Search

Abstract: Ion Beam Induced Charge (IBIC) is the basic mechanism of the operation of semiconductor detectors and it can lead to Single Event Effects (SEEs) in microelectronic devices. To be able to predict SEEs in ICs and detector responses one needs to be able to simulate the radiation-induced current as the function of time on the electrodes of the devices and detectors. There are analytical models, which work for very simple detector configurations, but fail for anything more complex. Technology Computer Aided Design (TCAD) programs can simulate this process in microelectronic devices, but these TCAD codes costs hundreds of thousands of dollars and they require huge computing resources. In addition, in certain cases they fail to predict the correct behavior. Here a simulation model based on the Gunn theorem was developed and used with the COMSOL Multiphysics framework, version 3.5. In the model, the induced current can be calculated both directly and in certain cases using the powerful adjoint method. A brief description of the model will be given in the paper with examples for detectors and microelectronic devices using both the direct and the adjoint method. [Copyright &y& Elsevier]

Abstract: The motion of ionizing-radiation-induced rogue charge carriers in a semiconductor can create unwanted voltage and current conditions within a microelectronic circuit. If sufficient unwanted charge or current occurs on a sensitive node, a variety of single event effects (SEEs) can occur with consequences ranging from trivial to catastrophic. This paper describes the application of heavy ion microprobes to assist with calibration and validation of SEE modeling approaches. [Copyright &y& Elsevier]

Pulse widths of single-event transients produced by alpha particles in a 65-nm bulk CMOS technology are reported. The experimental setup and calibration of the alpha particle experiment is described in detail. A focused-ion beam is also utilized to explore how pulse broadening in the test circuit impacts the alpha particle SET measurements. The results of this work show that alpha particles are able to induce transient signals with a width of about 25 ps in this technology. [ABSTRACT FROM AUTHOR]

We report heavy-ion microbeam and total dose data for a new complementary (npn + pnp) SiGe on thick-film SOI technology. Measured transient waveforms from heavy-ion strikes indicate a significantly shortened single-event-induced transient current, while maintaining the total dose robustness associated with SiGe devices. Heavy-ion broad-beam data confirm a reduced single event upset (SEU) cross-section in a high-speed shift register circuit. [ABSTRACT FROM AUTHOR]

We investigate the efficacy of mitigating radiation-based single event effects (SEE) within circuits incorporating SiGe heterojunction bipolar transistors (HBTs) built with an N-Ring, a transistor-level layout-based radiation hardened by design (RHBD) technique. Previous work of single-device ion-beam induced charge collection (IBICC) studies has demonstrated significant reductions in peak collector charge collection and sensitive area for charge collection; however, few circuit studies using this technique have been performed. Transient studies performed with Sandia National Laboratory's (SNL) 36 MeV ^16O microbeam on voltage references built with N-Ring SiGe HBTs have shown mixed results, with reductions in the number of large voltage disruptions in addition to new sensitive areas of low-level output voltage disturbances. Similar discrepancies between device-level IBICC results and circuit measurements are found for the case of digital shift registers implemented with N-Ring SiGe HBTs irradiated in a broadbeam environment at Texas A&M's Cyclotron Institute. The error cross-section curve of the N-Ring based register is found to be larger at larger ion LETs than the standard SiGe register, which is clearly counter-intuitive. We have worked to resolve the discrepancy between the measured circuit results and the device-level IBICC measurements, by re-measuring single-device N-Ring SiGe HBTs using a time-resolved ion beam induced charge (TRIBIC) set-up that allows direct capture of nodal transients. Coupling these measurements with full 3-D TCAD simulations provides complete insight into the origin of transient currents in an N-Ring SiGe HBT. The detailed structure of these transients and their bias dependencies are discussed, together with the ramifications for the design of space-borne analog and digital circuits using SiGe HBTs. [ABSTRACT FROM AUTHOR]

Delayed charge collection from ionizing events outside the deep trench can increase the SEU cross section in deep trench isolation technologies. Microbeam test data and device simulations demonstrate how this adverse effect can be mitigated through substrate engineering techniques. The addition of a heavily doped p-type charge-blocking buried layer in the substrate can reduce the delayed charge collection from events that occur outside the deep trench isolation by almost an order of magnitude, implying an approximately comparable reduction in the SEU cross section. [ABSTRACT FROM AUTHOR]

This paper presents the first 3D simulation of heavy-ion induced charge collection in a SiGe RBT,together with microbeam testing data. The charge collected by the terminals is a strong function of the ion striking position. The sensitive area of charge collection for each terminal is identified based on analysis of the device structure and simulation results. For a normal strike between the deep trench edges, most of the electrons and holes are collected by the collector and substrate terminals, respectively. For, an ion strike between the shallow trench edges surrounding the emitter, the base collects appreciable amount of charge. Emitter collects negligible amount of charge. Good agreement is achieved between the experimental and simulated data. Problems encountered with mesh generation and charge collection simulation are also discussed. [ABSTRACT FROM AUTHOR]