1:
2: \documentclass[12pt,preprint]{aastex}
3: %\documentclass[10pt,manuscript]{aastex}
4: \usepackage{graphicx}
5:
6: \newcommand\lsim{\lower0.5ex\hbox{$\; \buildrel < \over \sim \;$}}
7: \newcommand\gsim{\lower0.5ex\hbox{$\; \buildrel > \over \sim \;$}}
8: \newcommand{\nus}{\mbox{$\nu_{\rm s}$}}
9: \newcommand{\nul}{\mbox{$\nu_1$}}
10: \newcommand{\nuu}{\mbox{$\nu_2$}}
11: \newcommand{\dnu}{\mbox{$\Delta\nu$}}
12: \newcommand{\rin}{\mbox{$R_{\rm in}$}}
13: \newcommand{\ro}{\mbox{$R_0$}}
14: \newcommand{\win}{\mbox{$\omega_{\rm in}$}}
15: \newcommand{\ws}{\mbox{$\omega$}}
16: \newcommand{\bz}{\mbox{$B_{\rm z}$}}
17: \newcommand{\bfi}{\mbox{$B_{\phi}$}}
18:
19: \shorttitle{A Model for Twin kHz QPOs in NS LMXBs}
20: \shortauthors{Li \& Zhang}
21:
22:
23: \begin{document}
24: \title
25: {A MODEL FOR TWIN KILOHERTZ QUASI-PERIODIC OSCILLATIONS IN NEUTRON
26: STAR LOW-MASS X-RAY BINARIES}
27:
28: \author
29: {X.-D. Li\altaffilmark{1} and C.-M. Zhang\altaffilmark{2}}
30:
31: \affil{$^1$Department of Astronomy, Nanjing University,
32: Nanjing 210093, China; lixd@nju.edu.cn}
33: \affil{$^2$National Astronomical Observatories, Chinese Academy of
34: Sciences, Beijing 100012, China; zhangcm@bao.ac.cn}
35:
36: \date{}
37:
38:
39: %\maketitle
40:
41: \begin{abstract}
42: We suggest a plausible interpretation for the twin kiloHertz
43: quasi-periodic oscillations (kHz QPOs) in neutron star low-mass
44: X-ray binaries. We identify the upper kHz QPO frequencies to be
45: the rotational frequency and the lower kHz QPOs the standing kink
46: modes of loop oscillations at the inner edge of the accretion
47: disk, respectively. Taking into account the interaction between
48: the neutron star magnetic field and the disk, this model naturally
49: relates the twin QPO frequencies with the star's spin frequencies.
50: We have applied the model to four X-ray sources with kHz QPOs
51: detected simultaneously and known spin frequencies.
52:
53: \end{abstract}
54:
55: \keywords{accretion, accretion disk---stars: neutron---stars:
56: magnetic field---binaries: close---X-ray: stars}
57:
58: \section{INTRODUCTION}
59: The kiloHertz quasi-periodic oscillations (kHz QPOs) have been
60: measured in more than twenty neutron star low-mass X-ray binaries
61: (NS LMXBs) in their persistent emission with the {\em Rossi X-ray
62: Timing Explorer} ({\em RXTE}), which offered unique insights
63: into the physics of strong gravity and dense matter (see van der
64: Klis 2000, 2005 for reviews). In many cases the twin kHz QPOs
65: appear simultaneously and the correlations between the pair
66: frequencies have been investigated extensively (e.g., Psaltis et
67: al. 1998, 1999b; Belloni et al. 2005).
68: %
69: Moreover, their frequencies also follow rather tight correlations
70: with other timing features of the X-ray emission (Ford \& van der
71: Klis 1998; Psaltis et al. 1999a; Belloni et al. 2002).
72: %
73: There is currently no consensus as to the origin of these QPOs,
74: nor on what physical parameters determine their frequencies, which
75: have been identified with various characteristic frequencies in
76: the inner accretion flow (see e.g. Stella \& Vietri 1999;
77: Osherovich \& Titarchuk 1999; Lamb \& Miller 2001; Abramowicz et
78: al. 2003).
79:
80: With the discovery of the twin kHz QPOs in the accretion-powered
81: millisecond pulsar SAX J1808.4$-$3658, it was found that the
82: frequency separation $\dnu$ is almost half the spin frequency
83: (Wijnands et al. 2003). For other sources with detected spin
84: frequencies $\nus$ from the burst oscillations, $\dnu$ are shown
85: to be close to either the spin frequencies or half of them (van
86: der Klis 2005 and references therein).
87: %
88: %One of the puzzling issues related to the twin kHz QPOs is that,
89: %the peak separation $\dnu$ of the QPOs for a given source seems to
90: %be equal to or half the frequency of the nearly coherent
91: %oscillations observed during type I X-ray bursts that is thought
92: %to be the spin frequency $\nus$ of the neutron star,
93: %
94: These findings seem to hint some underlying mechanisms relating
95: $\nus$ to the upper and lower kHz QPO frequencies ($\nuu$ and
96: $\nul$). However, the more detailed measurements show that $\dnu$
97: is generally inconsistent with a constant value of $\nus$, but
98: varying with $\nuu$ or $\nul$ (van der Klis 2000, 2005 and
99: references therein), which cast doubts about the validity of the
100: simple beat-frequency interpretation (Miller, Lamb \& Psaltis
101: 1998; see Lamb \& Miller 2001 for a modified version). Osherovich
102: \& Titarchuk (1999) suggested the lower kHz QPO frequency $\nul$
103: to be the Keplerian frequency in the disk and the higher kHz QPO
104: frequency $\nuu$ the hybrid between $\nul$ and $2\nus$. Klu\'zniak
105: et al. (2004) showed that the twin kHz QPOs can be explained by a
106: nonlinear resonance in the epicyclic motion in the accretion
107: disks, which can lead to the 3:2 ratio for the two main resonances
108: (see also Abramowicz \& Klu\'zniak 2001; Abramowicz et al. 2003).
109:
110: %Osherovich \& Titarchuk (1999) suggested a possible relation between $\dnu$ and $\nus$.
111:
112:
113: In this paper we propose an alternative interpretation for the
114: origin of the twin kHz QPOs, by considering the interaction
115: between the neutron star magnetic field and the surrounding
116: accretion disk. We introduce the model in \S 2, and present its
117: applications to several NS LMXBs with the simultaneously detected
118: kHz QPOs and known spin frequencies in \S 3. The possible physical
119: implications and conclusions are given in \S 4.
120:
121:
122: \section{MODEL}
123: Neutron stars in LMXBs generally accretes via an accretion disk.
124: For most part of the disk, the plasma rotates in a Keplerian
125: orbit. Close to the NS surface, the stellar magnetic fields begin
126: to truncate the disk and control the motion of the plasma,
127: resulting in a non-Keplerian boundary layer lying between the
128: magnetosphere corotating with the star and the outer Keplerian
129: disk. We assume that the boundary layer is confined by the inner
130: and outer radii, $\rin$ and $\ro$ respectively. As conceivable,
131: the plasma corotates with the magnetosphere at $\rin$, and the
132: plasma's motion begins to deviate from Keplerian rotation and take
133: its maximum value at $\ro$ (see Fig.~1).
134:
135: As for the construction of the model, {\em we identify the upper
136: kHz QPO frequency $\nuu$ to be the rotational frequency at $\ro$},
137: i.e.,
138: \begin{equation}
139: \nuu\equiv\nu(\ro)\equiv\xi\nu_{\rm K}(\ro),
140: \end{equation}
141: where $\nu_{\rm K}(\ro)$ is the Keplerian rotation frequency at
142: $\ro$, and $0<\xi\lsim 1$. The value of $\xi$ depends on the
143: rotational frequency distribution inside the boundary layer.
144: %and
145: %hence on the spin frequency, magnetic field strength and mass
146: %accretion rate of the NS.
147: Unfortunately, there is no analytic solution to the structure of
148: the boundary layer, and the value of $\xi$ can be evaluated only
149: through numerical calculations.
150: %
151: However, based on the qualitative characteristics of the rotation
152: rate in the boundary layer, Campbell (1987) has suggested the
153: following form for the disk rotation profile close to the
154: magnetosphere
155: \begin{equation}
156: \nu(R)=\nu_{\rm K}(R)-\nu_{\rm K}(\rin)(1-\win)
157: \exp[\frac{3(R/\rin-1)}{2(1-\win)}],
158: \label{nur}
159: \end{equation}
160: where $\win=\nus/\nu_{\rm K}(\rin)$.
161: %
162: When $R\gg\rin$, $\nu(R)\rightarrow\nu_{\rm K}(R)$; when
163: $R\rightarrow\rin$, $\nu(R)\rightarrow\nus$.
164: %
165: Equation (2) gives a reasonable description of $\nu(R)$ close to
166: $\rin$, but our interest here focuses on the rotational behavior
167: around $\ro$, at which Eq.~(2) fails to be valid for a
168: sufficiently wide range of $\win$.
169: %
170: Instead, we take a slightly modified version of Eq.~(2) to
171: account for disk rotation,
172: \begin{equation}
173: \nu(R)=\nu_{\rm K}(R)-\nu_{\rm K}(\rin)(1-\win)
174: \exp[\frac{2(R/\rin-1)}{(1-\win)}].
175: \label{nur2}
176: \end{equation}
177: %
178: As an illustration, Fig.~1 shows $\nu(R)/\nu_{\rm K}(\rin)$
179: against $R/\rin$ for three selected values of $\win=0.2$, 0.5 and
180: 0.8 in solid curves. The dashed curve corresponds to Keplerian
181: rotation for comparison.
182: %
183: Although the hypothesized form (3) is phenomenological, it shares
184: the main features with the numerically calculated results (e.g.,
185: Ghosh \& Lamb 1979), and the following analysis also indicates
186: that its adequate for the disk rotation.
187: %
188:
189: From Eq.~(3) we can determine the location of $\ro$ using the
190: condition ${\rm d}\nu/{\rm d}R=0$ at $\ro$, and obtain the values
191: of $\xi\equiv\nu(\ro)/\nu_{\rm K}(\ro)$ for different values of
192: the so-called fastness parameter $\ws\equiv\nus/\nu_{\rm K}(\ro)$.
193: Figure 2 shows $\xi$ as a function of $\ws$.
194: %Equation (1) can then be written as
195: %\begin{equation}
196: %\nuu=(\xi/\ws)\nus\;.
197: %\label{nuu}
198: %\end{equation}
199: %
200:
201: Before discussing the origin of the lower kHz QPOs, we note that
202: in solar physics the QPOs of several minute periods in coronal
203: loops have been detected and successfully interpreted as the
204: standing kink modes of MHD waves (Aschwanden et al. 1999;
205: Nakariakov et al. 1999). Thus, following the similar ideas in
206: treating the QPOs in the coronal loops, we suppose that these
207: MHD waves could also be excited at the inner edge of the accretion
208: disks, where the poloidal field lines are dragged along the
209: azimuthal direction due to shear motion between the star and the
210: disk, and various types of MHD instabilities take place (Ghosh \&
211: Lamb 1979). The reconnection of the azimuthal magnetic field lines
212: could result in circular flux tubes (loops) in the boundary layer,
213: where plasma is confined along a magnetic field line with some
214: cross section. It is well known that the coronal loops may be set
215: into oscillations with various modes leading to brightness
216: oscillations (Roberts 2000 for a review). While the similar
217: oscillations may also occur in the accretion disks, generated by
218: MHD turbulence, here we focus on the fast kink mode of standing
219: MHD waves. These fast waves arise as a free mode only for high
220: density loops (Edwin \& Roberts 1982), which may be appropriate in
221: accretion disks in NS LMXBs. Assume the loop length being the
222: circumference of the magnetosphere, $2\pi\ro$, the oscillation
223: frequency is
224: \begin{equation}
225: \nu=\frac{Nc_{\rm k}}{4\pi\ro},
226: \end{equation}
227: where $c_{\rm k}$ is the kink speed. The wave number
228: $k=N\pi/(2\pi\ro)=N/(2\ro)$. The integer $(N-1)$ stands for the
229: node number of the vibration along the tube axis,
230: with $N=1$ being the principal mode. At the inner edge of
231: the accretion disk, the sound speeds are much smaller than the
232: azimuthal Alfv\'en speeds $c_{\rm A\phi}$, and the plasma inside
233: a loop is much denser than its surroundings for the comparable
234: magnetic field strengths, therefore we have the kink speed
235: $c_{\rm k}\simeq \sqrt{2}c_{\rm A\phi}$ (Roberts 2000),
236: and in turn the frequency of a standing
237: kink mode
238: \begin{equation}
239: \nu=\frac{Nc_{\rm A\phi}}{2\sqrt{2}\pi\ro}.
240: \end{equation}
241:
242: In the following, we estimate the azimuthal field strength at
243: $\ro$. We assume that the stellar magnetic field lines are
244: initially dipolar and penetrate the accretion disk. The
245: differential rotation between the star and the disk generates the
246: azimuthal field component $\bfi$ from the vertical component $\bz$
247: (Ghosh \& Lamb 1979).
248: %
249: According to Wang (1995), if the growth of $\bfi$ is limited by
250: the diffusive decay produced by the turbulent mixing within the
251: disk, $\bfi$ is given by\footnote{Other processes related to the
252: dissipation of the magnetic field give similar expressions of
253: $\bfi$ (Wang 1995).}
254: \begin{equation}
255: \bfi(R)=\gamma\frac{\nu(R)-\nus}{\nu(R)}\bz(R),
256: \end{equation}
257: %
258: where the parameter $\gamma\sim 1$ (Aly 1984; Uzdensky,
259: K$\ddot{o}$nigl, \& Litwin 2002).
260:
261: We assume that the magnitude of $\ro$ is close to that of the
262: Alf\'ven radius where the magnetic energy density equals the total
263: kinetic energy density (Davidson \& Ostriker 1973), i.e.,
264: \begin{equation}
265: \frac{\bz(R_0)^2}{8\pi}=\frac{1}{2}\eta\rho v_{\rm K}^2|_{R_0},
266: \end{equation}
267: where $\rho$ is the mass density, and $\eta\sim 1$.
268: %
269: Combining Eqs.~(6) and (7) we get the azimuthal Alf\'ven speed at
270: $\ro$,
271: \begin{equation}
272: c_{\rm A\phi}(\ro)=\gamma\eta^{1/2}\frac{(\nuu-\nus)}{\nuu}v_{\rm
273: K}(\ro).
274: \end{equation}
275:
276: %\begin{equation}
277: %\nu=N\gamma(\frac{\eta}{2})^{1/2}\frac{(\nuu-\nus)}{\nuu}\nu_{\rm
278: %K}(\ro).
279: %\end{equation}
280:
281: Furthermore, {\em we suggest the lower kHz QPOs to be the
282: principal fast kink mode of the standing MHD waves along the
283: $\bfi$ field lines at $\ro$}. Inserting Eq.~(8) into Eq.~(5) we
284: find the frequency to be
285: \begin{equation}
286: \nul=\gamma(\frac{\eta}{2})^{1/2}\frac{(\nuu-\nus)}{\nuu}\nu_{\rm
287: K}(\ro).
288: \end{equation}
289: Combining Eqs.~(1) and (9) we have
290: \begin{equation}
291: \nul=(\frac{\alpha}{\xi})(\nuu-\nus),
292: \end{equation}
293: where $\alpha=\gamma(\eta/2)^{1/2}\sim 1$ is taken as a free
294: parameter that absorbs the uncertainties in determining $\ro$ and
295: $\bfi(\ro)$. Since the $\xi(\ws)$ relation is known from Fig.~2,
296: Eq.~(10) suggests a unique relation between $\nul$ and $\nuu$ for
297: given values of $\nus$.
298:
299: Obviously, the MHD oscillations represent the continuum, i.e.,
300: the oscillation frequency is a continuous function of a wave
301: number. We ascribe the lower kHz QPOs to be the principal fast
302: kink mode with the loop length of $L=2\pi\ro$ due to the following
303: facts. Firstly, the global modes with $N=1$ or 2 are easiest to
304: excite in the loops (Roberts, Edwin, \& Benz 1984). Secondly, for
305: a standing wave the decay time of the MHD oscillations is
306: $\tau\propto (L/N)^2$ (Roberts 2000). Apparently, the oscillations
307: with the longer $L$ and smaller $N$ last the longer lifetime, and
308: hence are more easily to be detected. Thus, for the fast kink mode
309: at $\ro$, oscillations along $\bfi$ have the maximum loop length
310: and smallest wave number when $L=2\pi\ro$ and $N=1$.
311: %
312: There may exist the other oscillation modes (e.g. the fast sausage
313: mode of standing fast MHD waves, standing slow MHD waves, and
314: propagating MHD waves) in the accretion disks, and some of them
315: could be also shown as some kinds of QPOs. However, their
316: oscillation frequencies are either too high or too low compared to
317: those of the kHz QPOs in NS LMXBs. Additionally it seems that the
318: kink mode is most likely to be seen as a standing wave (Roberts
319: 2000) .
320:
321: \section{COMPARISON WITH OBSERVATIONS}
322: We have compared the predicted $\nul-\nuu$ relation with the
323: observations of kHz QPOs in four NS LMXBs (4U1608$-$52,
324: 4U1636$-$53, 4U1728$-$34, and 4U1915$-$05), in which both the
325: spin and twin kHz QPO frequencies have been measured (data were
326: provided by T. Belloni, M. M\'endez and D. Psaltis). In Fig.~3 the
327: crosses represent the measurements and the solid lines stand for
328: the theoretical relations. The spin frequency and the adopted
329: value of $\alpha$ for each source are also displayed. In all cases
330: the value of $\alpha$ is of order of unity as expected.
331: %
332: The theoretical predictions and the measured kHz QPO data match
333: quite well for 4U 1728$-$34 and 4U 1915$-$05, albeit the
334: approximated consistence is obtained for 4U 1608$-$52 and 4U
335: 1636$-$53, which may be caused by the fact that the real magnetic
336: field structure in these sources may be more complicated than in
337: the simplified model considered here.
338:
339:
340: \section{DISCUSSION AND CONCLUSIONS}
341: In this paper we have presented a plausible mechanism for the
342: production of twin kHz QPOs in NS LMXBs, invoking the magnetic
343: field-accretion disk interaction. Although NSs in LMXBs are
344: thought to have weak magnetic fields ($\sim10^{8-9}$ G), their
345: influence on the disk rotation has not been paid much attention in
346: the existed works on this subject. Titarchuk, Lapdius, \& Muslimov
347: (1998) has already suggested that a shock occurs in the transition
348: layer where the Keplerian disk adjust to sub-Keplerian flow. The
349: disk can undergo various types of oscillations under the influence
350: of the gas, radiation, magnetic pressure and gravitational force.
351: As to the origin of the oscillations, the boundary layer in our
352: work is similar to but closer to the NS compared with the
353: centrifugal barrier region in Titarchuk et al. (1998).
354: %Actually even for the non-magnetic
355: %accreting NSs or white dwarfs, it has been shown that the disk
356: %rotation becomes non-Keplerian when the disk extends close to the
357: %stellar surface (e.g. Kley 1991).
358:
359: One of the twin kHz QPO frequencies is usually interpreted as the
360: Keplerian rotation frequency at some preferred radii, most likely
361: the inner edge of the disk (e.g., Miller at al. 1998; Titarchuk \&
362: Osherovich 1999). We have followed this idea, but suggested that
363: the real (non-Keplerian) rotation at $\ro$ leads to the upper kHz
364: QPOs. Unfortunately, since there is no general analytic form for
365: the rotational profile within the boundary layer, we have to use a
366: phenomenological description for the rotation of disk plasma
367: around the inner edge\footnote{Even if we take $\xi=1$, that is,
368: Keplerian rotation at $\ro$, Eq.~(10) still fits the observational
369: data fairly well.}. Although it seems to be adequate, its validity
370: and accuracy should be testified more carefully in the future by
371: both observational and theoretical investigations.
372:
373: Moreover, we interpret the lower kHz QPOs to be the fast kink
374: modes of MHD oscillations in loops along the $\bfi$ field lines at
375: $\ro$, shared with the similar physical mechanism for coronal loop
376: oscillations. Since $\bfi$ is generated by $\bz$ through shear
377: motion between the NS and the disk, this naturally links the twin
378: kHz QPOs with the stellar spin. But it is distinct from the
379: traditional beat-frequency model by the fact that in our work the
380: twin OPOs originate from different physical processes. We note
381: that similar hypothesis was discussed by Muslimov (1995), who
382: suggested that the QPOs in LMXBs may be caused by the excitation
383: of the so-called nonlinear global Alfv\'en modes in the boundary
384: layer plasma. It is interesting that these modes were observed in
385: the studies of an ideal MHD spectrum of a toroidal plasma
386: (Goedbloed 1975). In addition, the detailed numerical
387: investigation of these modes was performed by Appert et al.
388: (1982), and their existence was confirmed experimentally by Behn
389: et al. (1984) and Evans et al. (1984).
390:
391: %Since the dynamics of the oscillations in the boundary layer is
392: %similar to that described in Titarchuk et al. (1998), we expect
393: %that the lower kHz QPOs may have relatively high quality factor.
394: %However,
395: Recent analysis by Barret, Olive, \& Miller (2005) showed that the
396: quality factor for the lower kHz QPOs in 4U 1636$-$53 increases
397: with frequency up to a factor of $\simeq 200$ when $\nul\simeq
398: 850$ Hz, then drops at higher frequencies. A ceiling of the lower
399: kHz QPO frequency at 920 Hz is also seen. In the frame of the
400: present work, we ascribe these features to the evolution of the
401: twisted field lines. Several theoretical studies of the star-disk
402: interaction (e.g. Aly 1984, 1985; Uzdensky et al. 2002) have shown
403: that as a dipole field is twisted due to differential rotation,
404: the field lines inflate and effectively open up when a critical
405: twist angle is attained (Uzdensky et al. 2002). This limits the
406: azimuthal pitch at $\ro$,
407: $|\bfi(\ro)/\bz(\ro)|=\gamma(\nuu-\nus)/\nuu$, to some critical
408: value, say, $\gamma_{\rm c}$, demonstrating that a steady state of
409: configuration could be established only if the rotation shear
410: $\nuu-\nus$ is small enough. These arguments suggest that there
411: may exist a maximum value of $\nuu$, beyond which the $\bfi$ field
412: becomes unstable, resulting in decreasing quality factor of the
413: lower kHz QPOs at higher frequencies and a saturation frequency of
414: $\nul$ when most of the field lines become open.
415:
416: We have presented a qualitative description of the kHz QPO
417: production mechanism and a crude quantitative expression of the
418: kHz QPO frequencies, to interpret the observed kHz QPO phenomena.
419: As a preliminary exploration, many physical details have not been
420: considered, such as in what condition and how much MHD wave energy
421: is produced to account for the observed Fourier power spectrum of
422: kHz QPOs. These should be investigated more carefully in the
423: future work.
424:
425: \acknowledgements This work was supported by NSFC under grant
426: number 10025314 and MSTC under grant number NKBRSF G19990754.
427: %
428: We are grateful for T. Belloni, M. M\'endez and D. Psaltis for
429: providing the QPO data, and T. P. Li and P. F. Chen for helpful
430: discussions.
431: %
432: %Thanks are also due to the language assistance and helpful
433: % suggestions by D. Menezes and P. Moodie.
434: The authors express thanks to an anonymous referee for the
435: critical comments that greatly helped improve the manuscript.
436:
437: \begin{thebibliography}{}
438:
439: \bibitem[2003]{ab03}
440: Abramowicz, M.~A., Bulik, T., Bursa, M., \& Klu\'zniak, W. 2003,
441: A\&A, 404, L21
442:
443: \bibitem[2001]{ab01}
444: Abramowicz, M.~A. \& Klu\'zniak, W. 2001, A\&A, 374, L19
445:
446: \bibitem[1984]{al84}
447: Aly, J. J. 1984, ApJ, 283, 349
448:
449: \bibitem[1985]{al85}
450: Aly, J. J. 1985, A\&A, 143, 19
451:
452: \bibitem[1982]{ap82}
453: Appert, K., Gruber, R., Troyon F., \& Vaclavik, J. 1982, Plasma
454: Phys., 24, 1147
455:
456: \bibitem[1999]{as99}
457: Aschwanden, M. J. et al. 1999, \apj, 520, 880
458:
459: \bibitem[2005]{ba05}
460: Barret, D., Olive, J.-F., \& Miller, M. C. 2005, MNRAS, 361, 855
461:
462: \bibitem[1984]{be84}
463: Behn, R. et al.
464: %de Chambrier, A., \& Collins1, G. A. et al.
465: 1984, Plasma Phys. Controlled Fusion, 26, 173
466:
467: \bibitem[2002]{be02}
468: Belloni, T., Psaltis, D., \& van der Klis, M., 2002, \apj, 572,
469: 392.
470:
471: \bibitem[2005]{be05}
472: Belloni, T., Mendez, M. \& Homan, J. 2005, A\&A, 457, 209
473: %The distribution of kHz QPO frequencies in bright LMXBs
474:
475: \bibitem[1987]{ca87}
476: Campbell, C. G. 1987, MNRAS, 229, 405
477:
478: \bibitem[1973]{da73}
479: Davidson, K. \& Ostriker, J. P. 1973, ApJ, 179, 585
480:
481: \bibitem[1982]{ed92}
482: Edwin, P. M. \& Roberts, B. 1982, Sol. Phys. 76, 239
483:
484: \bibitem[1984]{ev84}
485: Evans T. E. et al. 1984, in Proc. of the Fourth Joint
486: Varenna-Grenoble Internat. Symp. on Heating in Toroidal Plasma,
487: Rome, 1, 121
488:
489: \bibitem[1998]{fo98}
490: Ford, E. C. \& van der Klis, M. 1998, ApJ, 506, L39
491:
492: \bibitem[1979]{gh79}
493: Ghosh, P. \& Lamb, F. K. 1979, ApJ, 232, 259
494:
495: \bibitem[1975]{go75}
496: Goedbloed, J. P. 1975, Phys. Fluids, 18, 1258
497:
498: %\bibitem[1991]{kl91}
499: %Kley, W. 1991, A\&A, 247, 95
500:
501: \bibitem[2004]{kl04}
502: Klu\'zniak, W., et al. 2004, ApJ, 603, L89
503:
504: %\bibitem[Lamb, Pethick, \& Pines(1973)]{lam73}
505: %Lamb, F. K., Pethick, C. J., \& Pines, D. 1973, ApJ, 184, 271
506:
507: %\bibitem{}
508: %Lamb, F. K., Shibazaki, N., Alpar, M. A., \& Shaham, J. 1985,
509: %Nature, 317, 681
510:
511: \bibitem[2001]{la01}
512: Lamb, F. K., \& Miller, M. C. 2001, \apj, 554, 1210
513: % in one framework, why \mus or \mus/2
514:
515: %\bibitem[M\'endez \& van der Klis(1999)]{mk99}
516: %M\'endez, M. \& van der Klis, M. 1999, ApJ, 517, L51
517:
518: %\bibitem[M\'endez et al.(1999)]{m99}
519: %M\'endez, M., van der Klis, M. Ford, E. C., Wijnands, R., \& van
520: %Paradijs, J. 1999, ApJ, 511, L49
521:
522: \bibitem[1998]{mi98}
523: Miller, M.\ C., Lamb, F.\ K., \& Psaltis, D.\ 1998, \apj, 508, 791
524:
525: %\bibitem{}
526: %Muno, M.P. 2004, ``Millisecond Oscillations During Thermonuclear
527: %X-ray Bursts", review article for ``X-Ray Timing 2003: Rossi and
528: %Beyond", eds. P. Kaaret, F. K. Lamb, \& J. H. Swank (Melville, NY:
529: %American Institute of Physics)
530:
531: \bibitem[1995]{mu95}
532: Muslimov, A. G. 1995, in Physics of Neutron Stars, ed. A. M.
533: Kaminker, Nova Science Publishers, Inc., p. 277
534:
535: \bibitem[1999]{na99}
536: Nakariakov, V. M. et al. 1999, Science, 285, 862
537:
538: \bibitem[1999]{os99}
539: Osherovich, V. \& Titarchuk, L. 1999, ApJ, 522, L113
540:
541: %\bibitem{}
542: %Osherovich, V. \& Titarchuk, L. 1999b, ApJ, 523, L73
543:
544: \bibitem[1998]{ps98}
545: Psaltis, D. et al.
546: %Mendez, M., Wijnands, R., Homan, J., \& Jonker, P.,
547: %van der Klis, M., Lamb, F.\,K., Kuulkers, E., van Paradijs,
548: %J., \& Lewin, W.\,H.\,G.\
549: 1998, \apj, 501, L95
550:
551: \bibitem[1999a]{ps99a}
552: Psaltis, D., Belloni, T. \& van der Klis, M. 1999a, ApJ, 520, 262
553:
554: \bibitem[1999b]{ps99b}
555: Psaltis, D. et al.
556: %Wijnands, R., Homan, J., Jonker, P., \& van der Klis,
557: %M., Miller, M.C., Lamb, F.\,K., Kuulkers, E., van Paradijs,
558: %J., \& Lewin, W.\,H.\,G.\
559: 1999b, \apj, 520, 763
560:
561: \bibitem[2000]{ro00}
562: Roberts, B. 2000, Sol. Phys., 193, 139
563:
564: \bibitem[2000]{ro84}
565: Roberts, B., Edwin, P. M,, \& Benz, A. O. 1984, \apj, 279, 857
566:
567: \bibitem[1999]{st99}
568: Stella, L., \& Vietri, M., 1999, \prl, 82, 17 %(SV99)
569:
570: %\bibitem[Swank 2004]{sw04}
571: %Swank, J. 2004, to appear in the proceedings of ``X-Ray Timing
572: %2003: Rossi and Beyond", eds. P. Kaaret, F.K. Lamb, \& J.H. Swank
573: %(Melville, NY: AIP) (astro-ph/0402511)
574: %25 NS LMXBs kHz QPOs from Low-Mass X-Ray Binaries with Neutron Stars
575:
576: \bibitem[1998]{ti98}
577: Titarchuk, L., Lapidus, I., \& Muslimov, A. 1998, \apj, 499, 315
578:
579: \bibitem[1999]{ti99}
580: Titarchuk, L., \& Osherovich, V. 1999, ApJ, 518,
581:
582: %# Titarchuk, L., & Osherovich, V. & Kuznetsov, S. 1999, ApJ, 525, L129
583:
584: %\bibitem{}
585: %Titarchuk, L. \& Wood, K.S. 2002, ApJ, 577, L23
586:
587: \bibitem[2002]{uz02}
588: Uzdensky, D. A., K$\ddot{o}$nigl, A., \& Litwin, C. 2002, ApJ,
589: 565, 1191
590:
591: \bibitem[2000]{vk00}
592: van der Klis, M. 2000, ARA\&A, 38, 717
593:
594: \bibitem[2005]{vk05}
595: van der Klis, M. 2005, %A review of rapid X-ray variability in X-ray binaries
596: to appear in Compact stellar X-ray sources, eds. W. H. G. Lewin
597: \& M. van der Klis, Cambridge University Press (astro-ph/0410551)
598:
599: \bibitem[1995]{wa95}
600: Wang, Y. M. 1995, ApJ, 449, L153
601:
602: \bibitem[2003]{wj03}
603: Wijnands, R. et al.
604: %van der Klis, M., Homan, J., Chakrabarty, D.,
605: %Markwardt, C.B., \& Morgan, E.H.
606: 2003, \nat, 424, 44
607:
608: \end{thebibliography}
609:
610: \clearpage
611: \begin{figure}
612: \includegraphics{f1.eps}
613:
614: \caption{The solid lines show the dimensionless rotational
615: frequencies in the accretion disk for $\win=0.2$, 0.5, and 0.8
616: (from bottom to top). The dashed line represents the Keplerian
617: rotation. }
618: \end{figure}
619:
620: \clearpage
621: \begin{figure}
622: \includegraphics{f2.eps}
623: \caption{The parameter $\xi$ plotted against the fastness
624: parameter $\ws$.}
625: \end{figure}
626:
627: \begin{figure}
628: \includegraphics{f3.eps}
629: \caption{Predicted relations for twin kHz QPO frequencies and
630: measured frequencies for four NS LMXBs}
631: \end{figure}
632:
633: \end{document}
634: