Publications

Electrostatic discharge (ESD) and electrical overstress (EOS) damage of Micro-Electro-Mechanical Systems (MEMS) has been identified as a new failure mode. This failure mode has not been previously recognized or addressed primarily due to the mechanical nature and functionality of these systems, as well as the physical failure signature that resembles stiction. Because many MEMS devices function by electrostatic actuation, the possibility of these devices not only being susceptible to ESD or EOS damage but also having a high probability of suffering catastrophic failure due to ESD or EOS is very real. Results from previous experiments have shown stationary comb fingers adhered to the ground plane on MEMS devices tested in shock, vibration, and benign environments. Using Sandia polysilicon microengines, we have conducted tests to establish and explain the ESD/EOS failure mechanism of MEMS devices. These devices were electronically and optically inspected prior to and after ESD and EOS testing. This paper will address the issues surrounding MEMS susceptibility to ESD and EOS damage as well as describe the experimental method and results found from ESD and EOS testing. The tests were conducted using conventional IC failure analysis and reliability assessment characterization tools. In this paper we will also present a thermal model to accurately depict the heat exchange between an electrostatic comb finger and the ground plane during an ESD event.

One of the most significant challenges for technology characterization and future analysis is to keep instrumentation and techniques in step with the development of technology itself. Not only are dimensions shrinking and new materials being employed, but the rate of change is increasing. According to the 1999 International Technology Roadmap for Semiconductors (ITRS) the number and difficulty of the technical challenges continue to increase as technology moves forward. It could be argued that technology cannot be developed without appropriate analytical technique, nevertheless while much effort is being directed at materials and processes, only a small proportion is being directed at analysis. Whereas previous versions of the Semiconductor Industry Association roadmap contained a small number of implicit references to characterization and analysis, the 1999 ITRS contains many explicit references. It is clear that characterization is now woven through the roadmap, and technology developers in all areas appreciate the fact that new instrumentation and techniques will be required to sustain the rate of development the semiconductor industry has seen in recent years. Late in 1999, a subcommittee of the Sematech Product Analysis Forum reviewed the ITRS and identified a top-ten list of challenges which the failure analysis community will face as present technologies are extended and future technologies are developed. This paper discusses the PAF top-ten list of challenges, which is based primarily on the Difficult Challenges tables from each ITRS working group. Eight of the top-ten are challenges of significant technical magnitude, only two could be considered non-technical in nature. Most of these challenges cut across several working group areas and could be considered common threads in the roadmap, ranging from fault simulation and modeling to imaging small features, from electrical defect isolation to reprocessing.

The W88 Integrated Circuit Shelf Life Program was created to monitor the long term performance, reliability characteristics, and technological status of representative WR ICs manufactured by the Allied Signal Albuquerque Microelectronics Operation (AMO) and by Harris Semiconductor Custom Integrated Circuits Division. Six types of ICs were used. A total of 272 ICs entered two storage temperature environments. Electrical testing and destructive physical analysis were completed in 1995. During each year of the program, the ICs were electrically tested and samples were selected for destructive physical analysis (DPA). ICs that failed electrical tests or DPA criteria were analyzed. Fifteen electrical failures occurred, with two dominant failure modes: electrical overstress (EOS) damage involving the production test programs and electrostatic discharge (ESD) damage during analysis. Because of the extensive handling required during multi-year programs like this, it is not unusual for EOS and ESD failures to occur even though handling and testing precautions are taken. The clustering of the electrical test failures in a small subset of the test operations supports the conclusion that the test operation itself was responsible for many of the failures and is suspected to be responsible for the others. Analysis of the electrical data for the good ICs found no significant degradation trends caused by the storage environments. Forty-six ICs were selected for DPA with findings primarily in two areas: wire bonding and die processing. The wire bonding and die processing findings are not surprising since these technology conditions had been documented during manufacturing and were determined to present acceptable risk. The current reliability assessment of the W88 stockpile assemblies employing these and related ICs is reinforced by the results of this shelf life program. Data from this program will aid future investigation of 4/3 micron or MNOS IC technology failure modes.

Failure analysis is a critical element in the integrated circuit manufacturing industry. This paper reviews the changing role of failure analysis and describes major techniques employed in the industry today. Several advanced failure analysis techniques that meet the challenges imposed by advancements in integrated circuit technology are described and their applications are discussed. Future trends in failure analysis needed to keep pace with the continuing advancements in integrated circuit technology are anticipated.

In this paper we describe an algorithm by which obstructions and surface features in an underground storage tank can be modeled and used to generate virtual barrier functions for a real-time telerobotic system, which provides an aid to the operator for both real-time obstacle avoidance and for surface tracking. The algorithm requires that the slave's tool and every object in the waste storage tank be decomposed into convex polyhedral primitives, with the waste surface modeled by triangular prisms. Intrusion distance and extraction vectors are then derived at every time step by applying Gilbert's polyhedra distance algorithm, which has been adapted for the task. This information is then used to determine the compression and location of nonlinear virtual spring-dampers whose total force is summed and applied to the manipulator/teleoperator system. Experimental results using a PUMA 560 and a simulated waste surface validate the approach, showing that it is possible to compute the algorithm and generate smooth, realistic pseudo forces for the teleoperator system using standard VME bus hardware.

In this paper we describe an algorithm for generating virtual forces in a bilateral teleoperator system. The virtual forces are generated from a world model and are used to provide real-time obstacle avoidance and guidance capabilities. The algorithm requires that the slaves tool and every object in the environment be decomposed into convex polyhedral Primitives. Intrusion distance and extraction vectors are then derived at every time step by applying Gilbert`s polyhedra distance algorithm, which has been adapted for the task. This information is then used to determine the compression and location of nonlinear virtual spring-dampers whose total force is summed and applied to the manipulator/teleoperator system. Experimental results validate the whole approach, showing that it is possible to compute the algorithm and generate realistic, useful psuedo forces for a bilateral teleoperator system using standard VME bus hardware.

Before disposing of transuranic radioactive waste in the Waste Isolation Pilot Plant (WIPP), the United States Department of Energy (DOE) must evaluate compliance with applicable long-term regulations of the United States Environmental Protection Agency (EPA). Sandia National Laboratories is conducting iterative performance assessments (PAs) of the WIPP for the DOE to provide interim guidance while preparing for a final compliance evaluation. This volume of the 1992 PA contains results of uncertainty and sensitivity analyses with respect to the EPA`s Environmental Protection Standards for Management and Disposal of Spent Nuclear Fuel, High-Level and Transuranic Radioactive Wastes (40 CFR 191, Subpart B). Additional information about the 1992 PA is provided in other volumes. Results of the 1992 uncertainty and sensitivity analyses indicate that, conditional on the modeling assumptions, the choice of parameters selected for sampling, and the assigned parameter-value distributions, the most important parameters for which uncertainty has the potential to affect compliance with 40 CFR 191B are: drilling intensity, intrusion borehole permeability, halite and anhydrite permeabilities, radionuclide solubilities and distribution coefficients, fracture spacing in the Culebra Dolomite Member of the Rustler Formation, porosity of the Culebra, and spatial variability of Culebra transmissivity. Performance with respect to 40 CFR 191B is insensitive to uncertainty in other parameters; however, additional data are needed to confirm that reality lies within the assigned distributions.

A bilateral teleoperation system consists of a local master manipulator and a remotely located slave manuipulator. Generalized velocity commands are sent forward from the master to the slave, and generalized force/torque information is reflected'' back from the slave to the master. Often there is a transmission delay incurred when communicating between the two subsystems which causes instability in the bilateral teleoperator. Recently, a solution for this instability problem was found, based on mimicking the behavior of a lossless transmission line. Although this solution provides steady-state force and velocity tracking, it does not provide steady-state force and position tracking, as is desired for bilateral teleoperation. In this paper a modification is given to the basic control law which overcomes this difficulty. 13 refs., 6 figs., 1 tab.

The working group on electrical breakdown in vacuum was charged with considering all possible mechanisms by which electrical breakdown might occur either through the vacuum or along insulator bushings in large area electron beam emitter assemblies. It was understood that present systems need to be scaled up, by an order of magnitude or more in both beam area and total energy, to meet demands for higher power and larger size machines, and that increases in the e-beam current density and transport efficiency are also sought. A consideration of the consequences of such a scale-up was pertinent to many of the topics listed in the working-group agenda. Our group attempted to address each of these topics. 18 refs.

Over the past several decades, many researchers have contributed to present understanding of the flashover of electrically stressed insulators in vacuum, and a wealth of theories have been proposed to explain the many surprising attributes of this complex breakdown mechanism. Surface flashover appears to comprise at least two distinct phenomena which can be distinguished as being cathode-initiated or anode-initiated, with the former having received by far the most attention. Several models describing cathode-initiated flashover have been built on the pioneering work of Boersch and coworkers, published in 1963, and credit the breakdown mechanism to the action of an intense secondary-electron-emission avalanche on the insulator surface. Other researchers consider the electron avalanche to be only partially, if at all, responsible, and invoke various hot-carrier effects in the insulator bulk, the surface interfacial region, or in a layer of gas adsorbed on the insulator surface. Anode-initiated flashover, which contends with the cathode-initiated variety for the breakdown of insulators of conventional design, is thought to involve bulk breakdown in a way related to treeing failure. In spite of the considerable effort applied to understanding vacuum surface flashover, no single theory appears capable of explaining all the data, and new and often unexpected observations continue to be made. 42 refs., 6 figs.

Computed torque algorithms are used to compensate for the changing dynamics of robot manipulators in order to ensure that a constant level of damping is maintained for all configurations. Unfortunately, there are three significant problems with existing computed torque algorithms. First, they are nonpassive and can lead to unstable behavior; second, they make inefficient use of actuator capability; and third, they cannot be used to maintain a constant end-effector stiffness for force control tasks. Recently, we introduced a new control algorithm for robots which, like computed torque, uses a model of the manipulator's dynamics to maintain a constant level of damping in the system, but does so passively. This new class of passive control algorithms has guaranteed stability properties, utilizes actuators more effectively, and can also be used to maintain constant end-effector stiffness. In this paper, this approach is described in detail, implementation issues are discussed, and simulation results are given. 15 refs., 6 figs., 2 tabs.