京都大学宇宙物理学教室佐々木貴教 - Sasaki Takanori Online

地球の形成と初期進化：
生命誕生の場ができるまで
京都大学宇宙物理学教室
佐々木貴教
本日の発表内容
✤
太陽系形成論の簡単なレビュー!
✤
“ハビタブルゾーン” と “ハビタブルプラネット”!
✤
いかにして “地球” をつくるか？ −地球の初期進化についての新しいシナリオの提案−
太陽系形成論の簡単なレビュー
太陽系形成標準理論（京都モデル）
巨大氷惑星形成
©Newton Press
原始太陽系円盤の組成
一般に円盤質量の99%はガス（水素・ヘリウム）
残りの1%がダスト（固体成分）
最小質量円盤モデル（京都モデル）
・現在の太陽系の惑星の固体成分（約10-4M太陽）
→ すりつぶして円盤状にならす
・固体成分の約100倍の質量のガス成分を加える
原始太陽系円盤の初期質量は約10-2M太陽
重力と遠心力の釣り合いから半径は約100AU
微惑星の形成
Eimp = -
~
ダストの合体成長
d f
→ 微惑星形成
dN
V
V0
p dV
~
2.40 f N
3
2
Eimp
Eroll
Suyama et al. submitted to ApJ
微惑星の円盤が形成
[g/cm3]
微惑星の合体成長
数kmサイズの
微惑星が形成
↓
互いに衝突・合体
を繰り返し成長
暴走的成長
大きい粒子ほど成長が速い
!
秩序的成長
全ての粒子が同じ速度で成長
暴走的成長の様子
KOKUBO AND IDA
質量 [1023g]
KOKUBO AND IDA
軌道離心率
20
最大の天体
平均値
FIG. 4. Time evolution of the maximum mass (solid curve) and the mean
mass (dashed curve) of the system.
than this range are not statistically valid since each mass bin often
has only a few bodies. First, the distribution tends to relax to a
decreasing function of mass through dynamical friction among
(energy equipartition of) bodies (t = 50,000, 100,000 years).
Second, the distributions tend to flatten (t = 200,000 years). This
is because as a runaway body grows, the system is mainly heated
by the runaway body (Ida and Makino 1993). In this case, the
eccentricity and inclination of planetesimals are scaled by the
時間 [年]
FIG. 4. Time evolution of the maximum mass (solid curve) and the mean
mass (dashed curve) of the system.
微惑星の暴走的成長
軌道長半径 [AU]
FIG. 3. Snapshots of a planetesimal system on the a–e plane. The circles
represent planetesimals and their radii are proportional to the radii of planetesi-
→
原始惑星が誕生する
than this range are not statistically valid since each mass bin often
has only a few bodies. First, the distribution tends to relax to a
寡占的成長の様子
FORMATION OF PROTOPLANETS FROM PLANETESIMALS
23
＝
各場所で微惑星が暴走的成長
→ 等サイズの原始惑星が並ぶ
軌道離心率
寡占的成長とよぶ
各軌道での原始惑星
質量 [kg] 形成時間 [yr]
軌道長半径 [AU]
FIG. 7.
Snapshots of a planetesimal system on the a–e plane. The cir-
地球軌道
1×10
7×10
木星軌道
3×10
4×10
天王星軌道
8×10
2×10
FIG. 8.
The number of bodies in linear mass bins is plotted for t = 100,000,
原始惑星から惑星へ
snow line
)
原始惑星の質量( [地球質量]
地球型惑星
原始惑星同士の合体
!
巨大ガス惑星
原始惑星のガス捕獲
!
軌道長半径( [AU]
)
巨大氷惑星
原始惑星そのまま
ジャイアントインパクト
KOKUBO, KOMIN
軌道離心率
1134
軌道長半径 [AU]
Fig. 2.—Snapshots of the system on the a-e (left) and a-i (right) planes at t ¼ 0, 1
are proportional to the physical sizes of the planets.
長い時間をかけて原始惑星同士の軌道が乱れる
planets is hnM i ’ 2:0 Æ 0:6, which means that the typical resulting system consists of two Earth-sized planets and a smaller
planet. In this model, we obtain hna i ’ 1:8 Æ 0:7. In other words,
one or two planets tend to form outside the initial distribution of
protoplanets. In most runs, these planets are smaller scattered
planets. Thus we obtain a high efficiency of h fa i ¼ 0:79 Æ 0:15.
The accretion timescale is hTacc i ¼ ð1:05 Æ 0:58Þ ; 108 yr. These
results are consistent with Agnor et al. (1999), whose initial con-
→ 互いに衝突・合体してより大きな天体に成長
巨大天体衝突による月形成
原始地球に火星サイズの
原始惑星が衝突
飛び散った破片が地球の
周囲に円盤を形成
円盤中で月が誕生！
巨大ガス惑星の形成
原始惑星に円盤ガスが暴走的に流入 → ガス惑星へ
巨大氷惑星の形成
円盤散逸後に原始惑星が形成 → ガスを纏えず氷惑星へ
“ハビタブルゾーン” と
“ハビタブルプラネット”
「ハビタブルゾーン」
惑星表面に液体の水が存在できる領域
次に水の性質に注目してみよう．図1.2に 1 気圧に
おけるいろいろな化合物の融点と沸点を示した．これ
は大雑把に液体の状態をとる温度範囲を表すことにな
図1.1：太陽系の元素存在度．太陽組成ガス（Solar）と炭素質コ
ンドライト（CI）に含まれる元素の存在度を，珪素の存在
6
度を10 に規格化して示した．
Why H2O?
ハビタブルプラネットの起源と進化第１回／阿部
る．ここには化合物の分子量も示してある．一般に分
子量が大きい物質ほど融点・沸点とも高くなる傾向が
ある．その中にあって水は分子量が小さいにもかかわ
らず融点と沸点が高いことが分かるであろう．水並み
水素であり，次はヘリウムである．その次に酸素，炭
に融点・沸点が高い物質はどれもかなり複雑な物質で
素，ネオン，窒素と来る．固体の惑星の主成分である
ある．言い換えればそのような化合物は作りにくい．
マグネシウム，シリコン，鉄が次に続くが，酸素の存
195
こう見ていくと水は単純な物質，すなわち存在量が多
在量はこれらよりも 1 桁以上多い．元素合成の過程を
考えても水素と酸素が多い元素であることには違いが
ないであろう．反応性がないヘリウムを除けば，水は
最も多い二つの元素の組み合わせでできている．その
ことから考えても水という物質が非常に普遍性のある
物質であるということがわかる．なお，他の恒星系で
は酸素よりも炭素の方が多い，というようなこともあ
るかもしれない．この場合には，炭素がどのような形
態をとるかによっては，水は作りにくくなってしまう
かもしれない．
次に水の性質に注目してみよう．図1.2に 1 気圧に
おけるいろいろな化合物の融点と沸点を示した．これ
は大雑把に液体の状態をとる温度範囲を表すことにな
図1.1：太陽系の元素存在度．太陽組成ガス（Solar）と炭素質コ
ンドライト（CI）に含まれる元素の存在度を，珪素の存在
6
度を10 に規格化して示した．
る．ここには化合物の分子量も示してある．一般に分
子量が大きい物質ほど融点・沸点とも高くなる傾向が
図1.2： 1 気圧におけるさまざまな化合物の融点および沸点．線分で示された部分が，その化合物が液体の状態をとる温度範囲を
ある．その中にあって水は分子量が小さいにもかかわ
表している．また，その化合物の分子量を白丸で示している．水は分子量が小さい化合物の中で，ひと際融点と沸点が高い．
存在度の大きい単純な分子の中で
らず融点と沸点が高いことが分かるであろう．水並み
水素であり，次はヘリウムである．その次に酸素，炭
に融点・沸点が高い物質はどれもかなり複雑な物質で
素，ネオン，窒素と来る．固体の惑星の主成分である
ある．言い換えればそのような化合物は作りにくい．
マグネシウム，シリコン，鉄が次に続くが，酸素の存
こう見ていくと水は単純な物質，すなわち存在量が多
圧倒的に高い融点・沸点を持つ
H.Z. for Ocean Planets
・Habitable zone の内側境界@present S.S.
暴走温室条件：0.97A [Kopparapu et al. 2013]
・Habitable zone の外側境界@present S.S.
CO2 凝縮条件：1.70AU [Kopparapu et al.. 2013]
・Continuously habitable zone: 0.99AU-1.1AU
from Kasting et al. (1993)
436
!
次々に発見される太陽系外の惑星たち
宇宙は地球であふれてる！
(c) NASA
「第二の地球」を発見！？
Quintana et al., Science (2014)
いかにして “地球” をつくるか
All the Water on Planet Earth
Habitable Trinity
Dohm & Maruyama (2014)
その他の地球特有の事項
U. Mann et al. / Geochimica et Cosmochimica Acta 84 (2012) 593–613
on the parameters a, b and c in Eq. (20) as listed in Table 6.
In Fig. 6b–e the same relationships are shown as in Fig. 6a,
except that log D uncertainty envelopes are included for
each element. In each figure, results for Ru, Rh, Re and
Pt respectively are plotted together with extrapolated partition coeﬃcients for Ir (the most siderophile element) and Pd
(the least siderophile element). Including uncertainties, the
pressure range where the Pd mantle concentration could
have been produced by metal–silicate equilibration is 30–
45 GPa. Total uncertainties are largest for Ir, because its
concentration in the silicate melt was below the detection
limit in several samples and fewer data were available for
regression (Table 6). But even at the lower limit of uncertainty, the Ir mantle concentration could only have been
achieved at pressures of P53 GPa, based on the (unlikely)
assumption of oxidising redox conditions throughout accretion. Similarly, the Pt and Rh partition coeﬃcients could
core
only have the required D mantle
values at pressures of 50–
80 GPa and 50–100 GPa respectively if the oxygen fugacity
remained at a constant value of IW -2. The latter value is
inconsistent with the observed depletions of moderately siderophile elements (especially V and Cr) which require that
core formation initially occurred under highly-reducing
conditions, such as IW -4 (Wade and Wood, 2005; Rubie
et al., 2011). In the case of Ru and Re, at IW -2 even conditions of 75 and 85 GPa respectively could not have decreased their partition coeﬃcients suﬃciently (Figs. 6b, d).
At lower oxygen fugacities, the trends for all elements in
Fig. 6 would shift to even higher pressures such that at
2
60 GPa the partition coeﬃcients of Ru, Rh, Re, Ir and Pt
地球の軌道がほぼ真円
felsic melts, as indicated by multiple geochemical signatures6,8,9.
Oxygen isotopes, for example, demonstrate that the vast majority of
terrestrial Hadean zircons fall outside the zircon mantle equilibrium
field (d18OVSMOW 5 5.3 6 0.3 % (1s))5,20, which is a strong argument
for the input of supracrustal material into the source melts. Moreover,
most zircon crystallization temperatures are consistent with formation
in water-saturated or near-water-saturated melts7,8.
Conclusions about the oxidation state of the Hadean mantle based
on zircons of crustal affinity are admittedly indirect. However, a subset
of the Hadean zircon population does appear to be mantle-derived.
(We note that direct partial melts of the mantle cannot crystallize
zircon, but the fractionation products (residuals) of such melts do
eventually saturate in zircon21.) Five of the Hadean crystals have
d18O values that fall within the mantle equilibrium field, which has
been shown to be constant (60.2%) for the past 4.4 Gyr (ref. 20). If
these d18O values are primary, then these zircons crystallized from
uncontaminated, mantle-derived melts that did not interact with the
hydrosphere. The five zircons in the d18O mantle equilibrium field—
including two crystals with U–Pb ages approaching 4,400 Myr—give a
calculated oxygen fugacity of FMQ 1 1.4 (62) (Fig. 3). As a comparison, zircons crystallized from residual liquids of present-day
mantle-derived mid-ocean ridge basalts22–24 yield calculated fO2 values
of FMQ 1 0.4 (62.6), which agrees with estimates of the oxidation
state of the upper mantle10. This also implies that the redox conditions
in residual, zircon-saturated liquids of present-day mid-ocean ridge
systems did not undergo significant changes in fO2 (relative to the
buffer curve) along the liquid line of descent from basalt. This is also
consistent with observations from lunar zircons. Finally, we note that
Martian basalts (for example, shergottite meteorite Dar al Gani 476)
exhibiting little or no interaction with a crustal component are only
about 1 log unit higher than the estimated fO2 of the upper mantle25.
Furthermore, calculated oxygen fugacities from the same meteorite are
all within a log unit of the average (FMQ 2 2.5), even though phases
cover a crystallization range of .300 uC.
As in all studies in which mantle fO2 is estimated from mantlederived glasses or minerals3,9, it is important to emphasize that the
calculated fO2 may not directly reflect that of the mantle source region.
We note, however, that mantle-derived Hadean zircons would have had
to crystallize from residual melts that underwent ,4 log units of change
in fO2 to be reconcilable with mantle source regions in equilibrium
with the IW buffer. At present and in the Archaean, relative fO2
地球マントル
高!
圧!
実!
験
METHODS SUMM
Starting materials were co
was buffered by placing ru
direct physical contact wit
chamber separated by a
MMO). Quenched glasses
one measurement made
homogeneity. Synthesized
the glass was dissolved wi
order to obtain more accu
For the inset plot of F
represents a normalized v
natural zircons6,19,23–25,30 ar
text: (Ce/Ce*)CHUR 5 CeC
represents normalization t
concentration. Ce anomal
are $94% concordant, a
igneous chemistry5,6,17. Zir
assuming unity Ti-activity
systematic shifts of all data
the standard deviation of
polymict breccias 14304,
fO2 of IW at 700 uC, for di
distribution plot of Fig. 2
from zircon Ce anomalies
basaltic igneous rocks (IW
and average oxygen fugacit
pairs (NNO)27.
マントル中に強親鉄性元素が過剰に存在
18
Archaean and presentday upper mantle
207
硫黄同位体異常 → 2.5Ga まで低O , CO2濃度
3
changes along magma
order of ,1 log unit4
similarly constrained,
its present-day oxidat
pretation is consistent
and chemical isolation
formation28.
If our deductions re
are correct, then the spe
time would have been
An atmosphere of this
ance of sugars and esp
reduced atmosphere is
occurred exceptionally
outgassing of H2 coup
atmosphere out of equ
‘late veneer’ may have s
alternative scenarios no
into the oxidation state
the data presented here
tion state of zircon-bea
calibration will be espe
eons, for which zircon
information.
Fig. 7. Highly siderophile element concentrations, normalised to
CI chondrite (Fischer-Go¨dde et al., 2010), predicted for the
primitive upper mantle (PUM, data from Becker et al., 2006; Rh
from Fischer-Go¨dde et al., 2011) as a result of metal–silicate
6
Crust-derived
parititoning at various P–T conditions
in a magma ocean. Partition
Zircon in equil. with mantle (δ O)
coeﬃcients were calculated at
4 the pressures indicated and at the the
respective peridotite liquidus
temperatures (Herzberg and Zhang,
2
1996; Zerr et al., 1998) and at an oxygen fugacity of IW -3, using
Eqs. (19) and (20) together0 with the regression coeﬃcients from
Full Methods and any ass
the paper at www.nature.c
–2
Table 6. From these partition
coeﬃcients the magma ocean
Received 22 March; accep
concentrations were calculated
employing
a
mass
balance
equation
–4
1. Kasting, J. F. Earth’s ea
IW
2. Canil, D. Vanadium par
and assuming a chondritic bulk composition for the Proto-Earth
magmas. Nature 389, 8
–6
3. Delano, J. W. Redox hist
4,200
3,800
3,900of 4,000
4,100 for
prebiotic molecules. Or
and a constant molar proportion
16.6%
the4,300
core4,400(all data
4. Burgisser, A. & Scaillet
Pb/ Pb* age (Myr)
surface. Nature 445, 19
employed are based on molar
proportions).
5. Cavosie, A. J., Valley, J.
Figure 3 | Oxygen
fugacities of Hadean melts plotted against zircon
ΔFMQ
610
206
crystallization age. Errors for individual points are based on the standard
deviation of the entire data set (n 5 20; 1s 5 6 2.3 log units). Zircons with
mantle signatures are within error of the estimated oxidation state of the presentday and Archaean upper mantle3,10,26. On average, oxygen fugacities are lower for
younger zircons, though this trend is not robust given the present data set.
Magmatic d18O in 440
recycling of crust in the
Cavosie, A. J., Valley, J.
Correlated microanalys
constraints on the igne
Cosmochim. Acta 69, 6
マントルは酸化的@4.35Ga
6.
accretion of small amounts (e.g. for the Earth 0.09–0.5%
1 D
巨大天体衝突破片（GIF）
Giant Impact Fragments
微惑星
106$107
地球型惑星形成理論
巨大天体衝突
7
8
10 $10
原始惑星
地球型惑星（スーパーアース・地球たち）
sta
rt&
input&
input&
!
!
10
#
GIFによる円軌道化
10
e 0.01
#
#
!
!
!
!
!
!
!
!
原始海洋＋ GIF → 水素大気発生
金属鉄 Fe と原始海洋が反応して水素大気を生成
↓
原始海洋が大規模に失われる!
大量の水素大気をまとった原始地球の誕生
岩石
金属鉄
H2大気の形成
H2O+ Fe → FeO + H2
原始海洋
マントルへ
GIF降下後の大気進化
原始海洋と GIF が反応した後の水素の分配：f = H2O/(H2+H2O)!
→ GIF 降下後の原始地球における海洋質量が決まる!
!
H2 のハイドロダイナミックエスケープ：energy-limited escape!
（太陽 EUV のフラックスで散逸率が決まる [Zahnle et al. 1988]）
H2大気
太陽EUV
2H2 + CO2 → H2O + CH2O
CO2ガス
海洋質量と水素大気保持期間
1.0
A
0.9
byr
1
=
tion
n
e
t
r
H 2 re
2 by
r
3 by
0.7
ocean vo
0.6
0.5
0.4
2
lume =1
f = H2O / (H2+H2O)
0.8
0.5
0.3
A
0.2
0.1
0.0
B
0
10
20
30
Initial ocean volume [x present]
40
O2
of FMQ 1 0.4 (62.6), which agrees with estimates of the oxidation
state of the upper mantle10. This also implies that the redox conditions
in residual, zircon-saturated liquids of present-day mid-ocean ridge
systems did not undergo significant changes in fO2 (relative to the
buffer curve) along the liquid line of descent from basalt. This is also
consistent with observations from lunar zircons. Finally, we note that
Martian basalts (for example, shergottite meteorite Dar al Gani 476)
exhibiting little or no interaction with a crustal component are only
about 1 log unit higher than the estimated fO2 of the upper mantle25.
Furthermore, calculated oxygen fugacities from the same meteorite are
all within a log unit of the average (FMQ 2 2.5), even though phases
cover a crystallization range of .300 uC.
As in all studies in which mantle fO2 is estimated from mantlederived glasses or minerals3,9, it is important to emphasize that the
calculated fO2 may not directly reflect that of the mantle source region.
We note, however, that mantle-derived Hadean zircons would have had
to crystallize from residual melts that underwent ,4 log units of change
in fO2 to be reconcilable with mantle source regions in equilibrium
with the IW buffer. At present and in the Archaean, relative fO2
地球は還元的大気を長期間保持
6
Crust-derived
Zircon in equil. with mantle (δ18O)
4
ΔFMQ
2
Archaean and present-
0 day upper mantle
eons, for which
information.
METHODS SU
Starting materials w
was buffered by pla
direct physical cont
chamber separated
MMO). Quenched
one measurement
homogeneity. Synt
the glass was dissol
order to obtain mo
For the inset pl
represents a norma
natural zircons6,19,23
text: (Ce/Ce*)CHUR
represents normaliz
concentration. Ce a
are $94% concord
igneous chemistry5,
assuming unity Tisystematic shifts of
the standard devia
polymict breccias 1
fO2 of IW at 700 uC
distribution plot of
from zircon Ce ano
basaltic igneous roc
and average oxygen
pairs (NNO)27.
Full Methods and a
the paper at www.n
–2
Received 22 March
–4
–6
3,800
1.
2.
IW
3,900
4,000
207Pb/206Pb*
硫黄同位体異常 → 2.5Ga まで低O2, CO2濃度
[Farquhar et al., 2000]
4,200
4,100
4,300
4,400
age (Myr)
Figure 3 | Oxygen fugacities of Hadean melts plotted against zircon
crystallization age. Errors for individual points are based on the standard
deviation of the entire data set (n 5 20; 1s 5 6 2.3 log units). Zircons with
mantle signatures are within error of the estimated oxidation state of the presentday and Archaean upper mantle3,10,26. On average, oxygen fugacities are lower for
younger zircons, though this trend is not robust given the present data set.
3.
4.
5.
Kasting, J. F. Ear
Canil, D. Vanadi
magmas. Nature
Delano, J. W. Red
prebiotic molec
Burgisser, A. & S
surface. Nature
Cavosie, A. J., Va
Magmatic d18O
recycling of crus
Cavosie, A. J., Va
Correlated micr
constraints on t
Cosmochim. Act
マントルは[email protected]
6.
[Trail et al., 2011]
©2011 Macmillan Publishers Limited. All rights res
これまで：酸化的な原始地球マントル + 大気とマントルは平衡状態!
→ 2.5Ga まで大気を低O2, CO2濃度にしておくことは困難
本シナリオ：大量の H2 大気が系を支配（大気とマントルは非平衡）!
→ 長期間 H2 大気持続, 低O2, CO2濃度を維持
いかにして “地球” をつくるか
・大枠は太陽系形成論に則って説明される!
!
・巨大天体衝突過程で大量の破片（GIF）が形成!
!
・GIF による地球の円軌道化 & レイトベニア!
!
・原始海洋 + GIF → 大量の水素大気の発生!
!
・初期海洋質量が適度に減少（→ 1海洋）!
!
・酸化的な脱ガスの下で還元的な大気を長期間保持

Download Report