0809.3747/ms.tex
1: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
2: %
3: %  V1388 Ori (HD 42401)
4: %
5: %
6: %  Revision history:
7: %  June, 2008 -- first draft began by Stephen J. Williams
8: %
9: %  Sept. 17, 2008 -- Final revision before submission
10: %  
11: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
12: %
13: 
14: %Single column, submission format:
15: \documentclass[12pt,preprint]{aastex}
16: 
17: \usepackage{epsfig}
18: \usepackage{natbib}                % To format bibliographies.
19: 
20: \citestyle{aa}
21: 
22:         \shorttitle{HD 42401}
23:         \shortauthors{Williams}
24: 
25: \begin{document}
26: 
27: %%\received{}
28: %%\accepted{}
29: 
30: \title{System Parameters for the Eclipsing B-star Binary HD 42401}
31: 
32: \author{S. J. Williams\footnote{Visiting Astronomer, Kitt Peak National 
33: Observatory, National Optical Astronomy Observatory, which is operated by the 
34: Association of Universities for Research in Astronomy (AURA) under cooperative 
35: agreement with the National Science Foundation.}}
36: \affil{Center for High Angular Resolution Astronomy and \\
37:  Department of Physics and Astronomy,\\
38:  Georgia State University, P. O. Box 4106, Atlanta, GA 30302-4106; \\
39:  swilliams@chara.gsu.edu}
40: 
41: \begin{abstract}
42: 
43: I present results from an optical spectroscopic investigation of the
44: binary system HD~42401 (V1388 Ori; B2.5 IV-V + B3 V). A combined analysis of 
45: $V-$band photometry and radial velocities indicates that the system has an 
46: orbital period of 2.18706 $\pm$ 0.00005 days and an inclination of 75.5 $\pm$
47: 0.2 degrees. This solution yields masses and radii of $M_{\rm 1}$ = 7.42 $\pm$
48: 0.08 $M_{\odot}$ and $R_{\rm 1}$ = 5.60 $\pm$ 0.04 $R_{\odot}$ for the primary
49: and $M_{\rm 2}$ = 5.16 $\pm$ 0.03 $M_{\odot}$ and $R_{\rm 2}$ = 3.76 $\pm$ 0.03
50: $R_{\odot}$ for the secondary. Based on the position of the two stars plotted
51: on a theoretical H-R diagram, I find that the age of the system is 
52: $\gtrsim$~25 
53: Myr and that both stars appear overluminous for their masses compared to
54: single star evolutionary tracks. A fit of the spectral energy distribution
55: based on photometry from the literature yields a distance to HD~42401 of 832
56: $\pm$ 89 parsecs.
57: 
58: \end{abstract}
59: \keywords{binaries: spectroscopic -- stars: early-type -- stars: fundamental
60: parameters -- stars: individual (HD 42401)}
61: 
62: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
63: 
64: \section{Introduction}
65: 
66: Two fundamental parameters in stellar astrophysics are the masses and radii
67: of stars. Especially lacking are accurate ($\le2\%$) determinations of 
68: these quantities for O- and B-type systems. Data of this kind are important
69: for tests of stellar models, with far-reaching implications such as the 
70: modeling of stellar components of galaxies. Here I describe measurements of
71: these parameters for the B-star binary HD~42401. This effort is part of an 
72: ongoing program to determine the masses and radii of O- and B-type stars. 
73: Previously, I obtained parameters for the LMC O-star binary [L72]~LH~54-425 
74: \citep{wil08} and will continue with more analyses of eclipsing spectroscopic 
75: binary systems containing O- and B-type stars.
76: 
77: The star HD~42401 was first classified as B2 V by \citet{wal71}. Since 
78: then, it has been used as a spectral standard for the B2 V type stars in a 
79: number of publications \citep{wal90,de00,bag01}. The eclipsing nature of its 
80: light curve was first noticed with the \textit{Hipparcos} satellite 
81: (HIP 29321), where it was 
82: classified as an Algol-type eclipsing binary and given the moniker V1388 Ori 
83: \citep{kaz99}. Three radial velocity measurements of HD~42401 by \citet{feh97}
84: show a range from --103 to 36 km s$^{-1}$, but no further investigations were
85: made into this probable velocity variable. 
86: 
87: Presented here are the analyses of radial velocities from optical spectra 
88: obtained in 2008 January~(\S 2). Section 3 contains a description of the
89: measurement of radial velocities from the data. In \S 4 I describe tomographic
90: reconstructions of the individual component spectra. Covered in \S 5 is the
91: combined light curve and radial velocity curve solution. In \S 6 I 
92: finish with a discussion of the distance to the system, its fundamental
93: parameters, and other consequences of the analysis. 
94: 
95: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
96: 
97: \section{Observations}
98: 
99: I obtained 29 spectra of HD~42401 with the Kitt Peak National Observatory
100: (KPNO) 0.9-m coud\'e feed telescope during an observing run from 2008 
101: Jan 09 to 2008 Jan 18. These spectra were obtained with the long collimator and
102: grating A (632 grooves mm$^{-1}$ with a blaze wavelength of 6000 \AA~in 
103: second order with order sorting filter 4-96). The first two nights of
104: observations were made with the T1KB detector (1024$\times$1024 pixel array 
105: with 24 square $\mu$m pixels) and resulted in a resolving
106: power of $R = \lambda/\delta\lambda = $ 13,500 with wavelength coverage
107: from 4315--4490 \AA. The rest of the observing run made use of the F3KB 
108: detector (3072$\times$1024 pixel array with 15 square $\mu$m pixels) 
109: that resulted in a lower resolving power
110: of $R = \lambda/\delta\lambda = $ 11,500 with wavelength coverage from
111: 4250--4570~\AA. The wavelength coverage was chosen to include the 
112: He~\textsc{i}~$\lambda$4471 and Mg~\textsc{ii}~$\lambda$4481 lines that are
113: good temperature indicators for B-type stars plus the H$\gamma$ line that is
114: sensitive to gravity (linear stark effect). 
115: Exposure times were 1200 s giving a S/N $\simeq$ 
116: 100 pixel$^{-1}$. Numerous comparison spectra were obtained for wavelength
117: calibration, and many bias and flat field spectra were also obtained each 
118: night. The spectra were extracted and calibrated using standard routines
119: in IRAF\footnote{IRAF is distributed by the National Optical Astronomy
120: Observatory, which is operated by the Association of Universities for Research
121: in Astronomy, Inc., under cooperative agreement with the National Science
122: Foundation.}, and then each continuum rectified spectrum was transformed to a
123: common heliocentric wavelength grid in log $\lambda$ increments. 
124: 
125: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
126: 
127: \section{Radial Velocities}\label{sectrv}
128: 
129: The narrow range of wavelength coverage in our spectra limits the lines that 
130: may be used for radial velocity analysis. I measured radial velocities from
131: three lines, He~\textsc{i}~$\lambda$4387, He~\textsc{i}~$\lambda$4471, and 
132: Mg~\textsc{ii}~$\lambda$4481 via a template-fitting scheme \citep{gie02} that 
133: measures velocities by using model templates weighted 
134: by a flux ratio to match both the shifts and line depths in the observed 
135: spectra. There is no evidence of emission or intrinsic line asymmetries in 
136: these lines or H$\gamma$~$\lambda$4340. 
137: 
138: The BSTAR2006 grid of stellar models from \citet{lan07} was used to derive 
139: template spectra. These models are based on the line blanketed, non-LTE, 
140: plane-parallel, hydrostatic atmosphere code TLUSTY and the radiative transfer
141: code SYNSPEC \citep{hub88,hub95,hub98}. In finding templates, initial values
142: were used for temperatures, gravities, projected rotational velocities, and
143: flux contributions from each star. These parameters for model templates were
144: then checked by comparing the three lines used in radial velocity 
145: measurements against the tomographically reconstructed spectra of each star
146: (\S\ref{tomog}). The parameters were changed and new templates made 
147: after initial fits to the light curve and radial velocity curves (\S 5) 
148: indicated slightly different values were more appropriate. The velocity 
149: analysis was then performed again until the best fit was obtained. In this way,
150: I was able to obtain an estimate of the monochromatic flux ratio
151: in the blue spectra of $F_{\rm 2}/F_{\rm 1} = 0.25 \pm 0.05$ based upon the
152: relative line depths of the spectral components.
153: 
154: The template fitting scheme also needs preliminary estimates for the velocities
155: of each component. To obtain these, spectra which clearly showed two sets of
156: spectral lines were analyzed. Midpoints of spectral features were used to 
157: make crude estimates of radial velocities for each star. Preliminary orbital
158: parameters were obtained by using the nonlinear, least-squares fitting
159: program of \citet{mor74}. The preliminary velocities from this initial 
160: orbital solution were starting points for performing a nonlinear, least-squares
161: fit of the composite profiles with the template spectra and calculating the 
162: shifts for each star. The fitting scheme searches a region of $\pm$10 
163: \AA~around the rest wavelength of each line, so preliminary velocities are mere
164: starting points and need not be highly accurate. 
165: The three values for radial velocity from each spectrum
166: were averaged, and the standard deviation of the mean value was calculated. 
167: Each of these values are listed in Table \ref{rvs} for the primary and 
168: secondary stars. Also listed are the orbital phase for each observation and the
169: observed minus calculated values for each data point. Phase zero is defined
170: as the time of inferior conjunction of the primary star, $T_{\rm IC,1}$ 
171: (time of secondary minimum in the light curve). Figure \ref{fig1} shows two 
172: sample spectra at times of near quadrature according to this ephemeris. 
173: 
174: During eclipse phases, velocity measurements may be affected by the 
175: Rossiter-McLaughlin effect in which the center of light of the eclipsed star
176: will appear redshifted on ingress and blue shifted on egress because of the
177: rotational Doppler shifts of the visible portions. 
178: This effect will result in narrower line profiles as 
179: the red and blue shifted regions of the eclipsed star are blocked, and it is
180: more prominent in broad-lined, rapidly rotating systems. HD~42401 is a short 
181: period binary having relatively narrow lines, therefore the effect should be 
182: small. However, since the template method does not account for eclipses, 
183: deviations from orbital motion near eclipse phases are expected.
184: 
185: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
186: 
187: \section{Tomographic Reconstruction}\label{tomog}
188: 
189: The Doppler tomography algorithm of \citet{bag94} was used to separate the
190: primary and secondary spectra of HD~42401 for the F3KB spectra. This iterative 
191: method uses the 24 observed composite spectra, their velocity shifts,
192: and an assumed monochromatic flux ratio to derive individual
193: component spectra. The best flux ratio was the one that best matched line 
194: depths in the reconstructions with those in the model spectra. 
195: Figure~\ref{fig2} shows the reconstructed spectra for the primary and 
196: secondary with the best fit models over-plotted. The relative depths of
197: He~\textsc{i}~$\lambda$4471 and Mg~\textsc{ii}~$\lambda$4481 are good 
198: temperature indicators throughout the B-star sequence. Specifically, the 
199: He~\textsc{i}~$\lambda$4471 line gets weaker while the 
200: Mg~\textsc{ii}~$\lambda$4481 line gets stronger as temperature
201: decreases, as is seen in the spectrum of the secondary compared to that of
202: the primary. The set of spectra from the F3KB instrument has only one spectrum
203: that is close to an eclipse. Because there are many more spectra outside 
204: eclipse used in the tomographic reconstruction, the final reconstructed spectra
205: are insensitive to details of the Rossiter-McLaughlin effect. Indeed, 
206: tomographic reconstructions with the one eclipse spectrum omitted are 
207: negligibly different from those presented in Fig. \ref{fig2}.
208: 
209: The final reconstructed spectra were fit with TLUSTY/SYNSPEC model synthesis
210: spectra (see \S \ref{sectrv}). Fits to all three lines made in the velocity
211: analysis were used to estimate the projected rotational velocity $V \sin i$,
212: and the temperature and surface gravity of each star were estimated by 
213: comparing the reconstructed and model profiles for a grid of test values 
214: \citep[details are given in][]{wil08}. 
215: These parameter values are listed in Table \ref{tomo}.
216: 
217: The narrow wavelength range of the spectra is insufficient to attempt classical
218: spectral typing. However, I can arrive at an estimate of the spectral types 
219: for each 
220: star in the HD~42401 system by comparing the derived effective temperatures
221: with a spectral type versus effective temperature relation. 
222: According to Table 2 of \citet{boh81} the effective temperature and gravity 
223: (\S 5) of the primary of HD~42401 is most consistent with a B2.5 IV-V star 
224: while the secondary matches most closely with a B3 V star, and
225: these classifications are listed in Table \ref{tomo}.
226: 
227: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
228: 
229: \section{Combined Radial Velocity and Light Curve Solution}
230: 
231: The All Sky Automated Survey \citep[ASAS-3;][]{poj02} database contains light 
232: curves for $\sim$39,000 previously unknown variable stars. I extracted the
233: $V$-band light curve for HD~42401 (ASAS 061059+1159.7) from this catalog, 
234: removing points deemed lower quality by the data reduction pipeline used by 
235: the ASAS. 
236: 
237: I used the Eclipsing Light Curve (ELC) code \citep{oro00} to find orbital 
238: and astrophysical parameters for the HD~42401 system. ELC fits the radial 
239: velocity and light curves simultaneously, giving a joint orbital ephemeris 
240: based on both sets of data. The best fit for the radial velocity curve is shown
241: in Figure \ref{fig3}(upper panel) along with the observed minus calculated 
242: residuals (lower panel). ELC treats calculation of velocities during eclipse 
243: via a flux-weighted velocity centroid as described in \citet{wil76}. 
244: The Rossiter-McLaughlin effect is clearly seen in the 
245: best fit as a slight redshift going into eclipse and a slight blueshift 
246: leaving eclipse and the model velocity curves match the data well. 
247: The best fit for the $V$-band 
248: light curve is shown in Figure \ref{fig4}. Figures \ref{fig5} and \ref{fig6} 
249: show the details of the fit for the light curve in the region of the primary 
250: and secondary eclipses, respectively.
251: 
252: The genetic optimizer mode of ELC was used initially to
253: explore wide ranges of values for the period, epoch of inferior conjunction of
254: the primary $T_{\rm IC,1}$, inclination, mass ratio, primary velocity 
255: semiamplitude, and Roche lobe filling factor for each star. The Roche lobe
256: filling factor is defined by \citet{oro00} as the ratio of the radius of the 
257: star toward the inner Lagrangian point ($L_{\rm 1}$) to the distance to 
258: $L_{\rm 1}$ from the center of the star, $f \equiv x_{\rm point}/x_{L \rm 1}$.
259: 
260: The use of ELC included fixing the temperature of each star to the values
261: found in the tomographic reconstructions (\S \ref{tomog}) and also fixing 
262: the radius ratio ($R_1 / R_2$) based upon the temperatures, surface fluxes, and
263: monochromatic flux ratio of the two stars. Non-zero eccentricities for the 
264: HD~42401 system were explored during fitting but rejected based on the 
265: higher $\chi^2$ values for those fits. 
266: 
267: To estimate the uncertainties based on our best fit, the values of the seven
268: fitted parameters were varied in the calculation of $\sim 3~\times~10^6$ 
269: light and radial velocity curves. The well-explored $\chi^2$ surface was then
270: projected as a function of each fitted parameter or astrophysical parameter of
271: interest. The lowest $\chi^2$ value is found for each parameter, and the 
272: 1-$\sigma$ uncertainty may be estimated by the region where 
273: $\chi^2~\le~\chi^{2}_{\rm min}~+1$. These values and uncertainties are listed
274: in Table \ref{orb} for the orbital parameters and Table \ref{ELCparms} for 
275: the astrophysical parameters. Also listed are $R_{\rm eff}$, the effective
276: radius of a sphere with the same volume, $R_{\rm pole}$, the polar radius
277: of each star, and $R_{\rm point}$, the radius of the each star toward the inner
278: Lagrangian point. Our derived period of P= 2.18706 $\pm$ 0.00005 days for 
279: HD~42401 agrees well with the value found by the \textit{Hipparcos} satellite 
280: of 2.18709 $\pm$ 0.00050 days \citep{esa97}.
281: 
282: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
283: 
284: \section{Discussion and Conclusions}
285: 
286: The goal with this work was to determine the masses and radii of the stars
287: in the HD~42401 system to great accuracy. I have determined the masses of the
288: two stars to $1.1 \%$ and $0.6 \%$ and the radii of the stars to $0.7 \%$ and 
289: $0.8 \%$, for the primary and secondary respectively. Armed with these numbers 
290: and the results from Table \ref{ELCparms}, I can consider the details of the 
291: HD~42401 system.
292: 
293: As is seen in Table \ref{ELCparms}, both stars are well within their Roche
294: radii but experience tidal distortion that is evident in the light curve
295: (Fig. \ref{fig4}). The rotational velocities derived from the 
296: tomographic reconstructions (Table \ref{tomo}) match very well with the 
297: synchronous rotation values found by ELC (Table \ref{ELCparms}), indicating 
298: that this system has achieved synchronous rotation, and is therefore not very
299: young. 
300: 
301: To make an estimate of the age of the system, I used the effective 
302: temperatures from Table \ref{tomo} and radii from Table \ref{ELCparms} to plot
303: the two stars of HD~42401 on a theoretical H-R diagram and compare their 
304: locations to
305: evolutionary tracks. The result is shown in Figure \ref{fig7}, plotted against
306: evolutionary tracks for stars of 5, 7, and 9 $M_{\odot}$ from \citet{sch92} 
307: as well as isochrones from \citet{lej01} for solar metallicity with ages of
308: 21.9, 25.1, 27.5, and 31.6 Myr. The location of the stars is most consistent 
309: with an age of $\sim$25 Myr. The model tracks shown in Figure~\ref{fig7} are 
310: for non-rotating stellar models. The present ratios of spin angular velocity
311: to critical angular velocity are $\Omega / \Omega_{\rm crit} = 0.45$ and
312: 0.31 for the primary and secondary, respectively (assuming synchronous 
313: rotation), and the evolutionary tracks for such moderate rotation rates are
314: only slightly steeper and more extended in time than those illustrated
315: \citep{eks08}. Thus, our derived age may slightly underestimate the actual
316: value. 
317: 
318: Both stars appear (in Fig. \ref{fig7}) to be overluminous for the derived 
319: masses of 7.42 $M_{\odot}$ for the primary and 5.16 $M_{\odot}$ for the 
320: secondary. Table 3 of \citet{har88} lists astrophysical parameters for stars
321: as a function of spectral type for main sequence stars based on empirical data
322: from eclipsing binaries. The mass and effective 
323: temperature of the primary fit between the listed values for spectral types
324: B2 (mean of 8.6 $M_{\odot}$) and B3 (mean of 6.1 $M_{\odot}$), but the radius
325: is much larger than the means for comparable spectral types and matches a 
326: B0.5 star (mean of 5.5 $R_{\odot}$). The mass of the secondary 
327: is consistent with the B4 spectral type (mean of 5.1 $M_{\odot}$) while the
328: effective temperature and radius appear more consistent with the B3 spectral 
329: type. In a study of eclipsing 
330: binaries in the Small Magellanic Cloud, \citet{hil05} found several systems
331: of comparable mass that, like HD~42401, are overluminous compared to model
332: predictions. However, our findings for HD~42401 seem to conflict with the 
333: results of \citet{mal03} who shows that early B-type stars that are in close 
334: systems and rotate more slowly than single stars are on average smaller than 
335: those same single stars. In a subsequent paper, \citet{mal07} studied 
336: well-separated binaries in an effort to use the properties of their component 
337: stars for a more direct comparison with
338: single stars. His resulting mass-luminosity-radius relations, when applied to
339: our results for HD~42401, predict less luminous, hotter and smaller components.
340: This is perhaps not surprising, due to the age of HD~42401 and the evolution
341: of its components from the zero age main sequence.
342: 
343: I can also estimate the distance to the system by fitting a spectral energy
344: distribution (SED) to various photometric measurements. The spectra were not
345: flux calibrated, so to create an SED, I must rely on photometry performed on 
346: the system.
347: HD~42401 is a relatively bright $(V \sim 7.4)$ system and has thus been well
348: studied. The problem arises in estimating the phase at which a particular
349: observation in the historical literature was made. Two observations of HD~42401
350: were taken by $IUE$ at $\phi = 0.12 \pm 0.07$ based on our orbital ephemeris. 
351: The uncertainty in this estimate comes from the uncertainties in our values
352: for $T_{\rm IC,1}$ and period and the number of orbits between the $IUE$ 
353: observations and our observations. The range in uncertainty for the orbital
354: phase of the $IUE$ data
355: brings it close to eclipse, and this was supported by initial SED fits 
356: including the $IUE$ data that indicated a lower UV flux than indicated by other
357: measurements. There are various other measurements that also were 
358: sufficiently close to eclipses to be omitted in SED fitting. Fortunately, 2MASS
359: \citep{cut03} measurements were obtained at  $\phi = 0.16 \pm 0.03$ and are 
360: therefore far enough away from
361: eclipse, within uncertainties, to be used. The only other points used were 
362: Johnson $UBV$ magnitudes. To obtain these values, I used the value at 
363: quadrature from our light curve of $V = 7.40$ and applied colors of 
364: $(B-V) = -0.042$ and $(U-B) = -0.620$ which are averages of several 
365: observations that are listed by \citet{mer94}. The SED fit is shown in Figure
366: \ref{fig8} along with the $U, B, V, J, H, K_s$ magnitudes. I used two model 
367: spectra from \citet{lan07} matching our values in Table \ref{tomo} for the 
368: effective temperatures and gravities of each star, and scaled in the blue by 
369: the flux ratio of $0.25 \pm 0.05$ found from the tomographic reconstructions. 
370: This fit of the SED results in a limb darkened, angular 
371: diameter for the primary of $\theta_{\rm LD} = 62.7 \pm 1.1~\mu$as with a
372: reddening of $E(B-V) = 0.15 \pm 0.01$ mag and a ratio of total-to-selective 
373: extinction of $R = 3.24 \pm 0.10$. By directly comparing this angular 
374: diameter with the value for the radius of the primary, I estimate the distance
375: of the system to be $d = 832 \pm 89$ pc. This distance and reddening are 
376: in excellent agreement with the values in \citet{bow08} of 
377: $E(B-V) = 0.15 \pm 0.02$ mag and $d = 0.8$ kpc. 
378: 
379: HD~42401 has Galactic coordinates of $\ell = 197\fdg64$ and $b = -3\fdg33$ 
380: \citep{ree05}. This is close to the Galactic open cluster NGC~2169 at 
381: $\ell = 195\fdg61$ and $b = -2\fdg93$ (separation $\sim 2\fdg07$). 
382: \citet{abt77} found that the
383: earliest spectral type of the cluster members of NGC~2169 was B2 III. 
384: \citet{jef07} find a reddening value of $E(B-V) = 0.20 \pm 0.01$ mag and a 
385: distance of $\sim1060$ pc to NGC~2169. Proper motions for the objects are
386: similar as well, with HD~42401 having 
387: $\mu_{\alpha}~cos~\delta = -2.25 \pm 1.41$ mas yr$^{-1}$ and 
388: $\mu_{\delta} = -2.71 \pm 0.53$ mas yr$^{-1}$ \citep{esa97} and NGC~2169 
389: having $\mu_{\alpha}~cos~\delta = -2.17 \pm 1.41$ mas yr$^{-1}$ and 
390: $\mu_{\delta} = -2.52 \pm 0.53$ mas yr$^{-1}$ 
391: \citep{kha05}. The systemic velocity derived here for HD~42401 is $\sim$ 14 km
392: s$^{-1}$ and the average of 8 stars in NGC~2169 is 22 km s$^{-1}$ 
393: \citep{kha05}. 
394: While the proximity of HD~42401 to NGC~2169 is intriguing and the fact that 
395: the earliest star has a similar spectral type to the components of HD~42401, 
396: actual cluster membership is unlikely. The age of NGC~2169 in the literature
397: ranges from $9 \pm 2$ Myr \citep{jef07} to 7.8 Myr \citep{kha05}, too young to 
398: comfortably fit with the data for the components of
399: HD~42401 in Figure \ref{fig7}. \citet{kha05} determined the corona radius of
400: NGC~2169 by matching stellar densities of the cluster to that of the background
401: and found an angular size of 0.15 degrees. The distance on the sky of HD~42401
402: from NGC~2169 is about 14 times this radius. 
403: The difference in distances to the two objects 
404: and their separation on the sky make cluster membership even more unlikely. 
405: However, it is possible that a wave of star formation associated with the Orion
406: arm of the Galaxy swept through the region, leaving HD~42401 behind as it moved
407: on and eventually led to the creation of NGC~2169. 
408: 
409: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
410: 
411: \acknowledgments
412: 
413: I would like to thank the anonymous referee whose comments improved the
414: paper.
415: I would also like to thank Daryl Willmarth and the staff of KPNO for their 
416: assistance in making these observations possible and Jerry Orosz for 
417: clarifying some of the finer points of ELC's 
418: calculations. I am also greatly indebted to D.~R.~Gies for his suggestions that
419: improved the quality of this paper. This material is based on
420: work supported by the National Science Foundation under grants AST 05-06573
421: and AST 06-06861. I gratefully acknowledge support from the GSU College
422: of Arts and Sciences and from the Research Program Enhancement fund of the
423: Board of Regents of the University System of Georgia, administered through
424: the GSU Office of the Vice President for Research. This research has made 
425: use of the SIMBAD database, operated at CDS, Strasbourg, France and  
426: of the WEBDA database, operated at the Institute for Astronomy of the 
427: University of Vienna.
428: 
429: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
430: 
431: \bibliographystyle{apj}
432: 
433: \bibliography{apj-jour,paper}
434: 
435: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
436: 
437: \newpage
438: 
439: \begin{deluxetable}{cccccccc}
440: \tabletypesize{\scriptsize}
441: \tablewidth{0pt}
442: \tablecaption{HD~42401 Radial Velocity Measurements\label{rvs}}
443: \tablehead{
444: \colhead{Date}          &
445: \colhead{Orbital}       &
446: \colhead{$V_1$}         &
447: \colhead{$\sigma_{1}$}  &
448: \colhead{$(O-C)_1$}     &
449: \colhead{$V_2$}         &
450: \colhead{$\sigma_{2}$}  &
451: \colhead{$(O-C)_2$}     \\
452: \colhead{(HJD$-$2,400,000)}        &
453: \colhead{Phase}  &
454: \colhead{(km s$^{-1}$)} &
455: \colhead{(km s$^{-1}$)} &
456: \colhead{(km s$^{-1}$)} &
457: \colhead{(km s$^{-1}$)} &
458: \colhead{(km s$^{-1}$)} &
459: \colhead{(km s$^{-1}$)} }
460: \startdata
461:  54474.738 & 0.548 & \phn--36.3   & 5.4 & \phn6.9 & \phn\phn90.6 & \phn3.8 & \phn12.9   \\
462:  54474.836 & 0.593 & \phn--71.7   & 3.3 & --1.3   & \phn138.1    & 13.8    & \phn\phn4.9\\
463:  54476.690 & 0.441 & \phn\phn82.7 & 0.4 & \phn2.3 & \phn--45.7   & \phn6.3 & \phn20.4   \\
464:  54476.735 & 0.461 & \phn\phn65.6 & 3.3 & --1.6   & \phn--21.1   & \phn8.3 & \phn18.1   \\
465:  54476.775 & 0.480 & \phn\phn53.8 & 1.6 & \phn5.7 & \phn--35.8   & 21.7    & --20.9     \\
466:  54477.672 & 0.890 & \phn--84.6   & 1.8 & --3.8   & \phn152.6    & \phn6.9 & \phn\phn1.2\\
467:  54477.720 & 0.912 & \phn--70.4   & 4.0 & --6.5   & \phn129.4    & \phn4.7 & \phn\phn2.4\\
468:  54477.800 & 0.949 & \phn--41.4   & 3.0 & --8.9   & \phn\phn84.4 & \phn2.0 & --13.8     \\
469:  54478.660 & 0.342 & \phn139.3    & 0.9 & --3.9   & --160.9      & \phn5.4 & \phn\phn8.2\\
470:  54478.676 & 0.349 & \phn133.3    & 2.0 & --6.3   & --160.1      & \phn5.6 & \phn\phn3.6\\
471:  54478.718 & 0.368 & \phn125.0    & 1.9 & --3.3   & --156.2      & \phn5.4 & \phn--9.2  \\
472:  54478.779 & 0.396 & \phn109.6    & 4.1 & \phn0.3 & --114.4      & 11.0    & \phn\phn4.9\\
473:  54478.840 & 0.424 & \phn\phn87.6 & 2.0 & --1.6   & \phn--76.1   & 16.9    & \phn10.9   \\
474:  54479.667 & 0.802 & --128.4      & 0.9 & --1.9   & \phn218.7    & \phn4.6 & \phn\phn0.6\\
475:  54479.718 & 0.825 & --115.9      & 1.2 & \phn2.3 & \phn214.2    & \phn5.3 & \phn\phn8.3\\
476:  54479.760 & 0.845 & --109.8      & 0.9 & --0.8   & \phn194.8    & \phn2.0 & \phn\phn2.4\\
477:  54479.825 & 0.874 & \phn--90.2   & 0.5 & \phn1.3 & \phn174.5    & \phn2.6 & \phn\phn7.5\\
478:  54480.635 & 0.245 & \phn163.8    & 0.7 & --1.5   & --197.1      & \phn6.8 & \phn\phn6.6\\
479:  54480.684 & 0.267 & \phn163.2    & 3.8 & --1.6   & --196.5      & \phn5.7 & \phn\phn6.2\\
480:  54480.755 & 0.300 & \phn155.8    & 2.3 & --3.3   & --194.5      & \phn4.2 & \phn--1.0  \\
481:  54480.802 & 0.321 & \phn151.8    & 2.0 & --0.5   & --186.8      & \phn4.3 & \phn--3.9  \\
482:  54481.635 & 0.702 & --127.3      & 2.5 & \phn1.6 & \phn240.8    & \phn6.7 & \phn20.8   \\
483:  54481.680 & 0.722 & --135.7      & 2.4 & --2.7   & \phn226.5    & \phn6.5 & \phn\phn0.0\\
484:  54481.728 & 0.744 & --135.8      & 0.6 & --1.1   & \phn222.2    & \phn2.5 & \phn--7.4  \\
485:  54481.794 & 0.775 & --135.3      & 2.7 & --2.5   & \phn234.0    & \phn4.6 & \phn\phn6.9\\
486:  54483.646 & 0.622 & \phn--94.4   & 2.2 & --3.3   & \phn171.4    & \phn5.9 & \phn\phn8.2\\
487:  54483.687 & 0.640 & --108.5      & 1.9 & --5.8   & \phn180.2    & \phn2.5 & \phn--0.1  \\
488:  54483.757 & 0.672 & --120.3      & 2.0 & --1.5   & \phn200.8    & \phn2.9 & \phn--3.6  \\
489:  54483.839 & 0.710 & --129.4      & 2.0 & \phn1.3 & \phn223.4    & \phn3.6 & \phn\phn0.5
490: \enddata
491: 
492: \end{deluxetable}
493: 
494: \begin{deluxetable}{lcc}
495: \tablewidth{0pc}
496: \tablecaption{Tomographic Spectral Reconstruction Parameters\label{tomo}}
497: \tablehead{
498:   \colhead{Parameter} &
499:   \colhead{Primary}   &
500:   \colhead{Secondary}}
501: \startdata
502: Spectral Type\dotfill       & B2.5 IV-V\tablenotemark{a} & B3 V\tablenotemark{a}\\
503: $T_{\rm eff}$ (kK)\dotfill  & 20.5 $\pm$ 0.5 & 18.5 $\pm$ 0.5\\
504: log $g$ (cgs)\dotfill       & \phn3.75 $\pm$ 0.25 & \phn4.00 $\pm$ 0.25\\
505: $V~\sin~i$ (km s$^{-1}$)\dotfill & 125 $\pm$ 10 & \phn75 $\pm$ 15\\
506: $F_{\rm 2} / F_{\rm 1}$ (blue) & \multicolumn{2}{c}{0.25 $\pm$ 0.05}
507: \enddata
508: \tablenotetext{a}{These spectral types are estimated from derived values of 
509: $T_{\rm eff}$ and log $g$.}
510: \end{deluxetable}
511: 
512: \begin{deluxetable}{lc}
513: \tablewidth{0pc}
514: \tablecaption{Circular Orbital Solution for HD~42401\label{orb}}
515: \tablehead{
516:   \colhead{Element} &
517:   \colhead{Value}   }
518: \startdata
519: $P$~(days)\dotfill                      & \phn\phn\phn\phn2.18706 $\pm$ 0.00005\\
520: $T_{\rm IC,1}$ (HJD--2,400,000)\dotfill & 54477.9130 $\pm$ 0.0002\\
521: $K_{\rm 1}$ (km s$^{-1}$)\dotfill       & \phn\phn151.4 $\pm$ 0.3 \\
522: $K_{\rm 2}$ (km s$^{-1}$)\dotfill       & \phn\phn217.9 $\pm$ 1.0 \\
523: $\gamma_{\rm 1}$ (km s$^{-1}$)\dotfill  & \phn\phn\phn15.32 $\pm$ 0.07\\
524: $\gamma_{\rm 2}$ (km s$^{-1}$)\dotfill  & \phn\phn\phn12.9 $\pm$ 0.2\\
525: rms~(primary)~(km s$^{-1}$)\dotfill     & \phs\phs3.7 \\
526: rms~(secondary)~(km s$^{-1}$)\dotfill   & \phs\phs9.7 \\
527: rms~(photometry)~(mag)\dotfill          & \phs\phs0.007 \\
528: \enddata
529: \end{deluxetable}
530: 
531: \begin{deluxetable}{lcc}
532: \tablewidth{0pc}
533: \tablecaption{ELC Model Parameters for HD~42401\label{ELCparms}}
534: \tablehead{
535:   \colhead{Parameter}  &
536:   \colhead{Primary}    &
537:   \colhead{Secondary}  }
538: \startdata
539: Inclination (deg)\dotfill & \multicolumn{2}{c}{75.5 $\pm$ 0.2}\\
540: $M~(M_{\odot})$\dotfill   & \phn\phn7.42 $\pm$ 0.08 & \phn5.16 $\pm$ 0.03\\
541: $R_{\rm eff}~(R_{\odot})$\dotfill & \phn\phn5.60 $\pm$ 0.04 & \phn3.76 $\pm$ 0.03\\
542: $R_{\rm pole}$\tablenotemark{a}~($R_{\odot}$)\dotfill & \phn\phn5.40 $\pm$ 0.02 & \phn3.70 $\pm$ 0.01\\
543: $R_{\rm point}$\tablenotemark{b}~($R_{\odot}$)\dotfill & \phn\phn5.97 $\pm$ 0.02 & \phn3.87 $\pm$ 0.01\\
544: $V_{\rm sync}~\sin~i$ (km s$^{-1}$)\dotfill & 125.5 $\pm$ 0.9 & 84.3 $\pm$ 0.6\\
545: log $g$ (cgs)\dotfill & \phn\phn3.812 $\pm$ 0.005 & \phn3.999 $\pm$ 0.006\\
546: Filling factor\dotfill & \phn\phn0.674 $\pm$ 0.005 & \phn0.588 $\pm$ 0.010\\
547: $a_{\rm tot}~(R_{\odot})$\dotfill& \multicolumn{2}{c}{16.48 $\pm$ 0.05} \\
548: $F_{\rm 2} / F_{\rm 1}$ (blue)\dotfill& \multicolumn{2}{c}{\phn0.27 $\pm$ 0.04}
549: \enddata
550: \tablenotetext{a}{Polar radius.}
551: \tablenotetext{b}{Radius toward the inner Lagrangian point.}
552: \end{deluxetable}
553: 
554: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
555: 
556: \clearpage
557: 
558: %
559: % Figures!!
560: %
561: 
562: \input{epsf}
563: \begin{figure}
564: \begin{center}
565: {\includegraphics[angle=90,height=12cm]{f1.eps}}
566: \end{center}
567: \caption{Two spectra of HD~42401 obtained near quadrature phases (offset for 
568:   clarity). Solid lines indicate the positions of lines of the 
569:   primary component and dashed lines indicate those for the secondary 
570:   component. The lines marked are those used in the velocity analysis.}
571: \label{fig1}
572: \end{figure}
573: 
574: \input{epsf}
575: \begin{figure}
576: \begin{center}
577: {\includegraphics[angle=90,height=12cm]{f2.eps}}
578: \end{center}
579: \caption{Tomographic reconstructions of the components of HD~42401 based on
580:   24 spectra obtained with the F3KB instrument during January 2008 at KPNO. 
581:   The top solid line represents the reconstructed spectrum of the primary, 
582:   the bottom solid line is the reconstructed secondary spectrum offset by 0.3
583:   for clarity. Over-plotted for both are the model spectra for each (dashed 
584:   lines). The stellar parameters for the model spectra are given in 
585:   Table \ref{tomo}.}
586: \label{fig2}
587: \end{figure}
588: 
589: \input{epsf}
590: \begin{figure}
591: \begin{center}
592: {\includegraphics[angle=90,height=12cm]{f3.eps}}
593: \end{center}
594: \caption{Radial velocity curves for HD~42401. Primary radial velocities are
595:   shown by filled dots and secondary radial velocities are shown by filled
596:   triangles with associated uncertainties shown as line segments for both. 
597:   The solid line is the best fit solution for the primary and the dashed line
598:   is the same for the secondary. The lower panel shows the observed minus 
599:   calculated values for each measurement with uncertainties.}
600: \label{fig3}
601: \end{figure}
602: 
603: \input{epsf}
604: \begin{figure}
605: \begin{center}
606: {\includegraphics[angle=90,height=12cm]{f4.eps}}
607: \end{center}
608: \caption{$V$-band light curve for HD~42401. These data were taken from the All 
609:   Sky Automated Survey database \citep{poj02} and are presented here in phase 
610:   according to our best fit solution. The model is the thick gray line 
611:   and the data are the filled dots with $V$ uncertainties represented by line 
612:   segments. Phase zero corresponds to inferior conjunction of the primary 
613:   star.}
614: \label{fig4}
615: \end{figure}
616: 
617: \input{epsf}
618: \begin{figure}
619: \begin{center}
620: {\includegraphics[angle=90,height=12cm]{f5.eps}}
621: \end{center}
622: \caption{$V$-band light curve for HD~42401 around the time of primary eclipse. 
623:   As with Fig. \ref{fig4}, the model is the thick gray line and the data are 
624:   filled dots with line segments representing uncertainties. Residuals are 
625:   distributed as expected around the best fit.}
626: \label{fig5}
627: \end{figure}
628: 
629: \input{epsf}
630: \begin{figure}
631: \begin{center}
632: {\includegraphics[angle=90,height=12cm]{f6.eps}}
633: \end{center}
634: \caption{$V$-band light curve for HD~42401 around the time of secondary 
635:   eclipse. As with Fig. \ref{fig4}, the model is the thick gray line and the 
636:   data are filled dots with line segments representing uncertainties. Here, as 
637:   in Fig. \ref{fig5}, the residuals are distributed as expected around the 
638:   best fit.}
639: \label{fig6}
640: \end{figure}
641: 
642: \input{epsf}
643: \begin{figure}
644: \begin{center}
645: {\includegraphics[angle=90,height=12cm]{f7.eps}}
646: \end{center}
647: \caption{A theoretical H-R diagram showing the location of the primary star 
648:   (\textit{filled circle}) and secondary star (\textit{filled triangle}) of 
649:   HD~42401 including uncertainty regions for each. Also plotted are 
650:   evolutionary tracks for stars of various masses 
651:   from \citet{sch92} and isochrones (\textit{vertical dashed lines}) from 
652:   \citet{lej01} for solar metallicity with ages of 21.9, 25.1, 27.5, and 31.6 
653:   Myr going from left to right. The positions of the two stars are consistent 
654:   with an age of $\sim$25 Myr.}
655: \label{fig7}
656: \end{figure}
657: 
658: \input{epsf}
659: \begin{figure}
660: \begin{center}
661: {\includegraphics[angle=90,height=12cm]{f8.eps}}
662: \end{center}
663: \caption{The spectral energy distribution and fit for the combined light of the
664:   HD~42401 components (\textit{solid line}) to Johnson $U, B, V, J, H, K_s$ 
665:   photometry (\textit{open circles}).}
666: \label{fig8}
667: \end{figure}
668: 
669: \end{document}
670: