Cosmic Ray Tests of the Prototype TPC for the ILC Experiment
K. Ackermann, S. Arai, D. C. Arogancia, A. M. Bacala, M. Ball, T. Behnke, H. Bito, V. Eckardt, K. Fujii, T. Fusayasu, N. Ghodbane, H. C. Gooc Jr., T. Kijima, M. Hamann, M. Habu, R. -D. Heuer, K. Hiramatsu, K. Ikematsu, A. Kaukher, H. Kuroiwa, M.E. Janssen, Y. Kato, M. Kobayashi, T. Kuhl, T. Lux, T. Matsuda, S. Matsushita, A. Miyazaki, K. Nakamura, O. Nitoh, H. Ohta, R. L. Reserva, K. Sakai, N. Sakamoto, T. Sanuki, R. Settles, A. Sugiyama, T. Takahashi, T. Tomioka, H. Tsuji, T. Watanabe, P. Wienemann, R. Wurth, H. Yamaguchi, M. Yamaguchi, A. Yamaguchi, T. Yamamura, H. Yamaoka, T. Yazu, R. Yonamine
aa r X i v : . [ phy s i c s . i n s - d e t ] M a y Cosmic Ray Tests of the Prototype TPC for the ILC Experiment
K. Ackermann a , S. Arai b , D.C. Arogancia c , A.M. Bacala c , M. Ball d , T. Behnke d , H. Bito b , V. Eckardt a , K. Fujii e ,T. Fusayasu f , N. Ghodbane d , H.C. Gooc Jr. c , T. Kijima b , M. Hamann d , M. Habu b , R.-D. Heuer g , K. Hiramatsu h ,K. Ikematsu e , A. Kaukher i , H. Kuroiwa j , M.E. Janssen d , Y. Kato h , M. Kobayashi ∗ ,e , T. Kuhl d , T. Lux g , T. Matsuda e ,S. Matsushita b , A. Miyazaki b , K. Nakamura b , O. Nitoh b , H. Ohta b , R.L. Reserva c , K. Sakai b , N. Sakamoto j ,T. Sanuki k , R. Settles a , A. Sugiyama j , T. Takahashi l , T. Tomioka b , H. Tsuji j , T. Watanabe m , P. Wienemann d ,R. Wurth d , H. Yamaguchi j , M. Yamaguchi j , A. Yamaguchi n , T. Yamamura k , H. Yamaoka e , T. Yazu h , R. Yonamine o a Max-Planck-Institute for Physics, Munich, Germany b Tokyo University of Agriculture and Technology, Tokyo, Japan c Mindanao State University, Iligan City, Philippines d DESY, Hamburg, Germany e KEK, IPNS, Ibaraki, Japan f Nagasaki Institute of Applied Science, Nagasaki, Japan g Universit¨at Hamburg, Hamburg, Germany h Kinki University, Osaka, Japan i Universit¨at Rostock, Rostock, Germany j Saga University, Saga, Japan k University of Tokyo, ICEPP, Tokyo, Japan l Hiroshima University, Hiroshima, Japan m Kogakuin University, Tokyo, Japan n University of Tsukuba, Ibaraki, Japan o Graduate University for advanced Studies (KEK), Ibaraki, Japan
Abstract
A time projection chamber (TPC) is a strong candidate for the central tracker of the international linear collider(ILC) experiment and we have been conducting a series of cosmic ray experiments under a magnetic field up to 4 T, usinga small prototype TPC with a replaceable readout device: multi-wire proportional chamber (MWPC) or gas electronmultiplier (GEM). We first confirmed that the MWPC readout could not be a fall-back option of the ILC-TPC undera strong axial magnetic field of 4 T since its spatial resolution suffered severely from the so called E × B effect in thevicinity of the wire planes. The GEM readout, on the other hand, was found to be virtually free from the E × B effectas had been expected and gave the resolution determined by the transverse diffusion of the drift electrons (diffusionlimited). Furthermore, GEMs allow a wider choice of gas mixtures than MWPCs. Among the gases we tried so far amixture of Ar-CF -isobutane, in which MWPCs could be prone to discharges, seems promising as the operating gas ofthe ILC-TPC because of its small diffusion constant especially under a strong magnetic field. We report the measureddrift properties of this mixture including the diffusion constant as a function of the electric field and compare themwith the predictions of Magboltz. Also presented is the spatial resolution of a GEM-based ILC-TPC estimated from themeasurement with the prototype. Key words:
ILC, TPC, MWPC, GEM, CF , Spatial resolution PACS:
1. Introduction
One of the most important issues of the current highenergy physics is to find the Higgs boson and to reveal itsnature. The LHC will most likely find a Higgs candidateand the ILC is expected to follow it to complete the mis-sion [1][2]. To study the Higgs properties in detail we needa high performance central tracker. ∗ Corresponding author. Tel.: +81 29 864 5379; fax: +81 29 8642580.
Email address: [email protected] (M. Kobayashi)
A TPC is a natural candidate for the ILC central trackerbecause of its very good performance in the past colliderexperiments [3]. At the ILC, however, we need the highestpossible tracking efficiency in a jetty environment and amomentum resolution one order-of-magnitude better thanthose in the past. This requires very high 3-dimensionalgranularity (a small voxel size) and a space point resolu-tion in the r − φ plane better than 100 µ m throughout thesensitive volume. In order to realize a TPC with such un-precedented performance, intense R&D programs are nowon going in an international framework called the LC-TPCcollaboration. Three technologies have been considered as Preprint submitted to TIPP09 Proceedings in NIMA November 12, 2018 he gas amplification device for the ILC-TPC: MWPC,MicroMEGAS [4] and GEM [5].In order to compare the different readout technolo-gies, we prepared a small prototype TPC with a replace-able readout plane consisting of a gas amplification deviceand a pad plane. The results of the beam test at KEKfor a MicroMEGAS readout using a gas mixture of Ar-isobutane(5%) have already been published [6]. In thispaper our results of cosmic ray tests for an MWPC read-out using a TDR gas (Ar-CH (5%)-CO (2%)) [1] and for atriple GEM readout operated in a T2K gas (Ar-CF (3%)-isobutane(2%)) [7] are presented.
2. Experimental Setup
The GEM-based prototype is described in Ref. [8], wherethe preliminary results of the beam test for the GEM read-out using a TDR gas or a P5 gas (Ar-CH (5%)) are pre-sented. The sensitive volume of the prototype was ∼ ×
100 mm. The induction gap ofthe triple GEM stack was increased later from 1 mm to1.5 mm for the cosmic ray tests using a T2K gas, in orderto increase the charge spread after gas amplification, whilethe transfer gaps (1.5 mm) were maintained. A picture ofthe prototype is shown in Fig. 1. (cid:1)(cid:0)(cid:2)(cid:3)(cid:4)(cid:5)(cid:6)(cid:7)(cid:8)(cid:9)(cid:10)(cid:11)(cid:12)(cid:13)(cid:14)(cid:15)(cid:16)(cid:17)(cid:18)(cid:19)(cid:20)(cid:21)(cid:22)(cid:23)(cid:24)(cid:25)(cid:26)(cid:27) (cid:28)(cid:29)(cid:30)(cid:31) !"
Figure 1: Photograph of the prototype just before installation intothe gas vessel.
For the MWPC readout the GEM foils were replacedby two wire planes [9]. The sense-wire plane consisted of 20 µ m thick wires spaced with 2 mm pitch and was placed 1mm above the readout pad plane, on which pad rows werearranged in parallel to the wires. The drift region was de-limited by a grid wire plane. The grid wire plane consistedof 127 µ m thick wires strung 1 mm above the sense-wireplane with 2 mm spacing. It should be noted that thesense-wire pitch and the wires-to-pads distance were keptsmall as compared to those in conventional TPCs in orderto obtain the best spatial resolution and the highest two-track resolving power achievable with a practical MWPCreadout . The pad pitch was increased to 2.3 mm, from The width of the pad response function ( σ ) was about 1.3 mm. , correspondingto the increase in the width of the pad response functionwhile the pad-row pitch (6.3 mm) was retained.The MWPC-TPC was operated inside a 5-T solenoidat DESY and the GEM-TPC data was taken inside a 1-T solenoid at the KEK cryogenic center. A pair of longscintillation counters was used for cosmic ray triggers. Thereadout electronics system for the ALEPH TPC [10] wasused for the chamber signal processing and the traditionalMultiFit (DoubleFit) program [11] was used for the dataanalysis of the cosmic ray events.The hit coordinate along a pad row was determined bya simple charge centroid (barycenter) method using thepad signals within a charge cluster in the row. Amongthe tracks recorded in the TPC, those which were nearlyperpendicular to the pad rows (sense wires) and parallelto the readout plane were selected for later analysis.
3. MWPC Readout
The spatial resolution as a function of the drift distance( z ) obtained under a magnetic field ( B ) of 4 T is shown inFig. 2. The sense-wire potential was set to 1200 V whilethe grid wires were grounded. The drift field strength ( E d )was 220 V/cm, at which the transverse diffusion constant( D ) of drift electrons is about 60 µ m/ √ cm. Also shown inthe figure is the result of a Monte Carlo simulation. Thesimulation took into account the primary ionization statis-tics of incident tracks, the diffusion of drift electrons, thelateral displacement of the drift electrons (charge spread)near the wire planes due to the E × B effect, and finallythe avalanche fluctuation in gas amplification. Figure 2: Spatial resolution measured with the MWPC readout in amagnetic field of 4 T. The solid line shows the result of the simulation.
The simulation program Garfield/Magboltz [12] wasused to estimate the displacement of the drift electronsalong the wire direction ( dx ) due to the E × B effectas a function of the initial position along the track ( y ) The neighboring pad rows were staggered by half a pad pitch inthe GEM readout. E × B effect at short driftdistances and its reduction with increasing z due to thedeclustering effect caused by the diffusion of the drift elec-trons [13]. Figure 3: Simulated drift electron paths without diffusion in TDRgas for a track perpendicular to the wires ( B = 4 T).Figure 4: Displacement of the arrival position of drift electrons on thesense wire as a function of the initial position along the track for B =1, 2, 3 and 4 T. The coordinate y is measured from the point on thetrack right above the sense wire. The increase of displacement ( dx ),especially for large y , degrades the spatial resolution significantly. Fig. 5 shows the simulated spatial resolution of theILC-TPC equipped with the MWPC readout for B = 0,1 and 4 T. The resolution for B = 4 T is dominated bythe E × B effect throughout the drift region in spite of itsdecrease at long drift distances where the contribution ofthe electron diffusion increases. The situation is worse forT2K gas because of the larger charge spread near the wireplanes caused by the E × B effect and the smaller declus-tering effect due to its small diffusion constant ( D ∼ µ m/ √ cm at E d = 250 V/cm and B = 4 T). Therefore T2Kgas would not be a good choice for MWPC readout evenif the MWPC itself operated properly in this gas mixture. Figure 5: Simulated spatial resolution of the ILC-TPC equippedwith the MWPC readout operated in TDR gas (left) and in T2K gas(right) for tracks perpendicular to the wires and the pad rows.
4. GEM readout
The spatial resolutions measured under magnetic fieldsof 0 and 1 T are shown in Fig. 6. The high voltage appliedacross each GEM of the triple stack was 250 V and theelectric fields in the transfer and induction gaps were setto 1.6 kV/cm. The solid curves in the figure are the resultsof analytic calculation [6]. The resolution deterioration atshort drift distances prominent for the T2K gas at 1 T isnot caused by an E × B effect but by the finite pitch ofthe readout pads (1.27 mm), which is not small enoughcompared to the charge spread after gas amplification inthe GEM stack ( ∼ µ m) [6]. Figure 6: Spatial resolution as a function of the drift distance ob-tained with a T2K gas ( E d = 250 V/cm) for B = 0 and 1 T. Theresolutions obtained in the beam test with a P5 gas ( E d = 100 V/cm)are also shown for comparison. The spatial resolution ( σ X ) at long drift distances isgiven by σ = σ + D /N eff · z , σ X0 is the intrinsic resolution and N eff is the ef-fective number of electrons, which is determined by theprimary ionization statistics and the avalanche fluctuationfor a single drift electron [14].Since the diffusion constant is determined from the z -dependence of the pad response width [6][8] (or given byMagboltz) one can evaluate N eff from the z -dependence ofthe spatial resolution at long drift distances. The value of N eff thus obtained was about 21, which is in good agree-ment with a simple estimate of 22.2 [14], indicating thatpossible dissociative attachment of drift electrons by CF molecules [15][16][17] at the entrance to the GEM stack isnot significant if any.We also measured the diffusion constant and the driftvelocity in a T2K gas as functions of the drift field, cover-ing from the diffusion minimum ( E d = 100 V/cm) to thedrift velocity near-maximum ( E d = 250 V/cm). The driftvelocity was estimated from the full width of the drift timedistribution and the maximum drift length of the proto-type. The results are shown in Fig. 7 along with the cor-responding Magboltz predictions, which closely reproducethe data points. The reliability of Magboltz for this gasmixture was thus confirmed for the practical range of theelectric field for TPC operation. Figure 7: Transverse diffusion constants ( D T ) and drift velocities( W ) measured in T2K gas. The curves represent the Magboltz pre-dictions. . Fig. 8 shows the calculated resolution of a GEM-basedILC-TPC operated in a T2K gas, assuming the diffusionconstant given by Magboltz for E d = 250 V/cm.
5. Conclusion
MWPC readout turned out to be inappropriate for aTPC operated in a strong magnetic field because of a large E × B effect on its spatial resolution. Therefore it can notbe a fall-back option for the ILC-TPC. A GEM-based TPCusing a T2K gas is a good candidate for the central trackerof the ILC experiment. It operates stably and is expected Figure 8: Expected resolution of a GEM-based TPC operated inT2K gas for tracks perpendicular to the pad rows. to give the required spatial resolution of better than 100 µ m throughout the sensitive volume under a 4 T magneticfield. Acknowledgments
We are grateful to the KEK cryogenic center and theIPNS cryogenic group for the operation of the supercon-ducting magnet. This study is supported in part by theCreative Scientific Research Grant no. 18GS0202 of theJapan Society for Promotion of Science.
References [1] TESLA Technical Design Report, DESY 2001-011.[2] GLC Project Report, KEK Report 2003-7.[3] M.T. Ronan, Time-projection chambers, Review of ParticlePhysics, Phys. Lett. B592, 1 (2004) 261.[4] Y. Giomataris, et al., Nucl. Instr. and Meth. A 376 (1996) 29.[5] F. Sauli, Nucl. Instr. and Meth. A 386 (1997) 531.[6] D.C. Arogancia, et al., Nucl. Instr. and Meth. A 602 (2009) 403.[7] The T2K ND280 TPC Group, T2K ND280 TPC Technical De-sign Report (T2K Internal Document), January 2, 2007.[8] M. Kobayashi, et al., Nucl. Instr. and Meth. A 581 (2007) 265.[9] K. Ackermann, et al., papers on the MWPC and GEM readout,in preparation for publication in Nucl. Instr. and Meth. A.[10] M. Ball, N. Ghodbane, M. Janssen and P. Wienemann, ADAQ System for Linear Collider TPC Prototypes based on theALEPH TPC Electronics, arXiv:physics/0407120.[11] Matthias E. Janssen (on behalf of the FLC-TPC group atDESY), R&D Ongoing at DESY for a GEM based TPC at theILC: Resolution Studies; Techniques and Results, IEEE NuclearScience Symposium Conference Record (2006) 38, and refer-ences cited therein.[12] S.F. Biagi, Nucl. Instr. and Meth. A 421 (1999) 234.[13] W. Blum, U. Stiegler, P. Gondolo and L. Rolandi, Nucl. Instr.and Meth. A 252 (1986) 407.[14] M. Kobayashi, Nucl. Instr. and Meth. A 562 (2006) 136.[15] L.G. Christophorou, D.L. McCorkle, D.V. Maxey andJ.G. Carter, Nucl. Instr. and Meth. 163 (1979) 141.[16] L.G. Christophorou, P.G. Datskos and J.G. Carter, Nucl. Instr.and Meth. A 309 (1991) 160.[17] W.S. Anderson, et al., Nucl. Instr. and Meth. A 323 (1992) 273.[1] TESLA Technical Design Report, DESY 2001-011.[2] GLC Project Report, KEK Report 2003-7.[3] M.T. Ronan, Time-projection chambers, Review of ParticlePhysics, Phys. Lett. B592, 1 (2004) 261.[4] Y. Giomataris, et al., Nucl. Instr. and Meth. A 376 (1996) 29.[5] F. Sauli, Nucl. Instr. and Meth. A 386 (1997) 531.[6] D.C. Arogancia, et al., Nucl. Instr. and Meth. A 602 (2009) 403.[7] The T2K ND280 TPC Group, T2K ND280 TPC Technical De-sign Report (T2K Internal Document), January 2, 2007.[8] M. Kobayashi, et al., Nucl. Instr. and Meth. A 581 (2007) 265.[9] K. Ackermann, et al., papers on the MWPC and GEM readout,in preparation for publication in Nucl. Instr. and Meth. A.[10] M. Ball, N. Ghodbane, M. Janssen and P. Wienemann, ADAQ System for Linear Collider TPC Prototypes based on theALEPH TPC Electronics, arXiv:physics/0407120.[11] Matthias E. Janssen (on behalf of the FLC-TPC group atDESY), R&D Ongoing at DESY for a GEM based TPC at theILC: Resolution Studies; Techniques and Results, IEEE NuclearScience Symposium Conference Record (2006) 38, and refer-ences cited therein.[12] S.F. Biagi, Nucl. Instr. and Meth. A 421 (1999) 234.[13] W. Blum, U. Stiegler, P. Gondolo and L. Rolandi, Nucl. Instr.and Meth. A 252 (1986) 407.[14] M. Kobayashi, Nucl. Instr. and Meth. A 562 (2006) 136.[15] L.G. Christophorou, D.L. McCorkle, D.V. Maxey andJ.G. Carter, Nucl. Instr. and Meth. 163 (1979) 141.[16] L.G. Christophorou, P.G. Datskos and J.G. Carter, Nucl. Instr.and Meth. A 309 (1991) 160.[17] W.S. Anderson, et al., Nucl. Instr. and Meth. A 323 (1992) 273.