Detectors at ID1

It is a point detector normally used for energies from 5 keV to above 40 keV. It has a low noise (0.1 cps) and a relatively high maximum counting rate of ~ 2 x 10⁵ cps with 25% dead time.

For energies lower than 3 keV, the 200 microns Be window absorbs a substantial part of the incoming radiation (50% for 3 keV). Then, it can still be used, although care has to be taken when making comparison with absolute intensities taken at higher energies. 

A model specially developed for ID1 for vacuum compatibility allows working at atmospheric pressure and in vacuum (e.g. 1 x 10^-3 Torr) as well as in the difficult intermediate pressure range (the sealed detector head includes the photo multiplier high voltage source, to avoid problems associated with flashover at cabling connections). 

This detector is normally used for energies from 8 keV to 12 keV, where is > 50% efficient. For higher energies the X-ray absorption efficiency is low, but the detector is still usable. It has a detection area of 50 x 1 mm² and a spatial resolution of about 80 microns. 

As other gas filled detectors, it normally works filled with a mixture of Ar and CH₄ but it can also work filled with Xe, which extends considerably the range of higher energy accessible (up to 15 keV). An obvious inconvenient of Xe is its high price if used as a flow gas. A closed loop gas regeneration circuit for the filling gas is a possible alternative but technically difficult to operate.

Advantages of this detector are its good spatial resolution; its compactness and that it can be mounted in the detector arm. Therefore, it substitutes with advantage in many cases the bulky 2D gas filled detector when SAXS and WAXS are done simultaneously. Furthermore, the good collimation path (angular resolution is 0.1mrad) obtained with slits just after the sample and just before the detector defines a very small beam footprint in the sample surface (figure 4). In the collimation path itself there is primary vacuum. Air scattering is in this way drastically reduced, and by using a He flux cone as sample environment (figure 5), the air scattering is reduced by a factor of 20, thus making to pump the whole diffractometer vessel unnecessary. 

Fig. 1 (left) The Braun Position Sensitive Detector mounted in the detector arm. The collimation path has slits at both ends, and has primary vacuum.

Fig. 2 (right). The He flux cone containing the sample.

A serious drawback of this detector is the limited maximum counting rate before saturation (linear regime) of about 5 kcps. This means that above that counting rate, the detector dead time is higher than 10%. The maximum counting rate that it can stand is about 30 Kcps, above which detector failure is possible. It is obvious than there is an essential mismatch between having a beam of up to 1x10¹³ cps and a detector with only 5 kcps maximum count rate. In the medium or long run, an alternative to the linear PSD should be found. A serious candidate should be photodiode arrays, but which will require development to our specifications.

An important development of ID1 in the last years is the new Princeton CCD camera. Being a 2D detector that can be mounted in the detector arm, it records reciprocal space maps at small and wide angles in few seconds. It is also much less fragile with respect to radiation damage with respect to the gas filled detectors. 

The CCD camera from Princeton is intended for low energy scattering and is compatible with a 1.3x10^-3 mbar vacuum. The system comprises a CCD head with a 0.5 mm thick Be window and a 90 mm diameter optic fiber reducer coupled to a 1242x1152 pixels CCD with Peltier cooling to -60^oC, a controller, a temperature control unit, and a chiller. The system can be operated either from SPEC or as a standalone unit with PC/WinView control. The detection area is 67 x 62 mm² input field with a pixel size of 54 microns. The spatial resolution measured is 110 microns. The linear dynamic range is 15 bit. No dark signal was found at 100 s exposure. The readout rates are ~ 4 s / image (430 kHz pixel rate) and ~ 1.5 s / image (1 MHz pixel rate). We plot in figure 3 the efficiency as function of energy.

Fig. 3. Efficiency vs. energy for Princeton CCD camera.

This is the commercially available X-flash silicon drift diode normally used for energies from 3 keV to 15 keV. Its main advantage is that it is very fast; its maximum throughput count rate is about 300 kcps at 60% dead time loss. It has a resolution of about 180 - 200 eV at 6 keV and an aperture of 2.5 mm diameter.

It is based on a Si diode (300 microns thick) with Peltier cooling. As the Roentec and the CdZnTe detector below, it is a photon counter. Its energy range is from below 3 to ~ 20 keV. Its maximum count rate is quite low, about 10 kcps, but is has the advantage of being commercially available in a vacuum compatible package. Its resolution is about 180 eV and it is somehow fragile against radiation damage. 

This is a photon counter especially suitable for high energies. It can be used from 3keV to >100 keV (figure 7). The energy resolution is about 800 eV when the detector is cooled to –20C. The maximum count rate is about 10 kcps.

Fig. 4. Efficiency vs. energy for the CdZnTe detector. One can see the very large operation range of this detector (3-100 keV) with a very high efficiency.