田代 信

Radio and optical observations of the possible microlensed BL Lac source MS 0205.7+3509 were obtained simultaneously with ASCA x-ray measurements in February 1997. A single power law model, with a photon index of 2.61, is an adequate fit to the ASCA data, once hydrogenic absorption in excess of the Galactic value is permitted, confirming a previous ROSAT measurement. On the basis of our simultaneous data we have determined MS 0205.7+3509 to be a typical x-ray selected BL Lac, with alpha(xox) = -0.82. There is no indication of an inverse Compton (IC) component in the ASCA spectrum up to 10 keV. No evidence for variability on hour-long timescales is present in either the x-ray or the optical data. We discuss these results in the context of a gravitational microlensing scenario for MS 0205.7+3509.

Radio and optical observations of the possible microlensed BL Lac source MS 0205.7+3509 were obtained simultaneously with ASCA x-ray measurements in February 1997. A single power law model, with a photon index of 2.61, is an adequate fit to the ASCA data, once hydrogenic absorption in excess of the Galactic value is permitted, confirming a previous ROSAT measurement. On the basis of our simultaneous data we have determined MS 0205.7+3509 to be a typical x-ray selected BL Lac, with alpha(xox) = -0.82. There is no indication of an inverse Compton (IC) component in the ASCA spectrum up to 10 keV. No evidence for variability on hour-long timescales is present in either the x-ray or the optical data. We discuss these results in the context of a gravitational microlensing scenario for MS 0205.7+3509.

ASTRO-E Hard X-ray Detector (HXD) is characterized by well-type phoswich counters[1] in a compound-eye configuration which reduce the detector background to lower level than any other past hard X-ray mission. When operating in Low Earth Orbit, the expected background of the HXD is an order of 10 -5 counts/s/kcV/cm 2, mainly caused from the radioactivity induced within the detector materials by geomagnetically trapped protons. Results are presented from measurements of induced radioactivity in two phoswich scintillators, GSO (Gd 2SiO 5:Ce 0.5% mol) and BGO (Bi 4Ge 3O 12), irradiated by mono-energetic protons at an accelerator facility. Radiation transport computer codes are used to build the detector response functions for emissions from decays of spallation products. Based on the comparison between experimental and simulation results, the activation background level of HXD in the orbit is estimated. © 1999 IEEE.

ASTRO-E Hard X-ray Detector (HXD) is characterized by well-type phoswich counters[1] in a compound-eye configuration which reduce the detector background to lower level than any other past hard X-ray mission. When operating in Low Earth Orbit, the expected background of the HXD is an order of 10 -5 counts/s/kcV/cm 2, mainly caused from the radioactivity induced within the detector materials by geomagnetically trapped protons. Results are presented from measurements of induced radioactivity in two phoswich scintillators, GSO (Gd 2SiO 5:Ce 0.5% mol) and BGO (Bi 4Ge 3O 12), irradiated by mono-energetic protons at an accelerator facility. Radiation transport computer codes are used to build the detector response functions for emissions from decays of spallation products. Based on the comparison between experimental and simulation results, the activation background level of HXD in the orbit is estimated. © 1999 IEEE.

We present data for 18 blazars observed with the X-ray satellite ASCA, half of which were also observed contemporaneously with the EGRET instrument onboard Compton Gamma Ray Observatory as parts of multiwavelength campaigns. The observations show a clear difference in the spectra between three subclasses of blazars, namely, high-energy peaked BL Lacertae objects (HBLs), low-energy peaked BL Lac objects (LBLs), and quasar-hosted blazars (QHBs). The ASCA X-ray spectra of HBLs are the softest, with the power-law energy index alpha similar to 1-2, and they form the highest observable energy tail of the low-energy (LE, synchrotron) component. The X-ray spectra of the QHBs are the hardest (alpha similar to 0.6) and are consistent with the lowest observable energy end of the high-energy (HE, Compton) component. For LBLs, the X-ray spectra are intermediate. We find that the radiation process responsible for the HE peak for HBLs can be explained solely by Doppler-boosted synchrotron self-Compton (SSC) emission, with the Doppler factor delta consistent with the VLBI and/or gamma-ray variability data. For many QHBs, on the other hand, the gamma-rays cannot be solely a result of the SSC mechanism unless delta is significantly in excess of that inferred from VLBI data. We consider an alternative scenario consistent with the measured values of delta, where the SSC component is still present in QHBs and dominates in the X-ray band but is below the observed gamma-ray spectrum. With an assumption that the peak of the SSC emission is on the extrapolation of the X-ray spectrum, and adopting a value of 10 for delta, we infer the magnetic field B to be 0.1-1 G and Lorentz factors gamma(b) of electrons radiating at the peak of the nu F(nu) spectrum of similar to 10(3) for QHBs; this is much lower than gamma(b) similar to 10(5) for HBLs, even though the Values of B are comparable in the two subclasses. This difference of gamma(b) is most likely due to the large photon density expected in QHBs (e.g., from thermal components visible in these objects) as compared with that of HBLs; Compton upscattering of these photons may well provide the observed GeV flux.

We present data for 18 blazars observed with the X-ray satellite ASCA, half of which were also observed contemporaneously with the EGRET instrument onboard Compton Gamma Ray Observatory as parts of multiwavelength campaigns. The observations show a clear difference in the spectra between three subclasses of blazars, namely, high-energy peaked BL Lacertae objects (HBLs), low-energy peaked BL Lac objects (LBLs), and quasar-hosted blazars (QHBs). The ASCA X-ray spectra of HBLs are the softest, with the power-law energy index alpha similar to 1-2, and they form the highest observable energy tail of the low-energy (LE, synchrotron) component. The X-ray spectra of the QHBs are the hardest (alpha similar to 0.6) and are consistent with the lowest observable energy end of the high-energy (HE, Compton) component. For LBLs, the X-ray spectra are intermediate. We find that the radiation process responsible for the HE peak for HBLs can be explained solely by Doppler-boosted synchrotron self-Compton (SSC) emission, with the Doppler factor delta consistent with the VLBI and/or gamma-ray variability data. For many QHBs, on the other hand, the gamma-rays cannot be solely a result of the SSC mechanism unless delta is significantly in excess of that inferred from VLBI data. We consider an alternative scenario consistent with the measured values of delta, where the SSC component is still present in QHBs and dominates in the X-ray band but is below the observed gamma-ray spectrum. With an assumption that the peak of the SSC emission is on the extrapolation of the X-ray spectrum, and adopting a value of 10 for delta, we infer the magnetic field B to be 0.1-1 G and Lorentz factors gamma(b) of electrons radiating at the peak of the nu F(nu) spectrum of similar to 10(3) for QHBs; this is much lower than gamma(b) similar to 10(5) for HBLs, even though the Values of B are comparable in the two subclasses. This difference of gamma(b) is most likely due to the large photon density expected in QHBs (e.g., from thermal components visible in these objects) as compared with that of HBLs; Compton upscattering of these photons may well provide the observed GeV flux.

We present ASCA X-ray results on two SO galaxies, NGC 3065 and NGC 4203. In both galaxies, we detected hard X-ray emission from a pointlike source at the nucleus. A single power-law model having a photon index of similar to 1.8 well described the spectra of these sources, while thin thermal emission, which is common in SO galaxies, was not detected. The 2-10 keV luminosities of these nuclear sources (2.2 x 10(41) and 1.5 x 10(40) ergs s(-1)) are 1-2 orders of magnitude higher than those expected as an assembly of low-mass X-ray binaries in these galaxies. Our results strongly suggest that NGC 3065 and NGC 4203 host low-luminosity active galactic nuclei.

NGC 1316 (Fornax A) is a radio galaxy with prototypical double lobes, when the magnetic field intensity is accurately measured via the inverse Compton technique. The radio-emitting electrons in the lobes are inferred to have a synchrotron lifetime of similar to 0.1 Gyr. Considering the lobe energetics, we estimate the past nuclear X-ray luminosity of NGC 1316 to be at least similar to 4 x 10(34) W (4 X 10(41) ergs s(-1)). Thus, the nucleus was rather active at least 0.1 Gyr ago. In contrast, we confirmed with ASCA and ROSAT that the nucleus of NGC 1316 is very faint in X-rays at present, with the 2-10 keV luminosity of any active galactic nucleus-like hard component being less than 2 x 10(33) W (2 x 10(40) ergs s(-1)), even assuming a nuclear obscuration up to similar to 10(28) m(-2) (10(24) cm(-2)). This is at least an order of magnitude lower than the estimated past activity, indicating that the nucleus is presently very inactive. From these two results, we conclude that the nucleus of NGC 1316 has become dormant during the last 0.1 Gyr. This suggests the possible abundance of "dormant" quasars in nearby galaxies.

We present ASCA X-ray results on two SO galaxies, NGC 3065 and NGC 4203. In both galaxies, we detected hard X-ray emission from a pointlike source at the nucleus. A single power-law model having a photon index of similar to 1.8 well described the spectra of these sources, while thin thermal emission, which is common in SO galaxies, was not detected. The 2-10 keV luminosities of these nuclear sources (2.2 x 10(41) and 1.5 x 10(40) ergs s(-1)) are 1-2 orders of magnitude higher than those expected as an assembly of low-mass X-ray binaries in these galaxies. Our results strongly suggest that NGC 3065 and NGC 4203 host low-luminosity active galactic nuclei.

NGC 1316 (Fornax A) is a radio galaxy with prototypical double lobes, when the magnetic field intensity is accurately measured via the inverse Compton technique. The radio-emitting electrons in the lobes are inferred to have a synchrotron lifetime of similar to 0.1 Gyr. Considering the lobe energetics, we estimate the past nuclear X-ray luminosity of NGC 1316 to be at least similar to 4 x 10(34) W (4 X 10(41) ergs s(-1)). Thus, the nucleus was rather active at least 0.1 Gyr ago. In contrast, we confirmed with ASCA and ROSAT that the nucleus of NGC 1316 is very faint in X-rays at present, with the 2-10 keV luminosity of any active galactic nucleus-like hard component being less than 2 x 10(33) W (2 x 10(40) ergs s(-1)), even assuming a nuclear obscuration up to similar to 10(28) m(-2) (10(24) cm(-2)). This is at least an order of magnitude lower than the estimated past activity, indicating that the nucleus is presently very inactive. From these two results, we conclude that the nucleus of NGC 1316 has become dormant during the last 0.1 Gyr. This suggests the possible abundance of "dormant" quasars in nearby galaxies.

We observed the radio galaxy PKS 1343-601 (Centaurus B) in X-rays with ASCA and discovered diffuse hard X-ray emission from its radio lobe region. The obtained flux and spectrum strongly suggest that the X-rays are produced via an inverse-Compton (IC) process, in which the cosmic microwave background photons are boosted up to X-rays by synchrotron electrons. This is the second case, following the case of Fornax A, of the detection of the IC X-rays from radio lobes. The observed X-ray emission is brighter than expected in the case of minimum energy density, indicating particle-dominated lobes, Comparison of the X-ray and radio images reveals an outward increase in the magnetic energy density. This infers a gradual compression of magnetic fields toward the edge of the lobes.

We observed the radio galaxy PKS 1343-601 (Centaurus B) in X-rays with ASCA and discovered diffuse hard X-ray emission from its radio lobe region. The obtained flux and spectrum strongly suggest that the X-rays are produced via an inverse-Compton (IC) process, in which the cosmic microwave background photons are boosted up to X-rays by synchrotron electrons. This is the second case, following the case of Fornax A, of the detection of the IC X-rays from radio lobes. The observed X-ray emission is brighter than expected in the case of minimum energy density, indicating particle-dominated lobes, Comparison of the X-ray and radio images reveals an outward increase in the magnetic energy density. This infers a gradual compression of magnetic fields toward the edge of the lobes.

The results obtained from ASCA observations of the cluster of galaxies Abell 3627 are presented. This cluster, located behind the Milky Way, was recently found to be a nearby, X-ray bright, and very rich cluster. Pointed observations onto the central region of the cluster gave a gas temperature of similar to 7 keV and a metallicity of about 0.2 solar. An offset pointing to a substructure elongated to the south-east of the cluster center gave a significantly lower temperature of similar to 5 keV. The 2-10 keV luminosity within a radius of 40' (1.1 Mpc) is estimated to be 3.7 x 10(44) erg s(-1). The X-ray data imply a cluster mass of about 4 x 10(14) M-. within 40'.

The results obtained from ASCA observations of the cluster of galaxies Abell 3627 are presented. This cluster, located behind the Milky Way, was recently found to be a nearby, X-ray bright, and very rich cluster. Pointed observations onto the central region of the cluster gave a gas temperature of similar to 7 keV and a metallicity of about 0.2 solar. An offset pointing to a substructure elongated to the south-east of the cluster center gave a significantly lower temperature of similar to 5 keV. The 2-10 keV luminosity within a radius of 40' (1.1 Mpc) is estimated to be 3.7 x 10(44) erg s(-1). The X-ray data imply a cluster mass of about 4 x 10(14) M-. within 40'.

Reports have been made on the X-ray properties of the BL Lacertae object OJ 287, obtained with ASCA on 1994 November 18-19. Although the observation was made during a period of enhanced optical intensity, the obtained 0.5-10 keV flux falls below those from previous X-ray observations. Intensity variations on time scales of a few hours were detected, with a typical peak-to-peak amplitude of 20%. The X-ray spectra acquired with the SIS and the GIS are described consistently with a single power-law model with the galactic absorption, and the required photon index, similar to 1.7, is flatter than most of those previously observed from this object. The derived monochromatic X-ray flux at 1 keV, 0.76 mu Jy, exceeds the extrapolation from the lower frequency continua, which is based on near-simultaneous multi-band observations. A discussion is made concerning the origin of X-ray emission from OJ 287 in terms of the synchrotron and synchrotron-self-Compton mechanisms.

Reports have been made on the X-ray properties of the BL Lacertae object OJ 287, obtained with ASCA on 1994 November 18-19. Although the observation was made during a period of enhanced optical intensity, the obtained 0.5-10 keV flux falls below those from previous X-ray observations. Intensity variations on time scales of a few hours were detected, with a typical peak-to-peak amplitude of 20%. The X-ray spectra acquired with the SIS and the GIS are described consistently with a single power-law model with the galactic absorption, and the required photon index, similar to 1.7, is flatter than most of those previously observed from this object. The derived monochromatic X-ray flux at 1 keV, 0.76 mu Jy, exceeds the extrapolation from the lower frequency continua, which is based on near-simultaneous multi-band observations. A discussion is made concerning the origin of X-ray emission from OJ 287 in terms of the synchrotron and synchrotron-self-Compton mechanisms.

In 1994 May, the BL Lac object PKS 2155-304 was observed continuously for similar to 10 days with the International Ultraviolet Explorer and the Extreme Ultraviolet Explorer and for 2 days with ASCA, as well as with ROSAT and with ground-based radio, infrared, and optical telescopes. The light curves show a well-defined X-ray flare followed by a broader, lower amplitude extreme-ultraviolet flare similar to 1 day later and a broad, low-amplitude UV flare similar to 2 days later. X-ray fluxes obtained at three well-separated times the preceding week indicate at least one previous flare of comparable amplitude or perhaps ongoing stochastic X-ray variations, and additional rapid variability was seen at the beginning of the IUE observation, when extremely sharp changes in UV flux occurred. The X-ray flux observed with ASCA flared by a factor of similar to 2 in about half a day and decayed roughly as fast. In contrast, the subsequent UV flare had an amplitude of only similar to 35% and lasted longer than 2 days. Assuming that the X-ray, EUV, and UV events are associated, the lags, the decrease of amplitude with wavelength, and the broadening of the temporal profile with wavelength are all qualitatively as expected for synchrotron emission from an inhomogeneous, relativistic jet. Because of the high quality of the data, we can rule out that the observed flares were caused by either a Fermi-type shock acceleration event or a pair cascade in a homogeneous synchrotron-emitting region. A homogeneous region is still possible if there was an instantaneous (t much less than hours) injection of high-energy electrons that emit first at X-ray energies. Alternatively, the data are consistent with a compression wave or other disturbance crossing a region with stratified particle energy distributions. This kind of situation is expected to occur behind a shock front and/or in an inhomogeneous jet. The present light curves are in sharp contrast to the multiwavelength variability observed in 1991 November, when the amplitude was wavelength independent and the UV lagged the X-rays by less than similar to 3 hr. This means that the origin of rapid multiwavelength variability in this blazar is complex, involving at least two different modes.

We present the results of an ASCA observation of the radio-quiet quasar E1821 + 643. The 2-10 keV flux and the power-law index are consistent with the values reported previously by EXOSAT, Ginga, and BBXRT. The iron K-emission line feature discovered by the Ginga observations is confirmed, and its redshift-corrected, line-center energy is better determined to be 6.58 +/- 0.05 keV. This line feature shows marginal broadening with an upper limit of 180 eV in Gaussian sigma (in the quasar rest frame). The line equivalent width is 170 +/- 50 eV in the quasar rest frame, which is consistent with the Ginga measurement, and the contribution from the quasar-associated cluster of galaxies is estimated to be negligible.

In 1994 May, the BL Lac object PKS 2155-304 was observed continuously for similar to 10 days with the International Ultraviolet Explorer and the Extreme Ultraviolet Explorer and for 2 days with ASCA, as well as with ROSAT and with ground-based radio, infrared, and optical telescopes. The light curves show a well-defined X-ray flare followed by a broader, lower amplitude extreme-ultraviolet flare similar to 1 day later and a broad, low-amplitude UV flare similar to 2 days later. X-ray fluxes obtained at three well-separated times the preceding week indicate at least one previous flare of comparable amplitude or perhaps ongoing stochastic X-ray variations, and additional rapid variability was seen at the beginning of the IUE observation, when extremely sharp changes in UV flux occurred. The X-ray flux observed with ASCA flared by a factor of similar to 2 in about half a day and decayed roughly as fast. In contrast, the subsequent UV flare had an amplitude of only similar to 35% and lasted longer than 2 days. Assuming that the X-ray, EUV, and UV events are associated, the lags, the decrease of amplitude with wavelength, and the broadening of the temporal profile with wavelength are all qualitatively as expected for synchrotron emission from an inhomogeneous, relativistic jet. Because of the high quality of the data, we can rule out that the observed flares were caused by either a Fermi-type shock acceleration event or a pair cascade in a homogeneous synchrotron-emitting region. A homogeneous region is still possible if there was an instantaneous (t much less than hours) injection of high-energy electrons that emit first at X-ray energies. Alternatively, the data are consistent with a compression wave or other disturbance crossing a region with stratified particle energy distributions. This kind of situation is expected to occur behind a shock front and/or in an inhomogeneous jet. The present light curves are in sharp contrast to the multiwavelength variability observed in 1991 November, when the amplitude was wavelength independent and the UV lagged the X-rays by less than similar to 3 hr. This means that the origin of rapid multiwavelength variability in this blazar is complex, involving at least two different modes.

We present the results of an ASCA observation of the radio-quiet quasar E1821 + 643. The 2-10 keV flux and the power-law index are consistent with the values reported previously by EXOSAT, Ginga, and BBXRT. The iron K-emission line feature discovered by the Ginga observations is confirmed, and its redshift-corrected, line-center energy is better determined to be 6.58 +/- 0.05 keV. This line feature shows marginal broadening with an upper limit of 180 eV in Gaussian sigma (in the quasar rest frame). The line equivalent width is 170 +/- 50 eV in the quasar rest frame, which is consistent with the Ginga measurement, and the contribution from the quasar-associated cluster of galaxies is estimated to be negligible.

Two spiral galaxies, NGC 1365 and NGC 1386, were observed in X-rays with ASCA. Both galaxies exhibit hard X-ray emission from point-like sources at the nuclei. The continuous spectrum of NGC 1365 was fitted well with a power-law of photon index 0.8(-0.2)(+0.7), whose 2-10 keV luminosity is (3.8+/-0.5) x 10(40) erg s(-1) for an assumed distance of 18.4 Mpc. NGC 1386 has an absorbed [N-H=(2.8(-2.6)(+2.7)) x 10(23) cm(-2)] power-law continuum, whose 2-10 keV luminosity (at 18.4 Mpc) is (2.6(-2.0)(+3.2)) x 10(40) erg s(-1) after removing the absorption. A broad (sigma=0.20(-0.08)(+0.16) keV) Fe-K emission line centered at 6.61(-0.06)(+0.08) keV was detected from the NGC 1365 nucleus at an equivalent width of 2.1(-0.3)(+2.1) keV. The Fe-K line was also detected from NGC 1386 at an equivalent width of 1.6(-1.2)(+7.9) keV and at a similar energy as the NGC 1365 line. These spectral features indicate that both galaxies host low-luminosity active galactic nuclei (LLAGNs).

Two spiral galaxies, NGC 1365 and NGC 1386, were observed in X-rays with ASCA. Both galaxies exhibit hard X-ray emission from point-like sources at the nuclei. The continuous spectrum of NGC 1365 was fitted well with a power-law of photon index 0.8(-0.2)(+0.7), whose 2-10 keV luminosity is (3.8+/-0.5) x 10(40) erg s(-1) for an assumed distance of 18.4 Mpc. NGC 1386 has an absorbed [N-H=(2.8(-2.6)(+2.7)) x 10(23) cm(-2)] power-law continuum, whose 2-10 keV luminosity (at 18.4 Mpc) is (2.6(-2.0)(+3.2)) x 10(40) erg s(-1) after removing the absorption. A broad (sigma=0.20(-0.08)(+0.16) keV) Fe-K emission line centered at 6.61(-0.06)(+0.08) keV was detected from the NGC 1365 nucleus at an equivalent width of 2.1(-0.3)(+2.1) keV. The Fe-K line was also detected from NGC 1386 at an equivalent width of 1.6(-1.2)(+7.9) keV and at a similar energy as the NGC 1365 line. These spectral features indicate that both galaxies host low-luminosity active galactic nuclei (LLAGNs).

The Hard X-ray Detector (HXD) is one of three instruments on the fifth Japanese X-ray astronomy satellite, ASTRO-E, scheduled for launch in 2000. The HXD consists of a 4x4 = 16 modular assembly of identical counters, each consisting of a combination of YAP(or GSO)/BGO well-type phoswich counters and silicon PIN diodes. The field of view of the detector for high energy photons is restricted to 4 degrees x 4 degrees by active collimators made of BGO and the fov for low energy photons is restricted to 0.5 degrees x 0.5 degrees by the fine passive collimators made of phosphor bronze. The detector is characterized by a low background reaching severalx10(-6) c/s/cm(2)/keV. Combined with the other two instruments for soft X-ray observations (0.5-12 keV), the ASTRO-E mission will cover the entire range of soft and hard X-rays with the highest sensitivity ever achieved. Furthermore, thick EGO counters which surround the 4x4 matrix of well-type phoswich counters act as a gamma-ray burst detector in the energy band of 100-2000 keV.

The Hard X-ray Detector (HXD) is one of three instruments on the fifth Japanese X-ray astronomy satellite, ASTRO-E, scheduled for launch in 2000. The HXD consists of a 4x4 = 16 modular assembly of identical counters, each consisting of a combination of YAP(or GSO)/BGO well-type phoswich counters and silicon PIN diodes. The field of view of the detector for high energy photons is restricted to 4 degrees x 4 degrees by active collimators made of BGO and the fov for low energy photons is restricted to 0.5 degrees x 0.5 degrees by the fine passive collimators made of phosphor bronze. The detector is characterized by a low background reaching severalx10(-6) c/s/cm(2)/keV. Combined with the other two instruments for soft X-ray observations (0.5-12 keV), the ASTRO-E mission will cover the entire range of soft and hard X-rays with the highest sensitivity ever achieved. Furthermore, thick EGO counters which surround the 4x4 matrix of well-type phoswich counters act as a gamma-ray burst detector in the energy band of 100-2000 keV.

We observed the BL Lac object Mrk 421 with the X-ray satellite ASCA in 1994 as part of a multifrequency observation. The 24 hr observation was conducted 1 day after the onset of a TeV flare detected by the Whipple Observatory and detected an X-ray flare, with no apparent variability in the optical, UV, and EGRET GeV flux. The ASCA 2-10 keV flux peaked at 3.7 x 10(-10) ergs cm(-2) s(-1) and then decreased to 1.8 x 10(-10) ergs cm(-2) s(-1) with a doubling timescale of similar to 12 hr. The shape of the X-ray spectrum varied during the observation, such that the hard X-rays always led the soft X-rays, both in brightening and dimming of the source, with a lag of the 0.5-1 keV photons versus those in the 2-7.5 keV band of similar to 1 hr. The rapid TeV variability indicates a compact TeV-producing region, suggesting relativistic beaming with a Doppler factor delta greater than or equal to 5. The correlation of the flux in the X-ray and the TeV bands indicates that a high-energy tail of a single electron population is responsible for both X-rays and TeV gamma-rays, with radio, IR, UV and X-rays produced via the synchrotron process and GeV and TeV gamma-rays produced via Comptonization. Under the assumption that the ''soft lag'' observed in the X-ray band is due to the synchrotron-lifetime effects, with delta = 5, we calculate the magnetic field for the X-ray-producing region to be similar to 0.2 G. The Lorentz factors gamma(el) of the electrons responsible for the emission in the keV and TeV bands are similar to 10(6), consistent with the values implied by the Klein-Nishina limit.

We observed the BL Lac object Mrk 421 with the X-ray satellite ASCA in 1994 as part of a multifrequency observation. The 24 hr observation was conducted 1 day after the onset of a TeV flare detected by the Whipple Observatory and detected an X-ray flare, with no apparent variability in the optical, UV, and EGRET GeV flux. The ASCA 2-10 keV flux peaked at 3.7 x 10(-10) ergs cm(-2) s(-1) and then decreased to 1.8 x 10(-10) ergs cm(-2) s(-1) with a doubling timescale of similar to 12 hr. The shape of the X-ray spectrum varied during the observation, such that the hard X-rays always led the soft X-rays, both in brightening and dimming of the source, with a lag of the 0.5-1 keV photons versus those in the 2-7.5 keV band of similar to 1 hr. The rapid TeV variability indicates a compact TeV-producing region, suggesting relativistic beaming with a Doppler factor delta greater than or equal to 5. The correlation of the flux in the X-ray and the TeV bands indicates that a high-energy tail of a single electron population is responsible for both X-rays and TeV gamma-rays, with radio, IR, UV and X-rays produced via the synchrotron process and GeV and TeV gamma-rays produced via Comptonization. Under the assumption that the ''soft lag'' observed in the X-ray band is due to the synchrotron-lifetime effects, with delta = 5, we calculate the magnetic field for the X-ray-producing region to be similar to 0.2 G. The Lorentz factors gamma(el) of the electrons responsible for the emission in the keV and TeV bands are similar to 10(6), consistent with the values implied by the Klein-Nishina limit.

We obtained X-ray spectra of the Seyfert 1 galaxy NGC 3516 in 1995 March using the Japanese X-ray satellite, ASCA. Simultaneous far-UV observations were obtained with the Hopkins Ultraviolet Telescope on the Astro-2 shuttle mission. The ASCA spectrum shows a lightly absorbed power law of energy index 0.78. The low-energy absorbing column is significantly less than previously seen. Prominent O VII and O VIII absorption edges are visible, but, consistent with the much lower total absorbing column, no Fe K absorption edge is detectable. A weak, narrow Fe K alpha emission line from cold material is present as well as a broad Fe K alpha line. These features are similar to those reported in other Seyfert 1 galaxies. A single warm absorber model provides only an imperfect description of the low-energy absorption. In addition to a highly ionized absorber with ionization parameter U = 1.66 and a total column density of 1.4 x 10(22) cm(-2), adding a lower ionization absorber with U = 0.32 and a total column of 6.9 x 10(21) cm(-2) significantly improves the fit. The contribution of resonant line scattering to our warm absorber models limits the Doppler parameter to less than 160 km s(-1) at 90% confidence. Turbulence at the sound speed of the photoionized gas provides the best fit. None of the warm absorber models fitted to the X-ray spectrum can match the observed equivalent widths of all the UV absorption lines. Accounting for the X-ray and UV absorption simultaneously requires an absorbing region with a broad range of ionization parameters and column densities.

We obtained X-ray spectra of the Seyfert 1 galaxy NGC 3516 in 1995 March using the Japanese X-ray satellite, ASCA. Simultaneous far-UV observations were obtained with the Hopkins Ultraviolet Telescope on the Astro-2 shuttle mission. The ASCA spectrum shows a lightly absorbed power law of energy index 0.78. The low-energy absorbing column is significantly less than previously seen. Prominent O VII and O VIII absorption edges are visible, but, consistent with the much lower total absorbing column, no Fe K absorption edge is detectable. A weak, narrow Fe K alpha emission line from cold material is present as well as a broad Fe K alpha line. These features are similar to those reported in other Seyfert 1 galaxies. A single warm absorber model provides only an imperfect description of the low-energy absorption. In addition to a highly ionized absorber with ionization parameter U = 1.66 and a total column density of 1.4 x 10(22) cm(-2), adding a lower ionization absorber with U = 0.32 and a total column of 6.9 x 10(21) cm(-2) significantly improves the fit. The contribution of resonant line scattering to our warm absorber models limits the Doppler parameter to less than 160 km s(-1) at 90% confidence. Turbulence at the sound speed of the photoionized gas provides the best fit. None of the warm absorber models fitted to the X-ray spectrum can match the observed equivalent widths of all the UV absorption lines. Accounting for the X-ray and UV absorption simultaneously requires an absorbing region with a broad range of ionization parameters and column densities.

The X-ray binary pulsar GX 301-2 was observed with the Advanced Satellite for Cosmology and Astrophysics (ASCA) on 1994 February 13-14, at the binary orbital phase between 0.277 and 0.308. The source was found to be in a low-intensity phase, with an intrinsic luminosity L(X) = 6.6 x 10(35) ergs s(-1) in the 0.7-10.0 keV energy band. This is roughly 2 orders of magnitude lower than the peak luminosity generally observed at periastron. The spectral analysis shows the presence of a strong low-energy excess below about 4 keV and high photoelectric absorption (similar to 6 x 10(23) cm(-2)). Part of the soft excess may be caused by the scattering of X-rays around a region of increased density by the less dense ambient stellar wind, which would explain the absence of pulsations in the low-energy excess. The spectrum also shows a strong, narrow fluorescent iron line and an absorption iron K edge. The average energy of the line is 6.400 +/- 0.008 keV and that of the edge is 7.183 +/- 0.013 keV (at the 90% confidence level). The iron line equivalent width is 228 +/- 18 eV. The partial covering model does not describe the low-energy excess adequately. It underestimates the continuum between 2.0 and 5.0 keV but overestimates it below 1.5 keV. This, together with the absence of pulsations in the low-energy excess, rules out the partial covering model as a good description of the continuum spectrum. In addition, the spectra from the solid-state imaging spectrometers (SIS) aboard ASCA reveal two new features. First, it is seen that the fit with the scattering model is not acceptable below 2 keV and suggests an additional component as cause for this residual soft excess. The ultrasoft component has been observed in the spectra of GX 301-2 for the first time and can be fitted well by thermal emission from a hot plasma with a temperature kT of similar to 0.8 keV. The ultrasoft component may arise because of emission from X-ray-induced shocks in the gas trailing the neutron star. A partial ionization model with a warm absorber in series with a neutral absorber, including scattering, does not account for the ultrasoft excess below 1.5 keV and can be ruled out. Second, it is also seen that the SIS spectrum is dominated by line features at 0.80, 1.73, 2.30, 2.98, and 3.68 keV. These can be identified with the fluorescent line emission from neutral or near-neutral Ne, Si, S, Ar, and Ca, respectively. This implies that the line emission is coming from cold (less than or equal to 10(5) K) gas. The pulse phase spectroscopy shows that the photon index, the iron line intensity, and the equivalent width vary with pulse phase. The large equivalent width variations are, however, mainly due to the changes in the underlying continuum intensity.

The X-ray binary pulsar GX 301-2 was observed with the Advanced Satellite for Cosmology and Astrophysics (ASCA) on 1994 February 13-14, at the binary orbital phase between 0.277 and 0.308. The source was found to be in a low-intensity phase, with an intrinsic luminosity L(X) = 6.6 x 10(35) ergs s(-1) in the 0.7-10.0 keV energy band. This is roughly 2 orders of magnitude lower than the peak luminosity generally observed at periastron. The spectral analysis shows the presence of a strong low-energy excess below about 4 keV and high photoelectric absorption (similar to 6 x 10(23) cm(-2)). Part of the soft excess may be caused by the scattering of X-rays around a region of increased density by the less dense ambient stellar wind, which would explain the absence of pulsations in the low-energy excess. The spectrum also shows a strong, narrow fluorescent iron line and an absorption iron K edge. The average energy of the line is 6.400 +/- 0.008 keV and that of the edge is 7.183 +/- 0.013 keV (at the 90% confidence level). The iron line equivalent width is 228 +/- 18 eV. The partial covering model does not describe the low-energy excess adequately. It underestimates the continuum between 2.0 and 5.0 keV but overestimates it below 1.5 keV. This, together with the absence of pulsations in the low-energy excess, rules out the partial covering model as a good description of the continuum spectrum. In addition, the spectra from the solid-state imaging spectrometers (SIS) aboard ASCA reveal two new features. First, it is seen that the fit with the scattering model is not acceptable below 2 keV and suggests an additional component as cause for this residual soft excess. The ultrasoft component has been observed in the spectra of GX 301-2 for the first time and can be fitted well by thermal emission from a hot plasma with a temperature kT of similar to 0.8 keV. The ultrasoft component may arise because of emission from X-ray-induced shocks in the gas trailing the neutron star. A partial ionization model with a warm absorber in series with a neutral absorber, including scattering, does not account for the ultrasoft excess below 1.5 keV and can be ruled out. Second, it is also seen that the SIS spectrum is dominated by line features at 0.80, 1.73, 2.30, 2.98, and 3.68 keV. These can be identified with the fluorescent line emission from neutral or near-neutral Ne, Si, S, Ar, and Ca, respectively. This implies that the line emission is coming from cold (less than or equal to 10(5) K) gas. The pulse phase spectroscopy shows that the photon index, the iron line intensity, and the equivalent width vary with pulse phase. The large equivalent width variations are, however, mainly due to the changes in the underlying continuum intensity.

The Gas Imaging Spectrometer (GIS) system on board ASCA is described. The experiment consists of 2 units of imaging gas scintillation proportional counters with a sealed-off gas cell equipped with an imaging phototube. The performance is characterized by the high X-ray sensitivity (from 0.7 keV to over 10 keV), good energy resolution (7.8% FWHM at 6 keV following E(-0.5) as a function of X-ray energy E), moderate position resolution (0.5 mm FWHM at 6 keV with E(-0.5) dependence), fast time resolution down to 61 mu s, and an effective area of 50 mm diameter. The on-board signal processing system and the data transmitted to the ground are also described. The background rejection efficiency of the GIS is reaching the level achieved by the non-imaging multi-cell proportional counters.

The in-orbit performance and calibration of the Gas Imaging Spectrometer (GIS), located on the focal plane of the X-ray astronomy satellite ASCA, are described. An extensive in-orbit calibration has confirmed its basic performance, including a position resolution of 0.6 mm (FWHM) and an energy resolution of 7.8% (FWHM), both at 6 keV. When combined with the X-ray telescope, the GIS sensitivity range becomes 0.7-10 keV. The in-orbit non X-ray background of the GIS has been confirmed to be as low as (5-9) x 10(-4) cs(-1) cm(-2) keV(-1) over the 1-10 keV range. The long-term detector gain has been stable within a few ro for nearly 3 years. Extensive observations of the Crab Nebula and other sources have provided accurate calibrations of the position response, photometric capability, dead time, and timing accuracy of the GIS. Furthermore, the overall energy response, including the temporal and positional gain variations and the absolute gain scale, has been calibrated to similar to 1%. Thus, the GIS is working as an all-round cosmic X-ray detector, capable of X-ray imagery, fine X-ray spectroscopy, X-ray photometry With a flux dynamic range covering more than 5 orders of magnitude, and fast X-ray photometry with a time resolution up to 60 mu s.