High Energy Telescope onboard HXMT Satallite

  HXMT/HE adopts an array of NaI(Tl)/CsI(Na) PHOSWICH as the core detector. Its geometry area is 5096 cm2 and the field of view (FoV) is 5.2°x 5.2° (FWHM). HXMT/HE has the capabilities of spectral, timing, and imaging at the same time. HE will perform high sensitivity scanning in some small area regions, frequent and fast scanning on the Galactic plane to discover some X-ray bursting and transient sources. High sensitivity pointing observations are required for important X-ray sources (e.g. X-ray binary) to investigate their spectral and timing properties. The CsI(Na) in the PHOSWICH of HE can be used as a γ-ray burst (GRB) monitor. Its energy range in the normal mode is 40-600 keV and is changed to 200 keV-3 MeV in the GRB mode. It is expected to play an important role in the observation of GRB spectra and hunting for the electromag-netic counterpart of gravitational wave events.

  HE consists of main detector (HED), high energy collimator (HEC), auto-gain control detector (HGC), anti-coincidence shield detector (HVT), particle monitor (HPM), data processing and control box (HEB) and power box for HE (HEA). The major features of HE are listed in Table 1.

Table 1 main features of HE

  HED is the main detector of HE and responsible for the detection of X-ray in 20-250 keV energy band. 18 collimators are arranged in a certain pattern according to the inner grid orientations are adopted to confine the FOV of HED, suppress back-ground, and provide necessary spatial modulation for direct demodulation imaging. HGC achieves the auto-gain control on HED and spectral calibration. HVT will perform active shielding against charged particles and reduce background caused by charged particles. HPM will monitor the flux of charged particles and alert to close the high voltage of HEDes and HVTes in high flux region to avoid the damage to the PMTs of these detectors. The overview structure of HE is shown in Figure 1.


Figure 1 A sketch of the components of HE

  The diameter of the crystal in each HED is 190 mm and its area is 283.5 cm2. Thus, the total geometry area is about 5100 cm2. HED adopts NaI(Tl)/CsI(Na) PHOSWICH as the detector. The full energy of incident X-ray photon is deposited in the NaI(Tl) crystal, which forms a “good event”. CsI(Na) is used as active shielding to reject the background events from back-side within ~2π solid angle.

  Corresponding to each HED, there is a collimator in front of each HED. Taking fully into account the requirements of the spatial resolution of imaging, large FOV, and background subtraction, the FOV of 15 collimators is 1.1°x 5.2°, which are called “narrow-FOV collimators”; the FOV of 2 collimators is 5.2°x 5.2°, which are called “wide-FOV collimators”; 1 col-limator with the FOV of 1.1°x 5.2°is covered by a 2mm-thick tantalum lid, which is used to measure the local background of HE and called “Blind collimator”. 18 collimators are distributed along two concentric circles (6 in the inner circle and 12 in the outer circle).

  There is a HGC for each HED, which is installed in the corresponding collimator grid. It can adjust the high voltage of HED automatically in real time and thus ensure the gain stability in orbit.

  HVT system consists of 6 top detectors and 12 lateral detectors. They cover the front and side face of HEDs to shield the background induced by the incident charged particles from the corresponding direction. They are used as the active shielding detector of HED for the 2π solid angle in front side.

  HPM will monitor the flux of high energy protons and electrons in orbit. If the flux is higher than a pre-set threshold, the satellite platform will decrease the high voltage of PMTs of HED and HVT to avoid the damage to PMTs. There are three HPMs for HE and they are the backup for each other.

  The signals from all the HE detectors are dealt by HEB. The data is delivered to the satellite platform by LVDS and 1553B bus. HEB receives the time-synchronous signal, pulse-per-second signal of GPS, and high-accuracy signal of 5MHz crystal from the satellite platform at the same time. HEA provides secondary power for all detectors and HEB.

Figure 2 A sketch of a HED unit

  The flight model of HE has passed thermal and vibration test and completed system-level integration test to verify its in-terface and function. The ground calibration experiment has determined the performance of HE.

  The measurement of response is the main purpose of the ground calibration of HE. The experiments have been carried out both in normal and GRB mode. The experiment in normal mode was completed by a double-crystal monochromator (based on X-ray tube) and radioactive sources. Only radioactive sources were used in the calibration of GRB mode.

  Figure 3 presents the non-proportional efficiency of some NaI(Tl) experiments. The vertical axis stands for the pulse height per unit energy and all points are normalized at 60 keV. It can be found that the result of HE is consistent with other experiments, except for the low energy band of Fermi/GBM.

Figure 3 Nonlinearity response of NaI(Tl) for Z01-5, normalized to unity at 60keV 

  According to the fits of the full energy peak of the NaI events of HED, we can obtain the response of HED to different energies of X-ray and then the E-C relationship. The E-C relation of detector Z01-25 is shown in Figure 4. The residual of the fit is smaller than 1% at all points.


Figure 4 E-C relationship and corresponding residuals of Z01-5. Left: normal mode; right: GRB mode

  According to the calibration results, the detection energy range of NaI(Tl) is 16-350 keV with the 18 HEDs in normal mode. For CsI(Na), it is 130 keV–3 MeV in GRB mode.


Figure 5 Energy resolution of Z01-25 as a function of energy.  Left: normal mode; right: GRB mode. 

  In the normal mode, NaI events are good events. In the GRB mode, CsI events are good events. NaI or CsI events of the full energy peak were selected according to the signal width. Figure 6 (left) is the efficiency of the full energy peak of NaI of 18 HED in the normal mode. The red curve is the result of Monte-Carlo simulations. The blue lines are the experimental results of 18 flight models, which are well consistent with each other and consistent with trend of the simulation result. Figure 6 (right panel) is the efficiency of the full energy peak of CsI in the GRB mode. The red curve is the result of Monte-Carlo simulations. The blue lines are the experimental results of 18 flight models, of which the trend is consistent with the simula-tion result. In order to avoid wrong screening, the signals with the width of 91~110 ch were considered as pure CsI signals in the data processing. The low energy photons induced a broader distribution of pulse width and thus more low energy photons were excluded. As a result, the difference between the simulation and experimental results decreases with the increasing energy.


Figure 6 Efficiency of 18 HEDs derived from calibration experiments. Left: normal mode; right: GRB mode