1: %\documentclass[12pt,preprint]{aastex}
2: \documentclass[12pt]{emulateapj}
3: %\usepackage{rotating}
4: %\usepackage{epstopdf}
5: %\usepackage{graphicx}
6: %\usepackage[usenames]{color}
7: %
8: %\addtolength{\topmargin}{-0.5in}
9:
10: \newcommand{\MJ}{M_{\rm J}}
11: \newcommand{\RJ}{R_{\rm J}}
12: \newcommand{\Qp}{Q_{\rm p}}
13: \newcommand{\Mp}{M_{\rm p}}
14: \newcommand{\Rp}{R_{\rm p}}
15: \newcommand{\Fp}{F_{\rm p}}
16: \newcommand{\apla}{a_{\rm p}}
17: \newcommand{\ep}{e_{\rm p}}
18: \newcommand{\np}{n_{\rm p}}
19: \newcommand{\Mc}{M_{\rm c}}
20: \newcommand{\ac}{a_{\rm c}}
21: \newcommand{\ec}{e_{\rm c}}
22: \newcommand{\nc}{n_{\rm c}}
23:
24: \begin{document}
25: \title{On The Origins Of Eccentric Close-in Planets}
26: %{\bf [SHOULD THIS BE ``CLOSE-IN PLANETS?'' GJ436 IS NEPTUNE-SIZE]}}
27: %
28: \author{Soko Matsumura, Genya Takeda,}
29: \author{Frederic A.~Rasio}\affil{Department of Physics \& Astronomy,
30: Northwestern University, Evanston, IL, 60208}
31: %\email{soko@northwestern.edu, genya@u.northwestern.edu,
32: %rasio@northwestern.edu}
33: %
34: %
35: %--- Abstract ---
36: \begin{abstract}
37: Strong tidal interaction with the central star can circularize the
38: orbits of close-in planets. With the standard tidal quality factor
39: $Q$ of our solar system, estimated circularization times for
40: close-in extrasolar planets are typically shorter than the ages of
41: the host stars. While most extrasolar planets with orbital radii
42: $a\la0.1\,$AU indeed have circular orbits, some close-in planets
43: with substantial orbital eccentricities have recently been
44: discovered. This new class of eccentric close-in planets implies
45: that either their tidal $Q$ factor is considerably higher, or
46: circularization is prevented by an external perturbation. Here we
47: constrain the tidal $Q$ factor for transiting extrasolar planets by
48: comparing their circularization times with accurately determined
49: stellar ages. Using estimated secular perturbation timescales, we
50: also provide constraints on the properties of hypothetical second
51: planets exterior to the known ones.
52: \end{abstract}
53: %
54: \keywords{planetary systems}
55: %--- Introduction, Section 1
56: %----------------------------------------------------------------------
57: \section{Introduction}
58: The median eccentricity of the current sample of $\sim300$ planets
59: is 0.19, while it is 0.013 for close-in planets with semi-major axis
60: $a<0.1\,$AU. The circular orbits of close-in planets most likely
61: result from orbital circularization due to tides
62: \citep[e.g.,][]{Rasio96,Marcy97}. This requires the tidal
63: circularization time $\tau_{\rm circ}$ to be short compared to the
64: age of the system $\tau_{\rm age}$. Since $\tau_{\rm circ}$ is a
65: very steep function of $a$ (see Eq.~\ref{tcirc} or \ref{tcirc0}),
66: while $\tau_{\rm age} \sim 1-10\,$Gyr for most systems, a sharp
67: decline in eccentricity is expected below some critical value of
68: $a$. However, the observed transition seems to occur around
69: $0.03-0.04\,$AU, whereas the calculated \(\tau_{\rm circ}\) becomes
70: comparable to \(\tau_{\rm age}\) at $\sim 0.1\,$AU as we see below
71: (Figure~\ref{fig1}). Since almost one quarter (presently 16/68) of
72: planets within $0.1\,$AU have $e>0.1$, their high eccentricities
73: demand explanation.
74:
75: First, we calculate the circularization times for transiting
76: planets, and compare them with the estimated ages of the systems.
77: The circularization time $\tau_{\rm circ}=-e/\dot{e}$, where
78: $\dot{e} $ is the sum of the eccentricity change due to the tides
79: raised on the star by the planet and those raised on the
80: planet by the star, is
81: %written as
82: \citep{Goldreich66,Hut81,Eggleton98,Mardling02}:
83: %
84: \begin{equation}
85: \tau_{\rm circ}=
86: \frac{2}{81}\frac{\Qp^{\prime}}{n}\frac{\Mp}{M_*}
87: %\left(\frac{\Rp}{a}\right)^{-5}
88: \left(\frac{a}{\Rp}\right)^{5}
89: \left[\frac{\Qp^{\prime}}{Q_*^{\prime}}\left(\frac{\Mp}{M_*}\right)^2
90: %\left(\frac{\Rp}{R_*}\right)^{-5} F_* + \Fp \right]^{-1}
91: \left(\frac{R_*}{\Rp}\right)^{5} F_* + \Fp \right]^{-1}
92: \label{tcirc} \,
93: \end{equation}
94: %
95: The subscripts p and $*$ represent the planet and star,
96: respectively. The modified tidal quality factor for a planet is
97: defined as $\Qp^{\prime}\equiv3\Qp/2k_{\rm p}$, where $k_{\rm p}$ is
98: the Love number, and $\Qp$ is the specific dissipation function, which
99: depends on the planetary structure as well as the frequency and
100: amplitude of tides. We also define
101: %
102: \begin{eqnarray}
103: F_*&=&\left[f_1(e^2)-\frac{11}{18}f_2(e^2)\frac{\Omega_{*,{\rm
104: rot}}} {n}\right]
105: \\
106: \Fp&=&\left[f_1(e^2)-\frac{11}{18}f_2(e^2)\frac{\Omega_{{\rm
107: p,rot}}} {n}\right] \ ,
108: \end{eqnarray}
109: where $n=\sqrt{G(M_*+\Mp)/a^3}$ is the mean motion, $\Omega_{\rm
110: rot}$ is the rotational frequency, and
111: \begin{eqnarray}
112: f_1(e^2)&=&
113: \left(1+\frac{15}{4}e^2+\frac{15}{8}e^4+\frac{5}{64}e^6\right)/(1-e^2)
114: ^{13/2},
115: \\
116: f_2(e^2)&=&\left(1+\frac{3}{2}e^2+\frac{1}{8}e^4\right)/(1-e^2)^5 \
117: .
118: \end{eqnarray}
119: %
120: Generally $F_*$ and $F_p$ are comparable, and thus the stellar
121: damping is negligible unless the planet-to-star mass (radius) ratio
122: is large (small) or $Q_*^{\prime}\ll Q_p^{\prime}$. We define the
123: circularization time due to damping in the planet as
124: %
125: \begin{equation}
126: \tau_{{\rm
127: circ},0}=\frac{2}{81}\frac{\Qp^{\prime}}{n}\frac{\Mp}{M_*}
128: \left(\frac{\Rp}{a}\right)^{-5} \Fp^{-1} \ \label{tcirc0}.
129: \end{equation}
130: %
131: Note that $\tau_{{\rm circ},0}$ can be shorter or longer than
132: $\tau_{{\rm circ}}$, depending on the sign of $F_*$, which changes
133: at $(\Omega_{*,{\rm rot}}/n)_{\rm crit}= 18/11(f_1/f_2)$. In the
134: limit $e\rightarrow 0$, this equation leads to the standard
135: expression for the circularization time \citep[Eq.~4.198
136: of][]{Murray99}.
137:
138: Figure~\ref{fig1} compares the circularization times calculated from
139: Eq.~\ref{tcirc} and \ref{tcirc0} with the estimated stellar ages for
140: the systems in Table~\ref{tb1}
141: %%%% Footnote %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
142: \footnote{Note that, in Eq.~\ref{tcirc}, it is implicitly assumed
143: that the star and the planet both have zero obliquity. Currently
144: available measurements of the Rossiter--MacLaughlin effect show
145: that the planetary orbits in general are closely aligned with the
146: stellar equator \citep{Queloz00,Winn05}. Current exceptions may be
147: the HD\,17156 and XO-3 systems \citep{Narita07ap,Hebrard08ap}, which
148: we exclude from our analysis.}.
149: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
150: %
151: Here we assume $\Qp^{\prime}=10^5$, and $Q_*^{\prime}=10^6$, which
152: are the standard values motivated by measurements in our Solar
153: System \citep[e.g.][]{Yoder81,Zhang08}, and for main-sequence stars
154: \citep[e.g.][]{Carone07}. For $\Omega_{\rm p,rot}$, we assume that
155: planets with circular orbits are perfectly synchronized, i.e.,
156: $\Omega_{\rm p,rot}/n=1$, since the spin-orbit synchronization times
157: are $\sim10^{-3}\,\tau_{\rm circ}$ \citep {Rasio96}. On the other
158: hand, planets with eccentric orbits should spin down until they
159: reach quasi-synchronization \citep{DobbsDixon04}; in practice we
160: adopt a planetary spin frequency such that the rate of change of
161: spin frequency is zero \citep[Eq.~54 of][]{Mardling02}. For the
162: stellar spin, we assume typical periods derived from the observed
163: $v_{\rm rot}\sin i$, $P_{*,{\rm rot}}\sim 3\,$--\,70\,d
164: \citep{Barnes01}.
165:
166:
167: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
168: % FIGURE 1
169: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
170: \begin{figure}
171: \plotone{f1.eps} \caption[fig1]{ Circularization times calculated
172: from Eq.~\ref{tcirc0} (stars), Eq.~\ref{tcirc} with $\Omega_{*,{\rm
173: rot}}/n=3\,$d (orange circles) and 70\,d (blue circles),
174: compared with the estimated stellar ages (squares) for systems with
175: transiting planets. Here we assume $\Qp^{\prime}=10^5$ and
176: $Q_*^{\prime}=10^6$. For all transiting planets, the estimated
177: circularization time is shorter than the age of the host star.
178: \label{fig1}}
179: \end{figure}
180: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
181:
182:
183:
184: Figure~\ref{fig1} compares the circularization times calculated from
185: Eq.~\ref{tcirc} and \ref{tcirc0} with the estimated stellar ages for
186: the systems in Table~\ref{tb1}. For most systems the tidal damping
187: in the star is negligible. Two systems with non-negligible stellar
188: damping are WASP-14 and HAT-P-2. Both have a large planetary mass
189: (see Table~\ref{tb1}), and thus the first term in Eq.~\ref{tcirc} is
190: significant. Clearly, the estimated circularization times are always
191: shorter than stellar ages, which implies that all these planets
192: should have been circularized by now if their tidal $Q$ values were
193: similar to those of their solar-system analogues. However, our
194: sample contains at least 6 systems with a non-zero orbital
195: eccentricity. Possible explanations are that (1) the structure of
196: these planets is different and their actual $Q$ is greater than what
197: we assumed; or (2) they currently experience external perturbations
198: which maintain their orbital eccentricities against tidal
199: dissipation.
200: %
201: %--- Section 2
202: %------------------------------------------------------------------------
203: %------------
204: \section{Constraints on the Tidal $Q$ value of close-in planets}
205: %
206: Now we use Eq.~\ref{tcirc} to place constraints on the tidal
207: $Q$ values of the planets. An {\it upper\/} limit on $Q$ is provided
208: for planets with {\it circular\/} orbits since the circularization
209: must have occurred within the lifetime of the systems ($\tau_{\rm
210: circ} < \tau_{\rm age}$). Our assumption here is that these close-in
211: planets formed through tidal circularization of initially eccentric
212: orbits. \cite{Nagasawa08ap} showed that about one third of multiple
213: planetary systems could form close-in planets through tidal
214: circularization following a large eccentricity gain through
215: planet--planet scattering or Kozai-type perturbations. Direct
216: observational evidence for initially large orbital
217: eccentricities comes from the absence of
218: planetary orbits within {\it twice\/} the Roche limit around the
219: star \citep {Faber05,Ford06}.
220:
221: On the other hand, close-in, {\em eccentric\/} planets impose a {\em
222: lower\/} limit on $Q$ values, since $\tau_{\rm circ}\geq \tau_{\rm
223: age}$ is expected for these systems, provided that they are not
224: currently subject to any eccentricity excitation mechanism.
225:
226: For $P_{*,{\rm rot}}\sim 3-70\,$d, {\it all\/} planets in
227: Table~\ref{tb1} take $\Omega_{*,{\rm rot}}/n<(\Omega_{*,{\rm
228: rot}}/n)_{\rm crit}$, and hence $F_*>0$ and $\tau_{{\rm
229: circ},0}>\tau_{\rm circ}$. For planets with zero
230: (non-zero) eccentricity, we require $\tau_{\rm circ}<\tau_{{\rm
231: circ},0}<\tau_{\rm age}$ ($\tau_{\rm age}<\tau_{\rm circ}<\tau_{{\rm
232: circ},0}$). In other words, we assume that zero (non-zero)
233: eccentricity planets have been (have not been) circularized within
234: the lifetime of the system, independent of the rotation period of
235: the star. This gives the upper and lower limits for circular and
236: eccentric planets, respectively, as
237: \begin{eqnarray}
238: \Qp^{\prime}&<&
239: \frac{81}{2}n\left(\frac{M_*}{\Mp}\right)\left(\frac{\Rp}{a}\right)^5
240: \Fp \tau_{\rm age} \equiv Q_{\rm p, crit}^{\prime},
241: \end{eqnarray}
242: \begin{eqnarray}
243: \Qp^{\prime}&>& Q_{\rm p, crit}^{\prime}
244: \left(1-\frac{81}{2}\frac{n}{Q_*^{\prime}}\left(\frac{\Mp}{M_*}\right)
245: \left(\frac{R_*}{a}\right)^5 F_* \tau_{\rm age} \right)^{-1}.
246: %\frac{\Fp \tau_{\rm age}}
247: %{\frac{2}{81}\frac{\Qp^{\prime}}{n}\frac{\Mp}{M_*}\left(\frac{a}{\Rp}
248: %\right)^{5}
249: %- \tau_{\rm age}
250: %\frac{\Qp^{\prime}}{Q_*^{\prime}}\left(\frac{\Mp}{M_*}\right)^2\left
251: %(\frac{R_*}{\Rp}\right)^{5}
252: %F_*/Q_*^{\prime}} \ .
253: \end{eqnarray}
254: %
255: The latter also gives the lower limit for the {\it
256: stellar\/} tidal $Q$ factor, since the denominator must be positive:
257: %
258: \begin{equation}
259: Q_*^{\prime}>\frac{81}{2}n\left(\frac{\Mp}{M_*}\right)\left(\frac{R_*}
260: {a}\right)^5 F_* \tau_{\rm age} \equiv Q_{*,{\rm min}}^{\prime}
261: \label{stellarQmin} \ .
262: \end{equation}
263: %
264: This corresponds to a minimum stellar $Q$ value of $Q_*^{\prime}\sim
265: 3\times 10^4 - 4\times 10^7$, with a median value of
266: $0.4-1\times 10^6$ for $P_{*,{\rm rot}}=3-70\,$d,
267: which agree well with observations
268: \citep[e.g.][]{Carone07}.
269: %\citep[e.g.][]{Terquem98,Carone07}.
270:
271: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
272: % FIGURE 2
273: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
274: \begin{figure}
275: \plotone{f2.eps} \caption[fig2]{Estimated tidal $Q$ factors for the
276: case of slowly rotating stars ($\Omega_*/n<(\Omega_*/n)_{\rm
277: crit}$). Upper/lower limits calculated from Eq.~\ref{tcirc0} are
278: shown in black down/up triangles for planets with zero/non-zero
279: eccentricities. Open triangles are the corresponding estimates from
280: Eq.~\ref{tcirc}, which approach black ones as we take 2, 5, and 10
281: times the minimum stellar tidal $Q$s (blue, orange, and green
282: triangles, respectively) obtained from Eq.~\ref{stellarQmin}.
283: \label{fig2}}
284: \end{figure}
285: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
286:
287:
288: Figure~\ref{fig2} shows the upper and lower limits for the planets'
289: $Q$ values as a function of semi-major axis for
290: circular and eccentric orbits, respectively.
291: Since the lower limits on $Q$ values for eccentric planets depend on
292: $Q_*^{\prime}$, we take three different cases of $Q_*^{\prime}=2$,
293: $5$, and $10\, Q_{*,{\rm min}}^{\prime}$ as examples. Note that,
294: with this definition of $Q_*^{\prime}$, $\Qp^{\prime}$ becomes
295: independent of the stellar spin rate. Since circularization times
296: are shorter for planets with smaller orbital radii, we tend to
297: overestimate the maximum $Q$ values at the shortest-period end. All
298: transiting planets appear within the range
299: $10^5\lesssim \Qp^{\prime} \lesssim
300: 10^9$. The figure also shows that the high eccentricities of some
301: planets (marked with upper triangles) can be explained by assuming
302: relatively large ($\Qp^{\prime}\gtrsim 10^6$) but reasonable
303: ($\Qp^{\prime}\lesssim 10^9$) tidal $Q$ values.
304:
305: Although these estimated $Q$ values are larger than those of
306: Jupiter or Neptune, they cannot be excluded. Recent theoretical
307: studies of the excitation and dissipation of dynamical tides within
308: rotating giant planets have shown that tidal $Q$ values fluctuate
309: strongly depending on the tidal forcing frequency, and the effective
310: $Q$'s could go up to $\sim10^9$ depending on the spin rate and internal
311: structure of the planet \citep[e.g.,
312: presence/absence of a core, radiative envelope, or a density jump,
313: see][]{Ogilvie04,Wu05b}. According to these recent models, it
314: appears possible that some planets maintain large eccentricities
315: simply because of their larger $Q$ values.
316: %
317: %--- Section 3
318: %------------------------------------------------------------------------
319: %------------
320: \section{Dynamical Perturbations}
321:
322: Candidate perturbation mechanisms that could excite and maintain
323: planetary eccentricities include (1) tidal interaction with the
324: central star \citep{DobbsDixon04}, (2) quadrupole or higher-order
325: secular perturbation from an additional body, or (3) resonant
326: interaction with another planet. Here we discuss the effects of
327: these competing mechanisms against tidal eccentricity damping.
328:
329: Tidal dissipation inside the central star can increase the planet
330: eccentricity only when $\Omega_{*,{\rm rot}}>n$, or equivalently
331: $de/dt>0$ in Eq.~\ref{tcirc}. For a synchronized planet
332: ($\Omega_{p,{\rm rot}}\sim n$) with a small eccentricity ($e^2\ll
333: 1$), we obtain $\Omega_{*,{\rm
334: rot}}/n>18/11(1+7/18(Q_*^{\prime}/\Qp^{\prime})(M_*/\Mp)^2(\Rp/R_*)^5)$.
335: For a Jupiter-like planet around a main-sequence star, this yields
336: $\Omega_{*,{\rm rot}}>9.6n$ for $Q_*^{\prime}\sim \Qp^{\prime}$.
337: Since the rotation period for planet-hosting stars typically lies in
338: the range 3\,--\,70\,d, eccentricity excitation may occur for
339: planets only if their orbital periods are within 29\,--\,673\,d or
340: longer. For a $10 \MJ$ planet with radius $1 \RJ$, we have
341: $\Omega_{*,{\rm rot}}>1.7n$, which corresponds to an orbital period
342: greater than 5.1\,d. Therefore, this is unlikely to be responsible
343: for the eccentricity of observed planets within $a\sim
344: 0.06$--0.18\,AU.
345:
346: Another possibility is an undetected additional planet exciting the
347: eccentricity of the detected planet. If there is a large mutual
348: inclination angle ($i \ga 40^\circ$) between the two planets,
349: Kozai-type perturbations can become important \citep{Kozai62}. Such
350: highly non-coplanar orbits could result from
351: planet--planet scattering after dissipation of the gaseous disk.
352: \citet{Chatterjee07ap,Nagasawa08ap} have performed extensive
353: numerical scattering experiments and showed that the final
354: inclination of planets could be as high as $70^\circ$, with a median
355: of 10\,--\,$20^\circ$. If $i \la 40^\circ$, octupole perturbations
356: may still moderately excite the eccentricity of the close-in planet.
357: The secular interaction timescale of a pair of planets with small
358: mutual inclination can be derived from the classical
359: Laplace-Lagrange theory \citep{Brouwer61,Murray99}.
360:
361:
362: For this secular perturbation from an additional planet to be
363: causing the large eccentricity of the close-in planet, it must occur
364: fast enough compared to other perturbations causing orbital
365: precession. In particular, GR precession and tides are important
366: effects that would compete against the perturbation from the
367: additional body \footnote{Although stellar and planetary rotational
368: distortions cause additional precession \citep{Sterne39}, GR
369: precession dominates unless the stellar rotation is on the high
370: end.}. For detailed discussions see \cite{Holman97,Kiseleva98}.
371:
372: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
373: % FIGURE 3
374: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
375: \begin{figure}
376: \plotone{f3.eps}\caption[fig3]{Various secular eccentricity
377: excitation timescales for the planet GJ\,436\,b caused by a
378: hypothetical planetary (or stellar) companion GJ\,436\,c with mass
379: $\Mc$ and semi-major axis $\ac$. The solid black lines show the
380: predicted radial velocity amplitudes caused by the undetected
381: companion. The dot-dashed line shows the threshold right of which
382: the secular interaction between planets with $i\lesssim40$ deg is
383: suppressed by GR precession, while the three red dotted lines are
384: the similar thresholds for the Kozai mechanism with the assumed
385: orbital eccentricity of the companion $\ec=$ 0.01, 0.5, and 0.9 from
386: left to right. We use $\tau_{\rm pp}$, $\tau_{\rm Kozai}$, and
387: $\tau_{\rm GR}$ as in \cite{Takeda08ap,Fabrycky07}. Thresholds for
388: $\tau_{\rm circ}$ and $\tau_{\rm age}$ are also shown for
389: comparison. \label{fig3}}
390: \end{figure}
391: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
392:
393: Figure~\ref{fig3} illustrates the constraints on the mass and
394: orbital radius of the hypothetical outer planet in the GJ\,436
395: system. These are set by comparing the GR precession and secular
396: timescales. Similar results are seen for other systems with
397: eccentric close-in planets. Generally GR precession occurs faster
398: than any other perturbation mechanism. In order to induce Kozai
399: cycles in the inner planet (left of the dotted lines) while not
400: causing radial-velocity amplitudes above the detection limit of
401: $\sim5\,{\rm m} \,{\rm s}^{-1}$ (below black lines), the mass upper
402: limit of the hypothetical planet is $\sim 1 M_{\rm Neptune}$. For
403: near-coplanar systems even tighter constraints are placed on the
404: properties of the secondary planet (left of the dot-dashed line).
405:
406: However, one caveat is that our diagram only rules out the
407: possibility of hypothetical bodies being {\it currently\/}
408: responsible for the high eccentricity of GJ\,436\,b. Also, while
409: Kozai-type perturbations are almost always suppressed by GR
410: precession, eccentricity excitation through secular {\it octupole\/}
411: perturbations may be occasionally {\em enhanced\/} by GR effects
412: \citep{Ford00,Adams06}. For the case of the GJ\,436 system, we have
413: numerically tested the effect of a hypothetical secondary planet~c
414: on the eccentricity evolution of the inner planet~b. We have found
415: that, within the detectable radial-velocity limit $\la 5\,m/s$,
416: planet~c cannot excite the eccentricity of planet~b from 0.01 to the
417: observed 0.15, even if its eccentricity is as high as 0.5. This
418: result holds for other systems since they have even heavier planets.
419: Therefore, we can safely exclude the possibility that these planets
420: obtain their current high eccentricities through secular
421: perturbation from an undetected outer planet with $\lesssim M_{\rm
422: Neptune}$ if their orbits are initially near-circular.
423:
424: Yet another possibility is a resonant perturbation from an
425: undetected planet. Recently \citet{Ribas08} suggested that the
426: eccentricity of GJ\,436\,b might be caused by a mean-motion
427: resonance (MMR) with an unseen super-Earth, but there is little
428: observational support for this \citep[e.g.][]{Bean08ap}. Also, the
429: combined effects of GR precession and MMR are not fully understood
430: yet. In any case, it is unlikely that such resonances are
431: responsible for all the close-in eccentric planets, considering the
432: small fraction of extrasolar multiple planets in MMR.
433: %
434: %
435: %--- Section 4
436: %------------------------------------------------------------------------
437: %------------
438: \section{Summary}
439:
440: In this letter, we have investigated the origins of close-in planets
441: on an eccentric orbit. We place constraints on the tidal Q factor
442: of transiting planets by comparing the stellar age with the tidal
443: circularization time, and find that $10^5\lesssim \Qp^{\prime}
444: \lesssim 10^9$, which agrees well with current theoretical
445: estimates, can explain these eccentric planets. We also show that it
446: is difficult to explain the high eccentricities of these planets by
447: invoking a current interaction with an unseen second planet. Our
448: results suggest that at least some of the close-in eccentric planets
449: may be simply in the process of getting circularized.
450:
451: This work was supported by NSF Grant AST-0507727.
452: %
453: %\bibliographystyle{apj}
454: %\bibliography{REF}
455: %
456:
457: \begin{thebibliography}{33}
458: \expandafter\ifx\csname natexlab\endcsname\relax\def\natexlab#1{#1}\fi
459:
460: \bibitem[{{Adams} \& {Laughlin}(2006)}]{Adams06}
461: {Adams}, F.~C. \& {Laughlin}, G. 2006, ApJ, 649, 992
462:
463: \bibitem[{{Barnes}(2001)}]{Barnes01}
464: {Barnes}, S.~A. 2001, ApJ, 561, 1095
465:
466: \bibitem[{{Bean} \& {Seifahrt}(2008)}]{Bean08ap}
467: {Bean}, J.~L. \& {Seifahrt}, A. 2008, ArXiv e-prints, 806
468:
469: \bibitem[{{Brouwer} \& {Clemence}(1961)}]{Brouwer61}
470: {Brouwer}, D. \& {Clemence}, G.~M. 1961, {Methods of celestial mechanics} (New
471: York: Academic Press, 1961)
472:
473: \bibitem[{{Carone} \& {P{\"a}tzold}(2007)}]{Carone07}
474: {Carone}, L. \& {P{\"a}tzold}, M. 2007, P\&SS, 55, 643
475:
476: \bibitem[{{Chatterjee} {et~al.}(2007){Chatterjee}, {Ford}, {Matsumura}, \&
477: {Rasio}}]{Chatterjee07ap}
478: {Chatterjee}, S., {Ford}, E.~B., {Matsumura}, S., \& {Rasio}, F.~A. 2007, ArXiv Astrophysics
479: e-prints
480:
481: \bibitem[{{Dobbs-Dixon} {et~al.}(2004){Dobbs-Dixon}, {Lin}, \&
482: {Mardling}}]{DobbsDixon04}
483: {Dobbs-Dixon}, I., {Lin}, D.~N.~C., \& {Mardling}, R.~A. 2004, ApJ, 610, 464
484:
485: \bibitem[{{Eggleton} {et~al.}(1998){Eggleton}, {Kiseleva}, \&
486: {Hut}}]{Eggleton98}
487: {Eggleton}, P.~P., {Kiseleva}, L.~G., \& {Hut}, P. 1998, ApJ, 499, 853
488:
489: \bibitem[{{Faber} {et~al.}(2005){Faber}, {Rasio}, \& {Willems}}]{Faber05}
490: {Faber}, J.~A., {Rasio}, F.~A., \& {Willems}, B. 2005, Icarus, 175, 248
491:
492: \bibitem[{{Fabrycky} \& {Tremaine}(2007)}]{Fabrycky07}
493: {Fabrycky}, D. \& {Tremaine}, S. 2007, ApJ, 669, 1298
494:
495: \bibitem[{{Ford} {et~al.}(2000){Ford}, {Kozinsky}, \& {Rasio}}]{Ford00}
496: {Ford}, E.~B., {Kozinsky}, B., \& {Rasio}, F.~A. 2000, ApJ, 535, 385
497:
498: \bibitem[{{Ford} \& {Rasio}(2006)}]{Ford06}
499: {Ford}, E.~B. \& {Rasio}, F.~A. 2006, ApJL, 638, L45
500:
501: \bibitem[{{Goldreich} \& {Soter}(1966)}]{Goldreich66}
502: {Goldreich}, P. \& {Soter}, S. 1966, Icarus, 5, 375
503:
504: \bibitem[{{Hebrard} {et~al.}(2008){Hebrard}, {Bouchy}, {Pont}, {Loeillet},
505: {Rabus}, {Bonfils}, {Moutou}, {Boisse}, {Delfosse}, {Desort}, {Eggenberger},
506: {Ehrenreich}, {Forveille}, {Lagrange}, {Lovis}, {Mayor}, {Pepe}, {Perrier},
507: {Queloz}, {Santos}, {Segransan}, {Udry}, \& {Vidal-Madjar}}]{Hebrard08ap}
508: {Hebrard}, G., et~al.
509: %{Bouchy}, F., {Pont}, F., {Loeillet}, B., {Rabus}, M.,
510: % {Bonfils}, X., {Moutou}, C., {Boisse}, I., {Delfosse}, X., {Desort}, M.,
511: % {Eggenberger}, A., {Ehrenreich}, D., {Forveille}, T., {Lagrange}, A.~M.,
512: % {Lovis}, C., {Mayor}, M., {Pepe}, F., {Perrier}, C., {Queloz}, D., {Santos},
513: % N.~C., {Segransan}, D., {Udry}, S., \& {Vidal-Madjar}, A.
514: 2008, ArXiv e-prints, 806
515:
516: \bibitem[{{Holman} {et~al.}(1997){Holman}, {Touma}, \& {Tremaine}}]{Holman97}
517: {Holman}, M., {Touma}, J., \& {Tremaine}, S. 1997, Nature, 386, 254
518:
519: \bibitem[{{Hut}(1981)}]{Hut81}
520: {Hut}, P. 1981, A\&A, 99, 126
521:
522: \bibitem[{{Kiseleva} {et~al.}(1998){Kiseleva}, {Eggleton}, \&
523: {Mikkola}}]{Kiseleva98}
524: {Kiseleva}, L.~G., {Eggleton}, P.~P., \& {Mikkola}, S. 1998, MNRAS, 300, 292
525:
526: \bibitem[{{Kozai}(1962)}]{Kozai62}
527: {Kozai}, Y. 1962, AJ, 67, 591
528:
529: \bibitem[{{Marcy} {et~al.}(1997){Marcy}, {Butler}, {Williams}, {Bildsten},
530: {Graham}, {Ghez}, \& {Jernigan}}]{Marcy97}
531: {Marcy}, G.~W., {Butler}, R.~P., {Williams}, E., {Bildsten}, L., {Graham},
532: J.~R., {Ghez}, A.~M., \& {Jernigan}, J.~G. 1997, ApJ, 481, 926
533:
534: \bibitem[{{Mardling} \& {Lin}(2002)}]{Mardling02}
535: {Mardling}, R.~A. \& {Lin}, D.~N.~C. 2002, ApJ, 573, 829
536:
537: \bibitem[{{Murray} \& {Dermott}(1999)}]{Murray99}
538: {Murray}, C.~D. \& {Dermott}, S.~F. 1999, {Solar system dynamics} (Solar system
539: dynamics by Murray, C.~D., 1999)
540:
541: \bibitem[{{Nagasawa} {et~al.}(2008){Nagasawa}, {Ida}, \&
542: {Bessho}}]{Nagasawa08ap}
543: {Nagasawa}, M., {Ida}, S., \& {Bessho}, T. 2008, ArXiv e-prints, 801
544:
545: \bibitem[{{Narita} {et~al.}(2007){Narita}, {Sato}, {Ohshima}, \&
546: {Winn}}]{Narita07ap}
547: {Narita}, N., {Sato}, B., {Ohshima}, O., \& {Winn}, J.~N. 2007, ArXiv e-prints,
548: 712
549:
550: \bibitem[{{Ogilvie} \& {Lin}(2004)}]{Ogilvie04}
551: {Ogilvie}, G.~I. \& {Lin}, D.~N.~C. 2004, ApJ, 610, 477
552:
553: \bibitem[{{Queloz} {et~al.}(2000){Queloz}, {Eggenberger}, {Mayor}, {Perrier},
554: {Beuzit}, {Naef}, {Sivan}, \& {Udry}}]{Queloz00}
555: {Queloz}, D., et~al.
556: %{Eggenberger}, A., {Mayor}, M., {Perrier}, C., {Beuzit}, J.~L.,
557: % {Naef}, D., {Sivan}, J.~P., \& {Udry}, S.
558: 2000, A\&A, 359, L13
559:
560: \bibitem[{{Rasio} {et~al.}(1996){Rasio}, {Tout}, {Lubow}, \& {Livio}}]{Rasio96}
561: {Rasio}, F.~A., {Tout}, C.~A., {Lubow}, S.~H., \& {Livio}, M. 1996, ApJ, 470,
562: 1187
563:
564: \bibitem[{{Ribas} {et~al.}(2008){Ribas}, {Font-Ribera}, \&
565: {Beaulieu}}]{Ribas08}
566: {Ribas}, I., {Font-Ribera}, A., \& {Beaulieu}, J.-P. 2008, ArXiv e-prints, 801
567:
568: \bibitem[{{Sterne}(1939)}]{Sterne39}
569: {Sterne}, T.~E. 1939, MNRAS, 99, 451
570:
571: \bibitem[{{Takeda} {et~al.}(2007){Takeda}, {Ford}, {Sills}, {Rasio}, {Fischer},
572: \& {Valenti}}]{Takeda07}
573: {Takeda}, G., {Ford}, E.~B., {Sills}, A., {Rasio}, F.~A., {Fischer}, D.~A., \&
574: {Valenti}, J.~A. 2007, ApJS, 168, 297
575:
576: \bibitem[{{Takeda} {et~al.}(2008){Takeda}, {Kita}, \& {Rasio}}]{Takeda08ap}
577: {Takeda}, G., {Kita}, R., \& {Rasio}, F.~A. 2008, ArXiv e-prints, 802
578:
579: \bibitem[{{Winn} {et~al.}(2005){Winn}, {Noyes}, {Holman}, {Charbonneau},
580: {Ohta}, {Taruya}, {Suto}, {Narita}, {Turner}, {Johnson}, {Marcy}, {Butler},
581: \& {Vogt}}]{Winn05}
582: {Winn}, J.~N., et~al.
583: %{Noyes}, R.~W., {Holman}, M.~J., {Charbonneau}, D., {Ohta}, Y.,
584: % {Taruya}, A., {Suto}, Y., {Narita}, N., {Turner}, E.~L., {Johnson}, J.~A.,
585: % {Marcy}, G.~W., {Butler}, R.~P., \& {Vogt}, S.~S.
586: 2005, ApJ, 631, 1215
587:
588: \bibitem[{{Wu}(2005)}]{Wu05b}
589: {Wu}, Y. 2005, ApJ, 635, 688
590:
591: \bibitem[{{Yoder} \& {Peale}(1981)}]{Yoder81}
592: {Yoder}, C.~F. \& {Peale}, S.~J. 1981, Icarus, 47, 1
593:
594: \bibitem[{{Zhang} \& {Hamilton}(2008)}]{Zhang08}
595: {Zhang}, K. \& {Hamilton}, D.~P. 2008, Icarus, 193, 267
596:
597: \end{thebibliography}
598:
599: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
600: % TABLE 1
601: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
602:
603: %\begin{table*}
604: %\begin{center}
605: \begin{deluxetable}{lccccccc}
606: %{|l|c|c|c|c|c|c|c|}
607: %\caption{Planetary and Stellar Parameters of Systems with Transiting
608: %Planets}
609: %\begin{tabular}{|l|c|c|c|c|c|c|c|}
610: %\tableline\tableline
611: %\tablehead{
612: %\colhead{$\Mp$ [$M_J$]} & \colhead{$\Rp$ [$R_J$]} & \colhead
613: %{$a$ [AU]} & \colhead{e} & \colhead{$M_*$} [$M_{\odot}$]} &
614: %\colhead{$R_*$ [$R_{\odot}$]} & \colhead{Age [Gyr]}
615: %
616: %\tabletypesize{\footnotesize}
617: \tablecaption{Data are from {\tt http://exoplanet.eu/}. Ages are
618: computed using the stellar evolution database in \cite{Takeda07},
619: unless marked with $\ast$.
620: Median values of the derived posterior age probability distribution
621: functions are presented here, together with the 95\% credible
622: intervals in parenthesis. \label{tb1}}
623: %
624: \tablecolumns{8} \tablehead{ \colhead{Planet ID} & \colhead{$\Mp$
625: [$M_J$]} & \colhead{$\Rp$ [$R_J$]} & \colhead{$a$ [AU]} &
626: \colhead{e} & \colhead{$M_*$ [$M_{\odot}$]} & \colhead{$R_*$
627: [$R_{\odot}$]} & \colhead{Age [Gyr]} }
628: %\\
629: %\tableline \tableline
630: %
631: \startdata
632: OGLE-TR-56 b & 1.29 & 1.3 & 0.0225 &
633: 0 & 1.17 & 1.32 & 0.92 (0.20\,--3.00) \\
634: OGLE-TR-113 b & 1.32 & 1.09 & 0.0229 & 0 & 0.78 & 0.77
635: & 13.28 (11.00\,--\,13.92) \\
636: GJ 436 b $\ast$ & 0.072 & 0.38 & 0.02872 &
637: 0.15 & 0.452 & 0.464 & 6.00 (1.00\,--\,10.00) \\
638: OGLE-TR-132 b & 1.14 & 1.18 & 0.0306
639: & 0 & 1.26 & 1.34 & 0.96 (0.12\,--\,3.84) \\
640: HD 189733 b & 1.15 & 1.156 & 0.0312 &
641: 0 & 0.8 & 0.753 & 8.96 (1.04\,--\,13.72) \\
642: TrES-2 $\ast$ & 1.98 & 1.22 & 0.0367 &
643: 0 & 0.98 & 1 & 5.10 (2.40\,--\,7.80) \\
644: WASP-14 b $\ast$ & 7.725 & 1.259 & 0.037 & 0.095 &
645: 1.319 & 1.297 & 0.75 (0.5\,--\,1) \\
646: WASP-10 b $\ast$ & 3.06 & 1.29 & 0.0371 & 0.057 & 0.71 & 0.783
647: & 0.8 (0.6\,--\,1) \\
648: HAT-P-3 b $\ast$ & 0.599 & 0.89 & 0.03894 &
649: 0 & 0.936 & 0.824 & 0.40 (0.10\,--\,6.90) \\
650: %TrES-1 & 0.61 & 1.081 & 0.0393 & 0.135
651: %& 0.87 & 0.82 & 11.40 (3.20\,--\,13.84) \\
652: TrES-1 & 0.61 & 1.081 & 0.0393 & 0
653: & 0.87 & 0.82 & 11.40 (3.20\,--\,13.84) \\
654: HAT-P-5 b $\ast$ & 1.06 & 1.26 & 0.04075 &
655: 0 & 1.16 & 1.167 & 2.60 (0.80\,--\,4.40) \\
656: OGLE-TR-10 b & 0.63 & 1.26 & 0.04162
657: & 0 & 1.18 & 1.16 & 1.20 (0.16\,--\,4.64) \\
658: HD 149026 b $\ast$ & 0.36 & 0.71 & 0.0432 &
659: 0 & 1.3 & 1.45 & 2.00 (1.20\,--\,2.80) \\
660: HAT-P-4 b $\ast$ & 0.68 & 1.27 & 0.0446 &
661: 0 & 1.26 & 1.59 & 4.20 (3.60\,--\,6.80) \\
662: %HD 209458 b & 0.69 & 1.32 & 0.045 &
663: %0.07 & 1.01 & 1.12 & 4.00 (2.00\,--\,6.00) \\
664: %HD 209458 b $\ast$ & 0.69 & 1.32 & 0.045 &
665: %0 & 1.01 & 1.12 & 4.00 (2.00\,--\,6.00) \\
666: HD 209458 b & 0.69 & 1.32 & 0.045 &
667: 0.07 & 1.01 & 1.12 & 2.40 (0.48\,--\,4.60) \\
668: OGLE-TR-111 b & 0.53 & 1.067 & 0.047
669: & 0 & 0.82 & 0.831 & 5.17 (0.17\,--\,13.41) \\
670: %XO-3 b & 13.24 & 1.92 & 0.0476 & 0.219
671: %& 1.41 & 2.13 & 2.69 (2.51\,--\,2.96) \\
672: HAT-P-6 b $\ast$ & 1.057 & 1.33 & 0.05235 &
673: 0 & 1.29 & 1.46 & 2.30 (1.60\,--\,2.80) \\
674: %HAT-P-1 b & 0.53 & 1.36 & 0.0551 &
675: %0.09 & 1.12 & 1.15 & 3.60 \\
676: %HAT-P-1 b & 0.53 & 1.36 & 0.0551 &
677: %0 & 1.12 & 1.15 & 3.60 \\
678: HAT-P-1 b $\ast$ & 0.524 & 1.36 & 0.0553 &
679: 0.067 & 1.133 & 1.115 & 3.60 \\
680: HAT-P-2 b $\ast$ & 8.64 & 0.952 & 0.0677 & 0.517
681: & 1.298 & 1.412 & 2.70 (1.30\,--\,4.10) \\
682: %%\tableline
683: %%HD 17156 b & 3.111 & 0.964 & 0.1594 &
684: %%0.6717 & 1.2 & 1.47 & 5.7 & 4.4 & 7. \\
685: %\tableline
686: \enddata
687: %\end{tabular}
688: %\tablecomments
689: %\tablecaption{Data are from {\tt http://exoplanet.eu/}. Ages are
690: %computed using the stellar evolution database in \cite{Takeda07}.
691: %Median values of the derived posterior age probability distribution
692: %functions are presented here, together with the 95\% credible
693: %intervals in parenthesis. \label{tb1}}
694: \end{deluxetable}
695: %\end{center}
696: %\end{table*}
697:
698: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
699:
700:
701: \end{document}
702: