of 13

Neutron sensitivity ofthin gap chambers

Nuclear Instruments and Methods in Physics Research A 543 (25) Neutron sensitivity ofthin gap chambers H. Nanjo, T. Bando, K. Hasuko, M. Ishino, T. Kobayashi, T. Takemoto,
0 views13 pages
All materials on our website are shared by users. If you have any questions about copyright issues, please report us to resolve them. We are always happy to assist you.
Documenttranscript
Nuclear Instruments and Methods in Physics Research A 543 (25) Neutron sensitivity ofthin gap chambers H. Nanjo, T. Bando, K. Hasuko, M. Ishino, T. Kobayashi, T. Takemoto, S. Tsuno, B. Ye 1 International Center for Elementary Particle Physics, University of Tokyo, Tokyo , Japan Received 25 November 24; accepted 26 November 24 Available online 19 February 25 Abstract Thin gap chambers (TGC) will be used for triggering forward muons in the ATLAS detector for the LHC at CERN. A large amount ofneutron background is foreseen in the ATLAS experiment. This paper describes the measurements ofthe neutron sensitivities (detection efficiencies) ofthe TGCs. The sensitivities ofboth small and real size TGCs to 2.5 and 14 MeV mono-energetic neutrons were measured. For a small size TGC, sensitivities of.32% and.1% were measured to 2.5 and 14 MeV neutrons, respectively, whereas for a real size TGC, sensitivities of.48% and.13% were measured. These measured values were in reasonably good agreement with the simulations based on the Geant4. r 25 Elsevier B.V. All rights reserved. PACS: 29.4.Cs Keywords: LHC; ATLAS; Trigger; Thin gap chamber; TGC; Neutron; Geant4 1. Introduction The ATLAS detector [1] is one ofthe major detectors for the future 14 TeV proton collider, the Large Hadron Collider (LHC) at the European Organization for Nuclear Research (CERN). The event rate ofthe ATLAS experiment is expected to be 1 GHz [2] for the designed luminosity of the Corresponding author. Tel.: ; fax: address: (H. Nanjo). 1 On leave from USTC, Hefei, People s Republic of China. LHC-1 34 cm 2 s 1 : The event trigger is one ofthe important issues for the experiment. Thin gap chambers (TGC) [3] will be used for triggering forward muons in the ATLAS detector. The structure oftgcs is similar to that ofmulti-wire proportional chambers and their detection efficiency for minimum ionizing particles (MIP) is more than 99% within a 25 ns time gate [4] (time duration ofthis time gate is referred to as time jitter ), that satisfies the requirements ofthe ATLAS muon triggering. A large amount ofbackground radiation is predicted in the ATLAS experiment. In the /$ - see front matter r 25 Elsevier B.V. All rights reserved. doi:1.116/j.nima 442 H. Nanjo et al. / Nuclear Instruments and Methods in Physics Research A 543 (25) installation area ofthe TGCs, neutrons and photons are the primary components ofthe background. This may induce a high counting rate in the TGCs, thereby affecting stable operation and causing false muon triggers or the chamber aging. In order to estimate such effects, the sensitivities oftgcs to such background particles must be measured. The primary energy range in the case ofthe photon background ranges from 1 kev to 1 MeV according to simulation [5], where photons are primarily generated through the capture ofthermal neutrons. The sensitivity was measured in the energy range from 2 kev to 1:8 MeV and was found to be less than 1% [6]. In the case ofthe neutron background, it originates from the interaction of primary hadrons with the materials ofthe ATLAS detector and accelerator elements. Its energy spectra ranges primarily from :25 ev to 1 GeV with a gentle peak around the 5 kev region obtained from the simulation [5]. Recoil nuclei or fragments from neutron reactions can produce hits in the TGC. Photons generated through neutron reactions can produce electrons that can also be the cause ofhits in the TGC. We performed the first measurements on the neutron sensitivity (detection efficiency) of TGCs for mono-energetic neutrons of 2.5 and 14 MeV: The results ofthe measurements were evaluated with a Monte Carlo simulation that was based on the Geant4 [7] and a good understanding ofthe TGC response to neutrons was obtained. 2. Real size TGC and small size TGC In this measurement, two types ofthe TGCs were used to get a better understanding through comparing both results. One was a real size TGC, the structure and materials ofwhich were identical to that ofthe TGCs that will be used in the ATLAS experiment. The other was a small size TGC that had a smaller and a simpler structure than the real size TGC. The structure ofthe real size TGC is described in Ref. [8]. The crosssections ofboth real and small size TGCs are shown in Fig. 1. The anodes are gold-plated tungsten wires 5 mm in diameter uniformly spaced at 1:8mm: The gap between the anodes and the cathode is 1:4mm: The cathode surface is made ofa conductive layer ofapproximately 1 mm in thickness, which primarily comprises graphite and acrylic resin in order to achieve a surface resistivity of approximately 1 MO=square: The chamber walls are made offr4 2. The TGC is operated in the limited proportional mode with a gas mixture ofco 2 and n-pentane, the ratio of which is 55:45. The real size TGC is trapezoidal in shape with a height of125 mm and a base length of1529 mm: Approximately 2 wires are grouped together in order to obtain 32 channels for the anode readouts. There are 32 rows ofcopper strips, each with the thickness of3 mm on the FR4 boards, which are perpendicular to the wires. Two chambers compose a double layer module (doublet) with a 2-mm thick paper honeycomb between them to maintain mechanical rigidness. In addition, 5-mm thick paper honeycombs with 5-mm thick FR4 skins are glued on both the outer surfaces for protection and rigidness. The small size TGC is 1 cm in width and 12 cm in length. It is a single layer chamber without cathode strip readouts. Its wire spacing, the wire diameter and the gap between the wire and the cathode are identical to that ofthe real size TGC. The thickness ofone side ofthe chamber wall is 1:6 mm and that ofthe other side is :2mm: The thickness ofthe copper cladding on the wall is 1 mm: There are 16 anode wires, each ofwhich is 8 cm in length. The signals generated at each wire are individually read. The two edge wires are not used in order to eliminate the effect of a higher electric field and a larger drift space corresponding to the edge wires. Accordingly, the sensitive area was 8 cm in length and 2:52 cm in width. 3. Experimental setups The geometrical and electrical setups for the measurements with both 2.5 and 14 MeV neutrons are described in this section. 2 Flame retardant glass fabric base epoxy-resin laminate. H. Nanjo et al. / Nuclear Instruments and Methods in Physics Research A 543 (25) real size TGC Gold-plated tungsten wire : 5µmφ Graphite+Acrylic resin surface (~1µm, ~1MΩ/square) Copper strip (3µm) FR4 skin (.5mm) mm neutron (a) Paper honycomb (5mm) Paper honycomb (2mm) FR4 wall Copper clad (3µm) small size TGC Copper clad(1µm) Copper clad(1µm) mm neutron Graphite+Acrylic resin surface (~1MΩ /square) 1.8mm spacing 16 wires 25.2mm for 14 wires 8 12mm FR4 wall 1.6mm FR4 wall.2mm (b) Graphite+Acrylic resin surface (~1µm, ~1MΩ /square) Fig. 1. The structure ofthe (a) real size TGC and (b) small size TGC Experimental setup for the measurements of the sensitivities to 2.5 MeV neutrons Mono-energetic neutrons with energies ofapproximately 2:5 MeV were produced through d + D reactions 3. A Cockcroft Walton type accelerator in the Rikkyo University 4 was used to generate 3 D and d represent deuterium and deuteron, respectively. 4 Rikkyo University Nishi-Ikebukuro, Toshima, Tokyo , Japan. 97:5 kev d þ ions. The ions were transported to a TiD 5 target, :5 mm in thickness, through a collimator (a 15-mm thick aluminum with a hole, the diameter ofwhich was 6 mm). At the target, the mono-energetic neutrons with an energy of approximately 2:5 MeV were produced through a Dðd; nþ 3 He reaction, d þ D! nþ 3 He þ 3:27 MeV. 5 Deuterium storage titanium. 444 H. Nanjo et al. / Nuclear Instruments and Methods in Physics Research A 543 (25) x y z small size TGC 9 cm TiD target.5mm d 14.5 cm 13.5 cm 6mmφcollimator 5µm thick Al.8µm thick Al foil 8 cm sensitive region (along wire) n 78 Neutron Window.5mm thick SUS 9 3 He Si PIN Photodiode Vacuum Chamber Wall 8mm thick SUS Fig. 2. The geometrical setup for the measurements of the sensitivities to 2:5 MeV neutrons. The small size TGC was placed at a distance of9 cm from the target with its 1.6-mm thick FR4 wall near the target. The real size TGC was placed at a distance of4 cm from the target. Neutrons can be tagged with 3 He nuclei generated at the same time. The geometrical setup around the target is shown in Fig. 2, where the x, y, andz coordinates are indicated. A Si PIN photodiode of 1cm 2 ; S359-2, fabricated by Hamamatsu Photonics, was placed at an angle of9 with respect to the d þ beam axis and at a distance of14:5 cm from the target to detect 3 He nuclei. A collimator of 5-mm thick aluminum with a hole 6 mm in diameter were placed in front of the photodiode. The collimator was positioned in order to define the direction ofthe 3 He nuclei. A.8-mm thick aluminum foil was also positioned in front of the photodiode to stop deuterons coming from the target through Rutherford scattering in the target. All the apparatuses mentioned above were placed inside a vacuum chamber connected to the beam line. When a 3 He nucleon was detected at an angle of 9 ; the energy of 3 He nucleon was 8 kev and the corresponding neutron was emitted at an angle of78 with its energy of2:57 MeV in agreement with the two-body kinematics. Due to the energy loss ofthe deuteron in the target and the geometrical acceptance, the energy ofthe neutron ranged from 2.45 to 2:62 MeV at most. The full energy spread ofthe neutron was less than 7% and the emitting angle ofthe neutron in the x z plane ranged from 93 to 65 at most. There was a D(d,p)t reaction 6 besides Dðd; nþ 3 He; d þ D! p þ t þ 4:3 MeV. According to the two-body kinematics, the proton energy was approximately 3:1 MeV whereas the triton energy was approximately 99 kev: Such protons and tritons could be rejected by applying cuts on the energy distribution measured with the photodiode. The photodiode energy calibration was performed using three types of a sources 239 Pu; 241 Am; and 244 Cm: The energy resolution of.2% around 5 MeV with a good linearity of.1% was obtained. On the opposite side ofthe photodiode, a small size TGC was placed outside the vacuum chamber at a distance of9 cm from the target, whereas a real size TGC was placed at a distance of4 cm from the target. The wall of the vacuum chamber was made ofstainless steel (SUS) and was 8 mm in thickness. There was a.5-mm thick SUS neutron window at the side ofthe wall facing the TGC. From all the particles produced in the d þ D reactions, only neutrons could enter the TGC, whereas the other particles were stopped at the vacuum chamber wall or the neutron window. The loss ofthe neutrons at the TiD target or the neutron window was negligible according to the 6 t represents triton. H. Nanjo et al. / Nuclear Instruments and Methods in Physics Research A 543 (25) Geant4 simulation and the loss was less than 5% according to the total cross-sections. The TGC was set as its wires ran parallel to the z-axis and were spaced along the y-axis. The position ofthe TGC was designed such that it covered the cone of the neutrons corresponding to the 3 He nuclei detected by the photodiode. Events due to the neutron incidence on the TGC (this implies that the neutron was emitted toward the TGC sensitive volume) could be selected with a 3 He hit on the photodiode. Events due to the neutron hit on the TGC (this implies that the neutron generated the hit signals ofthe TGC) could be distinguished with the coincidence ofa 3 He hit at the photodiode and a hit on the TGC. The electrical setup was designed to measure both the energy deposited in the photodiode with an peak hold ADC and time interval between the signal ofthe photodiode and that ofthe TGC with a TDC. The signal from the TGC was digitized with Amplifier-Shaper-Discriminators (ASD) [9] to supply the stop timing ofthe TDC. The signal from the photodiode was used for serving its charge, making the gate ofthe ADC and making the start timing ofthe TDC. In order to serve these functions, two amplifiers shaping amplifier (SA) and timing filter amplifier (TFA) were used after a pre-amplifier. The energy deposited in the photodiode was measured with the ADC using the signal from the SA. The coincidence timing was measured with the TDC which began by the signal from the TFA and halted by the signals from the TGC Experimental setup for the measurements of the sensitivities to 14 MeV neutron In the case of14 MeV neutrons, mono-energetic neutrons were produced through the Tðd; nþ 4 He reaction, d þ T! nþ 4 He þ 17:5 MeV. There were no other d þ T reactions except for the Rutherford scattering. The electrical setup was identical to that for the 2:5 MeV neutrons. The geometrical setup was slightly modified. A TiT 7 7 Tritium storage titanium. target, instead ofa TiD target, was used. The T emitted 18:6 kev electron through beta decay. In order to avoid a high counting rate and pileups due to the beta rays, a 1-mm thick gold foil, instead ofthe :8 mm thick aluminum foil, was placed in front of the photodiode. It also stopped deuterons that were produced through Rutherford scattering. The 3.5-MeV 4 He nuclei (a particles) were detected using the photodiode. The energy ofthe neutrons ranged from 14. to 14:2 MeV and the emitting angle ofthe neutrons ranged from 78 to 91 at most. The full energy spread was less than 2%. 4. Experimental results Using the experimental setups described in the previous section, the measurements ofthe detection efficiencies to both 2.5 and 14 MeV neutrons were performed for both the small size TGC and the real size TGC. The analyses ofthe data are described separately for 2.5 and 14 MeV neutrons with the small size TGC, followed by the analysis with the real size TGC. The systematic uncertainties ofthe measurements are described in the last subsection Sensitivity of the small size TGC to 2.5 MeV neutrons The analysis for the sensitivity of the small size TGC to 2:5 MeV neutrons is described in this subsection. The energy distribution measured with the photodiode is shown in Fig. 3. Three peaks that corresponded to the 8 kev 3 He nuclei, 99 kev triton, and 3:1 MeV proton are clearly seen. The decrease in their energy was primarily due to the energy loss in the.8-mm aluminum foil placed in front of the photodiode. Deuterons scattered in the target through Rutherford scattering were well stopped in the aluminum foil. In this energy distribution, events in the region from 1 to 7 kev were selected to obtain events with 3 He detected by the photodiode. This selection was referred to as loose 3 He selection. The timing distribution ofthe coincidence for the selected events is shown in Fig. 4 as an open 446 H. Nanjo et al. / Nuclear Instruments and Methods in Physics Research A 543 (25) He t p loose 3 He selection 1-7 kev Energy Loss in the Photodiode[keV] Fig. 3. The energy distribution measured with the Si PIN photodiode for the d þ D reaction. The energy range corresponding to the loose 3 He selection is indicated Side Bands CoincidenceTDC[ns] Fig. 4. The timing distributions ofthe coincidence within 4 ns. The open histogram corresponds to the events after loose 3 He selection and the hatched histogram corresponds to the events after tight 3 He selection that is indicated in Fig. 5. The lower regions ofthe histograms are magnified at the bottom. The side bands region (former 5 ns and the latter 2 ns) is indicated and the points with error bars correspond to events after tight 3 He selection in the side bands regions. This was fitted with a constant value to estimate the accidental coincidence. histogram. The peak ofthe coincidence is clearly seen. The timing resolution ofthe photodiode mainly contributed to the width ofthe peak, while the time jitter ofthe TGC is 25 ns: In particular, the broadening ofthe peak in the 5 ns region in Fig. 4 was due to the time walk ofthe signals from the photodiode to start the TDC. In Fig. 5, the energy distribution measured with the photodiode corresponding to the loose 3 He selection is shown as an open histogram. The energy distribution for the events with a coincidence TDC from to 4 ns is also shown as the hatched histogram. The ratio between them (coincidence ratio) is also plotted at the bottom. The decrease in the coincidence ratio in an energy region around 25 kev was due to the large time walk corresponding to the small 3 He signals, which delayed the TDC start and the coincidence between the photodiode and the TGC was missed. In order to avoid such an effect, an energy region from 275 to 475 kev was selected for further analysis. This selection was referred to as tight 3 He selection. The number ofevents within the energy region for the open histogram was also referred to as N neutron ; which implied the number ofneutrons generated due to 3 He nuclei detected with the photodiode. The timing distribution for the events obtained with the tight 3 He selection is shown as a hatched histogram in Fig. 4. The effect of the time walk was reduced with the tight energy selection. The number ofevents within the region from 5 to 2 ns (coincidence region) in the distribution was referred to as N coincidence : The side bands for the distribution (the region ofthe former 5 ns and the latter 2 ns in Fig. 4) were used to evaluate the accidental coincidence and the events in the side bands were fitted with a constant value. The fitted value was multiplied by the total bin number in the coincidence region to obtain the number ofthe accidental coincidences, N accidental : The detection efficiency was evaluated as follows: Efficiency ¼ ðn coincidence N accidental Þð1 x contamination Þ N neutron =ð1 f loss ÞZ coverage ð1þ where f loss was defined as the loss ofthe neutron flux at the target or the neutron window, Z coverage H. Nanjo et al. / Nuclear Instruments and Methods in Physics Research A 543 (25) Loose 3 He Selection Coincidence(-4[ns]) CoincidenceRatio[%] tight 3 He selection Energy Loss in the Photodiode[keV] Fig. 5. The energy distributions measured with the photodiode for the events with the loose 3 He selection (open histogram) and further with the coincidence within 4 ns (hatched one) are shown at the top. The coincidence ratio is also shown at the bottom. The energy range corresponding to the tight 3 He selection is indicated. was defined as the TGC coverage for the neutron flux and x contamination was defined as the contamination ofthe hits ofgammas in the coincidence region, which were produced through the reactions ofthe incident neutrons in the surrounding materials (the target chambers and concrete walls ofthe experimental area). The f loss was set to, as such loss was estimated to be negligible, as previously mentioned in Subsection 3.1. The hit wire distribution ofthe TGC was used for the estimation of Z coverage : The hit wire distribution, which corresponds to the events in the coincidence region is shown in Fig. 6. The distribution was well restricted in the sensitive region which was 2:52 cm in width. This was confirmed by an analysis ofother runs where the TGC was shifted to approximately 1 cm in the y direction as shown in Fig. 2. The distribution was Wire No.[channel] Fig. 6. The hit wire distribution ofthe TGC is shown. The distribution was fitted with Gaussian plus a constant value calculated from the event number of accidental coincidence. The width ofsensitive area that corresponded to the 14 channels was 2:52 cm: 448 H. Nanjo et al. / Nuclear Instruments and Methods in Physics Research A 543 (25) Table 1 The measured sensitivities and the results ofthe simulation are summarized Measurement (%) Simulation (%) Small size 2:5 MeV :32 :1ðstatÞ þ:3 :4 ðsysþ.35 Real size 2:5 MeV :48 :1ðstatÞ þ:3 :5 ðsysþ.39 Small size 14 MeV :1 :2ðstatÞ þ:1 :1 ðsysþ.11 Real size 14 MeV :13 :2ðstatÞ þ:2 :2 ðsysþ.15 fitted with a Gaussian plus constant value that was fixed at the value calculated from the number of accidental coincidences mentioned abov
Advertisement
We Need Your Support
Thank you for visiting our website and your interest in our free products and services. We are nonprofit website to share and download documents. To the running of this website, we need your help to support us.

Thanks to everyone for your continued support.

No, Thanks
SAVE OUR EARTH

We need your sign to support Project to invent "SMART AND CONTROLLABLE REFLECTIVE BALLOONS" to cover the Sun and Save Our Earth.

More details...

Sign Now!

We are very appreciated for your Prompt Action!

x