In the light of recent observations in which short gamma-ray bursts are interpreted as arising from black-hole(BH), neutron-star(NS) or NS-NS mergings we would like to review our research on the evolution of compact binaries, especially those containing NS's. These were carried out with predictions for LIGO in mind, but are directly applicable to short gamma-ray bursts in the interpretation above. Most important in our review is that we show that the standard scenario for evolving NS-NS binaries always ends up with a low-mass BH (LMBH), NS binary. Bethe and Brown (1998) showed that this fate could be avoided if the two giants in the progenitor binary burned He at the same time, and that in this way the binary could avoid the common envelope evolution of the NS with red giant companion which sends the first born NS into a BH in the standard scenario. The burning of He at the same time requires, for the more massive giants such as the progenitors of the Hulse-Taylor binary NS that the two giants be within 4% of each other in ZAMS mass. Applying this criterion to all binaries results in a factor 5 of LMBH-NS binaries as compared with NS-NS binaries. Our scenario of NS-NS binaries as having been preceded by a double He-star binary is collecting observational support in terms of the nearly equal NS masses within a given close binary.

The discovery of the highly relativistic neutron star (NS) binary (in which both NS's are pulsars) not only increases the estimated merging rate for the two NS's by a large factor, but also adds the missing link in the double helium star model of binary NS evolution. This model gives $\sim 20$ times more gravitational merging of low-mass black-hole (LMBH), NS binaries than binary NS's, whatever the rate for the latter is.

In the hypernova/collapsar model of gamma-ray bursts, it is natural that radiation is emitted by the inner engine for some time after the burst. This has been discussed as a possible source of the X-ray line emission observed in some afterglows. We show here that the natural geometry of a hypernova --a source of radiation at the bottom of a deep funnel-- has very significant consequences for the shape and central energy of the observed emission lines: the lines acquire a very broad scattering wing on the low-energy side and a characteristic second peak one Compton wavelength away from the initial energy, due to the once- and twice-scattered photons. We suggest that this explains the large width of the observed lines. Furthermore, the downscattering lowers the central line energy (by up to 1 keV in the source rest frame, before the lines become unrecognizable), so that the observed line energies can become consistent with originating from cobalt and nickel, as expected in few-day-old supernova ejecta.

The discrepancy in the past years of many more black-hole soft X-ray transients (SXTs), of which a dozen have now been identified, had challenged accepted wisdom in black hole evolution. The sheer number of inferred SXTs requires that many, if not most, stars of ZAMS masses $20-35\msun$ end up as black holes (Portegies Zwart et al. 1997; Ergma & van den Heuvel 1998). In this paper we show that this can be understood by challenging the accepted wisdom that the result of helium core burning in a massive star is independent of whether the core is covered by a hydrogen envelope, or "naked" while it burns. The latter case occurs in binaries when the envelope of the more massive star is transferred to the companion by Roche Lobe overflow while in either main sequence or red giant stage. For solar metallicity, whereas the helium cores which burn while naked essentially never go into high-mass black holes, those that burn while clothed do so, beginning at ZAMS mass $\sim 20\msun$, the precise mass depending on the $^{12}C(\alpha,\gamma)^{16}$O rate as we outline. In this way the SXTs can be evolved, provided that the H envelope of the massive star is removed only following the He core burning. We also show that in order to evolve a black hole of mass $\gsim 10\msun$ such as observed in Cyg X-1, even employing extremely massive progenitors of ZAMS mass $\gsim 60\msun$ for the black hole, the core must be covered by hydrogen during a substantial fraction of the core burning. In other words, the progenitor must be a WNL star. We evolve Cyg X-1 in an analogous way to which the SXTs are evolved, the difference being that the companion in Cyg X-1 is more massive than those in the SXTs, so that Cyg X-1 shines continuously.

At least one, but more likely two or more, eccentric neutron-star, carbon-oxygen white-dwarf binaries with an unrecycled pulsar have been observed. According to the standard scenario for evolving neutron stars which are recycled in common envelope evolution we expect to observe $\gsim 50$ such circular neutron star-carbon oxygen white dwarf binaries, since their formation rate is roughly equal to that of the eccentric binaries and the time over which they can be observed is two orders of magnitude longer, as we shall outline. We observe at most one or two such circular binaries and from that we conclude that the standard scenario must be revised. Introducing hypercritical accretion into common envelope evolution removes the discrepancy by converting the neutron star into a black hole which does not emit radio waves, and therefore would not be observed.

We present a specific scenario for the link between GRB and hypernovae, based on Blandford-Znajek extraction of black-hole spin energy. Such a mechanism requires a high angular momentum in the progenitor object. The observed association of gamma-ray bursts with type Ibc supernovae leads us to consider massive helium stars that form black holes at the end of their lives as progenitors. We combine the numerical work of MacFadyen & Woosley with analytic calculations, to show that about 1E53 erg each are available to drive the fast GRB ejecta and the supernova. The GRB ejecta are driven by the power output through the open field lines, whereas the supernova is powered by closed filed lines and jet shocks. We also present a much simplified approximate derivation of these energetics. Helium stars that leave massive black-hole remnants in special ways, namely via soft X-ray transients or very massive WNL stars. Since binaries naturally have high angular momentum, we propose a link between black-hole transients and gamma-ray bursts. Recent observations of one such transient, GRO J1655-40/Nova Scorpii 1994, explicitly support this connection: its high space velocity indicates that substantial mass was ejected in the formation of the black hole, and the overabundance of alpha-nuclei, especially sulphur, indicates that the explosion energy was extreme, as in SN 1998bw/GRB 980425. (abstract shortened)

In this note we study the accretion disc that arises in hypercritical accretion of $\dot M\sim 10^8 M_{\rm Edd}$ onto a neutron star while it is in common envelope evolution with a massive companion. In order to raise the temperature high enough that the disc might cool by neutrino emission, Chevalier found a small value of the $\alpha$-parameter, where the kinematic coefficient of shear viscosity is $\nu=\alpha c_s H$, with $c_s$ the velocity of sound and $H$ the disc height; namely, $\alpha\sim 10^{-6}$ was necessary for gas pressure to dominate. He also considered results with higher values of $\alpha$, pointing out that radiation pressure would then predominate. With these larger $\alpha$'s, the temperatures of the accreting material are much lower, $\lsim 0.35$ MeV. The result is that neutrino cooling during the flow is negligible, satisfying very well the advection dominating conditions. The low temperature of the accreting material means that it cannot get rid of its energy rapidly by neutrino emission, so it piles up, pushing its way through the accretion disc. An accretion shock is formed, far beyond the neutron star, at a radius $\gsim 10^8$ cm, much as in the earlier spherically symmetric calculation, but in rotation. Two-dimensional numerical simulation shows that an accretion disc is reformed inside of the accretion shock, allowing matter to accrete onto the neutron star with pressure high enough so that neutrinos can carry off the energy.

We suggest Nova Sco 1994 (GRO 1655-40) as a possible relic of a Gamma Ray Burster (GRB) and Type Ib supernova (SN) explosion, showing that there is evidence both that the black hole was spun up by accretion and that there was a supernova explosion. We use the disc energy delivered from the rotational energy of the black hole to power the SN, and give arguments that roughly equal energy goes into the GRB and into the accretion disc to power the supernova.

In this article we consider the formation and evolution of black holes, especially those in binary stars where radiation from the matter falling on them can be seen. We consider a number of effects introduced by some of us, which are not traditionally included in binary evolution of massive stars. These are (i) hypercritical accretion, which allows neutron stars to accrete enough matter to collapse to a black hole during their spiral-in into another star. (ii) the strong mass loss of helium stars, which causes their evolution to differ from that of the helium core of a massive star. (iii) The direct formation of low-mass black holes ($M\sim2\msun$) from single stars, a consequence of a significant strange-matter content of the nuclear-matter equation of state at high density. We discuss these processes here, and then review how they affect various populations of binaries with black holes and neutron stars.

We suggest that high-mass black holes; i.e., black holes of several solar masses, can be formed in binaries with low-mass main-sequence companions, provided that the hydrogen envelope of the massive star is removed in common envelope evolution which begins only after the massive star has finished He core burning. Our evolution scenario naturally explains the gap (low probability region) in the observed black hole masses.

We use recent results in binary stellar evolution to argue that binaries with at least one black hole dominate the rate of compact-object mergers. Two phenomena generally attributable to such mergers, gamma-ray bursts and gravity-wave bursts, are therefore likely to originate from near the event horizon of a black hole. In addition to sheer numbers, black holes have an added advantage over neutron stars in both phenomena. For gamma-ray bursts, the presence of an event horizon eases the baryon pollution problem, because energy can be stored into rotation until most baryons have been swallowed, and then released into a cleaner environment via the Blandford-Znajek process. For gravity-wave bursts, black holes offer higher luminosities due to their higher masses, thus enabling detection out to larger distances, which leads to a 30-fold increase in the predicted LIGO event rate.

In this note we suggest that high-mass black holes; i.e., black holes of several solar masses, can be formed in binaries with low-mass main-sequence companions, provided that the hydrogen envelope of the massive star is removed in common envelope evolution which begins only after the massive star has finished He core burning. That is, the massive star is in the supergiant stage, which lasts only $\sim 10^4$ years, so effects of mass loss by He winds are small. Since the removal of the hydrogen envelope of the massive star occurs so late, it evolves essentially as a single star, rather than one in a binary. Thus, we can use evolutionary calculations of Woosley & Weaver (1995) of single stars. We find that the black holes in transient sources can be formed from stars with ZAMS masses in the interval $20-35\msun$. The black hole mass is only slightly smaller than the He core mass, typically $\sim 7\msun$.

We consider the merging of high-mass black-hole binaries of the Cyg X-1 type, in which the O-star companion becomes a neutron star following a supernova explosion. Our estimated rate of mergers, $\sim 2\times 10^{-5} yr^{-1}$ for the Galaxy, is relatively low because of the paucity of high-mass black holes. None the less, because of their high masses, these black-hole, neutron-star binaries could contribute importantly to the merging sought by LIGO. From stellar evolutionary calculations including mass loss, we estimate that a ZAMS mass of $\sim 80 \msun$ is necessary before a high-mass black hole can result in a massive binary. We estimate the detection rate for LIGO of high-mass black hole neutron-star mergers to be a factor $\sim 40$ greater than that for binary neutron stars. We suggest how our high cut-off mass can be reconciled with the requirements of nucleosynthesis, and show that a bimodal distribution with masses of black holes can account, at least qualitatively, for the many transient sources which contain high-mass black holes.

Beginning from massive binaries in the Galaxy we evolve black-hole, neutron-star binaries and binary neutron stars, such as the Hulse-Taylor system PSR 1913+16. The new point in our evolution is a quantitative calculation of the accretion of matter by a neutron star in common envelope evolution which sends it into a black hole. We calculate the mass of the latter to be $\sim 2.4\msun$. Our chief conclusion is that the production rate for black-hole, neutron-star binaries (in which the neutron star is unrecycled) is $\sim 10^{-4}$ per year per Galaxy, an order of magnitude greater than that of neutron star binaries. Not only should this result in a factor of $\sim$ 10 more mergings for gravitational wave detectors like LIGO, but the signal should be larger. We give some discussion of why black-hole, neutron-star binaries have not been observed, but conclude that they should be actively searched for.

None of the 1000 solar models in a full Monte Carlo simulation is consistent with the results of the chlorine or the Kamiokande experiments. Even if the solar models are forced artifically to have a \b8 neutrino flux in agreeement with the Kamiokande experiment, none of the fudged models agrees with the chlorine observations. The GALLEX and SAGE experiments, which currently have large statistical uncertainties, differ from the predictions of the standard solar model by $2 \sigma$ and $3 \sigma$, respectively.