CMU MEMS Laboratory Publication Abstract

 

in Ph.D. Thesis, August 2005, Carnegie Mellon University, Pittsburgh, PA.
Modeling and Simulation of Lab-on-a-Chip
Y. Wang
ABSTRACT:
This thesis presents a behavioral modeling and schematic simulation methodology for top-down Lab-on-a-Chip (LoC) design. The methodology involves decomposing a complex LoC system into a small set of elements. Each of the elements is associated with a parameterized behavioral model that describes its electric and biofluidic behavior. Key issues addressed include schematic representation, behavioral multi-physics modeling and numerical and experimental validation.

The modeling effort focuses on sample transport in LoC devices. Turn and Joule heating induced dispersion in electrophoretic separation chips are studied using the method of moments. The skew of the species band is effectively represented by a set of Fourier cosine series coefficients that are obtained analytically. These coefficients capture the effect of band skew on separation performance in various complex chip geometries (including multiple turns). Variations of sample concentration profiles in laminar diffusion-based micromixers are also derived using the Fourier cosine series representation. The model holds for arbitrary sample flow ratios and inlet concentration profiles, and accurately considers the overall effects of device topology, size and electric field on mixing performance. In addition, a simplified reaction model is developed and integrated with the separation and mixing models to perform system-level schematic simulations of an integrated LoC system. Simulation results at both element and system levels are validated against numerical and experimental data. Excellent accuracy (generally less than < 5% in relative error) and tremendous speedup (> 100 ×) have been achieved when compared with finite element analysis. The mixing model is also adapted to pressure driven flow and used to propose a novel concentration gradient generator design. The resulting modeling and simulation framework is a significant contribution to balancing the needs for efficiency and accuracy thus enabling iterative design of complex biofluidic LoC systems.

© 2005 Carnegie Mellon University, Department of Mechanical Engineering.
Full paper (PDF) (opens in new window).


This page was generated in 0.012222 seconds at 10:28:00 am EST on 20 Nov 2017.

overview | projects | people | publications | intranet | resources         © 1998-2009  Carnegie Mellon