1. Readout of a TPC with the MEDIPIX2 CMOS chip as direct anode Saclay/DapniaP. Colas Y. Giomataris NIKHEFA. Fornaini H. van der Graaf J. Timmermans J. Visschers CERN/MEDIPIX2E. Heijne Univ. Twente/MESA+J. Schmitz

2. Drift Space GEM foils MediPix CMOS pixel sensor Brass spacer block Printed circuit board Aluminium base plate Our GEM-equipped TPC We have constructed a small test TPC equipped with three GEM foils which can be read out by means of the MEDIPIX2 CMOS pixel sensor. MEDIPIX2 The GEM foils were obtained from the CERN/Sauli/GEM group; hole-to-hole distance (hexagonal geometry): 140 µm, hole diameter 85 µm, fiducial surface 100 mm x 100 mm, thickness 50 µm. The drift volume (vol. 100x100x100 mm 3 ) is surrounded by square wire loops, spaced 6.3 mm, put at decreasing potential. Three GEM foils are placed 7.4 mm behind the plane of the bottom wire loop; the distance between GEM foils is 1.6 mm. The anode plane, at ground potential, is 6.6 mm below the third GEM foil.

3. Top Drifter 5700 V Bottom Drifter 3473 V Top GEM 1 2813 V Bottom GEM 1 2462 V Top GEM 2 1876 V Bottom GEM 2 1524 V Top GEM 3 938 V Bottom GEM 3 586 V In the base plate of the chamber, a hole was cut out for the MEDIPIX2 chip: its pixel surface was flush with the (anode plane) base plate plane. The MEDIPIX2 chip contains 256 x 256 square pixels with pitch 55 µm x 55 µm giving a total fiducial sensitive area of 14.08 mm x 14.08 mm. Each pixel is equipped with a low-noise charge preamp, discriminator, two threshold DACs, a 13-bit counter and communication logic. Since a triggering system had not been implemented, we operated the MEDIPIX2 sensor by enabling the counters manually, and stop the counting after a pre-set time interval (0.1 - 10 s). After that, the counts of each pixel are read out. Drift length: 100 mm Distance between GEMs: 1.6/2.6 mm Distance bottom GEM/MEDIPIX: 6.6 mm

4. First events, recorded on March 29, 2003. Drift space irradiated with 55 Fe quanta

5. No source; exposed 0.01 sNo source; exposed 2 s No source; exposed 0.1 s 90 Sr source; exposed 0.01 s Fiducial field: 14 x 14 mm 2 Collected ionisation in 14 x 14 x 100 mm 3 during exposure time Feb 9, 2004

6. Dean Karlen: VRVS meeting Sept 17, 2003

7. August 2002 Jeju, Korea Dean Karlen: calculation of diffusion constants; Kirsten Sachs: measurment: offset 500  m in drift versus diffusion Sept 2003: VRVS meeting: Dean Karlen: Diffusion/defocussing too large at small drift distance Induced charge effect not ruled out

9. Anode strip Scope 55 Fe Test Pulse (edge)

10. Possible explanation for charge widening: Electrons end up in low E-field region Electrons are either ‘caught’ by drift field or diffuse to low field regions Electrons may migrate to regions associated with adjascent GEM holes

11. Conclusions - We have readout a TPC with triple GEM by means of a monolitic pixel chip as direct anode; - The distribution of the electrons leaving a GEM hole is much wider than the hole pitch. This may be explained by a large fraction of the avalanche not traveling straight to the anode, but arriving much later, after an effective de-focusing effect. - Due to the charge spread, the signal per pixel is very small, and the combination of three GEMs and the MediPix can not detect single primary electrons; - We expect much better results from a Micromegas/Pixel sensor combination

12. Thanks to: NIKHEF: Wim Gotink Joop Rovenkamp Jouri Ad Berkien, Frank de Zwart Edward Berbee, Ton Boerkamp Saclay: Arnoud Short term plans: - Micromegas/Pixel sensor tests; - Construction of Micromegas in post-wafer process; - definition of TimePix CMOS chip