0809.1890/ms.tex
1: \documentclass[10pt, preprint]{aastex}
2: %%\documentclass[dvips]{article}
3: %%\usepackage{graphicx}
4: %%\usepackage{color}
5: %%\usepackage{latexsym}
6: \usepackage{natbib,color,lscape}
7: \frenchspacing
8: \setlength{\textwidth}{17.0cm}
9: \setlength{\leftmargin}{0.01in}
10: %\setlength{\rightmargin}{0.2in}
11: \setlength{\oddsidemargin}{0.2in}
12: \setlength{\evensidemargin}{0.2in}
13: \setlength{\topmargin}{0.1in}
14: \setlength{\textheight}{8.7in}
15: \title{Population boundaries for compact white-dwarf binaries in
16: LISA's amplitude-frequency domain}
17: \author{Ravi kumar Kopparapu\altaffilmark{1}}
18: \altaffiltext{1}{Center for Gravitational Wave Physics, 104 Davey lab, Pennsylvania State
19: University, University Park, PA - 16802-6300, USA}
20: \begin{document}
21: \begin{abstract}
22: In an earlier investigation, we proposed population boundaries for 
23: both inspiralling and mass-transferring double
24: white dwarf (DWD) systems in the distance independent ``absolute'' amplitude-frequency domain
25: of the proposed space-based gravitational-wave (GW) detector, {\it LISA}.
26:  The degenerate zero temperature mass-radius (M-R) relationship of individual white dwarf stars that
27:  we assumed, in combination with the constraints imposed by Roche geometries, permits us to identify
28:  five key population boundaries for DWD systems in various phases of evolution.
29: % in this domain.
30: %of the proposed space-based gravitational-wave (GW) detector, {\it LISA}.
31: %this domain.  
32: Here we use the  non-zero entropy donor M-R relations of \cite{DB2003} to modify these boundaries for 
33:  both DWD and neutron star-white dwarf (NSWD) binary systems. We find that the mass-transferring
34: systems occupy a larger fraction of space in ``absolute'' amplitude-frequency domain compared to
35: the simpler $T=0$ donor model. We also discuss how these boundaries are
36: modified with the  new evolutionary phases found by \cite{Deloyeetal2007}. In the initial
37: contact phase, we find that the
38:  contact boundaries, which are the result of end of inspiral evolution,  would have some width, as opposed to an abrupt cut-off described in our earlier 
39: $T=0$ model.
40: % rather than a sharp cut-off.
41:  This will cause an overlap between a
42: DWDs $\&$ NSWDs evolutionary trajectories, making them indistinguishable with 
43: only LISA observations
44: within this region.  In the cooling phase of
45: the donor, which follows after the adiabatic donor evolution, the radius contracts,
46: mass-transfer rate drops and slows down the
47: orbital period evolution. Depending upon the entropy of the donor,
48:  these systems may then lie
49: inside the fully degenerate $T=0$ boundaries, but LISA may be unable to
50:  detect these systems as they might
51:  be below
52: the sensitivity limit or within the unresolved DWD background noise.
53: %Here we extend these boundaries to both inspiralling and mass-transferring 
54: %neutron star-white dwarf (NSWD) binary systems, which occupy distinct sub-domains
55: %than DWDs, in the ``absolute'' amplitude-frequency space.
56: %, assuming that the currently known 
57: %ultra-compact x-ray binaries (UCXBs) are NSWD systems in conservative mass transfer (CMT) 
58: %phase.
59: %Since the relative population of NSWDs in the field is small ($~10$) compared to DWD binary 
60: %systems, given one full year of uninterrupted operation, it is likely that
61: % {\it LISA} may be able to measure the rate at which the GW frequency $f$ is 
62: %changing for some of the DWD binaries
63: %, which we assume to be ultra-compact
64: %x-ray binaries (UCXBs), in our Galaxy and nearby globular clusters.
65: %Assuming that the currently known  
66: %ultra-compact x-ray binaries (UCXBs) are NSWD systems in conservative mass transfer (CMT) 
67: %phase,
68:  We assess the limits and applicability of our theoretical population boundaries
69: with respect to observations and find that a measurement of $\dot f$ by LISA at high frequencies 
70: (Log $[f] \geq 2$) would likely distinguish between DWD/NSWD binary. For low frequency
71: sources, GW observations
72: alone would unlikely tell us about the binary components, without the help of electromagnetic
73:  observations. 
74: %The fairly known issue of the apparent clustering of UCXBs
75: % in the orbital period range of $40 \sim 50$ minutes  can also be explained through
76: %our population boundaries, noting that the evolutionary time-scale for these systems
77: %is approaching the hubble time, hence more systems accumulate at longer orbital periods.
78: % We suggest that if LISA measures just the
79: %GW amplitude $h$ and frequency $f$ and if by independent means we know 
80: %the (DWD or NSWD) binary system is 
81: %undergoing conservative mass transfer,
82: % then the population boundaries proposed
83: %here can be used to set limits on the distance to these binaries.
84: \end{abstract}
85: \keywords{accretion, accretion disks -- binaries: close --- gravitational waves
86:  --- stars: white dwarfs, neutron stars}
87: 
88: \section{Introduction}
89: \label{sec1}
90: %One of the most promising GW sources for 
91: The proposed space-based Gravitational-wave (GW) detector, 
92: {\it LISA}\footnote{http://lisa.nasa.gov} ({\it Laser Interferometer Space Antenna})
93: \citep{FB84,EIS87,Bender98}, is sensitive to GWs in the $10^{-4} - 1$ Hz frequency range. 
94: Within this band, one of the most promising sources are double white dwarf (DWD)
95:  binary systems, as it is expected that a large fraction of
96:  main-sequence binaries 
97: end their lives as close DWDs \citep{IT84, IT86}. For this reason, the GWs emitted by these systems in our Galaxy 
98: may form a background noise in
99: the low frequency ($\leq 3 \times 10^{-3}$ Hz) band of {\it LISA}. The population of DWDs in 
100: our Galaxy is 
101: expected to be dominated by systems that undergo two distinct,
102: long-lived phases of evolution: an ``inspiral'' phase, where both the stars are detached from
103: their Roche lobes and the loss of angular momentum in the form of GW emission
104:  causes the two stars to slowly spiral in towards each other; and
105:  a ``stable mass transfer'' phase, where the less massive star fills its
106: Roche lobe initially and starts transferring mass steadily to its companion. An example of 
107: the stable mass transferring systems are the AM CVn type systems, of which 18
108:  \citep{Nelemans2005} are known through electromagnetic observations\footnote{Two
109: controversial candidate systems, RX J0806+15 and V407 Vul may change their number to $16$. See
110: \cite{Cropper1998, Wu2002, MS2002} for more details.}.
111: 
112: Apart from DWDs, neutron star white dwarf (NSWD)
113:  binary systems are also  one of the promising sources of GWs for LISA.
114: Various authors \citep{Kim2004, cooray2004, Nelemans2001} have estimated the GW background from
115:  these systems and concluded
116: that the number of NSWD systems detectable with LISA is 1-2 orders of magnitude less than DWD systems.
117: %they are order of magnitude smaller than the background signal produced by DWDs.
118:  Similar to DWDs, NSWD systems also undergo 
119: inspiral and stable mass transfer phases. Specifically, several studies \citep[see for example:][]{Nelson1986,
120:  BT2004a, Nelemans2006}
121:  have suggested that one of the possible formation scenarios of the so-called 
122: ultra compact x-ray binary (UCXB) systems,
123: with orbital periods  $\le 80$ minutes, is that a low mass
124: white-dwarf donor ($\le 0.1 M_\mathrm{\odot}$) transferring mass to an accreting 
125: neutron star (NS) primary in a short orbit.
126: In this scenario, a detached NSWD binary system initially evolves to a minimum orbital period as  angular
127: momentum is lost from the system due to GW radiation. At this minimum orbital period, the companion white dwarf (WD) star
128: starts filling its Roche lobe and transfers mass to the NS  and the system evolves to longer orbital periods.
129: At present, there are $12$ known UCXB systems with measured orbital periods, 
130: see Table\ref{table1}. Some of these systems have accreting millisecond pulsars
131: (XTE J1807-294, XTE J1751-305 \& XTE J0929-314: see \cite{Mark2002}, for example)
132:  and several of them are found in globular clusters as the stellar density and
133:  close encounters are more common \citep{clark1975,ivanova2007}. The total 
134: population of field UCXBs may be low, $\sim 10$ \citep{BT2004a, cooray2004} and 
135: theoretical studies by \cite{BT2004b} indicates that even at the Galactic center,
136:  accreting NS systems do not contribute much to the faint x-ray population.
137: % \cite{intzand2007} 
138: %Population synthesis models \citep{Nelemans??, BT2004} have predicted {\bf XXX} number 
139: %of UCXBs in the Galactic field. It is already clear that the observed systems exceed this predicted number, 
140: %indicating that there are more systems than expected. 
141: 
142: In general, the capabilities of {\it LISA} as a GW detector are usually discussed in
143: the context of the log $(h)$-- log $(f)$ domain.
144: %It was proposed
145:  \cite{KT2007} proposed  that an analogy can be drawn between the astronomy 
146: community's familiar color-magnitude (CM) diagram and {\it LISA}'s 
147:  amplitude-frequency  diagram.
148: % Astronomers can produce a CM diagram that is based on
149: %the apparent brightness (or apparent magnitude {\it m}) of a source from photometric
150: %measurements. However, the intrinsic brightness of each source can be 
151: %determined only if the distance $r$ is measured to the source 
152: %and the corresponding
153: %conversion of $m$ to an absolute magnitude $M$ is made.
154: % However, a determination of the intrinsic
155: %brightness of each source must await the determination of the distance
156: %$r$ to each source and the corresponding conversion of each measured
157: %value of $m$ to an absolute magnitude $M$.
158: % Similarly, LISA's measurement of
159: %$\log(h)$  for a given source only tells us how bright the
160: %GW source appears on the sky, analogous to the measurement of $m$. The 
161: %intrinsic brightness of each GW source can be determined only if the distance $r$
162: %is measured and the corresponding conversion of each
163: %measured value of (the apparent brightness) $h$ to a quantity that
164: %represents the ``absolute'' brightness of the source.
165:  For LISA
166: sources, an analogous quantity to absolute magnitude $M$ is $\log(rh)$, where $r$ is the 
167: distance to the source.
168: %the relevant quantity (analogous to $M$) is $\log(rh)$. 
169: %The underlying physical properties of
170: %individual stars, their evolution, and their relationship to one another in the
171: %context of stellar populations can only be determined from a CM
172: %diagram if $M$, rather than $m$, is used to quantify stellar
173: %magnitudes. By analogy, it should be clear that 
174: The underlying
175: physical properties of compact binary  systems such as DWDs and NSWDs,
176:  their evolution, and their
177: relationship to one another in the context of stellar populations
178: can be ascertained only if the observational properties of such
179: systems are displayed in a $\log(rh) - \log(f)$ diagram, rather than
180: in a plot of $\log(h)$ versus $\log(f)$. \cite{KT2007} discussed the DWD
181: binary systems in this context of ``absolute'' amplitude-frequency domain assuming that
182: the donors in these systems follow zero temperature mass-radius relation. Here, 
183: we will consider the effect of ``warm'' donors as proposed
184:  in \cite{DB2003, Deloyeetal2005, 
185: Deloyeetal2007}
186: and extend the discussion to NSWD binary systems and compare the population
187: boundaries between both of them. We also discuss the limits and applicability of
188: these population boundaries in the log$(rh)$ - log$(f)$ space within the context of
189:  observed AM CVn and UCXB systems.
190: 
191: 
192: %gravitational-wave (GW)
193: %equivalent of a ``color-magnitude'' (C-M) diagram can be constructed for compact binary systems
194: %in circular orbit, based purely on the system's intrinsic parameters such as mass ratio ($q$), total
195: % mass ($m_{tot}$) and orbital separation ($a$). 
196: %%in the
197: %%case of two white dwarfs orbiting each other, based
198: %% purely on  a binary
199: %%system's parameters such as mass ratio ($q$), total mass ($m_{tot}$) and the
200: %%separation ($a$).
201: % Similar to a C-M diagram that
202: %is generally used to describe the evolution of stars from main-sequence to their
203: %end stages, a GW C-M diagram allows us to understand
204: %the various evolutionary stages that a binary
205: %%Double White Dwarf (DWD)
206: % system undergoes
207: %in the distance-independent (``absolute'') amplitude-frequency parameter space.
208: %As an illustration, they considered Double White Dwarf (DWD) systems
209: %in circular orbits and
210: %identified various evolutionary tracks $\&$ population sub-domains that constrain
211: %inspiralling and mass-transferring systems to specific regions.
212: %These regions can be further sub-divided to restrict
213: %the parameter space occupied by stable/unstable mass-transferring systems and progenitors of
214: %%type Ia supernovae.
215: %%The rate at which these systems evolve is a function of the ``chirp'' time,
216: %% $\tau_\mathrm{chirp}$ and hence, lines of constant $\tau_\mathrm{chirp}$ provide 
217: %%an estimate of time spent by these systems at a particular frequency.
218: %% which in turn is an indication of
219: %% number density of systems at that frequency.
220: % Just as in the case of stellar C-M diagram, 
221: %in order to  transform from ``observational'' amplitude-frequency space of LISA to ``absolute''
222: %space, distance to the binary system from the Earth is needed. As shown in previous studies 
223: %\citep{Schutz86, KT2007} a measurement of rate of change of frequency, $\dot f$, by LISA would allow
224: % us to derive the distance to the binary.
225: %% on this 
226: %%parameter space.
227: 
228: % In this discussion, we will map similar sub-domains for Neutron Star (NS) - White Dwarf (WD)
229: %binary systems and in doing so, adopt the same notation and expressions used
230: %in \cite{KT2007}.
231: %We can extend the same argument to neutron star (NS)- white dwarf (WD) binary
232: %systems and identify different population sub-domains that are similar to
233: %DWD systems. In doing so, we will adopt the same notation and expressions used
234: %in \cite{KT2007} and trace out the evolutionary trajectories of NSWD systems
235: %on this GW C-M diagram.
236: 
237: \section{Evolution of WD binaries in the amplitude-frequency domain}
238: \label{sec2}
239: %In order to describe the evolution of NSWD systems in the ``absolute'' 
240: %amplitude-frequency domain of LISA,  we will assume that (1) all orbits are circular,
241: %(2) Kepler's third law holds true between orbital period and separation, (3) the
242: %total angular momentum of the system is given by the orbital angular momentum,
243: %assuming the spins of both the stars are ignored, (4) total mass of the system is 
244: %constant implying that a NSWD binary system undergoes conservative mass transfer 
245: %(CMT) phase and (5)  the angular momentum is lost from the system only through GW
246: %radiation and this loss is accurately described by a quadrupole radiation formula.
247: %Recalling the parameterization used in \cite{KT2007},
248: % the plus and cross polarizations of a gravitational-wave in the quadrupole 
249: %approximation \citep{PM63}
250: % generally takes the 
251: %form $h_\mathrm{+} = h_\mathrm{norm} \cos[\phi (t)]$ and
252: % $h_\mathrm{\times} = h_\mathrm{norm} \sin[\phi (t)]$, where $h_\mathrm{norm} $
253: %is the characteristic amplitude of a gravitational wave given by
254: %\begin{eqnarray}
255: %\nonumber
256: % h_\mathrm{norm} &=& \frac{G}{rc^{4}}\frac{4 \Omega^2 M_\mathrm{1} M_\mathrm{2} a^2}{M_\mathrm{tot}}\\
257: %\label{eq1}
258: %&=&\frac{4}{rc^{4}}\biggl(G M_\mathrm{tot}\biggr)^{5/3}\frac{q}{(1+q)^2}\biggl(\pi f \biggr)^{2/3}
259: %\end{eqnarray}
260: %and the phase angle $\phi(t)$ is a time-dependent function of the
261: % gravitational-wave frequency $f$ of the binary. In Eq.(\ref{eq1}),
262: % $r$ is the distance to the binary from Earth, $\Omega$ is the orbital frequency,
263: %$M_\mathrm{1}$ and $M_\mathrm{2}$ are the individual masses of the stars in the
264: %binary system and we assume that $M_\mathrm{2} < M_\mathrm{1}$, $a$ is the
265: % orbital
266: % separation, $M_\mathrm{tot}$  is the total
267: %mass of the system given by $M_\mathrm{tot} = M_\mathrm{1} + M_\mathrm{2}$ and
268: %$q=M_\mathrm{2}/M_\mathrm{1}$ is the mass ratio.
269: As mentioned in the introduction,
270: since the donor is a WD star and starts filling its Roche lobe to begin  mass transfer phase, 
271: %in a NSWD binary system,
272:  we will denote the donor WD with subscript {\it d} and the
273: accreting companion  as {\it a}. This notation
274: will be followed even during the detached inspiral phase of the evolution.
275:  Hence, the total mass $M_\mathrm{tot} = M_\mathrm{d} + M_\mathrm{a}$ and the mass ratio $q = M_\mathrm{d} / 
276: M_\mathrm{a}$. 
277:  Also we assume that the maximum mass of a WD to be Chandrasekhar mass, $M_\mathrm{ch}=1.44 M_\odot$, the
278: minimum mass of a NS  is $1.2  M_\odot$ and the maximum mass of NS to be $3  M_\odot$.
279: %Recalling the parameterization used in \cite{KT2007},
280: % the plus and cross polarizations of a gravitational-wave
281: % generally takes the
282: %form $h_\mathrm{+} = h_\mathrm{norm} \cos[\phi (t)]$ and
283: % $h_\mathrm{\times} = h_\mathrm{norm} \sin[\phi (t)]$, where $h_\mathrm{norm} $
284: %is the characteristic amplitude of a gravitational wave given by
285: %\begin{eqnarray}
286: %\nonumber
287: % h_\mathrm{norm} &=& \frac{G}{rc^{4}}\frac{4 \Omega^2 M_\mathrm{1} M_\mathrm{2} a^2}{M_\mathrm{tot}}\\
288: %\label{eq1}
289: %&=&\frac{4}{rc^{4}}\biggl(G M_\mathrm{tot}\biggr)^{5/3}\frac{q}{(1+q)^2}\biggl(\pi f \biggr)^{2/3}
290: %\end{eqnarray}
291: %where the phase angle $\phi(t)$ is a time-dependent function of the gravitational-wave frequency $f$
292: %of the binary,
293: % $r$ is the distance to the binary from Earth, $\Omega$ is the orbital frequency,
294: %$M_\mathrm{1}$ and $M_\mathrm{2}$ are the individual masses of the stars in the
295: %binary system and we assume that $M_\mathrm{2} < M_\mathrm{1}$, $a$ is the
296: % orbital
297: % separation, $M_\mathrm{tot}$  is the total
298: %mass of the system given by $M_\mathrm{tot} = M_\mathrm{1} + M_\mathrm{2}$ and
299: %$q=M_\mathrm{2}/M_\mathrm{1}$ is the mass ratio.
300: 
301: %\section{Evolution of NSWD binaries in the amplitude-frequency domain}
302: %\label{sec2}
303: In the case of a detached DWD or a NSWD system inspiralling as a result of loss of
304: angular momentum due to gravitational radiation, we can write,
305:  as shown in \cite{KT2007}, as
306: \begin{eqnarray}
307: rh_\mathrm{norm} &=& \biggl[ \frac{2^5\pi^2}{c^2} \biggl(
308: \frac{GM_\mathrm{ch}}{c^2} \biggr)^5  K^5 f^2 \biggr]^{1/3} = 5.38
309: ~[K^5 f^2]^{1/3}~\mathrm{m} \, , \label{h_f_relationship}
310: \end{eqnarray}
311: %where $M_\mathrm{ch} = 1.44 M_\odot$ is the Chandrasekhar mass limit and
312: %%$M_\mathrm{ch} = 1.44~M_\odot$ is the Chandrasekhar mass and
313: where the dimensionless mass parameter,
314: \begin{eqnarray}\label{K_definition}
315: K \equiv 2^{1/5}\biggl( \frac{\mathcal{M}}{M_\mathrm{ch}} \biggr) =
316: 2^{1/5} \biggl( \frac{M_\mathrm{tot}}{M_\mathrm{ch}}  \biggr)
317: Q^{3/5} = \biggl(\frac{M_\mathrm{a}}{M_\mathrm{ch}}\biggr) \biggl(
318: \frac{2 q^3}{1+q}\biggr)^{1/5} \, ,
319: \end{eqnarray}
320: $\mathcal{M} = M_\mathrm{tot}Q^{3/5}$ is the chirp mass and $Q \equiv q/(1+q)^2$.
321: Notice that, for DWDs,
322:  the maximum value of $K = 1$ occurs at $\mathcal{M} = 1.25 M_\odot$
323:  ($q=1, M_\mathrm{tot} = 2.88 M_\odot$)
324: but for NSWD 
325: binaries, the maximum value of $K$ is $1.39$ and occurs at $\mathcal{M} = 1.75 M_\odot$
326: ($q=0.5, M_\mathrm{tot} = 4.32 M_\odot$). This is purely due to the upper limit on the 
327: mass of the WD, $M_\mathrm{ch}$. Similar to a DWD system, where a $ K =1$
328:  creates a maximum inspiral trajectory in the log$[rh_\mathrm{norm}]-$log$[f]$
329: plane,  a
330:  $K = 1.39$ makes the limiting inspiral trajectory for NSWD binary systems 
331: in this plane.
332: This limiting trajectory, for both DWD $\&$ NSWD is plotted in  Figs. \ref{fig1} $\&$ \ref{fig2},
333: respectively, as a red line with a 
334: slope of $2/3$ (see Eq.(\ref{h_f_relationship}))
335:  beyond which a detached inspiralling DWD or NSWD system can not be found.
336: %representing the maximum (top green dashed line) and minimum (bottom green dashed line)
337: %allowable inspiral trajectories
338: % The minimum allowable inspiral
339: %trajectory is an artificial lower bound because, for the purpose of illustration
340: %it was plotted ``assuming'' that a WD
341: %can  not have a mass lower than $0.0036 M_\odot$ and the maximum mass for a NS is $3 ~ M_\odot$
342: %($q = 0.001$ and $M_\mathrm{tot} = 3 M_\odot$). 
343: % and
344: %the minimum allowable inspiral trajectory (MINIT, bottom green dashed line) below which a
345: % detached inspiralling NSWD can not be found.
346: %In the case of DWD systems,  \cite{KT2007} showed that MAXIT can be obtained for a system with
347: % $q=1$ and $M_\mathrm{tot} = 2.88 M_\odot$ ($M_\mathrm{1} =
348: %M_\mathrm{2} = 1.44 M_\odot$). Unlike DWDs, MAXIT for NSWD systems instead happens at
349: %$q=0.5$ and $M_\mathrm{tot} = 4.32 M_\odot$ because the parameter $K$ (or the chirp mass
350: %$\mathcal{M}$) reaches maximum at these values due to the constraints on NS and WD masses.
351: %The MINIT is sort of an artificial lower bound  because, for the purpose of illustration
352: %it was plotted ``assuming'' that a WD
353: %can  not have a mass lower than $0.0036 M_\odot$ and the maximum mass for a NS is $3 ~ M_\odot$
354: %($q = 0.001$ and $M_\mathrm{tot} = 3 M_\odot$).
355: 
356: 
357: The phase of a detached inspiral evolution terminates when the low mass WD companion comes
358:  into contact with it's
359: Roche lobe to initiate a phase of mass-transfer.  Since the
360:  donor is a WD star this contact period can be found by equating the radius of the
361:  star  with the Roche lobe radius.  Here, for the radius of the WD, we use 
362:  the mass-radius (M-R) relationship of non-zero entropy donor models of  \cite{DB2003}.
363: These models were initially developed
364: for ultra-compact x-ray binaries (UCXB) with WD donors.
365: Later \cite{Deloyeetal2005} applied them 
366:  to model the donors in AM CVn systems formed through WD channel.
367: For these models the donors are assumed to be fully convective and  corresponding
368: donor evolution is considered to be adiabatic in nature due to the large mass-
369:  transfer rates produced in AMCVn systems. However, \cite{Deloyeetal2007}
370:  showed that
371: the assumption of adiabatic evolution is applicable during the beginning of mass-transfer phases
372:  (when the mass-transfer rates are high) and may not work when the system evolves to larger 
373: orbital periods with low mass-transfer rates. Instead, they have identified new evolutionary
374: phases and we will also discuss the impact of this most recent study 
375:  on our population boundaries. We illustrate these boundaries by assuming that the donor
376:  composition is He, but it is straight forward to extend the discussion to
377: carbon (C) $\&$ oxygen (O) donors.
378: In this section, we will discuss in detail the population boundaries of both DWD 
379: $\&$ NSWD. In \S\ref{sec3} we will discuss the applicability of these boundaries
380: with respect to LISA observations, along  with the effects of new phases of evolution
381:  detailed in \cite{Deloyeetal2007}. Finally, we summarize in \S\ref{sec4}.
382: 
383: %Eggleton's mass-radius (M-R) relationship for zero-temperature white dwarfs\footnote{
384: %The effect of finite-temperature models, developed by \cite{DB2003}, on our 
385: %analysis
386: %is discussed in a later section. For the purpose of depicting these population 
387: %boundaries in the GW ``absolute'' amplitude-frequency domain and to draw parallels
388: % between  CM diagram that is generally used in astronomy, we will assume the 
389: %zero-temperature M-R relationsship for a white-dwarf.},
390: % as quoted by \cite{VR88}
391: %and also by \cite{MNS}:
392: % The Roche lobe radius is taken from \cite{EGG}.
393: % The expressions for the
394: %WD radius and the Roche lobe radius, respectively, are:
395: 
396: %\begin{eqnarray}
397: %\frac{R_d}{R_\mathrm{\sun}} = 0.0114
398: %\biggl[\biggl(\frac{M_d}{M_\mathrm{ch}} \biggr)^{-2/3} -
399: %\biggl(\frac{M_d}{M_\mathrm{ch}}\biggr)^{2/3}\biggr]^{1/2}
400: %\biggl[1+3.5\biggl(\frac{M_d}{M_\mathrm{p}}\biggr)^{-2/3}+
401: %\biggl(\frac{M_d}{M_\mathrm{p}}\biggr)^{-1}\biggr]^{-2/3} \, ,
402: %\label{mass_radius}
403: %\end{eqnarray}
404: 
405: %and for Roche lobe radius, we use the expression from \cite{EGG},
406: %\begin{eqnarray}
407: %R_\mathrm{L} \approx a \biggl[ \frac{0.49 \ q^{2/3}}{0.6 \ q^{2/3} +
408: %\ln(1 + q^{1/3})}\biggr].
409: %\label{roche_radius}
410: %\end{eqnarray}
411: %where $R_d$ and $M_d$ are the radii and mass of the WD respectively,
412: % $M_p = 0.00057 M_\odot$ and $R_L$ is the radius of the Roche lobe.
413: %Since there is a lower mass limit for a NS (assumed $1.44 m_\odot$ here)
414: %and hence a minimum $q$ there is a ``minimum contact boundary'' below which
415: %The two red curves shown in Fig. \ref{fig1} trace out the locus of termination (contact) points for 
416: %various detached inspiralling NSWD binary systems.
417: 
418: \subsection{Population boundaries for DWD systems}
419: In Fig. {\ref{fig1}},  We show the population boundaries for both inspiralling $\&$ 
420: mass-transferring phases of DWD systems. The (red) line with slope $=2/3$ shows the
421: maximum inspiral trajectory ($q=1, M_\mathrm{tot} = 2.88 M_\odot$) from a detached DWD system.
422: Above this boundary, no DWD binary can exist.
423: The lower (green cross) curve shows the locus of
424: the termination (contact) points of all detached inspiralling DWD binary systems, which
425: have $q=1$, assuming that the WD is a He donor and has a temperature of $T = 10^{4} K$ (which is 
426: equivalent to the boundary obtained with $T=0$ M-R relation; along this
427: isotherm, $M_\mathrm{tot}$ decreases from top to bottom). For 
428: comparison, we have also shown the contact boundary for $q=1$, generated by 
429: Eggleton's mass-radius (M-R) relationship for $T=0$ white dwarfs, (blue
430: dashed line), as quoted by \cite{VR88}
431: and also by \cite{MNS}.
432:  At higher amplitudes and frequencies, both the curves overlap each other.
433: This is due to the overlap of  M-R relations at higher masses as
434: these objects are supported by degeneracy pressure and the $T=0$ M-R relation that we
435: are using is applicable for fully degenerate He donors.
436: At lower amplitude and frequencies, the contact boundaries diverge slightly.
437: This is because there is a turn over in the M-R relation at low masses for $T=10^{4}K$ as 
438: the Coulomb interactions
439: dominate thermal contribution at that low temperatures and the  M-R relations deviate.
440: So this ($q=1$) contact curve represents the boundary beyond which no DWDs with He 
441: donors will be found. The corresponding boundaries for C $\&$ O would lie just below
442: this contact boundary, except that all three (He, C $\&$ O) boundaries merge at 
443: high frequencies and deviate at low frequencies.
444: 
445: Once the mass transfer phase is initiated, the orbital separation $a$ starts increasing
446:  and the system evolves to lower
447: amplitudes and frequencies. The evolutionary trajectories can be then traced out
448:  by assuming that
449: the donor WD star is marginally in contact with it's Roche lobe and conservative mass 
450: transfer (CMT) holds true.
451:  For DWDs, Since we assume that the maximum mass in a DWD system
452:  is $M_\mathrm{ch}$,   we can
453:  generate a upper boundary beyond which no mass-transferring DWD system can exist.
454: This limiting boundary for DWDs can be
455: obtained for systems with $M_\mathrm{tot} > M_\mathrm{ch}$ (because, then,
456:  eventually $M_\mathrm{a}$
457: will exceed $M_\mathrm{ch}$) when their $q$ drops below a critical $q_\mathrm{ch}$
458: \begin{eqnarray}
459: q_\mathrm{ch} = \frac{M_\mathrm{tot}}{M_\mathrm{ch}} - 1
460: \label{qch}
461: \end{eqnarray}
462:  Accordingly, the (green) dashed line 
463: in Fig. \ref{fig1} indicates an isotherm with Log$[T]=7.5$ and
464:  $M_\mathrm{a} = 1.44 M_\odot$. We note that Log$[T]=7.5$ 
465:  upper boundary is likely a
466: generous overestimate of the available phase space even if the donors were
467: to evolve adiabatically from contact.
468: Assuming that there are no donor WDs in a DWD system with Log$[T] > 7.5$, this curve
469:  represents the
470: boundary beyond which no mass transferring DWDs can exist.  Since the \cite{DB2003} 
471: donor models have two branch nature, with a
472: minimum attainable mass, it is  reflected in this boundary as it curves up,
473:  with a minimum attainable GW amplitude for a given isotherm. 
474: Corresponding boundary for donors with C $\&$ O composition would lie below this
475:  He curve because a higher temperature is
476: required for a C/O donor to fill the Roche lobe than for a  He donor.
477: % For donors
478: %with $Log[T] < 7.5$, the limiting boundaries with $M_\mathrm{a} = M_\mathrm{ch}$
479: %would lie below below the $Log[T] = 7.5$ isotherm.
480:   For 
481: comparison, a similar boundary that arises from $T=0$ M-R relation is also shown as
482: (light blue) dot-dashed curve.
483:  As is the case with $q=1$ contact boundary, these two
484: curves overlap at high frequencies.
485: 
486: From the above discussion, we note that all mass-transferring DWDs with different
487: donor masses, composition and temperatures  should be constrained within these two
488:  boundaries\footnote{For C $\&$ O composition donors, as discussed above, there is a
489: slight deviation at the lower contact boundary.}.
490:  Compared to T=0 model, as assumed in \cite{KT2007}, hot donors in mass-transferring
491: DWDs occupy larger fraction of space in $rh-f$ domain. This is because
492: hot donors are also more massive, therefore have higher intrinsic GW amplitude and come into
493: contact at lower frequencies because of larger radius.
494: 
495: 
496: 
497: 
498: 
499: 
500: % keeping the 
501: % minimum mass of a NS at $1.44 M_\odot$ with varying $M_\mathrm{tot}$.
502: % Similarly, the upper (blue) dashed curve
503: %is the locus of termination points for detached inspiralling NSWD systems with  NS 
504: %mass of $3 M_\odot$( assumed maximum mass for a NS here).
505: %% The region indicated as {\bf ``V''}, then, 
506: %%consists of all mass-transferring NSWD systems, beyond which none will be found. 
507: %Once the mass transfer phase is initiated, the orbital separation $a$ starts increasing and the
508: %system evolves to lower
509: %amplitudes and frequencies. The evolutionary trajectories can be then traced out by assuming that
510: %the donor WD star is marginally in contact with it's Roche lobe and conservative mass transfer (CMT)
511: %holds true.
512: % For DWDs, 
513: % an upper bound for these CMT trajectories can be obtained 
514: %which then defines a boundary within which mass-transferring DWDs
515: %will exist. This limiting boundary for DWDs can be 
516: %obtained for systems with $M_\mathrm{tot} > M_\mathrm{ch}$ (because, then,
517: % eventually $M_\mathrm{a}$
518: %will exceed $M_\mathrm{ch}$) when their $q$ drops below a critical $q_\mathrm{ch}$
519: %%\begin{eqnarray}
520: %%q_\mathrm{ch} = \frac{M_\mathrm{tot}}{M_\mathrm{ch}} - 1
521: %%\label{qch}
522: %%\end{eqnarray}
523: %Not surprisingly, this very same boundary
524: %becomes a {\it lower} termination bound (bottom green dashed curve in
525: % Fig. \ref{fig1})for detached inspiralling
526: % NSWD systems because the minimum mass for a NS is assumed here to be $M_\mathrm{ch}$. 
527: %%Hence, the {\it minimum} mass ratio a NSWD should have is given by
528: %%$q_\mathrm{ch}$.
529: %We can apply Eq.(\ref{qch}) for getting the upper termination boundary (top blue
530: %dashed  curve in
531: % Fig. \ref{fig1}) for NSWD systems with the only change being that $M_\mathrm{ch}$ in 
532: %Eq.(\ref{qch}) should be replaced by the maximum mass of a NS, $3 M_\odot$.
533: % In that case, for NSWD
534: %systems with $M_\mathrm{tot} > 3 M_\odot$, the critical mass ratio $q_\mathrm{NS}$ below which
535: %a NSWD binary can not exist, is
536: %\begin{eqnarray}
537: %q_\mathrm{NS} = \frac{M_\mathrm{tot}}{3 M_\odot} - 1
538: %\label{qNS}
539: %\end{eqnarray}
540: %% In case of NSWD systems, the CMT systems are
541: %%also bounded by a {\it minimum} and {\it maximum} boundaries because there is a minimum and
542: %%maximum mass for a NS. These two are shown in Fig.\ref{fig1} by two red
543: %% curves. The bottom redu curve is the locus of 
544: %% the bottom red curve
545: %%for the minimum and top red curve for the maximum mass of a NS, respectively. 
546: % The region indicated as {\bf ``V''} in Fig.\ref{fig1}, then, 
547: %consists of all mass-transferring NSWD systems, beyond which none will be found. 
548: 
549: The inspiral phase of a WD binary evolution is driven by the loss of the 
550: angular momentum due to GW radiation and the inspiral evolutionary time scale 
551: $\tau_\mathrm{chirp}$ can be written as:
552: % Even in the CMT phase, the driving mechanism is the 
553: %emission of GW radiation from the system, so that the donor stays in contact with its 
554: %Roche lobe.
555: % Consequently, the time dependent behavior
556: %of the orbital angular momentum $J_\mathrm{orb}$ can be written as
557: %\citep{PM63}
558: %\begin{equation}
559: %\label{JofTviaGR}
560: %J_\mathrm{orb}(t) = J_0 (1-t/\tau_\mathrm{chirp})^{1/8} \, ,
561: %\end{equation}
562: %where the inspiral evolutionary time scale is,
563: \begin{eqnarray}
564: \tau_\mathrm{chirp} &\equiv& \frac{5}{256} \frac{c^5 a^4}{G^3
565: M_\mathrm{tot}^{3}} \biggl[\frac{(1+q)^2}{q}\biggr] = \frac{ 5}{64
566: \pi^2 } \biggl( \frac{c}{rh_\mathrm{norm} f^2} \biggr) \, . \label{tau_chirp}
567: \end{eqnarray}
568: It is worth noting that the time scale of evolution in both inspiral and 
569: CMT phases is of the order of $\sim \tau_\mathrm{chirp}$ \citep{KT2007}. 
570: Therefore we have drawn ``chirp'' isochrones also in Fig.\ref{fig1} and each
571: isochrone has a slope of $-2$ in this log$(rh_\mathrm{norm})$ - log$(f)$ space,
572: as can be seen from Eq.(\ref{tau_chirp}) due to the dependence on the product
573: of $rh_\mathrm{norm}$ and $f^{2}$. This implies that a given WD binary system, 
574: in either inspiral or CMT phase, spends $\tau_\mathrm{chirp}$ amount of time
575:  corresponding to the GW frequency that the binary is emitting.
576:  At lower frequencies, the value of $\tau_\mathrm{chirp}$ is larger compared to
577: that of at higher frequencies. A consequence of this behavior is that more
578:  binary systems accumulate at lower frequencies. As mentioned in the
579:  introduction, the background noise 
580: arising due to millions of DWDs at the lower frequency band ($\leq 3 \times 
581: 10^{-3}$ Hz) of LISA is due to the fact that the value of $\tau_\mathrm{chirp}$
582: for these systems is $\sim 10^{10}$ years, approaching Hubble time (see 
583: Fig.\ref{fig2}). 
584: 
585: 
586: % Regione where no NSWD systems can be found (either detached or mass transferring)
587: % and chirp isochrones with a slope of $-2$ 
588: % indicating the timescale of evolution \citep[see][for details]{KT2007}
589: %are also shown in Fig.\ref{fig1}. 
590:  
591: \clearpage
592: \thispagestyle{empty}
593: \begin{figure}[!hbp|t]
594: %\vspace*{-25mm}
595: %\centering
596: \epsscale{0.8}
597: %\plotone{f1.eps}
598: \includegraphics[height=13.0909cm,width=15.8779cm,angle=360]{f1.eps}
599: \hfill \caption{ Population boundaries for DWD systems. The (red) line with positive
600: slope represents the maximum inspiral evolution trajectory for DWD systems, beyond which 
601: none will be found. This happens at $q = 1, M_\mathrm{tot} = 2.88 M_\odot$. The bottom
602: (green) curve with crosses is drawn using the M-R relation of isentropic donor models
603: from \cite{DB2003}. It represents the termination (contact) boundary for $q=1$ 
604: systems, assuming that the donors are of He composition with $T=10^{4} K$. For 
605: comparison, contact boundary derived using Eggleton's $T=0$ M-R relation for fully
606:  degenerate He donors is also shown as (blue) dashed line. The top (green) dashed 
607: curve represents the boundary at which $M_\mathrm{a} = M_\mathrm{ch}$, assuming that
608:  donors in DWD systems would have a  maximum $T = 10^{7.5} K$. No mass-transferring DWD
609: would be found above this boundary. For comparison, the (blue) dot-dashed curve shows 
610: a similar boundary for $T=0$ donor models. Also shown are constant $\tau_\mathrm{chirp}$
611: lines along with observed AMCVn systems (blue dots; see text). 
612: Explanation for vertical
613: (brown) dot-dashed line is given in \S\ref{sec3}.
614: % Unlike DWDs, where the limiting inspiral trajectory is for
615: % $q = 1, M_\mathrm{tot} = 2.88 M_\odot$, the 
616: %maximum inspiral trajectory for detached NSWDs is obtained when
617: % $q = 0.5, M_\mathrm{tot} = 4.32 M_\odot$ due to limit on the maximum 
618: %mass of the WD ($M_\mathrm{ch}$).
619: % The minimum 
620: %inspiral trajectory ($q = 0.001, M_\mathrm{tot} = 3 M_\odot$) is an artificial boundary
621: % because we assume that a WD can not have a mass lower
622: %than $0.0036 M_\odot$.
623: % The  
624: %region indicated as ``{\bf V}'' represents the region where mass-tranferring NSWDs can exist,
625: % the ( green) lower dashed curve and  the (blue) upper dashed curves are the locus of
626: % termination (contact)
627: % points of detached inspiralling
628: %NSWD systems with NS mass of $1.44 M_\odot$ and $3 M_\odot$,
629: % respectively.  The (orange) filled circle represents the
630: %contact point for the maximum inspiral evolution trajectory. Several chirp isochrones
631: %(see Eq.(\ref{tau_chirp}) and corresponding discussion), indicating the time scale of the
632: %evolution,  are also shown.
633: }
634: \label{fig1}
635: \end{figure}
636: \clearpage
637: 
638: \subsection{Population boundaries for NSWD systems}
639: In Fig. \ref{fig2}, population boundaries for both NSWD systems is shown in the 
640: log$(rh_\mathrm{norm})$ - log$(f)$ space. The top (red) curve represents the maximum
641: inspiral boundary for detached NSWD binary systems ($K = 1.39$) beyond which none will
642: be found. The bottom (green) curve with crosses represents the lower contact boundary,
643: drawn using \cite{DB2003} hot donor models, assuming that the donor is of He composition
644:  with $T=10^{4} K$. The mass of the 
645: NS on this boundary is set to the minimum mass assumed, $1.2 M_\odot$. For comparison,
646: a similar boundary using $T=0$ M-R relation for He donors
647: is also shown as (blue) dashed curve. As is the case with DWDs, these two boundaries
648: overlap at higher amplitudes and frequencies because at higher donor masses, the two
649: M-R relations match as they are supported by degeneracy pressure. Furthermore, there is an
650: overlap region between the lower NSWD contact boundary (green dashed curve in Fig. \ref{fig2})
651:  $\&$ the $M_\mathrm{a} = M_\mathrm{ch}$ DWD contact boundary
652: (green dashed curve in Fig. \ref{fig1}), indicating both types of systems can exist within
653: this region.  Similarly, an upper 
654: contact boundary can also be drawn for NSWDs, shown as (green) dashed curve in Fig. \ref{fig2}, 
655: assuming the maximum mass of the NS is $3.0 M_\odot$ and that the maximum donor 
656:  temperature in a NSWD system to be $T=10^{7.5} K$.
657:  The light blue dot-dashed curve is the upper contact boundary for $T=0$ donors.
658: Just like DWDs, for NSWD systems with $M_\mathrm{tot} > 3 M_\odot$, the critical
659:  mass ratio $q_\mathrm{NS}$ below which a NSWD binary can not exist, is
660: \begin{eqnarray}
661: q_\mathrm{NS} = \frac{M_\mathrm{tot}}{3 M_\odot} - 1
662: \label{qNS} 
663: \end{eqnarray}
664: So all the mass-transferring NSWD binary systems are bounded by the upper $\&$ lower
665:  (green) curves\footnote{The lower boundary is applicable for He donors.}
666:  and all the detached inspiralling NSWDs will be bounded between the 
667: top (red) curve and the bottom (green) curve.
668: Fig. \ref{fig2} also shows chirp isochrones,  indicating the evolutionary timescales.
669: 
670: % keeping the 
671: % minimum mass of a NS at $1.44 M_\odot$ with varying $M_\mathrm{tot}$.
672: % Similarly, the upper (blue) dashed curve
673: %is the locus of termination points for detached inspiralling NSWD systems with  NS 
674: %mass of $3 M_\odot$( assumed maximum mass for a NS here).
675: %% The region indicated as {\bf ``V''}, then, 
676: %%consists of all mass-transferring NSWD systems, beyond which none will be found. 
677: 
678: \clearpage
679: \thispagestyle{empty}
680: \begin{figure}[!hbp|t]
681: %\vspace*{-25mm}
682: %\centering
683: %\epsscale{0.6}
684: %\plotone{f2.eps}
685: %\plotone{f3.eps}
686: \includegraphics[height=13.0909cm,width=15.8779cm,angle=360]{f2.eps}
687: %\includegraphics[height=8cm,width=9.0909cm,angle=360]{f3.eps}
688: \hfill \caption{ Population boundaries for NSWDs in the
689:  log$[rh_\mathrm{norm}]$-log$[f]$ plane.
690: The red line represents the maximum inspiral trajectory from a 
691: detached NSWD binary system ($q=0.5, M_\mathrm{mtot} = 4.32 M_\odot$). The bottom
692: curve with (green) crosses represents the contact boundary, assuming minimum mass of
693: the NS to be $1.2 M_\odot$ with  He donors of $T=10^{4} K$. The blue dashed line is
694: drawn assuming a $T=0$ M-R relation for fully degenerate He donors. These two boundaries
695: overlap at high masses (high amplitude $\&$ frequency) and deviate slightly at lower 
696: masses. The top (green) dashed curve shows the upper bound for mass-transferring NSWD
697: systems, assuming that the maximum donor temperature in a NSWD system to be $T=10^{7.5} K$
698: and maximum mass of NS as $3.0 ~ M_\odot$. To compare, an upper boundary drawn
699: using $T=0$ donor model is also shown as (light blue) dot-dashed curve.  Also shown are
700: currently known UCXB systems (blue dots) along with chirp isochrones.}
701: % Region {\bf V} represents the space where mass-transferring
702: % NSWDs can exist. The blue stars are the known UCXBs (see Table \ref{table1} and
703: %\S\ref{UCXB_section}) which 
704: %we assume here to be mass-transferring NSWDs. The vertical dashed lines represent the clustering
705: %of UCXBs between orbital periods of $40 \sim 50$ minutes. The open (pink) circles represent
706: %the UCXB systems for which masses are unknown, so arbitrary values that
707: % would place these systems within Region {\bf V} were assigned without changing their known
708: % orbital periods.
709: %{\it Right panel:} Population boundaries for DWDs as shown in \cite{KT2007}. The top red line
710: %is the maximum inspiral trajectory ($q=1$ and $M_\mathrm{tot} = 2.88 M_\odot$) 
711: %beyond which inspiralling DWDs can not exist. The lower red line represents the locus of 
712: %termination (contact) points of all inspiralling systems with $q=1$, with different 
713: %$M_\mathrm{tot}$. Inspiralling systems can reside anywhere in region {\bf I}. The green
714: %dashed line represents the curve where $M_\mathrm{a} = M_\mathrm{ch}$ for DWD systems in 
715: %CMT phase. Hence mass transferring DWDs are constrained to region {\bf II}.
716: %The observed detached DWDs (cross) and AM CVn systems (plus) are also shown.}
717: \label{fig2}
718: \end{figure}
719: \clearpage
720: 
721: \section{Discussion}
722: \label{sec3}
723: The population boundaries for DWD $\&$ NSWD binary systems discussed in
724:  previous sections were generated using \cite{DB2003} non-zero entropy donor
725: models,  which assume that the donors are fully convective and undergo
726: adiabatic evolution throughout  the mass-loss phase. Recently, \cite{Deloyeetal2007}
727: showed that these assumptions may not properly estimate the donor's orbital
728: period evolution and specifically the donor's adiabatic evolution may not hold true
729:  for whole mass-transfer phase. Instead they identified three distinct 
730: evolutionary phases and here we discuss the implications of their new findings
731: on our population boundaries.
732: 
733: During the first phase, which happens during the mass-transfer ``turn-on'' phase
734: (when the donor comes into contact initially), the radius of the donor decreases
735: , the mass-transfer rate $\dot M_\mathrm{d}$ increases and 
736: the orbital period $P_\mathrm{orb}$ continues to decrease until the donor 
737: radius reaches it's minimum value ($\dot M_\mathrm{d}$ becomes maximum)
738: and starts expanding again.
739:  \cite{Deloyeetal2007} calculated that this turn on phase
740: lasts up to $\sim  10^{6}$ years. 
741: In the second phase, the donor responds to the mass loss
742: adiabatically and starts expanding, which is considered to be the normal 
743: AM CVn phase. 
744: But this phase of adiabatic expansion ends and a third phase of evolution 
745: begins at around $P_\mathrm{orb} \sim 45$ min,
746: when the mass-transfer rate (and the donors thermal time) drop enough for
747:  the donor to cool and start contracting
748: to a fully degenerate configuration, stalling the $P_\mathrm{orb}$ evolution. 
749: 
750: In the case of DWDs, the initial turn-on phase will result in the $q=1$ contact
751:  boundary (green curve with crosses in Fig. \ref{fig1}) to 
752:  have a width, instead of a sharp boundary as discussed in \cite{KT2007}.
753: % towards the higher frequencies.
754:  This is because it is calculated assuming that once the system comes into
755: contact, it would evolve towards lower amplitudes and frequencies, whereas
756:  $P_\mathrm{orb}$ decreases in the turn on phase, to a minimum even after the
757: initial contact. Accordingly, there will be a slight overlap of lower contact boundaries
758: for He, C $\&$ O donors. Once the system evolves off this initial contact phase, the donor
759: expands adiabatically in response to the mass-transfer and the system follows a
760: typical AM CVn evolution,  where the GW amplitude $\&$ frequency keeps decreasing
761: as $P_\mathrm{orb}$ increases. Assuming an adiabatic evolution means that the cooling
762: time of the donor is longer than the mass-transfer time-scale 
763: ($M_\mathrm{d} / \dot M_\mathrm{d}$), it will in turn affects the
764: orbital evolution time-scale. Accordingly, the donor follows a trajectory where it 
765: passes through different isotherms of decreasing $q$, after the initial contact. As \cite{KT2007} illustrate,
766: the lower right curve in Fig. \ref{fig1} shows the contact boundary for $q=1$ 
767: systems and similar contact boundaries for lower $q$'s would 
768: lie to the left of it, but the 
769: contact boundaries will shift towards lower amplitudes and frequencies.  
770: This phase of adiabatic evolution comes to an end between
771:  $P_\mathrm{orb} \approx 40-55$ min, when the donor starts to cool and contract 
772: eventually towards a fully degenerate  star. 
773: This will drastically (almost an order of magnitude, see \cite{Deloyeetal2007} Fig. 15)
774: reduce $\dot M_\mathrm{d}$ at these long orbital periods.
775:  Hence, after this range of $P_\mathrm{orb}$, the systems
776: GW frequency evolution slows down in accordance with the drop in $P_\mathrm{orb}$ evolution. If
777: the donor has cooled enough to approximate it as a $T=0$ degenerate model, then 
778: it may lie close to
779: one of the $T=0$ contact boundaries, depending upon its $q$. But these $T=0$
780: boundaries are bounded within the
781: region constrained by $M_\mathrm{a} = M_\mathrm{ch}$ (light blue dot-dashed) curve 
782:  and the $q=1$ contact boundary (green curve with crosses in Fig. \ref{fig1}), because these boundaries are drawn assuming
783: that the donor is a fully degenerate star with $T=0$ and such a system can not exist beyond
784: these curves. Therefore, we have drawn a vertical
785: (brown) dot-dashed line in Fig. \ref{fig1} at $P_\mathrm{orb} = 55$ min (Log$f = 3.21$) beyond which we would expect
786: these systems to be within the $T=0$ contact boundaries. Note that this vertical line
787: is not a sharp boundary: some systems may not reach a fully degenerate
788:  configuration by this $P_\mathrm{orb}$. Rather, a system's $P_\mathrm{orb}$
789: evolution slows down at a particular GW frequency
790: once they start cooling towards a degenerate configuration at the above mentioned $P_\mathrm{orb}$.
791: Fig. \ref{fig1} also shows the observed AM CVn type systems, for which the masses and $P_\mathrm{orb}$
792:  are taken from \cite{Deloyeetal2005} $\&$ \cite{Roelofs2007}. Couple of them are clearly outside $T=0$ 
793: region; for some of them, the system's mass function limits the minimum donor
794: mass to a value above that of a Roche-filling $T=0$ donor at the same 
795: $P_\mathrm{orb}$.  It is very difficult to know their exact temperature and/or
796: composition purely from GW observations, as there is a good overlap of systems with these 
797: characteristics\footnote{Out of these systems, 
798: CE 315 (the system with lowest GW amplitude and frequency in Fig. \ref{fig1})
799:  has $P_\mathrm{orb} = 65$ min and lies to the
800: left of the $P_\mathrm{orb} = 55$ line and below
801:  $T=0, M_\mathrm{a} = M_\mathrm{ch}$ boundary. This may seem to imply that this system may be
802:  cooling off and trying to reach a $T=0$ degenerate configuration. But in Fig 1. of 
803: \cite{Deloyeetal2005}, they give a temperature range of Log $T = 4.0-6.5$ and \cite{Bildsten2006} noted
804: that this system may have a hot donor.  If that is the case, \cite{Bildsten2006} note that this donor
805: may have evolved with constant entropy that it was born with or may have been heated by either the disk
806: or the accretor. However, if irradiation is the cause, then \cite{Deloyeetal2007} note that it will 
807: delay the onset of donor's cooling and  increases the temperature of the donor.
808: So though this system lies below the  $T=0, M_\mathrm{a} = M_\mathrm{ch}$
809:  boundary, it does not necessarily mean that the donor can be approximated
810: as a zero-temperature object. }.
811: 
812: For NSWDs, similar to DWDs, the  contact boundaries  with He composition drawn  assuming 
813:  NS mass is $1.2 M_\odot$ (green curve with crosses ) and
814:  $3.0 M_\odot$ (green dashed curve)
815:  will also have a width due to decrease in $P_\mathrm{orb}$ even after the contact. Furthermore, 
816: the lower contact boundary lies {\it below} the upper contact boundary for DWDs
817: ($M_\mathrm{a} = 1.44 M_\mathrm{\odot}$). This will make
818: the contact boundaries to ``overflow'' into the DWD region and LISA observations may not be able
819: to distinguish these two types of systems in this region. Here also the 
820: donor undergoes adiabatic evolution after $\dot M_\mathrm{d}$ reaches a maximum and consequently enters
821: the cooling phase at $P_\mathrm{orb}\approx 55$ min. Accordingly, in Fig. \ref{fig2}, the brown line shows
822: the orbital period of this transition phase. Beyond this line, the donors should start cooling and 
823: GW frequency slows down accordingly.
824: %inwards of the corresponding contact boundaries for $T=0$ (light blue dot-dashed and blue dashed lines).
825: Fig. \ref{fig2} also shows the currently observed UCXBs, which we assume to be mass-transferring NSWD
826: binary systems. Table \ref{table1} gives the values of the masses of donors and orbital periods that
827: we used. Some of the UCXBs shown in Fig. \ref{fig2} lie {\it below} the lower contact boundary (green
828: cross curve), indicating that probably the donors are not of He composition. Although, this contact
829: boundary has a width, and hence they may have He composition, it is unlikely
830:  that we
831:  can observe them during this relatively short lived phase. Moreover, these systems are plotted assuming
832: the minimum mass of the donors mass range 
833: derived from observations, so this provides additional uncertainty in
834: determining the composition of donors. 
835:  It is unlikely that LISA will be able to observe  cooling donors in either DWD or NSWD systems
836: because of the instrumental and/or DWD background noise.
837: In Fig.\ref{fig3}, we plot the known UCXBs and AM CVns on top of LISA's
838:  sensitivity curve\footnote{http://www.srl.caltech.edu/~shane/sensitivity/MakeCurve.html}
839:  (SNR = 1) to assess the detectability of these systems. In the case of known UCXBs,
840: it is clear that only one system (4U 1820-30) has enough signal strength to be visible to LISA,
841:  whereas some of the known AM CVns emit GWs above the instrumental and DWD background noise. 
842: 
843: 
844: %\subsection{Comparing population boundaries of NSWD and DWD binary systems.}
845: %\label{compare_NSWD_DWD}
846: %In Fig.\ref{fig2}, population boundaries 
847: % for both NSWD (left panel) and DWD systems (right panel) are shown in the 
848: %log$(rh_\mathrm{norm})$ - log$(f)$ space.
849: %% The three thick red curves
850: %% represent
851: %%the boundaries beyond which inspiralling (Region {\bf I}) or mass transferring
852: %%(Region {\bf II }) DWD system
853: %%can not exist.
854: % In the case of DWDs \citep{KT2007}, the top (red) line with positive slope 
855: %indicates the 
856: %maximum inspiral trajectory beyond which no DWD binary system can exist. From 
857: %Eqs.(\ref{h_f_relationship}) \& (\ref{K_definition}), this happens when $q=1$ and 
858: %$M_\mathrm{tot} = 2.88 M_\odot (M_\mathrm{a} = M_\mathrm{d} = 1.44 M_\odot)$.
859: % The bottom 
860: %thick (red) curve represents the locus of termination (contact) points of all detached 
861: %inspiralling DWD systems with $q = 1$, which is the maximum mass ratio, 
862: % for different total masses.
863: %% A DWD system with $q=1$ starts the phase of mass 
864: %%transfer from this curve and evolves to lower amplitudes and frequencies. 
865: %Therefore an inspiralling
866: %DWD system can exist anywhere between these two curves within the region
867: % indicated by {\bf I}. Once the detached inspiral phase of a DWD comes to an
868: % end, the donor WD fills its Roche lobe and mass transfer phase begins. 
869: %As explained in \S\ref{sec2}, then the orbital separation between both the WDs
870: % starts increasing and the system evolves to lower amplitudes and frequencies.
871: %Since, for DWDs, the accretor's mass can not exceed $M_\mathrm{ch}$, another 
872: %boundary arises beyond which mass-transferring DWDs will not exist. The 
873: %(green) dashed line in the right panel of Fig.\ref{fig2}, indicates this boundary
874: % where the 
875: %accretor's mass reaches $M_\mathrm{ch}$ for all DWD binary systems. Hence, the
876: %region bounded between the (green) dashed curve and bottom (red) curve 
877: %(termination boundary for $q=1$ systems) 
878: % indicates that the population of DWDs undergoing CMT can 
879: %exist only in this region indicated as {\bf II}. As in the case of Fig.\ref{fig1},
880: %constant ``chirp'' isochrones are also drawn indicating the 
881: %time scales of evolution for systems in different phases\footnote{Further
882: %sub-domains indicating the regions where stable/unstable mass transferring 
883: %systems and Type Ia progenitors reside, can also be identified, which are
884: %not shown in the right panel of Fig.\ref{fig2} but are included in
885: % \cite{KT2007}.}.
886: % Evolutionary boundaries for NSWD systems, discussed in the
887: %previous section, are also shown. Finally, some of the known short period 
888: %low mass x-ray binaries (LMXB\footnote{\cite{RKcat}, http://physics.open.ac.uk/RKcat/lmxbcat}), given in
889: %Table \ref{table1},
890: % are also shown (blue stars). It should be noted that the
891: % vertical location (log$(rh_\mathrm{norm})$ values) of these LMXBs are somewhat
892: % arbitrary because the masses of these systems are unknown. If we assume that
893: %LMXBs are mass transferring NSWD systems, then following previous section's discussion,
894: % they should be constrained to Region {\bf V}.
895: %%the cyan shaded region.
896: % Accordingly arbitrary values of log$(rh_\mathrm{norm})$, that would place these
897: %systems within this region, were assigned without changing their known orbital
898: %periods.
899: 
900: 
901: Recalling from \S\ref{sec1},  in order to transform from log$(h_\mathrm{norm})$ 
902:  to log$(rh_\mathrm{norm})$  space,
903: %space shown in Figs \ref{fig1} \& \ref{fig2},
904:  we need to know the distance $r$ to the binary system.
905: % This $r$ can be 
906: %determined either through independent electromagnetic (optical/x-ray etc.) 
907: %observations or through the measurement of rate of change of GW frequency,
908: %$\dot f$ by LISA.
909:  The relation between the unknown binary parameters $r$, $M_\mathrm{tot}$
910: and $q$ and the observables $h_\mathrm{norm}$, $f$ and $\dot f$
911:  can be written as \citep{KT2007}
912: \begin{eqnarray}
913: \label{mass_r_relation}
914: \frac{M_\mathrm{tot}^{5}}{r^{3}}\biggl[\frac{q}{(1+q)^{2}}\biggr]^{3} &=& \frac{c^{12}}
915: {2^{6}\pi^{2}G^{5}}\frac{h_\mathrm{norm}^{3}}{f^{2}} , \\
916: r(1 - 2 g) &=& \frac{5 c}{24 \pi^{2}} \frac{\dot f}{h_\mathrm{norm} 
917: f^{3}}
918: \label{r_fdot_relation}
919: \end{eqnarray}
920: where $g=0$ in the inspiral phase of evolution and hence, it is easy to determine $r$
921: from  $h_\mathrm{norm}$, $f$ and $\dot f$ through Eq.(\ref{r_fdot_relation}).
922:  For mass-transferring systems $g$  is a function of $M_\mathrm{tot}$ and $q$, and they can
923: be related to the observable $f$ by the requirement that in the mass
924:  transfer phase, $R_\mathrm{d} = R_\mathrm{L}$. 
925: 
926: %which are
927: %also unknown and Eq.(\ref{r_fdot_relation}) does not provide an explicit relation to
928: %solve for $r$. But, the requirement that in the mass transfer phase, $R_\mathrm{d} = 
929: %R_\mathrm{L}$ provides an additional relationship between the observable $f$ and the 
930: %system parameters $q$ and $M_\mathrm{tot}$. In combination with Eqs.(\ref{mass_r_relation})
931: % \& (\ref{r_fdot_relation}), we can solve for all the unknown system parameters 
932: %$r$, $M_\mathrm{tot}$ and $q$. In general, solving the three system of equations 
933: %analytically is 
934: %difficult due to the complexity of the function $g(q,M_\mathrm{tot})$. However,
935: %\cite{KT2007} considered the formulae that \cite{Paczynski67} adopted for mass-radius
936: %relation for a white dwarf and Roche lobe radius, to simplify the function $g$ and
937: %solved for $r$, $M_\mathrm{tot}$ and $q$ (see their Fig. 2).
938: 
939: %an $\dot f$ or the distance to
940: %the system should be known through an independent method, say an optical and/or x-ray observations.
941:  The determination of $r$ and/or the masses of the stars in DWD/NSWD binary system 
942: depends on the determination of $\dot f$.  If an $\dot f$ can not be measured for a system,
943:  then there is no way to tell whether that system
944: is a NSWD or DWD system based only on LISA observations.
945: % because, then, the transformation
946: %to an ``absolute'' magnitude from the ``apparent'' magnitude space would not be possible.
947:  If an $\dot f$ can be measured, and if it turns out to be negative, then it is possible 
948: that particular system is a mass-transferring system (DWD or NSWD). But as shown in Figs. \ref{fig1}
949:  $\&$ \ref{fig2}, it will still not
950: be possible to know the type of the system, at least for low frequency sources (Log$[f] \lessapprox -2$).
951:  There is a fairly good overlap in 
952: $\dot f$ between DWD $\&$ NSWD systems in this region because of the non-zero entropy nature
953: of the donors and also due to the lower limit on the mass of the NS ($1.2 M_\odot$).
954: But for high frequency mass-transferring sources (Log$[f] \gtrapprox -2$),
955:  it {\it may} still be possible to know the type of the system, as the overlap region 
956: reduces\footnote{There still will be some uncertainty for systems with NS mass lower than
957:  $M_\mathrm{ch}$, but for NS masses higher than  $M_\mathrm{ch}$,
958:  LISA should be able to distinguish both types of systems through the measurement of $\dot f$. }. 
959: % But since these two
960: %    types of systems fall in different regions on the log$(rh_\mathrm{norm})$-log$(f)$
961: % diagram (Regions
962: %{\bf II} \& {\bf V} in Fig.\ref{fig2}, respectively),
963: %    it is also possible to tell
964: %    whether it is a mass transferring NSWD or mass transferring DWD just
965: %    based on the ``value'' of the $\dot f$.
966:  The same thing can be said about inspiralling systems because
967:     there is a large  overlap of
968:     DWD and NSWD inspirals at low frequencies  and even a measurement of positive
969:  $\dot f$ would unlikely be able distinguish these two types of inspiralling systems.
970: 
971: 
972: % There may be a large number of individual DWD systems for which LISA
973: %is able to confidently measure $h_\mathrm{norm}$ and $f$, but not
974: %$\dot f$. Without a direct measure of $\dot f$, it will not be
975: %possible to determine distance to these systems.
976: %Nevertheless, the population boundaries that have been identified  in the
977: %present discussion for DWDs and NSWDs
978: %provide useful limits on the distance $r$ for such systems. Note that this can only be done
979: %for high frequency sources (Log$[f] \gesim 2$ for both DWDs) because both the donor models
980: %match at these frequencies\footnote{For these sources, there is a good chance
981: %of detecting $\dot f$ anyway, but the estimates that we provide here may still be useful
982: %for sources that have no measured $\dot f$ at these frequencies}. As we go to lower
983: %frequencies, the uncertainty will be large on these estimtes due to the consideration of 
984: %non-zero entropy donors. 
985: 
986: % Consider, for
987: %example, a system whose measured GW frequency is Log$[f] = -2.5~ (f = 3.16~ \mathrm {mHz}$) . 
988: %If there is an independent evidence that such a system is a DWD in the CMT
989: %phase of its evolution ({\it e.g.}, the system is known from optical
990: %observations to be accreting) then the system is restricted to Region {\bf II} in
991: %the ``absolute'' amplitude-frequency diagram. 
992: %It then tells us that $r$ must fall in the range,
993: %\begin{eqnarray}
994: %0.000575~ h_\mathrm{norm}^{-1} \leq r (\mathrm{meters}) \leq 0.00616~
995: %h_\mathrm{norm}^{-1} \, .
996: %\end{eqnarray}
997: %%If there is independent evidence that such a system is in the CMT
998: %%phase of its evolution ({\it e.g.}, the system is known from optical
999: %%observations to be accreting) then a tighter constraint on the
1000: %%distance can be obtained by using curve {\bf C} rather than curve
1001: %%{\bf A} as the upper population boundary. 
1002: %On the other hand, if it turns out that the system is a NSWD in CMT phase, then
1003: %$r$ must fall in the range, 
1004: %\begin{eqnarray}
1005: %0.00616 ~h_\mathrm{norm}^{-1} \leq r (\mathrm{meters}) \leq 0.01~
1006: %h_\mathrm{norm}^{-1} \, .
1007: %\end{eqnarray}
1008: %% Because curves {\bf A},
1009: %%{\bf B} and {\bf C} diverge from one another at lower frequencies,
1010: %%the uncertainty in such estimates of $r$ will necessarily be larger
1011: %%for DWD systems with lower GW frequencies.
1012: %% Notice that since mass-transferring NSWD systems  are constrained
1013: %%    in a much smaller region ({\bf V}) than DWDs ({\bf II}), a much tighter constraint on
1014: %% the distances to NSWDs can be obtained.
1015: %% (not really on masses, there is a
1016: %%    significant overlap in determining the masses of mass-transferring NSWD systems from the
1017: %%     measured $\dot f$.)
1018: % A curious issue is as we go down to low frequencies (larger $\tau_\mathrm{chirp}$),
1019: %%    mass-transferring NSWD systems,
1020: % the width of the region that constraints the
1021: %    mass-tranferring NSWD systems (region {\bf V}) becomes small. So if LISA measures
1022: % only $h$ and $f$ for a
1023: %    mass-transferring system  at these low frequencies (and mass-ratios),
1024: %    and if by independent
1025: %     means we know it is a mass-transferring NSWD system, then C-M diagram can give a
1026: % much sharper estimate of the distance than for a DWD.
1027: %    This is in contrast with mass-tranferring DWD systems (AMCVn) where, as we move towards
1028: % low frequencies,
1029: %    the error in distance actually increases due to the increase in the width of
1030: %    Region {\bf II}.
1031: %
1032: % The vertical position (log$(rh_\mathrm{norm})$) is
1033: % somewhat arbitrary for these observed systems because the masses are unknown. Assuming LMXBs are 
1034: %mass-transferring NSWD systems,  arbitrary values that
1035: % would place these systems within Region {\bf V} were assigned without changing their known
1036: % orbital periods.}
1037: %It seems there is a lower frequency limit on mass-transferring NSWD systems.
1038: %For mass-transferring NSWD systems in Fig.\ref{fig2}, the
1039: %    mass-transfer was stopped when the donor WD reached a minimum mass of 0.0036 $M_\odot$. This
1040: %assumes that the lowest mass of a WD is 0.0036 $M_\odot$. 
1041: % If LISA measures
1042: %    a negative $\dot f$ and finds a system below the minimum NSWD inspiral boundary (bottom green
1043: %dashed line in Fig.\ref{fig2}),
1044: %( log$(f) < -3.5$),
1045: % then most probably the donor is not a WD star because
1046: %    then the donor WD should have a mass less than 0.0036 $M_\odot$.
1047: 
1048: %    However, it is unlikely that an $\dot f$ can be measured at such low frequencies
1049: %    ( $< -3.5$).  So if a negative $\dot f$ is measured most probably we can tell
1050: %    what type of systems (NSWD or DWD) it is. On the other hand, since the 
1051: %    assumption is that an $\dot f$ can be measured only at little high frequencies,
1052: %    there is a good chance that the donor in a NSWD is a WD star rather than
1053: %    a Main-sequence/giant.
1054: 
1055: %\item Distances to the known LMXB's\footnote{http://physics.open.ac.uk/RKcat/lmxbcat} can be estimated 
1056: %with the help of Fig.\ref{fig2}. Specifically, for LMXBs above the frequency (log$(f) > -3.5$), a 
1057: %table of estimated distances can be made, if they are already not known.
1058: 
1059: %\subsection{Application to ultra compact x-ray binaries (UCXB)}
1060: %\label{UCXB_section}
1061: %The formation and evolution of semi-detached interacting binaries with low mass
1062: %secondaries has been studied extensively in the past \citep[CVs;][]{Warner1995};
1063: % \citep[x-ray binaries;][]{VandenHeuvel1983}. Of particular interest to our 
1064: %present discussion are a subclass of low mass x-ray binary (LMXB) systems called
1065: %ultra compact x-ray binaries (UCXB) whose orbital periods are $\leq 80$ minutes.
1066: %LMXBs are short period ($\approx$ hours) systems
1067: %in which the accreting star is a compact object such as a neutron star (NS) or a 
1068: %black hole (BH)  and the donor is 
1069: %a hydrogen rich, low mass Main-sequence (MS)/evolved star. On the other
1070: %hand, the donors in UCXBs can not have MS stars as their companions due to
1071: %their ultra short orbital periods and must be hydrogen poor \citep{Nelson1986}.
1072: %Some of them have millisecond pulsars \citep{Mark2002} and low mass white 
1073: %dwarfs  are considered as donors in these
1074: %systems \citep{Galloway2002, WC2004}. A population study of field UCXBs
1075: %in the context of x-ray sources \citep{BT2004a} and the contribution of Galactic
1076: %and extra-galactic 
1077: %systems to LISA's unresolved background noise
1078: %\citep{Kim2004,cooray2004} have been studied before. In this section, we mainly
1079: %concentrate on the resolvable sources of NSWD and DWD sytems for LISA and try to
1080: %ascertain the limits of applicability of our theoretical 
1081: %population boundaries, as motivated by observations of various compact binary systems.
1082: 
1083: 
1084: %In Fig.\ref{fig2}, we show the observed UCXB's (blue stars), detached DWDs (orange cross)
1085: %and AM CVn systems (blue plus) plotted
1086: %on top of the NSWD \& DWD population boundary diagrams. The distance independent amplitude,
1087: %$rh_\mathrm{norm}$, for these observed systems is calculated using 
1088: %Eq.(\ref{h_f_relationship}).
1089: % The donor masses  and GW frequency ($f=2/P_\mathrm{orb}$, where $P_\mathrm{orb}$ is the
1090: %orbital period) used to calculate $rh_\mathrm{norm}$ for all the observed
1091: % UCXBs are given in
1092: % table \ref{table1}, where we use the minimum mass value.
1093: % In the case of UCXBs, we assume the companion NS mass of $1.44 M_\mathrm{\odot}$. 
1094: %%Looking at Fig.\ref{fig2}, the following issues are apparent:
1095: % Since we assume that all mass-transferring NSWD systems are UCXBs and the donors in these 
1096: %systems follow zero temperature M-R relation, we would expect
1097: %these systems to be constrained to {\bf Region V}. Clearly, some of the UCXB systems 
1098: %do not reside in this region because  the donor masses for those systems
1099: %were derived using either a ``warm'' equation of state (EOS)
1100: % \citep{DB2003} or more simplified M-R relation by \cite{Paczynski67}.
1101: %Since our population boundaries are based on  
1102: %Eggleton's M-R relationship for zero-temperature white dwarfs (Eq.\ref{mass_radius}),
1103: %any system's donor mass  found utilizing this relation should be found within
1104: % {\bf Region V}.
1105: %
1106: %%The limitation to our CM diagram comes from assuming that the donors in UCXBs and AM CVns 
1107: %%are zero temperature objects and it is no surprise that some of the observed systems do not
1108: %%follow this assumption.
1109: %As pointed by \cite{Mark2002}, it is likely that some of the donors in UCXBs have ``warm'' 
1110: %low mass WD donors 
1111: %to conform with observations. \cite{DB2003} calculated finite entropy M-R relation 
1112: %models for low-mass donors in UCXBs with different compositions. While these more realistic
1113: %models indicate the limits of applicability of our population boundaries, assuming 
1114: %zero temperature M-R relationship for donor WDs would allow us to depict, in a concise 
1115: %and clear way,
1116: %the various evolutionary phases of compact binary systems (DWD/NSWDs) independent of the
1117: % distance.\footnote{We note that it is possible to derive these population
1118: % boundaries using even simple M-R relation of \cite{Paczynski67}. But using a zero
1119: %temperature M-R relation here is a choice to balance the issue of using an accurate model
1120: % that would allow us illustrate the evolutionary phases of compact binary systems, 
1121: %without making the accompanying discussion too complicated.}
1122: %%and we will discuss in a later section the affect of these
1123: %%finite entropy models on our CM diagram.
1124: %%how our CM diagram
1125: %%will be modified when we apply this finite entropy models.
1126: % The open circles in the left panel of Fig. \ref{fig2} represent the systems for which orbital
1127: % periods are measured but the masses are yet unknown. Since we assume that
1128: %UCXBs are mass transferring NSWD systems and the donors are zero temperature WDs, 
1129: % they should be constrained to Region {\bf V}.
1130: %Accordingly arbitrary values of log$(rh_\mathrm{norm})$, that would place these
1131: %systems within this region, were assigned without changing their known orbital
1132: %periods.
1133: %
1134: % It appears that many of the known UCXBs statistically 
1135: %cluster around orbital periods of $40 \sim 50$ minutes 
1136: %($Log[f] = -3.08 \sim -3.17$, vertical dashed lines in the left panel of Fig.\ref{fig2}).
1137: %  In general,
1138: %evolutionary arguments predict more systems at longer orbital periods \citep{DB2003, BT2004a,
1139: %Nelemans2006, KT2007} 
1140: %because the evolutionary 
1141: %time-scale, $\tau_\mathrm{chirp}$,  reaches Hubble time ($\sim 10^{10}$ years), as shown by
1142: % the constant 
1143: %$\tau_\mathrm{chirp}$ lines in Fig. \ref{fig2}. Furthermore, there are more number of systems
1144: %at around $\sim 42$ minutes than at other orbital periods. It is quite possible that since the 
1145: %donor mass and the mass-transfer rate decreases towards longer orbital periods \citep{DB2003},
1146: % the systems 
1147: %are fainter and are hard to detect\footnote{Although, according to \cite{DB2003}, systems with orbital
1148: %periods greater than $30$ minutes have transient behavior because of disk instabilities \citep{Dubus1999},
1149: % making them easier to detect, it appears that many of the long period
1150: %systems are actually persistent in nature \citep{Nelemans2006,intzand2007}}. There are three systems
1151: %at around orbital periods of $\sim 50$ minutes and it may happen that more systems may be found around
1152: %this period. 
1153: %Recently, six additional candidate UCXB systems were found \citep{intzand2007} for which orbital
1154: %periods are yet to be determined. 
1155: %
1156: %Unlike the observed UCXBs, it appears that AM CVn systems exhibit no preferential clustering
1157: % though most of them lie below $30$ minute orbital period ($Log[f] = -2.95$). The mass values are
1158: %taken from \cite{Deloyeetal2005} Table 1 and \cite{SV2006}. The detached DWD systems
1159: % \citep{Nelemans2005} are also 
1160: %plotted. Notice that, as mentioned in \S\ref{compare_NSWD_DWD}, if LISA observations are the only
1161: %information we have, i.e a measurement of $h$ and $f$ then, there is no way to tell 
1162: %whether these detached systems are inspiralling NSWDs or DWDs, because of the overlap of regions 
1163: %between these two systems.
1164: 
1165: \section{Summary}
1166: \label{sec4}
1167: In the previous sections, we have outlined the construction of population boundaries
1168:  to illustrate the various evolutionary phases
1169: that a DWD or a NSWD binary system would undergo in the distance
1170:  independent ``absolute'' amplitude-frequency domain (log$[rh_\mathrm{norm}]$ - log$[f]$) of LISA.
1171: %  and attempted to draw parallels
1172: %between the CM diagram that is generally used in astronomy.
1173: In an update to \cite{KT2007}, who assumed the donors to be fully degenerate $T=0$ He stars,  we 
1174: consider that the donors in these systems  follow the M-R relationship of 
1175: non-zero entropy donor models of \cite{DB2003} assuming He composition.
1176:  These models assume fully convective and adiabatically
1177: evolving donors during the whole episode of the mass-transfer phase. Figs. \ref{fig1} $\&$ \ref{fig2}
1178: show that these
1179: ``hot'' donors occupy a larger fraction of the $rh - f$ space, than the $T=0$ donors, because hot 
1180: donors are also more massive, increasing their intrinsic GW amplitude. At high frequencies, both the
1181: models match each other because at these high masses, the hot donors are supported by degeneracy pressure
1182: and the M-R relations match.
1183: At low frequencies, the $T=0$ model and \cite{DB2003} $T=10^{4} K$ full model (which is equivalent to
1184: $T=0$) diverge slightly because the Coulomb interactions dominate thermal contribution and M-R relations
1185: deviate. 
1186: %Similar boundaries for C $\&$ O would lie just below He boundaries, except they also match with
1187: %$T=0$ at high frequencies.
1188: 
1189:  We also discussed the implications
1190: of new evolutionary phases found by \cite{Deloyeetal2007} on our population boundaries. The initial
1191: ``turn on'' phase, where $\dot M_\mathrm{d}$ reaches it's maximum value and
1192:  $P_\mathrm{orb}$ decreases even after contact, will result in
1193:  contact boundaries having a width, instead of an abrupt cut-off.
1194:  This will cause some overlap
1195:  onto the C $\&$ O isotherms, which lie below He boundary. Soon after the system evolves from this 
1196: initial contact, the second phase starts where the donor undergoes adiabatic evolution in response to the
1197: mass-transfer and the system will follow a typical AM CVn evolutionary trajectory with decreasing GW 
1198: amplitude and frequency. This will continue
1199: until the donor begins to cool and the radius starts contracting. Accordingly, it  will reduce
1200:  $\dot M_\mathrm{d}$,  and $P_\mathrm{orb}$ (and GW frequency) evolution slows down and may stall
1201:  once the donor reaches fully degenerate configuration. But since LISA's
1202:  sensitivity in this range is
1203: limited by instrumental and DWD background noise, this part of the evolution (or cooling donors) will
1204: probably be not observable by LISA.
1205: 
1206: It is unlikely that determination of $\dot f$ will shed light on the type
1207:  (DWD/NSWD) of the low frequency systems 
1208: without the help of independent electromagnetic observations.
1209: This is because there is fairly a good overlap of NSWD and DWD systems within the resolvable frequency
1210: regime (Log$[f] > 3$) of LISA \footnote{Since this region is occupied by both
1211:  inspiralling $\&$ and mass-transferring systems, it may happen that more than one source can reside in
1212: a frequency bin, but a measurement of $\dot f$ would probably indicate the evolutionary phase.}.  But for high frequency sources, it might be possible to 
1213: distinguish between them.
1214: Combined with the expectation
1215:  that the relative population of NSWD is low compared to DWDs and short period systems
1216: do not stay longer at those periods, it is likely that
1217: LISA will be able to measure $\dot f$ for more number of DWDs than NSWDs. 
1218:  It will be
1219: interesting to see how the high frequency regions constrained by these
1220: boundaries would be populated through LISA observations and whether indeed
1221: we will know the nature of these systems.
1222: 
1223: % compared these
1224: % with DWD population boundaries, discussed in \cite{KT2007} assuming that the donor WDs 
1225: %follow zero temperature M-R relationship.  We noted that if LISA measures
1226: %only $h_\mathrm{norm}$ and $f$, then we would not be able to differentiate between a NSWD
1227: %or DWD, if no independent observations (optical/x-ray) are available. On the other hand if
1228: %an $\dot f$ is measured and if it turns out to be negative, then  the value of $\dot f$ would
1229: % indicate if it is a mass transferring NSWD/DWD system.  We also showed that if there is an
1230: %independent evidence that a particular NSWD/DWD system is undergoing CMT phase,
1231: % then it is possible to set limits on the distance to the system. Furthermore, a sharper 
1232: %estimate on the distance can be obtained for mass transferring NSWD systems than for DWDs
1233: %in CMT phase because of the regions they occupy in the log$(rh_\mathrm{norm})$ - log$(f)$
1234: %space (regions {\bf II} and {\bf V} in Fig.\ref{fig2}).
1235: 
1236: %Comparing the population boundaries with observed UCXBs, we found that some of them do not
1237: %reside in the restricted regions for mass-transferring NSWD systems because for those
1238: %systems the masses were derived using a ``warm''  M-R relation, in contrast to the 
1239: %zero temperature M-R relation used here. There are currently 12 UCXBs with known 
1240: %orbital periods \citep{intzand2007} and some of them are confirmed to be in globular clusters. 
1241: %In
1242: %Fig.\ref{fig3}, we plot the known UCXBs and AM CVns on top of LISA's sensitivity curve\footnote{http://www.srl.caltech.edu/~shane/sensitivity/MakeCurve.html}
1243: % (SNR = 1) to assess the detectability of these systems. In the case of known UCXBs, 
1244: %it is clear that only one system 
1245: %(4U 1820-30) has enough signal strength to be visible to LISA, whereas some of the known
1246: %AM CVns emit GWs above the instrumental and DWD background noise. Population synthesis
1247: % models indicate that the total number of UCXBs in the field is expected to be
1248: % $\geq 10$ \citep{BT2004a,cooray2004} but not too high and the extragalactic NSWD
1249: % background noise is not
1250: %expected to contribute much to the DWD noise \citep{Kim2004}. Combined with the expectation
1251: % that the relative population of NSWD is low compared to DWDs and short period systems
1252: %donot stay longer at those periods,
1253: % it is likely that 
1254: %LISA will be able to measure $\dot f$ for more number of DWDs than NSWDs. But on the 
1255: %brightside, although LISA would not be able to measure $\dot f$, any information on the
1256: %location of the source given by LISA in combination with follow-up observations in
1257: % optical/x-ray, to find
1258: %these systems would then be useful, as the population boundaries described here can then
1259: %be used to estimate the binary system parameters such as the distance and masses of
1260: % individual components of the binary.
1261: 
1262: 
1263: $\\$
1264: 
1265: \acknowledgements
1266: 
1267: We thank an anonymous referee whose suggestions have led to significant
1268:  improvements in the
1269: manuscript. R. K gratefully acknowledges the support of National Science
1270:  Foundation Grant
1271: No.~PHY 06-53462 and No.~PHY 05-55615, and NASA Grant No.~NNG05GF71G,
1272: awarded to The Pennsylvania State University.
1273: Many thanks to Joel Tohline (LSU) for the help and guidance provided in the 
1274: preparation of this work and manuscript.
1275: % We would also like to thank Deepto Chakrabarty for 
1276: %providing us with a list of known UCXBs.
1277: %Some of the mass-transferring NSWD/DWD systems are observed in optical and/or in x-ray
1278: %band and they are given in Tables (2) \& (3).  
1279: 
1280: \begin{deluxetable}{cccccc}
1281: \tablecaption{Observed and derived properties of some of the known UCXBs \citep{intzand2007} shown
1282:  in Fig.\ref{fig2}. $M_\mathrm{d}$ indicates minimum mass of the donor. 
1283: References: (1) \cite{Tarana2007} (1A) \cite{Cumming2003} (2) \cite{WC2004} 
1284: (2A) \cite{WC2004} (3) \cite{Sidoli2006} (3A) \cite{JC2005} (4) \cite{Dieball2005} (4A) \cite{McNamara2004} (5) \cite{Falanga2005} (5A) \cite{Campana2003} (6) \cite{Krauss2007} (6A) \cite{C1998} (7)\cite{Gierlinski2005} (7A) \cite{Papitto2008} (8) \cite{Galloway2002} (8A) \cite{Galloway2002} (10A)  \cite{C1998} 
1285: (11A) \cite{Nelemans2004} (12) \cite{Krimm2007} (12A) http://web.mit.edu/newsoffice/2007/pulsar-0913.html}
1286: \tablewidth{0pt}
1287: %
1288: \tablehead{ \colhead{Name} & \colhead{Orbital period} & \colhead{$\frac{M_\mathrm{d}}{M_\odot}$}& \colhead{Reference} & \colhead{distance} &\colhead{Reference} \\
1289: & \colhead{(min)} & & \colhead{for mass} & \colhead{(kpc)} & \colhead{for distance}}
1290: \startdata 4U 1820-30 & 11 & 0.06 & 1 & 7.6 & 1A\\
1291: 4U 1543-624 & 18 & 0.025 & 2 & 7.0 & 2A\\
1292: 4U 1850-087 & 21 & 0.04 & 3 & 8.2 & 3A\\
1293: M15 X-2 & 22.6 & 0.02 & 4 & 9.98 & 4A\\
1294: XTE J1807-294 & 41 & 0.0053 & 5 & 8.0 & 5A\\
1295: 4U 1626-67 & 42 & 0.04 & 6 & 5 & 6A\\
1296: XTE J1751-305 & 42.4 & 0.014 & 7 & 6.7 & 7A\\
1297: XTE J0929-314 & 43.6 & 0.008 & 8 & 5.0 & 8A\\
1298: NGC 6652B & 43.6 & ? & -- & ? & --\\
1299: 4U 1916-05 & 50 & ? & -- & 8.9 & 10A\\
1300: 4U 0614+091 & 50 & ? & -- & 3.0 & 11A\\
1301: SWIFT J1756.9-2508 & 54.7 & 0.0067 & 12 & 7.6 & 12A\\
1302:  \\
1303: \enddata
1304: \tablecaption{this is}
1305: \label{table1}
1306: \end{deluxetable}
1307: 
1308: 
1309: %\clearpage
1310: %\thispagestyle{empty}
1311: %\begin{figure}[!hbp|t]
1312: %%\vspace*{-25mm}
1313: %%\centering
1314: %%\epsscale{0.4}
1315: %%\plotone{ps_files/only_NSWD.eps}
1316: %\includegraphics[height=9cm,width=9.0909cm,angle=360]{../ps_files/only_NSWD_observed.eps}
1317: %\includegraphics[height=9cm,width=9.0909cm,angle=360]{../ps_files/only_DWD_observed.eps}
1318: %\hfill \caption{ Evolutionary trajectories of both NSWDs and DWDs: 
1319: %{\it left panel:} The red line represents the maximum inspiral trajectory from a 
1320: %detached NSWD binary system. Region {\bf V} represents the space where mass-transferring
1321: % NSWDs can exist. The blue stars are the known UCXBs (see Table \ref{table1} and
1322: %\S\ref{UCXB_section}) which 
1323: %we assume here to be mass-transferring NSWDs. The vertical dashed lines represent the clustering
1324: %of UCXBs between orbital periods of $40 \sim 50$ minutes. 
1325: %{\it Right panel:} Population boundaries for DWDs as shown in \cite{KT2007}.
1326: % Mass-transferring DWD systems are constrained to Region II. The 
1327: %observed detached DWDs (cross) and AM CVn systems (plus) are also shown.}
1328: %% The vertical position (log$(rh_\mathrm{norm})$) is
1329: %% somewhat arbitrary for these observed systems because the masses are unknown. Assuming LMXBs are 
1330: %%mass-transferring NSWD systems,  arbitrary values that
1331: %% would place these systems within Region {\bf V} were assigned without changing their known
1332: %% orbital periods.}
1333: %\label{fig2}
1334: %\end{figure}
1335: %\clearpage
1336: 
1337: \clearpage
1338: \thispagestyle{empty}
1339: \begin{figure}[!hbp|t]
1340: %%\vspace*{-25mm}
1341: \centering
1342: %\epsscale{0.4}
1343: %\plotone{f3.eps}
1344: \includegraphics[height=18cm,width=13.0909cm,angle=270]{f3.eps}
1345: \hfill \caption{ The known UCXBs (blue stars) and AM CVn systems (light blue
1346:  stars) 
1347: plotted on top of LISA's sensitivity curve for SNR = 1. In case of UCXBs, only one system
1348: (4U 1820-30) has enough signal strength to be visible to LISA, whereas couple of known
1349: AM CVns emit GWs above the instrumental and DWD background noise (green curve). }
1350: \label{fig3}
1351: \end{figure}
1352: \clearpage
1353: 
1354: 
1355: \begin{thebibliography}{}
1356: %\bibitem[Finn \& Chernoff(1993)]{FC1993}
1357: %Finn, L. S., \& Chernoff, D. F. 1993, Phys. Rev. D, 47, 2198
1358: 
1359: %\bibitem[Benacquista \& Holley-Bockelmann(2006)]{BK2006}
1360: %Benacquista, M., \& Holley-Bockelmann, K. 2006, ApJ, 645, 589
1361: \bibitem[Belczynski \& Taam(2004a)]{BT2004a}
1362: Belczynski, K., \& Taam, R. E. 2004a, ApJ, 603, 690
1363: 
1364: \bibitem[Belczynski \& Taam(2004b)]{BT2004b}
1365: ------------------. 2004b, ApJ, 616, 1159
1366: 
1367: \bibitem[Bender(1998)]{Bender98}
1368: Bender, P. L. 1998, \baas, 193, 48.03
1369: 
1370: \bibitem[Bildsten et al.(2006)]{Bildsten2006}
1371: Bildsten, L. et al. 2006, ApJ, 640, 466
1372: 
1373: \bibitem[Campana et al.(2003)]{Campana2003}
1374: Campana, S. et al. 2003, ApJ, 594, L39
1375: 
1376: \bibitem[Chakrabarty(1998)]{C1998}
1377: Chakrabarty, D. 1998, ApJ, 492, 342
1378: 
1379: \bibitem[Clark(1975)]{clark1975}
1380: Clark, G. W. 1975, ApJ, 199, L143
1381: 
1382: \bibitem[Cooray(2004)]{cooray2004}
1383: Cooray, A. 2004, MNRAS, 354, 25
1384: 
1385: \bibitem[Cropper et al.(1998)]{Cropper1998}
1386: Cropper, M. et al. 1998, MNRAS, 293, L57
1387: 
1388: \bibitem[Cumming(2003)]{Cumming2003}
1389: Cumming, A. 2003, ApJ, 595, 1077
1390: 
1391: \bibitem[Deloye \& Bildsten(2003)]{DB2003}
1392: Deloye, C. J. \& Bildsten, L. 2003, ApJ, 598, 1217
1393: 
1394: \bibitem[Deloye et al.(2005)]{Deloyeetal2005}
1395: Deloye, C. J. et al. (2005), ApJ, 624, 934
1396: 
1397: \bibitem[Deloye et al.(2007)]{Deloyeetal2007}
1398: Deloye, C. J. et al. (2007), MNRAS, 381, 525
1399: 
1400: \bibitem[Dieball et al.(2005)]{Dieball2005}
1401: Dieball, A. et al. 2005, ApJ, 634, 105
1402: 
1403: %\bibitem[Dubus et al.(1999)]{Dubus1999}
1404: %Dubus, G. et al. 1999, MNRAS, 303, 139
1405: 
1406: \bibitem[Eggleton(1983)]{EGG}
1407: Eggleton, P. P. 1983, ApJ, 268, 368
1408: 
1409: \bibitem[Evans, Iben, \& Smarr(1987)]{EIS87}
1410: Evans, C. R., Iben, I., Jr., \& Smarr, L. 1987, \apj, 323, 129
1411: 
1412: \bibitem[Falanga et al.(2005)]{Falanga2005}
1413: Falanga, M. et al. 2005, A\&A, 436, 647
1414: 
1415: \bibitem[Faller \& Bender(1984)]{FB84}
1416: Faller, J. E., \& Bender, P. L. 1984, in {\it Precision Measurement
1417: and Fundamental Constants II}, ed. B. N. Taylor \& W. D. Phillips
1418: (NBS Spec. Pub. 617)
1419: 
1420: \bibitem[Galloway et al.(2002)]{Galloway2002}
1421: Galloway, D. K. et al. 2002, ApJ, 576, L137
1422: 
1423: \bibitem[Gierlinski(2005)]{Gierlinski2005}
1424:  Gierlinski, M. \&  Poutanen, J. (2005), MNRAS, 359, 1261
1425: 
1426: \bibitem[Iben \& Tutukov(1984)]{IT84}
1427: Iben, I., Jr., \& Tutukov, A. V. 1984, ApJS, 54, 335
1428: 
1429: \bibitem[Iben \& Tutukov(1986)]{IT86}
1430: Iben, I., Jr., \& Tutukov, A. V. 1986, ApJ, 311, 753
1431: 
1432: \bibitem[in 't Zand et al.(2007)]{intzand2007}
1433: in 't Zand, J. J. M., et al. 2007, A\&A, 465, 953
1434: 
1435: \bibitem[Ivanova et al.(2007)]{ivanova2007}
1436: Ivanova, N., et al. 2007, MNRAS, 386, 553
1437: 
1438: \bibitem[Juett \& Chakrabarty(2005)]{JC2005}
1439: Juett, A, M. \& Chakrabarty, D. (2005), ApJ, 627, 926
1440: 
1441: 
1442: \bibitem[Kim et al.(2004)]{Kim2004}
1443: Kim, C. et al. 2004, ApJ, 616, 1109
1444: 
1445: \bibitem[Kopparapu \& Tohline(2007)]{KT2007}
1446: Kopparapu, R. K., \& Tohline, J. E. 2007, ApJ, 655, 1025
1447: 
1448: \bibitem[Krauss et al.(2007)]{Krauss2007}
1449: Krauss, M, I. et al. 2007, ApJ, 660, 605
1450: 
1451: \bibitem[Krimm et al.(2007)]{Krimm2007}
1452: Krimm, H. A. et al. (2007), ApJ, 668, L147
1453: 
1454: \bibitem[Markwardt et al.(2002)]{Mark2002}
1455: Markwardt, C. B et al. 2002, ApJ, 575, L21
1456: 
1457: \bibitem[Marsh \& Steeghs(2002)]{MS2002}
1458: Marsh, T. R., \& Steeghs, D. 2002, MNRAS, 331, L7
1459: 
1460: \bibitem[Marsh et al.(2004)]{MNS}
1461: Marsh T. R., Nelemans, G., \& Steeghs, D. 2004, MNRAS, 350, 113
1462: 
1463: %\bibitem[Nelemans et al.(2001)]{Nelemans2001}
1464: %Nelemans, G., Yungelson, L. R., Portegies Zwart, S. F. 2001, A \& A, 375, 890
1465: \bibitem[McNamara et al.(2004)]{McNamara2004}
1466: McNamara, B. J. et al. 2004, ApJ, 602, 264
1467: 
1468: \bibitem[Nelemans et al.(2001)]{Nelemans2001}
1469: Nelemans, G. et al. 2001, A\&A, 375, 890
1470: 
1471: \bibitem[Nelemans et al.(2004)]{Nelemans2004}
1472: Nelemans, G. et al. 2004, MNRAS, 348, L7
1473: 
1474: \bibitem[Nelemans(2005)]{Nelemans2005}
1475: Nelemans, G. 2005, ASPC, 330, 27
1476: 
1477: %\bibitem[Nelemans et al.(2005)]{Nelemansetal2005}
1478: %Nelemans, G. et al. 2005, A\&A, 440,1087
1479: 
1480: \bibitem[Nelemans et al.(2006)]{Nelemans2006}
1481: Nelemans et al. 2006, MNRAS, 370, 255
1482: 
1483: \bibitem[Nelson et al.(1986)]{Nelson1986}
1484: Nelson, L. A. et al. 1986, ApJ, 304, 231
1485: 
1486: \bibitem[Paczy\'nski(1967)]{Paczynski67}
1487: Paczy\'nski, B. 1967, Acta. Astr., 17, 287
1488: 
1489: \bibitem[Papitto(2008)]{Papitto2008}
1490: Papitto, A. et al. 2008, MNRAS, 383, 411
1491: 
1492: %\bibitem[Peters \& Mathews(1963)]{PM63}
1493: %Peters, P.C., \& Mathews, J. 1963, Phys. Rev., 131, 435
1494: 
1495: %\bibitem[Ritter \& Kolb(2003)]{RKcat}
1496: %Ritter, H., \& Kolb, U. 2003, A\&A, 404, 301
1497: 
1498: \bibitem[Roelofs et al.(2007)]{Roelofs2007}
1499: Roelofs, G. H. A. et al. 2007, ApJ, 666, 1174
1500: 
1501: %\bibitem[Schutz(1986)]{Schutz86}
1502: %Schutz, B. 1986, Nature, 323, 310
1503: 
1504: \bibitem[Sidoli et al.(2006)]{Sidoli2006}
1505: Sidoli, L. et al. 2006, A\&A, 460, 229 
1506: 
1507: %\bibitem[Stroeer \& Vecchio(2006)]{SV2006}
1508: %Stroeer, A. \& Vecchio, A. 2006, Classical Quantum Gravity, 23, 809
1509: 
1510: \bibitem[Tarana et al.(2007)]{Tarana2007}
1511: Tarana, A. et al. 2007,  ApJ, 654, 494
1512: 
1513: %\bibitem[Van den Heuvel(1983)]{VandenHeuvel1983}
1514: %Van den Heuvel, E. P. J. 1983, Accretion-driven stellar X-ray sources, Cambridge 
1515: %University Press, p303.
1516: 
1517: \bibitem[Verbunt \& Rappaport(1988)]{VR88}
1518: Verbunt, F., \& Rappaport, S. 1988, \apj, 332, 193
1519: 
1520: \bibitem[Wang \& Chakrabarty(2004)]{WC2004}
1521: Wang, Z \& Chakrabarty, D. 2004, ApJ, 616, L139
1522: 
1523: %\bibitem[Warner(1995)]{Warner1995}
1524: %Warner, B. 1995, Cataclysmic Variable stars
1525: 
1526: \bibitem[Wu et al.(2002)]{Wu2002}
1527: Wu, K. et al. 2002, MNRAS, 331, 221
1528: 
1529: \end{thebibliography}
1530: \end{document}
1531: