- Email: [email protected]
Results on diamond timing detector for the TOTEM experiment
Nuclear Instruments and Methods in Physics Research A ∎ (∎∎∎∎) ∎∎∎–∎∎∎
Contents lists available at ScienceDirect
Nuclear Instruments and Methods in Physics Research A journal homepage: www.elsevier.com/locate/nima
Results on diamond timing detector for the TOTEM experiment E. Bossini a,b,n, On behalf of the TOTEM Collaboration a b
Museo Storico della Fisica e Centro Studi e Ricerche E. Fermi, Rome, Italy Istituto Nazionale di Fisica Nucleare, Sezione di Pisa, Italy
art ic l e i nf o
a b s t r a c t
Keywords: Diamond detectors TOF Low-noise electronics
We describe the results and status of our R&D on diamond timing detectors for the TOTEM experiment at CERN. Tests with commercial devices have been done and here reported; the unsatisfactory results push us to design a new detector. We present test beams results and the front-end electronics, critical point of the design. Efficiency studies and timing performance dependence from detector capacitance will be also reported. & 2015 Published by Elsevier B.V.
1. Introduction Physics program of the TOTEM experiment at LHC will require precise time of flight detectors with a resolution better than 50 ps to be installed in the TOTEM Vertical Roman Pots (RPs) [1]. Moreover since the detector will work at few mm from the beam, high radiation hardness and low material are required. Efficiency close to unity is also needed. To achieve such result we exploited ultra pure sCVD diamond technology, which provides fast, low noise and radiation hard devices. Moreover, since one RP can host up to 4 diamond detection planes, timing constraints are relaxed by a factor 2. Although diamond noise is practically zero the output signal from MIP particles is very low (2 fC/particle), making the first stage of amplification critical.
2. Off-shelf devices First test with commercial detectors and electronic were performed at Paul Scherrer Institute (PSI, Zurich) in June 2014. PSI beam was made of a mixture of electrons, pions and muons with a selected momentum of 250 MeV/c2. Particle discrimination was performed through the particles TOF from the generation target to the detector. Trigger system was made of two plastic scintillator fingers (3 3 50 mm3) coupled to Hamamatsu SiPM. The trigger time resolution, below 1 ns, was enough for TOF calculation and discrimination. The detector under test were two ultra-pure sCVD diamond with a surface of 4.5 4.5 mm2 and 500 μm thickness from CIVIDEC. Signal amplification was provided by CIVIDEC fast charge amplifier (10 ns shaper, 750 ENC). A broadband current amplifier (2 GHz BW) was also tested with worse results. The
signal acquisition was performed with Agilent Infiniium Oscilloscope (2.5 GHz; 20 GSa/s) and offline analysis with different algorithm were used (fixed threshold, rising edge fit, CFD,…). Time resolution was obtained measuring the time difference between the two identical diamond sensor. Best result were obtained with CFD, but the detector resolution was found 4 300 ps. Same daq, offline algorithm and trigger have been used in later test beams.
3. Hybrid development and tests As mentioned before, diamond detector signal from MIP particle is 2 fC/particle, and thus dedicated very-low noise electronics must be developed to achieve better results. Capacitance of the detector is also critical and must be kept as low as possible. We moved from the work of the HADES collaboration [2] to develop a new Hybrid board (Fig. 1). The first prototype was successfully tested at SPS (CERN), with 250 GeV/c2 particles. The diamond hosted by the board can be metalized in up to 4 strips with a full amplification chain, described later, for each channel. To reduce parasitic capacitance pre-amplifiers are placed at 1 cm from the diamond. Channels are readout by SMA cable, while HV(up to 1 kV) and LV are provided through LEMO connectors. The core area can be shielded with aluminium boxes to reduce EM interferences. Results are showed in Fig. 2. The 90 ps result as been obtained making the time difference between one of our strips and the detector from the HADES collaboration. Resolution of the HADES detector as been previously measured using two identical detector. The result obtained was confirmed below the desired 100 ps even using different offline algorithm. 3.1. Amplification chain Signal amplification is performed through a three stage electronics.
n
Correspondence address: Museo Storico della Fisica e Centro Studi e Ricerche E. Fermi, Rome, Italy. E-mail address: [email protected]
Pre-Amplifier: 1 stage BFP840 SiGe BJT with low capacitive feedback (31 dB).
http://dx.doi.org/10.1016/j.nima.2015.10.099 0168-9002/& 2015 Published by Elsevier B.V.
Please cite this article as: E. Bossini, Nuclear Instruments & Methods in Physics Research A (2015), http://dx.doi.org/10.1016/j. nima.2015.10.099i
E. Bossini / Nuclear Instruments and Methods in Physics Research A ∎ (∎∎∎∎) ∎∎∎–∎∎∎
2
Fig. 4. Detector timing performance wrt strip capacitance.
Fig. 1. Hybrid board. General overview (bottom) and core detail (top). The Diamond in divided in four strips of different area.
Global signal amplification was of 93 dB. Output signals from MIP have a mean SNR of 25 and a rise time of 1.3 ns.
4. Latest results Prototype final validation has been performed in march 2015 at DESY (Hamburg), using electrons in the energy range 4–5.6 Mev/ c2. A tracker detector (DATURA telescope [3]), with 5 μm resolution was available on site and used to carry out efficiency studies, which are reported in Fig. 3. During efficiency measurements we used a low intensity beam to avoid pile up since the tracker integrate all tracks in a 112 μs window. Efficiency have been found 4 98%, also in the inter-strip area. Relation of detector capacity to time performance has also been investigated selecting coincidence between different area strips (alias detector capacitance) (Fig. 4). Fig. 2. Time resolution between HADES detector and TOTEM hybrid equipped with single pixel diamond.
5. Conclusion A prototype of the diamond timing detector has been designed and tested. The result obtained, below 100 ps, validate the electronic design. Timing performance dependence from the strip area has been investigated and very satisfactory results for efficiency has been proved. Final board, which will host 4 diamonds and 12 channels, will be tested during next months.
References [1] The TOTEM Collaboration, TOTEM upgrade proposal, CERN-LHCC-2013-009/ LHCC-P-007, 2013. [2] J. Pietrasko, L. Fabbietti, W. Koenig, M. Weberc, for HADES Collaboration, Nuclear Instruments and Methods A 618 (2010) 121. [3] 〈https://twiki.cern.ch/twiki/bin/view/MimosaTelescope/WebHome〉.
Fig. 3. Strip efficiency using DATURA tracker.
Amplifier: Monolithic microwave IC ABA-53563, near linear phase, absolute stable amplifier (22 dB).
Shaper: signal attenuator ( 10 dB) followed by two BFG425 in cascade Si BJT (50 dB)
Please cite this article as: E. Bossini, Nuclear Instruments & Methods in Physics Research A (2015), http://dx.doi.org/10.1016/j. nima.2015.10.099i
Contents lists available at ScienceDirect
Nuclear Instruments and Methods in Physics Research A journal homepage: www.elsevier.com/locate/nima
Results on diamond timing detector for the TOTEM experiment E. Bossini a,b,n, On behalf of the TOTEM Collaboration a b
Museo Storico della Fisica e Centro Studi e Ricerche E. Fermi, Rome, Italy Istituto Nazionale di Fisica Nucleare, Sezione di Pisa, Italy
art ic l e i nf o
a b s t r a c t
Keywords: Diamond detectors TOF Low-noise electronics
We describe the results and status of our R&D on diamond timing detectors for the TOTEM experiment at CERN. Tests with commercial devices have been done and here reported; the unsatisfactory results push us to design a new detector. We present test beams results and the front-end electronics, critical point of the design. Efficiency studies and timing performance dependence from detector capacitance will be also reported. & 2015 Published by Elsevier B.V.
1. Introduction Physics program of the TOTEM experiment at LHC will require precise time of flight detectors with a resolution better than 50 ps to be installed in the TOTEM Vertical Roman Pots (RPs) [1]. Moreover since the detector will work at few mm from the beam, high radiation hardness and low material are required. Efficiency close to unity is also needed. To achieve such result we exploited ultra pure sCVD diamond technology, which provides fast, low noise and radiation hard devices. Moreover, since one RP can host up to 4 diamond detection planes, timing constraints are relaxed by a factor 2. Although diamond noise is practically zero the output signal from MIP particles is very low (2 fC/particle), making the first stage of amplification critical.
2. Off-shelf devices First test with commercial detectors and electronic were performed at Paul Scherrer Institute (PSI, Zurich) in June 2014. PSI beam was made of a mixture of electrons, pions and muons with a selected momentum of 250 MeV/c2. Particle discrimination was performed through the particles TOF from the generation target to the detector. Trigger system was made of two plastic scintillator fingers (3 3 50 mm3) coupled to Hamamatsu SiPM. The trigger time resolution, below 1 ns, was enough for TOF calculation and discrimination. The detector under test were two ultra-pure sCVD diamond with a surface of 4.5 4.5 mm2 and 500 μm thickness from CIVIDEC. Signal amplification was provided by CIVIDEC fast charge amplifier (10 ns shaper, 750 ENC). A broadband current amplifier (2 GHz BW) was also tested with worse results. The
signal acquisition was performed with Agilent Infiniium Oscilloscope (2.5 GHz; 20 GSa/s) and offline analysis with different algorithm were used (fixed threshold, rising edge fit, CFD,…). Time resolution was obtained measuring the time difference between the two identical diamond sensor. Best result were obtained with CFD, but the detector resolution was found 4 300 ps. Same daq, offline algorithm and trigger have been used in later test beams.
3. Hybrid development and tests As mentioned before, diamond detector signal from MIP particle is 2 fC/particle, and thus dedicated very-low noise electronics must be developed to achieve better results. Capacitance of the detector is also critical and must be kept as low as possible. We moved from the work of the HADES collaboration [2] to develop a new Hybrid board (Fig. 1). The first prototype was successfully tested at SPS (CERN), with 250 GeV/c2 particles. The diamond hosted by the board can be metalized in up to 4 strips with a full amplification chain, described later, for each channel. To reduce parasitic capacitance pre-amplifiers are placed at 1 cm from the diamond. Channels are readout by SMA cable, while HV(up to 1 kV) and LV are provided through LEMO connectors. The core area can be shielded with aluminium boxes to reduce EM interferences. Results are showed in Fig. 2. The 90 ps result as been obtained making the time difference between one of our strips and the detector from the HADES collaboration. Resolution of the HADES detector as been previously measured using two identical detector. The result obtained was confirmed below the desired 100 ps even using different offline algorithm. 3.1. Amplification chain Signal amplification is performed through a three stage electronics.
n
Correspondence address: Museo Storico della Fisica e Centro Studi e Ricerche E. Fermi, Rome, Italy. E-mail address: [email protected]
Pre-Amplifier: 1 stage BFP840 SiGe BJT with low capacitive feedback (31 dB).
http://dx.doi.org/10.1016/j.nima.2015.10.099 0168-9002/& 2015 Published by Elsevier B.V.
Please cite this article as: E. Bossini, Nuclear Instruments & Methods in Physics Research A (2015), http://dx.doi.org/10.1016/j. nima.2015.10.099i
E. Bossini / Nuclear Instruments and Methods in Physics Research A ∎ (∎∎∎∎) ∎∎∎–∎∎∎
2
Fig. 4. Detector timing performance wrt strip capacitance.
Fig. 1. Hybrid board. General overview (bottom) and core detail (top). The Diamond in divided in four strips of different area.
Global signal amplification was of 93 dB. Output signals from MIP have a mean SNR of 25 and a rise time of 1.3 ns.
4. Latest results Prototype final validation has been performed in march 2015 at DESY (Hamburg), using electrons in the energy range 4–5.6 Mev/ c2. A tracker detector (DATURA telescope [3]), with 5 μm resolution was available on site and used to carry out efficiency studies, which are reported in Fig. 3. During efficiency measurements we used a low intensity beam to avoid pile up since the tracker integrate all tracks in a 112 μs window. Efficiency have been found 4 98%, also in the inter-strip area. Relation of detector capacity to time performance has also been investigated selecting coincidence between different area strips (alias detector capacitance) (Fig. 4). Fig. 2. Time resolution between HADES detector and TOTEM hybrid equipped with single pixel diamond.
5. Conclusion A prototype of the diamond timing detector has been designed and tested. The result obtained, below 100 ps, validate the electronic design. Timing performance dependence from the strip area has been investigated and very satisfactory results for efficiency has been proved. Final board, which will host 4 diamonds and 12 channels, will be tested during next months.
References [1] The TOTEM Collaboration, TOTEM upgrade proposal, CERN-LHCC-2013-009/ LHCC-P-007, 2013. [2] J. Pietrasko, L. Fabbietti, W. Koenig, M. Weberc, for HADES Collaboration, Nuclear Instruments and Methods A 618 (2010) 121. [3] 〈https://twiki.cern.ch/twiki/bin/view/MimosaTelescope/WebHome〉.
Fig. 3. Strip efficiency using DATURA tracker.
Amplifier: Monolithic microwave IC ABA-53563, near linear phase, absolute stable amplifier (22 dB).
Shaper: signal attenuator ( 10 dB) followed by two BFG425 in cascade Si BJT (50 dB)
Please cite this article as: E. Bossini, Nuclear Instruments & Methods in Physics Research A (2015), http://dx.doi.org/10.1016/j. nima.2015.10.099i