Paul J. McWhorter

Physics of the ferroelectric nonvolatile memory field effect transistor

The operation of the ferroelectric nonvolatile memory field effect transistor is theoretically examined extensively for the first time. The ferroelectric transistor device properties are derived by combining the silicon charge‐sheet model of metal‐oxide‐semiconductor field‐effect transistor device operation with Maxwell’s first equation which describes the properties of the ferroelectric film. The model we present describes ferroelectric transistor I‐V and C‐V behavior when time‐dependent voltages are applied which result in hysteresis due to ferroelectric switching. The theoretical results provide unique insight into the effects of geometrical and material parameters on the electrical properties of the transistor. These parameters include the ferroelectric spontaneous and remanent polarization, the coercive field, and dielectric layer thicknesses. We have found that the conventional concept of threshold voltage is no longer useful, and that increasing the spontaneous polarization has only a minor impact ...

Modeling the anneal of radiation-induced trapped holes in a varying thermal environment

The anneal of radiation-induced trapped holes in MOS transistors is found to be thermally activated. A quantitative, physical model based on thermal emission and tunneling is developed. It accurately predicts the anneal of radiation-induced trapped holes in constant or time-varying thermal environments. Data are presented which quantitatively verify the accuracy of the model for temperatures between 25 and 160 degrees C. This model provides the basis for developing accurate quantitative screens for the rebound failure mechanism. >

Embedded micromechanical devices for the monolithic integration of MEMS with CMOS

A flexible, modular manufacturing process for integrating micromechanical and microelectronic devices has been developed. This process embeds the micromechanical devices in an anisotropically etched trench below the surface of the wafer. Prior to microelectronic device fabrication, this trench is refilled with oxide, chemical-mechanically polished, and sealed with a nitride cap in order to embed the micromechanical devices below the surface of the planarized wafer. The feasibility of this technique in a manufacturing environment has been demonstrated by combining a variety of embedded micromechanical structures with a 2 /spl mu/m CMOS process on 6 inch wafers. A yield of 78% has been achieved on the first devices manufactured using this technique.

Modeling the memory retention characteristics of silicon‐nitride‐oxide‐silicon nonvolatile transistors in a varying thermal environment

The memory retention characteristics of silicon‐nitride‐oxide‐silicon nonvolatile memory devices are found to be strongly thermally activated. A model is developed based on thermal emission of charge from traps. This model accurately predicts the threshold voltage decay of transistors stored in varying thermal environments. The model is demonstrated to be accurate over 7 decades of time and for temperatures between −40 and 200 °C.

Electron paramagnetic resonance investigation of charge trapping centers in amorphous silicon nitride films

We have explored the nature of the silicon dangling‐bond center in amorphous hydrogenated silicon nitride (a‐SiNx:H) thin films, and its relationship to the charge trapping centers using electron paramagnetic resonance (EPR) and capacitance‐voltage (C‐V) measurements. We have investigated the quantitative relationship between the concentration of silicon dangling bonds using EPR and the concentration of charge traps, measured by C‐V measurements, for both UV‐illuminated and unilluminated a‐SiNx:H thin films subjected to both electron and hole injection sequences. A theoretical framework for our results is also discussed. These results continue to support a model in which the Si dangling bond is a negative‐U defect in silicon nitride, and that a change in charge state of preexisting positively and negatively charged Si sites is responsible for the trapping phenomena observed in these thin film dielectrics.

Friction in surface-micromachined microengines

Understanding the frictional properties of advanced Micro-Electro-Mechanical Systems (MEMS) is essential in order to develop optimized designs and fabrication processes, as well as to qualify devices for commercial applications. We develop and demonstrate a method to experimentally measure the forces associated with sliding friction of devices rotating on a hub. The method is demonstrated on the rotating output gear of the microengine recently developed at Sandia National Laboratories. In-situ measurements of an engine running at 18300 rpm give a coefficient of friction of 0.5 for radial (normal) forces less than 4 (mu) N. For larger forces the effective coefficient of friction abruptly increases, suggesting a fundamental change in the basic nature of the interaction between the gear and hub. The experimental approach we have developed to measure the frictional forces associated with the microengine is generically applicable to other MEMS devices.

Performance trade-offs for a surface micromachined microengine

An electromechanical model of Sandias microengine is developed and applied to quantify critical performance tradeoffs. This is done by determining how forces impact the mechanical response of the engine to different electrical drive signals. To validate the theoretical results, model- based drive signals are used to operate actual engines, where controlled operation is achieved for the following cases: 1) spring forces are dominant, 2) frictional forces are dominant, 3) linear inertial forces are dominant, 4) viscous damping forces are dominant, and 5) inertial load forces are dominant. Significant improvements in engine performance are experimentally demonstrated in the following areas: positional control, start/stop endurance, constant speed endurance, friction load reduction,and rapid actuation of inertial loads.

Extremal dynamics: A unifying physical explanation of fractals, 1/f noise, and activated processes

The properties of physical systems whose observable properties depend upon random exceedances of critical parameters are quantitatively examined. Using extreme value theory, the dynamical behavior of this broad class of systems is derived. This class of systems can exhibit two characteristic signatures: generalized activation when far from equilibrium and noise with a characteristic power spectrum (including 1/f ) when in quasiequilibrium. Fractal structures can also arise from these systems. It is thus demonstrated that generalized activation, noise, and fractals, in some cases, are simply different manifestations of a single common dynamical principle, which is termed ‘‘extremal dynamics.’’ Examples of physical processes governed by extremal dynamics are discussed, including data loss of nonvolatile memories and dielectric breakdown.

A practical predictive formalism to describe generalized activated physical processes

A predictive formalism is developed that is applicable to the large class of activated physical systems described by a differential equation of the generic form: ∂n(φ,t)/∂t =−n(φ,t)F(t) exp(−(φ−R(t))/A(t)). Practical techniques to predict the behavior of activated physical systems for arbitrary time‐dependent environments are both intuitively and mathematically developed. Useful techniques to experimentally determine the initial distribution of activation energies, utilizing arbitrary time‐dependent laboratory environments, are presented. A number of fundamental results regarding the correct use and interpretation of common diagnostic techniques, such as Arrhenius plots, are derived. It is shown how the predictive results significantly enhance the ability to quantitatively evaluate the reliability of physical systems whose rate‐limiting mechanisms are activated processes obeying the above differential equation. Specific issues regarding integrated circuit reliability are examined as potential applications...

Characterization of the embedded micromechanical device approach to the monolithic integration of MEMS with CMOS

Recently, a great deal of interest has developed in manufacturing processes that allow the monolithic integration of microelectromechanical systems (MEMS) with driving, controlling, and signal processing electronics. This integration promises to improve the performance of micromechanical devices as well as lower the cost of manufacturing, packaging, and instrumenting these devices by combining the micromechanical devices with a electronic devices in the same manufacturing and packaging process. In order to maintain modularity and overcome some of the manufacturing challenges of the CMOS-first approach to integration, we have developed a MEMS-first process. This process places the micromechanical devices in a shallow trench, planarizes the wafer, and seals the micromechanical devices in the trench. Then, a high-temperature anneal is performed after the devices are embedded in the trench prior to microelectronics processing. This anneal stress-relieves the micromechanical polysilicon and ensures that the subsequent thermal processing associated with fabrication of the microelectronic processing does not aversely affect the mechanical properties of the polysilicon structures. These wafers with the completed, planarized micromechanical devices are then used as starting material for conventional CMOS processes. The circuit yield for the process has exceeded 98 percent. A description of the integration technology, the refinements to the technology, and wafer- scale parametric measurements of device characteristics is presented. Additionally, the performance of integrated sensing devices built using this technology is presented.