Richard A. Kroeger

Gamma-Ray Spectral States of Galactic Black Hole Candidates

OSSE has observed seven transient black hole candidates: GRO J0422+32, GX 339-4, GRS 1716-249, GRS 1009-45, 4U 1543-47, GRO J1655-40, and GRS 1915+105. Two gamma-ray spectral states are evident, and based on a limited number of contemporaneous X-ray and gamma-ray observations, these states appear to be correlated with X-ray states. The former three objects show hard spectra below 100 keV (photon number indices Γ < 2) that are exponentially cut off with folding energy ~100 keV, a spectral form that is consistent with thermal Comptonization. This breaking gamma-ray state is the high-energy extension of the X-ray low, hard state. In this state, the majority of the luminosity is above the X-ray band, carried by photons of energy ~100 keV. The latter four objects exhibit a power-law gamma-ray state, with a relatively soft spectral index (Γ ~ 2.5-3) and no evidence for a spectral break. For GRO J1655-40, the lower limit on the break energy is 690 keV. GRS 1716-249 exhibits both spectral states, with the power-law state having significantly lower gamma-ray luminosity. The power-law gamma-ray state is associated with the presence of a strong ultrasoft X-ray excess (kT ~ 1 keV), the signature of the X-ray high, soft (or perhaps very high) state. The physical process responsible for the unbroken power law is not well understood, although the spectra are consistent with bulk-motion Comptonization in the convergent accretion flow.

Oriented Scintillation Spectrometer Experiment observations of Co-57 in SN 1987A

The Oriented Scintillation Spectrometer Experiment (OSSE) on the Compton Gamma Ray Observatory has observed SN 1987A for two 2 week periods during the first 9 months of the mission. Evidence for gamma-ray line and continuum emission from Co-57 is observed with an intensity of about 10 exp -4 gamma/sq cm/s. This photon flux between 50 and 136 keV is demonstrated by Monte Carlo calculations to be independent of the radial distribution of Co-57 for models of low optical depth, viz., models having photoelectric absorption losses of 122 keV photons no greater than several percent. For such models the observed Co-57 flux indicates that the ratio Ni-57/Ni-56 produced in the explosion was about 1.5 times the solar system ratio of Fe-57/Fe-56. When compared with nearly contemporaneous bolometric estimates of the luminosity for SN 1987A, our observations imply that Co-57 radioactivity does not account for most of the current luminosity of the supernova remnant in low optical depth models. We suggest alternatives, including a large optical depth model that is able to provide the SN 1987A luminosity and is consistent with the OSSE flux. It requires a 57/56 production ratio about twice solar.

Broadband X-Ray Spectra of the Black Hole Candidate GRO J1655–40

We present broad-band (2 keV to 2 MeV) X-ray spectra of GRO J1655-40, a luminous X-ray transient and occasional source of relativistic radio jets, obtained with RXTE and OSSE. In one observation, the luminosity is found to be 18% of the Eddington limit, which is one of the highest luminosities ever observed from GRO J1655-40. For this observation, we find that an adequate fit is obtained when a broad iron line and a reflection component are added to a model consisting of a power-law plus a soft excess component. The 95% confidence lower limit on the rms line width is 0.86 keV. The power-law component has a photon index of 2.72 and extends to at least 800 keV without a cutoff. After this observation, a significant drop in the (5-12 keV)/(1.5-5 keV) hardness ratio occurred on a timescale less than 2 hours. From an RXTE observation of GRO J1655-40 made after the hardness transition, we find that the power-law index is harder (2.415 +/- 0.011), the flux of the power-law component is lower, and the total luminosity is 10% of the Eddington limit. The change in the power-law component is consistent with the correlation between the spectral index and power-law flux previously reported for GRO J1655-40.

Considerations for the next Compton telescope mission

A high resolution Compton telescope has been identified by the Gamma Ray Astronomy Program Working Group (GRAPWG) as the highest priority major mission in gamma ray astrophysics following GLAST. This mission should provide 25–100 times improved sensitivity, relative to CGRO and INTEGRAL, for MeV gamma ray lines. It must have good performance for narrow and broad lines and for discrete and diffuse emissions. Several instrumental approaches are being pursued to achieve these goals. We discuss issues relating to this mission including alternative detector concepts, instrumental configurations, and background reduction techniques. We have pursued the development of position-sensitive solid-state detectors (Ge, Si) for a high spectral resolution Compton telescope mission. A ∼1 m2 germanium Compton telescope of position-sensitive germanium detectors was the basis for one of the GRAPWG concepts. Preliminary Monte Carlo estimates for the sensitivities of this instrument are encouraging. However, there are technic...

Germanium strip detector Compton telescope using three-dimensional readout

Compton telescopes using two germanium strip detectors with depth resolution have been demonstrated at the Naval Research Lab. Depth resolution allows interactions to be located to less than 1 mm, down from 1 cm with no depth resolution, which improves the imaging resolution of the telescope substantially. Compton images and reconstructed energy spectra of events in which gamma rays interact three times in the two detectors but did not deposit their full energy (i.e. Three-Compton) are examined. Finally, the multiple modes of the system including a standard Compton telescope and a Three-Compton telescope are compared.

Development and Applications of Position-Sensitive Solid-State Gamma Ray Detectors

The development of high-resolution position-sensitive, solid-state detectors will enable gamma ray detectors with improved sensitivity and imaging capabilities. The gamma ray astrophysics group at NRL has been developing germanium strip detectors for several years. We have shown that three-dimensional locations for gamma ray interactions can be determined with sub-millimeter accuracy, and have also demonstrated imaging capability within a single germanium strip detector. We have also initiated work on thick, silicon strip detectors. This was based on the fact that three sequential interactions can enable the energy and direction cone of the incident gamma ray to be determined, even without total energy deposition of the incident gamma ray. We are also working on low-power ASICs that are required to handle the large number of channels associated with arrays of strip detectors. Progress on this work will be presented, along with applications to high-energy astrophysics, medical imaging, nuclear physics, detection of fissile materials, and monitoring of environmental radioactivity.

Position resolution in a Ge-strip detector

We have investigated, both experimentally and theoretically, how to reconstruct in 3D the interaction positions for (gamma) -rays penetrating into a double-sized Ge cross trip detector. We found that when a suitable geometry is used, the 3D-reconstruction problem can be reduced to three 1D ones, which greatly simplifies the task. We report measurements on a 10mm thick detector with 2mm strip pitch, showing that at least 2mm position resolution can easily be achieved perpendicular to the detector plane. While the in- plane resolution is presently limited to the strip pitch we present work on progress in developing algorithms to improve this. This includes in particular the expected effects of the electronics and the interstrip capacitance on the signal shapes. Finally, we present captured waveforms that indicate the possibility of reconstructing more complex events such as Compton scattering.

Depth measurement in a germanium strip detector

We have demonstrated the ability to determine the depth of a gamma-ray interaction point over the full active volume of a thick germanium strip detector. This capability provides depth resolution of less than 0.5-mm full width half maximum (FWHM) at 122 keV in a device 11 mm thick with a 2-mm strip pitch. Fifty channels of electronics have been developed and tested with a 25 /spl times/ 25 germanium orthogonal strip detectors. Experiments examining the capabilities of the system and demonstrating a simple Compton telescope using a single detector have been performed.

Gamma Ray Observations of Cygnus X-1 With the Oriented Scintillation Spectrometer Experiment

Abstract : We report on ~120 days of observations of Cygnus X-1 with OSSE onboard the Compton Observatory. Emission is detected in the range 50 keV to 1 MeV and we find evidence for a continuum of hard X-ray flux levels rather than the existence of distinct flux states. Comparisons of the source spectra with various theoretical models show that an exponentially truncated power law best describes the average spectrum in the OSSE energy band. Although we have measured a new minimum in the hard X-ray flux from the source, no evidence was found for either a broad 1 MeV feature or a narrow 511 keV line previously reported in association with a low flux state. Upper limits on such emission features are an order of magnitude lower than earlier reported detections. The 5.6-day periodicity of the source measured at optical wavelengths was not detected with a sensitivity to the rms modulation fraction of 5% in the 60-140 keV energy band.

Hard X-rays from SN 1993J

The Oriented Scintillation Spectrometer Experiment (OSSE) on the Compton Observatory observed SN 1993J during three intervals centered approximately 12, 30, and 108 days after its outburst. Hard X-ray emission was detected in the first two of these intervals. No emission was seen in the third observation or in two earlier observations in 1991 and 1992. The coincidence of the observed excess with the outburst of SN 1993J and the consistency of the spectra and time evolution with those seen at lower energies by ROSAT and ASCA (Astro-D) argue that the observed emission is indeed from SN 1993J. It is probably due to the interaction of the fast supernova ejecta with circumstellar material. The luminosity, 5 x 10(exp 40) ergs/sec (50-150 keV) in the first interval, is significantly larger than predicted. Extrapolating the spectrum to a few keV accounts for most or all of the observed emission at low energy. The observed high temperature, 10(exp 9) K, is easily obtained in the shocked circumstellar matter, but a surprisingly high density is required there to give the observed luminosity, and little or no additional X-ray emission from denser shocked supernova ejecta is allowed. The hard emission might also be explained in terms of the shocked supernova ejecta itself with unexpectedly high temperature.