Closable Valves and Channels for Polymeric Microfluidic Devices

Abstract

This study explores three unique approaches for closing valves and channels within microfluidic systems, specifically multilayer, centrifugally driven polymeric devices. Precise control over the cessation of liquid movement is achieved through either the introduction of expanding polyurethane foam, the application of direct contact heating, or the redeposition of xerographic toner via chloroform solvation and evaporation. Each of these techniques modifies the substrate of the microdevice in a different way. All three are effective at closing a previously open fluidic pathway after a desired unit operation has taken place, i.e., sample metering, chemical reaction, or analytical measurement. Closing previously open valves and channels imparts stringent fluidic control-preventing backflow, maintaining pressurized chambers within the microdevice, and facilitating sample fractionation without cross-contamination. As such, a variety of microfluidic bioanalytical systems would benefit from the integration of these valving approaches.

Keywords: centrifugal; closable valving; contact heating; expandable foam; microfluidic; redeposition.

Figure 1

Schematic diagrams of the three putative closure methods. Method (A): Foam producing reagents are manually added to upstream reagent chambers within the device. Centrifugal pumping combines and mixes the two components. Upon mixing, the rapidly expanding polyurethane foam fills and blocks the target downstream channel. Method (B): Heat and pressure are applied directly to a previously opened laser hole. Controlled compression and melting of the polymeric layers and adhesives induce intermingling and permanent bonding of the polymeric materials. Upon cooling, a permanent, channel sealing weld is formed. Method (C): Chloroform is added to a previously opened laser patch, dissolving a portion of the xerographic toner. Rapid evaporation of the chloroform redeposits the toner into a uniform layer that covers the previously ablated laser hole. (D) Microdevices featured in this manuscript consisted of five laminated PeT films that were prepared and assembled using the “print-cut-laminate” (PCL) method of fabrication.

Figure 2

Blocking a channel with expanding polyurethane foam. (A) Schematic of the microchip prior to testing. The two reagents are separated, with architecture to mix them before sample is diverted. (B) Prior to channel blockage, dye solution freely flows through the open channel architecture. (C) Upon foam expansion and curing, the primary channel (2 mm diameter) is blocked. When new sample is added and the microdevice spun, fluid is diverted into the previously unfavorable flow path.

Figure 3

Schematic diagram and photographs of the gantry platform for bringing the heating element into contact with the disc. (A) Schematic diagram of the contact heating approach. (B) A gantry arm was 3D printed, outfitted with two self-aligning linear sleeve bearings (C), and suspended between two standard laboratory support stands. The soldering stylus was inserted into the gantry arm, held in place with set screws, and connected to the adjustable power supply. (D) Calibration of stylus temperature with a digital thermometer and a T-type thermocouple. (E) Close-up photograph of the soldering stylus in contact with a previously opened laser valve (2 × 2 mm).

Figure 4

Blocking a channel with expanding polyurethane foam: Percent fluid diversion as a function of elapsed time. When the foam mixture was allowed to expand and cure for ≥12 min, 100% fluid diversion was observed. n = 4 attempted closures per time point. Note: When curing time > 11 min, the error bars are quite small and are not readily visible at 12- and 14-min time points.

Figure 5

Main effect plots of the selected generalized linear model. The main effects for this model were the continuous predictor variables applied pressure, temperature, and contact time (power + height + time). Each of these marginal effects plots reflects the probability of successful channel closure when the other predictor variables are held constant at their respective means. Light blue traces indicate a 95% confidence interval for each probability plot.

Figure 6

Evaluation of valve closure via the contact heating method over a wider temperature range. (A) When pressure and time were held constant (336 ± 75 psi and 3 s, respectively) and as stylus temperature exceeded 230 °C, a sharp increase in channel closure success was noted. (B) Broadly speaking, this effects plot visualization of the generalized linear model suggests a rapid increase in the probability of successful closure as stylus temperature increases from 200 to 270 °C.

Figure 7

Chloroform redeposition of toner. A loaded four-domain chip prior to any fluid flow or valving events (A) and after following completion of all fluid flow and valving events (B). Valves were sealed using chloroform for dissolution and redeposition of xerographic toner patches. Scanned images were acquired with an Epson Perfection V100 Photo desktop scanner (Japan). Bar plot (C) comparing initial and final hue values for each colored dye. Differences in initial and final hue values were negligible, suggesting successful valve closures via chloroform redeposition of xerographic toner.