In recent years paper-based microfluidics have become an increasingly popular research topic, although it has been actively used for a long time, considering Litmus paper was reportedly used in the early 1800s [111]. There are several advantages of paper based microfluidic devices that make them attractive compared to microfluidic devices fabricated with more traditional materials such as glass, silicon and polymers, albeit these traditional materials have the advantage of mechanical robustness and reusability:
algorithm
___
20
Cellulose paper (or other porous materials such as woven and non-woven fabrics) is hydrophilic and can transport liquids through capillary action without active pumping action from complex integrated microfluidic pumps (see Figure 3) or external pumps, and without the need for external power supply to said pumps.
Figure 3 A planar, valveless, microfluidic pump using electrostrictive poly(vinylidene fluoride-trifluoroethylene), reprinted from Xia et al [112], Copyright 2006, with permission from Elsevier.
(1) Capillary action is relatively robust towards effects of gravity.
(2) There are several available techniques for fabrication of well-defined channels.
Hydrophobic patterns that constrain or direct liquids can be made by e.g.
photolithography, wax printing, screen printing and plasma treating. Alternatively cutting and lamination/encapsulation can be utilized to form a network of channels.
(3) Cellulose materials are inexpensive, and with simple inexpensive methods for forming channels, the paper-based substrate facilitates fabrication of low-cost disposable devices.
(4) A high surface area to volume ratio improves detection limits for colorimetric methods.
(5) Fibre networks in paper or related porous materials can store reagents in active form.
Although paper-based microfluidic devices can be fabricated with well-defined channels, fabrication of small dimensions is more problematic due to the fibrous nature of the
___
substrate compared to fabrication on e.g. silicon, but paper-based devices can process small sample volumes in the µL range, which facilitates low reagent consumption and chemical waste. On the other hand, transport in porous material is facilitated and restricted by the material properties of the material. It does not require an external power source, but the level of control in terms of flow speed and direction is limited, making, for example multistep operations and tuning difficult [113].
Many “Traditional” lab on chip devices and intricate microfluidic devices have been designed and created, but the success of bringing it from lab to commercialized product is still fairly limited [114]. This could be because of lack of communication between researchers and industry, development of modules that alone are not commercially viable or easily integrable in more complex systems, or that an ideal application for these system has yet to be discovered (an application in which microfluidic systems outperform more traditional disciplines both in terms of performance and cost efficiency), or that developed devices are still not simple and robust enough for practical commercial use.
Despite the success of urine dipsticks and pregnancy tests, and the promising nature of paper-based microfluidic diagnostic platforms, it has similar problems of realization into successful commercial products. Its simplicity in terms of relatively low cost materials and the absence of intricate external control mechanisms, translates well into feasibility of mass produced disposable devices, but does apparently not translate well into commercial success for new devices [115]. It is instead stuck in lab in the traditional two-dimensional form of hydrophilic paper with hydrophobic barriers either printed, stamped, cut, deposited, patterned, etc. with wax, polymers, photoresist [113], see Figure 4.
There has also been advances from 2D to 3D stacked/folded devices that facilitates packaging and multiplexing (see Figure 5) and advances in form of added functionality and enhanced control capability out-of-plane in terms of: Timing by including a tunable shunt to delay capillary flow [116], or by using a dissolvable bridge that functions as a timed off-switch (Figure 6) [117]. Timing can also be tuned in-plane by changing
algorithm
___
22
dimensions of the channel such as with varying width of the channel cross-section, but this is an “expensive” tuning procedure as it requires major changes to design.
Figure 4 examples of paper-based microfluidic 2D devices: a) ink jet printed hydrophilic-hydrophobic contrast patterns reprinted from Li et al. [118], Copyright 2010, with permission from Elsevier, b) Filter paper cut with CO2 laser, adapted from Evans et al.
[119] with permission of The Royal Society of Chemistry, c) Plasma treatment generation of microfluidic patterns, reprinted from Li et al. [120]
Figure 5 Three dimensional stacked or folded paper-based microfluidic device: a-d) layered paper-based microfluidic device with crossing paths, reprinted from Martinez et al. [121] Copyright 2008 National Academy of Sciences, e) Foldable paper-based microfluidic device for multiplexed sandwich chemiluminescence, adapted from Ge et al.
[122] with permission of The Royal Society of Chemistry.
___
Figure 6 Sideview of modified porous media channel: a) Shunt for tuning of capillary flow speed, adapted with permission from Toley et al. [116], Copyright 2013 American Chemical Society, b) dissolvable bridge, adapted with permission from Houghtaling et al.
[117], Copyright 2013 American Chemical Society.