1: %%%%%%%%%%%%%%%%%%%%%
2:
3:
4: %%
5: \documentclass[]{spie} %>>> use for US letter paper
6: %\documentclass[a4paper]{spie} %>>> use for A4 paper
7: \usepackage[dvips]{graphicx}
8:
9: \title{The distribution of mass ratios in compact object binaries}
10:
11: %>>>> The author is responsible for formatting the
12: % author list and their institutions. Use \skiplinehalf
13: % to separate author list from addresses and between each address.
14: % The correspondence between each author and his/her address
15: % can be indicated with a superscript in italics,
16: % which is easily obtained with \supit{}.
17:
18: \author{Tomasz Bulik\supit{a}, Krzysztof Belczynski\supit{b,c}
19: and Vassiliki Kalogera\supit{b}
20: \skiplinehalf
21: \supit{a}Nicolaus Copernicus Astronomical Center, Bartycka 18, 00176 Warsaw, Poland\\
22: \supit{b} Northwestern University,
23: 2145 Sheridan Rd., Evanston, IL 60208, USA\\
24: \supit{c}Lindheimer Fellow
25: }
26:
27: %>>>> Further information about the authors, other than their
28: % institution and addresses, should be included as a footnote,
29: % which is facilitated by the \authorinfo{} command.
30:
31: \authorinfo{E-mail:
32: T. B.: bulik@camk.edu.pl;
33: K. B.: kbelczyn@northwestern.edu;
34: V. K.: vicky@northwestern.edu}
35: %% NB: when using amstex, you need to use @@ instead of @
36:
37:
38: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
39: %>>>> uncomment following for page numbers
40: % \pagestyle{plain}
41: %>>>> uncomment following to start page numbering at 301
42: %\setcounter{page}{301}
43:
44: \begin{document}
45: \maketitle
46:
47: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
48: \begin{abstract}
49:
50: Using the StarTrack population synthesis code we compute the distribution
51: of masses of merging compact object (black hole or neutron star) binaries.
52: The shape of the mass distribution is sensitive to some of the parameters
53: governing the stellar binary evolution. We discuss the possibility of
54: constraining stellar evolution models using mass measurements obtained
55: from the detection of compact object inspiral with the upcoming
56: gravitational-wave observatories.
57:
58: \end{abstract}
59:
60: %>>>> Include a list of keywords after the abstract
61:
62: \keywords{compact objects, binaries, gravitational waves,stellar evolution}
63:
64: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
65: \section{INTRODUCTION}
66:
67: The operational phase of the gravitational
68: wave detectors LIGO\cite{1992Sci...256..325A} and VIRGO \cite{1990brada}
69: is quickly approaching, and
70: in a few years we anticipate the operation of the more
71: sensitive advanced LIGO detector. Mergers of compact objects
72: in binaries are the most promising sources of gravitational waves
73: to be detected by these detectors. The
74: expected rates of such mergers have been
75: a subject of intense research.
76: Recent analyses \cite{2001ApJ...556..340K,2002ApJ...572..407B,KKL}
77: have shown there is a little chance that
78: LIGO should see any mergers of compact object binaries, while LIGO~II
79: could detect a substantial number, up to hundreds or thousands of such
80: events. These estimated rates however, do vary strongly with the model of
81: stellar binary evolution.
82:
83: Once we begin to detect such events, the data analysis should yield
84: information about each system that merged. The merging of compact object
85: binaries proceeds in three phases: inspiral, merger, and ringdown. It is
86: the inspiral phase that will be used for detection of the stellar masses
87: involved. The analysis of the frequency change with time of the event
88: should yield the masses of the merging objects\cite{will}. However,
89: determining their type, i.e. a neutron star or a black hole will be
90: difficult\cite{blanchet} with the measurement during this phase. The
91: modifications of the gravitational wave signal due to the mass
92: distribution of the neutron star are small and may amount to one
93: in $10^4$ oscillations.
94:
95: The analysis of the observation of merger phase may lead to determination
96: of the type of the compact object i.e. will be different for mergers
97: involving neutron star, and may even lead to constraints on the neutron
98: star equation of state. Also, if the neutron stars are rotating there is a
99: chance of exciting stellar oscillations by the orbital motion which may
100: modify the gravitational waveform\cite{1999MNRAS.308..153H}.
101:
102: It is then important to ask what is the expected two dimensional
103: distribution of masses of compact object binaries that merge, and how does
104: this distribution depend on the assumed model of stellar binary evolution.
105: In \S\,2 we present the dependence of the mass ratio distribution on the
106: particular stellar evolution model. Next, \S\,3 is devoted to the
107: potential observations. The main observable derived from the
108: gravitational-wave signal is the chirp mass. We show the dependence of the
109: distribution of the observed chirp masses on the stellar evolution
110: parameters and estimate the possibility of distinguishing between the
111: models using the amount of data that we expect to be obtained. Finally in
112: \S\,4 we summarize our results.
113:
114:
115: \begin{figure}[t]
116: \centerline{
117: \includegraphics[width=0.9\textwidth]{ryson.ps}}
118: \caption{The distributions of intrinsic mass ratio in compact objects
119: produced in the standard model simulation (left panel), and
120: the contributions of different types of binaries (BH-BH thick line, NS-NS
121: thin line , and
122: BH-NS dotted line)to this
123: distribution. The distributions have been
124: calculated using a total of 106307 compact object binaries
125: and the bin width is $dq=0.01$.
126: Each distribution is normalized to unity.
127: }
128: \label{rysone}
129: \end{figure}
130:
131: \section{The distribution of mass ratio in compact object binaries.}
132:
133: In order to calculate theoretical distributions of the mass ratio of
134: compact object merging events we use the StarTrack population synthesis
135: code \cite{2002ApJ...572..407B}. The standard evolutionary scenario of the
136: code uses modified formulae of Ref.~\citenum{2000MNRAS.315..543H} to
137: evolve single stars along with their proposed wind mass loss rates. The
138: code takes into account different types of binary interactions:
139: dynamically stable and unstable mass transfer phases. The dynamically
140: stable events are modeled using the formalism of
141: Ref.~\citenum{1992ApJ...391..246P}, while unstable mass transfers are
142: treated following Ref.~\citenum{1984ApJ...277..355W}. Compact object
143: (black holes, neutron stars) formation in supernova (SN) explosion/core
144: collapse events, is described in the code following the results of
145: hydrodynamical calculations of Ref.~\citenum{1999ApJ...522..413F}. Neutron
146: stars receive natal kicks from the distribution of
147: Ref.~\citenum{1998ApJ...505..315C}; intermediate mass black holes are
148: formed through partial fall back and in attenuated SN explosions receiving
149: smaller kicks than neutron stars; while most massive black holes are
150: formed through direct collapse of their immediate progenitors with no
151: accompanying SN explosion and they do not receive any kicks at their
152: formation.
153:
154: The distribution of masses in compact object
155: binaries can be described by the probability density
156: $\rho ( m_1, m_2)$.
157: The population synthesis code provides a numerical
158: representation of this distribution
159: \begin{equation}
160: \rho( m_1, m_2) = {1\over N} \sum_i \delta(m_1^i)\delta(m_2^i)
161: \label{rho}
162: \end{equation}
163: where $\delta$ is the Dirac's function,
164: $N$ is the number of binaries obtained in a simulation.
165:
166: We present the mass ratios distributions
167: \begin{equation}
168: {dp\over dq} = \int dm_1\int dm_2 {\rho( m_1, m_2)}
169: \delta\left(q-{m_1\over m_2}\right) = {1\over N} \sum_i\delta(q_i)
170: \end{equation}
171: of different
172: types of compact object binaries in the right panel
173: of Figure~\ref{rysone}, and in the left panel we show the
174: total distribution summed over all the objects. Note that
175: the total distribution is dominated by
176: double NS binaries. They
177: are more numerous than the binaries containing black
178: holes simply because of the steep fall of the initial mass
179: function\cite{2002ApJ...567L..63B}.
180: The distributions are not flat and exhibit
181: distinctive features which are not numerical artifacts; the
182: statistical
183: fluctuation in the left panel of Figure~\ref{rysone} amount to less than 1\%.
184:
185:
186: The StarTrack code allows to test different
187: models of stellar binary evolution. We have run a set of simulations with
188: different models to show the dependence
189: of the observed distribution of mass ratio
190: on a given evolutionary model. We list the
191: models and their description in Table \ref{tone}.
192:
193:
194:
195:
196: \begin{table}
197: \caption{Description of different population synthesis models
198: for which the distributions of mass ratios have been calculated. For more
199: details see Ref.\ 3.}
200: \label{tone}
201: \begin{center}
202: \begin{tabular}{ll}
203: \hline \hline
204: Model & Description \\
205: \hline
206:
207: A & standard model described in Ref. 2. \\
208: B1,4,6,9 & zero kicks, single Maxwellian with \\
209: & $\sigma=30,50,300,$\,km\,s$^{-1}$, \\
210:
211: C & no hyper--critical accretion onto NS/BH in CEs \\
212: D1--2 & maximum NS mass: $M_{\rm max,NS}=2, 1.5$\,M$_\odot$ \\
213: E1--3 & $\alpha_{\rm CE}\times\lambda = 0.1, 0.5, 2$ \\
214: F1--2 & mass fraction accreted: f$_{\rm a}=0.1, 1$ \\
215: G1--2 & wind changed by\ $f_{\rm wind}=0.5, 2$ \\
216: H1--2 & Convective Helium giants: $M_{\rm conv}=4.0$,\, $0.0\,M_\odot$ \\
217: I & burst--like star formation history \\
218: J & primary mass: $\propto M_1^{-2.35}$ \\
219: L1--2 & angular momentum of material lost in MT: $j=0.5, 2.0$\\
220: M1--2 & initial mass ratio distribution: $\Phi(q) \propto q^{-2.7}, q^{3}$\\
221: N & no helium giant radial evolution\\
222: O & partial fall back for $5.0 < M_{\rm CO} < 14.0 \,M_\odot$\\
223: \hline
224: \end{tabular}
225:
226: \end{center}
227: \end{table}
228:
229: \begin{figure}[t]
230: \centerline{
231: \includegraphics[width=0.9\textwidth]{models.ps}}
232: \caption{
233: Distributions of mass ratio for a set of models listed in
234: Table~\ref{tone}. The distributions are normalized to unity, but for clarity
235: we added $5$ to each consecutive distribution.
236: }
237: \label{rystwo}
238: \end{figure}
239:
240: We present the distributions of the mass ratios for a range of stellar
241: evolutionary models in Figure~\ref{rystwo}. The distributions vary
242: significantly with the particular choice of the model. The kick velocities
243: (models B) change the relative strength of the peaks in the distribution.
244: These peaks are related to the neutron star in binaries, see
245: Figure~\ref{rysone}. Various parameters lead to specific changes in the
246: distribution. IN particular several parameters influence the shape of the
247: distribution very significantly: variation of the common envelope
248: efficiency $\alpha_{CE}\lambda$ (models E), changes of the stellar wind
249: strength (models G) etc. The distribution of masses, and consequently
250: mass ratios of compact object binaries does carry significant information
251: about the evolution of the massive binaries. It is therefore important to
252: ask the question: Can we constrain observationally this distribution? Is
253: it possible to learn about the evolution of massive stars by measuring
254: masses of merging binaries?
255:
256:
257: \section{What can be observed by GW detectors? }
258:
259: Mergers of compact object binaries will be observed
260: by the gravitational wave interferometers during
261: the inspiral phase, just minutes prior to the merger.
262: The data train will allow us to measure at least the chirp mass:
263: \begin{equation}
264: {\cal M}(m_1,m_2) = (m_1 +m_2) \left( {m_1\over m_2} +{m_2\over m_1}\right)^{-3/5} =
265: m_2 (1 +q) \left( {q +q^{-1}}\right)^{-3/5}
266: \end{equation}
267: The individual masses of merging compact objects, $m_1$ and $m_2$,
268: will be measured if higher
269: order correction due to general relativity are taken into account.
270: To find the
271: distribution of the observed chirp masses we need to
272: take into account also the different strength of the gravitational
273: wave signal from sources with different mass.
274:
275:
276: The strength of the gravitational wave signal
277: from coalescing compact sources has been estimated in
278: Refs~\citenum{1993PhRvD..47.2198F,1998PhRvD..57.4535F}. In the inspiral phase
279: the signal to noise $S/N$ ratio
280: for LIGO interferometers is proportional to ${\cal M}^{5/6}$.
281: In particular:
282: \begin{equation}
283: \left( {S\over N}\right) = k \left( {1 {\rm Gpc} \over D}\right)
284: \left( {4\mu \over M}\right)^{1/2}
285: \left( {(1+z) M\over 18 M_\odot}\right)^{5/6}
286: \label{SN}
287: \end{equation}
288: where $D$ is the distance, $M$ is the total mass
289: of the system, $\mu$ is the reduced mass,
290: $k= 0.58$ for the initial LIGO, and $k=15.3 $ for the advanced LIGO
291: configuration. Therefore the sampling distance for a constant
292: signal to noise ratio is proportional to
293: $ D\propto \left( {4\mu / M}\right)^{1/2} M^{5/6}$, and the sampling
294: volume $V\propto D^3$.
295: Thus the observed number of sources observed
296: with given masses $m_1$ and $m_2$ is
297: \begin{equation}
298: dN= V(m_1,m_2) \rho(m_1,m_2) dm_1 dm_2\, .
299: \end{equation}
300: and the total number of sources to be observed is
301: $N_{tot}=\int dN$. The normalized distribution of the observed
302: chirp masses is
303: \begin{equation}
304: p({\cal M})={1\over N_{tot}}{dN\over d{\cal M}}
305: ={1\over N_{tot}}
306: \int dm_1 \int dm_2 V(m_1,m_2) \rho(m_1,m_2)
307: \delta\left({\cal M}- {\cal M}(m_1,m_2)\right)\,.
308: \label{pq}
309: \end{equation}
310: The population synthesis code provides us with a numerical
311: representation of the $\rho(m_1,m_2)$ distribution from which we obtain the
312: numerical representation of the distribution of the mass
313: ratios $p({\cal M})$,
314: by inserting Equation~\ref{rho} into Equation \ref{pq}, and
315: replacing integrals with sums.
316:
317:
318:
319:
320:
321: \begin{figure}
322: \includegraphics[width=0.9\textwidth]{11.ps}
323: \caption{The expected distributions of the observed chirp mass
324: in selected models with varying kick velocity distributions (B1,B4,B6,B9), with
325: different common envelope efficiency $\alpha_{CE}\lambda$ (E1, E2, E3)
326: and without the hypercritical accretion onto the compact object (C). In each
327: panel we present for comparison the distribution of model A .}
328: \label{one}
329: \end{figure}
330:
331:
332: \begin{figure}
333: \includegraphics[width=0.9\textwidth]{22.ps}
334: \caption{The expected distributions of the observed chirp mass
335: in models D,F,G,H. In each
336: panel we present for comparison the distribution of model A . }
337: \label{two}
338: \end{figure}
339:
340: \begin{figure}
341: \includegraphics[width=0.9\textwidth]{33.ps}
342: \caption{The expected distributions of the observed chirp mass
343: in models I, L, M, N, O. In each
344: panel we present for comparison the distribution of model A .}
345: \label{three}
346: \end{figure}
347:
348:
349: We present the expected distributions of observed chirp masses in
350: Figures~\ref{one}, \ref{two}, and~\ref{three}. The dependence of the
351: observed distribution of mass ratios on the kick velocity is presented in
352: the top panels of Figure~\ref{one}. The observed distribution is dominated
353: by the BH-BH mergers, and BHs do not receive substantial kicks, especially
354: for the massive BH systems for which the sampling volume is the largest.
355: However, the contribution of binaries containing neutron stars at ${\cal
356: M}\approx 1.5\, M_\odot$ changes their number, see the strength of the
357: peak at ${\cal M} \approx 1-2\,M_\odot$.
358:
359: The bottom panels of Figure~\ref{one} show the dependence of the observed
360: distribution of chirp masses ratios on the common envelope efficiency
361: parameter $\alpha_{CE}\lambda$. Turning off the hypercritical accretion
362: onto compact sources (model C), is shown in the bottom right panel of
363: Figure~\ref{one}).
364:
365:
366: Models D1 and D2 shown in the top panels of Figure~\ref{two} serve as self
367: consistency checks of the calculations. In these models the maximum mass
368: of a NS is decreased, and since the character of the gravitational wave in
369: the inspiral phase does not depend on the nature (BH or NS) of a given
370: compact object they should not differ from the standard model A. In models
371: F1 and F2, shown also in Figure~\ref{two}, we vary the mass fraction
372: accreted onto a compact object in mass transfer events. While the
373: statistic for this models is poor we can clearly say that the observed
374: chirp mass distributions vary strongly. Models G1 and G2, shown in
375: Figure~\ref{two} correspond to the increasing and decreasing of the
376: stellar winds by a factor of two. The decreased wind strength shifts the
377: distribution to the higher values and decreases the number of small chirp
378: mass ratio mergers, while increasing the stellar winds makes NS-NS mergers
379: dominate the observed sample, and the entire distribution shifts to the
380: lower values. Models H1 and H2, where we test different values of
381: $M_{conv}$ of helium giants affect only the properties of the neutron
382: stars and therefore do not strongly change the overall shape of the
383: distribution, save for the peak at ${\cal M} \approx 1-2\,M_\odot$.
384:
385:
386: Models L, presented in Figure~\ref{three} correspond to variation of the
387: angular momentum fraction lost in mass transfer events, and affect strongly
388: the shape of the observed distribution of chirp masses. Models M, where we vary
389: the shape of the initial mass ratio distribution do not make a significant
390: difference. This is because the compact object binaries are formed from
391: binaries which initially had rather well defined properties (like the initial
392: mass ratio), the number of such binaries does change fore different initial
393: mass ratio distributions but their properties do not. Model N where we neglect
394: the helium giant radial evolution affects the properties of neutron stars and
395: does not change significantly the distribution of observed chirp masses.
396: Finally in model O we change the prescription for masses of the newly formed
397: compact objects, which has a strong influence on the shape of the distribution
398: of observed chirp masses.
399:
400:
401:
402:
403:
404: \begin{table}[t]
405:
406: \caption{Results of the simulations of
407: observation of $20$, $100$ and $500$ mergers. We list
408: KS-test probabilities that appeared in less than 1\% of
409: tries. The lower the probability the easier can a given model
410: be distinguished.
411: }
412: \begin{center}
413: \begin{tabular}{lrrr}
414: Model & 20 mergers & 100 mergers & 500 mergers \\
415: \hline\hline
416: B1 & $ 0.943 $ & $ 0.169 $ & $ 2.19\times 10^{-8} $\\
417: B2 & $ 0.945 $ & $ 0.261 $ & $ 3.56\times 10^{-9}$\\
418: B3 & $ 0.966 $ & $ 0.483 $ & $ 1.19\times 10^{-6}$\\
419: B4 & $ 0.956 $ & $ 0.523 $ & $ 1.11\times 10^{-6}$\\
420: B5 & $ 0.971 $ & $ 0.464 $ & $ 4.44\times 10^{-5}$\\
421: B6 & $ 0.980 $ & $ 0.327 $ & $ 9.20\times-07$\\
422: B7 & $ 0.990 $ & $ 0.891 $ & $ 5.28\times 10^{-2}$\\
423: B8 & $ 0.991 $ & $ 0.977 $ & $ 0.625 $\\
424: B9 & $ 0.991 $ & $ 0.989 $ & $ 0.823 $\\
425: B10 & $ 0.988 $ & $ 0.886 $ & $ 0.138 $\\
426: B11 & $ 0.984 $ & $ 0.580 $ & $ 2.45\times 10^{-5}$\\
427: B12 & $ 0.973 $ & $ 0.286 $ & $ 1.77\times 10^{-6}$\\
428: B13 & $ 0.995 $ & $ 0.990 $ & $ 0.870 $\\
429: C & $ 0.966 $ & $ 0.135 $ & $ 1.48\times 10^{-6}$\\
430: D1 & $ 0.992 $ & $ 0.993 $ & $ 0.750 $\\
431: D2 & $ 0.993 $ & $ 0.987 $ & $ 0.909 $\\
432: E1 & $ 0.010 $ & $ 1.0\times10^{-12}$ & $ 0.0$\\
433: E2 & $ 0.990 $ & $ 0.8248529 $ & $ 7.25\times 10^{-3}$\\
434: E3 & $ 0.915 $ & $ 0.0236 $ & $ 1.71\times 10^{-14}$\\
435: F1 & $ 0.148 $ & $ 1.15\times 10^{-9} $ & $ 0.0$\\
436: F2 & $ 0.401 $ & $ 1.07\times 10^{-4} $ & $ 3.25\times 10^{-23}$\\
437: G1 & $ 0.001 $ & $ 6.64\times 10^{-26} $ & $ 0.0$\\
438: G2 & $ 1.42\times 10^{-17}$ & $ 0.0 $ & $ 0.0$\\
439: H1 & $ 0.996 $ & $ 0.9941702 $ & $ 0.990 $\\
440: H2 & $ 0.990 $ & $ 0.9725315 $ & $ 0.420 $\\
441: I & $ 0.989 $ & $ 0.9746811 $ & $ 0.592 $\\
442: J & $ 0.992 $ & $ 0.9898050 $ & $ 0.953 $\\
443: L1 & $ 0.461 $ & $ 5.59\times 10^{-5} $ & $ 2.69\times 10^{-27}$\\
444: L2 & $ 0.188 $ & $ 4.95\times 10^{-9} $ & $ 0.0$\\
445: M1 & $ 0.995 $ & $ 0.9549476 $ & $ 0.111 $\\
446: M2 & $ 0.985 $ & $ 0.9328853 $ & $ 0.405 $\\
447: N & $ 0.995 $ & $ 0.9818770 $ & $ 0.365 $\\
448: O & $ 6.16\times10^{-12} $ & $ 0.00 $ & $ 0.0$\\
449: \end{tabular}
450:
451:
452: \end{center}
453: \label{wyn}
454: \end{table}
455:
456:
457:
458: \section{CONSTRAINING STELLAR EVOLUTIONARY PARAMETERS}
459:
460: In this section we discuss a way of quantifying the results presented
461: above. In order to test the usefulness of observations of chirp masses for
462: constraining stellar evolution models we performed the a Monte Carlo
463: simulation of such observations. We have assumed that the underlying
464: stellar population is described by a given model chosen from
465: Table~\ref{tone}. We estimated if one can reject a hypothesis that the
466: stellar population is described by the standard model A with an
467: observation of 20, 100, and 500 mergers. We chose ten thousand random
468: realizations of the observed sample of mergers. Each such realization was
469: compared with the distribution of model A using the Kolmogorov Smirnov
470: (KS) test. We then examined the distribution of the KS-test probabilities
471: and found the probability that appeared in less than 1\% of simulations. A
472: high value of this probability means that the given model can not be
473: distinguished from model A, while a low value means that the two models
474: can easily be distinguished. We present the results in Table~\ref{wyn}. We
475: first notice that models D do not differ from model A, just as expected.
476: In the case of models that affect the distribution of masses and
477: especially the range of their masses like e.g. models G and O, already a
478: small sample of several tens of mergers will be sufficient to recognize
479: them. With a larger sample of a hundred mergers one should be able to
480: constrain the common envelope parameter $\alpha_{CE}\lambda$ (models E).
481: Also it should be possible to determine the parameters describing mass
482: transfer events: the mass fraction accreted in mass transfer events
483: (models F), and the angular momentum of the material lost in mass
484: transfers. However, a larger sample of mergers will be required to
485: constrain the kick velocity distribution: even with 500 detections it is
486: going to be possible to put only rough constraints on the width of the
487: kick velocity distribution.
488:
489:
490:
491: \section{CONCLUSIONS}
492:
493: Using the StarTrack population synthesis code we show that the distribution of
494: masses of merging compact objects carries useful information about the ways
495: massive stars evolve. In particular the stellar wind strength, common envelope
496: efficiency, loss of angular momentum efficiency, and the fraction of mass
497: accreted in mass transfer events leave distinct tracks on the distribution of
498: observed chirp masses. The Monte Carlo simulations show that analysis of this
499: distribution shall yield useful constraints on the evolutionary parameters of
500: binaries containing massive stars. The advanced LIGO configuration will be
501: able to gather sufficient number of merging events to perform such analysis.
502:
503:
504:
505: \acknowledgments
506: This research was funded by KBN through grant number 5P03D01120. TB is
507: grateful for the hospitality of the Theoretical Astrophysics Group at
508: Northwestern University. KB acknowledges support by the Lindheimer Fund at
509: Northwestern University. VK acknowledges partial support by NSF grant
510: PHY-0121420.
511:
512:
513: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
514: %%%%% References %%%%%
515: \newcommand\apj{ApJ}\newcommand\mnras{MNRAS}\newcommand\apjl{ApJLett}
516: \newcommand\prd{Phys ReV D}
517: % \bibliography{gw} %>>>> bibliography data in report.bib
518:
519: \begin{thebibliography}{10}
520:
521: \bibitem{1992Sci...256..325A}
522: A.~{Abramovici}, W.~E. {Althouse}, R.~W.~P. {Drever}, Y.~{Gursel},
523: S.~{Kawamura}, F.~J. {Raab}, D.~{Shoemaker}, L.~{Sievers}, R.~E. {Spero}, and
524: K.~S. {Thorne}, ``{LIGO - The Laser Interferometer Gravitational-Wave
525: Observatory},'' {\em Science} {\bf 256}, pp.~325--333, Apr. 1992.
526:
527: \bibitem{1990brada}
528: Bradaschia {\em et~al.}, ``The virgo detector,'' {\em Nucl. Instrum and
529: Methods} {\bf A289}, p.~518, 1990.
530:
531: \bibitem{2001ApJ...556..340K}
532: V.~{Kalogera}, R.~{Narayan}, D.~N. {Spergel}, and J.~H. {Taylor}, ``{The
533: Coalescence Rate of Double Neutron Star Systems},'' {\em \apj} {\bf 556},
534: pp.~340--356, July 2001.
535:
536: \bibitem{2002ApJ...572..407B}
537: K.~{Belczynski}, V.~{Kalogera}, and T.~{Bulik}, ``{A Comprehensive Study of
538: Binary Compact Objects as Gravitational Wave Sources: Evolutionary Channels,
539: Rates, and Physical Properties},'' {\em \apj} {\bf 572}, pp.~407--431, June
540: 2002.
541:
542:
543:
544: \bibitem{KKL}
545: C.-L.\ Kim, V.\ Kalogera, D.R.\ Lorimer, ApJ, submitted, astro-ph/0207408
546: ``The Probability Distribution of Binary Pulsar Coalescence Rates. I.
547: Double Neutron Star Systems in the Galactic Field''
548:
549:
550: \bibitem{will}
551: C.~M. Will, ``The confrontation between general relativity and experiment,''
552: {\em Living Rev. Relativity} {\bf 4}, p.~4, 2001.
553: \newblock http://www.livingreviews.org/Articles/Volume4/2001-4will/.
554:
555: \bibitem{blanchet}
556: L.~Blanchet, ``Gravitational radiation from post-newtonian sources ans
557: inspiralling compact binaries,'' {\em Living Rev. Relativity} {\bf 5}, p.~3,
558: 2002.
559: \newblock http://www.livingreviews.org/Articles/Volume5/2002-3blanchet/.
560:
561: \bibitem{1999MNRAS.308..153H}
562: W.~C.~G. {Ho} and D.~{Lai}, ``{Resonant tidal excitations of rotating neutron
563: stars in coalescing binaries},'' {\em \mnras} {\bf 308}, pp.~153--166, Sept.
564: 1999.
565:
566: \bibitem{2000MNRAS.315..543H}
567: J.~R. {Hurley}, O.~R. {Pols}, and C.~A. {Tout}, ``{Comprehensive analytic
568: formulae for stellar evolution as a function of mass and metallicity},'' {\em
569: \mnras} {\bf 315}, pp.~543--569, July 2000.
570:
571: \bibitem{1992ApJ...391..246P}
572: P.~{Podsiadlowski}, P.~C. {Joss}, and J.~J.~L. {Hsu}, ``{Presupernova evolution
573: in massive interacting binaries},'' {\em \apj} {\bf 391}, pp.~246--264, May
574: 1992.
575:
576: \bibitem{1984ApJ...277..355W}
577: R.~F. {Webbink}, ``{Double white dwarfs as progenitors of R Coronae Borealis
578: stars and Type I supernovae},'' {\em \apj} {\bf 277}, pp.~355--360, Feb.
579: 1984.
580:
581: \bibitem{1999ApJ...522..413F}
582: C.~L. {Fryer}, ``{Mass Limits For Black Hole Formation},'' {\em \apj} {\bf
583: 522}, pp.~413--418, Sept. 1999.
584:
585: \bibitem{1998ApJ...505..315C}
586: J.~M. {Cordes} and D.~F. {Chernoff}, ``{Neutron Star Population Dynamics. II.
587: Three-dimensional Space Velocities of Young Pulsars},'' {\em \apj} {\bf 505},
588: pp.~315--338, Sept. 1998.
589:
590: \bibitem{2002ApJ...567L..63B}
591: K.~{Belczynski}, T.~{Bulik}, and W.~{\l}. {Klu{\' z}niak}, ``{Population
592: Synthesis of Neutron Stars, Strange (Quark) Stars, and Black Holes},'' {\em
593: \apjl} {\bf 567}, pp.~L63--L66, Mar. 2002.
594:
595: \bibitem{1993PhRvD..47.2198F}
596: L.~S. {Finn} and D.~F. {Chernoff}, ``{Observing binary inspiral in
597: gravitational radiation: One interferometer},'' {\em \prd} {\bf 47},
598: pp.~2198--2219, Mar. 1993.
599:
600: \bibitem{1998PhRvD..57.4535F}
601: E.~E. {Flanagan} and S.~A. {Hughes}, ``{Measuring gravitational waves from
602: binary black hole coalescences. I. Signal to noise for inspiral, merger, and
603: ringdown.},'' {\em \prd} {\bf 57}, pp.~4535--4565, 1998.
604:
605:
606:
607:
608: \end{thebibliography}
609: \bibliographystyle{spiebib} %>>>> makes bibtex use spiebib.bst
610:
611: \end{document}
612:
613: