Accelerator Based Infrared Sources and Detector Saturation

In order to perform various investigations by infrared synchrotron radiation, infrared detectors are important as well as source characteristics. In the next generation low emittance synchrotron plans, it is hard to expect that the infrared light becomes stronger than the current status. Therefore, detector systems become more important. In this study, characteristics of detector saturation are investigated by infrared synchrotron radiation (IR-SR) at infrared beamline BL43IR in SPring-8 and infrared free electron laser (IR-FEL) at KU-FEL in Kyoto University. HgCdTe (MCT) detector is an infrared semiconductor detector, which is most frequently used for mid-infrared spectroscopy measurements. A high sensitivity and a high response speed are the prominent features of MCT detector. However, the linearity is not good and the saturation causes serious problems to infrared spectra. The electron energy and the storage ring current of SPring-8 are 8 GeV and 100 mA, respectively. The Photon flux of the infrared light at SPring-8 is not higher than that at the other facilities. The problem of detector saturation, however, is serious. A photoconductive MCT detector mounted on BRUKER HYPERION2000 was used to evaluate the saturation. SPring-8 has eight types of bunch modes. The time structures are different in each bunch mode. We find that the detector saturation causes a decrease of the signal intensity and a deterioration of the signal-to-noise ratio, and degree of the saturation depends on the bunch mode. In order to exclude the time structure parameters and investigate the saturation effect in detail, we used IR-FEL. The macro pulse time width is about 2 microseconds at KU-FEL, that is similar to the train part duration time of IR-SR at SPring-8. Fig. 1 shows output voltage of photoconductive MCT detectors. The IR-FEL beam was split into two and they were injected into two detectors, MCT-1 and MCT-2. The intensity of the light into MCT-2 was enough low not to saturate. The intensity of the light into MCT-1 was increased from (a) to (c). The curve (a) shows a linear dependence, which means the detector is not saturated. The curve (b) has a flat part at the high voltage region, which is a typical saturation behavior. The injected intensity of the curve (c) is the highest among three, but the output voltage is found to be suppressed than that of (b). At the deep saturation condition, care must be taken in the analysis of the signal intensity. In order to reveal the mechanism of the saturation, we also observed response curves at various injection intensities and pre-amp gains using photoconductive and photovoltaic MCT detectors. Based on the results, appropriate use of MCT detector will be discussed.