The CCD detectors in the X-ray Imaging Spectrometers (XIS) aboard Suzaku have been equipped with a precision
charge injection capability. The purposes of this capability are to measure and reduce the detector degradation
caused by charged particle radiation encountered on-orbit. Here we report the first results from routine operation
of the XIS charge injection function. After 12 months' exposure of the XIS to the on-orbit charged particle
environment, charge injection already provided measurable improvements in detector performance: the observed
width of the 5.9 keV line from the onboard calibration source was reduced from 205 eV to less than 145 eV.
The rate of degradation is also significantly smaller with charge injection, so its benefit will increase as the
mission progresses. Measured at 5.9 keV, the radiation-induced rate of gain degradation is reduced by a factor
of 4.3 ± 0.1 in the front-illuminated sensors when injecting charge greater than 6 keV equivalent per pixel. The
corresponding rate of degradation in spectral resolution is reduced by a factor 6.5 ± 0.3. Injection of a smaller
quantity of injected charge in the back-illuminated XIS sensor produces commensurately smaller improvement
factors. Excellent uniformity of the injected charge pattern is essential to the effectiveness of charge injection in
The energy resolution of the X-ray CCDs onboard the Suzaku satellite (X-ray Imaging Spectrometer; XIS) has
been degraded since the launch due to radiation damage. To recover from this, we have applied a spaced-row
charge injection (SCI) technique to the Suzaku XIS in orbit. By injecting charge into CCD rows periodically,
the energy resolution 14 months after launch is improved from 210 eV to 150 eV at 5.9 keV, which is close to
the resolution just after the launch (140 eV). Additional information on these results is given in a companion
paper by the XIS team. In this paper, we report the details of CCD charge transfer inefficiency (CTI) in the
SCI mode, the correction method, and the implementation of it in ground analysis software for XIS data. In the
SCI mode, CTI depends on the distance of a charge packet from the nearest charge-injected row, and the gain
shows a periodic non-uniformity. Using flight data obtained with the onboard calibration sources, as well as a
cosmic source (the Perseus cluster of galaxies), we studied the non-uniformity in detail. We developed a method
to correct for the non-uniformity that will be valuable as the radiation damage progresses in future.
Suzaku is the fifth Japanese X-ray astronomical satellite and it was launched in July 2005. The Suzaku X-ray
Imaging Spectrometers (XISs) consist of four X-ray Charge-Coupled Device (CCD) cameras. Three of them are
front-illuminated (FI) CCD, and the other is back-illuminated (BI) CCD. The strong points of the XIS are a
high energy resolution, a large effective area, and a low and stable background. In particular, the background
level of the Suzaku/XIS is much lower than the other X-ray satellites, XMM-Newton/EPIC and Chandra/ACIS.
We investigated the background property of the XIS using the data obtained when the satellite is looking at the
night earth, and proved the low level and the stability of the XIS background. Non X-ray background (NXB)
consists of continuum component and some emission lines. The continuum component is very different between
the FI-CCD and the BI-CCD. We discussed the positional dependence of the continuum component and the line
components, and proved that the flux of the line components of the NXB is higher in the frame-store region than
the imaging area. Finally, we investigated the effects of magnetic cut-off rigidity (COR) upon the count rate of
The X-ray Imaging Spectrometer on the Suzaku satellite consists three front-illuminated (FI) and one back-illuminated (BI) CCD cameras. Using ground calibration data taken at Kyoto University and Osaka University, we obtained the energy response of the XIS, which consists of at least six components: 1. a main peak, 2. a sub peak, 3. a triangle component, 4. a Si escape, 5. a Si line, and 6. a constant component. The relation between the energy and the pulse height was also estimated, which is called as a gain. The relation cannot be represented with a single linear function. Then we divided the gain into two parts at the Si edge (1.839 keV) and each part can be described with a single linear function. Thus there is a discontinuity at 1.839 keV in the XIS gain. We have monitored the variation of the gain and energy resolution in orbit by observing the calibration source of 55Fe illuminating two corners of each CCD.
We report on the results of the ground calibration of Astro-E2/XIS with front-illuminated (FI) chips. The sensors have basically the same performance as that of Astro-E/XIS. However, there are some improved points: (1) A 55Fe radio isotope is equipped on a door, and (2) a charge injection (CI)capability (described below) is added. The FI sensors have been calibrated at Kyoto University, Osaka University, and MIT. At Kyoto University we focus on the high energy range (>1.5 keV). We measured the gain, energy resolution, and quantum efficiency as the function of energy by using characteristic X-rays for each sensor. An energy resolution of 130 eV@5.9 keV (FWHM) and a quantum efficiency of email@example.com keV are achieved. After XIS is launched, the Charge Transfer Inefficiency (CTI) increases due to the radiation damage by cosmic rays. Then XIS equips the CI capability to calibrate and compensate the increase of the CTI. In order to utilize the CI capability, the amount of charge injected into the CCDs is expected to be kept constant. The time variability of the amount of the injected charge is estimated.