pla
net m
ass
[M⊕
]
close-in planets gas giants
icy planets
terrestrial planets
1MJup= 320M⊕
● Mgas > 10Msolid● Mgas < Msolid a > aice● Mgas < Msolid a < aice
Prediction of Mp-a distribution Ida & Lin (2004a, ApJ, 604, 388)
G dwarfs
Pla
net m
ass
Mp
[M⊕
]
time106 y
Mp
100M⊕
gas disk
10M⊕
1M⊕
deficit
108 y107 y
Ida & Lin (2004a)
Planet Desert
final a [AU]
Prediction of Mp-a distribution Ida & Lin (2004a, ApJ, 604, 388)
detection probabilityIda & Lin (2004b, ApJ, 616, 567)
metallicity [Fe/H]
obs.Fischer&Valenti(2003)
% s
tars
with
de
tect
able
pla
nets
30
20
10
0
theoretical model- planets at 0.04AU: excluded
G dwarfs
0.5-0.5 0
Stellar mass dependenceF, G, K vs. M
Ida & Lin (2005, ApJ, in press)astro-ph/0502566
Stellar mass dependence of disksM
K
G
F
M dwarfs (mass M*=0.1-0.5 M )most abundant in a galactic diskon-going survey
gas giants are rare(~1/10 of FGK)isolated close-in neptune(GJ436; 0.03AU, 20M⊕)
Stellar mass dependence of disks
surface density distribution of disks
gasdust
• lifetime: indep. of M* (for M to F dwarfs)
depsungas or τΣ /
*21)/( teMM −∝21)/( *
orsundust MM∝Σ
Natta et al. 2004, Muzerolleo et al. 2004
Beckwith & Sargent 1996
logΣ
M* = 1M0.2M 0.4M
Min.MassSolar Nebula
x10x0.1 log normal
2* )/( sundisk MMM ∝&
yrsdep76 1010 −=τ
Planet mass [M⊕ ]
G
M
K
F
G
F
K
M
GJ436
M dwarfs- close-in Neptunes- Jupiters are rare
0.2M
0.4M
0.6M
1.0M
1.5M
distribution of planet mass([Fe/H]=0)Ida & Lin (2005, ApJ)
FGK dwarfs- close-in Jupiters- Jupiters are abundant
close-in:afinal<0.05AU distant:afinal=0.1-1AU0.2M
0.4M
0.6M
1.0M
1.5M
1 10 102 103 104 1 10 102 103 104
M
M
Ida & Lin (2005, ApJ, in press) astro-ph/0502566
M dwarfs vs. FGK dwarfs
• FGK: cores at > a few AU migrate after full gas accretion(>100M⊕)
close-in & distant Jupiter-mass planets
M: cores (~10M⊕) at ~1AUmigrate without gas accretion
low few Jupiterssmall aice icy cores at ~1AUgap opens at ~10M⊕[low T, weak stellar gravity]
at inner edge, gas accretion stops at ~10-20M⊕
close-in Neptune-mass planets (isolated, icy)
dustΣ
aice
10.3 3 [AU]0.1
10.3 3 [AU]0.1
aice=2.7(M*/M )2 AU
stellar mass [M ]
% s
tars
with
de
tect
able
pla
nets
(jup
iters
) 20
15
10
5
00.30.1 1
2sun* )/( MM∝Σ
1sun* )/( MM∝Σ
theoretical model- planets at 0.04AU: excluded
Σsmall icelarge a
detection probabilityIda & Lin (2005, ApJ, in press) astro-ph/0502566
[Fe/H]=0
M K G F
isolated close-in neptunes around M dwarfsicy planets in warm environment300-600K (at 0.04AU for M*=0.2-0.4M )
• habitable “ocean planets”? Leger et al. (2004)• marginally observable at present
habitable “ocean planets”?
They could be more abundant than Earth-like planets around G stars
habitable zone
tidal lock0.1 10.3 [AU]
icy planet
ocean planet ice boundary
The calibrated theory also predicts distribution of extrasolar terrestrial planets. (core accretion model)
SummaryMp-a distribution of extrasolar planetsbased on core accretion model many implications:- diversity of gas giant planets- metallicity dependence- stellar mass dependense (M stars)
disk surface density (a < 10AU) model is crucialgas & planetesimals(dust), stellar mass/metallicity dependence