1: \documentclass[12pt,preprint]{aastex}
2: \begin{document}
3:
4: %macro for invoking today's date (when TeX is run on your file)
5: \def\today{\number\year\space \ifcase\month\or January\or February\or
6: March\or April\or May\or June\or July\or August\or
7: September\or
8: October\or November\or December\fi\space \number\day}
9: %\fraction makes a nice fraction
10: \def\fraction#1/#2{\leavevmode\kern.1em
11: \raise.5ex\hbox{\the\scriptfont0 #1}\kern-.1em
12: /\kern-.15em\lower.25ex\hbox{\the\scriptfont0 #2}}
13: %\simlt and \simgt produce > and < signs with twiddle underneath
14: \def\spose#1{\hbox to 0pt{#1\hss}}
15: \def\simlt{\mathrel{\spose{\lower 3pt\hbox{$\mathchar''218$}}
16: \raise 2.0pt\hbox{$\mathchar''13C$}}}
17: \def\simgt{\mathrel{\spose{\lower 3pt\hbox{$\mathchar''218$}}
18: \raise 2.0pt\hbox{$\mathchar''13E$}}}
19: \def\etal{et al. }
20: %
21: \def\simlt{\mathrel{\hbox{\rlap{\hbox{\lower4pt\hbox{$\sim$}}}\hbox{$<$}}}}
22: \def\simgt{\mathrel{\hbox{\rlap{\hbox{\lower4pt\hbox{$\sim$}}}\hbox{$>$}}}}
23: \def\kms{${\rm km\;s^{-1}}$}
24: \def\mps{${\rm m\;s^{-1}}$}
25:
26: \title{\Large \bf Dispersing the Gaseous Protoplanetary Disc
27: and Halting Type II Migration}
28:
29: \author
30: {M. Lecar \& D. D. Sasselov}
31: \affil{Harvard-Smithsonian Center for Astrophysics, 60 Garden St.,
32: Cambridge, MA 02138}
33:
34: \begin{abstract}
35: More than 30 extra-solar Jupiter-like planets have shorter
36: periods than the planet Mercury. It is generally accepted that they
37: formed further out, past the 'snow line' ($\sim$1~AU), and migrated
38: inwards. In order to be driven by tidal torques from the gaseous disc,
39: the disc exterior to the planet had to contain about a planetary mass.
40: The fact that the planets stopped migrating means that their
41: outer disc was removed. We suggest, following the simulation by
42: Bate \etal (2003), that the outer disc was accreted by the planet. This
43: not only halts migration but removes the outer disc for planets interior
44: to about 2~AU. The disc further out could have been removed by
45: photoevaporation (Matsuyama \etal 2003). Furthermore, as also shown
46: by Bate \etal (op cit) this process also provides an upper limit to
47: planetary masses in agreement with the analysis of observed planetary
48: masses by Zucker \& Mazeh (2002).
49: In this scenario, the endgame is a race. The central star is accreting the
50: inner disc and the planet, while the planet is accreting the outer disc.
51: The planet survives if it accretes its outer disc before being accreted
52: by the star. The winner is determined solely by the ratio of the mass of
53: the outer disc to the local surface density of the disc. Some planets are
54: certainly eaten by the central star.
55: \end{abstract}
56: \keywords{extrasolar planets, Jupiter}
57:
58: \section{Introduction}
59:
60: Planets whose 'Roche Radii' are comparable to the disk scale height form a gap
61: in the disc. This was first suggested by Lin \& Papaloizou (1986) as the mechanism
62: to limit the growth of Jupiter. It was thought that the gap acted like semi-
63: permeable membrane preventing gas from flowing inwards.
64: Because the close-in extra-solar planets had migrated inwards, and then
65: stopped, they must have had an outer disc and then lost it.
66: Previously, we suggested (Lecar \& Sasselov 1999) that gas from the outer disc would stream across the
67: gap, when the gap width was less than the distance gas could travel in an
68: orbital period. We thought that the gas would
69: sneak across the gap when the planet was on the other side of the star, and
70: join the inner disc. We speculated that 3-D disc simulations
71: would confirm this. Bate \etal (2003) recently performed just
72: the simulation we wished for. Gas did, indeed, stream across the gap, almost
73: as if the gap wasn't there, but contrary to our speculation, did not reach the
74: inner disc. Almost all of the gas was accreted onto the planet.
75: However, this also halts migration.
76: The accretion of the outer disc by the planet solves two problems:
77: stopping the migration and removing the outer disc. A recent study
78: of removal of the disc by photoevaporation showed that that process
79: is only effective exterior to 2.4~AU (Matsuyama \etal 2003).
80: In the minimum
81: mass solar nebula (Hayashi 1981) there are about four Jupiter masses of
82: gas interior to 2.4~AU. In any case, we have to account for the more than
83: 50 extra-solar planets interior to 1 AU. Since the 'snow line' starts at
84: 1 AU, there is a possibility that planets outside of 1 AU formed in place,
85: or did not migrate much.
86:
87: An argument in favor of the planet accreting its outer disc is that it provides
88: a natural upper limit to the masses of planets of 10 Jupiter masses.
89: Zucker \& Mazeh (2002) found that to be the upper limit to planetary masses
90: and the lower limit to the masses of brown dwarfs. The formation of
91: gas giants by gravitational instability (Boss 2000) provides no natural
92: upper limit to the masses.
93:
94: We now discuss in more detail the migration process, following Lin \&
95: Papaloizou (1986). We wish to illustrate when migration switches
96: from Type~I (no gap) to Type~II (gap) (Ward 1997).
97: A planet with semi-major axis $a$ migrates at a rate
98: \begin{equation}
99: \dot{a}= -a{\Omega}{\mu}_{\rm pl}{\mu}_{\rm disk}I
100: \end{equation}
101: where
102: \begin{equation}
103: {\Omega}^2a^3= GM_{\odot};~~~{\mu}_{\rm pl}= \frac{M_{\rm pl}}{M_{\odot}};
104: ~~~{\mu}_{\rm disk}= \frac{2{\pi}{\Sigma}a^2}{M_{\odot}}.
105: \end{equation}
106: For a minimum mass solar nebula (Hayashi 1981), the surface mass density of the disk is
107: ${\Sigma}=1700~g~cm^{-2}$ at 1 AU, and ${\Sigma}\propto x^{-3/2}$, where $x=a/AU$.
108: The dimensionless integral, $I$, is given by
109: \begin{equation}
110: \int_{{\Delta}_{out}} dx{\frac{(1+x)^{5/2}}{x^4}} -
111: \int_{{\Delta}_{in}} dx{\frac{(1-x)^{5/2}}{x^4}} \cong
112: \frac{1}{3}({\frac{1}{{\Delta}_{out}^3}-\frac{1}{{\Delta}_{in}^3}}) +
113: \frac{5}{2}({\frac{1}{{\Delta}_{out}^2}+\frac{1}{{\Delta}_{in}^2}})
114: \end{equation}
115: where for ${\Delta}\ll 1$, $I$ is insensitive to the upper limits.
116: If there is no gap, ${\Delta}_{in}={\Delta}_{out}$ and
117: $I=\frac{5}{2}\frac{1}{{\Delta}^2}$, where ${\Delta}=\frac{h}{a}=\frac{c_s}{V}$,
118: $V$ is the circular orbital velocity, $c_s$ is the sound speed, and
119: $h$ is the scale height. Typically, ${c_s}/{V} \cong 0.06~x^{1/4}$.
120:
121: If there is a gap, the planet moves within the gap to equalize
122: the torques from the inner and outer discs. This is accomplished
123: by
124: \begin{equation}
125: {\Delta}_{out}= {\Delta}({1+\frac{5}{4}{\Delta}}), ~~~~~~~~
126: {\Delta}_{in}= {\Delta}({1-\frac{5}{4}{\Delta}})
127: \end{equation}
128: Henceforth, the planet responds to the inward migration of the
129: outer disc, which is driven by viscosity. Once a gap is opened,
130: the further evolution is controlled by viscous accretion and is referred to
131: as type~II migration (Ward 1997).
132:
133: The mass accretion rate is
134: \begin{equation}
135: \dot{M}_{acc}= 2{\pi}{\Sigma}a\dot{a}\equiv
136: 2{\pi}{\Sigma}a^2{\frac{\dot{a}}{a}} \equiv
137: 2{\pi}{\Sigma}a^2\frac{1}{t_{acc}},
138: \end{equation}
139: which by continuity is independent of $x$. If ${\Sigma}\propto x^{-3/2}$, then
140: $t_{acc}\propto x^{1/2}$. The time to accrete a Jupiter mass is
141: \begin{equation}
142: t_{M_{\rm J}}= \frac{M_{\rm J}}{\dot{M}_{acc}}=
143: \frac{M_{\rm J}}{2{\pi}{\Sigma}a^2}{t_{acc}}.
144: \end{equation}
145: If the outer disc has, say, a Jupiter mass, in order for a planet to accrete the
146: outer disc before being swept into the star, we require that
147: \begin{equation}
148: t_{M_{\rm J}}< t_{acc}
149: \end{equation}
150: or
151: \begin{equation}
152: \frac{M_{\rm J}}{2{\pi}{\Sigma}a^2}=
153: \frac{M_{\rm J}}{2{\pi}{\Sigma}_0a_0^2x^{1/2}}< 1
154: \end{equation}
155: The migration halts when
156: \begin{equation}
157: x^{1/2}> \frac{M_{\rm J}}{2{\pi}{\Sigma}_0a_0^2}.
158: \end{equation}
159:
160: For the minimum mass solar nebula, migration would halt
161: at $x=0.63$ or the orbital period $P=183$~days. For denser discs our
162: estimates (with ${\Sigma}_{min}=1700~g~cm^{-2}$ at 1 AU) are given in Table 1.
163: For surface densities larger than $4{\Sigma}_{min}$,
164: the planet is accreted by the star.
165:
166: More quantitatively, if the planet is at ${a}_0$, the ratio
167: $\frac{M_{\rm J}}{2{\pi}{\Sigma}_0a_0^2}$ can be written:
168: \begin{equation}
169: \int_{1}^{x} dx(x)^{1-n} = \frac{(x)^{2-n}}{2-n},
170: \end{equation}
171: if ${\Sigma}(x)= {\Sigma}(1)x^{-n}$ and $x=\frac{a}{a_0}$.
172: For the planet to accrete its outer disc before the star accretes the
173: planet, that quantity has to be less than 1.0, or $x\leq x_m$. This
174: is illustrated in Table 2.
175: For comparisom, if $a_0= 5.203~AU$ (Jupiter), Saturn is at $x=1.83$. If
176: $a_0= 1~AU$, photoevaporation is effective at $x> 2.4$.
177:
178: So far, we have avoided a discussion of the physical source of the
179: viscosity and the value of $\alpha$ (Shakura \& Sunyaev 1973),
180: because our result is independent of
181: this physics. But, to make contact with the literature, we note that
182: with our prescription (also used by Bate \etal, op cit), ${\alpha}\propto {\alpha}_0x^{1/2}$,
183: we require ${\alpha}_0\leq 3{\times}10^{-4}$ in order that the mass accretion be less
184: than $10^{-8}M_{\odot}{yr^{-1}}$. This is to limit the accretion luminosity
185: to yield
186: the so called 'passive disc' (Chiang \& Goldreich 1997; Sasselov \& Lecar 2000).
187: With this value of $\alpha$, we have $t_{acc}\approx 10^5x^{1/2}$~years,
188: which is short,
189: suggesting that ${\alpha}_0$ is smaller.
190:
191: We conclude with some speculations about our Jupiter. In the minimum
192: mass solar nebula, with ${\Sigma}\propto x^{-3/2}$, there is about 2.7 Jupiter
193: masses of gas between Jupiter and Saturn. Clearly, Jupiter (and Saturn)
194: did not accrete it all. However the investigation of photoevaporation
195: of the outer disc was motivated by the fact that Saturn has only $\frac{1}{3}$rd
196: the mass of Jupiter, and in any case, $x^{-3/2}$ yields a divergent mass
197: ($\propto x^{1/2}$).
198: The surface density profile must steepen. If we keep the minimum mass
199: surface density at Jupiter (143 g/cm$^2$), but allow the surface density
200: to decrease outwards at a steeper rate, say $x^{-7/2}$, then there is less
201: than a Jupiter mass between Jupiter and Saturn. If we allow Jupiter to start
202: accreting when its mass was $\geq 0.1~M_{\rm J}$, as suggested by Bate \etal
203: (2003), then the outer disc is in Jupiter.
204:
205:
206:
207: \bibliography{journals}
208:
209: \begin{table}[!h]
210: \begin{center}
211: \begin{tabular}{|c|c|c|}
212: \hline
213: $\Sigma$ & x & P \\
214: & (a/AU) & (days) \\
215: \hline\hline
216: ${\Sigma}_{min}$ & 0.630 & 183 \\
217: 2${\Sigma}_{min}$ & 0.157 & 23 \\
218: 3${\Sigma}_{min}$ & 0.070 & 7 \\
219: 4${\Sigma}_{min}$ & 0.039 & 3 \\
220: \hline
221: \end{tabular}
222: \end{center}
223: \caption[]{Orbital distances $x$ and periods $P$ at which a planet
224: accreting its outer disc would stop migrating, as a function of disc
225: surface density $\Sigma$.}
226: \end{table}
227:
228: \begin{table}[!h]
229: \begin{center}
230: \begin{tabular}{|c|c|c|}
231: \hline
232: $n$ &$\frac{(x)^{2-n}}{2-n}$ & $x_m$ \\
233: \hline\hline
234: 3/2 & $2(x^{1/2}-1)$ & 2.25 \\
235: 2 & $ln~x$ & 2.72 \\
236: 5/2 & $2(1-x^{1/2})$ & 4.0 \\
237: 3 & $(1-x^{-1})$ & $\infty$ \\
238: \hline
239: \end{tabular}
240: \end{center}
241: \caption[]{The minimum orbital distance, $x_m$, where migration halts,
242: for different types of discs.}
243: \end{table}
244:
245:
246: \end{document}
247: