Publications

Abstract not provided.

In many direct simulation Monte Carlo (DSMC) simulations, the majority of computation time is consumed after the flowfield reaches a steady state. This situation occurs when the desired output quantities are small compared to the background fluctuations. For example, gas flows in many microelectromechanical systems (MEMS) have mean speeds more than two orders of magnitude smaller than the thermal speeds of the molecules themselves. The current solution to this problem is to collect sufficient samples to achieve the desired resolution. This can be an arduous process because the error is inversely proportional to the square root of the number of samples so we must, for example, quadruple the samples to cut the error in half. This work is intended to improve this situation by employing more advanced techniques, from fields other than solely statistics, for determining the output quantities. Our strategy centers on exploiting information neglected by current techniques, which collect moments in each cell without regard to one another, values in neighboring cells, nor their evolution in time. Unlike many previous acceleration techniques that modify the method itself, the techniques examined in this work strictly post-process so they may be applied to any DSMC code without affecting its fidelity or generality. Many potential methods are drawn from successful applications in a diverse range of areas, from ultrasound imaging to financial market analysis. The most promising methods exploit relationships between variables in space, which always exist in DSMC due to the absence of shocks. Disparate techniques were shown to produce similar error reductions, suggesting that the results shown in this report may be typical of what is possible using these methods. Sample count reduction factors of approximately three to five were found to be typical, although factors exceeding ten were shown on some variables under some techniques.

Abstract not provided.

Modeling microscale heat transfer with the computational-heat-transfer code Calore is discussed. Microscale heat transfer problems differ from their macroscopic counterparts in that conductive heat transfer in both solid and gaseous materials may have important noncontinuum effects. In a solid material, three noncontinuum effects are considered: ballistic transport of phonons across a thin film, scattering of phonons from surface roughness at a gas-solid interface, and scattering of phonons from grain boundaries within the solid material. These processes are modeled for polycrystalline silicon, and the thermal-conductivity values predicted by these models are compared to experimental data. In a gaseous material, two noncontinuum effects are considered: ballistic transport of gas molecules across a thin gap and accommodation of gas molecules to solid conditions when reflecting from a solid surface. These processes are modeled for arbitrary gases by allowing the gas and solid temperatures across a gas-solid interface to differ: a finite heat transfer coefficient (contact conductance) is imposed at the gas-solid interface so that the temperature difference is proportional to the normal heat flux. In this approach, the behavior of gas in the bulk is not changed from behavior observed under macroscopic conditions. These models are implemented in Calore as user subroutines. The user subroutines reside within Sandia's Source Forge server, where they undergo version control and regression testing and are available to analysts needing these capabilities. A Calore simulation is presented that exercises these models for a heated microbeam separated from an ambient-temperature substrate by a thin gas-filled gap. Failure to use the noncontinuum heat transfer models for the solid and the gas causes the maximum temperature of the microbeam to be significantly underpredicted.

Abstract not provided.

A concurrent computational and experimental investigation of thermal transport is performed with the goal of improving understanding of, and predictive capability for, thermal transport in microdevices. The computational component involves Monte Carlo simulation of phonon transport. In these simulations, all acoustic modes are included and their properties are drawn from a realistic dispersion relation. Phonon-phonon and phonon-boundary scattering events are treated independently. A new set of phonon-phonon scattering coefficients are proposed that reflect the elimination of assumptions present in earlier analytical work from the simulation. The experimental component involves steady-state measurement of thermal conductivity on silicon films as thin as 340nm at a range of temperatures. Agreement between the experiment and simulation on single-crystal silicon thin films is excellent, Agreement for polycrystalline films is promising, but significant work remains to be done before predictions can be made confidently. Knowledge gained from these efforts was used to construct improved semiclassical models with the goal of representing microscale effects in existing macroscale codes in a computationally efficient manner.

The Microsystems Subgrid Physics project is intended to address gaps between developing high-performance modeling and simulation capabilities and microdomain specific physics. The initial effort has focused on incorporating electrostatic excitations, adhesive surface interactions, and scale dependent material and thermal properties into existing modeling capabilities. Developments related to each of these efforts are summarized, and sample applications are presented. While detailed models of the relevant physics are still being developed, a general modeling framework is emerging that can be extended to incorporate evolving material and surface interaction modules.