1232 IEEE TRANSACTIONS ON PLASMA SCIENCE, VOL. 30, NO. 3, JUNE 2002 Effects of Anode Materials on the Performance of Explosive Field Emission Diodes Donald A. Shiffler, John W. Luginsland, Member, IEEE, Ryan J. Umstattd, M. LaCour, K. Golby, Michael D. Haworth, M. Ruebush, D. Zagar, A. Gibbs, and Thomas A. Spencer Abstract—Explosive field-emission cathodes have been the electron emitter of choice, and often necessity, for high-power microwave (HPM) tubes for many years. The materials typically used for these cathodes range from polymer and cotton velvets, to metals such as stainless steel, and to carbon materials such as bulk carbon and carbon fibers. With several notable exceptions, the issues of the anode and its composition have been largely ignored. Generally, the diode performance, such as current levels, impedance collapse, and out-gassing, have been attributed to the cathode alone rather than to the combination of the cathode and anode. In this paper, we investigate the affects of various anode materials on the performance of explosive field emission cathodes. We show that bipolar flow significantly and rapidly alter diode performance at lower voltage and energy densities than usually observed. We show also the effects of anode material choice on out-gassing, and diode conditioning. Experiments have shown that bipolar flow is a significant issue in diode performance for even short pulses. The theoretical aspects of the diodes are discussed, with a comparison of experiment to theory. Index Terms—Anodes, electron beams, electron emission, elec- tron guns, plasmas. I. INTRODUCTION E XPLOSIVE field-emission cathodes have been the cathode of choice, driven in reality by necessity, for high-power microwave (HPM) tubes since their inception. HPM tubes typically require much greater amounts of electron current and power than lower power conventional microwave tubes [1]. While the highest power conventional microwave tubes operate at currents on the order of hundreds of amps and electron energies in the hundreds of of kiloelectronvolts, HPM tubes operate at least an order of magnitude higher, with beam currents in the thousands of amps and diode voltages approaching 500 keV–5 MeV. Hence, conventional thermal cathodes cannot currently meet the demands of an HPM environment. Manuscript received June 25, 2001; revised October 11, 2001. This work has been supported in part by the Air Force Office of Scientific Research. D. A. Shiffler, M. D. Haworth, M. Ruebush, D. Zagar, A.Gibbs, and T. A. Spencer are with the Directed Energy Directorate, Air Force Research Labora- tory, Kirtland, AFB, NM 87117 USA (e-mail: Don.Shiffler@kirtland.af.mil). J. W. Luginsland was with the Directed Energy Directorate, Air Force Re- search Laboratory, Kirtland AFB, NM 87117 USA. He is now with Science Applications International Corporation, Albuquerque, NM 87106 USA (e-mail: John.W.Luginsland@saic.com). R. J. Umstattd was with the Directed Energy Directorate, Air Force Research Laboratory, Kirtland AFB, NM 87117 USA. He is now with the Physics Department, Naval Postgraduate School, Monterey, CA 93943 USA (e-mail: rjumstat@nps.navy.mil). M. LaCour and K. Golby are with the Maxwell Technologies, Inc., Albu- querque, NM 87117 USA. Digital Object Identifier 10.1109/TPS.2002.802146 Further, many HPM tubes differ significantly from con- ventional tubes in the method by which the electron beam is deposited on the anode. In conventional tubes such as the klystron, the electron beam dump is largely separated from both the cathode and the RF interaction region. However, in HPM tubes such as the magnetron or the magnetically insulated transmission line oscillator (MILO), the cathode and anode are in the same interaction space as the microwaves which are being produced. Therefore, in HPM tubes, the effects of the anode are of more vital importance than for standard conventional microwave tubes. Historically a variety of materials have been used for explosive field-emission cathodes. These materials span the range from stainless steel, to polymer and cotton velvets, to bulk carbon and carbon fibers. In most cases, the overall diode performance is attributed to the cathode and the material of which it consists [2]. In this paper, we discuss a series of experiments which investigated the effects of the anode on the overall diode performance. It is well known that contaminating surface layers on the anode can change the diode performance [3], [4]. We investigate the performance of diodes with similar cathode and various anodes under two pulse length conditions. We discuss the effects of anode material on out-gassing. The diode performance is discussed in relation to theory and simulation. In particular, the results imply that contaminants in the anode, in the bulk of the material and on the surface, lead to bipolar flow in the diode. The bipolar flow can be initiated on very short time scales, on the order of 10 ns and is evident at lower fluences than previously observed [5]. The bipolar flow occurs without gap closure and over pulse lengths from 50 ns–1 s with similar diode geometries. Furthermore, the out-gassing in the diode can be influenced by the anode at least as much as by the cathode alone. The paper begins with a discussion of the experimental configuration, including the two pulsers used for the experi- ment. We discuss the theoretical and simulation aspects of the problem, followed by a review of the experimental results. We compare the experiment with theory and simulation and then finish with brief conclusions. II. EXPERIMENTAL CONFIGURATION Two pulsers were used in these experiments, both of which have been discussed elsewhere [5], [6]. The first, called the rep-rate test pulser (RTP) is a 1- s 100- 500-kV machine capable of operation at 2 Hz. The second pulser, the general rep-rate universal multipurpose pulser (GRUMP), is a 50-ns 5- 0093-3813/02$17.00 © 2002 IEEE