paper-based microﬂuidic devices: emerging themes and...

Paper-Based Microfluidic Devices: Emerging Themes andApplicationsYuanyuan Yang,† Eka Noviana,† Michael P. Nguyen,† Brian J. Geiss,‡ David S. Dandy,§

and Charles S. Henry*,†,§

†Department of Chemistry, Colorado State University, Fort Collins, Colorado 80523, United States‡Department of Microbiology, Immunology, and Pathology, Colorado State University, Fort Collins, Colorado 80523, United States§Department of Chemical and Biological Engineering, Colorado State University, Fort Collins, Colorado 80523, United States

■ CONTENTS

Fabrication 71Hydrophobic/Solvent Barrier 72Deposition 73Flow and Injection Control 74Three-Dimensional Devices 75Incorporating Nonsensing Electrodes 75

Colorimetric Detection 75Detectors and Readout 75

Reflectance-Based Measurement 75Transmittance-Based Measurement 77Instrument-Free Measurement 77

Biomedical Applications 77Enzymatic Methods 77Immunoassays 78Other 79

Environmental Applications 79Other Applications 80

Electrochemical Detection 80Electrodes 80

Carbon Electrodes 81Metallic Electrodes 81

Biological Applications 82Glucose Sensors 82Immunosensors 84Other Examples 84

Environmental Applications 84Other Technologies 85

Chemiluminescence and Electrochemilumines-cence 85Fluorescence 85Surface-Enhanced Raman Spectroscopy 85Separation 86Preconcentration 86

Conclusions and Future Directions 87Author Information 87

Corresponding Author 87ORCID 87Notes 87Biographies 87

Acknowledgments 88References 88

Since the first microfluidic paper-based analytical device(μPAD) was introduced by Whitesides and co-workers in

2007,1 the field has continued to develop at an exponential ratewith notable impacts to academic and industrial communities.Generally, μPADs fabricated from cellulose paper as a substratecan process 0.1 to 100 μL of liquids using fluidic channels withmillimeter dimensions. As an abundant, inexpensive, lightweight,and biodegradable material, cellulosic paper is a natural platformfor microfluidics. While μPADs may lack the high mechanicalrobustness of traditional microfluidic materials2 (i.e., glass,3

silicon,4 polymers5), they have other advantages. Cellulose’shydrophilicity and porosity generate flow via capillary actionwithout the need for external pumps or power supplies, andpatterning techniques exist to create well-defined channels usingany one of several fabrication methods. Application of μPADs isparticularly favored in point-of-need settings that require rapidanalysis with low cost and simple operation. While many of theseapplications are relevant to diagnostics in developing countrieswhere resources and experts are limited, numerous applicationsexist in the developed world as well. These advantages havedriven growth of the field toward applications of the portableanalytical devices for clinical diagnostics, environmentalmonitoring, and food safety assurance.The current Review serves as a comprehensive update from

our last review6 for emerging techniques and applications in theμPAD arena during the time range of November 2014 toOctober 2016. Fundamental paper microfluidic studies dealingwith topics of imbibition have been well discussed previously6−8

and are not elaborated upon here. We are also aware of the use ofpaper substrates in manufacturing energetic and electronicdevices (aka papertronics),9−11 but they are not discussed here ininterest of maintaining focus on analytical measurements. Asmuch as possible, we seek to highlight the most impactful workthat is searchable by Google Scholar, Web of Science, SciFinder,and PubMed. In the last 2 years, more than 1000 papers havebeen published and over 1500 citations have been made in thisfield (Figure 1). As a result, we are not able to cover all papers andapologize to those authors whose works we have not covered ininterest of space.

■ FABRICATION

Core to the novelty of any μPAD is device fabrication. Thenovelty of μPADs comes from the low-cost relative to

Special Issue: Fundamental and Applied Reviews in AnalyticalChemistry 2017

Published: December 7, 2016

Review

pubs.acs.org/ac

© 2016 American Chemical Society 71 DOI: 10.1021/acs.analchem.6b04581Anal. Chem. 2017, 89, 71−91

pubs.acs.org/ac

http://dx.doi.org/10.1021/acs.analchem.6b04581

conventional analytical methods. Fabrication methods oftentarget lowering the cost per device, increasing productionvolume, or adding assay functionality. Figure 2 shows cost-volume relationships for a number of different device materials,including several μPAD fabrication methods.12 A variety ofμPAD fabrication methods, however, have been reported in thelast 2 years with emphasis on improving upon the currenttechnology. These strategies include increasing the solventcompatibility, deposition of new reagents and barrier materials,improved designs for flow control, and new architectures andfunctionalities for new applications. Novel porous materials aswell as devices with enhanced properties including resistance toaggressive solutions have been reported.Hydrophobic/Solvent Barrier. Wax is the dominant

workhorse material for solution barriers in μPADs due to theease of printing designs using commercially available printers andthe hydrophobicity of the barriers that act to channel fluid flow.Once wax is printed on a paper substrate and melted, the waxcoats the cellulose fibers creating a barrier. While wax has manyadvantages, surfactants, and organic solvents can wet the surface,resulting in solution penetration over and/or through the wax

barrier and a loss of sample to the device. Alternative barriermaterials can provide a wider range of chemical compatibility.Jahanshahi et al. used pullulan, a polysaccharide polymer, as abarrier to organic solvents in tandem with wax barriers tofabricate omniphobic devices.13 To create that device, thecombination of a hydrophobic wax and an organic solventresistant pullulan were patterned in series. Laser direct-writing(LDW) techniques have also been shown to form hydrophobicbarriers by polymerizing photopolymer-impregnated filter paper,and this approach has been applied to fabricate time delaybarriers. He et al. created delay barriers in filter paper by adjustingthe write speed of a laser to control the barrier height, whichresulted in the delayed flows of different solutions throughout thedevice.14 Figure 3A shows a schematic for barrier fabrication andthe delay of liquid flow with barriers made with different laserwriting speeds. Silanes can also be used to create solvent barriers.Jiang et al. showed that hydrophobic paper generated bytreatment with octadecyltrichlorsilane (OTS) could be madehydrophilic with a hand-held corona generator.15 By using apoly(methyl metacrylate) (PMMA) mask, hydrophilic channelswere regenerated where the ozone came into contact with thesurface; they even used the corona generator as initiators forsingle-use turn-on valves. Other silicone barriers such astetrakis(dimethylsiloxy)silane and 1,3-dimethyltetramethoxysi-lane were used by Rajendra et al. to generate barriers showingresistance to aqueous, organic, and surfactant solutions.16 Withsimplicity in mind, Sameenoi et al. showed that a one-stepscreen-printing method for patterning hydrophobic polystyrenecould be used to create barriers.17 In another example of usingcommon hydrophobic materials, Sharpie ink was shown to workas an alternative to wax. Xu et al. showed that the Sharpie inkcould be harvested and loaded into cartridges and patterned withan ink jet printer.18 One disadvantage, noted by Xu et al., was thatthe ethanol solution used in the ink evaporated leading to poorshelf life.18 Gallibu et al. addressed the lifetime concern bycoupling Sharpie markers to an XY plotter to draw hydrophobicchannels.19 Other low-cost strategies for fabricating μPADbarriers utilized Parafilm in either embossing or photo-lithography processes.20 Both of these methods to fabricatehydrophobic channels, i.e., embossing and the photolithography,

Figure 1. Number of published items and citations with the topic ofpaper analytical devices in 2015 and 2016, respectively. Numbers arebased on results from the Web of Science (Thomson-Reuters) for thetopic “paper analytical device” on November 7, 2016.

Figure 2. Cost and volume comparison for common lab-on-a-chip fabrication technologies. Reprinted from Microf luidics and Nanof luidics, PrintedElectronics Integrated with Paper-basedMicrofluidics: NewMethodologies for Next-generation Health Care, 19, 2015, 251−261, Jenkins, G.;Wang, Y.;Xie, Y. L.; Wu, Q.; Huang, W.; Wang, L. H.; Yang, X. (ref 12), with permission of Springer.

Analytical Chemistry Review

DOI: 10.1021/acs.analchem.6b04581Anal. Chem. 2017, 89, 71−91

72


can be implemented with basic office supplies and laboratorybenchtop tools.Deposition. Automated inkjet printing methods for

deposition of reagents and hydrophobic barriers have becomevery popular. These printing methods are designed to providegood control of deposition volume (picoliter to nanoliter),accuracy of defined geometries (±10 μm), and potential for roll-to-roll processing.21,24 The use of three-dimensional (3D)printers has also grown recently.24−26 Asano et al. printed bothhydrophilic channels and hydrophobic barriers in conjunctionwith a patterned photomask.24 They succeeded by firstimpregnating chromatography paper with octadecyltrichlorosi-lane in n-hexane, leaving a hydrophobic surface once dried. Theyfollowed the deposition step by irradiating the surface withultraviolet (UV) light using a photomask to obtain selectiveregeneration of hydrophilic behavior. An advantage to 3D-printing the photomask is that 100 μm width channels can befabricated while maintaining sample flux. In other cases, 3Dprinters were modified to be used as wax printers by Pearce etal.25

A challenge with inkjet printing is the risk of contaminationwhen printing multiple inks, which necessitates meticulouscleaning of the cartridges and print heads. To address thecontamination issue, Li et al. fabricated a user-friendly custompiezoelectric mechanism that gave controllable deposits withpicoliter to nanoliter volumes.21 Figure 3B shows the piezo-electric actuator releasing fluid to the substrate. This cost-friendly

method utilized a unique cartridge-nozzle system that reducedthe risk of contamination by keeping the reagents from cominginto contact with the piezoelectric actuator and the cartridge.Advantages to using high-performance reagent printers such asthe Dimatix DMP-3000 were demonstrated by Jenkins et al.using flow-ready polydimetylsiloxane (PDMS) channels withwidths of 60 μm.12 The narrow channel width showed promisefor smaller sample volume applications where analytes might bedifficult to obtain or in low abundance. Jenkins et al. noted thattheir approach could be advantageous for high-volume roll-to-roll processing methods previously described.27 A simple PDMSprinted channel alternative was shown by Dornelas et al. usingstamps to fabricate hydrophobic barriers that also showed goodchemical compatibility with water, organic solvents (e.g.,acetonitrile, methanol, ethanol, dimethylformamide, and di-methyl sulfoxide), and surfactants (e.g., sodium dodecyl sulfate,cetyltrimethylamnonium bromide, Triton X-100).22 Figure 3Cshows an image of the stamp and a comparison of the chemicalcompatibility between PDMS and wax printed barriers withvarious organic solvents.Hand drawn deposition methods have advantages of ease of

use and transport in the field. Crayons were previously employedto draw hydrophobic barriers; however, innovations have beenrealized with custom reagent pencils and pens.28 Mitchell et al.showed that sensitive, biologically relevant reagents could becombined with graphite and pressed into pellets for use inmechanical pencils.23 As a comparison, a solution deposition

Figure 3. (A) (i) Variable depth inside the paper substrate, created using pulsed laser exposure and (ii) delay of liquid flow after the introduction of blueink in fluidic channels with barriers created using different speeds. Reproduced from He, P. J. W.; Katis, I. N.; Eason, R. W.; Sones, C. L. Lab Chip 2015,15, 4054−4061 (ref 14), with permission of The Royal Society of Chemistry. (B) Parts i and ii represent the printing states before and after mechanicalloads from the piezoelectric actuator. Reprinted from Li, B. Q.; Fan, J. Z.; Li, J. N.; Chu, J. R.; Pan, T. R. Biomicrof luidics 2015, 9, 11 (ref 21), with thepermission of AIP Publishing. (C) Stamp with geometry of hydrophobic PDMS barrier along with images of side-by-side comparison showingcompatibility of PDMS barriers and wax barriers with some organic solvents. Reprinted fromAnalytica Chimica Acta, Vol. 858, Dornelas, K. L.; Dossi, N.;Piccin, E. A simple method for patterning poly(dimethylsiloxane) barriers in paper using contact-printing with low-cost rubber stamps, pp 82−90 (ref22). Copyright 2015, with permission from Elsevier. (D) Reagent pencil with image when water is added to the sample zone of devices, it dissolves thereagents from the pencil traces and transports them into the test zone (top); the final appearance of the reagents in the test zone. Reproduced fromMitchell, H. T.; Noxon, I. C.; Chaplan, C. A.; Carlton, S. J.; Liu, C. H.; Ganaja, K. A.; Martinez, N. W.; Immoos, C. E.; Costanzo, P. J.; Martinez, A. W.Lab Chip 2015, 15, 2213−2220 (ref 23), with permission of The Royal Society of Chemistry.



73


method for horseradish peroxidase (HRP) exhibited signals for 3days while the pencil loaded HRP method showed signal for upto 10 days. Figure 3D shows the reagent pencil along with thedrawn reagents at the moment when water is added to the deviceand also after the dissolved sample reaches the detection zone.Oyola-Reynoso et al. employed a similar strategy but with inkpens.29 Their pens were loaded with a trichloro perfluorosilane/n-hexane ink and used to create hand drawn hydrophobicbarriers. Nuchtavorn et al. used Sharpie ink and bovine serumalbumin (BSA) bioink in a similar fashion but with the penscoupled to a Craft cutter to automate the drawing process.30

Carrasquilla et al. showed the feasibility of printing bioinkswithout the need for hydrophobic barriers.31 These megadalton-sized long-chain DNA aptamer bioinks were shown to have moreeffective print characteristics when applied with a thermal printhead instead of a piezoelectric head.Flow and Injection Control. While μPADs traditionally

utilize capillary action to create flow for sample transport, controlof that flow is important for temporally dependent trans-formations such as timed reagent additions; this principle is alsorelevant for applications such as serial injections and dilutions.Kinahan et al. described the use of slowly dissolving films fortemporally controlled reagent release on disk devices.32 Alkylketene dimer (AKD) is known to establish hydrophobic barriers.Salentijn et al. showed this as well as subsequent regeneration ofhydrophilic behavior in the paper after oxygen plasma treat-ment.33 The level of hydrophobicity could be controlled throughAKD loading in the paper. Additionally, the AKD depositioncould be manipulated to form gates that were impermeable toaqueous solutions but successfully wetted after the addition of

alcohol to the mixture. Songok et al. showed that the use ofconverging and diverging channel geometries provided controlover flow rate and enabled longer duration flows.34 Toley et al.programmed μPADs to automatically control turn-on valvesthrough the use of swellable sponge-like materials.35 Figure 4Ashows the turn-on action of the valve due to swelling of channel Aand the connection made with channel B. Thermally activatedrelease of sample from sponge-like materials was demonstratedby Niedl et al.36 Using hydrogels such as N-isopropylacrylamide97% (NIPAM), they were able to control the release of sample toa paper device. These hydrogels have the ability to swell to 85 wt%; however, the material does not release the liquid when incontact with the paper at room temperature. At any temperatureabove 30 °C, though, the hydrogel dispensed the entire sample.In contrast, acrylamide (AcAM) hydrogels can release knownfractions of the sample volume, in a manner that varies linearlywith temperature over the range from 30 to 60 °C. Figure 4Bshows the release of fluid with NIPAM and NIPAM-AcAm atdifferent temperatures. Akyazi et al. lowered the flow rate ofaqueous samples by employing two different sets of hydrogels.The first was Ionogels 1 (IL-1) and 2 (IL-2), and the second wasa mixure of 1-ethyl-3-methylimidazolium ethyl sulfate IL andtrihexyltetradecyl-phosphonium dicyanamide IL.37 Once theseionic gels were drop-cast and UV-photopolymerized, fluid flowrates were measured in straight-channel devices. Both devicesloaded with ionogels slowed the flow rate compared to theunmodified device, with the phosphonium gel (IL-2) exhibitingthe slowest flow rate.Traditionally, devices are sealed through lamination or packing

tape to lower contamination risk as well as to lower effects of

Figure 4. (A) Schematic of an on-switch. It consists of two flow channels, A and B, placed at two different levels. Upon actuation, one end of channel Alifts up to connect to one end of channel B. Reproduced from Toley, B. J.; Wang, J. A.; Gupta, M.; Buser, J. R.; Lafleur, L. K.; Lutz, B. R.; Fu, E.; Yager, P.Lab Chip 2015, 15, 1432−1444 (ref 35), with permission of The Royal Society of Chemistry. (B) Images displaying the switch-like liquid release from aNIPAM hydrogel at different temperatures (left) and images illustrating the gradual linear release of liquid from a composite NIPAM-AcAm hydrogel atdifferent temperatures (right). Reproduced fromNiedl, R. R.; Beta, C. Lab Chip 2015, 15, 2452−2459 (ref 36), with permission of The Royal Society ofChemistry. (C) Images of the manual tribocharging of PET sheets by rubbing on PTFE (i) and PMMA (ii) surfaces. Schematic representation of thetribosealing process by electrostatic “sandwich type” contact (iii) and the distance traveled by the fluid front inside the paper channel over time.Reproduced from da Silva, E.; Santhiago, M.; de Souza, F. R.; Coltro, W. K. T.; Kubota, L. T. Lab Chip 2015, 15, 1651−1655 (ref 38), with permission ofThe Royal Society of Chemistry.



74


humidity and evaporation.6 However, da Silva et al. showed newmethods for sealing that could affect flow rates; the triboelectriceffect was used as a means to electrostatically seal a μPADbetween two poly(ethylene terephthalate) (PET) films.38 Thepositively and negatively charged PET films were generated byrubbing against polytetrafluoroethylene (PTFE) and PMMA,respectively. When comparing the flow rates between theelectrostatically sealed devices and the unsealed devices, thecharged devices delayed the flow rate as a result of the attractionof the polar water molecules to the charged surfaces. Figure 4Cshows the tribocharging and tribosealing processes, and thesample’s distance traveled up the device.Three-Dimensional Devices. There are a number of three-

dimensional (3D) devices recently published that use the origamiconstruction method to measure analytes.39−42 Despite the largeflux of these origami devices published with new applications, thescope of this Review highlights new fabrication schemes in thissection; those origami devices will be discussed in followingsections. Wang et al. showed a unique greeting card “pop-up”ePAD that can control fluidic and electrical conductivity bysimply folding it. When the device was “open”, the sample wascontained in a reaction zone. Once the reaction was completeand the device was “closed”, the reaction zone came in contactwith the detection zone. This 3D device was fabricated from asingle sheet of paper and also allowed for control of the fluid’spath and timing.43 Another single layer 3D device was fabricatedby Jeong et al. using double-sided wax printing to create themicrofluidic network.44 The vertical and horizontal channelswere fabricated by varying the pattern on either side and thenmelting the wax and pressing the device with a laminator. Thisdouble-sided printing method has shown promise for highproduction volume since the 3D devices were fabricated rapidly(∼35.5 s). Another 3D device was developed using a two-layerconfiguration by Camplisson et al.45 The two-layer system’swicking rate was found to be significantly increased compared tothat of one-layer devices. The flow rates were measured at both100% and 35% relative humidity (RH) and compared withLucas-Washburn (LW) behavior for both the unsealed one- andtwo-layer devices. At 100% RH, the flow rates of both the singleand two-layer devices were accurately predicted by the LWequation. At 35% RH, the flow rates in both devices onlyfollowed the LW behavior for the first 5 min. The deviation frompredicted behaviors was accredited to solvent evaporation, whichis not accounted for in the LW equation.Incorporating Nonsensing Electrodes. Functional struc-

tures such as electrodes have been incorporated into μPADs andhave found broad application in addition to chemical analysis.Flow rates (distance/time) on paper were historically measuredcolorimetrically with dyes. Recently, Bathany et al. measured theflow rate by incorporating an electrode as an electrochemicalsensor for conductive aqueous samples.46 Here a polycarbonatedevice was assembled with inlet and outlet reservoirs and aworking electrode lying flat between them. When an electrolytesolution was added to the reservoir and a paper strip was addedinto the outlet, the rate of the fluid wicking up the paper wascalculated using the Cottrell equation based on the current beingmeasured. They also investigated the effects on flow rates bytreating the device with oxygen plasma and a DEX polymer.Mixing of reagents has also been explored by incorporating anelectrode-contained zigzag channel geometry.47 The behavior ofelectrokinetically driven flow in a zigzag channel was studied byDey et al. via the mixing of red and blue food coloring.47 Themixing effect was significantly improved by controlling the

electrokinetically mediated flow. Typically, electrodes with lowintrinsic resistance are added to devices to maximize the appliedcurrent, butMatsuda et al. incorporated more resistive electrodesto take advantage of Joule heating, which aided in controllingtemperature-sensitive reactions, concentrating the sample on agiven zone, as well as creating a stop valve.48

■ COLORIMETRIC DETECTIONMost μPADs employ colorimetric detection because it provideseasy readout of the generated chemical signals, opening the doorfor instrument-free measurements. Results of colorimetric assayscan be qualitative, semiquantitative or quantitative depending onthe assay goals. Qualitative assays on μPADs are commonlyachieved without any external instrumentation as they give yes/no results that can be readily determined by eye.49 Semi-quantitative analysis can involve the use of a color chart forestimating relative analyte amount based on a pre-establishedcalibration curve.50 More rigorous and carefully controlledmeasurements are required for quantitative analyses on μPADsin order to obtain accurate and precise analyte levels. Thesemeasurements are typically performed by employing instrumentsto acquire images and image-processing software to quantify thecolor signal intensities and/or hues.51,52

These device configurations work on the principle that visiblecolors result from the reaction between analyte and chromogenicreagents.49,53 The chromogenic reagents are chosen to selectivelyreact with the analyte with minimal interference from othersubstances. Colored products can also be generated uponenzymatic conversion of substrates.54−56 Metal nanoparticles(NPs) have high extinction coefficients, making them attractiveas colorimetric labels.57 The optical response of metal NPs istunable by controlling nanoparticle composition, morphology,size, and solution environment.58 Because of their unique surfaceplasmon properties, metal NPs such as AuNPs and AgNPs, arewidely utilized for colorimetric detection of ionic species such asmetal cations, anionic organic compounds (e.g., ascorbic acid),and proteins.59−61 Depending on size and charge of the ionicspecies, the surface charge of the NPs can be either stabilized ordestabilized, maintaining particle segregation or causingaggregation, respectively. NP aggregation causes a shift in colordue to changes in the plasmon resonance. Aggregation of metalNPs can also be induced by antigen−antibody binding.62

Detectors and Readout. To enable analyte quantification,detectors such as charge-coupled devices (CCD) and comple-mentary metal-oxide sensors (CMOS) embedded in cameras orflatbed scanners have been widely employed.51,52,63 Measure-ments using cameras and scanners (reflection mode) are basedon light reflectance whereby a light source is used to illuminate asample and the subsequent light reflected from the surface of thesample is then captured by a photodetector. Transmittance-based measurements such as those in traditional absorbancespectroscopy have also been utilized to provide more sensitiveanalyte quantification on μPADs.64,65 Some recent develop-ments on these reflectance- and transmittance-based methodsalong with several instrument-free quantification schemes arediscussed below.

Reflectance-Based Measurement. Smart device cameras andscanners have been utilized as digital aids to quantify colorintensities in colorimetric measurements due to their low costand simple operation. Conventional flatbed scanners are heavilyused in laboratory settings to capture paper device images due tohigh-precision light sources, making it unnecessary to employadditional features to obtain constant illumination inten-



75


sities.51,66−68 Less expensive portable scanners provide thesimilar uniform illumination benefit while giving moreconvenience for in-field analysis.66,69 Fiber optic devices canalso be used as portable and on-site tools for analyses on μPADs.Chen et al. described the use of such device for colorimetric Hg2+

assays.70 The device probe emitted a red laser, measured theintensity of reflected light, and presented the reflectance as adigital readout.Being equipped with both a light source and a detector,

smartphones can perform rapid measurements on paper devicesfor point-of-care (POC) applications. The last 2 years have seen arapid increase in smartphone use and network coverage aroundthe world. The trend is advantageous for the growth oftelemedicine and integrated global environmental screening.The use of smartphone-based paper sensors have beendemonstrated for monitoring air and water pollutions,52,71

disease outbreaks,60,63,72 hazardous metal presences,73 andalcohol levels in biological samples.74 A smartphone equipped

with an image processing application based on a pixel countingalgorithm was used by Sicard et al. to monitor pesticide levels inwater using colorimetric paper-based sensors.71 The algorithmanalyzed the images by taking ratios between different pixelvalues in red-green-blue (RGB) mode to minimize usersubjectivity. To reduce bias from varying ambient light, imagesof both test and control strips were taken simultaneously withouta flash while maintaining focus. In addition, image acquisitionvariability has also been reduced by fixing the phone camera on astand.73 A miniature attachment consisting of a light reflector, adiffuser, and a plano-convex (PCX) lens with 15 mm focal lengthheld in place using an acrylic holder was used by Jung et al. tocontrol sample positioning and provide constant illumination forimaging (Figure 5A).74 With the black acrylic material, theassembly attachment provided better contrast between thebackground and the test strip, thus better sensitivity than thewhite acrylic model. While RGB is the most common colorchannel used by investigators to analyze images, other channels

Figure 5. Various methods for colorimetric signal quantification on μPADs: (A) Camera phone equipped with attached sample holder and lightillumination adjuster for reflection-based measurement. Reprinted with permission from ref 74, The Optical Society. (B) Paper sensor (top) and hand-held transmittance reader (bottom) consisting of a photodetector and LEDs securely placed on a printed circuit board. Alignment pinhole was used toobtain repeatable placement of the paper device within the reader. Reprinted from Sens. Biosensing Res. Vol. 5, Swanson, C.; Lee, S.; Aranyosi, A. J.; Tien,B.; Chan, C.; Wong, M.; Lowe, J.; Jain, S.; Ghaffari, R. Rapid Light Transmittance Measurements in Paper-Based Microfluidic Devices, pp 55−61 (ref65). Copyright 2015 The Authors, published by Elsevier. (C) Semiquantitative measurement based on change in metal nanostructures properties in thepresence of analytes. Color chart (top) was used to estimate analyte concentration ranging from 0 (leftmost) to 100 μM(rightmost). Absorption spectracorresponding the colors were shown at the bottom. Reprinted from Talanta Vol. 148, Lin, T.; Li, Z.; Song, Z.; Chen, H.; Guo, L.; Fu, F.; Wu, Z. Visualand Colorimetric Detection of p-Aminophenol in Environmental Water and Human Urine Samples Based on Anisotropic Growth of Ag Nanoshells onAu Nanorods, pp 62−68 (ref 50). Copyright 2015, with permission from Elsevier. (D) Distance-based measurement for multiple metals determination.Reproduced fromCate, D.M; Noblitt, S. D.; Volckens, J.; Henry, C. S. Lab Chip 2015, 15, 2808−2818 (ref 53), with permission of The Royal Society ofChemistry.



76


such as hue-saturation-value (HSV) and cyan-magenta-yellow-key (CMYK) have also been applied to find an appropriate colorspace for optimal results.74,75 These methods have improved theefficiency and reproducibility of smartphone-based colorimetricmeasurements performed on μPADs.Transmittance-Based Measurement. Although reflectance is

the most common colorimetric detection mode used withμPADs, it only allows detection of the surface-bound analytesand often requires high analyte concentrations (i.e., highmicromolar) to give detectable signals.76 Saturation of signalsdue to the finite amount of analyte that can be accumulated onthe paper substrate surface also limits the linear range inreflectance-based measurements to typically 1 or 2 order(s) ofmagnitude.71 Transmittance-based measurements, on the otherhand, assess the cumulative density of light absorbing materialsthroughout the thickness of paper and may offer enhancedsensitivity. In addition, sensitivity can also be tuned bycontrolling the device thickness through using thicker paper ormultiple paper layers.64

Swanson and co-workers developed a portable transmittance-based reader consisting of photodetectors and light emittingdiodes (LEDs) placed on a rigid printed circuit board (Figure5B) to determine the concentration of alanine amino transferase(ALT), a biomarker for liver function assessment, in serumsamples.65 The paper device was sandwiched between LEDs anda photodiode to interrogate light transmission through the paper.Since light transmission depends on the water content of thepaper, measurements were done on wetted μPADs. The deviceswere sealed using self-adhesive plastic sheets to preventevaporation and maintain light transmission properties overthe assay period. A similar portable colorimeter was employed forascorbic acid quantification by Ferreira et al.61 By using singlelayer paper devices made of Whatman 1 chromatography paper,approximately 8 times lower detection limit was obtained frommeasurements using a transmittance reader compared to thoseperformed on a flatbed scanner. With the readout electronicsexpected to cost less than $20 in large-scale production,65

transmittance-based paper sensors offer an inexpensive andsensitive tool for colorimetric measurements. The currentdownside, however, is the relatively low throughput of thesystem where a single assay requires at least 10 min.Instrument-Free Measurement. One advantage of analytical

devices for POC applications is the ability to operate with simpleto no additional equipment. μPADs can also be designed toprovide interpretable analysis results without any externalequipment or reader. For example, external calibration with astandard colorimetric chart can be used in semiquantitativeanalyses. The chart allows users to estimate the amount ofanalyte present in the sample by visually comparing its color toassociated references. Lin et al. demonstrated the use of Agnanoshells that were anisotropically grown on the surface of Aunanorods in the presence of analyte to create a blue shift in theplasmon resonance band.50 The resulting colors were comparedto a standard chart where different colors represented varyinganalyte concentrations (Figure 5C). Karita and Kanetadeveloped μPADs for titrations of Ca2+ and Mg2+ that consistedof 10 separate reaction and detection zones all connected to asingle sample zone.77 The reaction zone on each arm containedincreasing concentrations of chelating agents. The detectionzones contained a constant concentration of metal indicators.When sample solution reached the reaction zone, the metalsreacted with the chelating ligands while the uncomplexed metalions traveled farther to the detection zone, causing color to

change from blue to pink. Hence, the metal concentration wascorrelated to the chelating agent concentration at the first zonewith no color change. A similar strategy was applied by Zhangand co-workers to develop equipment-free assays for adenosinedetection.78

Distance-based detection that relies on reading a visual signallength linearly corresponding to analyte concentration is alsopopular.78−82 Colored bands can be generated as analyte flowsalong a channel due to capillary force and reacts withpredeposited reagents to form colored products that remain onthe paper substrate,53,78 or based on the flow stopping resultingfrom channel constriction as molecular binding to surfacereceptors occurs.81,82 Cate et al. created distance-based multi-plexed paper sensors for simultaneous detection of Ni, Fe, andCu in welding fumes based on complexation reactions betweenthe metals and their corresponding chromogenic reagents.53 Thepaper devices were built with rulers printed next to the analysischannels to quickly quantify the colored region lengths andcorrelate these to the concentrations of the metals (Figure 5D).Aptamer cross-linked to a polyacrylamide hydrogel wasemployed to quantitatively measure cocaine levels in urineusing stop-flow distance-based μPADs.81 A series of enzymes andsubstrates in combination were trapped within the hydrogel ormodified on the hydrophilic channel on the paper devices forcolorimetric detection. As the target was captured by theaptamer, the hydrogel collapsed, triggering a cascade ofenzymatic reactions that produced a colored product. Since thetarget-induced cascade reaction is elicited by the bindingbetween the aptamer to the analyte, this stop-flow method issuitable for many analytes providing the aptamers arecommercially available or can be developed.

Biomedical Applications. Enzymatic Methods. Enzymaticreactions are widely used in detection schemes due to the abilityof enzymes to amplify the signal through their catalyticproperties. Enzymes can be immobilized on the paper surface,generating colored products when the substrate is present.Immobilization methods can significantly affect the shelf life ofenzyme-based μPADs, especially in applications where thesensors could be exposed to high temperatures during storageand transport. Nery and Kubota compared a variety ofimmobilization methods including adsorption, gel entrapment,layer-by-layer entrapment, microencapsulation and covalentlinkage using glucose oxidase (GOx) as a model enzyme.83

Storage stability at room temperature increased when theenzymes were entrapped in polymer matrixes such as starch ascompared to direct adsorption, possibly due to stabilization ofthe protein structure within the matrixes. Layer-by-layerentrapment methods where enzymes were mixed with bilayerionic polymer covered on the surface of the paper yielded higherenzymatic activity than the other strategies that wereinvestigated. Immobilization via covalent linkage has beenneeded to prevent the enzymes being washed out duringanalysis.84,85 However, these immobilization methods may giveconsiderably lower enzyme activities, likely due to enzymedeactivation by chemical residues from conjugation reactions anddisruption to the catalytic pockets of the enzymes.83

The applications of enzyme-based colorimetric μPADs havebeen demonstrated for detection of clinically important diseasebiomarkers such as glucose, creatinine, and phenylala-nine.54−56,86 Zhang et al. described the use of enzyme-functionalized starch to do simultaneous testing of glucose andlactate in blood samples.86 GOx and lactate oxidase (LOx) incombination with HRP were used for glucose and lactate assays,



77


while 4-aminoantipyrine andN-ethyl-N-(2-hydroxy-3-sulfoprop-yl)-m-toluidine were employed as the colorimetric reagents,respectively. Assay sensitivity was sufficient to determineclinically relevant concentrations of both analytes in blood (i.e.,0.5−10mM). The enzymes alsomaintained their activities over 6months in storage at room temperature, demonstrating thepotential for refrigeration-free long-term storage of the papersensors. The use of chitosan as an immobilization support forenzymes was reported for glucose and uric acid assays.54

Chitosan was impregnated into the paper substrate, followedby enzyme and substrate addition. Compared to native paperdevices, chitosan-modified paper gave a higher colorimetricresponse under similar reagent and analyte loading due to itsability to provide a suitable microenvironment for direct electrontransfer during the enzymatic reactions. The modified paper alsoyielded more homogeneous color intensity over the detectionzone area, confirming that the deposited reagents remained fixedin place, even during flow. That result is similar to one shown byMentele et al. for colorimetric metal detection.87 Detection limitsfor the glucose and uric acid assays were 23 and 37 μM,respectively. The use of foldable enzyme-based 3D μPADs forglucose determination was described by Choi et al. to quantify0−50 mM glucose concentrations in blood samples.88

Enzymes, which may also serve as disease biomarkers, cansimilarly be detected using paper devices with the appropriatechromogenic substrates to the target enzymes. Nosrati et al.developed μPADs for assessing male fertility based on a 3-(4,5-dimethyl-2-thiazolyl)-2,5-diphenyl-tetrazolium bromide (MTT)assay.89 Diaphorase flavoprotein enzyme present inmetabolicallyactive human sperm converts yellow-colored MTT into purpleformazan. μPADs that measured the activity of acetylcholineesterase (AChE) to evaluate organophosphate poisoning havealso been reported.90 AChE catalyzed the hydrolysis ofacetylcholine into choline and acetic acid. Choline was furtheroxidized by choline oxidase to produce H2O2 that reacted with3,5,3′,5′-tetramethylbenzidine (TMB) in the presence of HRP togive a blue colored product. Organophosphates inhibited theactivity of AChE, preventing formation of the blue color on theμPADs. Paper sensors employing pullulan were reported fordetection of lactate dehydrogenase (LDH), a biomarker ofcancer and organ failure, in serum samples.91,92 Nicotinamideadenine dinucleotide (NAD+), lactic acid, and iodonitrotetrazo-lium chloride (INT) were used as a coenzyme, substrate, andchromogenic reagent, respectively. NAD+ was reduced intoNADH as LDH catalyzed the conversion of lactate into pyruvate.The NADH product then converted INT into bright redformazan dye. Addition of pullulan helped stabilize the reagents,making it possible to store the paper strips for at least 10 weeks at4 °C with no loss in performance. Since the NAD+/NADHcouples are important coenzymes for many dehydrogenaseenzymes,93 NADH paper-sensors could be applicable fordiagnosis of diseases whose markers include dehydrogenases.Liang et al. reported paper-based devices for visual determinationof NADH based on dissolution of AuNPs.94 The assays werebased on inhibition of AuNP dissolution in Au3+-cetyltrimethy-lammonium bromide (CTAB) solution in the presence ofNADH. Thus, it was possible to span the range from colorless tointense red, depending on the amount of undissolved AuNPsassociated with increasing concentrations of NADH.Immunoassays. Capitalizing on antibody−antigen binding

properties, μPADs can be designed to perform sensitive andspecific detection of biomarkers in complex biological matrixes.Similar to enzymes, antibody immobilization on μPADs may

involve direct adsorption to the native cellulose substrate or theuse of substrate modification to stabilize the attachedcomplex.95−97 In most cases, the capture antibodies must bephysically attached to the sensor surface and should not detachduring washing or any flow steps. The analyte binds to theimmobilized capture antibody and is later sandwiched with adetection antibody. To generate colors on paper devices,detection antibodies are conjugated to either enzymes or NPssuch as those in conventional enzyme-linked immunosorbentassays (ELISAs) and lateral flow assays (LFAs). Blocking agentssuch as bovine serum albumin (BSA) and fat-free milk are oftenadded to minimize nonspecific adsorption of biomolecules to thepaper substrate.96,98 Aptamers have also been employed asreceptors in immunoassays due to their relatively easierproduction and better stability compared to monoclonalantibodies.99 A number of aptamer-based immunoassays onpaper sensors were reported for monitoring cancer cells, diseasediagnosis, and detection of biologically relevant analy-tes.78,81,82,100−102

Paper-based sandwich immunoassays were reported by Lei etal. for detection of influenza A H1N1 and H3N2 viruses.96

Capture antibodies were adsorbed on filter paper and sample wasloaded to allow for virus particles to bind to the antibodies,followed by the addition of primary antibodies and HRP- orAuNP-linked secondary antibodies to complete the sandwich. Awashing procedure was carried out between each solutionaddition to remove unbound analytes/reagents. AuNPs withgold enhancement (GE), a process to enlarge and fuse theAuNPs to increase color intensity,103 gave better colorimetricresponse than enzyme-based assays using HRP. Sensitivity of thepaper-based immunoassay was comparable to conventionalELISA and detection limits for H1N1 and H3N2 viruses were2.7 × 103 and 2.7 × 104 particle forming units (PFU)/assay,respectively. Device cost and reagent consumption are lowerthan the conventional ELISA. However, the assays are still time-consuming to perform (i.e., >1 h) due to the multiple wash stepsrequired and incubation of reagents for binding reactions. Similarpaper-based sandwich ELISAs using HRP were developed forbacteriophage detection.98 A nonenzymatic AuNP-based paperimmunoassay was also reported for the detection of choleratoxin.62 An antibody to the toxin was conjugated to the AuNPs,and the particles became aggregated to produce a color change inthe presence of the toxin.Badu-Tawiah and co-workers described a paper-based

immunoassay utilizing polymerization-based amplification(PBA) for rapid detection of Plasmodium falciparum histidine-rich protein 2 (PfHRP2) in malaria.95 Capture antibodies werecovalently coupled to the aldehyde groups on oxidizedchromatography paper and detection antibodies were conjugatedwith eosin, a photoinitiator. Upon the binding of target protein tothe antibodies, eosin was localized at the capture site (Figure 6).Amplification solution containing polymerizable monomers andphenolphthalein indicator at pH 7.9 were added, followed bylight illumination to initiate polymerization. As the hydrogelformed, phenolphthalein became trapped in the polymernetwork. Addition of NaOH solution to the paper increasedthe pH and changed the color of pH indicator from colorless topink. The assay detection limit was 5.8 nM and the visualizationsteps could be completed within minutes. Lathwal and Sikesassessed performances of different amplification methodsincluding enzymatic amplification using HRP and alkalinephosphatase (ALP), PBA, and silver enhancement (SE) onAuNPs for PfHRP2 immunoassays on paper.104 Using a



78


smartphone camera and ImageJ for color processing, limits ofdetection (LOD) of the enzyme-based immunoassays werefound to be an order of magnitude lower than assays using eitherPBA or SE/AuNPs.Other. A number of colorimetric μPADs that did not involve

the use of large biomolecules such as enzymes, and antibodieshave also been reported for biomedical applications. Kumar et al.fabricated paper devices for uric acid detection based on metalNP-catalyzed colorimetric reactions.105 In the presence ofcharged AuNPs, H2O2 reacted with TMB to produce a blue/green color that became colorless upon reaction with uric acid.The LOD of the assays was 8.1 ppm or 48 μM, which is alsocomparable to the enzyme-based assays on μPADs.54,106 Papersensors based on the Griess reaction have been reported foridentification of S-nitrosothiols (RSNOs), which are carriers ofnitric oxide (NO) involved in several physiological andpathological functions.52 Decomposition of RSNOs by Hg2+ orlight yielded products that reacted with sulfanilamide to producea diazonium salt.N-(1-Naphthyl)ethyldiamine then reacted withthe diazonium salt to generate a pink-colored azo-dye. LODsvaried slightly depending on decomposition methods and were 4,6, and 11 μM for Hg2+, UV, and Vis LED-induced

decompositions, respectively. A paper-based skin patch forquantitative measurement of anions in sweat was employed fordiagnostic screening of cystic fibrosis byMu et al.72 The assay wasbased on anion exchange via diethylaminoethyl moietiesmodified on the cellulose fiber. Excreted anions were absorbedonto the paper and triggered a release of OH− that alkalized thepaper and caused color changes.

Environmental Applications. Many colorimetric μPADshave been developed to monitor environmental contaminationdue to their versatility for in-field testing. Environmentalpollutants may be of inorganic (metals and nonmetals), organic,or biological (bacteria etc.) origin. Human exposure guidelinesfor a large number of toxic compounds have been established byregulatory agencies such as U.S. Environmental ProtectionAgency (EPA) and the Occupational Safety and HealthAdministration (OSHA). These compounds often havesignificant negative effects on the organisms in aquatic andterrestrial ecosystems.Multiple colorimetric paper-based sensors for analyses of

metals, including Ni2+, Cu2+, Fe3+, Cr6+, Ag+, Hg2+, and Pb2+ havebeen recently reported that offer attractive alternatives to themore sophisticated, expensive, laboratory-extensive techniquessuch as atomic absorption spectroscopy (AAS) and inductivelycoupled plasma mass spectrometry (ICPMS).49,53,70,107−110

Many of these colorimetric sensors involve metal complexationwith ligands. Li et al. designed qualitative periodic table-styledpaper sensors for reporting the presence of Ni2+, Cu2+, and Cr6+

in water samples by patterning chromatography paper to formhydrophilic symbols of the metal surrounded by wax barriers.49

Colorimetric reagents for each metal were added to thehydrophilic region and allowed to dry. To analyze water samples,paper sensors were dipped into the sample to allow for dissolvedmetal ions to react with the predeposited reagents. The visualLODs were 0.8, 0.3, and 0.5 ppm for Cu, Cr, and Ni, respectively.The assays were useful for Cu and Ni; however, their LOD for Crexceeded themaximum contaminant level for total Cr in drinkingwater (0.1 ppm).111 More quantitative paper-based sensors formetals analyses in welding fumes were fabricated by Cate et al.using distance-based measurements with LOD at approximately10 ppm.53 Sensitive paper sensors have also been developed forCu2+ detection based on the catalytic etching of Ag nanoplates(AgNPls) by thiosulfate (S2O3

2−).107 The Cu2+ accelerated theetching rate of AgNPls to form Ag(S2O3)2

3− and Cu(S2O3)35−,

reducing the size of the NPls and causing a blue-shift in theabsorption peak. The sensors were able to report as low as 1.0ppb metal by visual observation, making the μPADs alsoapplicable for more sensitive applications such as measuring Cuin biological fluids.112 Metal nanomaterials were also employedby Chen and co-workers to construct a colorimetric paper-basedsensor for Hg2+ determination.70 PtNPs catalyzed the oxidationof 3,3,5,5-tetramethylbenzidine (TMB) to produce a blueproduct. The oxidation was suppressed when Hg2+ was presentdue to interactions between the Hg2+ ions and the PtNPs,reducing the color signal intensity. The intensity was monitoredusing a fiber optic device andHg2+ concentrations as low as 2 ppbcould be detected. In both nanomaterial-based sensors, targetmetals could be selectively measured in the presence of othermetals due to the unique catalytic properties of the metals on thecolorimetric reactions. Other NPs-based sensors were reportedby Nath et al. for Pb2+ and Cu2+ detection using a AuNP-thiocticacid-dansylhydrazine conjugate and Guo et al. for Cr6+ detectionbased on the leaching of AuNPs.109,113

Figure 6. Polymerization-based amplification in paper immunoassay.Acronyms: poly(ethylene glycol)diacrylate (PEGDA), triethanolamine(TEA), 1-vinyl-2-pyrrolidinone (VP). Reproduced from Badu-Tawiah,A.K.; Lathwal, S.; Kaastrup, K.; Al-Sayah, M.; Christodouleas, D. C.;Smith, B.S.;Whitesides, G.M.; Sikes, H. D. Lab Chip 2015, 15, 655−659(ref 95), with permission of The Royal Society of Chemistry.



79


In addition to metals detection, colorimetric μPADs have beenapplied to detect nonmetal inorganic pollutants. Paper sensorsbased on the Griess reaction were reported to detect nitrite inenvironmental samples in low micromolar concentrations.75

Chen et al. designed colorimetric μPADs for analysis of dissolvedNH3 and CO2 using multiple pH-sensitive dyes.67 The sensorshave the potential for distinguishing the two target analytes insolution. A gas diffusion-based μPAD for quantifying NH3 inwastewater was described by Jayawardane et al.66 The deviceconsisted of two layers of paper sheets sandwiching gaspermeable Teflon tape sealed with laminating film. A NaOHsolution was introduced to one layer to convert the ammoniumions dissolved in water into NH3 that diffused through the Teflonmembrane to the second layer of paper where NH3 reacted withbromothymol blue or 3-nitrophenol to yield a color. The assaysdo not require any sample preparation and the paper sensors canserve as an inexpensive alternative to measure NH3 concen-trations higher than 5 ppm in wastewater.Pesticides are among a number of organic pollutants that can

contaminate water sources, including groundwater used forlivestock and human consumption and thus are the subject ofcareful monitoring. Separation techniques such as gas and liquidchromatography are frequently used for pesticide identificationand quantification. However, these techniques require expensiveinstrumentation, are relatively time-consuming, and not suitablefor on-site monitoring. Sicard et al. developed low-cost papersensors for on-site quantification of organophosphate pesticidesbased on enzymatic conversion of colorless indoxyl acetate into ablue-colored product by AChE.71 A range of signals, from intenseblue to colorless, was obtained with increasing concentration ofpesticide. They also created a smartphone application thatenabled image acquisition, signal quantification, and datareporting to a centralized web server where a map of the areawith tested water quality can be accessed. The LODwas 3 ppb forthe tested organophosphates (malathion and paraoxon). Similarpesticide sensors using AChEwere reported with detection limitsranging from 3 to 600 ppb for several organophosphates.114 Thecombination of enzymes and CeO2 NPs to generate coloredproducts for pesticide analysis was also reported.115 Alkasir et al.fabricated portable sensors for measuring bisphenol A (BPA), anestrogen disruptor associated with the plastics industry, in indoordust that employed interchangeable paper sensors interfacedwith a sampling cassette for collecting air samples.51,116

Tyrosinase enzyme was entrapped using layer-by-layer deposi-tion of chitosan and alginate polyelectrolytes on the papersubstrate. BPA was enzymatically converted into a quinone thatlater cross-linked to the chitosan layer and changed the papercolor to blue-green. The LOD of the assay was found to be 0.28ppm. They also carried out an interference study usingsubstances that could affect the enzyme activity including otherphenolic compounds, ascorbic acid, and organophosphates, andno significant effect was observed with interference concen-trations lower than 50 ppm.There are a large number of recently developed μPADs for

detection of bacterial contamination in environmental samples,particularly in food and water, with the goals of providingdiagnostic tools that are faster to answer and lower in cost thantraditional methods such as culture analysis and polymerasechain reaction (PCR).117−121 A NP-based paper sensor wasdescribed by Shafiee et al. for detection of Escherichia coli, ageneric indicator of fecal contamination.117 AuNPs weremodified with lipopolysaccharide binding protein (LBP) tocapture bacteria in the solution and subsequently transferred to

paper to allow for the distribution of aggregated NPs.Concentrations as low as 8 colony forming units (CFU)/mLwere detectable, demonstrating that the assays can be applieddirectly to the samples without prior enriching steps. Paper-based ELISA using HRP/TMB was also reported to detect E. coli≥ 105 CFU/mL.118 While these assays offer high specificity indetecting certain bacterial strains, they do not provideconfirmation of pathogen viability. Jin et al. reported Salmonellalive cell detection on μPADs based on quantification ofadenosine triphosphate (ATP).120 However, the technique isnot selective since many other cell types such as mammal, plant,and yeast cells also release ATP.

Other Applications. The use of μPADs has also beendemonstrated in fields such as pharmaceutics, food safety, andforensics. Identification of multiple pharmaceutical ingredientsand dietary supplements has been carried out using papersubstrates.61,122−125 Weaver and Lieberman developed a 12-lanetest card to simultaneously quantify multiple antimalarial activepharmaceutical ingredients and excipients.123 Colorimetricreagents for each tested ingredient were dried on the lanes andcrushed dosages were applied to the card, followed by dippingthe card into water to allow for analyte transport through theporous channel and subsequent reaction with the depositedchemicals. The investigators also explored whole-cell yeastpaper-based sensors for similar applications.124 Quantification ofvarious food contaminant levels, notably heavy metals, toxicorganic and inorganic compounds, and food-borne pathogenshave also been reported.75,108,126 Peters et al. utilized paper baseddevices to visually detect inorganic and organic explosivematerials.76 Detection limits ranging from 0.39 to 19.8 μg ofexplosive compound were reported. Paper sensors were alsoemployed to quantify mustard gas concentrations, a chemicalwarfare agent.127

■ ELECTROCHEMICAL DETECTIONOne challenge with colorimetric μPADs is achieving lowdetection limits with high sensitivity and selectivity. Electro-chemical paper-based analytical devices (ePADs), benefitingfrom the intrinsic advantages of providing more stable andquantifiable signals, have become one of the most extensivelyinvestigated detection motifs. Since Henry and co-workersreported the first application of ePAD in 2009,128 a wide range ofelectrodes, paper platforms, and detectors have been integratedtogether for the analysis of biological, environmental, and othersamples. Moreover, electrochemical sensing techniques includ-ing amperometry, voltammetry, potentiometry, electrochemicalimpedance, and capacitance have been explored to generate thedesired analytical measurements. Several recent reviewsdescribed the development of ePADs129,130 as well as paper-based electronic sensing devices and energy devices.9−11 In thissection, we provide a review of the latest development of ePADs,including innovative electrode fabrication methods, optimizationstrategies, electrochemical techniques, and their major use assensors.

Electrodes. Electrochemical detection is normally performedwith a three-electrode system containing working, reference, andcounter electrodes. The working electrode is especiallyimportant because it primarily dictates analytical performanceand functionality of ePAD systems. Carbon and noble metals arethe most common working electrode materials used in ePADs,while other materials such as indium tin oxide (ITO)131 have alsobeen reported. Recently, there is an increasing interest inemploying nanomaterials for electrode modification to achieve



80


more specific recognition and/or improved sensitivity of analytedetection. New patterning and assembly methods, as well asnovel reference electrode materials, have also been investigatedin the pursuit of making ePADs with better analyticalperformance and at lower cost.Carbon Electrodes. Carbon-based electrodes are inexpensive,

easy to fabricate, and have a relatively wide potential window inmost aqueous electrolyte solutions. In the creation of carbonelectrodes on ePADs, screen/stencil-printing is recognized as asimple and effective approach as evidenced by its use in the massproduction of the test strips utilized in commercial hand-heldglucometers.132 The performance of screen/stencil-printingrelies on the material of screen/stencil as well as the compositionof the carbon ink. Several methods for customizing the patteringscreen/stencil include photolithography, crafting, and lasercutting. The most common carbon inks are composed ofgraphite and/or other conductive forms of carbon, but additionalchemical modifiers can be incorporated for additional function-ality. In a recent study, Rungsawang et al. mixed cellulose acetatewith graphite powder to create customized carbon ink, which wasused for fabrication of screen-printed electrodes.133 Thesensitivity of electrochemical detection was significantlyimproved due to the network and porous structure of cellulose.NP modified screen-printed carbon electrodes have also shownimproved performance,134 specifically toward the detection ofglucose, which is a benchmark analyte in biosensor development.Both enzyme-modified electrode133 and enzymeless elecrode135

configurations have been investigated. In the latter case, enzymesare normally still involved but immobilized on paper instead ofthe electrode. More complex chemical modification strategiesinvolved biomaterials are detailed in the following sections.Carbon nanotubes (CNTs) are an attractive carbon source for

these electrodes. CNTs have larger surface areas relative totraditional carbon electrodes and can be functionalized beforebeing used to create electrodes. Functionalized CNTs can bedeposited on precursor electrodes;136 recently, CNTs have beendirectly patterned as electrodes on ePADs. Lei et al. depositedbiotin-modified CNTs on paper as an electrode for detection ofavidin.137 Similarly, Koo et al. used chemical vapor deposition(CVD) to grow highly aligned CNTs and integrated them intoan origami paper device.138 They found that 200-layer CNTworking electrodes had higher conductivity and higher electro-chemical sensitivity compared to 100-layered electrodes.Pencil drawing is a straightforward method to produce

graphite trace electrodes with desired geometries on paper.This can be done using commercially available pencils, whichaffords great convenience for making ePAD devices in resource-limited areas. Fully drawn pencil electrodes were integrated intothe detection zone of an origami-type μPAD that containedferrocenecarboxylic acid as an electron transfer mediator for thecatalytic oxidation of glucose.40 Conventional hand-drawnelectrodes demonstrate limited reproducibility due to humanfactors as well as differences between manufacturers. Recently, Liet al. reported a pressure-assisted ball pen device for directwriting electrodes on paper as shown in Figure 7.139 The devicewas composed of two parts: a pre-emptied ballpoint pen as theink reservoir and a 5 mL syringe with two PMMA boards as thepressure-assisted accessory (Figure 7A). Carbon ink and silverink were loaded into the ballpoint pens and hand-applied to thepaper to create a working/counter electrode or quasi-referenceelectrode (Figure 7B). The strategy was extended to fabricatemultielectrode arrays (Figure 7C) and to write electrodes onnonplanar surfaces (Figure 7D). However, the surface roughness

was still slightly higher than that of conventional printingmethods due to the use of manual application.Much of the focus has been on fabricating working electrodes,

but a stable all-solid-state carbon reference electrode has beendeveloped by Hu et al.140 They chose colloid-imprintedmesoporous (CIM) carbon as the solid contact and filled themesopores with a polymeric sensing matrix. As a proof-of-concept, the resulting electrode was incorporated in an ePAD forpotentiometric Cl− sensing. The CIM carbon-based referenceelectrode exhibited excellent resistance to common interferingagents such as light and O2, with outstanding potential stability incontinuous potentiometric measurements. The potential driftwithout redox couple was as low as 1.7 μV/h over 110 h.

Metallic Electrodes. Noble metals (e.g., Au, Ag, Pt) areexcellent electrode materials due to their excellent electrontransfer capabilities and ready surface modification. Thin filmdeposition is among the most common methods for creatingmetallic electrodes on ePADs and can be achieved byevaporation, sputtering, or spraying metals on paper through amask. While electrodes produced from metallic ink can beadvantageous in yielding high-quality electrodes, they aretypically limited to expensive fabrication equipment. Therefore,simpler fabrication methods and other metallic electrode formatshave been explored.Freestanding metal microwire electrodes were first introduced

and characterized in hollow paper-based devices by Crooks andco-workers.141 Metal microwires generally have low resistanceand are easier to clean and/or modify prior to assembly. Adkins

Figure 7. (A) Schematic of direct writing electrodes on paper using apressure-assisted ball pen. The inset represents the relationship betweenthe pressure inside the syringe measured by a pressure meter with asyringe scale. (B-D) Photographs of the fabricated ePAD: (B) on an A4paper with a three-electrode system; (C) on an A4 paper with an eight-electrode array of eight carbon working electrodes, a communal Agreference electrode, and a communal carbon counter electrode; (D) on apaper cup with two three-electrode systems similar to (A) except withdifferent electrode shapes. Reproduced from Li, Z.; Li, F.; Hu, J.; Wee,W. H.; Han, Y. L.; Pingguan-Murphy, B.; Lu, T. J.; Xu, F. Analyst 2015,140, 5526−5535 (ref 139), with permission of The Royal Society ofChemistry.



81


et al. then characterized the electrochemical behavior ofmicrowires in contact with paper in both static solution142 andquasi-steady flow conditions.143 In static conditions, themicrowires’ electrochemical behavior was between that ofcylindrical and hemicylindrical microelectrodes due to itsproximity to the paper surface.142 A semithin layer behaviorwas also observed which is likely caused by the confined solutionvolume in the ePAD. They expanded the investigation into aflow-through ePAD system as shown in Figure 8.143 A 270° fan-shaped passive pump was connected to an inlet channel thatcompensated for the decay in flow rate within the channel(Figure 8A). Because the height of the device was constant, theflow behavior within the channel followed Darcy’s Law and wasvalidated colorimetrically. A quasi-steady flow was subsequentlygenerated in the inlet channel where three metal microwires wereembedded between two paper layers for electrochemical analysis(Figure 8B). The double-layer structure increased the current 2.5times over that of a static saturated double-layer channel and 1.7times over a flowing single-layer channel (Figure 8C). Anenzyme kinetic study of β-galactosidase with p-aminophenyl-galactopyranoside was carried out in the device. A broaderimpact can also be expected for automating flow or washing stepsin an ePAD system.Another inexpensive, simple alternative electrode is a metal

pin.144 Glavan et al. showed that stainless steel pins could be usedas electrodes in devices fabricated with embossed omniphobicfluoroalkylated paper or cotton thread. The pins easily passedthrough the paper and provided 3D electronic connections to apotentiostat or other instrument. The configuration isparticularly beneficial in generating multiwell electrochemicalplates/substrates that can interface with existing instrumentationfor multiplexed electrochemical readout.Incorporation of NPs is an effective strategy for amplifying the

electrochemical signals, thereby enhancing detection sensitivity.Metal NPs can provide a high surface-to-volume ratio, highconductivity, and good stability. Recently, Au nanoflowers were

fabricated using a seed-mediated growth approach and served asan effective matrix for antibody attachment.145 In another report,Li et al. used a nanoporous Pt particle modified paper workingelectrode as the matrix and L-cysteine (Cys) capped flower-likeAuNPs as the signal amplification label.146 The Au-Cys structureoffered an active interface to carry metal ions such as Pb2+ andAg+ to form flower-like Au-Cys/metal ion tracers for directantibody labeling.

Biological Applications. Glucose Sensors. Blood glucosemonitoring has served as a valuable tool for screening, diagnosis,and long-term management of diabetes. Even though gluc-ometers are readily purchased by consumers and patients, itremains a challenge to achieve rapid, accurate, reliable, andsensitive glucose measurements, particularly for the developingworld.132 GOx, a major enzyme catalyzing the oxidation of β-D-glucose to gluconic acid and hydrogen peroxide (H2O2), isnormally immobilized on the surface of electrochemicaltransducers. The current produced from oxidation of H2O2quantifies glucose level. Advances in GOx-based ePADs havebeen made by investigating new redox mediators (e.g., 4-aminophenylboronic acid)133 and electrode surface modificationstrategies (e.g., ZnO nanowire deposition147).The modifier and modification process involved in enzyme

immobilization onto electrodes may cause fouling and impedeelectron transfer. Therefore, immobilizing enzymes into thecellulose network of paper adjacent to electrodes has become anattractive alternative. Sekar et al. improved Prussian Blue (PB)-modified screen-printed carbon electrodes by physicallyadsorbing GOx within the porous structure of paper discs thatwere placed on top of the electrodes.135 It required only 0.5 μL ofsample to perform a satisfactory analysis. Similarly, Yang et al.used a GOx-immobilized cellulose paper coupled with a PtNP-modified screen-printing carbon electrode for sensitive glucosemeasurement.134

3D and origami-type devices have also been exploited toseparate the reaction and detection zones. A paper-based

Figure 8. (A) Quasi-stationary flow ePAD fabrication showing the device layers, the top view of the device design and the device image with electrodeleads attached, (B) image of channel crosssection, cut down the center, showing a 25 μm Au electrode sandwiched between two paper layers with anenlarged inset of the cut, circular end of the Aumicrowire electrode in the upper right corner, (C) the linear-sweep voltammetry current profiles for one-and two-layer devices either with quasi-steady flow or without flow in a saturated channel using 5mMK4Fe(CN)6/K3Fe(CN)6 in 0.5MKCl, detected at25 μm diameter Au microwire electrodes (Pt counter and reference). Reproduced from Adkins, J. A.; Noviana, E.; Henry, C. S. Anal. Chem. 2016, 88,10639−10647 (ref 143). Copyright 2016 American Chemical Society.



82


electrochemical device was inspired by pop-up greeting cards andused to detect beta-hydroxybutyrate (BHB) in combination witha commercial glucometer (Figure 9).43 BHB is a biomarker fordiabetic ketoacidosis and its measurement requires a specificBHBmeter and test strips, which are more costly than those usedto measure glucose. By simply folding and unfolding the device,this reconfigurable 3D structure enabled functionalities includingtemporal control, fluidic handling and path programming,control over complex sequences of steps, and alterations inelectrical connectivity. On the basis of a similar concept, Li et al.developed an origami-type ePAD composed of two zonesseparated by a central crease. One detection zone with pencil-drawn electrodes also contained ferrocenecarboxylic acid as anelectron transfer mediator for the catalytic oxidation of glucose.

Another immobilization zone contained GOx and was broughtinto contact with the electrodes by folding the paper along thecentral crease.40

Dias et al. developed the first paper-based enzymatic reactors(PERs) coupled with a 3D printed batch injection analysis (BIA)cell for amperometric detection of glucose in artificial serumsamples.148 The paper surface of PERs was oxidized with sodiumperiodate solution, then perforated with a punch to createmicrodisks, and was activated through EDC/NHS bioconjuga-tion to immobilize GOx. The reaction product H2O2 wascollected with an electronic micropipet in a microtube andanalyzed in the 3D BIA cell coupled with screen-printedelectrodes. The proposed method exhibited excellent repeat-ability and reproducibility (% RSD < 2%) as well as a high

Figure 9. (A) Schematic of the pop-up-ePAD in the “open” format. Enzymes are stored in the reaction zone. The detection zone contains printedelectrodes designed specifically for the commercial electrochemical reader used in this study. IE, indicator (filling) electrode; CE/RE, common counterand reference electrode; WE, working electrode. (B) Photographs showing the valve capabilities of the pop-up-ePAD when it is used with a glucometer.Reproduced fromWang, C. C.; Hennek, J. W.; Ainla, A.; Kumar, A. A.; Lan, W. J.; Im, J.; Smith, B. S.; Zhao, M. X.; Whitesides, G. M. Anal. Chem. 2016,88, 6326−6333 (ref 43). Copyright 2016 American Chemical Society.

Figure 10. (A) Schematic illustration of the approach of electrochemical immunoassay in hydrophobic paper microwells. (B) Effect of concentration ofPf HRP2 on the square wave voltammetry peak current in a sandwich ELISA. The average value of the blank was subtracted from each measurement.Each datum is the mean of 7 replicates, and the error bars are standard deviations of the measurements. The limit of detection is estimated to be 4 ngmL−1. The R2 value of the curve fit to the data using the Hill equation is 0.988. The curve is approximated as linear between the concentrations of 10 and1000 ng mL−1 of PfHRP2 per well (R2 = 0.989). Reproduced from Glavan, A. C.; Christodouleas, D. C.; Mosadegh, B.; Yu, H. D.; Smith, B. S.; Lessing,J.; Fernandez-Abedul, M. T.; Whitesides, G. M. Anal. Chem. 2014, 86, 11999−12007 (ref 149). Copyright 2014 American Chemical Society.



83


sampling rate (30 injections h−1) under low injection volume (10μL) with 0.11 mmol L−1 LOD and 1−10 mmol L−1 linear range.Immunosensors. Immunological detection uses immuno-

assay techniques to detect humoral antibodies or antigenicsubstances through an antigen−antibody reaction. Immunosen-sor coupling with ePAD provides promising opportunities inrapid, sensitive, and low-cost clinical diagnosis. Enzyme-linkedimmunosorbent assay (ELISA) based ePADs are an attractivealternative to colorimetric ELISA in clinical diagnostics since theelectrochemical measurement can be output in simple numericreadouts with low-cost electronics. Glavan et al. developed adevice for electrochemical ELISA on paper and the fabricationprocess is shown in Figure 10A.149 These devices were appliedfor the detection of rabbit IgG as a model antigen and malarial PfHRP2 in spiked human serum. C10

H paper, cellulose paper thatwas rendered hydrophobic by vapor-phase silanization with decyltrichlorosilane, was identified as an optimal platform forelectrochemical ELISA. Its low surface free energy allowed theprinting of highly reproducible electrodes, and its highhydrophobicity and surface area facilitated the immobilizationof antibodies or antigens. In another report, Wang et al.integrated ELISA ePAD with a hand-held multichannelpotentiostat for detection of carcinoembryonic antigen(CEA).150 Amino functional graphene/thionine/AuNP nano-composites were synthesized and coated on screen-printedelectrode in order to immobilize antibodies and increase thedetection sensitivity. This platform was capable of performingELISA simultaneously on eight samples within 20 min.While enzyme-based immunosensors are subject to interfer-

ence from various environmental factors such as temperature,oxygen, pH, and coexisting chemicals, enzyme-free immuno-sensors have been developed to lessen the complicationsassociated with enzyme purification, immobilization, andstabilization. A diversity of metal nanoparticles (NPs) has beenutilized as signal amplification labels in nonenzymatic andquantitative ePAD immunosensors. AgNPs can act as goodelectron conductors that accelerate the electron transfer andcatalyze redox reactions. Cunningham et al. reported the use ofAgNPs linked to a magnetic microbead support in animmunosandwich for the detection of ricin at picomolar levels.42

The method included two simple but effective amplificationsteps: magnetic preconcentration of AgNP labels at the workingelectrode surface and 250 000-fold electrochemical amplificationof the target concentration by combining 20 nm AgNP labelsupon anodic stripping voltammetry. They have also extended thework to include galvanic exchange detections.151 Recently Sun etal. reported using ZnO-modified reduced graphene oxide (rGO)paper electrodes combined with highly catalytic BSA-stabilizedAgNPs as an signal amplification label for H2O2 reduction.

152

The enzyme-free property was due to the high conductivity ofZnO nanorods, the large surface area of rGO, and the highcatalytic activity of BSA-stabilized AgNPs. AuNPs have also beenextensively used in modifying the electrodes in enzyme-freeimmunosensors. Li et al. reported a sandwich immunosensingsystem for detecting α-fetoprotein (AFP).153 It containednoncovalent, ultrathin Au nanowire functionalized graphenesheets as the sensing platform, in situ grown CuS NPs on thesurface of graphene sheets, and AFP antibody covalently boundCuS/graphene composites. A wide calibration range of 0.001−10ng mL−1 and a low detection limit of 0.5 pg mL−1 were achievedfor serum samples from healthy people and cancer patients. Maet al. explored an Au nanorod-modified paper working electrodeas the sensor platform and metal ion-coated Au/BSA nano-

spheres as tracing tags for detecting cancer antigens.154 Theyachieved LODs down to 0.08 pg mL−1 for CEA and 0.06 mUmL−1 for cancer antigen 125 using differential pulse voltammetrywithout metal preconcentration.

Other Examples. In vitro detection of neurotransmitters (e.g.,dopamine) on ePADs has been achieved by Feng et al. in a linearrange of 10 nM to 1 μM through differential pulsevoltammetry.136 The platform consisted of a carbon tape withspecific working regions regulated by a plastic adhesive tape witha punched hole as the working electrode and the pure cellulosepaper as the electrochemical reservoir. They also demonstratedthat multiwall carbon nanotube (MWCNT)-Nafion modifica-tion and oxygen plasma treatment could significantly improvethe electrochemical responses. The modified electrode enabledanalysis of dopamine extracted from the striatum of rat with onlya 10 μL sample volume. Another online detection of dopaminereleased from PC12 cells was achieved by growing cells in a 3Dmatrix deposited on patches of filter paper and quantifiedthrough amperometry.155 Yu et al.131 developed a label-freeimpedance spectrometry based ePAD for the enumeration ofcells and the evaluation of drugs’ toxic effects on living cells. Aunanorod-modified ITO electrodes provided a biocompatiblesurface for immobilizing living cells and maintaining theirbioactivities. The impedance results exhibited good correlationto the logarithmic value of cell numbers ranging from 7.5× 102 to3.9 × 106 cells mL−1 with a LOD of 500 cells mL−1.

Environmental Applications. ePADs are an attractiveanalytical tool for rapid monitoring of environmental pollutantsin the field. Although colorimetric detection allows for simplevisual interpretation, electrochemistry is a more sensitive andquantitative technique for point-of-need testing. Recently,Ruecha et al. developed a graphene-polaniline nanocompositeelectrode for simultaneous detection of Zn(II), Cd(II), andPb(II) on both paper and transparency film.156 They concludedthat the screen-printed electrodes on plastic yielded overallbetter electrochemical performance than those on printingpaper. Simultaneous detection of Pb(II), Cd(II), and Cu(II) wasreported by Chaiyo et al. through dual electrochemical andcolorimetric detection on paper.157 There, the μPADwas dividedinto two parts. One part was electrochemical detection of Pb(II)and Cd(II) using a Bi-modified, B-doped diamond electrode.The LOD was 0.1 ng mL−1 for both metals. Another part wascolorimetric detection of Cu(II) owing to thiosulfate catalyticetching of Ag nanoplates (AgNPls). Another dual electro-chemical and colorimetric detection on ePADs was reported fordiscrimination of procaine in seized cocaine samples.158

A halide ion sensor was reported by Cuartero et al. for absolutedetermination of halides in a range of diverse water samples andfood supplements.159 A thin layer made of cellulose paper, acation exchange Donnan exclusion membrane, and a silver-foilworking electrode was selected as an optimum material forcomposing the electrochemical cell. Cyclic voltammetry waschosen to impose the sequential oxidation/plating of the halideat the silver electrode, followed by regeneration with an invertedpotential. A mixture of Cl−, Br−, and I− ions was assessed in awide range of concentrations from 10−4.8 to 0.1 M for Br− and I−

and from 10−4.5 to 0.6 M for Cl− with a LOD of 10−5 M. Anotherpaper-based ion sensor was developed by Hu et al. by embeddinga potentiometric cell into paper and completed with an ion-selective electrode, a reference electrode, and a paper-basedmicrofluidic sample zone.160 As proof-of-concept, they fabricatedCl− and K+ sensors that required only 20 μL of sample, withhighly reproducible and accurate responses in aqueous and



84


biological media. Moreover, an ionophore-based ion-selectiveoptode platform on paper was described for the first time with aNa+ optode as the example.161 The selective elements(chromoionophore, ionophore, and ion-exchanger) weredirectly adsorbed on cellulose paper via intermolecular forces.The resulted hydrophobic microenvironment essentially enabledaqueous solution spreading on the optical sensing area, whichwas in analogy to classical ion selective electrode/optode.

■ OTHER TECHNOLOGIESChemiluminescence and Electrochemiluminescence.

Chemiluminescence (CL) has emerged as a sensitive andefficient analytical technique for μPADs because it can providehigh sensitivity and wide dynamic range with minimal skilledlabor. In CL, a chemical reaction generates excited molecules,which is advantageous relative to fluorescence because it does notrequire an external light source that can result in increasedbackground noise. Digital cameras and smartphones are low-cost, reliable readout devices that can be coupled with CLμPADs, and they have shown successful application inbiochemical studies, including the detection of salivarycortisols,162 H2O2,

163 and the genotyping of single nucleotidepolymorphisms.164 An inexpensive CCD imaging device wasintegrated into a μPAD system as a detector for CL DNAsensing.165 The device detected long DNA amplicons on thebasis of hybridization reactions with a covalently immobilizedDNA probe and biotin-labeled signal DNA strands. The CLreactions were catalyzed by a HRP-streptavidin conjugate. TheLOD was 3−5 orders of magnitude lower than those obtainedfrom other CCD-based microfluidic CL DNA assays. In anotherexample, Wang et al. illustrated a CL μPAD setup using aterminal digital multimeter detector and a paper supercapacitorfor signal amplification for measuring CEA concentration atpicomole levels.166 CL analysis was also expanded for environ-mental samples such as Cr(III)167 and dichlorvos (DDV).168

As an alternative to CL, electrochemiluminescence (ECL)uses electrochemical reactions to generate luminescence. ECLconfers some advantages over CL, including lower backgroundoptical signals, easier to control reagent generation throughpotential control and improved selectivity through potentialcontrol at the electrodes. The most common ECL reagents aretris(bipyridine)-ruthenium(II) (Ru(bpy)3

2+) and luminol.169,170

Recently Gao et al. synthesized Au nanocages that can adsorbRu(bpy)3

2+ as an ECL signal amplification label in a μPAD-basedimmunoassay.171 Their proposed device showed highly sensitivedetection toward CEA in a linear range of 0.001−50 ngmL−1 anda LOD of 0.0007 ng mL−1. Furthermore, there have been a fewbreakthroughs integrating ECL with other techniques forbroader applications. Wu et al. reported the first ECL origamipaper device combined with rolling circle amplification (RCA)on oligonucleotide functionalized carbon dots for sensingmacromolecules such as proteins.172 Real-time visual determi-nation of the flux of H2S from multiple cancer cells was alsoachieved by the use of paper-based ECL cyto-device with hollow-channels.173 Open bipolar electrode (BPE)-ECL was alsoreported by Liu et al. toward visual detection.174 Theydemonstrated two designs: a hydrophilic paper channel tohosting a single band-shaped BPE and a platform containing 11BPEs stridden across the two legs of the U-shaped paper channel.The single band-shaped configuration exhibited better overallanalytical performances, especially in image sensing.Fluorescence. Paper-based fluorescence assays often

encounter a few difficulties because commercially available

paper substrates contain additives that also self-fluoresce and leadto high background noise. Fluorescence resonance energytransfer (FRET)-based ratiometric probes involve the observa-tion of target-triggered ratiometric changes in the fluorescenceintensity at two different wavelengths, which could providecorrection for eliminating environmental effects, leading tominimization of any possible noise and increase in the dynamicresponse.175 Kim et al. fabricated a paper-based sensor stripcontaining carbon nanodots (C-dots) conjugated with arhodamine moiety for the detection of Al3+.176 The detectionwas based on that Al3+ induced ring-opening of rhodamine on C-dots through chelation and the rhodamine moieties showedmore emission under excitation. The device selectively detectedAl3+ from ametal ionmixture with increased emission intensity atlong-wavelength emission via FRET. The sensing based onFRET of C-dots was color-tunable, could be recognized with anaked eye, and provided a new platform for specific metal-ionsensing on paper devices. FRET assays were also applied to thedetection of macromolecules. Graphene oxide (GO) is afrequently used quencher. It is a two-dimensional material withhigh adsorption and quenching ability. It also has good solubilityin aqueous media due to the surface hydroxyl and carboxylicgroups. Li et al. recently reported a GO-based paper chip (glassfiber) paired with fluorescently labeled single-stranded DNA(ssDNA).177 They used a labeled ssDNA as the reporter and GOas the quencher on the glass fiber paper for data acquisition. Thischip could monitor fluorescence recovery during FRET activity.One significant advantage is that it did not require any compleximmobilization or washing to treat the surface and its imagescould be captured from a digital camera for analysis. Otherapplications combining FRET and GO electrode include thedetection of cancer cells178 and fluoride ions.179

Analysis of electroneutral Lewis bases and anions in aqueousmedia is challenging for routine instrumentation; however, theanalysis was recently realized by incorporating fluorescent zincand copper complexes in a μPAD.180 Caglayan et al. synthesizeda fluorescent ligand, which favors the coordination of d9 or d10

metal ions that can adopt square planar geometries or distortedtetrahedral geometries. They selected Coumarin 343 (C343) as afluorescent indicator. The fluorescence was quenched when theindicator was coordinated to the complex and recovered uponexposure to an analyte. The concept was carried out on papermicrozone plates and verified by the qualitative discrimination ofadrafinil (a banned performance-enhancing drug).

Surface-Enhanced Raman Spectroscopy. Surface-en-hanced Raman spectroscopy (SERS) creates a vibrationalfingerprint of individual molecules down to the single-moleculelevel when coupled with plasmonic nanomaterials as thesubstrate matrix. Generally, SERS signals are highly dependenton the nature of plasmonic NPs, the extent of NP aggregation ingenerating electromagnetic hot spots and the effective local-ization of Raman probes in the hot spots. Implementing SERSinto μPADs can be problematic because pre-embedding of theplasmonic NPs or analytes in paper may induce particleaggregation and restrict hot spot generation.A series of functionalization strategies have been exploited in

order to improve paper as a SERS platform. AgNP-based filterpaper was modified with Cl− ions because Cl− could overcomethe electrostatic repulsion between AgNPs and paper thusimprove chemisorption.181 Polyelectrolytes such as poly-(allylamine hydrochloride) (PAH) and poly(sodium 4-styrene-sulfonate) (PSS) were also added to the surface of AgNP-basedfilter paper and applied for the separation, preconcentration, and



85


detection of colorants in drinks with complex components.182

Recently Zheng et al. reported that after simply dip-coating filterpaper in a concentrated AuNP dispersion in toluene, it exhibitedstrong SERS activity that allowed real-time monitoring ofchemical reactions.183 Bare AuNPs were also employed by Douet al. for measuring β-agonists in swine hair extract withoutchemical labeling, purification, or separation steps.184 Com-parably, 4-mercaptophenylboronic acid (4-MBA) and 1-decanethiol (1-DT) modified rod-shaped Au particles wereembedded as SERS probe for blood glucose determination in anitrocellulose-membrane μPAD.185 Another promising paper-based SERS platform for μPADs was created by modifying paperwith nanofibrous poly(L-lactic acid) (PLLA) adsorbed withplasmonic nanostructures.186 The PLLA nanofibrous paperexhibited a porous structure and a clean surface with goodhydrophobicity (contact angle 133.4°), which assisted inreducing background interference under Raman laser excitation.NP transport within the cellulose matrix was characterized by

acquiring both Raman and SERS spectral maps.187 The mappingprovided informative images regarding differentiation ofcellulose fibers and characterization of analyte depositionpatterns. Furthermore, Saha et al. showed that SERS could beadapted in simple μPADs where both the plasmonic nanoma-terials and analyte were used in the mobile phase.188 Thisapproach enabled analyte-induced, controlled particle aggrega-tion, and electromagnetic hot spot generation inside themicrofluidic channel. The resultant SERS signal was highlysensitive and could detect pico- to femtomolar protein levels. Inorder to facilitate a more manageable sample manipulation forSERS-immunoassay measurement, a switch-on-chip multilay-ered μPAD was fabricated to provide operators with manualswitches on the interactions between different fluids for thedetection of clenbuterol (a banned feed additive in manycountries) from swine hairs.189

Separation. While μPADs have traditionally been used forassays, there has been a recent effort to adapt separation methodsbased on decades old paper chromatography and electrophoresismethods190−192 where μPADs can add new functionality.Strategies for manipulating fluids in μPADs include designinggeometry with varying channel length, width, and shape,193

changing chemical properties in the flow path to adjust flowrate,194,195 or using mechanical means to maintain or stop thefluid flow. Electrophoresis on paper has been of interest forseparation and subsequent detection owing to paper’s inherentadvantages such as hydrophilicity and self-pumping. Forexample, Chagas et al. presented the first electrophoresismicrochips using pencil drawn electrodes (PDE) on paperplatforms for capacitive coupled contactless conductivitydetection (C4D) of inorganic cations in human tear samples.196

Macromolecule separation was achieved on paper-based electro-phoretic devices that were composed of multiple thin (180 μm/layer) folded paper layers as the supporting medium.197 A highelectric field (∼2 kV/m) was applied between the anode andcathode over an extremely short distance. The multilayer designenabled convenient sample introduction by the use of a lip layerand recover by unfolding the origami paper and cutting outdesired layers. It achieved rapid (∼5 min) separation of proteinsin bovine serum at low voltages (∼10 V). As a way to build amicro total analysis system on μPAD, a laminated cross-channelpaper-based separation device was fabricated by cutting a filterpaper sheet followed by lamination.198 By applying potential atdifferent reservoirs for sample, sample waste, buffer, and bufferwaste, multiple function units including sample loading, sample

injection, and electrophoretic separation were integrated on asingle μPAD for the first time. This concept was subsequentlyextended to a isotachophoresis (ITP) μPAD for focusing DNAsamples with certain lengths.199

Alternative separation techniques have been investigated onμPADs to replace conventional microfluidic separation devices.Zhong et al. developed a size-based extraction method forseparating free dye from a mixed protein and dye solution.200

They demonstrated the separation of fluorescein isothiocyanatefrom fluorescein isothiocyanate-labeled bovine serum albumin.Separation and direct quantitation of volatile and nonvolatilecompounds was also achieved on μPADs by Phansi et al.201 Theirdevices consisted of 3 layers: a “donor layer” (with donorreservoir) in the bottom, a “spacer layer” (with a cylindrical airgap) in the middle, and an “acceptor layer” (with acceptorreservoir) on the top. A volatile compound could diffuse from thedonor reservoir to the upper acceptor layer where it reacted witha reagent in the acceptor reservoir leading to a change in color.

Preconcentration.As an effective preconcentrationmethod,ion concentration polarization (ICP) using ion selectivemembranes has found use in ion-depletion and ion-enrichmentprocesses. Recent work has incorporated paper as a lower cost,disposable platform for one-time use.202,203 Nafion, as a cation-selective polymer, has been used in μPADs for concentratinganionic samples. As shown in Figure 11A, a paper-based ICPdevice for direct DNA analysis was reported.204 The device usednitrocellulose paper with a sample channel, two reservoirs, and aregion coated with nanoporous Nafion. The main functionalcomponent was Nafion, which enabled nanoscale electrokinetictransport of ions. A voltage was applied to induce ICP and the netmovement of anions was ultimately dictated by electrophoreticmigration (EPH) toward the anode and electroosmotic flow(EOF) toward the cathode (Figure 11B). Thus, multiple anionicanalytes were separated by their varied electrophoretic mobility.The proposed devices were used to detect HBV DNA in humanserum and assess sperm DNA integrity in raw human semen.Figure 11C shows that intact and heavier dsDNA wereconcentrated at the depletion boundary, while fragmented andlighter ssDNA were concentrated further downstream. DNAsensors of this type enable inexpensive (materials cost <1 USDper test) and rapid (∼10 min) analysis in a simple format withcomparable performance to the conventional gold standards. Inanother Nafion-based ICP device, two-sided wax-printingprocess was found to be effective in reducing the μPAD channeldepth thus enhancing the ICP preconcentration.205 Han et al.demonstrated that Nafion-patterned μPAD channels couldensure passive and steady lateral flows with a high preconcentra-tion factor up to 340-fold for protein samples.206

μPADs with multiple channels have also been investigated fortheir preconcentration capabilities. Hong et al. fabricated apassive fluidic device with a straight channel divided intomultipledaughter channels and expanded regions.207 Steady flowsthrough the straight and daughter channels were created dueto the larger hydrodynamic resistance in the straight channel.Upon optimization, the steady flow continued for 10min withoutexternal pumping devices. Combining this bifurcation systemwith ICP, they developed a continuous-flow preconcentrator anda field-flow separation system. Hung et al. investigated thepreconcentration performance of μPADs comprising one, two,and three convergent channels.208 The single-channel μPADachieved a 20-fold improvement in the sample concentration inapproximately 2 min, while the double and triple channel μPADsachieved preconcentration factors of 60- and 140-fold,



86


respectively. Finally, a portable concentrator device wasproposed by the use of four 16-V batteries, yielding a practicalPOC diagnostic device. Aside from ICP, electrokineticstacking,209 evaporation,210 and extraction211,212 based techni-ques have also been incorporated in μPADs to enrich samplesand enhance the sensitivity of downstream detection methods.

■ CONCLUSIONS AND FUTURE DIRECTIONSThe research progress highlighted in this Review demonstratesthat μPADs provide great opportunities for simple but effectiveanalytical measurements in a variety of areas includingbiomedical, environmental, and food safety. The advent of newmaterials and technologies represent major avenues for researchand development of μPADs. These innovations are madepossible by new microfabrication techniques, surface modifica-tions, and a variety of new materials for improved properties andapplications. The continued growth of applications for μPADs isassured by the properties that attracted the initial users: the lowcost for creating a lightweight, portable, and energy efficientdevice, the simple fabrication without clean room facilities, the

minimum volume of sample required, and the possibility ofsimultaneous measurement on integrated devices.While significant progress has been made in the field, new

trends are emerging. For example, despite the growth of μPADresearch in university settings, few devices have been extensivelytested in the field. Moving a laboratory technology to the point-of-need can be complicated; however, utilization in theseconditions is the ultimate goal for developing many of theseμPAD platforms. A recent report by Whitesides illustrated theprocess of moving low-cost diagnostics from the bench to thefield in developing countries.213 The experience drawn fromthese cases provides valuable information for researchersinterested in the design, fabrication, and assessment of newμPAD technologies. In the meantime, the World HealthOrganization, the Environmental Protection Agency, and othersimilar organizations have actively updated public health andenvironmental concerns all around the world, which areinformative in guiding the development of new devices forinnovative applications.To fully deploy μPADs as reliable point-of-care analytical

tools, it is essential but often overlooked to investigate theanalytical performance under different test conditions (e.g.,temperature, humidity, ambient light, complexity of the samplematrixes), perform stability studies (especially when sensitivereagents or materials are involved in the device architecture), andcarry out interference studies. Additionally, inexpensive self-reporting devices that allow easy interpretation of results foruntrained users could significantly broaden the impact of μPADs.The use of text reporting devices and color indicators to alertwhen a toxic substance or disease biomarker has exceeded certainpermissible or normal levels would be some alternatives toimprove the user-friendliness of the sensors.

■ AUTHOR INFORMATIONCorresponding Author*E-mail: [email protected]. Phone: +1-970-491-2852.ORCIDCharles S. Henry: 0000-0002-8671-7728NotesThe authors declare no competing financial interest.BiographiesYuanyuan Yang received her Ph.D. in Analytical Chemistry from theWayne State University in 2016. She is currently a postdoctoral fellow inthe Chemistry Department at Colorado State University. Her researchfocuses on the development of electrochemical and colorimetric paper-based analytical devices for biological and environmental applications.

Eka Noviana received her B.S. in Pharmaceutical Science from GadjahMada University in 2012 and M.S. in Chemistry from the University ofArizona in 2015. Currently, she is a graduate student in the ChemistryDepartment at Colorado State University. Her research focuses on thedevelopment of paper-based sensors for quantification of biologicallyrelevant analytes.

Michael P. Nguyen received his B.S. in Chemistry and Biochemistryfrom San Diego State University in 2013. He is currently a graduatestudent in the Chemistry Department at Colorado State University. Hisresearch focus is on developing new geometries and construction ofμPADs tailored for inorganic environmental analysis.

Brian J. Geiss received his Ph.D. in Cell and Molecular Biology from theSaint Louis University School of Medicine followed by microbiology-focused postdoctoral studies at Washington University School of

Figure 11. Design and operation of the paper-based ICP device. (A)Device schematic with reservoirs (blue), sample channel (purple), andNafion-coated region (white). (B) Schematic of ICP enrichment anddepletion effects under applied voltage. Net movement of analytes isgoverned by electrophoretic migration (EPH) and electroosmotic flow(EOF), in response to the local electric field strength, E. (C) Intensityprofile of the separated DNA. The percent DNA Fragmentation Index(% DFI) is calculated using areas under the normal fits. Reproducedfrom Gong, M. M.; Nosrati, R.; Gabriel, M. C. S.; Zini, A.; Sinton, D. J.Am. Chem. Soc. 2015, 137, 13913−13919 (ref 204). Copyright 2015American Chemical Society.



87

mailto:[email protected]

http://orcid.org/0000-0002-8671-7728


Medicine and Colorado State University (CSU). He is currently anassociate professor in the Department of Microbiology, Immunology,and Pathology at CSU and is also is an associate faculty member in theCSU School of Biomedical Engineering and professeur associe in theDepartement de Biochimie at the Universite de Sherbrooke.

David S. Dandy received his Ph.D. in chemical engineering from theCalifornia Institute of Technology. Following his graduate studies, Dr.Dandy was a senior member of technical staff at Sandia NationalLaboratories, working first on combustion research and then onchemical vapor deposition. He came to Colorado State University in1992 as an assistant professor, eventually being promoted to fullprofessor in 2002. Since 2007, he has served as department head ofchemical and biological engineering. He also has a joint appointment asa professor of biomedical engineering.

Charles S. Henry received his Ph.D. in Analytical Chemistry from theUniversity of Arkansas followed by postdoctoral studies as an NIHpostdoctoral fellow at the University of Kansas. He started his academiccareer at Mississippi State University before moving to Colorado StateUniversity in 2002. He is currently a full professor and department chairof Chemistry at Colorado State University and also serves as a facultymember in Chemical & Biological Engineering and BiomedicalEngineering.

■ ACKNOWLEDGMENTSThis work was supported by funding from Colorado StateUniversity to the Coalition for the Development andImplementation of Sensing Systems.

■ REFERENCES(1) Martinez, A. W.; Phillips, S. T.; Butte, M. J.; Whitesides, G. M.Angew. Chem. 2007, 119, 1340−1342.(2) Nge, P. N.; Rogers, C. I.; Woolley, A. T. Chem. Rev. 2013, 113,2550−2583.(3) Grover, W. H.; Skelley, A. M.; Liu, C. N.; Lagally, E. T.; Mathies, R.A. Sens. Actuators, B 2003, 89, 315−323.(4) Harris, N. R.; Hill, M.; Beeby, S.; Shen, Y.; White, N. M.; Hawkes, J.J.; Coakley, W. T. Sens. Actuators, B 2003, 95, 425−434.(5) Becker, H.; Locascio, L. E. Talanta 2002, 56, 267−287.(6) Cate, D. M.; Adkins, J. A.; Mettakoonpitak, J.; Henry, C. S. Anal.Chem. 2015, 87, 19−41.(7) Lopez-Marzo, A. M.; Merkoci, A. Lab Chip 2016, 16, 3150−3176.(8) Xia, Y. Y.; Si, J.; Li, Z. Y. Biosens. Bioelectron. 2016, 77, 774−789.(9) Mahadeva, S. K.; Walus, K.; Stoeber, B. ACS Appl. Mater. Interfaces2015, 7, 8345−8362.(10) Safdar, M.; Janis, J.; Sanchez, S. Lab Chip 2016, 16, 2754−2758.(11) Sharifi, F.; Ghobadian, S.; Cavalcanti, F. R.; Hashemi, N.Renewable Sustainable Energy Rev. 2015, 52, 1453−1472.(12) Jenkins, G.; Wang, Y.; Xie, Y. L.; Wu, Q.; Huang, W.;Wang, L. H.;Yang, X. Microfluid. Nanofluid. 2015, 19, 251−261.(13) Jahanshahi-Anbuhi, S.; Pennings, K.; Leung, V.; Kannan, B.;Brennan, J. D.; Filipe, C. D. M.; Pelton, R. H. ACS Appl. Mater. Interfaces2015, 7, 25434−25440.(14) He, P. J. W.; Katis, I. N.; Eason, R.W.; Sones, C. L. Lab Chip 2015,15, 4054−4061.(15) Jiang, Y.; Hao, Z. X.; He, Q. H.; Chen, H. W. RSC Adv. 2016, 6,2888−2894.(16) Rajendra, V.; Sicard, C.; Brennan, J. D.; Brook, M. A. Analyst2014, 139, 6361−6365.(17) Sameenoi, Y.; Nongkai, P. N.; Nouanthavong, S.; Henry, C. S.;Nacapricha, D. Analyst 2014, 139, 6580−6588.(18) Xu, C. X.; Cai, L. F.; Zhong, M. H.; Zheng, S. Y. RSC Adv. 2015, 5,4770−4773.(19) Gallibu, C.; Gallibu, C.; Avoundjian, A.; Gomez, F. A.Micromachines 2016, 7, 6.(20) Yu, L.; Shi, Z. Z. Lab Chip 2015, 15, 1642−1645.

(21) Li, B. Q.; Fan, J. Z.; Li, J. N.; Chu, J. R.; Pan, T. R. Biomicrofluidics2015, 9, 054101.(22) Dornelas, K. L.; Dossi, N.; Piccin, E. Anal. Chim. Acta 2015, 858,82−90.(23)Mitchell, H. T.; Noxon, I. C.; Chaplan, C. A.; Carlton, S. J.; Liu, C.H.; Ganaja, K. A.; Martinez, N. W.; Immoos, C. E.; Costanzo, P. J.;Martinez, A. W. Lab Chip 2015, 15, 2213−2220.(24) Asano, H.; Shiraishi, Y. Anal. Chim. Acta 2015, 883, 55−60.(25) Pearce, J. M.; Anzalone, N. C.; Heldt, C. L. Jala 2016, 21, 510−516.(26) He, Y.; Gao, Q.; Wu, W. B.; Nie, J.; Fu, J. Z.Micromachines 2016,7, 108.(27) Focke, M.; Kosse, D.; Muller, C.; Reinecke, H.; Zengerle, R.; vonStetten, F. Lab Chip 2010, 10, 1365−1386.(28) Zhong, Z. W.; Wang, Z. P.; Huang, G. X. D. Microsyst. Technol.2012, 18, 649−659.(29) Oyola-Reynoso, S.; Heim, A. P.; Halbertsma-Black, J.; Zhao, C.;Tevis, I. D.; Cinar, S.; Cademartiri, R.; Liu, X. Y.; Bloch, J. F.; Thuo, M.M. Talanta 2015, 145, 73−77.(30) Nuchtavorn, N.; Macka, M. Anal. Chim. Acta 2016, 919, 70−77.(31) Carrasquilla, C.; Little, J. R. L.; Li, Y. F.; Brennan, J. D.Chem. - Eur.J. 2015, 21, 7369−7373.(32) Kinahan, D. J.; Kearney, S. M.; Faneuil, O. P.; Glynn, M. T.;Dimov, N.; Ducree, J. RSC Adv. 2015, 5, 1818−1826.(33) Salentijn, G. I. J.; Hamidon, N. N.; Verpoorte, E. Lab Chip 2016,16, 1013−1021.(34) Songok, J.; Toivakka, M. Microfluid. Nanofluid. 2016, 20, 9.(35) Toley, B. J.; Wang, J. A.; Gupta, M.; Buser, J. R.; Lafleur, L. K.;Lutz, B. R.; Fu, E.; Yager, P. Lab Chip 2015, 15, 1432−1444.(36) Niedl, R. R.; Beta, C. Lab Chip 2015, 15, 2452−2459.(37) Akyazi, T.; Saez, J.; Elizalde, J.; Benito-Lopez, F. Sens. Actuators, B2016, 233, 402−408.(38) da Silva, E.; Santhiago, M.; de Souza, F. R.; Coltro, W. K. T.;Kubota, L. T. Lab Chip 2015, 15, 1651−1655.(39) Ding, J.; Li, B.; Chen, L.; Qin, W. Angew. Chem. 2016, 128,13227−13231.(40) Li, W.; Qian, D.; Wang, Q.; Li, Y.; Bao, N.; Gu, H.; Yu, C. Sens.Actuators, B 2016, 231, 230−238.(41)Wu, L. D.;Ma, C.; Ge, L.; Kong, Q. K.; Yan,M.; Ge, S. G.; Yu, J. H.Biosens. Bioelectron. 2015, 63, 450−457.(42) Cunningham, J. C.; Scida, K.; Kogan, M. R.; Wang, B.; Ellington,A. D.; Crooks, R. M. Lab Chip 2015, 15, 3707−3715.(43) Wang, C. C.; Hennek, J. W.; Ainla, A.; Kumar, A. A.; Lan, W. J.;Im, J.; Smith, B. S.; Zhao, M. X.;Whitesides, G.M.Anal. Chem. 2016, 88,6326−6333.(44) Jeong, S. G.; Lee, S. H.; Choi, C. H.; Kim, J.; Lee, C. S. Lab Chip2015, 15, 1188−1194.(45) Camplisson, C. K.; Schilling, K. M.; Pedrotti, W. L.; Stone, H. A.;Martinez, A. W. Lab Chip 2015, 15, 4461−4466.(46) Bathany, C.; Han, J. R.; Abi-Samra, K.; Takayama, S.; Cho, Y. K.Biosens. Bioelectron. 2015, 70, 115−121.(47) Dey, R.; Kar, S.; Joshi, S.; Maiti, T. K.; Chakraborty, S.Microfluid.Nanofluid. 2015, 19, 375−383.(48) Matsuda, Y.; Shibayama, S.; Uete, K.; Yamaguchi, H.; Niimi, T.Anal. Chem. 2015, 87, 5762−5765.(49) Li, M.; Cao, R.; Nilghaz, A.; Guan, L.; Zhang, X.; Shen, W. Anal.Chem. 2015, 87, 2555−2559.(50) Lin, T.; Li, Z.; Song, Z.; Chen, H.; Guo, L.; Fu, F.; Wu, Z. Talanta2016, 148, 62−68.(51) Alkasir, R. S. J.; Rossner, A.; Andreescu, S. Environ. Sci. Technol.2015, 49, 9889−9897.(52) Ismail, A.; Araujo,M. O.; Chagas, C. L. S.; Griveau, S.; D’Orlye, F.;Varenne, A.; Bedioui, F.; Coltro, W. K. T. Analyst 2016, 141, 6314.(53) Cate, D. M.; Noblitt, S. D.; Volckens, J.; Henry, C. S. Lab Chip2015, 15, 2808−2818.(54) Gabriel, E. F. M.; Garcia, P. T.; Cardoso, T. M. G.; Lopes, F. M.;Martins, F. T.; Coltro, W. K. T. Analyst 2016, 141, 4749−4756.(55) Talalak, K.; Noiphung, J.; Songjaroen, T.; Chailapakul, O.;Laiwattanapaisal, W. Talanta 2015, 144, 915−921.



88


(56) Robinson, R.; Wong, L.; Monnat, R. J.; Fu, E. Micromachines2016, 7, 28.(57) Jain, P. K.; Lee, K. S.; El-Sayed, I. H.; El-Sayed, M. A. J. Phys. Chem.B 2006, 110, 7238−7248.(58) Noguez, C. J. Phys. Chem. C 2007, 111, 3806−3819.(59) Apilux, A.; Siangproh, W.; Praphairaksit, N.; Chailapakul, O.Talanta 2012, 97, 388−394.(60) Lai, T. S.; Chang, T. C.; Wang, S. C. Sens. Actuators, B 2017, 239,9−16.(61) Ferreira, D. C. M.; Giordano, G. F.; Soares, C. C. D. S. P.; DeOliveira, J. F. A.; Mendes, R. K.; Piazzetta, M. H.; Gobbi, A. L.; Cardoso,M. B. Talanta 2015, 141, 188−194.(62) Khan, S. A.; DeGrasse, J. A.; Yakes, B. J.; Croley, T. R. Anal. Chim.Acta 2015, 892, 167−174.(63) Jain, S.; Rajasingham, R.; Noubary, F.; Coonahan, E.; Schoeplein,R.; Baden, R.; Curry, M.; Afdhal, N.; Kumar, S.; Pollock, N. R. PLoS One2015, 10, e0128118.(64) Ellerbee, A. K.; Phillips, S. T.; Siegel, A. C.;Mirica, K. A.;Martinez,A. W.; Striehl, P.; Jain, N.; Prentiss, M.; Whitesides, G. M. Anal. Chem.2009, 81, 8447−8452.(65) Swanson, C.; Lee, S.; Aranyosi, A. J.; Tien, B.; Chan, C.; Wong,M.; Lowe, J.; Jain, S.; Ghaffari, R. Sensing and Bio-Sensing Research 2015,5, 55−61.(66) Jayawardane, B. M.; McKelvie, I. D.; Kolev, S. D. Anal. Chem.2015, 87, 4621−4626.(67) Chen, Y.; Zilberman, Y.; Mostafalu, P.; Sonkusale, S. R. Biosens.Bioelectron. 2015, 67, 477−484.(68) Pena-Pereira, F.; Lavilla, I.; Bendicho, C.Talanta 2016, 147, 390−396.(69) Wongwilai, W.; Jayawardane, B. M.; Kolev, S. D.; Grudpan, K.;McKelvie, I. D. Chiang Mai Journal of Science 2016, 43, 176−182.(70) Chen, W.; Fang, X.; Li, H.; Cao, H.; Kong, J. Sci. Rep. 2016, 6,31948−31948.(71) Sicard, C. m.; Glen, C.; Aubie, B.;Wallace, D.; Jahanshahi-Anbuhi,S.; Pennings, K.; Daigger, G. T.; Pelton, R.; Brennan, J. D.; Filipe, C. D.M. Water Res. 2015, 70, 360−369.(72) Mu, X.; Xin, X.; Fan, C.; Li, X.; Tian, X.; Xu, K.-F.; Zheng, Z.Chem. Commun. 2015, 51, 6365−6368.(73) Kang, S. M.; Jang, S. C.; Huh, Y. S.; Lee, C. S.; Roh, C.Chemosphere 2016, 152, 39−46.(74) Jung, Y.; Kim, J.; Awofeso, O.; Kim, H.; Regnier, F.; Bae, E. Appl.Opt. 2015, 54, 9183−9183.(75) Cardoso, T. M. G.; Garcia, P. T.; Coltro, W. K. T. Anal. Methods2015, 7, 7311−7317.(76) Peters, K. L.; Corbin, I.; Kaufman, L. M.; Zreibe, K.; Blanes, L.;McCord, B. R. Anal. Methods 2015, 7, 63−70.(77) Karita, S.; Kaneta, T. Anal. Chim. Acta 2016, 924, 60−67.(78) Zhang, Y.; Gao, D.; Fan, J.; Nie, J.; Le, S.; Zhu, W.; Yang, J.; Li, J.Biosens. Bioelectron. 2016, 78, 538−546.(79) Berry, S. B.; Fernandes, S. C.; Rajaratnam, A.; DeChiara, N. S.;Mace, C. R. Lab Chip 2016, 16, 3689−3694.(80) Chen, Y. T.; Yang, J. T. Biomed. Microdevices 2015, 17, 52.(81)Wei, X.; Tian, T.; Jia, S.; Zhu, Z.; Ma, Y.; Sun, J.; Lin, Z.; Yang, C. J.Anal. Chem. 2016, 88, 2345−2352.(82)Wei, X.; Tian, T.; Jia, S.; Zhu, Z.; Ma, Y.; Sun, J.; Lin, Z.; Yang, C. J.Anal. Chem. 2015, 87, 4275−4282.(83) Nery, E. W.; Kubota, L. T. J. Pharm. Biomed. Anal. 2016, 117,551−559.(84) Dias, A. A.; Cardoso, T. M. G.; Cardoso, R. M.; Duarte, L. C.;Munoz, R. A. A.; Richter, E. M.; Coltro, W. K. T. Sens. Actuators, B 2016,226, 196−203.(85) Cao, R.; Guan, L.; Li, M.; Tian, J.; Shen,W. Sensing and Bio-SensingResearch 2015, 6, 13−18.(86) Zhang, L.; Cao, X.; Wang, L.; Zhao, X.; Zhang, S.; Wang, P.Analyst 2015, 140, 4105−4113.(87) Mentele, M. M.; Cunningham, J.; Koehler, K.; Volckens, J.;Henry, C. S. Anal. Chem. 2012, 84, 4474−4480.(88) Choi, S.; Kim, S. K.; Lee, G. J.; Park, H. K. Sens. Actuators, B 2015,219, 245−250.

(89) Nosrati, R.; Gong, M. M.; Gabriel, M. C. S.; Pedraza, C. E.; Zini,A.; Sinton, D. Clin. Chem. 2016, 62, 458−465.(90) Yen, T. H.; Chen, K. H.; Hsu, M. Y.; Fan, S. T.; Huang, Y. F.;Chang, C. L.; Wang, Y. P.; Cheng, C. M. Talanta 2015, 145, 66−72.(91) Kannan, B.; Jahanshahi-Anbuhi, S.; Pelton, R. H.; Li, Y.; Filipe, C.D. M.; Brennan, J. D. Anal. Chem. 2015, 87, 9288−9293.(92) Augoff, K.; Hryniewicz-Jankowska, A.; Tabola, R. Cancer Lett.2015, 358, 1−7.(93) Popov, V.; Lamzin, V. Biochem. J. 1994, 301, 625−643.(94) Liang, P.; Yu, H.; Guntupalli, B.; Xiao, Y. ACS Appl. Mater.Interfaces 2015, 7, 15023−15030.(95) Badu-Tawiah, A. K.; Lathwal, S.; Kaastrup, K.; Al-Sayah, M.;Christodouleas, D. C.; Smith, B. S.; Whitesides, G. M.; Sikes, H. D. LabChip 2015, 15, 655−659.(96) Lei, K. F.; Huang, C. H.; Kuo, R. L.; Chang, C. K.; Chen, K. F.;Tsao, K. C.; Tsang, N. M. Anal. Chim. Acta 2015, 883, 37−44.(97) Zhao, M.; Li, H.; Liu, W.; Guo, Y.; Chu, W. Biosens. Bioelectron.2016, 79, 581−588.(98) Khan,M. S.; Pande, T.; van de Ven, T. G.M.Colloids Surf., B 2015,132, 264−270.(99) Nezlin, R. Mol. Immunol. 2016, 70, 149−154.(100) Tian, T.; Wei, X.; Jia, S.; Zhang, R.; Li, J.; Zhu, Z.; Zhang, H.; Ma,Y.; Lin, Z.; Yang, C. J. Biosens. Bioelectron. 2016, 77, 537−542.(101) Dirkzwager, R. M.; Liang, S.; Tanner, J. A. ACS Sensors 2016, 1,420−426.(102) Liang, L.; Su, M.; Li, L.; Lan, F.; Yang, G.; Ge, S.; Yu, J.; Song, X.Sens. Actuators, B 2016, 229, 347−354.(103) Rodríguez, M. O.; Covian, L. B.; García, A. C.; Blanco-Lopez, M.C. Talanta 2016, 148, 272−278.(104) Lathwal, S.; Sikes, H. D. Lab Chip 2016, 16, 1374−1382.(105) Kumar, A.; Hens, A.; Arun, R. K.; Chatterjee, M.; Mahato, K.;Layek, K.; Chanda, N. Analyst 2015, 140, 1817−1821.(106) Chen, X.; Chen, J.; Wang, F.; Xiang, X.; Luo, M.; Ji, X.; He, Z.Biosens. Bioelectron. 2012, 35, 363−368.(107) Chaiyo, S.; Siangproh, W.; Apilux, A.; Chailapakul, O. Anal.Chim. Acta 2015, 866, 75−83.(108) Zhang, Y.; Zuo, P.; Ye, B. C. Biosens. Bioelectron. 2015, 68, 14−19.(109) Nath, P.; Arun, R. K.; Chanda, N. RSC Adv. 2015, 5, 69024−69031.(110) Satarpai, T.; Shiowatana, J.; Siripinyanond, A.Talanta 2016, 154,504−510.(111) United States Environmental Protection Agency. Table ofregulated drinking water contaminants, https://www.epa.gov/ground-water-and-dr inking-water/table-regulated-dr inking-water-contaminants (accessed December 2016).(112) Linder, M. C.; Hazegh-Azam, M. Am. J. Clin. Nutr. 1996, 63,797S−811S.(113) Guo, J.-f.; Huo, D.-q.; Yang, M.; Hou, C.-j.; Li, J.-j.; Fa, H.-b.;Luo, H.-b.; Yang, P. Talanta 2016, 161, 819−825.(114) Apilux, A.; Ayudhya, C. I. N.; Tantimongcolwat, T.;Prachayasittikul, V. EXCLI Journal 2015, 14, 307−319.(115) Nouanthavong, S.; Nacapricha, D.; Henry, C. S.; Sameenoi, Y.Analyst 2016, 141, 1837−1846.(116) Alkasir, R. S. J.; Ornatska, M.; Andreescu, S. Anal. Chem. 2012,84, 9729−9737.(117) Shafiee, H.; Asghar, W.; Inci, F.; Yuksekkaya, M.; Jahangir, M.;Zhang, M. H.; Durmus, N. G.; Gurkan, U. A.; Kuritzkes, D. R.; Demirci,U. Sci. Rep. 2015, 5, 8719−8719.(118) Shih, C. M.; Chang, C. L.; Hsu, M. Y.; Lin, J. Y.; Kuan, C. M.;Wang, H. K.; Huang, C. T.; Chung, M. C.; Huang, K. C.; Hsu, C. E.;Wang, C. Y.; Shen, Y. C.; Cheng, C. M. Talanta 2015, 145, 2−5.(119) Liu, F.; Liu, H.; Liao, Y.; Wei, J.; Zhou, X.; Xing, D. Biosens.Bioelectron. 2015, 74, 778−785.(120) Jin, S. Q.; Guo, S. M.; Zuo, P.; Ye, B. C. Biosens. Bioelectron. 2015,63, 379−383.(121) Kim, J.-Y.; Yeo, M.-K. Mol. Cell. Toxicol. 2016, 12, 101−109.(122) Xiang, X.; Zhang, Z.; Shi, J.; Huang, F. RSC Adv. 2015, 5, 2615−2619.



89

https://www.epa.gov/ground-water-and-drinking-water/table-regulated-drinking-water-contaminants




(123) Weaver, A. A.; Lieberman, M. Am. J. Trop. Med. Hyg. 2015, 92,17−23.(124) Weaver, A. A.; Halweg, S.; Joyce, M.; Lieberman, M.; Goodson,H. V. Anal. Bioanal. Chem. 2015, 407, 615−619.(125) Myers, N. M.; Kernisan, E. N.; Lieberman, M. Anal. Chem. 2015,87, 3764−3770.(126) Jokerst, J. C.; Adkins, J. A.; Bisha, B.;Mentele, M.M.; Goodridge,L. D.; Henry, C. S. A paper-based analytical device for the colorimetricdetection of foodborne pathogenic bacteria. In 15th InternationalConference On Miniaturized Systems For Chemistry and Life Sciences 2011,Microtas 2011, 2011, 2116−2118.(127) Bidmanova, S.; Steiner, M. S.; Stepan, M.; Vymazalova, K.;Gruber, M. A.; Duerkop, A.; Damborsky, J.; Prokop, Z.; Wolfbeis, O. S.Anal. Chem. 2016, 88, 6044−6049.(128) Dungchai, W.; Chailapakul, O.; Henry, C. S. Anal. Chem. 2009,81, 5821−5826.(129) Adkins, J.; Boehle, K.; Henry, C. Electrophoresis 2015, 36, 1811−1824.(130) Mettakoonpitak, J.; Boehle, K.; Nantaphol, S.; Teengam, P.;Adkins, J. A.; Srisa-Art, M.; Henry, C. S. Electroanalysis 2016, 28, 1420−1436.(131) Yu, C. M.; Wang, Q. H.; Li, W. B.; Li, Y. B.; Liu, S. X.; Bao, N.;Gu, H. Y. Nanotechnology 2015, 26, 325501.(132) Yoo, E.-H.; Lee, S.-Y. Sensors 2010, 10, 4558−4576.(133) Rungsawang, T.; Punrat, E.; Adkins, J.; Henry, C.; Chailapakul,O. Electroanalysis 2016, 28, 462−468.(134) Yang, J.; Nam, Y. G.; Lee, S. K.; Kim, C. S.; Koo, Y. M.; Chang,W. J.; Gunasekaran, S. Sens. Actuators, B 2014, 203, 44−53.(135) Sekar, N. C.; Shaegh, S. A. M.; Ng, S. H.; Ge, L.; Tan, S. N. Sens.Actuators, B 2014, 204, 414−420.(136) Feng, Q. M.; Cai, M.; Shi, C. G.; Bao, N.; Gu, H. Y. Sens.Actuators, B 2015, 209, 870−876.(137) Lei, K. F.; Yang, S.-I.; Tsai, S.-W.; Hsu, H.-T. Talanta 2015, 134,264−270.(138) Koo, Y.; Shanov, V. N.; Yun, Y. Micromachines 2016, 7, 72.(139) Li, Z.; Li, F.; Hu, J.; Wee, W. H.; Han, Y. L.; Pingguan-Murphy,B.; Lu, T. J.; Xu, F. Analyst 2015, 140, 5526−5535.(140) Hu, J.; Ho, K. T.; Zou, X. U.; Smyrl, W. H.; Stein, A.; Buhlmann,P. Anal. Chem. 2015, 87, 2981−2987.(141) Fosdick, S. E.; Anderson, M. J.; Renault, C.; DeGregory, P. R.;Loussaert, J. A.; Crooks, R. M. Anal. Chem. 2014, 86, 3659−3666.(142) Adkins, J. A.; Henry, C. S. Anal. Chim. Acta 2015, 891, 247−254.(143) Adkins, J. A.; Noviana, E.; Henry, C. S. Anal. Chem. 2016, 88,10639−10647.(144) Glavan, A. C.; Ainla, A.; Hamedi, M. M.; Fernandez-Abedul, M.T.; Whitesides, G. M. Lab Chip 2016, 16, 112−119.(145) Wang, X.; Deng, W. P.; Shen, L.; Yan, M.; Yu, J. H.New J. Chem.2016, 40, 2835−2842.(146) Li, L.; Kong, Q. K.; Zhang, Y.; Dong, C. M.; Ge, S. G.; Yu, J. H. J.Mater. Chem. B 2015, 3, 2764−2769.(147) Li, X.; Zhao, C.; Liu, X. Microsyst. Nanoeng. 2015, 1, 15014.(148) Dias, A. A.; Cardoso, T. M. G.; Cardoso, R. M.; Duarte, L. C.;Munoz, R. A. A.; Richter, E. M.; Coltro, W. K. T. Sens. Actuators, B 2016,226, 196−203.(149) Glavan, A. C.; Christodouleas, D. C.; Mosadegh, B.; Yu, H. D.;Smith, B. S.; Lessing, J.; Fernandez-Abedul, M. T.; Whitesides, G. M.Anal. Chem. 2014, 86, 11999−12007.(150)Wang, Y.; Xu, H.; Luo, J.; Liu, J.; Wang, L.; Fan, Y.; Yan, S.; Yang,Y.; Cai, X. Biosens. Bioelectron. 2016, 83, 319−326.(151) Cunningham, J. C.; Kogan, M. R.; Tsai, Y.-J.; Luo, L.; Richards,I.; Crooks, R. M. ACS Sensors 2016, 1, 40−47.(152) Sun, G.; Zhang, L.; Zhang, Y.; Yang, H.; Ma, C.; Ge, S.; Yan, M.;Yu, J.; Song, X. Biosens. Bioelectron. 2015, 71, 30−36.(153) Li, L.; Zhang, L.; Yu, J.; Ge, S.; Song, X. Biosens. Bioelectron. 2015,71, 108−114.(154)Ma, C.; Li, W.; Kong, Q.; Yang, H.; Bian, Z.; Song, X.; Yu, J.; Yan,M. Biosens. Bioelectron. 2015, 63, 7−13.(155) Trouillon, R.; Gijs, M. A. M. RSC Adv. 2016, 6, 31069−31073.

(156) Ruecha, N.; Rodthongkum, N.; Cate, D. M.; Volckens, J.;Chailapakul, O.; Henry, C. S. Anal. Chim. Acta 2015, 874, 40−48.(157) Chaiyo, S.; Apiluk, A.; Siangproh, W.; Chailapakul, O. Sens.Actuators, B 2016, 233, 540−549.(158) Silva, T. G.; de Araujo, W. R.; Munoz, R. A. A.; Richter, E. M.;Santana, M. H. P.; Coltro, W. K. T.; Paixao, T. Anal. Chem. 2016, 88,5145−5151.(159) Cuartero, M.; Crespo, G. A.; Bakker, E. Anal. Chem. 2015, 87,1981−1990.(160) Hu, J. B.; Stein, A.; Buhlmann, P. Angew. Chem., Int. Ed. 2016, 55,7544−7547.(161) Wang, X. W.; Qin, Y.; Meyerhoff, M. E. Chem. Commun. 2015,51, 15176−15179.(162) Zangheri, M.; Cevenini, L.; Anfossi, L.; Baggiani, C.; Simoni, P.;Di Nardo, F.; Roda, A. Biosens. Bioelectron. 2015, 64, 63−68.(163) Lebiga, E.; Fernandez, R. E.; Beskok, A. Analyst 2015, 140,5006−5011.(164) Spyrou, E. M.; Kalogianni, D. P.; Tragoulias, S. S.; Ioannou, P.C.; Christopoulos, T. K. Anal. Bioanal. Chem. 2016, 408, 7393−7402.(165) Liu, F. F.; Zhang, C. S. Sens. Actuators, B 2015, 209, 399−406.(166) Wang, Y. H.; Liu, H. Y.; Wang, P. P.; Yu, J. H.; Ge, S. G.; Yan, M.Sens. Actuators, B 2015, 208, 546−553.(167) Alahmad, W.; Uraisin, K.; Nacapricha, D.; Kaneta, T. Anal.Methods 2016, 8, 5414−5420.(168) Liu, W.; Guo, Y.; Luo, J.; Kou, J.; Zheng, H.; Li, B.; Zhang, Z.Spectrochim. Acta, Part A 2015, 141, 51−57.(169) Richter, M. M. Chem. Rev. 2004, 104, 3003−3036.(170) Miao, W. Chem. Rev. 2008, 108, 2506−2553.(171) Gao, C. M.; Su, M.; Wang, Y. H.; Ge, S. G.; Yu, J. H. RSC Adv.2015, 5, 28324−28331.(172) Wu, L.; Ma, C.; Zheng, X.; Liu, H.; Yu, J. Biosens. Bioelectron.2015, 68, 413−420.(173) Li, L.; Zhang, Y.; Liu, F.; Su, M.; Liang, L. L.; Ge, S. G.; Yu, J. H.Chem. Commun. 2015, 51, 14030−14033.(174) Liu, R.; Zhang, C. S.; Liu, M. Sens. Actuators, B 2015, 216, 255−262.(175) Fan, J.; Hu, M.; Zhan, P.; Peng, X. Chem. Soc. Rev. 2013, 42, 29−43.(176) Kim, Y.; Jang, G.; Lee, T. S. ACS Appl. Mater. Interfaces 2015, 7,15649−15657.(177) Li, H.; Fang, X. E.; Cao, H. M.; Kong, J. L. Biosens. Bioelectron.2016, 80, 79−83.(178) Liang, L. L.; Su, M.; Li, L.; Lan, F. F.; Yang, G. X.; Ge, S. G.; Yu, J.H.; Song, X. R. Sens. Actuators, B 2016, 229, 347−354.(179) Chen, X.; Yu, S.; Yang, L.; Wang, J.; Jiang, C. Nanoscale 2016, 8,13669−13677.(180) Caglayan, M. G.; Sheykhi, S.; Mosca, L.; Anzenbacher, P. Chem.Commun. 2016, 52, 8279−8282.(181) Hasi, W.-L.-J.; Lin, X.; Lou, X.-T.; Lin, S.; Yang, F.; Lin, D.-Y.;Lu, Z.-W. Appl. Phys. A: Mater. Sci. Process. 2015, 118, 799−807.(182) Zhu, Y.; Zhang, L.; Yang, L.Mater. Res. Bull. 2015, 63, 199−204.(183) Zheng, G.; Polavarapu, L.; Liz-Marzan, L. M.; Pastoriza-Santos,I.; Perez-Juste, J. Chem. Commun. 2015, 51, 4572−4575.(184) Dou, B.; Luo, Y.; Chen, X.; Shi, B.; Du, Y. G.; Gao, Z. G.; Zhao,W. J.; Lin, B. C. Electrophoresis 2015, 36, 485−487.(185) Torul, H.; Ciftci, H.; Cetin, D.; Suludere, Z.; Boyaci, I. H.;Tamer, U. Anal. Bioanal. Chem. 2015, 407, 8243−8251.(186) Shao, J.; Tong, L.; Tang, S.; Guo, Z.; Zhang, H.; Li, P.; Wang, H.;Du, C.; Yu, X.-F. ACS Appl. Mater. Interfaces 2015, 7, 5391−5399.(187) Lahr, R. H.; Wallace, G. C.; Vikesland, P. J. ACS Appl. Mater.Interfaces 2015, 7, 9139−9146.(188) Saha, A.; Jana, N. R. ACS Appl. Mater. Interfaces 2015, 7, 996−1003.(189) Zheng, T. T.; Gao, Z. G.; Luo, Y.; Liu, X. M.; Zhao, W. J.; Lin, B.C. Electrophoresis 2016, 37, 418−424.(190) Katz, A. M.; Dreyer, W. J.; Anfinsen, C. B. J. Biol. Chem. 1959,234, 2897−2900.(191) Bennett, J. C. In Methods in Enzymology; Academic Press: NewYork, 1967; pp 330−339.



90


(192) Block, R. J.; Durrum, E. L.; Zweig, G. Soil Sci. 1955, 79, 409.(193) Kar, S.; Maiti, T. K.; Chakraborty, S. Analyst 2015, 140, 6473−6476.(194) Murphy, A.; Gorey, B.; de Guzman, K.; Kelly, N.; Nesterenko, E.P.; Morrin, A. RSC Adv. 2015, 5, 93162−93169.(195) Nilghaz, A.; Shen, W. RSC Adv. 2015, 5, 53172−53179.(196) Chagas, C. L. S.; Duarte, L. C.; Lobo, E. O.; Piccin, E.; Dossi, N.;Coltro, W. K. T. Electrophoresis 2015, 36, 1837−1844.(197) Luo, L.; Li, X.; Crooks, R. M. Anal. Chem. 2014, 86, 12390−12397.(198) Xu, C. X.; Zhong, M. H.; Cai, L. F.; Zheng, Q. Y.; Zhang, X. J.Electrophoresis 2016, 37, 476−481.(199) Li, X.; Luo, L.; Crooks, R. M. Lab Chip 2015, 15, 4090−4098.(200) Zhong, Z. W.; Wu, R. G.; Wang, Z. P.; Tan, H. L. J. Chromatogr.B: Anal. Technol. Biomed. Life Sci. 2015, 1000, 41−48.(201) Phansi, P.; Sumantakul, S.; Wongpakdee, T.; Fukana, N.;Ratanawimarnwong, N.; Sitanurak, J.; Nacapricha, D. Anal. Chem. 2016,88, 8749−8756.(202) Phan, D. T.; Shaegh, S. A. M.; Yang, C.; Nguyen, N. T. Sens.Actuators, B 2016, 222, 735−740.(203) Yang, R. J.; Pu, H. H.; Wang, H. L. Biomicrofluidics 2015, 9,014122.(204) Gong, M. M.; Nosrati, R.; Gabriel, M. C. S.; Zini, A.; Sinton, D. J.Am. Chem. Soc. 2015, 137, 13913−13919.(205) Yeh, S. H.; Chou, K. H.; Yang, R. J. Lab Chip 2016, 16, 925−931.(206) Han, S. I.; Hwang, K. S.; Kwak, R.; Lee, J. H. Lab Chip 2016, 16,2219−2227.(207) Hong, S.; Kwak, R.; Kim, W. Anal. Chem. 2016, 88, 1682−1687.(208) Hung, L. H.;Wang, H. L.; Yang, R. J.Microfluid. Nanofluid. 2016,20, 9.(209) Ma, B.; Song, Y. Z.; Niu, J. C.; Wu, Z. Y. Lab Chip 2016, 16,3460−3465.(210) Wong, S. R. Y.; Cabodi, M.; Rolland, J.; Klapperich, C. M. Anal.Chem. 2014, 86, 11981−11985.(211) Rodriguez, N. M.; Linnes, J. C.; Fan, A.; Ellenson, C. K.; Pollock,N. R.; Klapperich, C. M. Anal. Chem. 2015, 87, 7872−7879.(212) Ueland, M.; Blanes, L.; Taudte, R. V.; Stuart, B. H.; Cole, N.;Willis, P.; Roux, C.; Doble, P. Journal of Chromatography A 2016, 1436,28−33.(213) Kumar, A. A.; Hennek, J. W.; Smith, B. S.; Kumar, S.; Beattie, P.;Jain, S.; Rolland, J. P.; Stossel, T. P.; Chunda-Liyoka, C.; Whitesides, G.M. Angew. Chem., Int. Ed. 2015, 54, 5836−5852.



91


paper-based microﬂuidic devices: emerging themes and...

Documents