1: \documentclass[12pt,preprint]{aastex}
2:
3: \usepackage{color}
4: \newcommand{\vdag}{(v)^\dagger}
5: \newcommand{\myemail}{i.hunter@qub.ac.uk}
6:
7:
8: \shorttitle{Rotation and nitrogen enrichment in massive stars}
9: \shortauthors{Hunter et al.}
10:
11:
12: \begin{document}
13:
14: \title{The VLT-FLAMES survey of massive stars: rotation and nitrogen
15: enrichment as the key to
16: understanding massive star evolution}
17:
18:
19: \author{I. Hunter\altaffilmark{1,2}, I. Brott\altaffilmark{3},
20: D.J. Lennon\altaffilmark{4},
21: N. Langer\altaffilmark{3}, P.L. Dufton\altaffilmark{1},
22: C. Trundle\altaffilmark{1}, S.J. Smartt\altaffilmark{1},
23: A. de Koter\altaffilmark{5},
24: C.J. Evans\altaffilmark{6} and R.S.I. Ryans\altaffilmark{1}}%
25:
26:
27: \altaffiltext{1}{Astrophysical Research Centre, School of Mathematics and
28: Physics, The Queen's University of Belfast, Belfast, BT7 1NN,
29: Northern Ireland, UK}
30: \altaffiltext{2}{The Isaac Newton Group of Telescopes, Apartado de Correos 321, 38700
31: Santa Cruz de La Palma, Canary Islands, Spain}
32: \altaffiltext{3}{Astronomical Institute, Utrecht University, Princetonplein 5,
33: NL-3584CC, Utrecht, Netherlands}
34: \altaffiltext{4}{Instituto de Astrof\'{i}sica de Canarias, Calle V\'{i}a L\'{a}ctea,
35: E-38200 La Laguna, Spain}
36: \altaffiltext{6}{Astronomical Institute Anton Pannekoek, University of Amsterdam, Kruislaan 403, 1098
37: SJ Amsterdam, Netherlands}
38: \altaffiltext{5}{UK Astronomy Technology Centre, Royal Observatory, Blackford Hill,
39: Edinburgh, EH9 3HJ}
40:
41: \begin{abstract}
42: Rotation has become an important element in evolutionary models
43: of massive stars, specifically via the prediction of rotational mixing.
44: Here, we study a sample of stars, including rapid rotators, to constrain
45: such models and use nitrogen enrichments as a probe of the mixing
46: process. Chemical compositions (C, N, O, Mg and Si) have been estimated for 135
47: early B-type stars in the Large Magellanic Cloud with projected
48: rotational velocities up to $\sim$300\,km\,s$^{-1}$ using a non-LTE {\sc
49: tlusty} model atmosphere grid.
50: Evolutionary models, including rotational mixing, have been generated
51: attempting to reproduce these observations by adjusting the overshooting and
52: rotational mixing parameters and produce reasonable agreement with
53: 60\% of our core hydrogen burning sample.
54: We find (excluding known binaries) a significant population
55: of highly nitrogen enriched intrinsic slow rotators
56: ($v \sin i \lesssim$50\,km\,s$^{-1}$) incompatible with our models ($\sim$20\% of the sample).
57: Furthermore, while we find fast rotators with enrichments
58: in agreement with the models, the observation of evolved ($\log g < 3.7$dex)
59: fast rotators that are relatively unenriched (a further $\sim$20\% of the sample)
60: challenges the
61: concept of rotational mixing. We also find that 70\% of our blue supergiant sample
62: cannot have evolved directly from the hydrogen burning main-sequence.
63: We are left with a picture where invoking binarity and perhaps fossil magnetic fields
64: are required to understand the surface properties of a population of massive main sequence stars.
65:
66:
67:
68:
69: %find two groups of core hydrogen burning objects in conflict with
70: %these evolutionary tracks; a group of evolved fast rotators ($\sim$250\,km\,s$^{-1}$) with
71: %little evidence of mixing, and a group of slow rotators ($\lesssim$50\,km\,s$^{-1}$)
72: %which show up to 0.7\,dex increase in the surface nitrogen abundance.
73: %While both groups may be populated by close binaries,
74: %we suggest that the latter may represent predominantly magnetic stars.
75: %%Taken out by Norbert (see ack. email).
76: %%{\bf Additionally we find a large faction of highly enriched supergiants
77: %%that are not reproduced by the models.}
78: %While rotational mixing could be responsible for a small fraction of
79: %or sample stars (about 15), the presence of about 5 rapidly rotating
80: %($v \sin i \simeq 150 ...300$\,km\,s$^{-1}$) evolved ($\log g \lt 3.7$dex)
81: %core hydrogen buring stars challenges the concept of rotational mixing.
82: %We conclude that rotational mixing \textcolor{blue}{may} not the dominant
83: %process responsible for the nitrogen enhancements
84: %of {\bf a significant faction of} our sample.
85: \end{abstract}
86:
87:
88: \keywords{stars: early-type --- stars: rotation --- stars: abundances --- stars: evolution --- Magellanic Clouds}
89:
90:
91: \section{Introduction} \label{s_intro}
92:
93: Stellar rotation in massive stars has been adopted in recent theoretical
94: models to, for example, predict the correct blue to red supergiant ratio \citep{mae01},
95: the progenitors of gamma-ray bursts through homogeneous evolution \citep{yoo05} and
96: Wolf-Rayet populations as a function of metallicity (\citealt{mey05}; \citealt{vin05}).
97: However, the consequent
98: surface enrichment of helium and nitrogen through rotational mixing is poorly constrained
99: by observations \citep{daf01}, with the mixing typically calibrated from
100: either studies of main-sequence
101: stars with low projected rotational velocities \citep{kor02}
102: or evolved supergiant stars \citep{ven99}. The former are biased towards slow rotators while the latter
103: have evolved beyond the core hydrogen burning phase. The analysis
104: of an unbiased sample of fast rotating core-hydrogen
105: burning stars in order to properly calibrate the predicted rotational mixing
106: efficiency was a strong driver for
107: a large survey of O- and
108: early B-type stars in our
109: Galaxy and the Magellanic Clouds
110: \citep{eva05}. \citet{hun07a} and \citet{tru07}, hereafter Paper I and II
111: respectively,
112: have presented chemical
113: abundances for the narrow lined objects (predominately slow rotators)
114: in the LMC sample. Here we extend the study
115: with the chemical analysis of the faster rotating stars.
116:
117:
118: \section{Observations and model atmosphere analysis}\label{s_obs}
119:
120: High-resolution ($R\sim20\,000$)
121: spectra from the Fibre Large Array Multi-Element Spectrograph (FLAMES) at the European
122: Southern Observatory Very Large Telescope
123: were obtained for some 750 O- and early B-type stars located towards clusters
124: in our Galaxy and the Large and Small Magellanic Clouds (LMC and SMC respectively).
125: A discussion of target selection, observational details and initial data reduction
126: can be found in \citet{eva05}. In this letter we discuss the LMC early B-type
127: stars, using nitrogen as a probe of chemical enrichment.
128:
129: A grid of non-LTE {\sc TLUSTY} model atmospheres
130: \citep{hub95} has been used to calculate the
131: atmospheric parameters and chemical compositions (C, N, O, Mg and Si) of our
132: targets as described
133: in Papers I, II and \citet{hun07b}, hereafter Paper III.
134: We have fitted the lines with
135: rotationally broadened profiles to estimate the equivalent widths (EWs), since at
136: significant rotational velocities ($>$50\,km\,s$^{-1}$) the line shape is rotationally dominated.
137: For those objects in which no nitrogen features could be identified, an upper limit
138: to the nitrogen abundance was estimated by placing an upper limit on the equivalent width
139: of the strongest N\,{\sc ii} line in our spectral range, which is located at 3995\AA\ (Paper I).
140:
141: The mean abundances of C, O, Mg and Si are in excellent agreement (within 0.05\,dex)
142: with the LMC baseline abundances given in Paper I. However the mean nitrogen abundance is
143: 0.36\,dex higher than its baseline abundance indicating that nitrogen enrichment
144: has occurred in many of the stars.
145: Additionally the scatter in the nitrogen abundances is over a factor of two larger
146: than that of the other elements, indicating that different levels of enrichment explain the
147: mean nitrogen abundance, rather than
148: systematic errors. Since nitrogen is an important element in the CNO-cycle,
149: the surface nitrogen enrichments can be used as a measure of the mixing efficiency.
150: Note that although a corresponding carbon depletion
151: of up to $\sim$0.2\,dex
152: would be expected, within the uncertainties in determining carbon abundances
153: (see Paper I), such
154: an effect would be difficult to observe.
155: In Fig.~\ref{f_nvsini}
156: the nitrogen abundances are plotted as a function of the projected rotational velocity with
157: the sample being split into two groups; core hydrogen burning objects with
158: surface gravities $\geq$3.2\,dex and supergiants
159: having surface gravities $<$3.2\,dex (see Paper III).
160:
161: \section{Stellar evolution models} \label{s_models}
162:
163:
164: Comparison with published stellar evolution models such as the Geneva models \citep{mae01}
165: is complicated as no rotating models at the LMC metallicity are currently available.
166: Additionally, the initial chemical composition for evolutionary models is often scaled from
167: solar composition, and this is not appropriate for all elements, in particular nitrogen.
168:
169: New evolutionary models have been calculated (Brott et al. in prep.)
170: using the stellar evolution code of \citet{yoo06},
171: which includes rotation
172: \citep{heg00b} and angular momentum transport by magnetic torques \citep{spr02}.
173: Only two differences apply here:
174: we updated the mass loss prescription and now use the recipe of \citet{vin01}, and we disregard
175: the magnetically induced chemical diffusion term of \citep{spr02},
176: which --- in contrast to the magnetic angular momentum transport --- is not
177: observationally supported, and appears controvercial at present \citep{spr06}.
178:
179: %the mass-loss prescription of \citet{vin01}.
180: %Angular momentum transport is achieved by magnetic fields, however we
181: %decided not to use the transport
182: %of chemical elements by magnetic fields as described in \citep{spr02}.
183: %We cannot constrain the efficiencies of
184: %rotational and magnetic mixing processes simultaneously with only one
185: %set of observations. Therefore, we decided to describe and calibrate
186: %the chemical element transport by the much better understood rotational mixing only.}
187: %%%New evolutionary models
188: %%%have been calculated (Brott et al. in prep.)
189: %%%based upon the stellar evolution code of \citet{yoo06} which
190: %%%includes rotation, angular momentum transport by magnetic fields \citep{spr02}
191: %%%and the mass-loss prescription of \citet{vin01}.
192:
193: As initial composition we adopted the LMC C, N, O, Mg and Si abundances given in Paper I and
194: all the other metal abundances decreased by 0.4\,dex from the solar composition of \citet{asp05}.
195: Based on the recent discussion of the primordial helium abundance \citep{pei07},
196: we have updated the initial helium mass fraction for our LMC models to Y= 25.6\%, which together
197: with the metallicity of the adopted chemical composition (Z= 0.5\%), results in a hydrogen
198: mass fraction of X= 73.9\%.
199:
200: In Paper III we determined
201: that the end of core hydrogen burning occurs at a logarithmic surface gravity of
202: $\sim$3.2\,dex. We find that a convective core overshooting of 0.335
203: pressure scale heights is required to reproduce this result. While this value
204: is larger than what is typically assumed, with consequences
205: for the post-main sequence
206: evolution which still remain to be investigated, it provides the only way to
207: understand the sharp drop in the rotational velocity
208: distribution of our sample at a surface gravity of 3.2\,dex.
209: Evolutionary models neglecting overshooting
210: indicate that the surface nitrogen
211: enrichment is smaller by less than 0.15\,dex at core hydrogen
212: exhaustion for a rapid rotator. Hence, our principal conclusions
213: are not significantly affected by the adopted amount of overshooting.
214:
215: The efficiency of rotationally induced mixing in our models
216: is controlled by the parameter $f_{\rm c}$,
217: which is the ratio
218: of the turbulent viscosity to the diffusion coefficient that describes the transport
219: of angular momentum
220: by rotationally induced hydrodynamic
221: instabilities \citep{heg00b}. The mean projected rotational velocity
222: of the stars shown in Fig.~\ref{f_nvsini} is 110\,km\,s$^{-1}$,
223: with the mean surface nitrogen abundance being 7.2\,dex.
224: Although this mean
225: projected rotational velocity is lower than that observed for cluster stars
226: (see, for example, \citealt{hua06}) there is no bias towards stars with low
227: rotational velocities. Our stars are a relatively unbiased sample of the true populations,
228: as explained in \citet{eva06}. This population is dominated by field stars, which are
229: known to rotate slower than cluster stars, and our velocity distribution is
230: comparable to the Galactic field stars of \citet{abt02}. Additionally Be-type stars
231: are excluded, hence our mean velocity is representative only for normal
232: B-type stars and there is no velocity bias for these B-type stars.
233:
234: The parameter $f_{\rm c}$ has been calibrated to reproduce the mean
235: surface nitrogen abundance of the core-hydrogen burning stars
236: at core hydrogen exhaustion for a 13\,M$_\sun$ model (the mean mass of our non-supergiant stars)
237: initially rotating at 140\,km\,s$^{-1}$ (110$\times$4/$\pi$ to account for random angles of inclination).
238: We find $f_{\rm c}$=2.28$\times$10$^{-2}$, compared to 3.33$\times$10$^{-2}$ adopted previously
239: (\citealt{heg00b}). Using these parameters, we computed a grid of models for masses
240: representative of our sample and a range of initial rotational velocities.
241:
242:
243: \section{Discussion and conclusions}
244:
245: In Fig.~\ref{f_nvsini} the nitrogen abundances are plotted against
246: the projected rotational velocities for our sample stars, and compared with the evolutionary models.
247: Given the large number of objects in our sample, it is reasonable to assume a
248: random distribution of inclination angles,
249: and hence the rotational velocity of the tracks has been scaled by $\pi$/4.
250: Although we have attempted to contrain the evolutionary
251: tracks to the observations, we are able to reproduce agreement for only 60\% of the data
252: (excluding known radial velocity variables).
253: %\textcolor{blue}{\bf Fig.~1a shows that most core hydrogen burning stars in our sample
254: %have a rather low $v \sin i$ ($\lesssim 100$\,km\,s$^{-1}$) and a nitrogen
255: %abundance compatible with no enrichment. }
256:
257: %%%{\bf By contraining the evolutionary
258: %%%tracks to the observations we are able to reproduce good agreement for xx\% of the data.}
259: %are a good fit to the majority of our data, both in
260: %the unenriched low rotational velocity regime as well
261: %as the highly enriched fast rotators. This is simply a consequence of the tracks being calibrated to
262: %the mean of the sample.
263:
264: However, two groups of core hydrogen burning stars in Fig.~\ref{f_nvsini}(a)
265: stand out as being in conflict with the evolutionary models.
266: Group~1 contains fast rotators which have little chemical mixing.
267: In particular, this group includes fast rotating single stars with
268: surface gravities indicating that they are near the end of core hydrogen burning (excluding radial
269: velocity variables this is $\sim$20\% of the
270: core hydrogen burning sample.)
271: The gravities used to discriminate these stars have not been
272: corrected for the fact that we may observed them almost equator-on, i.e.
273: their true polar gravities could be larger, meaning that they are less evolved than our
274: derived gravities imply. However, applying the surface gravity corrections described in \citet{hua06b}
275: would increase the gravities by up to $\sim$0.3\,dex for our faster rotators.
276: As the zero age main-sequence
277: gravity of our 13\,M$_{\sun}$ models is about 4.3\,dex, and since most
278: of the stellar expansion on the main sequence occurs towards the end of core
279: hydrogen burning, this gravity uncertainty can not reconcile the situation.
280:
281: %{\bf However, if we view fast
282: %rotators close to their equatorial plane we would estimate lower gravities than
283: %their evolutionary status would imply. Although it is not possible to obtain
284: %the polar gravity if the angle of inclination is unknown, in order to test the effect
285: %in an average way, we have taken the surface gravity corrections
286: %described in \citet{hua06b} and examined the effect on Fig.~\ref{f_nvsini}. While this
287: %increases the gravities by up to 0.3\,dex for the fast rotators
288: %(i.e. making them younger with less enrichment expected), if we take into account that the
289: %evolutionary status is continuous and not simply split at a gravity of 3.7\,dex as we show
290: %in Fig.~\ref{f_nvsini} the discrepancy between the Group~1 stars (xx\% of the sample)
291: %and the evolutionary tracks
292: %remains valid.}
293:
294: The second discrepant group of stars in Fig.~1a are the 15 slow rotators
295: ($v \sin i < $50\,km\,s$^{-1}$) that show significant nitrogen enrichment (Group~2). This group
296: also forms $\sim$20\% of the non-binary core hydrogen burning sample.
297: Although it could be argued that they may be fast rotators observed
298: pole-on, statistically this is unlikely.
299: In order to reproduce their mean nitrogen abundance requires
300: a rotational velocity of $\sim$200\,km\,s$^{-1}$.
301: Using simple geometry to calculate the solid angle
302: restriction that we must impose, i.e. $4\pi(1-cos\theta)$,
303: %Using the method described in \citet{gre96}
304: and assuming that we need 15 stars
305: to have this velocity, we expect less than one star to appear with
306: such a low rotational velocity due to
307: random inclinations.
308: The simple corollary of this is that if we have 15 stars with
309: velocities $\sim$200\,km\,s$^{-1}$ populating
310: Group~2 through random angles of inclination, over 300
311: similary enriched stars with $v\sin i$ values between 150 and 200\,km\,s$^{-1}$ would be expected.
312: Such a population clearly does not exist.
313:
314: Evolved rapid rotators with low nitrogen enrichment (Group~1) may be produced by close binaries
315: with an initial period small enough to ensure tidal locking and slow rotation but large
316: enough for highly non-conservative mass transfer,
317: i.e. with an amount of mass accreted
318: which is sufficient to spin-up the star but insufficient to enrich it significantly
319: \citep{pet05}. However, for many of the evolved stars in Group~1,
320: there is no indication of binarity.
321: The single star nature for these objects would pose a serious challenge to the
322: theory of rotational mixing.
323:
324: While binaries
325: %which have sufficiently short initial periods to remain tidally locked
326: %after the mass transfer, could also produce stars in
327: might be able to populate Group~2 \citep{lan08}, another explanation may be more likely.
328: \citet{mor06} have analyzed ten slowly rotating Galactic early $\beta$-Cephei B-type stars
329: and found a highly enriched group (four out of ten) of which three have detected
330: magnetic fields. Although pulsations in these stars may
331: cause an apparent surface enrichment, \citet{bou07} report that only slight nitrogen enrichments
332: would be expected. As such, assuming that the effect of pulsations is negligible,
333: there appears to be a correlation between magnetic fields and nitrogen enrichment.
334: Indeed, \citet{wol07} have attributed the large number of slow
335: rotators often seen in massive star populations to
336: magnetic locking of the star to the accretion disk during the star formation process and \citet{ale07}
337: also show a correlation between slow rotation and magnetic fields for magnetic A- and B-type stars.
338: Hence we postulate that the highly enriched slow rotators in the LMC are analogs
339: of the enriched Galactic magnetic stars. This would imply that, independent of metallicity,
340: a significant fraction of early B-type stars are intrinsically magnetic, analogous to
341: the well known situation for lower mass stars. In this context, the
342: identification of three He-rich slowly rotating OVz stars in NGC\,346 \citep{mok06} suggests
343: that the phenomenon of intrinsic magnetic fields in massive stars may not be confined to the
344: B-star regime. If these magnetic fields are of fossil origin,
345: i.e. possessing long-term stability \citep{bra04}, in contrast
346: to the fields produced by the \citep{spr02} mechanism,
347: one may speculate that stars in Group~2, and their Galactic counterparts,
348: might be the progenitors of magnetars, analogous to the suggestion of \citet{fer06}
349: that magnetic white dwarfs evolve from magnetic A, F and late B stars.
350: While the fossil field hypothesis might well explain the slow rotation,
351: the physical process which leads to the enrichment of nitrogen in
352: the stars of Group~2 remains to be identified.
353: %{\bf We performed tests with models that use only magnetic mixing as rotational
354: %mixing is very week at these velocities.
355: %These models are also not able to reproduce the stars observed in Group 2.
356: %We suggest that these stars are mixed by a different magnetic mechanism than the
357: %Taylor-Spruit-Dynamo. }
358:
359:
360: %In Fig.~\ref{f_nvsini}(b) the distribution of nitrogen abundances is plotted for the supergiants. They
361: The nitrogen abundances of the supergiants (Fig.~\ref{f_nvsini}(b))
362: appear to fall into two distinct groups, one group having a level of nitrogen enhancement
363: consistent
364: with that seen for the majority of the core hydrogen burning objects ($<$7.2\,dex; Group 3),
365: and a second having a much greater level of enrichment ($>$7.6\,dex; Group 4). It should be
366: noted that although the Group~4 objects appear to be
367: well fitted by evolutionary tracks with initial rotational
368: velocities of 200-300\,km\,s$^{-1}$,
369: such a rotational velocity distribution for supergiants is inconsistent
370: with that for core-hydrogen burning objects and rotation does not increase the blue-supergiant
371: lifetime. Hence, the enrichments of the supergiants in Group~4 (70\% of the
372: non-binary supergiant sample) are incompatible with the
373: theory of rotational mixing. Additionally it should be noted that the lowest gravity stars
374: ($<$2.8\,dex) are all enriched in nitrogen, implying that some evolutionary process not
375: accounted for in the models is responsible for the enrichment.
376:
377: The simplest way to interprete this would be
378: to characterize Group~3 as pre-red supergiant objects, and
379: Group~4 as post-red supergiant objects.
380: The nitrogen abundances and rotational velocities in Group~4 are indeed consistent with the
381: predictions for a blue loop stage \citep{heg00}. However, we note that our
382: models do not return to the high effective temperatures
383: of the stars in Group~4 and hence their position on the H-R diagram
384: cannot be reproduced. As several of the enriched objects show evidence of binarity,
385: mass transfer may also be important \citep{wel01}.
386:
387: To conclude, our study can not provide unambigous evidence for rotational mixing
388: acting in massive stars. The fast rotators in our sample can be interpreted in two ways.
389: Firstly as rotationally mixed stars, if the relatively non-enriched
390: stars (Group~1) can be explained by
391: binary effects. However, if it were confirmed that these Group~1 stars
392: are single stars, then the enriched rapid rotators may need to be
393: understood as binary products --- and rotational mixing would be inefficient,
394: or much more complex than described by the present-day shellular models.
395: Additionally, the population of intrinsically slow rotators with nitrogen enhancements
396: of up to a factor of $\sim$6 implies that studies of nitrogen enhancements with stellar samples
397: which are restricted to low projected rotational velocities --- which applies to most
398: previous works --- are not suited for constraining, or motivating, the adoption of
399: rotational mixing in massive stars. Finally our supergiant sample implies that
400: these stars cannot be considered as representative of the amount of mixing that occurs
401: during the hydrogen burning main-sequence and hence should not be used to constrain this
402: process.
403: In summary, our study provides a challenge to rotational mixing, and provides evidence
404: for two other processes affecting the rotation and the surface abundances of
405: populations of massive stars, likely binarity and magnetic fields.
406:
407:
408:
409: %To conclude, we have calibrated the evolutionary models to fit the {\bf observations
410: %and xx\% of the data points can be explained in the context of rotational
411: %mixing. }
412: %%observed
413: %%trend of nitrogen abundances in the majority of our LMC sample.
414: %However, the observation
415: %of unenriched fast rotating stars is clearly in conflict with such models. Additionally
416: %a group of intrinsically slow rotators with nitrogen enhancements
417: %of up to a factor of~6, which we interpret as magnetic stars, implies that previous
418: %studies of nitrogen-rich stars with a restriction to low projected rotational velocities
419: %are not suited for constraining rotational mixing. {\bf The nitrogen abundances
420: %of many of the supergiants also cannot be reproduced by the models.}
421: %The analysis
422: %of complete populations, as presented here, is clearly necessary for the
423: %calibration of theoretical models. These results show that while rotational
424: %mixing may occur in the atmospheres of massive stars, it is not the only process
425: %affecting their surface abundances,
426: %and certainly not the most significant at low rotational velocities.
427:
428:
429:
430:
431: \acknowledgments
432:
433: We are grateful to staff from the European Southern Observatory for assistance in
434: obtaining the data, and to STFC (UK Science and Technology Facilities
435: Council) and NWO (Netherlands Science Foundation)
436: for financial support. IH acknowledges financial support from
437: DEL (Dept. of Employment \& Learning
438: Northern Ireland). SJS acknowledges the European Heads of Research Councils and European
439: Science Foundation EURYI Awards scheme and the EC Sixth Framework Programme. We would
440: also like to thank the referee for highly constructive comments which significantly
441: improved this manuscript.
442:
443:
444: {\it Facilities:} \facility{ESO (VLT-FLAMES, 2.2m-WFI)}.
445:
446: \begin{thebibliography}{}
447: \bibitem[Abt et al.(2002)]{abt02} Abt, H.A., Levato, H. \& Grosso, M. 2002,
448: \apj, 573, 359
449: \bibitem[Alecian et al.(2007)]{ale07} Alecian, E., Wade, G.A., Catala, C. et al. 2007,
450: To appear in the proceedings of `SF2A-2007: Semaine de l'Astrophysique Francaise', ed. J. Bouvier,
451: A. Chalabaev, \& C. Charbonnel, astro-ph/0710.1780
452: \bibitem[Asplund et al.(2005)]{asp05} Asplund, M., Grevesse, N.
453: \& Sauval, A.J. 2005, in
454: `Cosmic Abundances as Records of Stellar Evolution and Nucleosynthesis',
455: ed. T.G. Barnes III \& F.N. Bash, ASP Conf. Ser., 336, 25
456: \bibitem[Bourge et al.(2007)]{bou07} Bourge, P.-O., Th\'{e}ado, S. \& Thoul, A. 2007,
457: Communications in Astroseismology, 150, 203
458: \bibitem[Braithwaite \& Spruit(2004)]{bra04} Braithwaite, J. \& Spruit, H.C. 2004, \nat, 431, 819
459: \bibitem[Brott et al.(in prep)]{bro07} Brott, I. et al. 2007, in prep.
460: \bibitem[Daflon et al.(2001)]{daf01} Daflon, S., Cunha, K., Butler, K. \& Smith, V.V. 2001, ApJ, 563 325
461: \bibitem[Dufton et al.(2005)]{duf05} Dufton, P.L., Ryans, R.S.I.,
462: Trundle, C., et al. 2005, \aap, 434, 1125
463: \bibitem[Evans et al. (2006)]{eva06} Evans, C.J., Lennon, D.J.,
464: Smartt, S.J. \& Trundle, C. 2006, \aap, 456, 623
465: \bibitem[Evans et al.(2005, 2006)]{eva05} Evans, C.J., Smartt, S.J.,
466: Lee, J.-K., et al. 2005, \aap, 437, 467
467: \bibitem[Ferrario \& Wickramasinghe(2006)]{fer06} Ferrario, L. \& Wickramsinghe, D. 2006, MNRAS, 367, 1323
468: %\bibitem[Grebel et al.(1996)]{gre96} Grebel, E.K., Roberts, W.J. \& Brandner, W. 1996, A\&A, 311, 470
469: \bibitem[Huang \& Gies(2006a)]{hua06} Huang, W. \& Gies, D.R. 2006a, \apj, 648, 580
470: \bibitem[Huang \& Gies(2006b)]{hua06b} Huang, W. \& Gies, D.R. 2006b, \apj, 648, 591
471: \bibitem[Heger \& Langer(2000)]{heg00} Heger, A. \& Langer, N. 2000, \apj, 544, 1016
472: \bibitem[Heger et al.(2000)]{heg00b} Heger, A., Langer, N. \& Woosley, 2000,\apj, 528, 368
473: \bibitem[Hubeny \& Lanz(1995)]{hub95} Hubeny, I. \& Lanz, T. 1995, \apj, 439, 875
474: \bibitem[Hunter et al.(2007a)]{hun07a} Hunter, I., Dufton, P.L.,
475: Smartt S.J., et al. 2007a, \aap, 466, 277, Paper\,I
476: \bibitem[Hunter et al.(2007b)]{hun07b} Hunter, I., Lennon, D.J.,
477: Dufton, P.L., et al. 2007b, \aap, submitted, Paper III
478: \bibitem[Korn et al.(2002)]{kor02} Korn, A.J., Keller, S.C., Kaufer, A., et al.
479: 2002, A\&A, 385, 143
480: \bibitem[Langer (2008)]{lan08} Langer, N. 2008, in: IAU~Symp.~250, in press
481: \bibitem[Maeder \& Meynet(2001)]{mae01} Maeder, A. \& Meynet, G.
482: 2001, \aap, 373, 555
483: \bibitem[Meynet \& Maeder(2005)]{mey05} Meynet, G. \& Maeder, A. 2005, \aap, 429, 581
484: \bibitem[Mokiem et al.(2006)]{mok06} Mokiem, M.R., de Koter, A., Evans, C.J., et al. 2006, \aap, 456, 1131
485: \bibitem[Morel et al.(2006)]{mor06} Morel, T. Butler, K., Aerts, C.,
486: Neiner, C. \& Briquet, M. 2006, \aap, 457, 651
487: \bibitem[Peimbert et al.(2007)]{pei07} Peimbert, M., Luridiana, V. \& Peimbert, A. 2007, \apj, astro-ph/0701580
488: \bibitem[Petrovic et al.(2005)]{pet05}Petrovic, J., Langer, N., van der Hucht, K.A. 2005, \aap, 435, 1013
489: \bibitem[Spruit(2002)]{spr02} Spruit, H.C. 2002, \aap, 381, 923
490: \bibitem[Spruit(2006)]{spr06} Spruit, H.C. 2006, astro-ph/0607164
491: \bibitem[Trundle et al.(2007)]{tru07} Trundle, C., Dufton, P.L.,
492: Hunter, I., et al. 2007, \aap, 471, 625
493: \bibitem[Venn(1999)]{ven99} Venn, K.A. 1999, ApJ, 518, 405
494: \bibitem[Vink \& de Koter(2005)]{vin05} Vink, J.S. \& de Koter 2005, A\&A, 442, 587
495: \bibitem[Vink et al.(2001)]{vin01} Vink, J.S., de Koter, A. \& Lamers, H.J.G.L.M 2001, A\&A,
496: 369, 574
497: \bibitem[Wellstein et al.(2001)]{wel01} Wellstein, S., Langer, N.
498: \& Braun, H. 2001, \aap, 369, 939
499: \bibitem[Wolff et al.(2007)]{wol07}
500: Wolff, S.C, Strom, S.E., Dror, D. \& Venn, K. 2007, AJ, 133, 1092
501: %\bibitem[Woosley \& Heger(2006)]{woo06} Woosley, S.E. \& Heger, A. ApJ, 637, 914
502: \bibitem[Yoon \& Langer(2005)]{yoo05} Yoon, S.-C. \& Langer, N. 2005, \aap, 443, 643
503: \bibitem[Yoon et al.(2006)]{yoo06} Yoon, S.-C., Langer, N. \& Norman, C. 2006, \aap, 460, 199
504: \end{thebibliography}
505:
506:
507: \newpage
508:
509: \begin{figure*}[ht]
510: %\epsfig{file=f1.eps, height=75mm, angle=0}
511: \plotone{f1.eps}
512: \caption{Nitrogen abundance (12 + log [N/H]) against the projected
513: rotational velocity ($v\sin i$) for core hydrogen burning (a) and supergiant (b) objects.
514: Open symbols: radial velocity variables; downward arrows: abundance upper limits;
515: dotted line: LMC baseline nitrogen
516: abundance. The mean uncertainty in the nitrogen abundance is 0.25\,dex while that in $v\sin i$ is
517: 10\% and these are illustrated. These errors are largely independant
518: of rotational velocity since the systematic uncertainties are comparable to the
519: measurement errors.
520: The bulk of the core hydrogen burning objects occupy a region at low
521: $v\sin i$ and show little or modest nitrogen enrichment.
522: The tracks are computed
523: for an initial mass of 13\,M$_\sun$ (a) and 19\,M$_\sun$ (b),
524: corresponding to the average mass of our non-supergiant and supergiant stars, respectively,
525: and their rotational velocity has been multiplied by $\pi$/4. Although the plot
526: contains stars with a range of masses, comparison of the tracks shown in (a) and (b)
527: show that any mass effect is negligible compared to the abundance uncertainties.
528: In (a) the surface gravity has been used as indicator of the evolutionary status and the objects
529: (see legend) and tracks have been split into red and blue to
530: indicate younger and older stars respectively. However, this is
531: illustrative only, since the evolutionary status is obviously continuous and not discrete.
532: Gray shading in panel~(a) highlights two groups of stars which remain unexplained
533: by the stellar evolution tracks. In panel~(b), gray shading highlights the apparent
534: division of the supergiants into two distinct groups. The
535: surface gravity estimates of many
536: of the objects in Group~3, and the two apparantly rapidly rotating unenriched
537: supergiants, are consistent
538: with being in the core hydrogen phase within their uncertainties. }
539:
540:
541: \label{f_nvsini}
542: \end{figure*}
543:
544:
545:
546:
547: \end{document}
548: