N3B3  Synchrotron Radiation and FEL Instrumentation

Wednesday, Nov. 4  10:30-12:30  Golden West

Session Chair:  Gabriella Carini, SLAC National Accelerator Laboratory, United States; D Peter Siddons, National Synchrotron Light Source, United States

Show/Hide All Abstracts

(10:30) N3B3-1, ePixS: a High Channel Count X-Ray Spectroscopy Detector for X-Ray FELs

A. Dragone, P. Caragiulo, B. Markovic, G. Blaj, J. Hasi, J. Segal, A. Tomada, K. Nishimura, R. Herbst, P. Hart, S. Osier, J. Pines, C. Kenney, G. Haller

SLAC National Accelerator Laboratory, Menlo Park, CA, United States

ePixS is a novel multichannel X-ray spectroscopy detector that can deliver Silicon Drift Detector (SDD) like performance at strobed x-ray sources (X-ray FEL), where it can measure a spectrum on a pulse-by-pulse basis. It is based on hybrid architecture with p-i-n diodes flip-chipped bonded to mixed signal integrated circuits. The ePixS ASIC from which the detector takes the name, is the latest addition to the class of x-ray signal processors named ePix and has an equivalent input noise of 8 e- r.m.s. at room temperature. The full detector is typically operated at -40° C where it shows a noise floor of 12 e- r.m.s. at nominal 32µs integration time. The first generation offers 4 10x10 pixel sub-arrays with a pitch of 500 µm. The system is scalable to 10,000 pixels and can sustain a frame rate of 30kHz. Since it has an electronic shutter, sub arrays can also be phased to achieve an 100 kHz effective frame rate. Although intended for LCLS, it is a natural fit to the requirements of other FELs and offers significant advantages over an array of SDDs in terms of cost, data acquisition, and system complexity, while providing similar spectral resolution. The large channel count drastically reduces the time required to take a two or three-dimensional scan of a sample allowing more experiments to be performed in a given period. The detector can be also operated at synchrotrons, showing performances comparable to SDDs in terms of resolution and, because of the fine segmentation, with higher throughput. The system and results from prototypes will be presented.

(10:50) N3B3-2, Experimental Characterization of the PERCIVAL Soft X-Ray Detector

A. Marras1, C. B. Wunderer1, M. Bayer1, J. Correa1, P. Goettlicher1, S. Lange1, I. Shevyakov1, S. Smoljanin1, M. Viti1, Q. Xia1, M. Zimmer1, D. Das2, N. Guerrini2, B. Marsh2, I. Sedgwick2, R. Turchetta2, G. Cautero3, D. Giuressi3, A. Khromova3, R. Menk3, L. Stebel3, R. Fan4, J. Marchal4, U. Pedersen4, N. Rees4, P. Steadman4, M. Sussmuth4, N. Tartoni4, H. Yousef4, H. Hyun5, K. Kim5, S. Rah5, S. Reza1,6, H. Graafsma1,6

1Deutsches Elektronen-Synchrotron (DESY), Hamburg, Geremany
2Science & Technology Faculties (STFC), Didcot, UK
3ELETTRA Sincrotrone, Elettra, Italy
4Diamond Light Source (DLS), Didcot, UK
5Pohang Accelerator Lab (PAL), Pohang, South Korea
6Mittsweden University, Sundsvall, Sweden

High-brilliance X-ray synchrotron sources and X-ray Free-Electron Lasers impose demanding constraints on detectors. PERCIVAL is a soft-X-ray detector under development as a collaboration between DESY, STFC, ELETTRA, DLS and PAL to answer those needs (high dynamic range, fast acquisition rate, low noise, good QE, good spatial resolution). When produced in its final form the PERCIVAL system will be a multi-mega-pixel imager; before final production, test prototypes have been built on a reduced scale: Monolithic Active Pixel Sensor arrays (~33kpixel), with ~25um pixel pitch and a multiple-gain circuit (capacitor overflow) to increase the dynamic range. They are back-illuminated (to achieve 100% fill factor) and post-processed to minimize entrance window. Correlating Double Sampling, Analogue-to-Digital conversion (to 12+3 bits), fast digital readout (up to 120 frames/s) are performed on-chip. The detector is included in a vessel: a FR4 board glued to a custom flange allows access to output data, while keeping the sensor and front-end electronics in vacuum. Digitized data are passed to a fast readout board capable of handling several parallel 10Gb ethernet links. Experimental results of detector characterization with 400eV photons (at the I10 beamline of the DLS synchrotron ring) will be presented here. Beam images were acquired, through pinholes; diffraction pattern analysis confirmed that the detector response is dominated by low energy photons (400eV), rather than by higher harmonics components of the beam (which would be more likely to pass through an entrance window of inert material). Charge Collection Efficiency was evaluated (~85%), by comparing the collected charge with the estimation coming from a calibrated photodiode: System noise has been measured and its main contributors identified: a revised prototype is under manufacturing. Leakage current has also been measured, as well linearity of the detector response in low flux condition.

(11:10) N3B3-3, The Development of the DSSC Detector for the European XFEL: Toward the First Ladder Camera

M. Porro

Max Planck Institut fuer Extraterrestrische Physik, Garching, Germany

On behalf of the DSSC Collaboration

The DSSC collaboration is developing an X-ray camera for the European XFEL. Among the developments for the European XFEL, the DSSC will be the only 2D large-area high-speed detector able to achieve single photon resolution in the low energy range down to 0.25 keV. The camera is based on Si-sensors and is composed of 1024x1024 pixels for a total active area of 210 x 210 mm^2. 256 bump-bonded ASICs provide full parallel readout, comprising analog filtering, 8-bit digitization and data storage. The challenge of having high-dynamic range and single photon detection simultaneously requires a non-linear response of the system front-end. The DEPFET pixels, which we intend to use for the advanced version of the camera system, provide signal compression at the sensor level. For the simplified mini-SDD array that will be used to build the first mega-pixel camera for the day-zero of the European XFEL operation, the dynamic range compression is provided by the ASIC front-end. The very high frame rate and the low noise specifications imply considerable power consumption and sophisticated thermal designs have been conceived. The camera head electronics is able to cope with a total data transfer of 144 Gbit/s for the 1Mpixel device. After five years of development, all the components necessary to build the first detector submodule (128 x 512) are available and fully tested. We will give an overview of the DSSC system with its main components from the sensor to the daq electronics. The latest experimental results on the two type of sensors coupled with the prototype ASICs will be presented. Operating the system at 1 MHz, a noise down to 18 el. rms. and of 50 el. rms has been achieved with the DEPFET and with the mini-SDD sensor respectively. For the first time the DEPFET and the Mini-SDD solutions will be compared in terms of noise and dynamic range on the basis of experimental results. We will show the measured key characteristics of the first full-format 64 x 64 readout ASIC.

(11:30) N3B3-4, Validation of Proton Tests in Air for Detector Calibration over a Wide Range of Charge Injection Levels

A. Castoldi1,2, C. Guazzoni1,2, G. V. Montemurro1,2, L. Carraresi3,4, S. Schlee5, G. Weidenspointner5, M. Porro6

1Dip. Elettronica, Informazione e Bioingegneria, Politecnico di Milano, Milano, Italy
2Sezione di Milano, INFN, Milano, Italy
3Dip. Fisica e Astronomia, Università degli Studi di Firenze, Firenze, Italy
4sezione di Firenze, INFN, Firenze, Italy
5European XFEL, Hamburg, Germany
6Max-Planck-Institut für extraterrestrische Physik, Garching, Germany

We successfully evaluated the possibility of using a pulsed monoenergetic proton beam as a diagnostic tool for semiconductor detectors? response mapping at high charge densities. In order to ease the setup of the detector under test we explored the opportunity of performing tests with protons in air. We qualified a polyimide film window (Upilex-S, 7.5 ?m nominal thickness) as proton extraction window and the energy loss in air as a function of distance. The tests have been carried out in vacuum first, in order to evaluate the energy loss due to the window only, followed by in-air tests aimed at the investigation of the total energy degradation of the extracted proton beam

(11:50) N3B3-5, Design of the High Dynamic Range Pixel Array Detector (HDR-PAD)

J. T. Weiss1, K. S. Shanks1, H. T. Philipp1, J. Becker1, M. W. Tate1, S. M. Gruner1,2

1Department of Physics, Cornell University, Ithaca, NY, USA
2Cornell High Energy Synchrotron Source (CHESS), Cornell University, Ithaca, NY, USA

X-ray detector capabilities must advance in concert with synchrotron radiation light source technology to fully realize experimental possibilities. At present light source technology is outpacing detector development, and there is a dire need for new x-ray detectors capable of capturing wide dynamic range images [1]. X-ray free electron lasers (XFELs) place particularly demanding requirements on detectors, with x-rays being delivered in intense pulses of femtosecond durations. Global efforts are underway to meet these needs, and detectors capable of quantifying up to 103-105 photons in a single XFEL pulse are being developed [2, 3, 4, 5]. The High-Dynamic Range Pixel Array Detector (HDR-PAD) is being developed at Cornell with the goal of extending dynamic range to 106 photons/pixel in a single XFEL pulse while preserving single photon sensitivity. To satisfy detector needs at 3rd generation synchrotron sources, the detector will also be capable of tolerating a sustained flux of 1011 ph/pixel/s. These goals will be met by building on the architecture of the MM-PAD [6]. The MM-PAD uses a mixed analog and digital pixel to achieve a full well depth of 4x107 photons, but it is not suitable for use at XFELs. In addition to utilizing charge-removal techniques similar to those developed for the MM-PAD, plasma effects in silicon diodes are being exploited to delay the arrival of photocurrent, effectively stretching XFEL pulses to more manageable durations on the sensor level [7]. Our investigation of high charge carrier density plasmas in silicon sensors demonstrates potential benefits of this approach. We have completed extensive simulations investigating the most promising pixel architectures for the HDR-PAD. Our first pixel front-end prototypes have been fabricated. The HDR-PAD concept, design, and test results will be described.

(12:10) N3B3-6, AGIPD: a High Frame Rate Detector for the European XFEL

D. Mezza1, A. Allahgholi2, L. Bianco2, R. Dinapoli1, P. Goettlicher2, H. Graafsma2,3, D. Greiffenberg1, H. Hirsemann2, S. Jack2, R. Klanner4, A. Klyuev2, H. Krueger5, S. Lange2, A. Mozzanica1, A. Marras2, R. Seungyu2, B. Schmitt1, J. Schwandt4, I. Sheviakov2, X. Shi1, S. Smoljanin2, U. Trunk2, Q. Xia2, J. Zhang2, M. Zimmer2

1Paul Scherrer Institute, Villigen, Switzerland
2Deutsches Elektronensynchrotron DESY, Hamburg, Germany
3Mid Sweden University, Sundsvall, Sweden
4University of Hamburg, Hamburg, Germany
5University of Bonn, Bonn, Germany

The AGIPD (Adaptive Gain Integrating Pixel Detector) collaboration - consisting of Deutsches Elektronensysnchrotron (DESY), University of Hamburg, University of Bonn and the Paul-Scherrer-Institute (PSI) - is currently developing a 2D hybrid pixel detector system capable to fulfill the requirements of the European XFEL (Eu-XFEL) that is currently being built in Hamburg (Germany). At the Eu-XFEL photons will arrive in bunch trains every 100 ms (or at a rate of 10Hz). Each train consists of 2700 bunches that arrive within 600 µs (i.e. a bunch spacing of 220 ns, meaning 4.5 MHz frame rate) followed by 99.4 ms without pulses. Each single pulse consists of 10^12 X-ray photons arriving in less than 100 fs and in an energy range between 250 eV up to 25 keV. In order to cope with the large dynamic range, the first stage of each pixel in the AGIPD ASIC is a charge sensitive preamplifier with three different gain settings that are dynamically switched during charge integration. Dynamic gain switching allows to have single photon resolution in the high gain stage and to cover a large dynamic range of 10^4·12.4 keV photons in the low gain stage with a linearity better than 1%. The high frame rate (4.5 MHz) requires a storage of the signal in the pixel before the readout takes place during the gap between bunch trains. The full scale chip (AGIPD1.0), received at the end of 2013, is a 64 x 64 pixel matrix. Each pixel (Area 200 x 200 µm^2) is equipped with 352 storage cells. The single module system is composed by 8 x 2 AGIPD chips and the 1M system, that is actually in construction and is foreseen to be ready during summer, consists of 4 quadrants of 4 modules each, for a total of 16 modules. In this presentation a general overview of the AGIPD 1M system will be given. The focus will be on the characteristics of the ASIC including first experimental results. We will present the current status and give an overview over the foreseen upgrade of the readout chip.