Formation and evolution of compact binaries in globular clusters: II. Binaries with neutron stars
N. Ivanova, C. Heinke, F.A. Rasio, K. Belczynski, J. Fregeau

TL;DR
This study models the formation and evolution of neutron star binaries in dense globular clusters, explaining observed low-mass X-ray binaries and millisecond pulsars, and predicting merger rates relevant for gravitational wave detection.
Contribution
It introduces a comprehensive simulation of neutron star binary formation in globular clusters, incorporating electron-capture supernovae and detailed pulsar recycling processes.
Findings
Reproduces observed LMXB formation rates with ECS supernovae assumptions.
Predicts weak metallicity dependence of quiescent LMXB numbers.
Identifies conditions for producing merging double neutron stars in dense clusters.
Abstract
In this paper, the second of a series, we study the stellar dynamical and evolutionary processes leading to the formation of compact binaries containing neutron stars (NSs) in dense globular clusters (GCs). For this study, 70 dense clusters were simulated independently, with a total stellar mass ~2x10^7Msun, exceeding the total mass of all dense GCs in our Galaxy. We find that, in order to reproduce the empirically derived formation rate of low-mass X-ray binaries (LMXBs), we must assume that NSs can be formed via electron-capture supernovae (ECS) with typical natal kicks smaller than in core-collapse supernovae. Our results explain the observed dependence of the number of LMXBs on ``collision number'' as well as the large scatter observed between different GCs. We predict that the number of quiescent LMXBs in different GCs should not have a strong metallicity dependence. In our…
Click any figure to enlarge with its caption.
Figure 1
Figure 2
Figure 3
Figure 4
Figure 5
Figure 6
Figure 7
Figure 8
Figure 9
Figure 10
Figure 11
Figure 12| MB | Runs (number of stars) | |||||||||
|---|---|---|---|---|---|---|---|---|---|---|
| standard | 1.0 | 0.005 | 10 | 40 | 100% | 1.0 | IT03 | 5 with , 5 with | ||
| metal-poor | 0.0005 | 5 with | ||||||||
| high-dens | 5 with | |||||||||
| med-dens | 5 with | |||||||||
| low-dens | 5 with | |||||||||
| low- | 5 | 20 | 5 with | |||||||
| long- | 3.0 | 5 with | ||||||||
| BF05 | 50% | 5 with | ||||||||
| fast-MB | RVJ | 5 with | ||||||||
| CE-reduced | 0.1 | 5 with | ||||||||
| oldkicks | 5 with | |||||||||
| 47 Tuc | 3.0 | 0.0035 | 11.5 | 57 | 5 with | |||||
| Terzan 5 | 0.93 | 0.013 | 11.6 | 49 | 5 with |
| CC | EIC | AIC | MIC | ||||
|---|---|---|---|---|---|---|---|
| single | |||||||
| Z=0.0005 | 3354 | 594 | - | - | 2.1 | 251 | 584848 |
| Z=0.001 | 3255 | 581 | - | - | 1.9 | 245 | 565168 |
| Z=0.005 | 2833 | 570 | - | - | 1.2 | 243 | 498513 |
| Z=0.02 | 2666 | 400 | - | - | 1.2 | 172 | 424696 |
| oldkicks | 3104 | - | - | - | 19.3 | 21.9 | 339669 |
| binary | |||||||
| Z=0.0005 | 3079 | 545 | 59 | 14 | 19 | 329 | 249775 |
| Z=0.001 | 3056 | 553 | 60 | 15 | 15 | 332 | 242601 |
| Z=0.005 | 2750 | 576 | 58 | 16 | 14 | 335 | 219236 |
| Z=0.02 | 2463 | 406 | 33 | 20 | 12 | 240 | 180582 |
| oldkicks | 2929 | - | - | - | 58 | 58 | 237705 |
| model | Core | Halo | ||||||
|---|---|---|---|---|---|---|---|---|
| CC | EIC | AIC | MIC | CC | EIC | AIC | MIC | |
| standard | ||||||||
| metal-poor | ||||||||
| high-den | ||||||||
| med-den | ||||||||
| low-den | ||||||||
| lowsig | ||||||||
| longtrh | ||||||||
| BF05 | ||||||||
| fast-MB | ||||||||
| CE-reduced | ||||||||
| oldkicks | ||||||||
| Tuc-47 | ||||||||
| Terzan-5 | ||||||||
| Z | after NS formation | at 11 Gyr | ||||
|---|---|---|---|---|---|---|
| CC | EIC | AIC | CC | EIC | AIC | |
| 0.0005 | 2.9% | 3.0% | 92.3% | 0.28% | 1.38% | 78.6% |
| 0.001 | 2.8% | 3.4% | 93.6% | 0.25% | 1.41% | 81.8% |
| 0.005 | 2.6% | 3.4% | 96.0% | 0.18% | 1.54% | 88.0% |
| 0.02 | 1.4% | 4.2% | 95.2% | 0.12% | 1.49% | 85.9% |
| Z | DCE | TC | ||||
|---|---|---|---|---|---|---|
| km/s | Tot | MT | Tot | MT | ||
| standard | 0.005 | 10 | 3.68% | 1.75% | 2.35% | 1.20% |
| metal-poor | 0.0005 | 10 | 3.25% | 1.30% | 1.93% | 0.95% |
| low- | 0.05 | 5 | 6.80% | 3.61% | 9.24% | 4.37% |
| Cluster | All pulsars | Single | BD | He WD | MS | heavy WD | d | ecc | high-B |
|---|---|---|---|---|---|---|---|---|---|
| High-den model | |||||||||
| All high-den | 19 | 0.71 | 0.06 | 0.0 | 0.12 | 0.12 | 0.06 | 0.18 | 0.12 |
| Terzan 5 model | |||||||||
| All Ter5 | 49 | 0.43 | 0.08 | 0.31 | 0.12 | 0.06 | 0.08 | 0.16 | 0.04 |
| 47Tuc model | |||||||||
| All 47Tuc | 37 | 0.43 | 0.19 | 0.30 | 0.11 | 0 | 0.03 | 0.08 | 0.03 |
| Standard model | |||||||||
| all stand | 8 | 0.25 | 0.25 | 0.50 | 0.0 | 0 | 0.25 | 0.25 | 0.0 |
| Med-den model | |||||||||
| All med-den | 11 | 0.18 | 0.18 | 0.64 | 0 | 0 | 0.18 | 0.18 | 0 |
| Field | 77 | 0.25 | 0.04 | 0.53 | 0 | 0.13 | 0.30 | 0.01 | - |
Peer Reviews
No public reviews on file for this paper yet. If you reviewed it on a platform where reviews are public (OpenReview, ICLR, NeurIPS, ICML), you can paste yours below so the community can read it here.
Videos
No videos yet. Explain this paper in a talk, walkthrough, or lecture? Add one.
Formation and evolution of compact binaries in globular clusters:
II. Binaries with neutron stars.
N. Ivanova 1, C. 0. Heinke2,3, F. A. Rasio2, K. Belczynski4,5, & J. M. Fregeau2
1Canadian Institute for Theoretical Astrophysics, University of Toronto, 60 St. George, Toronto, ON M5S 3H8, Canada
2Northwestern University, Dept. of Physics & Astronomy, 2145 Sheridan Rd, Evanston, IL 60208, USA
3University of Virginia, Dept. of Astronomy, PO Box 400325, Charlottesville, VA 22904, USA
4Los Alamos National Lab, P.O. Box 1663, MS 466, Los Alamos, NM 87545
5New Mexico State University, Department of Astronomy, 1320 Frenger Mall, Las Cruces, New Mexico 88003-8001, USA E-mail:[email protected], Tremaine FellowLindheimer FellowOppenheimer FellowChandra Fellow
Abstract
In this paper, the second of a series, we study the stellar dynamical and evolutionary processes leading to the formation of compact binaries containing neutron stars (NSs) in dense globular clusters. For this study, 70 dense clusters were simulated independently, with a total stellar mass , exceeding the total mass of all dense globular clusters in our Galaxy.
We find that, in order to reproduce the empirically derived formation rate of low-mass X-ray binaries (LMXBs), we must assume that NSs can be formed via electron-capture supernovae with typical natal kicks smaller than in core-collapse supernovae. Our results explain the observed dependence of the number of LMXBs on “collision number” as well as the large scatter observed between different globular clusters. We predict that the number of quiescent LMXBs in different clusters should not have a strong metallicity dependence.
We compare the results obtained from our simulations with the observed population of millisecond pulsars (MSPs). We find that in our cluster model the following mass-gaining events create populations of MSPs that do not match the observations (either they are inconsistent with the observed LMXB production rates, or the inferred binary periods or companion masses are not observed among radio bMSPs): (i) accretion during a common envelope event with a NS formed through electron capture supernovae, and (ii) mass transfer (MT) from a WD donor. Some processes lead only to a mild recycling – physical collisions or MT in a post-accretion induced collapse system. In addition, for MSPs, we distinguish low-magnetic-field (long-lived) and high-magnetic-field (short-lived) populations, where in the latter NSs are formed as a result of accretion induced collapse or merger induced collapse. With this distinction and by considering only those mass-gaining events that appear to lead to NS recycling, we obtain good agreement of our models with the numbers and characteristics of observed MSPs in 47 Tuc and Terzan 5, as well as with the cumulative statistics for MSPs detected in globular clusters of different dynamical properties.
We find that significant production of merging double NSs potentially detectable as short -ray bursts occurs only in very dense, most likely core-collapsed clusters.
keywords:
stellar dynamics – binaries: close – binaries: general – stars: neutron – pulsars: general – globular clusters: general – X-rays: binaries.
1 Introduction
Neutron stars (NSs) are seen in globular clusters (GCs) via their (current or past) participation in interacting binary systems. From the earliest observations of X-ray binaries in GCs it has been noted that their abundance per unit mass is 100 times greater in GCs than in the Galaxy as a whole (Katz, 1975). This was understood to be a consequence of the high stellar density of GCs, which may lead to the creation of compact NS binaries in close stellar encounters (Clark, 1975). NS binaries are often relatively easy to identify in GCs (compared to other binaries), and they allow us to constrain the dynamical history of clusters and the evolution of NS properties in binary systems. They have been observed so far in three guises: (i) as bright low-mass X-ray binaries (LMXBs); (ii) as quiescent LMXBs (qLMXBs); and (iii) as binary or single millisecond pulsars (MSPs). In the future, they could also be detected as sources of gravitational radiation.
Bright LMXBs have X-ray luminosities of typically – ergs s*-1*. In bright LMXBs, the NS is actively accreting from a companion star via Roche-lobe overflow and an accretion disk. Thirteen sources are known in Galactic GCs, of which 12 are known to contain NSs through the detection of thermonuclear bursts. Six are transient systems (two have remained in outburst for more than a decade; the others have outburst time-scales of weeks), while the other seven seem to be persistent. Orbital periods have now been identified for seven systems (Verbunt & Lewin, 2006; Dieball et al., 2005; Altamirano et al., 2007), of which three are below one hour, indicating ultracompact systems composed of NSs with white dwarf companions are common in globular clusters (Deutsch et al., 2000). Efficient surveys of the X-ray sky give us confidence that we have detected all bright LMXBs that have existed in Galactic GCs during the past 10 years.
Quiescent LMXBs have lower X-ray luminosities ( ergs/s), indicating that accretion is not active or is significantly reduced during quiescence. Their X-ray luminosity is thought to derive at least in part from the re-radiation of heat from the core of the NS, which was generated during accretion outbursts (Brown et al., 1998), though other mechanisms are also suggested (see, e.g., Campana et al. (1998) ). They can be identified through their positional coincidence with the location of a bright transient LMXB in outburst (Wijnands et al., 2005), and/or through their characteristically soft X-ray spectrum (Rutledge et al., 2002). 111Not all qLMXBs show the characteristic soft X-ray spectrum (Wijnands et al., 2005), though it seems that the majority do. The total number of qLMXBs in the Galactic GC system is inferred to be 100-200 (Heinke et al., 2003), although only a fraction of the Galactic clusters have been studied sufficiently to identify qLMXBs.
As of this writing 129 millisecond radio pulsars have been observed in GCs, with more than 10, 20 and 30 seen from the clusters M28 (Stairs et al., 2006), 47 Tuc (Camilo et al., 2000), and Terzan 5 (Ransom et al., 2005), respectively222See http://www.naic.edu/ pfreire/GCpsr.html for an updated list.. These millisecond pulsars (MSPs) are thought to be descendants of LMXBs (for reviews see Bhattacharya & van den Heuvel, 1991; Phinney & Kulkarni, 1994). It is estimated that of order 1000 MSPs are present in the Galactic GC system (Heinke et al., 2005b), of which half or more are potentially detectable. Current pulsar searches have reached the bottom of the pulsar luminosity function in very few clusters (e.g., 47 Tuc and M15). As an example, the total radio luminosity from Terzan 5 suggested that this cluster may contain 100 pulsars (Fruchter & Goss, 1995)—this number is 3 times more than the number of identified pulsars.
Since NSs in GCs appear to be excellent tests for our understanding of both stellar dynamics and neutron star evolution, it is not surprising that this issue has attracted significant continuing attention. The largest challenge for these studies is computational. It has been estimated that a cluster consisting of stars () will contain only a few hundred NSs, and models with a smaller number of stars will not therefore have sufficient statistics. Direct -body methods are unlikely to address the million-body problem before 2020 (Hut, 2006). Monte Carlo methods now allow million-body simulations for spherically symmetric systems with single stars (Giersz, 2006) and with large fractions of primordial binaries, including direct integrations of all binary encounters (Fregeau & Rasio, 2007), but a full implementation of binary star evolution is still lacking (Rasio et al., 2006). As a result, most previous theoretical studies of NSs in clusters were performed using analytic esimates (e.g., Hut et al., 1991), or no binaries (e.g., Drukier, 1996), or simple scattering experiments, with a static background of single or binary stars (e.g., Sigurdsson & Phinney, 1995; Rasio et al., 2000; Kuranov & Postnov, 2006). We chose instead to use a method that treats in detail the stellar evolution of a large population of stars and binaries in a dense cluster environment, but with a highly simplified description of the stellar dynamics which is not fully self-consistent.
In this paper we present the results of our ongoing study of compact binary formation in GCs, concentrating on the formation of binaries with NSs. For the first time we have used cluster simulations that include a dynamically evolving system and represent consistently an entire stellar population with numbers of stars and binaries comparable to those in observed GCs. In addition, several statistically independent simulations were performed for each cluster model, to show how different the outcomes can be with slightly different but equivalent initial conditions. Some preliminary results on NS binaries from our work were reported in Ivanova & Rasio (2004); Ivanova et al. (2005b); Ivanova & Rasio (2005c); Rasio et al. (2006), and detailed results for the formation of binaries containing white dwarfs (WDs) were presented in Paper I (Ivanova et al., 2006). We refer the reader to Paper I for a complete description of the physical processes of formation and destruction of mass-transferring compact binaries, as well as our numerical methods. We outline updates to the method and our assumptions that are especially relevant to NSs in Section 2. Section 3 outlines the scenarios for formation of MSPs in GCs. In Section 4 we describe the range of GC models that we consider. Formation and retention of NSs are analyzed in Section 5, and the following section is devoted to the formation of binaries containing NSs. We discuss the formation of LMXBs in Section 7, and MSPs are considered in Section 8. Section 9 briefly describes double neutron stars. We conclude by summarizing the connection between our results and the observations, as well as giving predictions for future observations and discussing constraints on NS formation.
2 Method
Our numerical methods generally remain the same as described in Ivanova et al. (2005a, 2006). To study the population of NSs, we introduced a number of updates to the population synthesis model specifically for NS formation, and we revised the treatment of dynamical events with NSs. We also optimized numerical aspects of our code for more efficient use of parallel supercomputers.
2.1 Population synthesis updates
We adopt the binary evolution model from the population synthesis code StarTrack (Belczynski et al., 2002; Belczynski et al., 2008). In our first study of binary populations in GC cores (Ivanova et al., 2005a) we used a previous version of StarTrack described in Belczynski et al. (2002). In our study of close binaries with white dwarfs (Ivanova et al., 2006, paper I) we employed a more recent version of StarTrack available at the time (as in Belczynski et al., 2008, but including updates only for WD evolution compared to the previous version). In this paper we incorporated all the latest updates of StarTrack, as described in Belczynski et al. (2008). Below we outline several of the most important changes in StarTrack that affect the formation and evolution of NSs compared to our previous studies of NSs in GCs (Ivanova & Rasio, 2004; Ivanova et al., 2005b; Ivanova et al., 2006). We also provide detailed descriptions of further modifications to the StarTrack model, beyond those in Belczynski et al. (2008), which we developed for the treatments of electron-capture supernovae and common-envelope events.
2.1.1 Natal kicks and supernova disruptions
At the moment of formation, both NSs and BHs receive additional speed (a natal kick), most likely due to asymmetric supernova ejecta (Fryer, 2004). Although most recent simulations are in relatively good agreement with the measured distribution of pulsar velocities, the agreement is not yet firmly established (Scheck et al., 2006). In this paper we adopt the most recently derived pulsar kick velocity distribution from Hobbs et al. (2005), which is a Maxwellian distribution with one-dimensional RMS velocity km s*-1* (the mean three-dimensional velocity is km s*-1*). In contrast to some earlier studies (e.g., see Arzoumanian et al., 2002), no evidence for a bimodal velocity distribution is present.
At the moment of the explosion, the supernova progenitor can be a binary companion. Because of the kick, the binary can become unbound (and the binary components will separate) or the entire binary can be affected by the natal kick. In our previous studies, in order to find the velocities of the components of an unbound binary, we used the derivation by Tauris & Takens (1998), where the pre-supernova orbit is assumed to be circular. The approach used in the current version of StarTrack allows us to properly calculate cases with an arbitrary pre-explosion orbital eccentricity (for full details, see §6.3 in Belczynski et al., 2008).
2.1.2 Electron-capture supernovae
Through our studies of accreting WDs (Ivanova et al., 2006) we found that electron-capture supernovae (ECSe) could be the dominant source of retained NSs in clusters. Similar results were also recently obtained in Kuranov & Postnov (2006). We have therefore considered in great detail all possible types of ECSe for this work.
It has been argued that when a degenerate ONeMg core reaches , its collapse is triggered by electron capture on 24Mg and 20Ne before neon and subsequent burnings start and, therefore, before the iron core formation (Miyaji et al., 1980; Nomoto, 1984, 1987; Timmes & Woosley, 1992; Timmes et al., 1994). The explosion energy of such an event is significantly lower than that inferred for core-collapse supernovae (Dessart et al., 2006; Kitaura et al., 2006). There are several possible situations when a star can develop a degenerate ONeMg core that will eventually reach .
First, this can occur during the normal evolution of single stars. It was initially proposed by Barkat et al. (1974) that a star of 7–10 , after non-explosive carbon burning, develops an ONeMg core. If the initial core mass is less than required for the neon ignition, , the core becomes strongly degenerate. Through the continuing He shell burning, this core grows to . In more massive stars, , carbon, oxygen, neon and silicon burnings progress under non-degenerate conditions, and, in less massive stars, ONeMg cores never form. The formation of degenerate or non-degenerate ONeMg cores depends on the He core mass at the start of the asymptotic giant branch, where is a rough boundary between the cases (Nomoto, 1984). If the initial He core mass is below , no off-centre ignition will happen—the carbon core burning will occur when the degenerate core reaches the Chandrasekhar mass, resulting in a thermonuclear explosion (Hurley et al., 2000, also J. Eldridge 2006, priv. comm.).
The conditions for ECS were shown to occur in single stars of initial mass 8–10 by Nomoto (1984). This critical mass range depends on the properties of the He and CO cores, which, in turn, are highly dependent on the mixing prescription (semiconvection, overshooting, rotational mixing, etc.) as well as on the adopted opacities. As a result, the range varies between different evolutionary codes (see discussion in Podsiadlowski et al., 2004; Siess, 2006), and is generally reduced compared to that proposed initially by Nomoto (1984, 1987). In the code that we use for our cluster simulations, a non-degenerate ONeMg core is formed when the initial He core mass is about (Pols et al., 1998; Hurley et al., 2000) and the range of initial masses for single stars of solar metallicity that leads to the formation of such a core is 7.66 to 8.26 . The equivalent mass ranges are from 6.85 to 7.57 and from 6.17 to 6.76 for single stars with GC-like metallicities, and , respectively.
The initial stellar mass is not the only parameter that defines whether a star will form a degenerate ONeMg core. It was recently pointed out by Podsiadlowski et al. (2004) that the range of progenitor masses for which an ECS can occur depends also on the mass transfer history of the star, and therefore may be different for binary stars, making it possible for more massive progenitors to collapse via ECS. We will refer to the single and binary scenarios for ECS in non-degenerate stars, as described above, as evolution-induced collapse (EIC).
The second possibility for an ECS to occur is by accretion on to a degenerate ONeMg WD in a binary: accretion-induced collapse (AIC). In this case, a massive ONeMg WD steadily accumulates mass until it reaches the critical mass (for more detail on the adopted model of the accretion on the WD see Ivanova & Taam, 2004; Belczynski et al., 2008). In addition, ONeMg WDs can be formed from CO WDs via off-centre carbon ignition if the mass of CO WDs is above 1.07 and the accretion rate is high, (Kawai et al., 1987) so that eventually this also leads to AIC.
The third case that we consider is coalescing double WDs with a total mass exceeding . To distinguish it from AIC, we will refer to this case as merger-induced collapse (MIC). The nature of NS formation is the same as in the case of AIC (accumulation of mass by a massive ONeMG WD). In the case of coalescing CO WDs, then, as in the case of single stars, off-centre carbon ignition occurs first, which quiescently converts the star into an ONeMg core, and this proceeds with an ECS (Saio & Nomoto, 1985, 2004). Although the approach above was proposed for coalescing WDs in binaries only, we assume that in the case of collisions between two WDs it is applicable as well: except for the rare cases of near head-on collisions, a less massive WD will be tidally disrupted and accreted at high rate on to the more massive WD.
We therefore assume in our simulations that a NS will be formed via ECS in the following cases:
- •
if at the start of the AGB the He core mass is (EIC)
- •
if an accreting ONeMg WD reaches (AIC)
- •
if the total mass of two merging or colliding WDs exceeds (MIC).
For NSs formed via ECS we assume that the accompanying natal kick is 10 times smaller than in the CC case. This assumption follows from the results of numerical simulations which find that the SASI instability, required by current understanding for the large explosion asymmetry in the case of core-collapse supernovae, fails to develop and resulting kicks are significantly smaller and on overall do not exceed 100 km/s (Buras et al., 2006; Kitaura et al., 2006). We will provide the separate statistics for NSs created by different channels (core-collapse, EIC, AIC or MIC), so that if one of the channels later turns out to have been overestimated it will be easy to recalibrate.
The gravitational mass of the newly formed NS is . This is of its baryonic mass (which cannot exceed the pre-collapse core mass) after subtracting the binding energy and is calculated as in Timmes et al. (1996):
[TABLE]
where the masses are in solar masses. As the angular momentum of the binary is conserved, in the case of no kick, the binary widens.
The condition for ECS to occur depends essentially on the central density of the object. As a consequence, a rapidly rotating WD can reach a much higher mass than the Chandrasekhar limit before the central density becomes high enough for electron captures on 24Mg and 20Ne to occur. For example, its mass can be as high as if the ratio of the WD rotational energy to the gravitational binding energy is 0.0833 and the star is highly oblate (Dessart et al., 2006). Such rapidly rotating heavy WDs can be formed, e.g., during the coalescence of two WDs with total mass exceeding the Chandrasekhar mass. During such a merger, less than 0.5 per cent of mass will be lost from the system (Guerrero et al., 2004). The collapse of a rapidly rotating WD can therefore lead to the formation of a more massive and very fast spinning NS.
2.1.3 Convective and radiative envelopes
The presence or absence of a deep outer convective envelope influences a star’s behavior during various phases of close binary interactions (e.g., angular momentum loss via magnetic braking, tidal interactions, stable or unstable mass transfer). In particular, it has been pointed out in Ivanova (2006) that the absence of a deep outer convective zone in low-metallicity GCs may explain the preferential formation of LMXBs in metal-rich clusters (Grindlay, 1993; Bellazzini et al., 1995; Zepf et al., 2006).
In this study, for main-sequence stars we adopt the metallicity-dependent mass range to develop a convective envelope described by (eq.(10) in Belczynski et al., 2008) In addition, in our previous work, all giant-like stars were assumed to have outer convective envelopes. Now stars crossing the Hertzsprung gap and stars in the blue loop are not assumed to have convective envelopes if their effective temperature satisfies .
2.1.4 Common-envelope phases
For common-envelope (CE) events we use the standard energy formalism (Livio & Soker, 1988). In this approach, the outcome of the CE phase depends on the adopted efficiency for orbital energy transfer into envelope expansion energy and on the donor envelope central concentration parameter . The commonly used prescription is , and this is adopted in our models as well. However, for He stars with an outer convective envelope we determine from the set of detailed He star evolution models calculated by Ivanova et al. (2003). In this case, , where is the radius of the He star in solar radii. As many He stars end their lives as NSs, the CE evolution of He stars is an important channel for the formation of close binaries with NSs. Indeed, the prescription above was recently used in a study of double neutron star formation as short -ray burst progenitors (Belczynski et al., 2006).
2.2 Dynamical events
2.2.1 Binary formation via physical collisions with giants
We consider in our models the possiblity of (eccentric) binary formation via physical collisions involving a red giant (RG). For binaries formed through NS–RG collisions, the final binary separation and eccentricity depend on the closest approach distance (Lombardi et al., 2006) and can be estimated using the results of hydrodynamic calculations as
[TABLE]
[TABLE]
More details on our treatment of these collisions can be found in Ivanova et al. (2006). In this paper, we refer to this event as to a dynamical common envelope (DCE).
2.2.2 Collisions and mergers with NSs
We assume that the amount of matter accreted by a NS following a merger or collision depends on the evolutionary stage of the other star. If a NS collides or merges with a MS or a He MS star, we assume that it accretes no more than 0.2 (Rosswog, 2006). In the case of a collision with a giant-like star, the maximum accreted mass is 0.01 (Lombardi et al., 2006). When the other star is a WD, all the WD mass is accreted. In essentially all mergers and collisions, the accreted mass must have gone through an accretion disk, implying that, if a significant amount of matter is accreted, a recycled pulsar will be formed.
2.2.3 Triples and their treatment
Stable hierarchical triples can be formed through binary–binary encounters. Although these triples would be stable in isolation, it is likely that they will be destroyed during their next dynamical encounter. As there are no developed population synthesis methods for handling triple star evolution, we cannot keep these triples in our simulations and instead we have to break them immediately after formation into a binary and a single star. We do this while conserving energy: the energy required to eject the outer companion is balanced by shrinking the inner binary orbit. The outer companion is released unless the required energy is such that the inner binary merges. In the latter case the inner system is allowed to merge and the outer companion is kept in a new, wider orbit to form the final binary system.
It is possible that in a triple the inner orbit’s eccentricity will be increased via the Kozai mechanism (Kozai, 1962). This secular coupling causes large variations in the eccentricity and inclination of the orbits and could drive the inner binary of the triple system to merge before the next dynamical interaction. The maximum eccentricity in systems with large initial inclinations , such that () is (e.g., Innanen et al., 1997; Eggleton & Kiseleva-Eggleton, 2001):
[TABLE]
The period of the cycle to achieve is (Innanen et al., 1997; Miller & Hamilton, 2002)
[TABLE]
where is the initial eccentricity of the inner binary, , and are the masses of the inner binary companions and the mass of the outer star, and are initial orbital separations for the inner and outer orbits, and is the semiminor axis of the outer orbit.
We compare the Kozai time-scale with the collision time (computed as in Ivanova et al., 2005a):
[TABLE]
Here is the triple period in days, is the total triple mass in , is the mass of an average single star in , and , where is the stellar number density.
If , the Kozai mechanism does not affect the triple evolution before the next encounter occurs, and we break the triple as describe above. However, for triples with , we consider the circularization time-scale (see §3.3 in Belczynski et al., 2008). The characteristic time-scale for the inner binary separation to decrease through tidal dissipation is (Mazeh & Shaham, 1979). If , the inner binary will shrink until one of the components overfills its Roche lobe or its separation will small enough to cause the Darwin instability resulting in a merger (Eggleton & Kiseleva-Eggleton, 2001). In the opposite case, when , we assume that mass transfer starts if is such that at least one of the stars in the inner binary overfills its Roche lobe at pericentre. As a result, triple formation may enhance the formation of mass-transferring binaries with a NS.
For sufficiently compact inner binaries, relativistic effects may play an important role. The relativistic precession period of the inner binary, to first post-Newtonian order, is (see Weinberg, 1972, p.197):
[TABLE]
If , Kozai cycles are strongly suppressed (Holman et al., 1997).
We expect that some dynamically formed triples can also undergo significant secular eccentricity evolution even for orbital inclinations smaller than the Kozai angle (see e.g. Ford et al., 2000), but here we neglect this possibility.
3 Radio Pulsar Modeling
3.1 Slow, non-recycled radio pulsars
The time over which a pulsar will spin down from its initial period to a current period is
[TABLE]
where [s]. With the assumption that the magnetic field does not decay, remains constant with time.
Several observed radio pulsars in GCs are slowly spinning (s) and have high s (higher than can be produced by gravitational acceleration in the cluster potential) implying relatively high fields, Gauss. These include NGC 6342A (van Kerkwijk et al., 2000), NGC 6624B (Biggs et al., 1994), and NGC 6440A (Lyne et al., 1996). Their inferred characteristic ages range from yr to yr, compared to yr for “normal” MSPs. These pulsars cannot have been formed in their current state through core-collapse SNe, which only occur within the first ten million years. The inferred birthrate for this class of objects is similar to that of “normal” MSPs, but observationally they seem to be concentrated in the densest clusters, and 2 of 3 are isolated pulsars.
A strong magnetic field of -gauss can be produced by the collapse, assuming flux conservation, of a typical magnetic WD with -gauss. Observed magnetic WDs typically have higher masses than nonmagnetic WDs(Wickramasinghe & Ferrario, 2005), indicating that they may dominate the population of WDs that undergo AIC. Stellar dynamo theory supports the link between strong magnetic fields and massive WDs (Thompson & Duncan, 1993). It is also possible that a strong magnetic field may be generated during the merger of a WD pair, where one or more WDs had a strong magnetic field, leading to the production of -gauss magnetic fields, or even magnetar field strengths of -gauss (King et al., 2001; Levan et al., 2006; Chapman et al., 2006). MSPs formed with such high magnetic fields will have relatively short lifetimes, and thus will make up only a small portion of the observed globular cluster MSP population. In our simulations, we do not assign a specific magnetic field to formed neutron stars. However we keep a record of their formation and therefore can test whether the population of AIC and MIC NSs is formed at the required rates to explain the population of slow GC radio pulsars, assuming that their life-time is only years.
We refer to young, recently formed pulsars that have not yet been recycled as “1a” and “1b” in Fig. 1. As a standard picture, we consider that the origin of a NS does not affect its further evolution if a MT event occurred in a dynamically formed binary (c.f. post-AIC evolution), but we note that it is possible that MIC NSs might have a higher than usual magnetic field and their subsequent evolution might be different – they can be seen only for a shorter time after their formation as young slow pulsars and they might be not able to accrete and be spun up.
3.2 Mild recycling in post-AIC systems
Very little is known about whether a magnetic field is always reduced during mass accretion, independently of its initial strength. E.g., the UCXB and 7.7 second X-ray pulsar 4U 1626-67 has a strong (-gauss) magnetic field (Orlandini et al., 1998). On the other hand, known magnetic fields in accreting MSPs are usually - gauss (e.g., -gauss in XTE J1751-305, Di Salvo & Burderi (2003); -gauss in IGR J00291+5934 Torres et al. (2007); -gauss in XTE J0929-314, Galloway et al. (2002)) and are consistent with the magnetic fields of MSPs. Even though 4U 1626-67 is evolving through the accretion phase and has passed the stage when the MT rate is maximum, the NS has kept a high magnetic field and has not been spun up to millisecond periods. To explain this, it has been suggested that the NS in 4U 1626-67 was recently formed via AIC (Yungelson et al., 2002). Several radio pulsars in the Galaxy possess relatively high fields and long periods, with low-mass companions in low-eccentricity orbits (Breton et al., 2007). They argue that these systems can be well-explained through AIC.
It is unclear why post-AIC binaries may yet fail to produce MSPs. Reasons could include the high initial field and moderately fast spin of newly-formed AIC NSs, which could disrupt spin-up; or simply that after sufficient accretion onto the WD to reach the Chandrasekhar limit, there is usually not enough remaining mass in the companion to spin the NS up to millisecond periods. We will show below that producing MSPs in the same binaries in which AIC occurred contradicts observations. For our final results, we therefore adopt that NSs formed via AIC or MIC have short lifetimes ( years), and that NSs recycled in their original AIC systems are only mildly recycled, also with lifetimes of years (case “2a” in 1).
3.3 Recycling via stable MT
A common understanding of recycled pulsar formation is that the NS is recycled through disk accretion and has its magnetic fields reduced during the accretion phase (van den Heuvel, 1984). Since the characteristic lifetime of a MSP is years and is significantly larger than a GC life-time ( years), we do not consider in our simulations the evolution of MSPs towards a death zone after they are formed.
The amount of material that needs to be accreted on to a NS to form a MSP is however not firmly established. In our simulations, we record for further analysis any event through which a NS acquires at least 0.01 . We also distinguish mass gain through steady accretion from that which can result from a physical collision with a red giant (DCE); a CE event during binary evolution; or a merger or physical collision with a MS star or a WD.
The connection between stable mass-transfer and the binary MSP production in the field seems to be reasonably well-understood and is reviewed by Deloye (2007). We outline only the main branches:
- •
If the donor is a RG or an asymptotic-giant branch companion, then the resulting binary MSP has a wide orbital separation (“4b” at the Fig. 1).
- •
If the MT occurred in a NS-MS binary, then the evolution depends on the period at the start of the MT. Some systems will evolve towards shorter periods(“3a” at the Fig. 1), and others to longer (“4d” at the Fig. 1) (for a review of the current understanding of NS-MS LMXBs evolution see Deloye, 2007). The progenitors to the short-period binary MSPs evolve slowly enough that they may be detected as (bright or quiescent) LMXBs; systems with these orbital periods are indeed seen in globulars (Heinke et al., 2005a; Altamirano et al., 2007).
- •
If a binary system with a NS has evolved through a CE resulting in a significant shrinkage of a binary orbit, MT with a WD donor can start. This leads to the formation of an ultra-compact X-ray binary (UCXB) (“3b” at the Fig. 1).
We do not discuss here the case of thermal-time scale MT that occurs with an intermediate-mass donor () in a close orbit at the start of the MT (0.3-2 days), as we find it unlikely for a NS in a GC environment to acquire an intermediate mass donor (c.f. Rasio et al., 2000). We will return to this point later in §8 and only note here that the evolutionary life-time of an intermediate mass donor is not much longer than the cluster age when ECS NSs are formed.
Even if the MT leads to spin-up, a NS does not appear as a radio MSP immediately even if has accumulated sufficient mass. While the MT rate remains high, systems continue to appear as LMXBs. In general, LMXBs with a likely MS donor do not show up as radio MSPs while the companion mass is above (see Fig. 1 in Deloye, 2007).
3.4 Short-period radio bMSPs
There are discrepancies between the observed orbital periods of short-period radio binary MSPs (bMSPs) with very low-mass companions and the modeled periods. Using a standard mass-radius relation for WDs during the MT evolution of a NS-WD binary gives periods several times smaller than those of observed bMSPs with companions of similar masses. It has been shown that even the consideration of more realistic WD models does not change the result strongly – hotter donors don’t, by themselves, produce the long periods needed to explain the short-period bMSPs (Deloye & Bildsten, 2003; Deloye, 2007). If the mass-losing companion is a WD, some poorly understood process must be invoked – e.g., strong tidal heating of the WD during MT (Rasio et al., 2000). In that case, it is not possible to accurately predict the bMSP periods during the MT or at its end.
There is no clear termination point for MT from a degenerate donor. As a result, the MT continues uninterrupted and within the Hubble time the companion mass can reach below . No observed radio bMSPs have been shown to have such low-mass companions, in globular clusters or the field.
Several accretion-powered MSPs with minutes have been identified, consistent with (partly) degenerate He or C/O donors (e.g. Krimm et al., 2007), and several are known in globular clusters (e.g. Dieball et al., 2005). However, no radio bMSPs have been observed with such short periods. These systems will spend Gyrs as transient LMXBs, with continually decreasing mass transfer rates (“3b” and “6a” in Fig. 1). It is possible that the UCXBs in clusters remain in this configuration, never turning on as millisecond pulsars, for a Hubble time. A sizable number of such systems could remain hidden in globular clusters if their accretion outburst intervals are sufficiently long. Alternatively, partly degenerate (due to irradiation) donors will have a minimum companion mass for a given temperature, indicating that evaporation of the donor star and production of isolated millisecond pulsars is a likely outcome (Bildsten, 2002; Deloye & Bildsten, 2003). Other possibilities include very strong tidal heating of the companion leading to strong orbital expansion, producing the low-mass eclipsing pulsars (Rasio et al., 2000), or spin-down of the NS due to decreasing mass transfer rates (see discussion in Deloye, 2007). Or, current pulsar detection algorithms may not yet be sufficiently sensitive to detect such short-period systems (e.g. Hessels et al., 2007). We do not know which of these fates befalls ultracompact LMXBs, although we know observationally that ultracompact radio bMSPs (“6a” in Fig. 1) have not yet been seen. We therefore count the numbers of ultracompact systems that have evolved past 55 minutes (the longest orbital period yet observed for an ultracompact LMXB, Krimm et al., 2007) and tabulate them as ultracompact MSPs, but do not plot their final binary periods and masses; their true final fate is uncertain.
In the case of MT in a NS-MS binary, if the resulting binary has a short period, it will have a (partly) degenerate hydrogen companion. Such a companion is larger than a hydrogen-poor WD, and better matches the observed periods of known low-mass eclipsing radio bMSPs than systems with WD donors. It is thought that the MT becomes unstable and leads to a merger of the two stars when the hydrogen-rich companion becomes less massive than (for analytic estimates see King et al., 2005), producing a single MSP (“4a” in fig. 1). However, current binary evolution codes lack the equation of state appropriate for low temperatures. Therefore, as of today, there are no direct mass transfer calculations that can proceed below donor masses of 0.04 Msun (van der Sluys et al., 2005a, b), which is well above the masses of observed bMSPs with very low-mass companions. The main reason is that an equation of state that is appropriate for extremely low-mass stars (e.g., Saumon et al., 1995) is not yet incorporated in current binary evolutionary codes.
Our code evolves a NS-MS binary down to very low masses assuming a simple mass-radius dependence for a hydrogen-rich degenerate donor. We find that this evolution ends as a merger when the donor mass is , where this value varies a bit depending on the initial donor mass. While evolving, the system appears in a globular cluster as first a qLMXb and then as a bMSP for substantial lengths of time. The values of the donor mass when the instability occurs, and the orbital periods during the qLMXB stage, could use refinement when improved evolutionary simulations are available.
3.5 Recycling during fast (“dynamical”) events
We consider here mass gain which can result from a physical collision with a red giant (DCE,case “4e”, see Fig. 1) ; a CE event during binary evolution (case “4c”); or a merger or physical collision with a MS star or WD (“2b” at the Fig. 1).
During a physical collision with a giant, the NS retains of the giant’s envelope (Lombardi et al., 2006). This matter has high specific angular momentum and the resulting accretion must proceed through an accretion disk, implying that a recycled pulsar can be formed.
A CE event is also likely to lead to spin-up. As an example, it is believed that the double neutron star system PSR J0737-3039 is a post-common envelope remnant (Dewi & van den Heuvel, 2004), where the first-formed NS has been spun up during a CE event to millisecond periods.
It is less clear whether possible accretion as a result of a merger leads to recycling. Such mergers have been considered to produce only mild spin up (Lyne et al., 1996). For a standard scenario we assume that physical collision of NSs with other stars leads only to the formation of short-living slow (isolated) pulsars.
3.6 Post-recycling phase
Once recycled, a single or binary MSP can participate again in dynamical encounters, and form new binary systems whose properties will depend on the type of binary formation: binary exchange (“5a”), tidal capture (“5b”) or another DCE (“5c”). This type of binary millisecond pulsars, along with post-DCE “4e”, should not be formed in the field population. Newly dynamically formed binaries with MSPs may evolve again through stable MT.
4 Globular cluster models
Our “standard” cluster model has an initial binary fraction of 100 per cent. The distribution of initial binary periods is constant in the logarithm between contact and d and the eccentricities are distributed thermally. We emphasize that about 2/3 of these binaries are soft initially (the initial binary fraction for hard binaries is about per cent if the 1-D velocity dispersion is 10 km/s) and most very tight binaries are destroyed through evolutionary mergers. Our initial binary fraction is therefore comparable to the initial binary fractions that are usually adopted in -body simulations, 10-15 per cent in hard binaries (soft binaries are short-lived in dense clusters but slow down the calculations considerably). For a more detailed discussion of the choice of primordial binary fraction, see Ivanova et al. (2005a).
For single stars and primaries we adopt the broken power law initial mass function (IMF) of Kroupa (2002) and we assume a flat mass-ratio distribution for secondaries. The initial core mass is 5 per cent of the total cluster mass and no initial mass segregation is assumed (an average object in the core is as massive as an average cluster star). At the age of 11 Gyr, which we adopt as a typical age of a Galactic GC, the mass of such a cluster in our simulations is for the metallicity of a typical metal-rich cluster () and is comparable to the mass of typical GCs in our Galaxy. In the case of a metal-poor cluster (Z=0.0005), the mass is . To make a comparison between cluster models of different metallicities we will show all results scaled to a cluster mass . As the metal-rich case observationally is more important – most bright LMXBs and MSPs detected so far are located there – we take for our standard model the value .
For our standard model we fix the core density at . The link to an observed luminosity density can be found using the number density-to-mass density ratio, which in our simulations is at the ages of 7-14 Gyr, and mass-to-light ratio, which on average is and is for 47 Tuc (Pryor & Meylan, 1993). We adopt a half-mass relaxation time Gyr (Harris, 1996). The characteristic velocities are computed for a King model with dimensionless central potential and the total model mass, giving a one-dimensional velocity dispersion km/s and a central escape velocity km/s. If, after an interaction or SN explosion, an object in the core acquires a velocity higher than the recoil velocity km/s, the object is moved from the core to the halo. The ejection velocity for objects in the halo is km/s. This standard model represents a typical dense globular cluster in our Galaxy.
In addition to this “standard” model we also considered cluster models with the following modifications:
- •
a metal-poor cluster with – “metal-poor”;
- •
different central densities: , and – “high-dens”, “med-dens” and “low-dens” respectively;
- •
a cluster with smaller velocity dispersion and smaller ejection velocity – “low-”;
- •
shorter half-mass relaxation time “long-”;
- •
reduced initial binary fraction, per cent – “BF05”;
- •
magnetic braking prescription of Rappaport et al. (1983) – “fast MB”;
- •
reduced efficiency of the common envelope, – “CE-reduced”;
- •
natal kick distribution is doubled peaked Maxwellian as in Arzoumanian et al. (2002) and NSs are produced only via core-collapse and double WD collisions (MIC), but not through EIC, AIC or evolutionary issued MIC – “oldkicks”
- •
47 Tuc-type cluster, characterized by a higher density (with M/L=2 as in (Pryor & Meylan, 1993) it corresponds to as in Harris (1996)) , metallicity , Gyr (Harris, 1996) km/s (Pryor & Meylan, 1993), (Webbink, 1985), with the recoil velocity of 51 km/s and km/s (King model ) – “47 Tuc”;
- •
Terzan 5 cluster, ( as in Heinke et al. (2003)), metallicity (Origlia & Rich, 2004), Gyr (Harris, 1996) km/s, (Webbink, 1985), with the recoil velocity of 39 km/s and km/s (King model ) – “Terzan 5”.333The recent study of Ortolani et al. (2007) suggests a shorter distance, and thus higher central density for Terzan 5 (/pc3). If this study is correct, then Terzan 5’s characteristics may be more closely approximated by our “high-density” cluster.
For the complete list of models see Table 1.
We expect that only a few LMXBs can be formed in every cluster with a typical mass of 200,000 through 11 Gyr of dynamical evolution, and therefore we perform 5 runs for each model with initial stars and for the standard models - five runs with stars, to check that statistics for rare objects like LMXBs are not different when the resolution is increased (for most of our results presented below we did not find statistical differences between high and low resolution models, and fluctuations for the rate events are smaller in the case of high resolution). For Terzan 5 and 47 Tuc models, though, we do all runs at high resolution. In particular, in the case of Terzan 5 stars provide at 11 Gyr the mass comparable to the estimated observed mass of 370,000 . In the case of 47 Tuc the reason for a higher resolution is that the relative core mass is smaller than in the case of a standard cluster, due to longer .
5 Neutron star formation and retention
5.1 Formation
To estimate how many NSs can be formed and retained in GCs, we first studied stellar populations without dynamics, considering separately populations with only primordial single stars and only primordial binaries. We considered several metallicities – from solar (as in the Galactic field) to that of a typical metal-poor cluster: Z=0.02, 0.005, 0.001, 0.0005. In addition, for comparison with studies done by different researchers in the past, we examined a model (with Z=0.005) with the previously widely accepted natal kick distribution, 2 Maxwellians with a lower peak for kick velocities at 90 km/s (Arzoumanian et al., 2002). In this model we adopt that NSs are produced only via core-collapse or through collisional WD mergers, no evolutionary ECS. The results for NS production via different channels are shown in Table 2.
The production of NSs via core-collapse SNe (CC NSs), per unit of total stellar mass at 11 Gyr, decreases as metallicity increases. The difference between the Galactic field case and a metal-poor case is about 20 per cent, while the difference between metal-rich and metal-poor clusters (Z=0.005 and Z=0.0005) is about 10-15 per cent, for primordial single or binary populations.
EIC in single stars for metal-rich populations occurs in stars of higher masses. The range of initial masses is, however, the same for both populations and is about . As a result, the number of NSs produced via EIC (EIC NSs) from the population of single stars is smaller in the case of a Galactic field population than in the case of a GC population by 30 per cent, while the difference between metal-rich and metal-poor clusters is only 5 per cent. This is in complete agreement with the adopted power law for the IMF. In binaries, the mass interval where EIC could occur is expanded compared to single stars of the same metallicity.
NSs produced via AIC and MIC also have similar production rates in metal-rich and metal-poor populations. We note that MIC gives the smallest contribution to the production of NSs in populations evolved without dynamics.
It can be seen that once we adopt that a NS can be created via an ECS, the number of formed NSs via CC is depleted – some stars would be evolving via core-collapse, but have experienced the condition for a EIC before that.
5.2 Retention
In Fig. 2 we show the retention fraction of NSs at formation as a function of fixed escape velocity the whole population, considering populations of metal-poor primordial singles and binaries; CC NSs and the whole NS population are shown separately. It can be seen that, with the adopted natal kick distribution, CC NSs from a single star population are rarely retained (see also Table 2). In a population of primordial binaries, the fraction of retained CC NSs is higher than for single stars. None the less, the resulting number of retained CC NSs is significantly smaller than we get from the other NS formation channels. We conclude that if indeed EIC, AIC or MIC are not accompanied by a large natal kick, the total number of retained NSs is high and only a negligible number of CC NSs is present. As the EIC/AIC NS masses are only , this may lead to a NS mass function skewed towards low values.
With the old distribution for natal kick velocities, reasonable numbers of NSs can be produced and retained in a typical GC, while a massive cluster like 47 Tuc could retain as many as 600 NSs (see also Fig. 2). This result agrees with the retention fractions obtained in Pfahl et al. (2002).
The reference list from the paper itself. Each links out to its DOI / PubMed record.
- 1Altamirano et al. (2007) Altamirano D., Casella P., Patruno A., Wijnands R., van der Klis M., 2007, Ar Xiv e-prints, 708
- 2Anderson et al. (1990) Anderson S. B., Gorham P. W., Kulkarni S. R., Prince T. A., Wolszczan A., 1990, Nature, 346, 42
- 3Arzoumanian et al. (2002) Arzoumanian Z., Chernoff D. F., Cordes J. M., 2002, Ap J, 568, 289
- 4Barkat et al. (1974) Barkat Z., Reiss Y., Rakavy G., 1974, Ap JL, 193, L 21
- 5Belczynski et al. (2002) Belczynski K., Kalogera V., Bulik T., 2002, Ap J, 572, 407
- 6Belczynski et al. (2008) Belczynski K., Kalogera V., Rasio F. A., Taam R. E., Zezas A., Bulik T., Maccarone T. J., Ivanova N., 2008, Ap J Supp, 174
- 7Belczynski et al. (2006) Belczynski K., Perna R., Bulik T., Kalogera V., Ivanova N., Lamb D. Q., 2006, Ap J, 648, 1110
- 8Belczynski et al. (2007) Belczynski, K., O’Shaughnessy, R., Kalogera, V., Rasio, F., Taam, R., & Bulik, T. 2007, Ar Xiv e-prints, 712, ar Xiv:0712.1036, Science, submitted
