Simulation Tools for Detector Performance and Calibration at European XFEL Ashley Joy, Markus Bohlen, Steffen Hauf, Burkhard C. Heisen, Andreas Koch, Markus Kuster, Jolanta Sztuk–Dambietz, Monica Turcato, Matthew Wing and Christopher Youngman Abstract—The detectors to be used at the European XFEL have to deal with the unique time structure of the machine, delivering up to 2700 pulses, with a repetition rate of 4.5 MHz, ten times per second, the very high photon flux of up to 10 12 photons/pulse and the need to combine single-photon sensitivity and a large dynamic range. These machine properties present a challenge for the large-area 2D imaging detectors to be used at European XFEL. In order to thoroughly characterize the detectors, optimize their performance and the required calibration concepts, as well as give estimates of the expected scientific performance in a wide range of experimental scenarios, we are currently pursuing different simulation projects. I. I NTRODUCTION T HE European X-ray Free Electron Laser (XFEL) dis- tinguishes itself from existing facilites like the SLAC Linac Coherent Light Source (LCLS), USA and the SPring- 8 Angstrom Compact free electron LAser (SACLA), Japan, by its high repetition rate and high peak brilliance. The machine will deliver up to 27000 pulses/s amounting to 5 × 10 33 photons/s/mm 2 /mrad 2 (0.1% bandwidth, energy range: 0.25 keV - 25 keV) at 4.5 MHz repetition rate. This poses unique challenges in terms of performance and radiation hardness for the 2D X-ray detectors which are to be used at the instrumental stations. In order to thoroughly characterize these detectors, optimize their performance and the required calibration concepts, as well as give estimates of the expected scientific performance in a wide range of experimental scenarios, we are currently pursuing different simulation projects. Initially, small self-contained simulation tools were devel- oped, e.g. for estimating the performance of calibration sources or for estimates on the performance of the large 2D pixel detectors; the results of which are presented here. Additionally, we are currently developing a large-scale simulation toolkit, which will be integrated into the European XFEL software framework, Karabo [1]. It is envisaged that this framework will be available to XFEL facility users and allow the simulation of the 2D pixel detectors used at European XFEL – starting from the photons incident on the detector down to the level of the read-out electronics. Manuscript received May 9, 2014 A. Joy and M. Wing are at University College London, London, U.K (e- mail: a.joy@ucl.ac.uk). M. Bohlen, S. Hauf, B.C. Heisen, A. Koch, M. Kuster, J. Sztuk-Dambietz, M. Turcato and C. Youngman are with European X-ray Free Electron Laser Facility GmbH, Hamburg, Germany (e-mail: steffen.hauf@xfel.eu). [keV] γ E 0 20 40 Counts/mA*ns (all X-rays) -2 10 -1 10 1 10 2 10 3 10 4 10 C @ 50kV Al @ 50kV Cu @ 50kV Mo @ 50kV Ag @ 50kV Au @ 50kV Fig. 1: Geant4 simulation of the X-ray spectra generated by an electron gun based X-ray source using different target materials: C, Al, Cu, Mo, Ag and Au. Each simulation consisted of 40 Million primary electrons at 50 keV. The target angle was set to 18 ◦ at which X-ray production was found to be most efficient. target Z 0 10 20 30 40 50 60 70 80 Eff 0.002 0.004 0.006 0.008 0.01 0.012 0.014 0.016 0.018 0.02 = 100 keV e E = 50 keV e E = 30 keV e E Fig. 2: Conversion efficiency of incident primary electrons into Bremsstrahlung X-rays depending on target material. Estimates for incident electrons at energies of 30 keV, 50 keV and 100 keV and elements C (Z=6), Al (Z=13), Cu (Z=29), Mo (Z=42), Ag (Z=47) and Au (Z=79) are shown. Each simulation consisted of 40 Million primary electrons. The target angle was set to 18 ◦ at which X-ray production was found to be most efficient. II. X- RAY CALIBRATION SOURCE SIMULATIONS Calibration of the 2D Mega-pixel detectors with 4.5 MHz readout rate will pose unique challenges for the laboratory