Fabricating PDMS Microfluidic Devices with Integrated Circuits

Dielectrophoresis (DEP) is a technique that manipulates particles using non-uniform electric fields by inducing dipoles in particles based on their dielectric properties relative to the surrounding medium; the Clausius-Mossotti factor determines whether particles experience positive DEP (attracted to high-field regions) or negative DEP (repelled from high-field regions), enabling applications in cell separation, particle sorting, and manipulation in microfluidic devices.

The dielectrophoretic force is F_DEP = 2π ε₁ K_CM a³ ∇(E₀²), where K_CM = (ε₂-ε₁)/(ε₂+2ε₁). Key dependencies: (1) Proportional to a³ (third power of size), (2) Depends on field gradient squared, (3) Independent of electric field sign, (4) Sign depends on K_CM. Since F_DEP ∝ a³, larger particles experience stronger forces than smaller ones. In microfluidic channels with electric field gradients, larger particles move toward regions of higher field intensity while smaller particles remain relatively stationary, enabling size-based separation of cells and particles. The DEP force direction depends on the Clausius-Mossotti factor sign. For particle trapping toward an electrode, F_DEP must be positive, requiring ε₂ > ε₁ (particle more dielectric than fluid). In microfluidic channels with asymmetric electrodes (hemispherical bottom, flat top), particles experience competing forces: DEP force (pulling toward electrode) and drag force (carrying with flow). For trapping to occur, F_DEP must exceed F_drag. The velocity profile in parallel-plate channels is parabolic: u(z) = (6u_avg/H²) z(H-z), where u_avg is average velocity.

Dielectrophoresis manipulates particles using non-uniform electric fields based on differences in polarizability between particles and surrounding medium. Particles experience forces proportional to the gradient of the square of the electric field. Positive dielectrophoretic (p-DEP) particles move toward high-field regions, while negative dielectrophoretic (n-DEP) particles move toward low-field regions. The force magnitude depends on particle shape, conductivity, permittivity, and field gradients. COMSOL implements dielectrophoresis through particle tracing physics combined with AC/DC electromagnetics, enabling separation of particles with different electrical properties for bioseparation and cell manipulation applications.

Dielectrophoresis is an electrokinetic phenomenon that enables the manipulation of neutral particles (with zero net charge) by applying a non-uniform electric field, where the dielectrophoretic force F_DP = P · ∇E acts on particles based on their electric dipole moment P and the spatial gradient of the electric field ∇E; unlike electrophoresis which relies on net particle charge, dielectrophoresis exploits the polarization properties of particles relative to their surrounding medium, making it particularly useful for separating and trapping biological cells and particles without requiring them to carry a net electrical charge.

This section introduces fundamental principles of particle manipulation in microfluidic systems. Electrophoresis moves charged particles in DC fields proportional to charge and field strength, while dielectrophoresis moves neutral particles in nonuniform fields proportional to dipole moment and electric field gradient squared. The Clausius-Mossotti factor (beta) determines particle response: positive beta moves particles to high-field regions, negative beta to low-field regions. This enables particle separation based on dielectric properties and contactless trapping at channel centers. Multiple particles experience electrostatic interactions causing chaining phenomena where particles form elongated chains along walls. The point-dipole approximation treats particles as point dipoles, valid when particles are much smaller than channel dimensions. The Maxwell stress tensor approach accounts for field deformation by particles, necessary for larger particles.

EIS measurements are sensitive to electrode geometry, particularly the distance between working and counter electrodes. Increasing spacer length raises uncompensated solution resistance, which affects observed Rct values even with identical electrode materials. Beyond material characterization, EIS enables development of impedimetric biosensors for detecting specific analytes. These biosensors use recognition elements (like antibodies) that bind target molecules, creating barriers that increase Rct. The change in Rct correlates with analyte concentration, enabling quantitative detection. This principle has been applied to detect viral proteins, where Rct increases proportionally with viral concentration, producing linear calibration curves for sensitive biosensing applications.

The Agilent xCELLigence RTCA esight platform employs impedance-based label-free biosensor technology to enable real-time, automated cell analysis, allowing researchers to monitor cellular events such as cell death, adhesion, and morphology changes continuously without requiring fluorescent labels or endpoint measurements, thereby significantly reducing experimental time while providing comprehensive kinetic data for applications in immunotherapy and drug discovery.

Electrochemical Impedance Spectroscopy (EIS) is a technique where a potentiostat applies a sinusoidal potential or current to an electrochemical system and measures the corresponding sinusoidal current or potential response; the technique uses Fourier transform to convert time-domain data to frequency-domain data, calculating impedance (the AC equivalent of resistance) from the ratio of potential and current amplitudes, along with phase shifts, which are then plotted as Bode plots (magnitude and phase vs. frequency) or Nyquist plots (real vs. imaginary impedance), allowing researchers to deconvolute different electrochemical phenomena occurring at characteristic frequencies (such as double-layer charging, solution resistance, charge transfer resistance, and molecular diffusion) that occur simultaneously but on different timescales, enabling both qualitative and quantitative analysis of electrochemical systems through equivalent circuit modeling.

Electrochemistry offers versatile real-time detection of molecules released from living cells without chemical pre-treatments or labeling. Using carbon fiber electrodes held at oxidizing potentials above serotonin's redox potential, researchers observe current spikes corresponding to serotonin release. Concentrations range from sub-micromolar to tens of micromolars, with release events occurring once per second or at varying frequencies depending on stimulation. However, traditional approaches only measure serotonin from the apical (top) side of gut epithelium, whereas the basolateral (bottom) side releases serotonin that stimulates enteric nerves. Gut-on-chip platforms provide convenient access to the basolateral surface. The developed platform integrates electrochemical impedance spectroscopy (EIS) on the top membrane surface to monitor cell coverage by detecting changes in electrical properties as cells attach, combined with cyclic voltammetry electrodes on the bottom surface for serotonin detection, enabling simultaneous physical and molecular monitoring.

Electrochemical impedance spectroscopy models electrode surfaces using equivalent circuits containing solution resistance, double-layer capacitance, and charge transfer resistance. At low frequencies, current flows through solution resistance and capacitive elements; at high frequencies, current preferentially flows through capacitors. Intermediate frequencies reveal contributions from both. Software analyzes impedance data across frequency ranges to determine equivalent circuit parameters. Charge transfer resistance changes proportionally to molecular binding extent, enabling quantitative detection of biomolecules. Carbon electrodes with pyrene NHS coupling chemistry provide cost-effective alternatives to gold electrodes for immunosensor development.

Roll-to-roll manufacturing represents a fundamental paradigm shift from conventional batch processing in display manufacturing. Unlike Gen 10 glass facilities costing billions, roll-to-roll uses printing-press-like continuous web handling. The key challenge is adapting photolithography to flexible plastic substrates that deform by hundreds of parts per million. HP's approach involved starting fresh without legacy glass-processing assumptions, leading to imprint lithography as the enabling technology. Imprint lithography solves alignment problems by encoding all patterning information into a single monolithic 3D stamp molded at the beginning of the process, eliminating multiple photolithography steps. This achieves 40nm resolution on 50-micron plastic at 5m/min throughput. Sustainability benefits include reduced mass (plastic weighs 50x less than glass), lower temperatures, and compact equipment. The zeroth-order sustainability rule states that reducing mass proportionally reduces energy consumption. Cleanroom requirements are dramatically reduced to localized environments. Equipment cost comparisons show roll-to-roll achieves approximately order-of-magnitude cost reductions compared to conventional photolithography.

Injection molding imposes constraints including the need for draft angles and the inability to create undercuts without expensive slide molds. Small, thin parts may appear strange on computer screens but become acceptable when manufactured. 3D printing offers an alternative for complex components with undercuts, allowing companies to provide data for customers to print their own versions. This approach requires additional design work to remove undercut sections but provides more flexibility and can reduce costs for certain parts.

Master durability improves through metallic conversion: template stripping deposits thick metal layers (750 nm silver) then mechanically strips away the film; electroplating coats patterns with seed layers and plates metals (nickel, gold, copper) to desired thicknesses. Both methods preserve dimensional accuracy, enabling long-lasting masters for industrial replication. Injection molding transforms nanoimprint masters into precision plastic parts at industrial scale, producing tens to hundreds of thousands of units daily. Current patterning speed is ~1,000 square microns per minute, comparable to electron beam lithography. Future developments include multi-tip arrays (10-tip linear arrays, commercial availability planned), 8-inch wafer-capable systems, high-speed laser systems for interconnects, and fully automated platforms with tip/sample exchange modules to enable large-scale manufacturing adoption.

3D printing and injection molding have different economic and practical advantages. Injection molding requires expensive molds (costing significant money) and minimum production runs to be economical. Mold design errors require complete remaking. 3D printing eliminates mold costs entirely, allows design changes without remaking, and provides faster turnaround times. For certain production volumes and object types, 3D printing can be more advantageous in terms of time, practicality, and design flexibility, though it may not completely replace injection molding for all applications.

This video explains why 3D printing is amazing but will likely never replace injection molding for mass production. While hobby 3D printers have become more affordable and capable than ever before, making it easier for individuals to create objects from their ideas, the existing manufacturing process of injection molding remains superior for high-volume production due to its monstrous efficiency.