堂谷 忠靖

The bursting pulsar GRO J1744-28 was observed with ASCA on February 26-27, 1996. The source was detected at a persistent flux level of 0.9 Crab with the X-ray bursts followed by a shallow dip. Clear pulsation with apparent barycentric period of 467.054 msec was also detected. Energy spectrum of GRO 51744-28 shows large photoelectric absorption, which is consistent with the location close to the Galactic center. We detected structures in the energy spectrum of the persistent emission around 7 keV. The structure is most probably due to a partial covering of the X-ray emission region. (C) 1998 COSPAR. Published by Elsevier Science Ltd.

We report the results of a programme of analysis of the LMXB dipping sources, recently concentrating on the spectral changes taking place on rapid timescales during dip ingress and egress in XB 1916-053 using the high quality ASCA data. We tested the hypothesis that the source can be fitted by the same model that we have previously shown gives very good explanations of the two very different sources X 1755-338, the energy-independent dipper and X 1624-490, the Big Dipper. This complex continuum model consists of a blackbody identified with emission from the surface of a neutron star, plus a power law seen as Comptonised emission from an ADC. Analysis of the ASCA observation of XB 1916-053 shows this source to be remarkable in that all emission components are completely absorbed in dips, and that the rapid spectral evolution during dip ingress and egress is well fitted by our model. On the basis that the model fits these 3 very different sources we now propose that the complex continuum model will be able to fit all of the dipping sources. (C) 1998 COSPAR. Published by Elsevier Science Ltd.

Structure of the accretion disk is compared between the soft and hard states of Cyg X-l using the ASCA data. Large uncertainty of the disk parameters in hard state prevent us from drawing clear conclusion, but the data are consistent with a factor of 3 larger (optically thick) inner disk boundary in the hard state than in the soft state.

We report results of analysis of the ASCA observation of the low-mass X-ray binary dipping source XB 1916-053 made on 1993 May 2, during which dipping was very deep, so that in the deepest parts of dips, the X-ray intensity in the band 0.5-12.0 keV fell to zero, demonstrating that all emission components were completely removed. The best-fit orbital period of the binary system, determined from the X-ray data, was found to be 3005 +/- 10 s. The high-quality ASCA data allowed spectral evolution in dipping to be systematically investigated by spectral analysis in intensity bands covering the full range of dipping from intensities close to zero to nondip values. We have shown that the spectra can be well fitted by the same two-component model previously used to give good explanations of the very different dip sources X1755-338 and X1624-490, consisting of point-source blackbody emission from the neutron star and extended Comptonized emission, probably from the accretion disk corona. In the case of XB 1916-053, we show that all levels of dipping can be fitted using kT(bb) = 2.14 +/- 0.28 keV and power-law photon index = 2.42 +/- 0.21, which are the best-fit values for nondip data, together with the corresponding nondip normalizations. Dipping is shown to result from large increases of column density for the pointlike blackbody, combined with the extended power-law component being progressively covered by the absorber, until, in the deepest parts of dips, the partial covering fraction approaches unity. This approach differs radically from the ''absorbed-plus-unabsorbed'' approach previously used in spectral modeling of XB 1916-053 and similar sources, in which the normalization of the unabsorbed component is allowed to decrease markedly in dipping, behavior that is generally attributed to the effects of electron scattering. Thus, we have shown that spectral evolution in XB 1916-053 can be explained simply in terms of photoelectric absorption, without the need for substantial electron scattering. This explanation is supported by calculation of the relative importance of photoelectric absorption and electron scattering in the absorbing region, which shows that little electron scattering is expected in the ASCA energy band.

We report here the discovery of a 65 ms X-ray pulsar in the supernova remnant G11.2-0.3 with ASCA. The object is known to be a composite supernova remnant that shows both an extended shell component and a compact plerionic component. We have detected a sinusoidal pulsation with an amplitude of similar to 19% in the hard X-ray band (3-10 keV). The object has been proposed to be the historical remnant of the supernova of A.D. 386, which makes it the second candidate for the secure association of a pulsar and a historical supernova after the Crab Nebula and its pulsar.

We describe the ASCA observation of the black hole candidate Cyg X-l in the soft state of 1996 May and report the mass estimation of the central object from X-ray spectral analysis. The energy spectrum in 0.7-10 keV can be described by the sum of a power law (Gamma = 2.3 +/- 0.1) and an ultrasoft component; the latter is most probably the emission from an optically thick accretion disk. Through the spectral analysis of the ultrasoft component, we estimated the size of the accretion disk, which is closely related to the Schwarzschild radius and, hence, to the mass of the central object. Taking the general relativistic effects into account, the most probable mass of the compact object is estimated as 12(-1)(+3) M.. This value is independent of, yet consistent with, the mass estimate based on the orbital kinematics information.

We report the detection using Advanced Satellite for Cosmology and Astrophysics (ASCA) data of a hard X-ray diffuse emission around the gamma-ray pulsar PSR 1055-52. The pulsar is a middle-aged radio pulsar (P = 0.197 s, tau = P/(P) over dot approximate to 5.3 x 10(5) yr) with a pulse profile similar to that of the Crab pulsar and is one of the EGRET gamma-ray pulsars. In the soft X-ray band, ROSAT detected the pulsar as a point source indicating radiation from the neutron star. The ASCA hard band image (2-8 keV), however, showed that the diffuse emission around the pulsar dominates the pulsar itself. Morphology suggests a ring around the pulsar with a cavity in between, although the source is clumpy. The size of the suggested ring is 20' (5.8 D-kpc pc, where D-kpc denotes the distance in kiloparsecs) in diameter. The luminosity of the diffuse source is approximate to 2.0 x 10(32) D-kpc ergs s(-1), which corresponds to 6.8 x 10(-3) of the rotation power of the pulsar. Spectral analysis has been performed for the two brightest clumps. The spectrum can be fitted by a Raymond-Smith thermal plasma model with temperatures of 9 keV and 3 keV, respectively, for the two clumps and by a power-law model with photon indices of 1.7 and 2.2. Since the thermal model gives a very high pressure, we prefer the power model. We suggest that the discovered X-ray emission, together with morphology is the synchrotron nebula excited by the pulsar wind of PSR 1055 - 52.

In this Letter, we present the Advanced Satellite for Cosmology and Astrophysics data from EXO 0748 - 676 that indicate that a previously unnoted soft X-ray excess exists in the spectrum of this source. We have modeled the intensity-selected spectra with typical candidate low-mass X-ray binary spectral models. As expected, the best performing model was the generalized thermal (GT) model and an additional component that is thought to arise from the obscuration of the X-ray radiator by dipping material within the system. The GT model, however, has an unacceptable value of chi(nu)(2) (=3.16). We find that a previously undetected soft component exists (0.4-1.5 keV) in all intensity-selected data, which is the cause of the large value of chi(nu)(2) in our GT model fits. We modeled the new soft excess by adding an additional component to the GT model. A variety of soft component models were examined, and the best fitting was a steep power law with one or two local components (either edges or Gaussian lines), with the number of local components needed being dependent on the intensity-selected spectrum being fit. An edge feature and power-law model best fitted the quiescent soft spectral component, while Gaussian-line and power-law models best fitted the soft spectral component in all dipping spectra. We evaluate these results in terms of the expected physical geometry of this system.

We report the first clear detection of X-ray pulsation of magnetospheric origin from a millisecond pulsar, PSR B1821-24, with the Advanced Satellite for Cosmology and Astrophysics. The photon arrival time shows a periodicity of 3.05 ms period, as observed in the radio band. The observed X-ray pulse is double peaked. The pulses are characterized by a sharp temporal profile and hard power-law spectrum with a phase-averaged photon index of similar to 1.9. These two features are quite similar to the X-ray/y-ray pulses from the Crab pulsar and characteristic of the nonthermal emission from the magnetosphere of the neutron star, but contradictory to the thermal emission from its surface. Since this pulsar has physical parameters completely different from those of young pulsars, we expect these observations to provide significant constraints on models of pulsar magnetospheric emission, which have been developed mostly based on the observations of young pulsars.

Recent studies with ASCA have shown very complicated, strange iron K features in the spectra or galactic jet systems. SS 433, the "classic" jet, was found to have pairs of Doppler-shifted lines, contrary to the previous belief that the receding X-ray jet is short and hidden behind the accretion disk, The transient jets, GRS 1915+105 and GRO J1655-40, show spectral dips, which have never been observed in ally other source and are interpreted as absorption lines or Doppler-shifted absorption edges. K they are resonant absorption lines of helium-like iron, they would be the evidence of highly ionized, anisotropically distributed plasma near the jet engine. These features peculiar to galactic jet systems are expected to be explained in terms of the nature of the sources and the jet-formation mechanisms. Since ASCA was proved to be an excellent tool for diagnostics of jets, observation campaigns of the jet systems were planned and performed. SS 433 was observed about thirty times in the three years of Che campaign, covering the phase space of the 162.5-day precession and the 13-day orbital motion. The extracted physics of the system, such as X-ray-jet length ten times longer than previous estimations, jet kinetic luminosity exceeding 10(40) erg s(-1), etc., draw a highly energetic and stormy, new picture of SS 433. The transient jets, GRS 1915+105 and GRO J1655-40, were also observed several times. GRS I915+105 was found to be active: in ASCA band even months after onsets of outburst. Violent variations were not seen. GRO J1655-40 was observed to be transit between high and low states, and the low state is consistent to occultation of a component, We review ASCA Observations of galactic jet systems and present some topics from recent progresses.

We report on the variability of the iron K emission line in the Seyfert 1 galaxy MCG-6-30-15 during a four-day ASCA observation. The line consists of a narrow core at an energy of about 6.4 keV, and a broad red wing extending to below 5 keV, which are interpreted as line emission arising from the inner parts of an accretion disc. The narrow core correlates well with the continuum flux whereas the broad wing weakly anticorrelates. When the source is brightest, the line is dominated by the narrow core, whilst, during a deep minimum, the narrow core is very weak and a huge red tail appears. However, at other times when the continuum shows rather rapid changes, the broad wing is more variable than the narrow core, and shows evidence for correlated changes contrary to its long-time-scale behaviour. The peculiar line profile during the deep minimum spectrum suggests that the line-emitting region is very close to a central spinning (Kerr) black hole where enormous gravitational effects operate.

We observed the Rapid Burster in quiescence and detected a significant X-ray flux for the first time. Neither a type-I nor type-II burst was observed during the observation. The luminosity was estimated to be (3(-1)(+2)) x 10(33) erg s(-1) (2-10 keV), assuming a power-law with a photon index of 2 and a distance of 10 kpc. This luminosity is comparable to that of neutron-star transients in quiescence. If mass accretion is responsible for the X-ray emission, we can constrain the magnetic-field strength and spin period of the neutron star. A highly magnetized, rapidly rotating neutron star may be excluded.

We observed soft X-ray transients; X1608-52 and Cen X-4, in quiescence with ASCA at X-ray luminosities on the order of 10(32-33) erg s(-1). The energy spectra were rather soft for both sources. Blackbody fits to the data required characteristic temperatures of 0.2-0.3 keV and emission regions of similar to 10 km(2). A conspicuous hard tail remained after fitting Cen X-4 with a blackbody curve. An extremely soft spectrum in quiescence seems to be a common property of soft X-ray transients.

ACTIVE galactic nuclei and quasars are probably powered by the accretion of gas onto a supermassive black hole at the centre of the host galaxy(1), but direct confirmation of the presence of a black hole is hard to obtain. As the gas nears the event horizon, its velocity should approach the speed of light; the resulting relativistic effects, and a gravitational redshift arising from the proximity to the black hole, should be observable, allowing us to test specific predictions of the models with the observations. Here we report the detection of these relativistic effects in an X-ray emission line (the K alpha line) from ionized iron in the galaxy MCG-6-30-15. The line is extremely broad, corresponding to a velocity of similar to 100,000 km s(-1), and asymmetric, with most of the line flux being redshifted, These features indicate that the line mast probably arises in a region between three and ten Schwarzschild radii from the centre, so that we are observing the innermost region of the accretion disk.

Two X-ray bursts from GX 13 + 1 were detected with the Large Area Counters on board Ginga in 1989 September. They show clear cooling during decay, and are therefore most likely type-I X-ray bursts, i.e., the result of thermonuclear flashes. The bursts have a relatively large gamma-value (similar to 0.3; ratio of the persistent flux to the net peak flux) and short durations (similar to 15 s) which is typical of bursts from bright sources. GX 13 + 1 is probably a burst source close to the critical luminosity, above which burst activity ceases, like 1820-30 and GX 3 + 1.

The dipping low-mass X-ray binary X1916-05 was observed with the large area proportional counters (LAG) on board Ginga. Although X-ray dips appeared every 50.00 min on the average, it was found that the recurrence period slightly varied on a time scale of 6 d. In association with the period change, the duration and depth of dips also varied on the same time scale. These time variations suggest dip properties with a period of 5-6 d. The energy spectrum during the dip was also investigated, and was found to be generally reproduced by a model spectrum comprising two components having the same intrinsic spectrum, but different regarding the degree of X-ray absorption from each other: one free and the other suffering from absorption. Based on a timing analysis, it is suggested that such a spectrum originates from a partial covering of the X-ray emission region by absorbing matter. Assuming that the absorbing matter is in Keplerian motion at the outer edge of the accretion disk, the sire of the X-ray emission region was derived to be (1.2+/-0.9) x 10(8) cm, 100-times larger than the radius of a neutron star, based on the mean value, 3.6+/-2.5 s, of the distribution of the ingress/egress time of the dips. This result suggests that the optically thin hot plasma around the neutron star extends out to 10(7)-10(8) cm distance from the center.

A bright X-ray nova, GS 1124-68 (Nova Muscae 1991, GRS 1124-68), which later turned out to be a promising black-hole binary with evidence of large mass, was observed with the Large Area Counters onboard Ginga from its discovery in 1991 January to its quiescent state in 1991 September. The X-ray intensity reached a maximum (similar to 8 Crab) eight days after its discovery, then decreased exponentially with an e-folding time of about similar to 30 d. Temporary increases in the luminosity were found similar to 80 d and similar to 200 d after the outburst. The source was last detected at similar to 2 mCrab on the 239th day after the outburst; although subsequent observations were carried out on days 260 and 282, the source was below the detection limit (similar to 0.3 mCrab). GS 1124-68 exhibited such a drastic state transition that was very reminiscent of the high-low transition of black-hole candidates Cyg X-l and GX 339-4, between the 131st day (May 18) and 157th day (June 13) after the outburst. Before the transition, the energy spectrum was characterized by a thermal soft component which was dominant below similar to 8 keV, and a power-law like hard-tail component, which was significant above that energy. After the transition, the energy spectrum became much harder. The spectrum was approximately represented by a single power-law function with a photon index of similar to 1.7. Intense short time variations, which were quite similar to those in the low state of Cyg X-l, GX 339-4 and GS 2023+338, became prominent down to a timescale of milliseconds. The outburst mechanism, the origin of the soft component and the state transition can be favorably explained in terms of an optically thick accretion-disk model around a similar to 6M. black-hole at a distance of similar to 2.5 kpc.

We investigated spectral variations of the Z source GX 5-1 from the high-intensity end of the normal branch (NB) through the flaring branch (FB). We detected a clear hard-energy tail, which can be approximated by a power-law with a photon index of 1.8 and its intensity decreased by an order of magnitude from the NB to the FB. Therefore the hard tail may be intrinsic to the source. The energy spectrum in the FB shows a significant deviation from the continuous spectra of models: i.e., the sum of a blackbody and an unsaturated Comptonization spectrum, or a sum of a blackbody and multi-color disk model. The deviation can be approximated by a broad Gaussian profile (FWHM approximately 3 keV) centered at around 10 keV. The X-ray light curve in the FB is characterized by a dipping behavior on time scales of a few tens of seconds. The cross-correlation coefficient of the time variations in various energy bands against those in the 1.7-4.0 keV shows a peak structure centered at around 11 keV. This means that the Gaussian-like structure is not an artifact due to a particular selection of the continuum model. We discuss the origin of the hard-energy tail and consider several possible origins of the Gaussian component in the energy spectrum.

We have made simultaneous Ginga LAC and ROSA T PSPC observations of 4U 1820-30. The 685-s orbital light curves obtained with the two instruments are very similar, indicating that the energy dependence of the orbital modulation is small. Our measurements extend the baseline over which the period variations can be measured to 15 yr. The previous possibility, that the changes in the period are themselves periodic, with a period of about 8 yr, is no longer preferred over a constant P. Over the interval 1976-91 the period has decreased, rather than increasing as predicted by the standard model for the orbital evolution of the binary. The average period derivative P/P was (-0.88 +/- 0.16) x 10(-7) yr-1, different by 11sigma from the predicted value. Under the assumption that there are no intrinsic changes in the light curve that mimic a period change, we discuss three possible explanations. The possibility that the observed P is due to spin-orbit coupling (with a companion that is out of corotation due to stellar radius changes) is rejected, as there is no known mechanism that could cause these radius changes. The possibility of acceleration of the system by the gravitational potential of the cluster is reinvestigated in the light of a new detailed model of the mass distribution of NGC 6624. We conclude that it is unlikely that acceleration by the cluster can fully explain the observed P. Acceleration by a distant triple companion or in a chance encounter with another star in the cluster remains a possibility. Finally, we investigate the possibility that the companion is a helium-burning star. This could explain the observed P, but the likelihood of this scenario depends strongly on unknown aspects of the stellar population of NGC 6624.

The bright Galactic bulge source GX 3+1 (4U 1744-26) was observed with the Ginga satellite on several occasions from 1987 through 1990. The source has been found to show flux variations of a factor of 2 on the time scales of years. The hardness ratio, which is a measure of the spect;al shape, stayed almost constant during the variations. Two type-I X-ray bursts were detected near to the lower end of the hu( variations. When the variabilities were investigated on a shorter time scale of days, upper and lower banana branches were found to clearly appear in the hardness-intensity diagram. The branches become smeared out if data are accumulated for much longer than a few days in the hardness-intensity diagram. As. indicated by the presence of a banana branch, the short-term variations are correlated variations between the intensity and the hardness and, hence, their nature is very different from that of the long-term variations. This means that the long- and short-term variations have different origins. If the burst detection with Hakucho and Ginga in a low state reflects a red change in the mass accretion rate, the long-term variations may be determined by changes in the mass accretion rate. This would lead to the conclusion that the change in the mass accretion rate may not be responsible for the formation of the banana branches. We also discuss another possible interpretation that the mass accretion rate determines the branches, not the long-term variations.

We present the results of coordinated, partly simultaneous, optical and X-ray (Ginga) observations of the low-mass X-ray binary Sco X-1. We find that the division between the optically bright and faint state, at a blue magnitude B = 12.8, corresponds to the change from the normal to the flaring branch in the X-ray colour-colour diagram as proposed by Priedhorsky et al. (1986). From archival Walraven data we find that in both optical states the orbital light curve is approximately sinusoidal, and have a similar amplitudes.

Scorpius X-1 was observed with the Japanese X-ray satellite Ginga on 1989 March 9-11 as part of a multiwavelength campaign. The source was observed to be on the flaring branch for the first 31 hr, on the lower normal branch for the next 7 hr, and on the flaring branch during a brief observation 17 hr after the final normal branch observation. We constructed a standard color-color, or Z-, diagram for Sco X-1. We then parameterized the state of Sco X-1 by determining s(Z), the distance along the Z-diagram from the flaring branch-normal branch vertex, as a function of time. Sco X-1 moved smoothly and continuously through the Z-diagram and was never observed to jump from one branch to another. All observed spectral characteristics (intensity, X-ray hardness ratios) are well correlated with s(Z), indicating that the source changes its physical state smoothly as it moves along the Z-diagram. The tight correlation of observables with s(Z) is consistent with variation of a single physical parameter being responsible for the changes in state. There is no correlation between s(Z), the rate of change of s(Z), and observed spectral characteristics indicating that there is little, if any, hysteresis or memory in the spectral state of Sco X-1 and that motion in the Z-diagram is essentially stochastic. The range of observed values of s(Z) depends on s(Z):s(Z) always changes slowly along the flaring branch when it is near the flaring branch-normal branch vertex, but can change rapidly during the excursions up and down the flaring branch which correspond to large flares. A self-similarity is observed in the temporal profile of flares which is independent of flare amplitude. The temporal characteristics observed-including quasi-periodic oscillations, high-frequency noise, and very low frequency noise-are consistent with previous observations of Sco X-1. Quasi-periodic oscillations are observed on both the normal branch (6.3 Hz mean frequency) and the lower flaring branch (14.4 Hz mean frequency), but not on the upper flaring branch. Limits are placed on the fractional rms variation of quasi-periodic oscillations on the upper flaring branch of less than or similar 2% of the total intensity. The characteristics of the observed red noise components are found to vary along the flaring branch. The fractional rms variation of very low frequency noise increases from less than or similar 2% to greater than or similar to 6% as Sco X-1 moves from the flaring branch-normal branch vertex to the upper end of the flaring branch; the power-law index for very low frequency noise, alpha, remains constant along the flaring branch (alpha = 1.72 +/- 0.01). Both the rms fractional variation and the width of the high-frequency noise component vary as Sco X-1 moves out the flaring branch, with the rms fractional variation decreasing from approximately 3% to less than or similar 1% and the Gaussian rms width decreasing from approximately 15 Hz to approximately 5 Hz. No dependence on s(Z) is observed for any of the temporal characteristics. A search for coherent pulsations was made in both high time resolution optical and X-ray data. Limits on the 95% confidence limit pulsed amplitude are set at 0.10% (optical) and 0.19% (X-ray) for nu < 50 Hz and 0.26% (X-ray) for nu < 400 Hz. A search was made for optical quasi-periodic oscillations in data obtained while normal branch oscillations were observed in the Ginga data; none were seen with an upper limit to the fractional rms variation of 0.23% of the total intensity.

During approximately 6 hr of Ginga observations of the Rapid Burster in 1988 August, the source emitted type II bursts with durations between approximately 3 and approximately 35 s and with peak fluxes which varied by factors of approximately 5. The Rapid Burster was active in two bursting modes as defined by Marshall et al.: mode I characterized by a burst energy distribution which is 'double-peaked' and mode II by a distribution which is 'single-peaked'. We report a study of the relationship between the maximum bolometric burst flux, F(max), and the associated blackbody temperature, T(c). We find that mode I bursts consistently show higher T(c) values than bursts of mode II. Our results, together with previous work, suggest that the relation between F(max) and T(c) is dependent on mode. In mode I the colour temperature is approximately independent of the burst peak flux, whereas in mode II the two properties are correlated.

We report on a detailed study of the time variability in the Z source GX 5-1 using Ginga. The source was observed in the horizontal branch (HB) and in the normal branch (NB). Our five-day observations provide the most extensive continuous data set obtained to date. Intensity-dependent high-frequency quasi-periodic oscillations (QPO) (approximately 13-48 Hz) were observed in the HB in which the source flux varied by a factor of 1.9. At the low-intensity end of the hardness-intensity diagram, the HB shows an upward bend which is very strong in the colour-colour diagram (CCD). This bend forms essentially a new spectral branch, giving the CCD the shape of an 'S' since there is no flaring branch in GX 5-1. The QPO frequency in this portion of the HB ranges from approximately 13 to 17 Hz which is lower than previously observed in any Z source. It is possible that other Z sources will also exhibit the sharp upward turn of the HB in the CCD if their intensities (and thus the QPO frequencies) become sufficiently low. The upward turn of the HB indicates that our understanding of the spectral dependence on the mass accretion rate is rather poor. High-frequency (approximately 48 Hz) and low-frequency (approximately 6 Hz) QPO were simultaneously observed in the NB. The strength of both forms of QPO and the strength of the low-frequency noise increase rapidly with increasing photon energy. We discuss our results in connection with existing models.

We detected an X-ray outburst in the Vela-Puppis region with the All Sky Monitor (ASM) aboard Ginga on 1990 November 22. Following observations by the Large Area proportional Counter (LAC) revealed that this outburst consisted of two transient sources, GS0834-430 and GS0836-429, accidentally in the same field of view of the LAC by a separation of only 0.4 degree. These two transient X-ray sources were subsequentially monitored with the LAC until 1991 May. GS 0834-430 was identified by an X-ray pulsar with a period of 12.s3. This pulsar shows a power-law energy spectrum with a photon index of approximately 0.9, and with a high-energy cutoff that is typical of X-ray pulsars. GS 0836-429 was identified by an X-ray burster, which shows typical type-I bursts approximately every 2 h.

GX 339-4 was observed with the large area counters (LAC) onboard Ginga in its very high state, where the X-ray intensity was about a factor of 2-3 larger than its high state and it showed very rapid variations on time scales of less than several minutes, which had not been observed earlier in the high state of this source. The X-ray energy spectrum was very soft; it consisted of a low-energy component and a high-energy tail. The low-energy component could be interpreted as being due to disk blackbody radiation (the disk blackbody component) and the high-energy tail as being due to Compton-scattering radiation (the Comptonized blackbody component). The X-ray energy spectrum also showed K-edge absorption of highly ionized iron atoms of about 10(19) cm-2 and an iron emission line with an equivalent width of about 60-100 eV. On short time scales of less than several minutes, the X-rays showed rapid time variations. For instance, in power spectral density functions, 6 Hz quasi-periodic oscillations (QPOs), very low frequency noise (VLF noise), and low-frequency noise (LF noise) were recognized. There are three types of power spectral density functions. Time variations such as dips and flip-flops were also observed. These rapid time variations are due to the Comptonized blackbody component. On long time scales larger than hours, the disk blackbody component and the Comptonized blackbody component changed independently. However, these changes took place within some restricted regions in an X-ray hardness ratio (color)-intensity or a color-color diagrams. These are a hardness ratio increasing (with the X-ray flux) branch, a hardness ratio decreasing branch, and their crossing region. These two energy spectral branches and their crossing region have three different types of power density spectra: these different branches and regions have different time variations on time scales of less than several minutes. In the hardness ratio increasing branch, the hard Compton-scattering component is the main cause of the long-term time variation of the X-ray flux, and in the hardness ratio decreasing branch, the disk blackbody component is the main cause of the long-term time variation of the X-ray flux. On a time lag versus Fourier period diagram, the time lag between time variations of different energy X-rays showed shoulder structures in addition to the large time lags at large Fourier periods similar to those observed in Cygnus X-1. The time variations of the X-rays of 2.3-4.6 keV were most advanced: time variations of X-rays with both lower and higher energy than 2.3-4.6 keV showed time lags relative to those of 2.3-4.6 keV X-rays. These facts together with the X-ray energy spectrum can be explained in terms of a high-energy component being due to a Compton-scattering cloud of size of about 10(9) cm, temperature kT(e) of about 30 keV, and Thomson scattering depth of about 0.5-1.0; the variable blackbody radiation with kT = approximately 1 keV is incident to the large Compton-scattering cloud.

The Hard X-ray Telescope (HXT) is a Fourier-synthesis imager; a set of spatially-modulated photon count data are taken from 64 independent subcollimators and are Fourier-transformed into an image by using procedures such as the maximum entropy method (MEM) or CLEAN. The HXT takes images of solar flares simultaneously in four energy bands, nominally 15 (or 19)-24, 24-35, 35-57, and 57-100 keV, with an ultimate angular resolution as fine as ∼ 5 arc sec and a time resolution 0.5 s. Each subcollimator has a field of view wider than the solar disk. The total effective area of the collimator/detector system reaches ∼ 70 cm2, about one order of magnitude larger than that of the HINOTORI hard X-ray imager. Thanks to these improvements, HXT will for the first time enable us to take images of flares at photon energies above ∼ 30 keV. These higher-energy images will be compared with lower-energy ones, giving clues to the understanding of nonthermal processes in solar flares, i.e., the acceleration and confinement of energetic electrons. It is of particular importance to specify the acceleration site with regard to the magnetic field figuration in a flaring region, which will be achieved by collaborative observations between HXT and the Soft X-ray Telescope on board the same mission. © 1991 Kluwer Academic Publishers.

An outburst of the transient X-ray pulsar X0115+634 was detected with the All Sky Monitor (ASM) on board Ginga on 1990 February 5. Follow-up observations with the large-area proportional counters (LACs) revealed complex changes in the energy spectrum which depend on the phase of the 3.6 s pulsation. We find that characteristic structures in the spectra above 10 keV can be best interpreted as two dips at ∼12 and ∼23 keV, although not at all phases. The center energies of the two dips are consistent with the harmonic relation of 1:2, showing phase-dependent ±10% variations with pulse phase. The results strongly suggest that the structures in the spectra are due to cyclotron resonant scattering and the two apparent absorption lines are ascribed to the fundamental and second harmonics. This indicates a magnetic field strength on the neutron star surface of ∼1 × 1012 G. Equivalent widths of the second harmonic line are about 2 times larger than those of the first harmonic line, depending on the pulse phase.

Observations made in 1989 February and April with the Ginga satellite show that the times of maximum light for the 685 s X-ray intensity variations of 4U 1820-30 occur 71 +/- 21 and 94 +/- 21 s earlier than would be expected for the constant-period ephemeris based on previous observations. The times of maximum light observed between 1976 and 1989 are consistent with a constant rate period decrease of 0.074 +/- 0.013 ms yr-1 (P/P = -1.08 +/- 0.19 10(-7) yr-1). We find the decrease is significant at the 99.9% level. If this reflected a true change in the orbital period, this result would be inconsistent with the standard model in which the 685 s period is that of the orbit of a binary star consisting of a neutron star whose companion is a low-mass (approximately 0.07 M.) degenerate helium star, and mass transfer is driven by loss of orbital angular momentum through gravitational radiation. The apparent period change may be explained in two ways: either the standard model is incorrect, or 4U 1820-30 is being accelerated toward us by the gravitational field of another body. Acceleration of the binary by a distant third companion in a hierarchical triple, or by the cluster potential, are both possibilities.

Ginga observations of quasi-periodic oscillations (QPO) and the low-frequency noise (LFN) from GX 5-1 in its horizontal-branch spectral state are presented. Power spectral fits were attempted using model functions based on simple oscillating shot models. In order to obtain acceptable fits, separate life distribution functions are necessary in the shot models representing QPO and LFN. Furthermore, a larger amplitude of shots is necessary for QPO than for LFN for the longer lifetime shots. A clear second-harmonic peak of QPO was detected. Variations in the powers of QPO and LFN on timescales of 8-256 were also studied. These variations were significant for all of the timescales studied, and were uncorrelated with each other on timescales shorter than a few tens of seconds, and correlated on longer timescales. From simulations based on a simple shot model, it was found that the variation amplitude and the lack of correlation on short timescales are not inconsistent with the oscillating shot models. A more complex model is necessary to fully explain the observed properties.

The Hard X-ray Telescope (HXT), now under fabrication for the SOLAR-A mission (scheduled for launch in August 1991), is an advanced Fourier-synthesis imager. An overview of the HXT instrument is given together with its scientific objectives, that is, the electron acceleration and confinement mechanisms in solar flares. Scientific return from HXT will be greatly increased if worldwide collaboration with other space and ground-based observations is well organized. © 1991.

We report on a Ginga observation of the globular-cluster source 4U 2129+11 in M15, during which we detected a very energetic X-ray burst. This burst showed a precursor, separated from the main event by approximately 6 s, which indicates that strong expansion of the neutron star photosphere occurred. During the first 30 s of the main part of the burst a series of slow radius oscillations were observed, whose termination may signify the transition from an outflowing neutron star wind to a static extended (but very slowly shrinking) envelope. From a time-resolved spectral analysis of the burst data we find evidence for deviations of the burst spectra from a blackbody, which cannot be described in terms of a spectral hardening alone. We show that, in principle, the results of the analysis may be used to derive information on the mass and radius of the neutron star, as well as on the anisotropy of the burst emission (as long as this remains the same during the burst). The applicability of this method is presently limited by uncertainties in the interpretation of the burst spectra.

X-ray pulsars1,2 are magnetized, spinning neutron stars accreting matter from their binary companions. Their pulse periods P, ranging over four orders of magnitude, increase and decrease in complex ways1,3,4. The more luminous ones tend to show faster spin-up1,5. A puzzle is that the spin-up timescales of many X-ray pulsars are much shorter than their binary-evolution timescales, thus apparently violating the steady-state condition. It has there-fore been suspected6 that there exist many 'turned-off X-ray pulsars currently spinning down undetected. An excellent test for this hypothesis became available using the X-ray pulsar GX1 +4, which used to show the fastest spin-up over a decade1,7-10and then faded away11. Using the X-ray satellite Ginga12, we detected GX1+4 at ∼1/40 the previous intensity, and found that it now has an average spin-down trend. This discovery apparently supports the above hypothesis. © 1988 Nature Publishing Group.