Investigation of Switching Time in GaN/AlN Resonant Tunneling Diodes by Experiments and P-SPICE Models W-D. Zhang, Senior Member, IEEE, T. A. Growden, D. F. Storm, D. J. Meyer, P. R. Berger, Fellow, IEEE and E. R. Brown, Fellow, IEEE Abstract—The experimental and simulated switching behav- ior across the negative differential resistance (NDR) region of GaN/AlN double-barrier resonant tunneling diodes (RTDs) is presented. The shortest 10-90% experimental switching time was ∼55 ps. The experimental results are also studied with P-SPICE circuit models, which show that the relatively low peak-to-valley current ratio (∼1.5) and relatively high specific series resistance (≥ 1 × 10 -5 Ω-cm 2 ), both limit the switching time. Index Terms—GaN/AlN, heterostructure, RTD, switching time, measurement, modeling, P-SPICE. I. I NTRODUCTION G N-based heterostructures have broad applications, with notable examples being field effect transistors [1] and light emitting devices [2]. These successes and the large bandoffset have motivated researchers to begin exploration of vertical GaN-based heterostructures as a platform for other types of devices. Recently, GaN/AlN/GaN resonant tunneling diodes (RTDs) have demonstrated highly repeatable operation at room temperature while suffering no obvious degradation from traps. This success is attributable to better design method- ology that addresses the high polarization fields, high-quality growth methods and improved fabrication techniques [3]-[4]. However, questions remain on whether GaN/AlN RTDs are capable of being implemented as oscillators and switches with speeds that can compete with earlier RTDs based on GaAs/AlGaAs, InGaAs/AlAs, and InAs/AlSb materials. The wide bandgap and the excellent thermal conductivity of GaN allows very high current density operation of RTDs [5], but can these merits provide any advantages for GaN-based RTDs in high-speed applications? This letter reports our latest investigation into this issue, specifically the first switching time measurements on GaN/AlN RTDs. Additionally, we have W-D. Zhang is with Departments of Physics and Electrical Engi- neering, Wright State University, Dayton, OH 45435, USA. Email: wz- zhang@fastmail.fm T. A. Growden was Department of Electrical and Computer Engineering, The Ohio State University, Columbus, OH 43210, USA. Currently he is NAS-NRC Postdoctoral Research Fellow at the Naval Research Laboratory, Washington, DC 20375, USA. D. F. Storm and D. J. Meyer are with U.S. Naval Research Laboratory, Washington, DC 20375, USA. P. R. Berger is with Department of Electrical and Computer Engineering, The Ohio State University, Columbus, OH 43210, USA. E. R. Brown is with Departments of Physics and Electrical Engi- neering, Wright State University, Dayton, OH 45435, USA . Email: el- liott.brown@wright.edu Manuscript received August 20, 2019; revised August 26, 2019. Fig. 1. (a) GaN/AlN heterostructure stack used for fabrication of GaN/AlN RTDs. (b) Experimental setup for switching time measurements. conducted P-SPICE simulations to better explain the mea- sured data. Together, the experiments and models presented here provide insight for designing next-generation high-speed, high-power GaN/AlN RTDs. II. SWITCHING TIME MEASUREMENTS The GaN/AlN/GaN RTD heterostructure studied in this research is shown in Fig. 1(a). Its growth and fabrication have been described in detail elsewhere [3], [6]. Figure 1 (b) displays the experimental set-up which was used to investigate the switching time of the GaN/AlN RTDs. This same setup was previously used to successfully measure the switching time of In 0.53 Ga 0.47 As/AlAs RTDs [7]-[8]. At the center of the set-up is an Anritsu bias tee with a risetime of t 1 ∼ 7 ps. The DC port of the bias tee is connected to a ramp signal generator, which produces a DC offset bias along with a triangular waveform that triggers the switching events. The RF port of the bias tee is connected to a fast oscilloscope, in this case, an Infiniium MSOS804A 8-GHz oscilloscope, via a SMA coaxial cable. The Infiniium oscilloscope has an internal