Session J06: Microscale Flows: Devices (8:00am - 8:45am CST)

Optimised hyperbolic microchannels for the mechanical characterisation of bio-particles

The transport of bio-particles in viscous flows exhibits complex dynamics such as reorientation, deformation and morphological transitions. Characterizing such behavior under controlled flows is key to understanding the mechanics of biological particles and the rheological properties of their suspensions. Here, we propose an innovative approach coupling numerically optimized design of microfluidic converging-diverging channels with a microscope-based tracking method to characterize the individual dynamics of bio-particles under homogeneous straining flow with high-quality images. We demonstrate experimentally the ability of the optimized microchannels to provide linear velocity streamwise gradients along the centerline, allowing for extended consecutive regions of homogeneous elongation and compression. We selected three test cases - DNA, actin filaments and protein aggregates - to highlight the ability of our approach for investigating the dynamics of objects from the biological world with a wide range of sizes, characteristics and behaviors.   

Design and Fabrication of Membrane-Based Sensors for In-Situ Capillary Pressure Measurement in Microfluidic Channels

Pressure is a fundamental quantity in virtually all problems in fluid dynamics from macroscale to micro/nano-scale flows. Although technologies are well developed for its measurement at the macro-scale, pressure quantification at the microscale is still not trivial. Yet precise pressure mapping at the microscale such as in microfluidics is imperative in a variety of applications, including porous medium flows and biomedical engineering. In particular, pore-scale capillary pressure is a defining variable in multiphase flow in porous media and has rarely been directly measured. Herein we aim to design and fabricate an on-chip sensor that enables quantification of capillary pressure in microfluidic porous media, called micromodels. The micromodel is fabricated in PDMS using soft lithography with a thin membrane incorporated that deflects with pressure variations in the fluid flow. Employing a microscope coupled with a high-speed camera, 2D pressure fields can be inferred from membrane deflections based on a pre-calibrated relation, along with other flow parameters, such as velocity fields and phase distribution. This study provides a novel method for in-situ quantification of capillary pressure and a renewed understanding of pore-scale physics of multiphase flow in porous media.   

3D Printing Aided Fluid Substance Dispensing in Micro-scale

Fluid dispensing is common in medical applications, spanning from marco- to micro-scale. While controlled dispensing is crucial to suitable dosage and optimized therapy, it remains challenging to clinical application in many cases due to constrains in fluid properties, material selection and device design. Recent development in 3D printing technology has intrinsically revolutionized the fabrication of medical device with tunable properties to enable precise microfluidic control. In this talk, we will first present the novel disperser design made with this technique and demonstrate the state-of-the-art control in micro-flow achieved by tuning the interfacial tension and viscosity of the fluid and the surface roughness of the device. We will further propose the dispensing mechanism tailored to representative fluids in clinical use with typical viscosity on the order of 1 mPa.s. Finally, the potential applications in microfluidic drug loading implant will be explored.   

Towards virtual channels in a microfluidic device

Manipulating particles is of interest in diverse fields of engineering and are of interest to the oil, food, and medical industries. Generally, manipulation activities carried out in micro-devices have a fixed design tailored for a specific task. It makes different analyses unfeasible on a single device. To address this issue, we designed a Hele-Shaw flow cell with "virtual" channels generated by uniform flow in the transverse direction and three inlets in the longitudinal axis. These three inlets can inject or withdraw fluid in the flow cell to deviate the streamlines. We use hydrodynamic forces as a way for non-contact particle manipulation because other non-contact techniques rely on complicated control systems which could interact with the particle properties. Since the depth-averaged velocity over the channel in a Hele-Shaw cell is irrotational, we use potential flow theory to predict the flow field. We have developed a particle path optimization algorithm to apriori determine the optimized path while confining the flow rate bounds and the variation in the flow rate. This device provides us the opportunity to integrate multiple functionalities such as particle separation, bringing particles close to each other, trapping onto a single device.   

Actuating Surface-Attached Post (ASAP) Arrays as High Aspect Ratio Actuators

Actuatable micropillar arrays are used in a wide range of applications, including controlling surface wettability, manipulating light refraction, and pumping and mixing fluids. Non-invasive magnetic actuation of pillars is favorable for many applications. Magnetically actuatable micropillars have been developed by molding magnetic elastomer nanocomposites. This fabrication route faces significant challenges including uncontrolled aggregation of particles and low optical clarity of the base layer for high quality imaging. We developed a fabrication method that uses centrifugation to control the amount of magnetic material in the micropillars while maintaining optical clarity in the base layer. We created posts with cross-sectional areas of less than 1 square micron and aspect ratios as high as 23:1, but demolding these fragile structures proved challenging. Our demolding method uses solvents to gently swell and de-swell cured elastomer to minimize high strain on the structures. We used our fabrication flexibility to create novel array designs including ``herringbone'' arrays of adjacent rectangular paddles that can be actuated with controlled phase. Such structures have been shown theoretically to evade the scallop theorem and pump fluids with reciprocal motions.   

Experimental and numerical study of microfluidic label-free viability cell sorting

Cell biomechanical properties often change in predictable ways with important cell phenotypes changes, such as cell loss of viability. We propose a biophysical approach for cell viability sensing, enumeration, and purification that is label-free and continuous. Experimentally using microfluidics with periodic cell-compressing constrictions, we show that we can separate viable cells from nonviable cells based on the difference in their stiffness with an enrichment factor of \textgreater 5 and an overall recovery of 95{\%}. We numerically study the effect of cell elasticity and adhesion on cell motion in the microchannel using lattice Boltzmann and lattice spring models. The sorting technology consists of a microfluidic channel with diagonal ridges that direct cells along different paths in a manner dependent on cell biomechanical properties. As a result, the sorted viable and nonviable cells are collected at different microchannel outlets. The platform can be used for cell characterization and purification either in-line with cell bioreactors or after cell manufacture and prior to administration to improve outcomes.   

A small-volume microcapillary rheometer with sample recovery

Rheology of small volumes is necessary for high value fluids such as biological samples, but many rheometers require milliliter volumes. A capillary device is constructed by connecting a single silica microcapillary to a round glass millimeter diameter capillary tube. Sample liquid is driven pneumatically to fill the microcapillary and partially fill the larger glass capillary. The glass capillary is mounted on an optical linear sensor array and the meniscus is tracked in real time to measure flow rate and enable flow reversal by switching the pressure differential with a pneumatic valve. The flow rate and the pressure drop are used to determine viscosity as a function of shear rate using capillary rheology equations. A given sample of at least 50 \textmu L can be measured over 2 to 3 decades within the shear rate range of 10 s$^{\mathrm{-1}}$ to 10$^{\mathrm{5}}$ s$^{\mathrm{-1}}$, and be essentially fully recovered. Validation is performed by comparing measurements of Newtonian and non-Newtonian fluids with reference measurements. The operational limits and sources of uncertainty are analyzed, including instrumentation error, meniscus effects, and inertial effects. Future work towards automation in well plates and temperature control are discussed.   

Computational optimization of inertial focusing microfluidics utilizing immersed boundary methods

Inertial focusing microchips have gained momentum in the past decade for size separation of particles and cells. Unlike many other techniques for size separation, inertial focusing is simple and passive allowing for greater applicability for point-of-care systems. One such chip design relies on an expanded channel to produce microvortices to capture and retain particles within a size range. Although these chips show promise for an elegant solution there are numerous parameters impacting the efficiency of the chip that require optimization including particle structure, flow conditions, cavity geometries, and particle concentration. Immersed boundary methods are implemented to model not only the complex geometry of the chip, but also allows for the forces on the particle to be resolved by integrating the total stress tensor over the surface of the particle as well as the impact of particles on the flow. Our simulations determined the optimal cavity geometry, as well as the relationship between the Reynolds number and the size of particles captured. With the two-way coupling of the fluid-structure interface considered we gained insight into the impact of concentration on particle capture.