Untitled Document

High-Pressure DC Pumps and High Throughput Nano-Spray

We have developed a nanoporous silica substrate that is ideal for DC pumps and DC sprays. The silica pore size is roughly the double layer thickness and hence represents the optimum pore size such that the entire interstitial space is charged. The monolithic structure of the silica substrate, which can be fused with a silica capillary during synthesis, also provides an integral monolithic structure to sustain the large (>4 atm) back pressure in a DC osmotic pump. In addition, a Nafion housing design prevents bubble invasion into the pumping channel. We are able to produce very large pressure that can be used to pump fluid in chip scale channels and chromatographs with large hydrodynamci resistance. The same high pressure is used to pressurize the meniscus at the tip of the capillary such that a DC spray can be ejected from the same chip. This represents the easiest interface to transfer samples between a chip and a mass spectrometer. It also allows us to carry out nanotube coating of surfaces by directing the charged drops to the target with an applied potential.

Relevant publications: Chen. Z.. Wang, P. and Chang, H.-C., "An Electro-osmotic Micropump based on Monolithic Silica for Micro-flow Analyses and Electrosprays", Anal. Bioanal. Chem., 382, 817(2005).

A silica monolith DC pump. The capilllary between the two reservoirs is filled with the monolith. The ground electrode is placed in the reservoir on the left and the polarized electrode in the right reservoir. A Nafion housing connects the two capillaries in the right reservoir to prevent bubble and ion pentration into the pumping chamber. Electrospray is emitted to the right.

Micron-sized fiber spunned by the AC spray. A polymer solution with a volatile solvent is sprayed out the of pump/spray unit above. Solvent evaporation in mid air results in polymer fibers or capsules depending on whether tip streaming or microjet occurs at the spray (see electrospray section).

In anticipation of nanoscale devices using nanotubes and nanoporous material, the group has begun to investigate fundamental transport laws at such small scales. Due to atomic and molecular interaction and because the transport time scale is comparable to the surrounding thermal bath time scale, the usual Fickian diffusion mechanism breaks down completely and new ones must be derived. We utlize MD simulation and coarse-graining strategies to derive large-scale continuum transport continuum equations for diffusion in nanotubes and zeolites and surface diffusion on metals and crystals. We have uncovered a new deterministic diffusion mechanism that is not driven by thermal noise. Instead, a Hamiltonian chaotic trajectory results due to coupling between a confined transversed degree of freedom and one or multiple longitudinal degrees of freedom. The projection of the chaotic trajectory in the longitudinal direction is noise like but it is not driven by any stochastic force, as shown below for methane transporting in a cylindrical nanopore in a zeolite. We use KAM type theories to estimate the transport rate of such deterministic diffusive mechanisms. In cases when both deterministic and stochastic effects are important, we have developed an RRKM type Kramer statistical theory to capture long-time tranport rate due to rare-juimps through bottle necks. This latter approach requires a model for the stochastic noise which we develop through short-time simulations or an analytical theory for the linear lattice thermal vibrations.