James C. Higdon

Cosmic-Ray Acceleration from Supernova Ejecta in Superbubbles

We suggest that the cosmic rays are accelerated primarily out of the supernova ejecta-enriched matter in the interiors of superbubbles. These hot, low-density superbubbles, which reach dimensions of several hundred parsecs, are generated by the winds and ejecta of supernova explosions of massive stars formed in giant molecular cloud OB associations that last for tens of megayears. Since these bubbles expand with shell velocities that are much faster than the dispersion velocities of the O and B star progenitors of the supernovae that power the bubbles, the bulk of the supernovae occur in their cores. The expanding remnants of each of these supernovae fill only less than 1% of this core before they have slowed to sonic velocities. Thus, the bulk of these supernovae remnants, together with their metal-rich grain and gas ejecta and their cosmic-ray-accelerating shocks, are well confined within the cores of superbubbles. These cores can thus provide a source of cosmic-ray matter of essentially constant metallicity throughout the age of the Galaxy, which is required to account for the constancy of cosmic-ray-produced Be relative to supernova-produced Fe observed in halo stars formed in the early Galaxy. The interactions of the grains and gas in metal-rich superbubbles, with recurrent supernova shocks every ~3×105 yr, also reconcile the requirement of a supernova ejecta source of cosmic rays with the recent observations that require a greater than 105 yr delay between nucleosynthesis and acceleration for the cosmic-ray metals. Supernova-enriched bubble metallicity may also explain the X-ray emission from the interiors of superbubbles in the Large Magellanic Cloud.

Nature versus Nurture: The Origin of Soft Gamma-Ray Repeaters and Anomalous X-Ray Pulsars

Soft gamma-ray repeaters (SGRs) and anomalous X-ray pulsars (AXPs) are young and radio-quiet X-ray pulsars that have been rapidly spun-down to slow spin periods clustered in the range 5-12 s. Most of these unusual pulsars also appear to be associated with supernova shell remnants (SNRs) with typical ages less than 30 kyr. By examining the sizes of these remnants versus their ages, we demonstrate that the interstellar media that surrounded the SGR and AXP progenitors and their SNRs were unusually dense compared to the environments around most young radio pulsars and SNRs. We explore the implications of this evidence on magnetar and propeller-based models for the rapid spin-down of SGRs and AXPs. We find that evidence of dense environments is not consistent with the magnetar model unless a causal link can be shown between the development of magnetars and the external interstellar medium. Propeller-driven spin-down by fossil accretion disks for SGRs and AXPs appears to be consistent with dense environments since the environment can facilitate the formation of such a disk. This may occur in two ways: (1) formation of a pushback disk from the innermost ejecta pushed back by prompt reverse shocks from supernova remnant interactions with massive progenitor wind material stalled in dense surrounding gas or (2) acquisition of disks by a high-velocity neutron stars, which may be able to capture sufficient amounts of comoving outflowing ejecta slowed by the prompt reverse shocks in dense environments.

OB Associations, Supernova-generated Superbubbles, and the Source of Cosmic Rays

We have considered the effects of both the spatial and temporal clustering of OB stars and their subsequent core-collapse supernovae on their generation of superbubbles and their resultant role as the primary source of cosmic rays. Employing a wide range of astronomical and astrophysical observations, we determine quantitatively the fraction of Galactic core-collapse supernovae that occur in superbubbles. We show that the fraction of core-collapse supernovae occurring in superbubbles is high, ranging from ~80% (solely temporal correlations) to ~90% (only spatial correlations). In addition, we find that the singleton end of our stellar cluster distribution is sufficient to reproduce the observed relative number of OB field stars. Core-collapse supernovae (Types II and Ib/c) constitute 85% of Galactic supernovae; only a small fraction of the remaining class of supernovae, Type Ia, occur in superbubbles. Thus, ~75% of all Galactic supernovae are expected to occur within superbubbles. The occurrence of the great majority of Galactic supernovae in superbubbles has major implications for cosmic-ray acceleration. Acceleration of cosmic-ray nuclei heavier than He in enriched (ZSB ? 3 Z?) superbubble interiors can consistently explain the anomalous cosmic-ray 22Ne/20Ne ratio, the cosmic-ray actinde/Pt group and UPuCm/Th ratios, and the constant LiBeB/(C+O) ratio observed in very old, metal-poor stars. Finally, although only ~75% of supernovae occur in superbubbles, ~88% of the cosmic-ray heavy particles are accelerated there because of the factor of ~3 enhanced superbubble core metallicity.

Actinides in the Source of Cosmic Rays and the Present Interstellar Medium

The abundances of the actinide elements in the cosmic rays can provide critical constraints on the major sites of their acceleration. Using recent calculations of the r-process yields in core collapse supernovae, we have determined the actinide abundances averaged over various assumed time intervals for their supernova generation and their cosmic-ray acceleration. Using standard Galactic chemical evolution models, we have also determined the expected actinide abundances in the present interstellar medium. From these two components, we have calculated the U/Th and other actinide abundances expected in the supernova-active cores of superbubbles, as a function of their ages and mean metallicity resulting from dilution with interstellar cloud debris. Then, using observations of the fractions of Galactic supernovae that occur in superbubbles and in the rest of the interstellar medium, we calculate the expected actinide abundances in cosmic rays accelerated by Galactic supernovae. We find that the current measurements of actinide/Pt-group and preliminary estimates of the UPuCm/Th ratio in cosmic rays are all consistent with the expected values if superbubble cores have mean metallicities of around 3 times solar. Such metallicities are quite comparable to the superbubble core metallicities inferred from other cosmic-ray observations. Future, more precise measurements of these ratios with experiments such as ECCO (Westphal, Weaver, & Tarle 2001) are needed to provide a better measure of the mean source metallicity sampled by the local Galactic cosmic rays. Measurements of the cosmic-ray actinide abundances have been favorably compared with the protosolar ratio, inferred from present solar system abundances, to infer that the cosmic rays are accelerated from the general interstellar medium. We suggest, however, that such an inference is not valid because the expected actinide abundances in the present interstellar medium are very different from the protosolar values, which sampled the interstellar medium 4.5 Gyr ago and included an additional fresh ejecta component from a neighboring supernova.

Cosmic Rays, Dust, and the Mixing of Supernova Ejecta into the Interstellar Medium in Superbubbles

Most of the galactic core-collapse supernovae occur in OB associations that produce superbubbles, and thus, the bulk of the cosmic rays are accelerated in the cores of such superbubbles by the shock waves from these supernovae. Here we show that the initial mixing of the freshly synthesized elements from such supernovae with the gas and dust in the interstellar medium also occurs in the cores of these superbubbles, and that the unique composition of the galactic cosmic rays can be produced there by suprathermal ion injection from interactions of this mix of dust, gas, and supernova shocks. We further show that the basic features of the cosmic-ray composition provide a unique measure of the mixing ratio of the fresh supernova ejecta and the old interstellar medium in this initial phase of interstellar mixing.

THE GALACTIC SPATIAL DISTRIBUTION OF OB ASSOCIATIONS AND THEIR SURROUNDING SUPERNOVA-GENERATED SUPERBUBBLES

The Galactic spatial distribution of OB associations and their surrounding superbubbles (SBs) reflect the distribution of a wide range of important processes in our Galaxy. In particular, it can provide a three-dimensional measure not only of the major source distribution of Galactic cosmic rays, but also the Galactic star formation distribution, the Lyman continuum ionizing radiation distribution, the core-collapse supernova distribution, the neutron star and stellar black hole production distribution, and the principal source distribution of freshly synthesized elements. Thus, we construct a three-dimensional spatial model of the massive-star distribution based primarily on the emission of the H II envelopes that surround the giant SBs and are maintained by the ionizing radiation of the embedded O stars. The Galactic longitudinal distribution of the 205 μm N II radiation, emitted by these H II envelopes, is used to infer the spatial distribution of SBs. We find that the Galactic SB distribution is dominated by the contribution of massive-star clusters residing in the spiral arms.

HEAO 3 limits on the Ti-44 yield in Galactic supernovae

Data fron the high-resolution gamma-ray spectroscopy experiment on HEAO 3 have been searched for line emission from the decay of Ti-44 created in recent, as yet unobserved, Galactic supernova explosions, where the ages and locations are unknown. Because the 78 yr mean life of Ti-44 is comparable to the average time between Galactic supernovae, the gamma-ray line emission from its decay should appear as Galactic point sources. No evidence was found for such emission from a point source anywhere in the Galactic plane, with a 1-sigma limit of 8.3 x 10 exp -5 photons/sq cm per sec. Detailed models were developed to simulate the Galactic gamma-ray emission from the decay of Ti-44 produced in both type I and type II supernovae. These models were used with the measured gamma-ray line limits to constrain the supernova yields and recurrence periods.

Cosmic ray acceleration in superbubbles and the composition of cosmic rays

We review the evidence for cosmic ray acceleration in the superbubble/hot phase of the interstellar medium, and discuss the implications for the composition of cosmic rays and the structure and evolution of the interstellar medium (ISM). We show that the bulk of the galactic supernovae, their expanding remnants, together with their metal-rich grain and gas ejecta, and their cosmic ray accelerating shocks, are all confined within the interiors of hot, low-density superbubbles, generated by the multiple supernova explosions of massive stars formed in giant OB associations. This superbubble/hot phase of the ISM provides throughout the age of the Galaxy a cosmic ray source of essentially constant metallicity for acceleration by the shocks of many supernovae over time scales of a few Myr, consistent with both Be/Fe evolution and ACE observations of 59Ni/59Co. We show that the expected metallicity (>2 times Solar) and filling factor (>50%) of the superbubble/hot phase is high enough that the composition of cosm...

The Galactic 26Al Problem and the Close Binary Type Ib/c Supernova Solution?

The origin of the long-lived (1.07 Myr mean life) radioactive 26Al, which has been observed in the Galactic interstellar medium from its 1.809 MeV decay gamma-ray line emission, has been a persistent problem for over 20 years. Wolf-Rayet (W-R) winds were thought to be the most promising source, but their calculated 26Al yields are not consistent with recent analyses of the 1.809 MeV emission from the nearest W-R star and nearby OB associations. The expected 26Al yield from the W-R star exceeds, by as much as a factor of 3, that set by the 2 ? upper limit on the 1.809 MeV emission, while the W-R yields in the OB associations are only about of that required by the 1.809 MeV emission. We suggest that a solution to these problems may lie in 26Al from a previously ignored source: explosive nucleosynthesis in the core-collapse Type Ib/c supernovae (SNe Ib/c) of W-R stars that have lost most of their mass to close binary companions. Recent nucleosynthetic calculations of SNe Ib/c suggest that their 26Al yields depend very strongly on the final pre-SN mass of the W-R star and that those with final masses around 6-8 M? are expected to produce as much as 10-2 M? of 26Al per SN. Such binary SNe Ib/c make up only a small fraction of the current SNe Ib/c and only about 1% of all Galactic core-collapse SNe. But they appear to be such prolific sources that the bulk of the present 26Al in the Galaxy may come from just a few hundred close binary SNe Ib/c, and the intense 1.809 MeV emission from nearby OB associations may come from just one or two such SNe. More extensive SN Ib/c calculations of the 26Al yields versus pre-SN mass are clearly needed to test this possibility.

Near-infrared observations of GRS 1915+105

During 9 July 1995 and 28 April 1996 we carried out J, H, and Ks observations of the X-ray transient and Galactic black hole candidate GRS 1915+105 using the Cassegrain IR Camera on the 5-meter telescope at Mt. Palomar. Between the two observations the infrared intensity increased by approximately one magnitude to one of the highest levels yet seen, again confirming the highly variable nature of the emission. Our second observation was made during a time when the source was being continuously monitored by CGRO (20–100 keV), RXTE (2–12 keV), and the Ryle Telescope (15 GHz), allowing simultaneous multiwavelength spectra covering a very broad energy range.