Jichul Kim

Comparative study of various release methods for polysilicon surface micromachining

This paper presents an objective comparison of various release techniques for polysilicon surface micromachining using identical test structures made by MCNC Multi-User MEMS Processes (MUMPs). Test structures of varying width and length were prepared using five different releasing procedures-evaporation drying with deionized water and methanol as final rinsing liquids, sublimation drying using p-dichlorobenzene and t-butyl alcohol, and CO/sub 2/ supercritical drying. Both sublimation drying methods as well as supercritical drying rendered good results with 2 /spl mu/m thick, 2 /spl mu/m gap polysilicon cantilevers up to 700 /spl mu/m in length. In addition, the systematic test in this study reveals, for the first time, that the maximum beam length obtainable increases as the beam width increases for the case of sublimation, opposite to the well known case of evaporation drying. In the course, we also introduce a new setup that considerably improves the way sublimation is used for releasing.

Development and characterization of a microfluidic chamber incorporating fluid ports with active suction for localized chemical stimulation of brain slices

We report a novel microfluidic chamber incorporating fluid ports with active suction to achieve localized chemical stimulation of brain slices. A two-level soft-lithography process is used to fabricate fluid ports with integrated injection and suction holes that are connected to underlying microchannels. Fluorescence imaging, particle tracking velocimetry, and cell staining are used to characterize flows around a fluid port with or without active suction to validate effective localization of injected chemicals. To demonstrate biological applicability of the chamber, we show an induction of cortical spreading depression (CSD) waves in mouse brain slices through controlled focal delivery of potassium chloride solution.

Capacitively induced high mobility two-dimensional electron gas in undoped Si/Si1-xGex heterostructures with atomic-layer-deposited dielectric

The authors demonstrate that a high mobility two-dimensional electron gas can be capacitively induced in an undoped Si∕Si1−xGex heterostructure using atomic-layer-deposited Al2O3 as the dielectric. The density is tuned up to 4.2×1011∕cm2, limited by the gate leakage current. The mobility increases with the density rapidly and reaches 5.5×104cm2∕Vs at the highest density. The observation of well developed quantum Hall states and two-dimensional metal-insulator transition shows that the devices are suitable for two-dimensional electron physics studies.

Exploration of thermolithography for micro- and nanomanufacturing

Lithography is a critical enabling technology for manufacturing micro- and nanoscale devices and structures. The present work explores alternative lithography techniques that pattern photoresist layers through selective thermochemical cross-linking. Microfabricated thin-film heaters are used as precisely defined heat sources to determine the thermal transport properties of photoresist layers and study the kinetics of cross-linking reactions. The present work identifies heating temperature, heating duration, and UV exposure dose as independent control parameters in thermolithography and demonstrates its potential for three-dimensional micro- and nanomanufacturing.

Brownian Microscopy for Simultaneous In Situ Measurements of the Viscosity and Velocity Fields in Steady Laminar Microchannel Flows

Brownian microscopy is an intriguing technique that enables in situ determination of the local fluid properties and velocity fields in microfluidic devices. We report application of Brownian microscopy to steady laminar microchannel flows and demonstrate accurate measurements of the viscosity of glycerin/water mixtures, as well as the velocity profiles. A theoretical model is developed to evaluate statistical errors in the measured Brownian diffusivity while accounting for complications associated with the electronic noise and finite exposure time of an imaging system. The model is validated using both Monte Carlo simulations and experiments.

Experimental Study of Heat Conduction in Aqueous Suspensions of Aluminum Oxide Nanoparticles

We perform a systematic experimental study of heat conduction in aqueous suspensions of aluminum oxide nanoparticles at volume concentrations up to 10%. We develop a micro-hotwire device to reduce experimental errors resulting from spatial or temporal temperature inhomogeneity within a sample. The volume concentration dependence of the thermal conductivity can be explained using the effective medium model with a physically reasonable set of parameters. The average particle size as well as the thermal conductivity is measured as a function of sample sonication time and temperature. The size of particles/aggregates in our nanofluid samples is much greater than the nominal particle size reported by the manufacturers and does not change appreciably with sonication for up to 24 hours. Our data do not reveal any anomalous enhancement in the thermal conductivity or strong temperature dependence reported in other previous studies. The discrepancy may reflect subtle differences in nanopowders or nanofluid preparation procedures that result in drastic difference in the size or shape of suspended particles/aggregates.

Microfluidic chamber with active suction ports for localized chemical stimulation of brain slices

We report a novel microfluidic chamber with active suction ports to locally deliver soluble neuromodulators into a brain slice in a precisely controlled manner. A two-level soft-lithography process is developed to fabricate injection/suction port assemblies integrated with microchannels. Fluorescence imaging is used to characterize the flows around an injection port while switching the suction port on and off. To demonstrate biological applicability of the chamber, we use a focal delivery of potassium chloride solution to induce cortical spreading depression (CSD) in mice brain slices.

The proximity effect of the regrowth interface on two-dimensional electron density in strained Si

We studied the density of two-dimensional electron gas (2DEG) in Si∕SiGe heterostructures as a function of the distance between the substrate-epilayer interface and the 2DEG layer. The 2DEG sheet density was observed to change from 2.2×1011to3.5×1011cm−2. Theoretical simulations are shown to be consistent with the experimental results within experimental errors. The slight deviations of the experimental results possibly come from temperature variation of the Sb dopant source during the growth of the Sb doping layer.

Analysis and Experimental Characterization of Statistical Errors in Brownian Microscopy

The Brownian diffusivity is a key parameter characterizing the stochastic motion of particles suspended in a liquid, which in turn reflects the rheological properties and thermodynamic states of the liquid. Fundamental understanding of statistical uncertainty in the Brownian diffusivity is critical since the number of Brownian steps tracked in practical experiments is limited. We develop an approximate analytic expression for statistical errors in the Brownian diffusivity, while accounting for electronic noise and other imperfections in a video microscopy system. The expression is validated using Monte Carlo simulations and microchannel based experiments. The present model allows us to make a priori estimate of overall experimental uncertainty and thereby helps improve the accuracy and reliability of various microrheological and biosensing techniques based on Brownian microscopyCopyright

On-chip characterization of the transport properties of liquids using microfluidic channel-based Brownian microscopy

We report techniques to characterize the transport properties as well as velocity fields of liquids flowing in microchannels. Microscopic particles of a precisely defined diameter are dispersed in a liquid of interest and their cross-stream Brownian motions are analyzed to determine the local Brownian diffusivity, which is in turn a function of the viscosity or mass concentration. We apply the proposed techniques to determine the viscosity and mass diffusivity of mixtures of glycerin and water, which agree well with the literature values. The techniques developed in this work will enable massively parallel characterization of novel liquids and their mixtures as well as in situ performance evaluation of microfluidic devices.