1: \documentclass[12pt,preprint]{aastex}
2:
3: \def\sun{\odot}
4: \def\earth{\oplus}
5: %\usepackage{epsfig}
6: \begin{document}
7: \title{The Ratio of Retrograde to Prograde Orbits: A Test for Kuiper
8: Belt Binary Formation Theories} \shortauthors{Schlichting & Sari}
9: \shorttitle{Prograde and Retrograde Orbits of Kuiper Belt Binaries}
10: \author{Hilke E. Schlichting\altaffilmark{1} and Re'em Sari\altaffilmark{1,2} }
11: \altaffiltext{1}{California Institute of Technology, MC 130-33, Pasadena, CA
12: 91125}
13: \altaffiltext{2}{Racah Institute of Physics, Hebrew University,
14: Jerusalem 91904, Israel}
15: \email{hes@astro.caltech.edu, sari@tapir.caltech.edu}
16:
17:
18: \begin{abstract}
19: With the discovery of Kuiper Belt binaries that have wide separations and
20: roughly equal masses new theories were proposed to explain their
21: formation. Two formation scenarios were suggested by Goldreich and
22: collaborators: In the first, dynamical friction that is generated by a sea
23: of small bodies enables a transient binary to become bound ($L^2s$ mechanism);
24: in the second, a transient binary gets bound by an encounter with a third body
25: ($L^3$ mechanism).
26:
27: We show that these different binary formation scenarios leave their own unique
28: signatures in the relative abundance of prograde to retrograde binary
29: orbits. This signature is due to stable retrograde orbits that exist much
30: further out in the Hill sphere than prograde orbits. It provides an excellent
31: opportunity to distinguish between the different binary formation scenarios
32: observationally.
33:
34: We predict that if binary formation proceeded while sub-Hill velocities
35: prevailed, the vast majority of all comparable mass ratio binaries have
36: retrograde orbits. This dominance of retrograde binary orbits is a result of
37: binary formation via the $L^2s$ mechanism, or any other mechanism that
38: dissipates energy in a smooth and gradual manner. For super-Hill velocities
39: binary formation proceeds via the $L^3$ mechanism which produces a roughly
40: equal number of prograde and retrograde binaries. These predictions assume
41: that subsequent orbital evolution due to dynamical friction and
42: dynamical stirring of the Kuiper belt did not alter the sense of the binary
43: orbit after formation.
44: \end{abstract}
45:
46: \keywords {Kuiper Belt --- planets and satellites: formation}
47:
48: \section{INTRODUCTION}
49: The detection of comparable mass binaries with wide separations in the Kuiper
50: Belt called for new theories explaining their formation
51: \citep[e.g.][]{W02,GLS02,F04,A05,A07}. Their existence cannot be explained
52: with a formation scenario involving a collision and tidal evolution, as has
53: been proposed for the formation of the Moon and Charon \citep{HD75,CW76,MK89},
54: since it cannot provide the current angular momentum of the binary system. In
55: a formation scenario proposed by \citet{W02} two Kuiper Belt objects (KBOs)
56: collide with each other inside the Hill sphere of a third. However, in the
57: Kuiper Belt, gravitational scattering between the two intruders is about 100
58: times\footnote{For this estimate we used $\alpha\sim 10^{-4}$ and assumed that
59: the velocity dispersion of the KBOs at the time of binary formation is less
60: than their Hill velocity, see \S 2 for details} more common than a
61: collision. Therefore, three body gravitational deflection ($L^3$ mechanism),
62: as proposed by \citet{GLS02}, should dominate the binary formation over such
63: collisional scenario. A second binary formation scenario that has been
64: suggested by \citet{GLS02} consists of the formation of a transient binary
65: that gets bound with the aid of dynamical friction from a sea of small
66: bodies. We call this the $L^2s$ mechanism. In the formation scenario of
67: \cite{A05} and \cite{A07} the existence of long lived transient binaries that
68: spend a long time in their mutual Hill sphere, near a periodic orbit, is
69: responsible for the creation of Kuiper Belt binaries (KBBs). Finally,
70: \citet{F04} proposed a binary formation mechanism that involves a collision
71: between two large KBOs. This collision creates a small moon that is replaced
72: in an exchange reaction by a massive body with high eccentricity and large
73: semi-major axis.
74:
75: In this paper, we show that the $L^2s$ and $L^3$ mechanism leave unique
76: signatures in the relative abundance of prograde to retrograde binary
77: orbits. The $L^2s$ mechanism dominates over the $L^3$ mechanism for sub-Hill
78: velocities \citep{SR07}. We argue that binaries that form from dynamically
79: cold KBOs by the $L^2s$ mechanism have retrograde orbits. This is due to the
80: existence of stable retrograde binary orbits with modified Jacobi constants
81: similar to that of unbound KBOs on circular orbits that have impact parameters
82: that correspond to distances of closest approach of less than the Hill
83: radius. No equivalent prograde orbits exist
84: \citep[e.g.][]{H70,I79,ZI88,HB91,HK97}. Since dynamical friction only
85: gradually increases the modified Jacobi constant (for a binary this
86: corresponds to gradually increasing the absolute value of the binding energy),
87: all binaries that form via the $L^2s$ mechanism, or any other mechanism that
88: dissipates energy in a smooth and gradual manner, will start with modified
89: Jacobi constants close to that of unbound KBOs that penetrate the Hill sphere
90: and hence have retrograde orbits. For super-Hill KBO velocities, only the
91: $L^3$ mechanism can form tight binaries that tend to survive \citep{SR07}. The
92: fact that retrograde orbits are stable for larger semi-major axes is no longer
93: of importance since only tight binaries are saved from break up. This,
94: therefore, leads to the formation of a roughly equal number of prograde and
95: retrograde binaries for super-Hill KBO velocities.
96:
97: Our paper is structured as follows: In \S 2 we outline our assumptions,
98: explain our choice of parameters and define variables that will be used
99: throughout this paper. We calculate the ratio of prograde to retrograde binary
100: orbits for the $L^2s$ and $L^3$ mechanism and predict the relative abundance
101: of prograde to retrograde orbits for sub-Hill and super-Hill KBO velocities in
102: \S 3. We compare our predictions with observations in \S 4. Discussion and
103: conclusions follow in \S 5.
104:
105:
106: \section{DEFINITIONS AND ASSUMPTIONS}
107: The Hill radius denotes the distance from a body at which the tidal forces due
108: to the Sun and the gravitational force due to the body, both acting on a test
109: particle, are in equilibrium. It is given by
110: \begin{equation}\label{e1}
111: R_{H} \equiv a_{\sun} \left( \frac{m_1+m_2}{3 M_{\sun}}\right) ^{1/3}
112: \end{equation}
113: where $m_1$ and $m_2$ are the masses of the two KBOs, $a_{\sun}$ is their
114: semi-major axis around the Sun and $M_{\sun}$ the mass of the Sun. Our
115: definition of the Hill radius differs from that used by \citet{SR07} since we
116: include the combined mass of both KBOs here. We chose to do so since it will
117: make comparisons with works by other authors easier.
118:
119: We use the `two-group approximation' \citep{GLS02,GLS04} which consists of the
120: identification of two groups of objects, small ones, that contain most of the
121: total mass with surface mass density $\sigma$, and large ones, that contain
122: only a small fraction of the total mass with surface mass density $\Sigma \ll
123: \sigma $. We assume $\sigma \sim 0.3 \rm{g~cm^{-2}}$ which is the
124: extrapolation of the minimum-mass solar nebular \citep{H81} to a heliocentric
125: distance of $40\rm{AU}$. Estimates from Kuiper Belt surveys
126: \citep{TJL01,TB03,PK08,F08,FH08} yield $\Sigma \sim 3 \times 10^{-4}
127: \rm{g~cm^{-2}}$ for KBOs with radii of $R \sim 100~\rm{km}$. We use this value
128: of $\Sigma$, assuming that $\Sigma$ during the formation of KBBs was the same
129: as it is now. Our choice for $\Sigma$ and $\sigma$ is also consistent with
130: results from numerical coagulation simulations by \citet{KL99}.
131:
132: Large bodies grow by the accretion of small bodies. Large KBOs viscously stir
133: the small bodies, increasing the small bodies' velocity dispersion $u$. As a
134: result $u$ grows on the same timescale as $R$ provided that mutual collisions
135: among the small bodies are not yet important. In this case, $u$ is given by
136: \begin{equation}\label{e2}
137: \frac{u}{v_H} \sim \left( \frac{\Sigma}{\sigma \alpha} \right)^{1/2} \sim 3
138: \end{equation}
139: where $\alpha = R/R_{H}\sim 10^{-4}$ at $40\rm{AU}$ \citep{GLS02}. $v_H$ is
140: the Hill velocity of the large bodies which is given by $v_H = \Omega R_H$
141: where $\Omega$ is the orbital frequency around the sun. The velocity $v$ of
142: large KBOs increases due to mutual viscous stirring, but is damped by
143: dynamical friction from a sea of small bodies such that $v < u$. Balancing
144: the stirring and damping rates of $v$ and substituting for $u$ from equation
145: (\ref{e2}), we find
146: \begin{equation}\label{e3}
147: \frac{v}{v_H} \sim \alpha^{-2} \left(\frac{\Sigma}{\sigma}\right)^{3} \sim
148: 0.1.
149: \end{equation}
150: For our choice of parameters, we have sub-Hill KBO velocities during the epoch
151: of formation of bodies with $R\sim 100\rm{km}$. We therefore focus our work on
152: the shear-dominated velocity regime ($v \ll v_{H}$). However, we discuss how
153: our results would be modified if $v \gg v_H$.
154:
155:
156: \section{PROGRADE VERSUS RETROGRADE BINARY ORBITS}
157:
158: \subsection{Sub-Hill velocities: $v \ll v_H$}
159: The disk of KBOs is effectively two-dimensional in the shear-dominated
160: velocity regime ($v \ll v_H$), since the growth of inclinations is suppressed
161: \citep{WS1993,R2003,GLS04}. We therefore restrict our calculations for the
162: shear-dominated velocity regime to two dimensions. Since we are interested in
163: close encounters among the KBOs, their interaction is well described by Hill's
164: equations \citep{H78,GT80,HP86}. In Hill coordinates the equations of motion
165: of the two KBOs can be decomposed into their center of mass motion and their
166: relative motion with respect to one another. The modified Jacobi constant is
167: exactly conserved in the Hill formalism, but the Hill formalism itself is an
168: approximation to the general three body problem. It assumes that the masses of
169: body 1 and 2 (in our case the two KBOs) are much less than that of the Sun. We
170: use the standard Hill coordinate system and reference frame as in \citet{HP86}
171: and \citet{Ida90}. In this rotating frame the direction of the $x$-axis is
172: given by the line connecting the Sun and the center of mass of the two KBOs
173: such that the positive $x$ direction is pointing away from the Sun. The
174: $y$-axis is perpendicular to the $x$-axis pointing in the direction of motion
175: of the KBOs' center of mass around the Sun. In Hill coordinates the modified
176: Jacobi constant is
177: \begin{equation}\label{e4}
178: J_C=3x^2 + \frac{6}{(x^2+y^2)^{1/2}} -\dot{x}^2 -\dot{y}^2
179: \end{equation}
180: where $x$ and $y$ correspond to the relative separation between the two KBOs
181: in the $x$ and $y$ direction respectively \citep{HP86}. Length has been scaled
182: by $R_{H}$ and time by $\Omega^{-1}$. In Hill coordinates the Lagrangian
183: points $L_1$ and $L_2$ are located at $(-1,0)$ and $(+1,0)$ respectively,
184: where we define $L_1$ as the Lagrangian point located between the KBO and the
185: Sun. Their modified Jacobi constants are $J_C(L_1)=J_C(L_2)=9$. From equation
186: \ref{e4} we can see that tight binaries with small separations have $J_C \gg
187: 9$. We call a binary orbit prograde if its angular momentum about the binary
188: center of mass, as viewed in the non rotating frame, is in the same direction
189: as the orbital angular momentum of the binary around the Sun. If the binary
190: angular momentum is in the opposite direction to the orbital angular momentum
191: of the binary around the Sun, the orbit is called retrograde. Several authors
192: recognized that planar retrograde orbits are stable for larger semi-major axes
193: than prograde orbits \citep[e.g.][]{H70,I79,ZI88,HB91,HK97}. A prograde
194: binary with an initially circular orbit becomes unbound for $a \gtrsim 0.49
195: R_H$ where $a$ is the initial semi-major axis of the mutual binary orbit
196: \citep{HB91}. This implies that prograde orbits with modified Jacobi constants
197: less than that of the Lagrangian points $L_1$ and $L_2$ are unbound. In
198: contrast to the prograde case, there exist stable retrograde binary orbits
199: with $J_C \lesssim J_C(L_1)=J_C(L_2)=9$. This result is also shown in Figures
200: \ref{fig1}. Figure \ref{fig1} shows histograms of $J_C$ for prograde and
201: retrograde binaries that formed by $L^3$ mechanism from KBOs with initially
202: circular orbits around the Sun. In the reminder of this paper we discuss the
203: stability of prograde and retrograde orbits in terms of $J_C$ and not
204: semi-major axis since the latter is not well defined (i.e. it is not a
205: constant of motion) for wide orbits with $a \sim R_H$. The modified Jacobi
206: constant for two KBOs that approach each other from infinity is
207: \begin{equation}\label{e14}
208: J_C=3x^2 -\dot{x}^2 -\dot{y}^2=\frac{3}{4}b^2 - e^2
209: \end{equation}
210: where $b$ is the initial separation between the two KBOs in the $x$ direction
211: and $e$ is the relative eccentricity in Hill units given by $\vert
212: \mathbf{e_1}-\mathbf{e_2} \vert$ where $\mathbf{e_1}$ and
213: $\mathbf{e_2}$ are the eccentricity vectors of body 1 and body 2
214: respectively. Only KBOs with $b$ ranging from $1.7 R_H$ to $2.5 R_H$ penetrate
215: each others Hill sphere if started on circular orbits. From equation
216: (\ref{e14}) we have therefore that only KBOs with $2.2 \le J_C \le 4.7$ have a
217: distance of closest approach of $R_H$ or less provided that they started on
218: circular orbits around the Sun.
219:
220:
221: \subsubsection{$L^2s$ Mechanism}
222: In the $L^2s$ mechanism KBBs form from transient binaries that become bound
223: with the aid of dynamical friction from a sea of small bodies. This dynamical
224: friction provides a gentle force that damps the random velocity of large
225: KBOs. For typical parameters, the dynamical friction force only extracts a
226: small fraction of energy over an orbital timescale. Therefore, KBBs that form
227: via the $L^2s$ mechanism, or any other mechanism that dissipates energy
228: gradually, have initially modified Jacobi constants similar to that of the
229: unbound KBOs that penetrate within the Hill sphere. As mentioned above, for
230: KBOs that started on circular orbits around the Sun this corresponds to $2.2
231: \le J_C \le 4.7$. However, only stable retrograde orbits exist with $J_C
232: \lesssim 9$. This implies that all KBB that form this way must have retrograde
233: orbits since no stable prograde orbits exist for $J_C \lesssim 9$. Once a
234: binary is formed, dynamical friction increases the modified Jacobi constant and
235: the absolute value of the binary binding energy. We confirm that all binaries
236: that form from KBOs on initially circular orbits around the Sun via the
237: $L^2_s$ mechanism are retrograde by numerical integrations that are presented
238: below.
239:
240: Since it is not feasible to examine the interactions with each small body
241: individually, their net effect is modeled by an averaged force which acts to
242: damp the large KBOs' non-circular velocity around the Sun. We parameterize the
243: strength of the damping by a dimensionless quantity $D$ defined as the
244: fractional decrease in non-circular velocity due to dynamical friction over a
245: time $\Omega^{-1}$:
246: \begin{equation}\label{e5}
247: D\sim \frac{ \sigma}{\rho R} \left(\frac{u}{v_H}\right)^{-4} \alpha^{-2} \sim
248: \frac{\Sigma}{\rho R} \alpha^{-2} \left(\frac{v}{v_H}\right)^{-1}.
249: \end{equation}
250: The first expression is simply an estimate of dynamical friction by a sea of
251: small bodies assuming $u>v_H$. The second expression describes the mutual
252: excitation among the large KBOs for $v \ll v_H$. $v$ achieves a quasi-steady
253: state on a timescale shorter than at which $R$ grows since only a subset of
254: the deflected bodies are accreted. The stirring among the large KBOs can
255: therefore be equated to the damping due to dynamical friction (for a detailed
256: derivation see \citet{GLS04}).
257:
258: Since the growth of inclinations is suppressed in the shear-dominated velocity
259: regime the disk of KBOs is effectively two-dimensional
260: \citep{WS1993,R2003,GLS04}. We therefore restrict this calculation to two
261: dimensions. In Hill coordinates the relative motion of two equal mass KBOs,
262: including the dynamical friction term, is governed by
263: \begin{equation}\label{e6}
264: \ddot{x}-2 \dot{y}
265: -3x=-\frac{3 x}{(x^2+y^2)^{3/2}}-D \dot{x}
266: \end{equation}
267: \begin{equation}\label{e7}\ddot{y}+2
268: \dot{x}=-\frac{3 y}{(x^2+y^2)^{3/2}} -D (\dot{y}+1.5x).
269: \end{equation}
270: Length has been scaled by $R_{H}$ and time by $\Omega^{-1}$. Equations
271: (\ref{e6}) and (\ref{e7}) are integrated for different values of $D$ and
272: impact parameters ranging from $1.7 R_H$ to $2.5 R_H$ with equal step
273: size. Impact parameters outside this range result in a distance of closest
274: approach between the two KBOs of more than $R_H$.
275:
276: For $D=0.01$, we performed 20000 integrations. About two percent of these
277: integrations resulted in the formation of a binary. Figure \ref{fig2} shows
278: three examples of the evolution of the specific angular momentum and $J_C$ of
279: binary formation events from our integrations for $D=0.01$. In addition, we
280: performed integrations for values of $D$ ranging from 0.1 to 0.0004 and find
281: that, just like in the $D=0.01$ case, only retrograde binaries form. We define
282: $h$ as the specific angular momentum of the binary in the non rotating
283: frame. It can be written as $h=x\dot{y}-y\dot{x}+x^2+y^2$ and is related to
284: the total binary orbital angular momentum, $L$, by $h=(1/m_1+1/m_2) L$. The
285: time $t=0$ corresponds to the time at which $y=0$ if the relative KBO velocity
286: is solely due to the Keplerian sheer (i.e. ignoring the actual gravitational
287: interaction between the bodies). The evolution of $h$ and $J_C$ is shown until
288: the binary separation has decreased to $0.1 R_H$ or less. Binaries with
289: separations of $0.1 R_H$ or less are sufficiently tight that perturbations
290: from the Sun are too weak to flip the sign of the angular momentum. As
291: expected from our discussion above, the angular momenta of the binaries are
292: negative corresponding to retrograde binary orbits. In fact all binaries that
293: form via the $L^2s$ mechanism in our numerical integrations display retrograde
294: orbits. Dynamical friction shrinks the binary separation. As a result, the
295: magnitude of the binary angular momenta decreases with time. The right hand
296: side of Figure \ref{fig2} shows the evolution of the modified Jacobi
297: constant. Newly formed binaries initially have a modified Jacobi constant $<9$
298: which is possible only for retrograde binaries. Dynamical friction shrinks the
299: semi-major axes of the binaries which leads to an increase of $J_C$ with time
300: while keeping the sense of rotation, i.e. the sign of $h$, fixed. Eventually
301: the modified Jacobi constant grows to values above $J_C(L1)=J_C(L2)= 9$. For
302: $J_C\gtrsim 9$ prograde orbits can exist; however all binaries that formed
303: with the aid of dynamical friction started out with $J_C < 9$ for which only
304: retrograde orbits are stable. Therefore, all KBB that form via the $L^2s$
305: mechanism, or any other mechanism that gradually removes energy from transient
306: binaries, orbit each other in the retrograde sense since otherwise they would
307: not be able to form in the first place. Figure \ref{fig3} shows the evolution
308: of $h$ and $J_C$ as a function of time for KBO encounters that did not lead to
309: the formation of a binary. These examples show that KBOs encounter each other
310: and leave each other with positive angular momenta. This is a result of the
311: Keplerian sheer and follows from the definition of $h$.
312:
313: We have assumed here that all KBOs are initially on circular orbits around the
314: Sun and have shown that this leads to the formation of exclusively retrograde
315: binaries in the $L^2s$ mechanism. If, however, the velocity dispersion of the
316: KBOs is sufficiently large, such that $e$ is of the order of the Hill
317: eccentricity, bigger impact parameters allow the KBOs to penetrate each others
318: Hill sphere. In this case, there now exist KBOs that have an initial $J_C$
319: just a little below $9$ (see equation \ref{e14}) in which case only a small
320: change in $J_C$ is sufficient for the formation of retrograde and prograde
321: binaries. Therfore, prograde binaries can form with the aid of dynamical
322: friction provided that the velocity dispersion of the KBOs is about $v_H$.
323:
324: Our prediction for the sense of the binary orbit relies on the assumption that
325: dynamical friction does not alter the sense of the binary orbit in the
326: subsequent binary evolution. Although we have shown in our simulations that for
327: our dynamical friction model this is indeed the case, it might be that the
328: actual behavior of dynamical friction differs from the model implemented here.
329:
330:
331: \subsubsection{$L^3$ Mechanism}
332: A transient binary forms when two large KBOs penetrate each other's Hill
333: sphere. This transient binary must lose energy in order to become
334: gravitationally bound. In the $L^3$ mechanism the excess energy is carried
335: away by an encounter with a third massive body. This encounter can provide a
336: significant change in energy which corresponds to a considerable change in
337: $J_C$. The modified Jacobi constants of KBBs that form via the $L^3$
338: mechanism are therefore not constraint to values similar to that of their
339: initial $J_C$; their orbits can therefore be both prograde and retrograde. We
340: show that this is indeed the case with numerical integrations discussed below
341: and determine the ratio of prograde to retrograde orbits for binary formation
342: via the $L^3$ mechanism.
343:
344: Our calculation is performed in the shear-dominated velocity regime in two
345: dimensions. As initial condition, we assume that all bodies are on circular
346: orbits. We modify Hill's equations \citep{H78,GT80,PH86} to include three
347: equal mass bodies besides the Sun. The equations of motion, with length scaled
348: by $R_{H}$ and time by $\Omega^{-1}$, for body 1 are given by
349: \begin{equation}\label{e8}
350: \ddot{x}_{1}-2
351: \dot{y}_{1} -3x_{1}=-\frac{3(x_{1}-x_{2})}
352: {2((x_{1}-x_{2})^2+(y_{1}-y_{2})^2)^{3/2}}-\frac{3(x_{1}-x_{3})}
353: {2((x_{1}-x_{3})^2+(y_{1}-y_{3})^2)^{3/2}} \end{equation}
354: \begin{equation}\label{e9}
355: \ddot{y}_{1}+2\dot{x}_{1}=-\frac{3
356: (y_{1}-y_2)} {2((x_{1}-x_2)^2+(y_{1}-y_2)^2)^{3/2}}-\frac{3(y_{1}-y_{3})}
357: {2((x_{1}-x_{3})^2+(y_{1}-y_{3})^2)^{3/2}}.
358: \end{equation}
359: The subscripts 1, 2 and 3 label the $x$- and $y$-coordinates of KBO 1, 2, and
360: 3 respectively. Similar equations of motion can be obtained for bodies 2 and
361: 3. Resulting binary orbits are calculated by numerically integrating the
362: equations of motion. We refer the reader to \citet{SR07}
363: for the exact details of these calculations.
364:
365: Figure \ref{fig1} shows histograms of the modified Jacobi constant of prograde
366: and retrograde binaries that formed via the $L^3$ mechanism. Both histograms
367: are normalized to unity. As discussed above, we indeed find that prograde
368: orbits only exist for $J_C \gtrsim 9$. The stability of retrograde orbits
369: extends below $J_C=9$ down to $J_C \sim -10$. It therefore includes the range
370: $2.2< J_C <4.7$ with orbits that penetrate the Hill sphere from circular
371: heliocentric orbits. Unlike the $L^2s$ mechanism, the $L^3$ mechanism does
372: produce retrograde {\em and} prograde binaries for $v \ll v_H$. We find that
373: 65\% of all binary orbits are retrograde and 35\% prograde (see Figure
374: \ref{fig4}). Here, we only considered binary formation from three equal mass
375: bodies that started on initially circular orbits around the Sun. We therefore
376: caution, that the ratio of prograde to retrograde orbits due to the $L^3$
377: mechanism might differ for other mass ratios and velocity dispersions.
378:
379:
380: \subsubsection{The Ratio of Retrograde to Prograde Orbits}
381: \citet{SR07} have shown that for sub-Hill KBO velocities the ratio of the
382: $L^3$ to $L^2s$ binary formation rate is
383: \begin{equation}\label{e10}
384: \frac{FR_{L^3}}{FR_{L^2s}}=0.05\frac{v}{v_H}.
385: \end{equation}
386: Therefore, for sub-Hill KBO velocities, binaries in the Kuiper Belt form
387: primarily due to dynamical friction. For our estimate of $(v/v_H)\sim 0.1$,
388: we have that $ FR_{L^3}/FR_{L^2s} \sim 0.005 $, in which case $\sim 0.5 \%$
389: of all binaries form directly by the $L^3$ mechanism. Since prograde binaries
390: can only form via the $L^3$ mechanism, they make up a negligible fraction of
391: the total binaries. Below we discuss how a somewhat larger fraction of
392: prograde binaries can arise due to exchange reactions with unbound KBOs.
393:
394: Once a binary is formed its semi-major axis shirks due to dynamical friction
395: provided by a sea of small bodies. Dynamical friction decreases the orbit of
396: a KBB that has an orbital velocity $v_B$ at a rate
397: \begin{equation}\label{e20}
398: \mathcal{R}_{shrink} \sim D \Omega \sim \frac{\Sigma}{\rho R} \alpha^{-2} \left(\frac{v}{v_H}\right)^{-1}
399: \end{equation}
400: where we assume that $v_B<u$. Exchange reactions or binary break up by passing
401: KBOs occurs at a rate given by
402: \begin{equation}\label{e12}
403: \mathcal{R}_{exchange} \sim \frac{\Sigma }{\rho R} \alpha^{-2} \Omega
404: \left(\frac{v_B}{v_H}\right)^{-1}.
405: \end{equation}
406: The ratio of these two rates is given by
407: \begin{equation}\label{e13}
408: \frac{\mathcal{R}_{shrink}}{\mathcal{R}_{exchange}}\sim
409: \left(\frac{v_B}{v}\right)
410: \end{equation}
411: where $v \ll v_H$ and $v_B \gtrsim v_H$. Break up or exchange reactions are
412: most likely for wide binaries, in which case $v_B \sim v_H$ since $v_B$
413: increases as the semi-major axis of the mutual binary orbit
414: decreases. Therefore we have from equation (\ref{e13}) that
415: $\mathcal{R}_{shrink}/\mathcal{R}_{exchange} \sim (v_H/v) \sim 10$ for our
416: estimate of $(v/v_H)\sim 0.1$. This implies that $\sim$ 10\% of all binaries
417: that formed will suffer an exchange reaction or break up. We performed
418: numerical integrations of binary break up and exchange reactions to obtain a
419: more accurate estimate and find that only about 3\% of the binaries suffer an
420: exchange reaction and/or break up. Our order of magnitude calculation,
421: therefore, slightly over estimates the number of binaries that experience an
422: exchange reaction and/or break up. Moreover, only a fraction of the these
423: binaries will end up as binaries with prograde orbits. In conclusion, we
424: predict that the vast majority ($\gtrsim 97$\%) of comparable mass ratio
425: binaries will have retrograde orbits if KBO velocities of $v \lesssim 0.1 v_H$
426: prevailed during binary formation. This prediction assumes that subsequent
427: orbital evolution due to dynamical friction did not alter the sense of the
428: binary orbit after formation.
429:
430:
431: \subsection{Super-Hill Velocity: $v \gg v_H$}
432: There is some uncertainty in what the actual values of $\sigma$ and $\Sigma$
433: were during binary formation. For a few times larger value of $ \Sigma $ with
434: $ \sigma $ unchanged, we enter the regime in which $v$ exceeds the Hill
435: velocity (this can be seen from equation (\ref{e3})). We discuss here briefly
436: how this would affect the ratio of prograde to retrograde binary orbits.
437:
438: \citet{SR07} have shown that, for $v \gg v_H$, only binaries that form with a
439: binary separation of $R_{crit}=R_H(v_H/v)^2$ or less tend to be saved from
440: break up. The $L^2s$ mechanism fails in creating binaries with separations
441: $\sim R_{crit}$ or less since dynamical friction is not able to dissipate
442: sufficient energy for tight binaries to form. Therefore, the $L^2s$ mechanism
443: is not important if KBOs have super-Hill velocities. Tight binaries (with
444: separations less than $\lesssim R_{crit}$), can form via the $L^3$
445: mechanism. However in this case, the binary formation cross section is
446: significantly reduced with respect to the sub-Hill velocity regime (see
447: \citet{N07} and \citet{SR07} for details). The fact that retrograde orbits are
448: stable for larger semi-major axes is no longer of importance since only tight
449: binaries tend to survive. We therefore predict that a roughly equal number of
450: prograde and retrograde binaries form if super Hill velocities prevail. This
451: prediction is supported by Figure \ref{fig4}. Figure \ref{fig4} shows the
452: ratio of retrograde binaries with a modified Jacobi constant of $J_C^{min}$ or
453: larger to the total number of binaries that formed via the $L^3$ mechanism for
454: $v \ll v_H$. When all binaries are included we find that about $2/3$ have
455: retrograde orbits. More retrograde than prograde binaries form because
456: retrograde binary orbits are stable further out in the Hill sphere than
457: prograde ones. As $J_C^{min}$ increases the fraction of retrograde binaries
458: decreases reaching a minimum of about $1/3$ for $J_C^{min} \sim9$. This may be
459: due to the Keplerian sheer which increases the duration of a prograde
460: encounter between unbound KBOs compared to a retrograde encounter. The
461: fraction of prograde and retrograde binaries becomes comparable for $J_C^{min}
462: \gg 9$ because for such binaries neither the Keplerian sheer nor the increased
463: stability of retrograde orbits are important. This is the relevant regime for
464: binaries that form for $v \gg v_H$ since these large modified Jacobi constants
465: correspond to tight binaries, which are the only binaries that are saved from
466: break up if super-Hill velocities prevail.
467:
468:
469: \section{COMPARISON WITH OBSERVATIONS}
470: To date more than a dozen KBBs have well determined orbits
471: \citep[e.g.][]{N07}. Unfortunately due to projection effects, the prograde and
472: retrograde orbital solutions of the KBBs are nearly degenerate. This
473: degeneracy can usually only be broken after several years once the viewing
474: angle of the KBBs has changed sufficiently. Very recently, after the
475: submission of our original manuscript, two groups reported unique orbital
476: solutions for KBBs Typhon-Echidna \citep{GN08} and $2001~QW_{322}$
477: \citep{P08}. \citet{GN08} find a prograde orbit for Typhon-Echidna and
478: \citet{P08} report a retrograde orbit for $2001~QW_{322}$. $2001~QW_{322}$ has
479: such a large binary separation that, even in the current Kuiper belt, it
480: experiences significant dynamical interactions with other large KBOs. It is
481: early to draw conclusions for the whole binary population, but if a comparable
482: number of retrograde and prograde binaries is found it would imply that KBBs
483: formed from a dispersion dominated KBO disk, which would also be consistent
484: with observed binary inclinations. Dispersion dominated KBO velocities would
485: imply that the value of $\Sigma/\sigma$ was larger during binary formation
486: than what we used in equation \ref{e3}. However, the velocity dispersion
487: during binary formation cannot have exceeded $v_H$ significantly since the
488: binary formation timescales would otherwise become excessively long
489: \citep{N07,SR07}.
490:
491:
492: \section{DISCUSSION AND CONCLUSIONS}
493: The relative abundance of prograde to retrograde orbits enables us to
494: differentiate between various proposed binary formation scenarios
495: observationally.
496:
497: We predict that the vast majority ($\gtrsim 97$ \%) of comparable mass ratio
498: binaries will have retrograde orbits if KBO velocities of $\lesssim 0.1 v_H$
499: prevailed during their formation. This dominance of retrograde over prograde
500: binary orbits is due to the fact that for sub-Hill velocities binaries form
501: primarily via the $L^2s$ mechanism instead of $L^3$ mechanism. Since dynamical
502: friction only gradually increases the modified Jacobi constant, all binaries
503: that form via the $L^2s$ mechanism, or any other mechanism that dissipates
504: little energy over an orbital timescale, will start with modified Jacobi
505: constants close to that of unbound KBOs. Only stable retrograde orbits exist
506: for modified Jacobi constants similar to that of KBOs with initially circular
507: orbits around the Sun that penetrate inside the Hill sphere. Therefore, KBBs
508: have retrograde orbits provided that they form from dynamically cold KBOs via
509: the $L^2s$ mechanism.
510:
511: As the KBO velocities approach $v_H$ the preference of retrograde orbits
512: decreases. Further, we predict a comparable number of
513: prograde and retrograde binaries form for super-Hill KBO velocities. This is
514: because only the $L^3$ mechanism can form tight binaries that tend to
515: survive if super Hill velocities prevail \citep{SR07}. The fact that
516: retrograde orbits are stable for larger semi-major axes is no longer of
517: importance since only tight binaries tend to survive. This therefore leads to
518: the formation of a roughly equal number of prograde and retrograde binaries
519: for super-Hill KBO velocities.
520:
521: The analysis presented here has also implications for some of the other
522: proposed binary formation scenarios. \citet{W02} suggested that KBBs form by a
523: collision among two KBOs inside the Hill sphere of a third. Although the $L^3$
524: mechanism dominates over such a collisional binary formation scenario we
525: briefly discuss our predictions for this collisional binary formation
526: mechanism. For sub-Hill velocities more retrograde than prograde binaries form
527: because retrograde binary orbits are stable further out in the Hill sphere
528: than prograde ones (i.e. the phase space for forming retrograde binaries is
529: larger than that for prograde binaries). For super-Hill velocities a
530: comparable number of prograde and retrograde binaries form because the fact
531: that retrograde orbits are stable for larger semi-major axes is no longer of
532: importance since only tight binaries are saved from break up. In the formation
533: scenario of \citet{A05} the existence of long lived transient binaries that
534: spend a long time in their mutual Hill sphere, near a periodic orbit, is
535: responsible for the creation of KBBs. \citet{A07} find an excess of prograde
536: over retrograde binaries and suggest that this is a signature of their binary
537: formation process. Our work indicates that an excess of prograde over
538: retrograde binaries might simply be the result of the velocity regime (i.e.$v
539: \sim v_H$) in which the binaries form (see Figure \ref{fig4}).
540:
541: All of the above predictions rely on the assumption that subsequent
542: orbital evolution due to dynamical friction and dynamical stirring of the
543: Kuiper belt did not alter the sense of the binary orbit. The Kuiper Belt has
544: undergone a phase of dynamical excitation which probably modified the orbital
545: properties of KBBs. A detailed study on how dynamical stirring of the Kuiper
546: Belt and dynamical friction affects binary inclinations would be very
547: worthwhile to determine whether they were able to reverse the binary orbit
548: from prograde to retrograde rotation.
549:
550:
551: \acknowledgments
552: Some of the numerical calculations presented here were performed
553: on Caltech's Division of Geological and Planetary Sciences Dell
554: cluster. R. S. is an Alfred P. Sloan Fellow and a Packard Fellow.
555:
556:
557: \begin{thebibliography}{25}
558: \expandafter\ifx\csname natexlab\endcsname\relax\def\natexlab#1{#1}\fi
559:
560: \bibitem[{{Astakhov} {et~al.}(2005){Astakhov}, {Lee}, \& {Farrelly}}]{A05}
561: {Astakhov}, S.~A., {Lee}, E.~A., \& {Farrelly}, D. 2005, \mnras, 360, 401
562:
563: \bibitem[{{Cameron} \& {Ward}(1976)}]{CW76}
564: {Cameron}, A.~G.~W. \& {Ward}, W.~R. 1976, in Lunar and Planetary Institute
565: Conference Abstracts, Vol.~7, Lunar and Planetary Institute Conference
566: Abstracts, 120--+
567:
568: \bibitem[{{Fraser} {et~al.}(2008){Fraser}, {Kavelaars}, {Holman}, {Pritchet},
569: {Gladman}, {Grav}, {Jones}, {MacWilliams}, \& {Petit}}]{F08} {Fraser},
570: W.~C., {Kavelaars}, J., {Holman}, M.~J., {Pritchet}, C.~J., {Gladman},
571: B.~J., {Grav}, T., {Jones}, R.~L., {MacWilliams}, J., \& {Petit},
572: J.~-. 2008, ArXiv Astrophysics e-prints
573:
574: \bibitem[{{Fuentes} \& {Holman}(2008)}]{FH08}
575: {Fuentes}, C.~I. \& {Holman}, M.~J. 2008, ArXiv Astrophysics e-prints
576:
577: \bibitem[{{Funato} {et~al.}(2004){Funato}, {Makino}, {Hut}, {Kokubo}, \&
578: {Kinoshita}}]{F04}
579: {Funato}, Y., {Makino}, J., {Hut}, P., {Kokubo}, E., \& {Kinoshita}, D. 2004,
580: \nat, 427, 518
581:
582: \bibitem[{{Goldreich} {et~al.}(2002){Goldreich}, {Lithwick}, \& {Sari}}]{GLS02}
583: {Goldreich}, P., {Lithwick}, Y., \& {Sari}, R. 2002, \nat, 420, 643
584:
585: \bibitem[{{Goldreich} {et~al.}(2004){Goldreich}, {Lithwick}, \& {Sari}}]{GLS04}
586: ---. 2004, \araa, 42, 549
587:
588: \bibitem[{{Goldreich} \& {Tremaine}(1980)}]{GT80}
589: {Goldreich}, P. \& {Tremaine}, S. 1980, \apj, 241, 425
590:
591: \bibitem[{{Grundy} {et~al.}(2008)}]{GN08}
592: {Grundy}, W.~M., {Noll}, K.~S., {Virtanen}, J., {Muinonen}, K., {Kern}, S.~D.,
593: {Stephens}, D.~C., {Stansberry}, J.~A., {Levison}, H.~F., \& {Spencer},
594: J.~R. 2008, ArXiv Astrophysics e-prints
595:
596: \bibitem[{{Hamilton} \& {Burns}(1991)}]{HB91}
597: {Hamilton}, D.~P. \& {Burns}, J.~A. 1991, Icarus, 92, 118
598:
599: \bibitem[{{Hamilton} \& {Krivov}(1997)}]{HK97}
600: {Hamilton}, D.~P. \& {Krivov}, A.~V. 1997, Icarus, 128, 241
601:
602: \bibitem[{{Hartmann} \& {Davis}(1975)}]{HD75}
603: {Hartmann}, W.~K. \& {Davis}, D.~R. 1975, Icarus, 24, 504
604:
605: \bibitem[{{Hayashi}(1981)}]{H81}
606: {Hayashi}, C. 1981, Progress of Theoretical Physics Supplement, 70, 35
607:
608: \bibitem[{{Henon}(1970)}]{H70}
609: {Henon}, M. 1970, \aap, 9, 24
610:
611: \bibitem[{{Henon} \& {Petit}(1986)}]{HP86}
612: {Henon}, M. \& {Petit}, J.-M. 1986, Celestial Mechanics, 38, 67
613:
614: \bibitem[{{Hill}(1878)}]{H78}
615: {Hill}, G.~W. 1878, Am. J. Math., 1, 5
616:
617: \bibitem[{{Ida}(1990)}]{Ida90}
618: {Ida}, S. 1990, Icarus, 88, 129
619:
620: \bibitem[{{Innanen}(1979)}]{I79}
621: {Innanen}, K.~A. 1979, \aj, 84, 960
622:
623: \bibitem[{{Kenyon} \& {Luu}(1999)}]{KL99}
624: {Kenyon}, S.~J. \& {Luu}, J.~X. 1999, \aj, 118, 1101
625:
626: \bibitem[{{Lee} {et~al.}(2007){Lee}, {Astakhov}, \& {Farrelly}}]{A07}
627: {Lee}, E.~A., {Astakhov}, S.~A., \& {Farrelly}, D. 2007, \mnras, 379, 229
628:
629: \bibitem[{{McKinnon}(1989)}]{MK89}
630: {McKinnon}, W.~B. 1989, \apjl, 344, L41
631:
632: \bibitem[{{Noll} {et~al.}(2007){Noll}, {Grundy}, {Chiang}, {Margot}, \&
633: {Kern}}]{N07}
634: {Noll}, K.~S., {Grundy}, W.~M., {Chiang}, E.~I., {Margot}, J.-L., \& {Kern},
635: S.~D. 2008, The Solar System Beyond Neptune, 345
636:
637: \bibitem[{{Petit} \& {Henon}(1986)}]{PH86}
638: {Petit}, J.-M. \& {Henon}, M. 1986, Icarus, 66, 536
639:
640: \bibitem[{{Petit} {et~al.}(2008)}]{P08}
641: {Petit}, J.-M., {Kavelaares}, J.~J., {Gladman}, B.~J.,\& {Laredo}, T. 2008,
642: The Solar System Beyond Neptune, 71
643:
644: \bibitem[{{Petit} {et~al.}(2008)}]{PK08}
645: {Petit}, J.-M., {Kavelaares}, J., {Gladman}, B., {Margot}, J.-L.,
646: {Nicholson}, P.~D., {Jones}, R.~L., {Parker}, J., {Ashby}, M. 2008, in
647: Asteroids, Comets, Meteors Conference Abstracts, 8354
648:
649: \bibitem[{{Rafikov}(2003)}]{R2003}
650: {Rafikov}, R.~R. 2003, \aj, 125, 942
651:
652: \bibitem[{{Schlichting} \& {Sari}(2008)}]{SR07}
653: {Schlichting}, H.~E. \& {Sari}, R. 2008, \apj, 673, 1218
654:
655: \bibitem[{{Trujillo} \& {Brown}(2003)}]{TB03}
656: {Trujillo}, C.~A. \& {Brown}, M.~E. 2003, Earth Moon and Planets, 92, 99
657:
658: \bibitem[{{Trujillo} {et~al.}(2001){Trujillo}, {Jewitt}, \& {Luu}}]{TJL01}
659: {Trujillo}, C.~A., {Jewitt}, D.~C., \& {Luu}, J.~X. 2001, \aj, 122, 457
660:
661: \bibitem[{{Weidenschilling}(2002)}]{W02}
662: {Weidenschilling}, S.~J. 2002, Icarus, 160, 212
663:
664: \bibitem[{{Wetherill} \& {Stewart}(1993)}]{WS1993}
665: {Wetherill}, G.~W. \& {Stewart}, G.~R. 1993, Icarus, 106, 190
666:
667: \bibitem[{{Zhang} \& {Innanen}(1988)}]{ZI88}
668: {Zhang}, S.-P. \& {Innanen}, K.~A. 1988, Icarus, 75, 105
669:
670: \end{thebibliography}
671:
672:
673: \clearpage
674:
675: \begin{figure}[htp]
676: \centerline{
677: \includegraphics[ scale=1.6]{f1.eps}
678: }
679: \caption{Histogram of modified Jacobi constants, $J_C$, of prograde and
680: retrograde KBBs that formed via three body gravitational deflection, $L^3$
681: mechanism, for $v \ll v_H$. Each histogram is normalized to unity, but
682: overall retrograde orbits are twice as abundant as prograde orbits. Note,
683: prograde binaries exist only for $J_C \gtrsim 9$ whereas retrograde binaries
684: exist also for $J_C \lesssim 9$.}
685: \label{fig1}
686: \end{figure}
687:
688: \begin{figure} [htp]
689: \centerline{\includegraphics[ scale=1.2]{f2.eps}}
690: \caption{Three examples of KBO encounters in the $L^2s$ mechanism for $v \ll
691: v_H$ and $D=0.01$ that result in the formation of a binary. The plots on the
692: left and right hand side show the evolution of the specific angular
693: momentum, $h$, and modified Jacobi constant, $J_C$, as a function of time
694: respectively. The time $t=0$ corresponds to the time at which $y=0$ if the
695: relative KBO velocity is solely due to the Keplerian sheer (i.e. ignoring
696: the actual gravitational interaction between the bodies). The evolution of
697: $h$ and $J_C$ is shown until the binary separation has decreased to $0.1
698: R_H$ or less. These examples show that the sense of rotation is practically
699: preserved. $h$ displays large variations right after capture caused by solar
700: tides. The most extreme case of angular momentum sign change for bodies that
701: form binaries found in our simulations is displayed in the second of the
702: three examples. The angular momenta of the binaries are all negative
703: corresponding to retrograde binary orbits. In fact all binaries that form
704: via the $L^2s$ mechanism in our numerical integrations display retrograde
705: orbits. Dynamical friction shrinks the binary separation leading to a
706: decrease in the magnitude of $h$ and an increase of $J_C$ with time. The
707: modified Jacobi constant of the newly formed binaries is smaller than
708: $J_C(L1)=9$ which explains why all their orbits are retrograde (see \S
709: 3.1.1. for details).}
710: \label{fig2}
711: \end{figure}
712:
713: \begin{figure} [htp]
714: \centerline{\includegraphics[ scale=1.2]{f3.eps}}
715: \caption{Same as in Figure \ref{fig3} but for two examples of KBO encounters
716: in the $L^2s$ mechanism for $v \ll v_H$ and $D=0.01$ that do not result in
717: the formation of a binary. As a result of the Keplerian sheer, KBOs
718: encounter and leave each other with positive $h$.}
719: \label{fig3}
720: \end{figure}
721:
722: \begin{figure} [htp]
723: \plotone{f4.eps}
724: \caption{The ratio of retrograde binaries, $N_{ret}$, with a modified Jacobi
725: constant of $J_C^{min}$ or larger to the total number of binaries, $N_{total}$,
726: that formed via the $L^3$ mechanism for $v \ll v_H$. For small $J_C^{min}$,
727: i.e. when all binaries are included, about $2/3$ have retrograde orbits. More
728: retrograde than prograde binaries form because retrograde binary orbits are
729: stable further out in the Hill sphere than prograde ones. As $J_C^{min}$
730: increases the fraction of retrograde binaries decreases reaching a minimum of
731: about $1/3$ for $J_C^{min} \sim9$. This may be due to the Keplerian sheer
732: which increases the duration of a prograde encounter between unbound KBOs
733: compared to a retrograde encounter. The fraction of prograde and retrograde
734: binaries becomes comparable for $J_C^{min} \gg 9$ because for such binaries
735: neither the Keplerian sheer nor the increased stability of retrograde orbits
736: are important.}
737: \label{fig4}
738: \end{figure}
739:
740:
741: \end{document}
742: