Optical technology is rapidly finding novel applications in several exiting bioanalytical, biological, and biomedical applications. Optical beams are increasingly used for bio-fluidic sample manipulation in BioMEMS devices replacing convectional mechanical, electrostatic, and electrokinetic methods. This paper presents novel multiphysics computational approach for modeling optical interaction with fluidic, thermal, mechanical, and biological processes. We present a model of optical manipulation of particles and biological cells with laser beams. Computational results are compared to available experimental data from laboratory experiments and from practical engineered optical bio microdevices. The modeling approach is demonstrated on selected specific applications of optical manipulation of micro spheres, micro cylinders, and optical manipulation and sorting of biological cells in microfluidic cytometers.
The application of the frequency domain and steady-state diffusive optical spectroscopy (DOS) and steady-state near infrared spectroscopy (NIRS) to diagnosis of the human lung injury challenges many elements of these techniques. These include the DOS/NIRS instrument performance and accurate models of light transport in heterogeneous thorax tissue. The thorax tissue not only consists of different media (e.g. chest wall with ribs, lungs) but its optical properties also vary with time due to respiration and changes in thorax geometry with contusion (e.g. pneumothorax or hemothorax). This paper presents a finite volume solver developed to model photon migration in the diffusion approximation in heterogeneous complex 3D tissues. The code applies boundary conditions that account for Fresnel reflections. We propose an effective diffusion coefficient for the void volumes (pneumothorax) based on the assumption of the Lambertian diffusion of photons entering the pleural cavity and accounting for the local pleural cavity thickness. The code has been validated using the MCML Monte Carlo code as a benchmark. The code environment enables a semi-automatic preparation of 3D computational geometry from medical images and its rapid automatic meshing. We present the application of the code to analysis/optimization of the hybrid DOS/NIRS/ultrasound technique in which ultrasound provides data on the localization of thorax tissue boundaries. The code effectiveness (3D complex case computation takes 1 second) enables its use to quantitatively relate detected light signal to absorption and reduced scattering coefficients that are indicators of the pulmonary physiologic state (hemoglobin concentration and oxygenation).
The response of the biological cells to optical manipulation in the bio-microfluidic devices is strongly influenced by the flow and motion inertia. There is a variety of microfluidic architectures in which both the cell-fluid interaction and the optical field are driving forces for segregation and manipulation of the cells. We developed a computational tool for analysis/optimization of these devices. The tool consists of two parts: an optical force library generator and the computational fluid dynamics solver with coupled optical force field. The optical force library can be computed for spherical and non-spherical objects of rotational symmetry and for complex optical fields. The basic idea of our method is to a) represent an incident optical field at the biological cell location as an angular spectrum of plane waves; b) compute the scattered field, being a coherent superposition of the scattered fields coming from each of the incident plane waves, with the powerful T-matrix method used to compute the amplitude matrix; c) use the incident and computed scattered fields to build a spatial map of optical forces exerted on biological cells at different locations in the optical beam coordinate system, and d) apply the library of optical forces to compute laser beam manipulation in microfluidic devices. The position and intensity of the optical field in the microfluidic device may be dynamic, thus optical forces in microfluidic device are based on the instantaneous relative location of the cell in the beam coordinate system. The cell is simulated by the macroparticle that undergoes mutual interactions with the fluid. We will present the exemplary applications of the code.
The design of the next generation of vertical-cavity surface-emitting lasers (VCSELs) will greatly depend on the availability of accurate modeling tools. Comprehensive models of semiconductor lasers are needed to predict realistic behavior of various laser devices, such as the spatially nonuniform gain that results from current crowding. Advanced physics models for VCSELs require benchmark quality experimental data for model validation. This paper presents preliminary results of a collaborative effort at ARL to fabricate and experimentally characterize test optoelectronic structures and VCSEL devices, and at CFDRC to develop comprehensive multiphysics modeling, design and optimization tools for semiconductor lasers and photodetectors. Experimental characterization procedure and measurements of optical and electrical data for oxide-confined intracavity VCSELs are presented. A comprehensive multiphysics modeling tools CFD-ACE+ O’SEMI has been developed. The modeling tool integrates electronic, optical, thermal, and material gain data models for the design of VCSELs and edge emitting lasers (EELs). This paper presents multidimensional simulation analysis of current crowding in oxide-confined intracavity VCSELs. Computational results helped design the test structures and devices and are used as a guide for experimental measurements performed at ARL.
Simulation and design of microfluidic systems requires various level models: high-fidelity models for design and optimization of particular elements and devices as well as system-level models allowing for VLSI-scale simulation of such systems. For the latter purpose, reduced or compact models are necessary to make such system simulations computationally feasible. In this paper, we present a design methodology and practical approach for generation of compact models of microfluidic elements. In this procedure we use high-fidelity 3D simulations of the microfluidic devices to extract their characteristics for compact models, and subsequently, to validate the compact model behavior in various regimes of operation. The compact models are generated automatically in the formats that can be directly used in SPICE or SABER. As an example of a nonlinear fluidic device, the generation of compact model for 'Tesla valve' is described in detail. Tesla valve is one of the no-moving- parts valves used in micropumps in MEMS. Its principle of operation is based on the rectification of the fluid, so it may be considered as a 'fluidic diode'.
In this work, a new fully automated geometrical modeling and meshing tool is described. It imports standard layout formats, images, and 3D boundary representations s. A 3D model is then generated by simulating 3D operations specified by the process data or the user. A 3D finite element mesh with tagged boundary and volume conditions is then automatically created. The automatic generation of 3D model and mesh takes typically a couple of minutes on a current PC machine. The paper will present the geometry/meshing engines, user interfaces, and will demonstrate them on a range of microsystem applications.
This paper presents a framework for modeling essentially 1D devices and components embedded in multi-dimensional spaces. The main characteristic and main advantage of the new methodology is that the 1D and multi-dimensional objects or domain are meshed completely independently of each other, without regard to their relative alignment or location, and subsequently combined into a single, unified composite mesh. The coupling of the solution between the different domains is handled fully-automatically in the solver, entirely through exchange of source terms between these domains of differing dimensionality. The source terms are evaluated locally on a cell-by-cell basis, depending on the solution values in these domains and the manner in which the 1D grids intersect the multi-dimensional grids. The capabilities and usefulness of the method are demonstrated with several examples.
Modern microsystems use integrated sensors, controllers and actuators, and involve multi-physics phenomena. Detailed and accurate multi-physics based simulations are a key to device optimization and successful designs. In recent years, CFD- ACE+, a fluid flow solver has been validated and demonstrated on different MEMS devices involving coupled fluid flow, heat transfer, structural mechanics and electrostatics. Presented here are results of dynamic devices such as micropumps with dynamic valves and membrane micropumps as well as priming of a capillary pump and novel valves that use fluid surface tension for operation. Comparisons with experimental and other data are also presented to demonstrate the accuracy of multi-physics simulations. The capabilities of this state-of-the-art software and its usefulness in MEMS design environment is demonstrated.
Computer aided design (CAD) systems are a key to designing and manufacturing MEMS with higher performance/reliability, reduced costs, shorter prototyping cycles and improved time- to-market. One such system is CFD-ACE+MEMS, a modeling and simulation environment for MEMS which includes grid generation, data visualization, graphical problem setup, and coupled fluidic, thermal, mechanical, electrostatic, and magnetic physical models. The fluid model is a 3D multi- block, structured/unstructured/hybrid, pressure-based, implicit Navier-Stokes code with capabilities for multi- component diffusion, multi-species transport, multi-step gas phase chemical reactions, surface reactions, and multi-media conjugate heat transfer. The thermal model solves the total enthalpy from of the energy equation. The energy equation includes unsteady, convective, conductive, species energy, viscous dissipation, work, and radiation terms. The electrostatic model solves Poisson's equation. Both the finite volume method and the boundary element method (BEM) are available for solving Poisson's equation. The BEM method is useful for unbounded problems. The magnetic model solves for the vector magnetic potential from Maxwell's equations including eddy currents but neglecting displacement currents. The mechanical model is a finite element stress/deformation solver which has been coupled to the flow, heat, electrostatic, and magnetic calculations to study flow, thermal electrostatically, and magnetically included deformations of structures. The mechanical or structural model can accommodate elastic and plastic materials, can handle large non-linear displacements, and can model isotropic and anisotropic materials. The thermal- mechanical coupling involves the solution of the steady state Navier equation with thermoelastic deformation. The electrostatic-mechanical coupling is a calculation of the pressure force due to surface charge on the mechanical structure. Results of CFD-ACE+MEMS modeling of MEMS such as cantilever beams, accelerometers, and comb drives are discussed.
Computational design of MEMS involves several strongly coupled physical disciplines, including fluid mechanics, heat transfer, stress/deformation dynamics, electronics, electro/magneto statics, calorics, biochemistry and others. CFDRC is developing a new generation multi-disciplinary CAD systems for MEMS using high-fidelity field solvers on unstructured, solution-adaptive grids for a full range of disciplines. The software system, ACE + MEMS, includes all essential CAD tools; geometry/grid generation for multi- discipline, multi-equation solvers, GUI, tightly coupled configurable 3D field solvers for FVM, FEM and BEM and a 3D visualization/animation tool. The flow/heat transfer/calorics/chemistry equations are solved with unstructured adaptive FVM solver, stress/deformation are computed with a FEM STRESS solver and a FAST BEM solver is used to solve linear heat transfer, electro/magnetostatics and elastostatics equations on adaptive polygonal surface grids. Tight multidisciplinary coupling and automatic interoperability between the tools was achieved by designing a comprehensive database structure and APIs for complete model definition. The virtual model definition is implemented in data transfer facility, a publicly available tool described in this paper. The paper presents overall description of the software architecture and MEMS design flow in ACE + MEMS. It describes current status, ongoing effort and future plans for the software. The paper also discusses new concepts of mixed-level and mixed- dimensionality capability in which 1D microfluidic networks are simulated concurrently with 3D high-fidelity models of discrete components.
A versatile electro-thermal and optical numerical simulation tool has been developed that can handel various types of vertical-cavity surface-emitting laser designs. As an example, we consider a cylindrically symmetric top-emitting mesa laser with an oxide window. The control-volume-based model provides spatial distributions of carrier density in the active region plane as well as electrical potential, current density and temperature distributions in the entire device. We focus our attention on current redistribution processes, and find out that the current crowding effect is strongly affected by the presence of oxide window. In addition to the geometry-related crowding, we describe the current self-distribution (CSD) effect related to spatially non-uniform stimulated recombination. The CSD is shown to counteract the spatial hole-burning caused by carrier consumption through intense stimulated emission.
An integrated device and package 3D model is developed to computationally investigate the thermal crosstalk in arrays of proton-implanted top-surface emitting lasers. A self- consistent electro-thermo-opto model is employed for the device. The anisotropic thermal property is considered for the package model. Temperature dependency of critical device and material properties is included, as well as multiple heat generation mechanisms. Effects of spacing on lasing performance and non-uniformity of VCSEL arrays are found significant. Thermal crosstalk becomes worse for increased sizes and packaging densities of laser arrays. Degraded laser performance is found due to the thermal crosstalk, especially for the lasers closest to the center of the array package.