Three-dimensional hydrodynamic flow and particle focusing using four vortices Dean flow
We present a hydrodynamic 3D flow and particle focuser built on a single layer microfluidic platform. In contrast to prior work, our device is high throughput and enables the control of location and size of the sample core but requires only single sheath flow. In addition to flow focusing, 3D focusing of 3.2 um particles is demonstrated.
- Design and operating principles
The design and mechanism of the 3D hydrodynamic focusing device is illustrated in Fig. 1. The device has two inlets, one for the sheath flow (slice 1, width 200 ¥ìm) and the other for the sample flow (slice 2, width 100 ¥ìm). The channel height is 100 ¥ìm throughout. The sheath fluid flowing through the curve meets the sample flow at 60 degrees of the curve where a pair of counter-rotating vortices (Dean vortex) is developed in the main channel and starts pulling the sample flow in between towards the opposite side of the channel, focusing it vertically (slices 3-5). Next, as the fluid turns more than 90 degrees of the curve, the profiles of Dean vortex evolves into a combination of multiple vortices. In our system with flow rates of 288 ml hr-1 and 18 ml hr-1 for sheath and sample flow respectively, there builds up another pair of small vortices near the outer wall of the channel in addition to the big pair (slice 6). These small vortices generate a counter flow against a transverse drift, create a stagnation point near the center of the channel, and eventually focus the sample horizontally (slices 7-8). As the fluids escape from the curve and the vortices fade out through diffusion, the sample core maintains its position at the center of the channel being three dimensionally focused (slice 9).
Schematic diagram showing the 3D hydrodynamic focusing process. Slices 1-9 show the cross-sectional profiles of the rhodamine concentration in the device. The three insets show the velocity vector fields in the channel cross-sections along with the contours of the rhodamine concentration. The blue arrows in the middle inset indicate that a pair of small vortices formed near the outer wall of the channel.
The cross-sectional images of sample (rhodamine dissolved water) concentration profile taken from z-stack scanning of confocal microscopy are shown in Fig. 2-4. These not only prove the 3D flow focusing ability of our device but also shows that location of the sample core can be controlled by channel curvature (Fig. 2) and sheath flow rate (Fig. 4) and narrow focusing can be achieved by reducing sample flow rate (Fig. 3). 3.2 ¥ìm particle focusing and its distribution are illustrated in Fig. 5 and 6, respectively. All the experimental results are compared with numerical simulation.
The left-hand panel shows LIF confocal microscopy images of the 3D flow focusing under each curvature condition (a, d) 0.44, (b, e) 0.27, and (c, f) 0.15. The top views (a-c) and cross-sectional views (d-f) of the fluorescence intensity profiles of rhodamine are shown. The right-hand panel shows the CFD-aided simulated distributions of the rhodamine concentration under the same geometric conditions. Shown are the top views (g-i) and cross-sectional views (j-l) under each curvature condition: (g, j) 0.44, (h, k) 0.27, and (i, l) 0.15.
Cross-sectional images of the rhodamine intensity profiles measured by LIF confocal microscopy (left column) and the rhodamine concentrations calculated using the CFD simulations (right column) of the 3D flow focusing. The curvature of the channel was 0.44.
Cross-sectional images of the rhodamine intensity profiles measured by LIF confocal microscopy (left column), and the rhodamine concentrations calculated using CFD simulations (right column), for the 3D flow focusing applications.
Demonstration of the particle focusing effects in (a) the experimental study, and (b) the numerical simulation. A 3D view of the simulation results clearly demonstrates that the particles were three-dimensionally focused. Image (b) shows the overlapping images of the particles in the cross-sectional planes over the length of the stream. Image (a) shows the overlapping images of fluorescent particles as measured by LIF confocal microscopy over an 850 ¥ìm length of the final stream. The color bar represents the magnitude of the particle velocity.
Number distribution of 3.2 ¥ìm particles on (a) x-axis and on (b) y-axis.
Continuous separation of particles in a PDMS microfluidic channel via travelling surface acoustic waves (TSAW)
We demonstrate a simple and efficient device for the continuous label-free separation of microparticles using travelling surface acoustic waves (TSAW). A focusing interdigitated unidirectional transducer released high frequency (133.3MHz) TSAW normal to the fluid flow direction to segregate 3 µm particles from 10 µm particles with a separation efficiency of 100%. The TSAW based separator does not necessitate a tight alignment of PDMS microchannel with the transducer.
(a) Schematic diagram showing a TSAW-based particle separator. (b) The top and side views of the separation region are shown. Particles travel over a lateral deflection distance, depending on their size. Acoustic energy radiates from the substrate surface at the Rayleigh angle (¥è) when a TSAW interacts with the fluid. (c) Plotted is the TSAW absolute amplitude for a 133.3 MHz actuation frequency. (d) Fabricated TSAW device alongside a US one cent coin. The PDMS microchannel was bonded to the LiNbO3 chip. The inset shows a close-up view of the interlocking metal electrodes. Red dye (Rhodamine) is used to highlight the microchannel and plastic tubing.
- Experimental setup
The CAPS chip was placed on the stage of a microscope (Olympus BX53) outfitted with a high speed camera (pco. 1200 hs). A micro-syringe pump (Cetoni GmbH neMESYS) was used to simultaneously inject sample and sheath flow into separate inlets of the microfluidic channel. A radio frequency (RF) signal generator (Agilent N5181A) was used to produce a 133.3 MHz and -11.1 dBm sinusoidal signal that was then amplified by a power amplifier (miniCircuits ZHL-1-2W). The amplified signal of 23.7 dBm (235 mW) was fed into FUT. TSAW originating from the transducer passed through the acoustic window and interacted with the fluid inside the microchannel.
Experimental setup includes a high frequency signal generator connected with a power amplifier; a DC power supply to run the amplifier; a micro syringe pump to inject fluid inside the microfluidics channel.
Microchannel alignment with the trasducer: L-shape markers are aligned with the side walls of acoustic window. A misalignment of the microchannel shifts the centerline 200 µm in the upward direction.
Schematic diagram of the PDMS microchannel is shown in the center. The separation of 3 µm particles and 10 µm particles was achieved under a flow rate of 100 µL/h (particle sample flow: 25 µL/h, sheath flow: 75 µL/h) and an input power of 235mW. When the TSAW were turned OFF, no separation distance was observed between the particles of the two sizes. Once the TSAW were turned ON, a distinct separation distance could be observed. Images are stacked to show the trajectories followed by the particle.
TSAW alignment with the separation zone is shown. (a) TSAW amplitude profile (b) Separation zone along with time steps (c) Trajectories followed by 3µm (left) and 10 µm (right) particles (d) Acoustic streaming flow traced by 1 µm particle. The arrows show the vortices formed by the acoustic streaming flow (e) PIV measurements. Image (c-e) are obtained through an image processing software (ImageJ.).
(a) Deflection of 10 µm particles is plotted against the input power for variable flow rate. Inset shows the deflection against flow rate in microchannel for input power of 0 and 150 mW. Power fitted trend lines are also shown. (b) Deflection of 3 µm and 10 µm is plotted against power for a constant flow rate of 100 µL/h. For a power input of approx. 230 mW and flow rate of 100 µL/h, all the 10 µm particles travel a maximum deflection distance making them flow in a line next to the side wall.
When the TSAW were turned OFF (0mW), all of the particles flowed through outlet 1. When the TSAW were turned ON (151mW), 3 µm particles were collected at outlet 1, whereas 100% of the 10 µm particles passed through outlet 3.
Acoustic streaming induced by the high frequency (133.3MHz) TSAW is shown. Particles of diameter 1µm are used as the acoustic radiation force on smaller particles is minimal compared to the Stokes drag force. Particles are shown circulating with the acoustic streaming flow.
Separation of 3µm particles from 10µm particles is successfully demonstrated in this short clip. In the first part when TSAW are OFF, the particles are flowing together but later when TSAW are ON, the separation of particles is evident. The video is captured by a high speed camera (pco. 1200 hs) and slowed down for better visualization.
Refractive index driven optofludic particle manipulation
- Objective -
To describe optofluidic particle manipulation based on the refractive index contrast between the particle and the surrounding medium. High-refractive-index and low-refractive-index particles could be separated in terms of their refractive indices because the direction of the gradient forces acting on such a particle is opposite the direction of the forces acting on a high-refractive-index particle.
- Method -
Schematic diagram of the experimental setup. The laser wavelength was 532 nm. The inset shows the principle underlying the present optofluidic particle manipulation system. The laser beam center axis was aligned adjacent and parallel to the sidewall of the microfluidic channel.
The gradient forces acting on a particle were calculated to predict and interpret the particle behavior. High-refractive-index and low-refractive-index particles, prepared from polystyrene latex (PSL) and hollow glass particles with refractive indices of 1.59 and 1.22, respectively, were employed.
The gradient force acting on PSL (high-refractive-index) and hollow glass (low-refractive-index) particles under various laser beam powers. The inset illustrates the principles underlying the positive and negative gradient forces.
Trajectories of a PSL (high-refractive-index) particle.
Trajectories of a hollow glass (low-refractive-index) particle.
Trajectories of the doubly attached hollow glass particles. The motions of these aggregates were similar to the motions of a single particle, possibly because particles near the laser beam axis dragged the other particles along under the higher gradient force.
To describe the optical mobilities of blood cell components. Blood cells are heterogeneous, and their optical behaviors depend on size, morphology, and other optical properties.
A schematic diagram of the experimental setup. The optical mobility could be expressed in terms of the
retention distance, the flow, and the laser conditions.
Despite the sample heterogeneity, the optical behaviors could be predicted using the instrumental parameters, including the flow velocity (U), laser power (P), laser beam waist radius (x0), and refractive index of the outer media (n0).
Snapshots of the cell trajectories for (a) RBCs, (b) lymphocytes, (c) granulocytes, and (d) monocytes.
Retention distance of the blood cell components in the COPS.
Optical mobilities of lymphocytes (N¨ù59), granulocytes (N¨ù36), monocytes (N¨ù42), and RBCs (N¨ù81). Error bars represent the standard deviations across each cell population.