Research

Research and development is also being carried on to extend the applications of our detectors optimizing both the sensors and the readout techniques.

Our detectors are normally used in the energy range 5-30 keV. In this range the absorption efficiency of standard 0.3-1 mm silicon sensors is acceptable. Silicon sensors of thickness up to 2 mm have been tested in terms of spatial resolution and signal-to-noise ratio in order to improve the absorption efficiency for hard X-rays.

In order to achieve an acceptable absorption efficiency for harder X-rays either thicker silicon of high-Z sensor materials should be used. Test measurements are being carried out in collaboration with other institution to develop strip and pixel CdTe, CdZnTe and GaAs sensors for our detectors.

For energies lower than 2 keV, most of the photons are absorbed at the sensor surface, therefore the entrance window i.e. the backplane of the sensor must be optimized. We collaborate with silicon foundries in order to optimize the charge collection efficiency at the entrance window of the sensor.

The charge produced by the X-rays in the silicon sensors diffuses in the bulk while drifting to the collecting electrodes. This means that if a photon is absorbed in a region between two electronics channels, it signal will be splitted between them (charge sharing effect). Even if it is a disadvantage for single photon counting detectors, the charge sharing can be exploited by charge integrating ones to improve the spatial resolution by inter-channel interpolation.

This method has been tested using GOTTHARD, MOENCH . The hit position of single photons can be interpolated by exploiting the fix correlation between the hit position and the ratio of the charge collected by neighboring channels. Once this relation has been measured, with the help of a uniform photon field, the position of incoming photons can be recostructed with a (sub-)micron precision.

The main limitation of state-of-the-art counting systems appears at high photon fluxes. In fact, if a second photon arrives during the time required to register the previous one, it is lost and causes a loss of efficiency and of linearity. Although the data can be partially corrected, this effect sets a maximum limit on the count rate for the detector and consequently the data throughput is reduced. Higher fluxes can only be measured with charge integrating (CI) detectors, which normally have disadvantages like limited dynamic range or a resolution that is not single-photon, sensitivity to sensor dark current and contribution to the background arising from the fluorescent radiation possibly emitted by the samples. The ideal detector would have the noise level, dynamic range and background suppression capability of a SPC detector, but the flux capability of a CI device.

A method to extend the count-rate capability is to operate SPC devices using the time-over-threshold (ToT) acquisition mode while still preserving their outstanding dynamic range. It allows the energy deposited by a particle in the detector element to be estimated by measuring the time during which the signal generated by the detected particle remains above a comparator threshold. The ToT method directly converts the signal pulse height into a digital value in the early stage of the front-end electronics in parallel for all the channels of the detector, which greatly simplifies the system compared with analog detectors with serial readout through one or several ADCs.

The rate capability of MYTHEN is extended by at least a factor of two when using the ToT acquisition mode and this can further be optimized by using a different shaping of the analogue signal. MYTHEN can be operated with high gain settings (low noise) with improved performances at high rates compared with the standard or fast settings of traditional SPC operation and, therefore, experiments can be performed at low energies also with high radiation fluxes, without the need to attenuate the beam as would be required in SPC mode.

Another method to extend the count rate capability is pileup tracking by using multiple comparator threshold. In this case, one threshold is kept as normal at 50% of the photon energy, while additional threshold are tuned at higher values in order to detect photons which pile up at high rates. This mehtod has been tested using MYTHEN II where, together with an outstanding shaping time shorter than 30 ns, it allows to achieve count rates higher than 25 MHz per strip with 80% efficiency using three counters.

Hybrid detector can be used not only to detect X-rays, but also different type of ionizing particles.

Applications like PEEM and electron microscopy also benefit from the large dynamic range, fast frame rate capability and low noise level of ourdetectors.

It is very challenging to obtain single photon resolution for energies lower than 1 keV since the charge created by single photons is comparable with the electronic noise of our readout electronics and is difficult to discriminate.

Low Gain Avalanche Detectors (LGADs) are novel sensors offering internal multiplication of the signal. Proof-of-principle measurements using both single photon counting and charge integrating microstrip readout show that the single photon resolution can be pushed to much lower energies. We have demonstrated that by using of a multiplication factor of about 7, the minimum detectable energy of the MYTHEN single photon counting microstrip detector could be reduced of a factor of 5. 

Based on these encouraging results, we are collaborating with silicon foundries in order to optimize the performance of this  technology, which was originnally developed for timing for high energy physics, for soft X-ray detection. In particular, we need to optimize the segmentation and the quantum efficiency of these devices.