1: % For ApJ submission -----------------------------------------

2: %\documentclass[12pt,preprint]{aastex}

3: %\def\baselinestretch{1.3}

4: 

5: \documentclass{emulateapj}

6: \usepackage{apjfonts}

7: \lefthead{KIM ET AL.}

8: \righthead{LIMITS OF MICROLENING BINARIES}

9: 

10: 

11: %==== CUSTOMIZED LATEX MACROS ========================================

12: 

13: \newcommand{\vect}[1]{\ensuremath{\mbox{\boldmath $#1$}}}

14: 

15: 

16: % Equation align

17: \def\eqalign#1{\null\,\vcenter{\openup\jot

18:         \ialign{\strut\hfil$\displaystyle{##}$&$

19:         \displaystyle{{}##}$\hfil \crcr#1\crcr}}\,}

20: 

21: %=======================================================================

22: 

23: %\lefthead{HAN} 

24: %\righthead{SEVERE BINARY DEGENERACY}

25: 

26: \begin{document}

27: \title{Limits of Binaries That Can Be Characterized by Gravitational Microlensing}

28: 

29: 

30: \author{

31: Doeon Kim\altaffilmark{1},

32: Yoon-Hyun Ryu\altaffilmark{2},

33: Byeong-Gon Park\altaffilmark{3},

34: Heon-Young Chang\altaffilmark{2},

35: Kyu-Ha Hwang\altaffilmark{1},\\

36: Sun-Ju Chung\altaffilmark{3},

37: Chung-Uk Lee\altaffilmark{3},

38: and

39: Cheongho Han\altaffilmark{1,4}\\

40: }

41: 

42: \altaffiltext{1}{Department of Physics, Institute for Basic Science

43: Research, Chungbuk National University, Chongju 361-763, Korea}

44: \altaffiltext{2}{Department of Astronomy and Atmospheric Sciences, 

45: Kyungpook National University, Daegu 702-701, Korea}

46: \altaffiltext{3}{Korea Astronomy and Space Science Institute, Hwaam-Dong,

47: Yuseong-Gu, Daejeon 305-348, Korea}

48: \altaffiltext{4}{corresponding author}

49: 

50: 

51: \submitted{Submitted to The Astrophysical Journal}

52: 

53: \begin{abstract}

54: Due to the high efficiency of planet detections, current microlensing 

55: planet searches focus on high-magnification events. High-magnification 

56: events are sensitive to remote binary companions as well and thus a 

57: sample of wide-separation binaries are expected to be collected as a 

58: byproduct.  In this paper, we show that characterizing binaries for 

59: a portion of this sample will be difficult due to the degeneracy 

60: of the binary-lensing parameters.  This degeneracy arises because the 

61: perturbation induced by the binary companion is well approximated by 

62: the Chang-Refsdal lensing for binaries with separations greater than 

63: a certain limit.  For binaries composed of equal mass lenses, we find 

64: that the lens binarity can be noticed up to the separations of $\sim 60$ 

65: times of the Einstein radius corresponding to the mass of each lens.

66: Among these binaries, however, we find that the lensing parameters can 

67: be determined only for a portion of binaries with separations less than 

68: $\sim 20$ times of the Einstein radius.

69: \end{abstract}

70: 

71: \keywords{gravitational lensing}

72: 

73: % ==================================================================

74: 

75: 

76: \section{Introduction}

77: 

78: 

79: 

80: Searches for extrasolar planets by using microlensing are being 

81: carried out by observing stars located toward the Galactic bulge 

82: field.  The lensing signal of a planet is a short-duration perturbation 

83: to the smooth standard light curve of the primary-induced lensing 

84: event occurring on a background star \citep{mao91, gould92}.  For 

85: the detections of the short-duration planetary lensing signals, 

86: these searches are using a combination of survey and follow-up 

87: observations, where the survey observations (e.g., OGLE, 

88: \citet{udalski03}; MOA, \citet{bond02}) aim to maximize the 

89: lensing event rate by monitoring a large area of sky and the 

90: follow-up observations (e.g., PLANET, \citet{albrow01}; MicroFUN, 

91: \citet{dong06}) intensively monitor the events alerted by the survey 

92: observations.  However, the limited number of telescopes restricts 

93: the number of events that can be followed at any given time and thus 

94: priority is given to those events that will maximize the planetary 

95: detection probability.  Currently, the highest priority is given to 

96: high-magnification events because the source trajectories of these 

97: events always pass close to the perturbation region around the 

98: planet-induced caustic located near the primary lens \citep{griest98}.  

99: In addition, follow-up observations can be prepared for these events 

100: because the perturbation typically occurs near the peak of the event, 

101: which can be predicted from the data on the rising part of the event 

102: light curve.  As a result, six (OGLE-2005-BLG-071Lb, OGLE-2005-BLG-169Lb, 

103: OGLE-2006-BLG-109Lb,c, MOA-2007-BLG-400b, MOA-2007-BLG-192) of the 

104: eight reported microlensing planets were detected through the channel 

105: of high-magnification events \citep{bond04, udalski05, beaulieu06, 

106: gould06, gaudi08, dong08, bennett08}.

107: 

108: 

109: In addition to planets, high-magnification events are sensitive to 

110: wide-separation binary companions as well.  Similar to the planetary 

111: case, the companion of a wide-separation binary induces a small 

112: caustic close to the primary lens and thus can produce a short-duration 

113: perturbation near the peak of a high-magnification event.  Due to the 

114: different nature of the companions, however, the perturbations induced 

115: by a wide-separation binary companion and a planet can be distinguished

116: \citep{albrow02, han08}. Therefore, under the current planetary lensing 

117: strategy focusing on high-magnification events, a considerable number 

118: of wide-separation binaries are expected to be detected as a byproduct 

119: and this sample might provide useful information about the physical 

120: distribution of binaries. 

121: 

122: 

123: Unlike this expectation, however, we find that characterizing binaries 

124: for a significant portion of the binary lens sample will be difficult 

125: due to the degeneracy of the binary-lensing parameters.  This degeneracy 

126: arises because the perturbation induced by the binary companion is 

127: well approximated by the Chang-Refsdal lensing (hereafter C-R lensing) 

128: for binaries with separations greater than a certain value.  The C-R 

129: lensing represents single-mass lensing superposed on a uniform 

130: background shear.  For a wide-separation binary, the shear results 

131: from the combination of the binary-lens parameters and thus the 

132: individual parameters cannot be separately determined. 

133: 

134: 

135: The paper is organized as follows.  In \S\ 2, we briefly describe 

136: the properties of binary and C-R lensing. In \S\ 3, we demonstrate 

137: the proximity between the binary and C-R lensing for binaries with 

138: separations beyond a certain limit.  We then set the range in the 

139: binary-lensing parameter space where the degeneracy of binary-lensing

140: parameters occurs.  We discuss the meaning of this degeneracy in the 

141: studies of binaries by using microlensing. We summarize the results 

142: and conclude in \S\ 4.

143: 

144: 

145: 

146: 

147: 

148: \section{Lensing Properties}

149: 

150: If a source is lensed by a single point mass, the lens mapping 

151: from the lens plane to the source plane is expressed by the lens 

152: equation

153: \begin{equation}

154: \zeta = z - {1\over \bar{z}},

155: \label{eq1}

156: \end{equation}

157: where $\zeta=\xi+i\eta$ and $z=x+iy$ are the locations of the 

158: source and the lens in complex notations, respectively, and 

159: $\bar{z}$ denotes the complex conjugates of $z$.  Here all 

160: angles are normalized by the angular Einstein radius of the 

161: lens, which is related to the physical parameters of the lens by

162: \begin{equation}

163: \theta_{\rm E}=\left({4GM\over c^2}\right)^{1/2}

164: \left({1\over D_L}-{1\over D_S}\right)^{1/2},

165: \label{eq2}

166: \end{equation}

167: where $M$ is the mass of the lens and $D_L$ and $D_S$ are the 

168: distances to the lens and source, respectively.  Solving the 

169: lens equation results in two solutions of the image position.  

170: The lensing process conserves the source surface brightness and 

171: thus the magnifications of the individual images correspond to 

172: the area ratios between the images and source.  Mathematically, 

173: this is obtained by solving the Jacobian of the mapping equation 

174: evaluated at the image position, i.e., 

175: \begin{equation}

176: A_i=\left\vert 

177: \left( 

178: 1-{\partial\zeta\over \partial\bar{z}}

179: {\overline{\partial\zeta}\over \partial\bar{z}}\right)_{z=z_i}^{-1}

180: \right\vert.

181: \label{eq3}

182: \end{equation}

183: Then, the total magnification corresponds to the sum of the individual 

184: images, i.e., $A=\sum_i A_i$, and this results in 

185: \begin{equation}

186: A={u^2+2\over u(u^2+4)^{1/2}},

187: \label{eq4}

188: \end{equation}

189: where $u=|\zeta|$ represents the normalized lens-source separation.

190: For a rectilinear motion, the lens-source separation is related to 

191: the lensing parameters of the Einstein time scale $t_{\rm E}$, the 

192: closest lens-source separation $u_0$, and the time at the moment $t_0$ 

193: by 

194: \begin{equation}

195: u=\left[ u_0^2+\left( {t-t_0\over t_{\rm E}}\right)^2\right]^{1/2}. 

196: \label{eq5}

197: \end{equation}

198: The light curve of a single-lensing event is characterized by a 

199: smooth and symmetric shape \citep{paczynski86}.

200: 

201: 

202: If a source is lensed by a binary lens, on the other hand, the 

203: mapping equation is expressed as \citep{witt90}

204: \begin{equation}

205: \zeta = z - {m_1/M \over \bar{z}-\bar{z}_{L,1}} 

206: - {m_2/M \over \bar{z}-\bar{z}_{L,2}},

207: \label{eq6}

208: \end{equation}

209: where $z_{L,1}=x_{L,1}+iy_{L,1}$ and $z_{L,2}=x_{L,2}+iy_{L,2}$ 

210: are the positions of the lens components, $m_1$ and $m_2$ are 

211: their masses, and $M=m_1+m_2$ is the total mass of the binary.  

212: One important characteristic of binary lensing that differentiates 

213: from single lensing is the formation of caustics, which represent 

214: source positions at which the magnification of a point source becomes 

215: infinite. The set of caustics form closed curves, each of which is 

216: composed of concave curves (fold caustic) that meet at points (cusps).  

217: Binary lenses can have one, two, or three closed caustic curves. If 

218: the two masses are separated by approximately an Einstein radius, 

219: then there is a single six-cusp caustic.  If the masses are much 

220: closer than an Einstein ring, there is a central four-cusp caustic 

221: and two outlying three-cusp caustics.  If they are separated by much 

222: more than an Einstein ring (wide-separation binary), then there are 

223: two four-cusp caustics, where each of the caustics is associated with 

224: each member of the binary.  In addition to the parameters of single 

225: lensing, modelling a binary-lensing event requires two additional 

226: lensing parameters of the separation $s=|z_{L,1}-z_{L,2}|$ and the 

227: mass ratio $q=m_2/m_1$ between the lens components.

228: 

229: 

230: The Chang-Refsdal lensing represents single lensing superposed 

231: on a uniform background shear \citep{chang79, chang84}.  The lens 

232: equation for the C-R lensing is represented by 

233: \begin{equation}

234: \zeta=z-{1\over\bar{z}} + \gamma\bar{z},

235: \label{eq7}

236: \end{equation}

237: where $\gamma$ is the shear.  The shear induces a single set of 

238: caustics, which form around the lens.  The caustic has a shape 

239: of a hypocycloid with four cusps.  Its size as measured by the 

240: separation between two confronting cusps are related to the 

241: shear by

242: \begin{equation}

243: \Delta\zeta_c =4\gamma. 

244: \label{eq8}

245: \end{equation}

246: 

247: 

248: In the limiting case of a binary lens where the binary separation 

249: is much larger than the Einstein radius ($s\gg 1$), the lensing 

250: properties in the vicinity of each of the binary lenses is well 

251: described by that of the C-R lensing \citep{dominik99}.  The shear 

252: exerted by the companion is related to the binary parameters by

253: \begin{equation}

254: \gamma \sim {q\over \hat{s}^2};\qquad \hat{s}=(1+q)^{1/2}s,

255: \label{eq9}

256: \end{equation}

257: where $\hat{s}$ is the binary separation normalized by the Einstein 

258: radius corresponding to the mass of the lens close to which the 

259: caustic is located.  The shear decreases as $1/\hat{s}^2$ and 

260: thus the caustics of a wide-separation binary lens shrink rapidly 

261: as the binary separation increases.  In the limiting case where 

262: $\hat{s} \rightarrow \infty$, the shear and the caustics vanish 

263: and the individual lens components behave as if they are two 

264: independent single lenses.  The positions of the lens components 

265: effectively working as single lenses, effective lens position 

266: $z_{L,eff}$, are slightly shifted from their original positions 

267: with an offset 

268: \citep{distefano96}

269: \begin{equation}

270: \Delta z_L = z_L - z_{L,eff}

271: \sim -{q\over \hat{s}} {z_{L,2}-z_{L,1} \over |z_{L,2}-z_{L,1}|}.

272: \label{eq10}

273: \end{equation}

274: 

275: 

276: % Figure 1 ----------------------------------------------------

277: \begin{figure*}[t]

278: \epsscale{1.0}

279: %\epsscale{0.8}

280: \plotone{fig1.eps}

281: \caption{\label{fig:one}

282: Maps of the fractional magnification residual from that of 

283: single lensing as a function of the source position in the 

284: central region of a component of binary lenses with various 

285: separations and mass ratios.  All lengths are scaled by the 

286: Einstein radius corresponding to the mass of the binary 

287: component around which the map is constructed (primary).  

288: The coordinates are centered at the effective position of 

289: the primary lens (see the definition of the `effective lens 

290: position' in the text of \S\ 2).  The width of each map 

291: corresponds to 8 times of the width of the caustic.  In each 

292: map, the regions with blue and brown-tone colors represent 

293: the areas where the binary-lensing magnification is lower 

294: and higher than the single-lensing magnification, respectively.  

295: For each tone, the color changes into darker scales when the 

296: residual is $|\epsilon| \geq 1\%$, 2\%, 5\%, and 10\%, 

297: respectively.  The straight lines with arrows represent the 

298: source trajectories and the light curves of the resulting 

299: events are presented in the corresponding panels of 

300: Fig.~\ref{fig:three}.  We note that maps are not presented 

301: if the perturbation from single lensing is severely washed 

302: out by the finite-source effect and thus the magnification 

303: pattern of a binary-lensing case is difficult to be 

304: distinguished from that of a single-lensing case.

305: }\end{figure*}

306: % -------------------------------------------------------------

307: 

308: 

309: % Figure 2 ----------------------------------------------------

310: \begin{figure*}[t]

311: \epsscale{1.0}

312: %\epsscale{0.8}

313: \plotone{fig2.eps}

314: \caption{\label{fig:two}

315: Same as Fig.~\ref{fig:one} except that the magnification residual 

316: is from that of C-R lensing.  The panels blocked by thick lines 

317: represent the cases where the binary signal can be detected, 

318: but characterization of the binary is difficult due to the 

319: degeneracy of the binary-lensing parameters. 

320: }\end{figure*}

321: % -------------------------------------------------------------

322: 

323: 

324: 

325: % Figure 3 ----------------------------------------------------

326: \begin{figure*}[t]

327: \epsscale{1.0}

328: %\epsscale{0.8}

329: \plotone{fig3.eps}

330: \caption{\label{fig:three}

331: Example light curves of high-magnification events caused by 

332: wide-separation binaries.  The source trajectories responsible 

333: for the individual events are marked in the corresponding panels 

334: of Fig.~\ref{fig:one}.  In the upper part of each panel, there 

335: are three light curves: blue curve for single lensing, red curve 

336: for C-R lensing, and black curve for binary lensing.  The lower 

337: part shows the fractional residuals from the single lensing 

338: magnification (blue curve) and the C-R lensing approximation 

339: (red curve).

340: }\end{figure*}

341: % -------------------------------------------------------------

342: 

343: 

344: 

345: \section{Degeneracy}

346: 

347: 

348: The fact that the lensing behavior of a wide-separation binary in 

349: the region close to each lens component is well described by the 

350: C-R lensing implies that although lens binarity can be noticed 

351: from the perturbation to the light curve of a high-magnification 

352: event, it may be difficult to determine the binary-lensing parameters 

353: from the analysis of the perturbation.  This is because the C-R 

354: lensing is described by a single parameter of $\gamma$, which 

355: results from the combination of the two binary-lensing parameters 

356: of $s$ and $q$.

357: 

358: 

359: Then, naturally rising questions are (1) how serious this degeneracy

360: is and (2) in which region of the binary-lensing parameter space 

361: this degeneracy occurs.  To answer these questions, we construct 

362: two sets of perturbation pattern maps as a function of the source 

363: position in the central region of a component of binary lenses 

364: with various separations and mass ratios.  In the first set, we 

365: construct the maps of the fractional magnification residual from 

366: that of single lensing, i.e., 

367: \begin{equation}

368: \epsilon_s = {A-A_s\over A_s},

369: \label{eq11}

370: \end{equation}

371: where $A$ and $A_s$ are the magnifications of binary and single 

372: lensing without the companion, respectively.  In the second set, 

373: we construct the maps of the residual from that of C-R lensing, i.e., 

374: \begin{equation}

375: \epsilon_{\rm C-R} = {A-A_{\rm C-R}\over A_{\rm C-R}},

376: \label{eq12}

377: \end{equation}

378: where $A_{\rm C-R}$ is the magnification obtained from C-R 

379: lensing by approximating the shear exerted by the companion as 

380: $\gamma=q/\hat{s}^2$.

381: 

382: 

383: Figure~\ref{fig:one} and \ref{fig:two} show the two sets of maps. 

384: In each map, the coordinates are centered at the effective position 

385: of a binary component.  The width of each map is set such that it 

386: corresponds to 8 times of the width of the C-R lensing caustic 

387: computed by equations (\ref{eq8}) and (\ref{eq9}).  The separation 

388: and the mass ratio of the binary for each map are marked at the 

389: top and on the right side of the panel, respectively.  The mass 

390: ratio is such that $q<1.0$ when the companion is less massive 

391: than the lens component around which the map is constructed and 

392: vice versa.  In each map, the regions with blue and brown-tone 

393: colors represent the areas where the binary-lensing magnification 

394: is lower and higher than the single or C-R lensing magnification, 

395: respectively.  For each tone, the color changes into darker scales 

396: when the residual is $|\epsilon| \geq 1\%$, 2\%, 5\%, and 10\%, 

397: respectively. Figure~\ref{fig:three} shows light curves of events 

398: resulting from the source trajectories marked in Figure~\ref{fig:one}.  

399: In the upper part of each panel, there are three curves: blue curve 

400: for single lensing, red curve for C-R lensing, and black curve for 

401: binary lensing.  The lower part shows the fractional residuals from 

402: the single-lensing magnification (blue curve) and the C-R lensing 

403: approximation (red curve).

404: 

405: 

406: The caustic induced by a wide-separation binary is small and 

407: thus perturbations induced by the caustic are vulnerable to 

408: finite-source effect.  The finite-source effect is parameterized 

409: by the ratio of the source radius $r_\star$ to the Einstein 

410: radius.  For a lensing event toward the Galactic bulge field, 

411: this ratio is scaled by the physical parameters of the lens by

412: \begin{equation}

413: \rho_\star = 9\times 10^{-4} 

414: \left( {r_\star\over R_\odot}\right)

415: \left[

416: \left( {m_1\over 0.3\ M_\odot}\right)

417: \left( {D_L\over 6\ {\rm kpc}}\right)

418: \left( 1-{D_L\over D_S}\right)\right]^{-1/2}.

419: \label{eq13}

420: \end{equation}

421: Then, the magnification affected by the finite-source effect

422: becomes 

423: \begin{equation}

424: A={\int_0^{\rho_\star} I(r)A_p(|{\bf r}-{\bf r}_L|)r dr

425: \over\int_0^{\rho_\star} I(r)r dr},

426: \label{eq13}

427: \end{equation}

428: where ${\bf r}_L$ is the displacement vector of the source center 

429: with respect to the lens, ${\bf r}$ is the vector to a position 

430: on the source star surface with respect to the center of the 

431: source star, $I(r)$ represents the source brightness profile, 

432: and $A_p$ is the point-source magnification.  We consider the 

433: finite-source effect by assuming that the source star has a 

434: uniform disk with a radius equivalent to the Sun and the physical 

435: parameters of the lens system are $m_1=0.3\ M_\odot$, $D_L=6$ kpc, 

436: and $D_S=8$ kpc by adopting the values of a typical event being 

437: detected toward Galactic bulge field.  This results in $\rho_\star 

438: = 1.8\times 10^{-3}$.  In Figure \ref{fig:one} -- \ref{fig:three}, 

439: we do not present maps and light curves if the perturbation from 

440: single lensing is severely washed out by the finite-source 

441: effect and thus the magnification pattern of a binary-lensing 

442: case is difficult to be distinguished from that of a single-lensing 

443: case.

444: 

445: 

446: The residual map from single lensing shows that although the 

447: perturbation becomes weaker as the distance to the companion 

448: increases, it lasts for a considerable distance.  The maximum 

449: distance for the detection of the companion signal depends on 

450: the companion/primary mass ratio and the observational precision.  

451: By adopting a threshold residual of $\epsilon_s=5\%$, we find 

452: that companions can be detected up to the separations of 

453: $\hat{s}\sim 31$, 50, 63, and 100 for binary companions with 

454: mass ratios of $q=1/5$, 1/2, 1.0, and 2.0, respectively.  

455: Considering that the physical radius of the Einstein ring is 

456: $r_{\rm E}\sim 1.9$, AU, these limits correspond to the physical 

457: separations of $d\sim 59$, 95, 120, and 190 AU, respectively, 

458: implying that high-magnification events provide an efficient 

459: channel to detect wide-separation binaries.

460: 

461: 

462: However, the condition for detection is different from the 

463: condition for characterization.  The condition for binary 

464: characterization depends on various factors such as real 

465: experimental data, photometric errors, sampling rate, 

466: observation duration, and telescopes performance.  Therefore, 

467: it is not easy to define a single observable that can be 

468: used for an indicator to judge the characterizability of 

469: binary-lens events.  However, if a light curve significantly 

470: deviates from C-R lensing approximation, it would be 

471: possible to determine the binary-lensing parameters.  The 

472: photometric error achieved by the current follow-up observations 

473: reaches down to $\sim 1-2\%$ level for high-magnification events 

474: (Gould 2008, private communication). We therefore set the 

475: condition for binary characterization such that light curves

476: should be distinguished from the C-R lensing approximation with 

477: $\epsilon_{\rm C-R}\geq 5\%$.  With this criterion, we find 

478: that the upper limits of binary separation for characterization 

479: are $\hat{s}\sim 13$, 20, 22, and 25 for binary companions with 

480: mass ratios of $q=1/5$, 1/2, 1.0, and 2.0, respectively.  From 

481: the comparison of the limits for detection, the ranges of binary 

482: separation for characterization are substantially narrower than 

483: the ranges for detection. This implies that for a significant 

484: fraction of events for which signals of wide-separation binary 

485: companions are detected, it will be difficult to characterize 

486: the properties of the binary by determining the lensing parameters.  

487: In Figure~\ref{fig:two}, we mark this region of degeneracy by 

488: blocking the panels with thick solid lines.  We also list the 

489: detection and degeneracy regions in Table~\ref{table:one}.

490: 

491: 

492: 

493: 

494: \begin{deluxetable}{lll}

495: \tablecaption{Detection and Degeneracy Zones\label{table:one}}

496: \tablewidth{0pt}

497: \tablehead{

498: \multicolumn{1}{c}{mass } &

499: \multicolumn{1}{c}{detection} &

500: \multicolumn{1}{c}{degeneracy}\\

501: \multicolumn{1}{c}{ratio} &

502: \multicolumn{1}{c}{zone} &

503: \multicolumn{1}{c}{zone}

504: }

505: \startdata

506: $q=1/10$ &  $s \lesssim 24$  ($d \lesssim 46\ {\rm AU}$)  &  $11 \lesssim s \lesssim 24$  ($21\ \lesssim d \lesssim 46\ {\rm AU}$)  \\

507: $q=1/5$  &  $s \lesssim 31$  ($d \lesssim 59\ {\rm AU}$)  &  $13 \lesssim s \lesssim 31$  ($25\ \lesssim d \lesssim 59\ {\rm AU}$)  \\

508: $q=1/3$  &  $s \lesssim 40$  ($d \lesssim 76\ {\rm AU}$)  &  $16 \lesssim s \lesssim 40$  ($20\ \lesssim d \lesssim 76\ {\rm AU}$)  \\

509: $q=1/2$  &  $s \lesssim 50$  ($d \lesssim 95\ {\rm AU}$)  &  $20 \lesssim s \lesssim 50$  ($38\ \lesssim d \lesssim 95\ {\rm AU}$)  \\

510: $q=1.0$  &  $s \lesssim 63$  ($d \lesssim 120\ {\rm AU}$) &  $22 \lesssim s \lesssim 63$  ($41\ \lesssim d \lesssim 120\ {\rm AU}$) \\ 

511: $q=2.0$  &  $s \lesssim 100$ ($d \lesssim 190\ {\rm AU}$) &  $25 \lesssim s \lesssim 100$ ($48\ \lesssim d \lesssim 190\ {\rm AU}$)

512: \enddata

513: \tablecomments{

514: Detection and degeneracy zones of wide-separation binaries.

515: The detection zone represents the range of binary separation

516: where the lensing signature of a companion can be detected.

517: The degeneracy zone indicates the range where the companion 

518: can be detected but the characterization of the binary is 

519: difficult due to the degeneracy of the binary-lensing parameters. 

520: The absolute physical ranges are given assuming a microlensing 

521: system with $D_S=8\ {\rm kpc}$, $D_L=6\ {\rm kpc}$, and $m_i=

522: 0.3\ M_\odot$.

523: }

524: \end{deluxetable}

525: 

526: 

527: 

528: \section{Conclusion}

529: 

530: By investigating the patterns of central perturbations produced 

531: by wide-separation binary companions, we found that the perturbation 

532: can be noticed and thus lens binarity can be identified for 

533: binaries with considerable separations.  However, we also found 

534: that perturbations for a significant portion of these binaries 

535: can be well approximated by the C-R lensing, implying that the 

536: individual binary-lensing parameters cannot be separately determined 

537: and thus characterization of the binary is difficult.  We set the 

538: range in the binary-lensing parameter space where this degeneracy 

539: occurs.  From this, we found that the lens binarity can be noticed 

540: up to the separations of $\sim 60$ times of the Einstein radius 

541: corresponding to the mass of each lens of a binary composed of 

542: equal mass lenses.  Among these binaries, however, we found that 

543: the lensing parameters can be determined only for a portion of 

544: binaries with separations less than $\sim 20$ times of the Einstein 

545: radius.

546: 

547: 

548: 

549: 

550: \acknowledgments 

551: C.\ H. acknowledge support from the Astrophysical Research Center 

552: for the Structure and Evolution of the Cosmos (ARCSEC) of the 

553: Korea Science and Engineering Foundation (KOSEF) through the 

554: Science Research Program (SRC).  H.-Y.\ C. and B-G.\ P. were 

555: supported by the grant (KRF-2006-311-C00072) from Korea Research 

556: Foundation.

557: 

558: 

559: 

560: 

561: \begin{thebibliography}{99}

562: \frenchspacing

563: 

564: 

565: \bibitem[Albrow et al.(2001)]{albrow01}

566: Albrow, M.\ D., et al.\ 2001, \apj, 556, L113

567: 

568: \bibitem[Albrow et al.(2002)]{albrow02}

569: Albrow, M.\ D., et al.\ 2002, \apj, 572, 1031

570: 

571: \bibitem[Beaulieu et al.(2006)]{beaulieu06}

572: Beaulieu, J.\ P., et al.\ 2006, Nature, 439, 437

573: 

574: \bibitem[Bennett et al.(2008)]{bennett08}

575: Bennett, D.\ P., et al.\ 2008, \apj, submitted

576: 

577: \bibitem[Bond et al.(2002a)]{bond02}

578: Bond, I.\ A., et al.\ 2002, \mnras, 331, L19

579: 

580: \bibitem[Bond et al.(2004)]{bond04}

581: Bond, I.\ A., et al.\ 2004, \apj, 606, L155

582: 

583: \bibitem[Chang \& Refsdal(1979)]{chang79}

584: Chang, K., \& Refsdal, S.\ 1979, 282, 561

585: 

586: \bibitem[Chang \& Refsdal(1984)]{chang84}

587: Chang, K., \& Refsdal, S.\ 1984, 132, 168

588: 

589: \bibitem[Dominik(1999)]{dominik99}

590: Dominik, M.\ 1999, \aap, 349, 108

591: 

592: \bibitem[Di Stefano \& Mao(1996)]{distefano96}

593: Di Stefano, R., \& Mao, S.\ 1996, \apj, 457, 93

594: 

595: \bibitem[Dong et al.(2006)]{dong06}

596: Dong, S., et al.\ 2006, \apj, 642, 842

597: 

598: \bibitem[Dong et al.(2008)]{dong08}

599: Dong, S., et al.\ 2008, \apj, submitted

600: 

601: \bibitem[Gaudi et al.(2008)]{gaudi08}

602: Gaudi, B.\ S., et al.\ 2008, Science, 319, 927

603: 

604: \bibitem[Gould et al.(2006)]{gould06}

605: Gould, A., et al.\ 2006, \apj, 644, L37

606: 

607: \bibitem[Gould \& Loeb(1992)]{gould92}

608: Gould, A., \& Loeb, A.\ 1992, \apj, 396, 104

609: 

610: \bibitem[Griest \& Safizadeh(1998)]{griest98}

611: Griest, K., \& Safizadeh, N.\ 1998, \apj, 500, 37

612: 

613: \bibitem[Han \& Gaudi(2008)]{han08}

614: Han, C., \& Gaudi, B.\ S.\ 2008, \apj, in press

615: 

616: \bibitem[Mao \& Paczy\'nski(1991)]{mao91}

617: Mao, S., \& Paczy\'nski, B.\ 1991, \apj, 374, L37

618: 

619: \bibitem[Paczy\'nski(1986)]{paczynski86}

620: Paczy\'nski, B.\ 1986, \apj, 304, 1

621: 

622: \bibitem[Udalski(2003)]{udalski03}

623: Udalski, A.\ 2003, Acta Astron., 53, 291

624: 

625: \bibitem[Udalski et al.(2005)]{udalski05}

626: Udalski, A., et al.\ 2005, \apj, 628, L109

627: 

628: \bibitem[Witt(1990)]{witt90}

629: Witt, H.\ J.\ 1990, \aap, 236, 311

630: 

631: \end{thebibliography}

632: 

633: \end{document}

634: