2932 IEEE TRANSACTIONS ON NUCLEAR SCIENCE, VOL. 52, NO. 6, DECEMBER 2005 A New Coplanar-Grid High-Pressure Xenon Gamma-Ray Spectrometer Scott D. Kiff, Student Member, IEEE, Zhong He, Senior Member, IEEE, and Gary C. Tepper Abstract—High-pressure xenon (HPXe) gas is a desirable radia- tion detection medium for many reasons, including its large atomic number, high density, low mean energy to produce an electron-ion pair, and the ability to produce devices with large detection vol- umes. While past work in HPXe has produced relatively successful detectors with energy resolution at 662 keV as good as approxi- mately 2% FWHM, an expected limitation of these chambers in field operation is resolution degradation due to the vibration of their Frisch grids. Progress on a detector without a Frisch grid is reported in this submission; it is expected that the proposed anode design will provide competitive energy resolution with minimal degradation from mechanical vibration. Simulations accounting for charge carrier statistics, changes in the charge induced on the anode as a function of interaction location, and electronic noise predict a best-case energy resolution of 2.3% FWHM at 662 keV. Experimental data is compared with these simulations. Index Terms—Coplanar anodes, gas detectors, Geant, ionization chambers, single polarity charge sensing, xenon. I. INTRODUCTION H IGH-PRESSURE XENON (HPXe) gas is a desirable radiation detection medium for many reasons, including its large atomic number, high density, low mean energy to produce an electron-ion pair, and the ability to produce devices with large detection volumes. In the past, three main categories of geometries have been developed: parallel plate with a Frisch grid [1], cylindrical without a grid [2], and finally, gridded cylindrical chambers [3]. Cylindrical detectors without a Frisch grid generally have the poorest energy resolution, as expected: the best energy resolution reported is around 4% full width at half maximum (FWHM) at 662 keV [4]. The spectroscopic per- formance of both geometries of gridded detectors is generally better and has improved steadily, with the best reported results approaching 2% at 662 keV for both the parallel plate [5], [6] and cylindrical [7] geometries. A limitation of these chambers in practice is rather extreme microphonic noise due to Frisch grid vibration. In addition, the construction of Frisch grids can be complicated. The purpose of this paper is to find a successful alternative to the Frisch grid; the technique of coplanar anode grids first reported by Luke [8] is employed. Manuscript received November 15, 2004; revised September 22, 2005. This work was supported in part by a National Science Foundation Graduate Re- search Fellowship, and by the U. S. Department of Energy NEER program of- fice. S. D. Kiff and Z. He are with the University of Michigan, Ann Arbor, MI 48109 USA (e-mail: kiff@umich.edu; hezhong@umich.edu). G. C. Tepper is with Virginia Commonwealth University, Richmond, VA 23284 USA (e-mail: gtepper@vcu.edu). Digital Object Identifier 10.1109/TNS.2005.862804 In the coplanar anode technique, the detector’s anode is seg- mented into many strips (or wires in this case), and alternating strips are directly connected to one another to form two inde- pendent anodes. Each anode, then, is made of one-half the total number of strips, and while one anode is constructed from strips 1, 3, 5, and so on, the other anode contains strips 2, 4, 6, etc. Usually one of the anodes is biased higher than the other, so that the electrons will always be collected on one set of anode wires, termed the collecting anode. By constructing and oper- ating the detector in such a manner, the induced charge on each of the two anodes is equal through most of the detection volume except very near the anodes, and subtracting the noncollecting anode’s preamplifier output signal from the collecting anode’s will give a final pulse amplitude that is independent of interac- tion location inside the detector, assuming electron-ion recom- bination can be neglected. Previously-reported HPXe designs employing the coplanar anode technique include: (i) a cylindrical detector with an insulating support rod along the central axis, around which alternating collecting and noncollecting anode wires were wound in a double-helical fashion; (ii) a planar device with anode strips spiraling out of the center of a circular electrode plate toward its outer radius; and (iii) a concept using two anode wires stretched parallel to the detector’s central axis, termed the dual-anode cylindrical ionization chamber (DACIC) by its de- velopers, which has a reported energy resolution of 3% FWHM for 662-keV gamma rays while operated in single-anode mode [9], [10]. Even though the DACIC has exhibited some potential for performing spectroscopy without a Frisch grid, there are some aspects of its design and operation that can be improved. For example, because there are only two anode wires, charge in- duction on the two anodes is not very uniform for large portions of the detection volume. Two other concerns with the DACIC design stem from both of its anodes being at the same potential: if an event occurs near a point equidistant from the two anode wires, charge diffusion could lead to electrons being collected on both anodes, resulting in a reduced pulse amplitude after signal subtraction; and for multi-vertex events, which are quite probable for high gamma-ray energies, interactions occurring on opposite sides of the detector will be collected by opposing anodes, and the true energy deposition will be diminished by the subtraction step. The focus of this paper is an expansion of the DACIC con- cept, extending it to an anode structure formed by numerous wires connected into two sets; increasing the number of wires makes the charge induction as a function of interaction position more uniform. The wires run parallel to the central axis of the cylindrical detector, are all positioned at the same radius with re- 0018-9499/$20.00 © 2005 IEEE