Dynamics of Capillary-Driven Flow in Paper-Based Microfluidic Devices

摘要

Microfluidics has shown considerable potential for improving diagnostics, food safety and environmental analysis. Characteristics such as rapid sample processing, portability and precise control of samples have made microfluidic technologies attractive and alternative to conventional laboratory-based approaches. The recent emergence of paper as a substrate for construction of low-cost sensors has been propelled by the promise of getting the best compromise between affordability, performance and simplicity. Paper is inexpensive and abundant and can transport fluids without the need for external pumps, making it a potential substrate for construction of disposable and equipment-free sensors. Fundamental to fabricating highly effective and sensitive paper-based devices is the ability to transport and manipulate fluids within the devices. Two main approaches for capillary-driven liquid transport have been presented in literature: flow in porous paper matrix, guided by printed barriers1,2, and (ⅱ) flow on paper surface, confined into hydrophilic pathways on otherwise hydrophobic surface3,4. A mix of these, channel flow in porous pigment coating defined by printed hydrophobic boundaries, has also been proposed5. While numerous functioning paper-based microfluidic demonstrators have been reported, some limitations of the devices, including large required sample volumes, sample evaporation and retention in paper matrix as well as low liquid flow rates, have often been overlooked. The current work analyzes the capillary-driven flow dynamics of paper-based microfluidic flow and proposes a new approach to speed up the flow and reduce sample evaporation. Herein, the acceleration of the flow is achieved by forcing the sample liquid into a narrow gap geometry formed by two surfaces separated by spacers, as shown in Figure 1. The top surface is hydrophobic, while the bottom one has either a planar microfluidic channel on paper, or a channel fabricated inside the paper. The closed-channel flow system showed increased spreading distances and accelerated liquid flow. An average flow rate improvement of 200% was obtained for planar microchannels on paper, and a 100% increase for in-paper flow in filter papers. The increase was attributed to an increase in the driving force for the flow, brought about by the forced bending of the meniscus by the hydrophobic top surface at the advancing liquid front.The analysis of flow dynamics revealed a linear dependence of the meniscus position to the square root of time: I = Zt~(0.5). For open channel flow on paper, Z ≈ 0.9(σV~(2/3))/(μw), where σ is liquid surface tension,μ liquid viscosity, V the deposited liquid volume, and w the channel width. The closed channel flow on paper surface was also found to be square root of time dependent, but the pre-factor Z is more complicated and includes parameters such h (the gap size), Θ_T and Θ_B (the liquid contact angles against the top and bottom surfaces), ε (the channel aspect ratio), and R_d (the deposited droplet radius):I= {((16σh~2)/(π~4μ))[((cosθ_T+cosθ_B)/h)-(2/w-1/(R_d))](1-(1/(π/(2∈)))tanh(π/(2∈)))t}~(1/2)For the in-paper fabricated microfluidic channels with gap flow, further modifications to the pre-factor Z were needed to account for porosity and thickness of the paper.The current work presents an approach for fluid manipulation in paper-based sensors, which accelerates the flow in microfluidic channels, reduces the required samples amounts, and reduces the sample evaporation if that is an issue for the end-use application. Following the proposed approach, a simple parallel assay demonstrator fabricated on paper is shown in Figure 2.