Nuclear Instruments and Methods in Physics Research A 348 (1994) 600-606 North-Holland NUCLEAR INSTRUMENTS & METHODS IN PHYSICS RESEARCH SeCtFOn A Operation of a BaF2-TMAE prototype detector K. Wells a,,, D. Visvikis a, R.J. Ott a, J.E. Bateman b, R. Stephenson b, j. Connolly G. Tappern b a Dept. of Physics, Institute of Cancer Research, Royal Marsden Hospital, Belmont, Surrey, SM2 5PT, UK b Rutherford Appleton Laboratory, Chilton, Didcot, Oxon, UK b A fully optimised BaF 2-TMAE detector has been developed for use in Positron Emission Tomography (PET). As part of this work we have attempted to optimise the performance of this technology with respect to efficiency, timing (resolution and deadtime), and spatial resolution for 511 keV photons. We report for the first time on the detailed operation of our detector. Under stable operating conditions over an 8 hour period, we consistently obtain an absolute detection efficiency of 23.5% for an 8 mm thick crystal, simultaneously with 4 ns FWHM and 9.5 ns FWTM timing resolution and a spatial resolution of 6 mm FWHM. We also discuss how these parameters affect the performance of a full-size camera based on BaF 2-TMAE detectors. 1. Introduction 2. Detector operation Despite promising early results [1,2], PET using Multi-Wire Proportional Chambers (MWPCs) has had little widespread impact. Although one such design, the MUP-PET system, has had considerable success in an oncology setting (see e.g. refs. [3-6]) there are several severe disadvantages associated with this type of technology. The principal limitations are low sensi- tivity, high random fraction and high scatter fraction. Bateman [7] describes in detail the problems associ- ated with imaging under these conditions. The net result is that in a practical imaging situation, the over- all statistical quality of the data is compromised by the limited maximum count rate (~ 2 cps). This is a conse- quence of saturation in the readout circuitry from singles and random events at an activity ~ 15 MBq in the field of view. By capitalising on experience gained with the MUP- PET camera and utilising the BaFz-TMAE wire cham- ber technique first demonstrated by Anderson [8] we have developed a detector which addresses these limi- tations. The two major attractions of this technique are the increase in efficiency available, which is simply propor- tional to the BaF z crystal thickness, and the fast time resolution resulting from low-pressure operation asso- ciated with using TMAE. * Corresponding author. The detector is divided into three main sections: the primary detection region, the transfer gap and the MWPC readout (see Fig. 1) The primary detection region comprises four 50 x 50 mm 2 crystals, a 0.5 mm wide reverse bias region, 3.5 mm wide conversion and a 10 mm preamplification region (Fig. 1). The scintillator holder accommodates four 50 x 50 mm 2 crystals, up to 16 mm thick, in a 2 X 2 array as shown in Fig. 1 (inset). The electric fields are defined by wire planes consisting of 50 ixm beryllium-copper (Be-Cu) wire wound at a pitch of 500 txm. Annihilation photons enter the crystal and either Compton scatter or photoelectrically absorb, producing scintillation light in the crystal. Photoelectrons are produced by photoionisation of the TMAE from the UV scintillation light emitted directly below the crystal. Any photoelectrons created within the conversion re- gion are immediately subjected to gas amplification. (The reverse bias gap shields the crystal surface from the positive ions produced within the primary ava- lanche. The chamber becomes unstable if these ions are allowed to adhere, due to the resulting charge presence on the crystal's highly insulating surface. Chamber performance deteriorates within 1-2 hours unless the reverse bias is used.) The initial small electron shower passes into the preamplification gap where further multiplication takes place. The pulse induced as the electrons pass plane 3 0168-9002/94/$07.00 © 1994 - Elsevier Science B.V. All rights reserved SSDI 0168-9002(94)00112-K