Design of an interface to allow microfluidic electrophoresis chips to drink from the fire hose of the external environment

TLDR

The study presents an interface that enables automated sample introduction into an electrokinetic microchip without disturbing the device’s internal liquids. The interface uses a low‑resistance flow channel (300 µm deep, 1 mm wide, 15–20 mm long) etched in glass, whose resistance is 0.54–4.1 × 10⁶ times lower than that of the electrokinetic manifold, allowing high‑rate sample delivery. Experiments showed that up to 1 mL/min could be pumped through the interface without perturbing the electrokinetic channels, with peak heights and on‑chip reactions remaining unchanged and injection reproducibility between 1.5–4 %.

Abstract

An interface design is presented that facilitates automated sample introduction into an electrokinetic microchip, without perturbing the liquids within the microfluidic device. The design utilizes an interface flow channel with a volume flow resistance that is 0.54—4.1 × 106 times lower than the volume flow resistance of the electrokinetic fluid manifold used for mixing, reaction, separation, and analysis. A channel, 300 μm deep, 1 mm wide and 15—20 mm long, was etched in glass substrates to create the sample introduction channel (SIC) for a manifold of electrokinetic flow channels in the range of 10—13 μm depth and 36—275 μm width. Volume flow rates of up to 1 mL/min were pumped through the SIC without perturbing the solutions within the electrokinetic channel manifold. Calculations support this observation, suggesting a leakage flow to electroosmotic flow ratio of 0.1:1% in the electrokinetic channels, arising from 66—700 μL/min pressure-driven flow rates in the SIC. Peak heights for capillary electrophoresis separations in the electrokinetic flow manifold showed no dependence on whether the SIC pump was on or off. On-chip mixing, reaction and separation of anti-ovalbumin and ovalbumin could be performed with good quantitative results, independent of the SIC pump operation. Reproducibility of injection performance, estimated from peak height variations, ranged from 1.5—4%, depending upon the device design and the sample composition.