We have produced thick-foil and fine-pitch gas electron multipliers (GEMs) using a laser etching technique. To improve production yield we have employed a new material, Liquid Crystal Polymer, instead of polyimide as an insulator layer. The effective gain of the thick-foil GEM with a hole pitch of 140 µm, a hole diameter of 70 µm, and a thickness of 100 µm reached a value of 10 4 at an applied voltage of 720 V. The measured effective gain of the thick-foil and fine-pitch GEM (80 µm pitch, 40 µm diameter, and 100 µm thick) was similar to that of the thick-foil GEM. The gain stability was measured for the thick-foil and fine-pitch GEM, showing no significant increase or decrease as a function of elapsed time from applying the high voltage. The gain stability over 3 h of operation was about 0.5%. Gain mapping across the GEM showed a good uniformity with a standard deviation of about 4%. The distribution of hole diameters across the GEM was homogeneous with a standard deviation of about 3%. There was no clear correlation between the gain and hole diameter maps.

Hercules X-1 was observed with Suzaku in the main-on state from 2005 to 2010. The 0.4–100 keV wide-band spectra obtained in four observations showed a broad hump around 4–9 keV in addition to narrow Fe lines at 6.4 and 6.7 keV. The hump was seen in all the four observations regardless of the selection of the continuum models. Thus it is considered a stable and intrinsic spectral feature in Her X-1. The broad hump lacked a sharp structure like an absorption edge. Thus it was represented by two different spectral models: an ionized partial covering or an additional broad line at 6.5 keV. The former required a persistently existing ionized absorber, whose origin was unclear. In the latter case, the Gaussian fitting of the 6.5-keV line needs a large width of σ = 1.0–1.5 keV and a large equivalent width of 400–900 eV. If the broad line originates from Fe fluorescence of accreting matter, its large width may be explained by the Doppler broadening in the accretion flow. However, the large equivalent width may be inconsistent with a simple accretion geometry.