Flex. Print. Electron 1 (2016) 044001 doi:10.1088/2058-8585/1/4/044001

PAPER

Integrated NFC power source for zero on-board power in fluorescentpaper-based lateral flow immunoassays

VishakVenkatraman1, Ralph Liedert2, KennethKozak3 andAndrew J Steckl1

1 Nanoelectronics Laboratory, University of Cincinnati, Cincinnati, OH, 45221USA2 VTTTechnical ResearchCentre of Finland Ltd,Oulu, Finland3 Meridian Bioscience Inc. Cincinnati, OH, 45244USA

E-mail: a.steckl@uc.edu

Keywords:NFC, LFIA,fluorescent, quantumdots

AbstractA lab-on-chip vehicle was explored combining RFnearfield communication (NFC)harvested powerand light sources for paper-based lateralflow immunoassay systems (LFIA)with quantumdots (QD)asfluorophores. Such a vehicle has potential applications in point-of-care systems requiring highsensitivity while also being low-cost, disposable and easy to use.Micro-LEDswhich provided theexcitation source for the test line of aQD-LFIAwere surfacemounted on plastic substrates using aprinted hybrid electronics approach for the fabrication of power harvestingNFC antenna, chipassembly and electronics integrationwith the LFIA strip. The LFIA is a rotavirus assay kit with Aunanoparticles intowhichQDs emitting at 655 nmwere also incorporated. A digital camerawas usedfor detecting thefluorescence fromQDs and the reflected signal fromAuNPs. The signals werecompared using grayscale analysis. TheNFC-powered LED light source integrated with theQD-LFIAdemonstrated~9× higher sensitivity compared to conventional Au-NP based assays. Such anintegrated system can be potentiallymassmanufactured using roll-to-roll processingmaking thedevice cost effective, as well as having high sensitivity.

1. Introduction

Lab-on-chip (LOC) technology is focused on minia-turizing laboratory processes to the chip level, thusenabling portable and cost effective diagnostic sys-tems. This also enables the use of reduced samplevolume, which is critical in several applications [1, 2].Microfluidics platforms have been developed to han-dle fluids of such low volume and typically form thecore of LOC systems [3–6]. Though plastic (PDMS)-based soft lithography is widely utilized in microflui-dics [7], paper-based devices are also very attractive,owing to their low cost and ability for capillarytransport [8]. Lateral flow immunoassays (LFIA) areone such example of successful application of paper-based devices that utilize immunoreactions of antigenand antibody (Ab) molecules at detection zones,known as test and control lines [9, 10]. Transport ofthe analytes in the sample solution is by the capillaryaction of paper, which makes such devices attractiveover PDMS-based devices that typically require exter-nal pumps to drive liquids [8]. The outcome of such

devices are typically qualitative (yes/no), which isinferred visually from the change of color whencolloidal gold nanoparticles (Au-NP) collect on thetest line [11–13]. One such widely utilized LFIA in thecommercial world is the pregnancy test that detectsthe presence of human chronic gonadotropin (hCG)hormones in urine specimens [14]. Applications ofpaper-based devices in electronic displays and sensorshave also been pursued by several groups [15, 16].Such devices also have an inherent advantage of beingamenable for large volume and low unit cost usingroll-to-rollmanufacturing.

Given all the desirable characteristics of LFIAs,they still fall behind on sensitivity and limit-of-detec-tion (LOD) in comparison to enzyme-linked immu-nosorbent assay (ELISA) [17]. A possible approach toimprove LFIA sensitivity is the use of fluorescent parti-cles over conventional Au nanoparticles (NP) as thecolorimetric indicator in the assays [18–21]. Specifi-cally, quantum dots (QD) as fluorescent particles havegained considerable interest in medical applications[22–24]. Some of their desirable characteristics over

RECEIVED

25May 2016

REVISED

8 September 2016

ACCEPTED FOR PUBLICATION

11October 2016

PUBLISHED

7November 2016

© 2016 IOPPublishing Ltd

other assay particles are high photoluminescence andsimple synthesis using colloidal water solublemethods[25, 26]. The incorporation ofQDs in LFIA devices hasbeen reported by several groups. Particularly note-worthy for the application targeted in this article arethose of Yang et al [27]whohave shown a 10× increasein visual signal and Gui et al [20] who used CCD sen-sors to integrate the photo signal for best sensitivityand LOD. Several groups investigated making theassay more quantitative, along with being more sensi-tive, by using QDs in conjunction with external readersystems [24, 28, 29]. Major limitations in the work ofthese groups for point-of-care (POC) applicationswere the requirements for external readers, light sour-ces and power supplies. We have previously reported[30] one solution to this limitation by integratingorganic light emitting diodes (OLEDs) as a light sourceand successfully produced similar improvements insensitivity. However, there remains the need to inte-grate the power source into the device aswell.

Thin film batteries are possible candidates for thisapplication as they can be flexible, thin and biodegrad-able [31, 32]. However, an approach having a zero on-board power and harvesting energy from ubiquitoussourcesmay bemore desirable [33]. RF harvesting [34]on board is a possibility, however a limitation can bethe availability of the source of RF power. The incor-poration of smartphones into everyday existence hasmade their image collection/transmission and com-putation available for much more than communica-tion. Smartphones have increasingly gainedsubstantial interest as adjunct platforms in diagnosticapplications [35–37]. Shen et al [38] used a smart-phone’s camera and CPU for detection and signal pro-cessing applied to colorimetric assays. Many modernsmartphones possess near field communications(NFC) which uses RF technology for connecting devi-ces. Lee et al [39] used this technology for commu-nicating data to/from the diagnostic device. In ourapproach we focus on power harvesting from NFC-equipped smartphones. NFC harvesting antennas canbe hybrid manufactured on flexible substrates, whichimplies that they can ultimately bemassmanufacturedusing large scale roll-to-roll technology, potentiallybeing cheaper than thin film battery options. Ofcourse, in addition to serving as the power source,NFC may also be used for communication betweenthe diagnostic device and smartphone for processingand cloud connectivity.

In this manuscript we explore NFC-poweredgreen LEDs as light sources in QD-based LFIAs. WhileOLEDs have the capability for large emission area thatcan bematched to LFIA geometries, they currently stillhave emission and lifetime limitations compared toinorganic point-source LEDs. Suitable optical filtershave been selected for maximum sensitivity. A con-ventional CCD digital camera was used for detection,providing the ability to integrate the fluorescent signalfor the best sensitivity. Figure 1 shows an overall

sketch of the LFIA operation, illustrating the integra-tion of the LFIA with the antenna, LEDs and opticalfilters.

For the LFIA, a conventional Au-NP based com-mercial test kit was used. QDs introduced into the testkit with the sample solution are captured on the testline along with Au NPs. The presence of both types ofdetector particle in the same test line results in aneasier and fair comparison. Since the major focus is toobtain improved sensitivity, the lower concentrationregion is of more interest. This makes interference ofone detector type over the other negligible. To quan-tify the signal, a grayscale plot of the image was takenand the contrast was calculated.

2.NFCLED chip fabrication

The electronic backbone circuitry was fabricated usinga hybrid electronics manufacturing approach forreduced device cost, waste and footprint. Copper-coated PET foil (Prinel Ltd, Finland) with a thicknessof 50 μmwas patterned by direct printing of an etchingpaste to create the antenna structure and electroniccircuit design. Four bare-die green LEDs (ES-CEGHV15B, Epistar Corp., China, thickness 100 μm)and electronic components responsible for the powertransfer betweenmobile phone and the LED chip wereassembled onto the flexible circuit using anisotropicconductive adhesive (ACA) bonding on a custommade roll-to-roll hybrid assembly machine at VTT’sprinted and hybrid electronics pilot manufacturingenvironment (Datakon EVO 2200, Austria, equippedwith custom roll-to-roll feeder unit). The electroniccomponents used for harvesting the RF energy fromthe antenna structure are a RFID chip responsible forphone-to-device communication (Mifare UltralightMF01CU2101DUD, 120 μm thickness), and twocapacitors for impedance matching with energy har-vesting antenna and stabilization of the LED illumina-tion (2×50 pF capacitors, 100 μm thickness). All

Figure 1. Schematic showingNFCpower harvesting andintegrationwith LEDs, optical filters and LFIA.

2

Flex. Print. Electron 1 (2016) 044001 VVenkatraman et al

components are bare-die silicon chips to reduce thedevice thickness to a total thickness of 170 μmincluding the PET substrate.

Figure 2(a) illustrates the powering principle of theflexible LED chip using the NFC module of a smart-phone. The LEDs were separately characterized elec-trically and optically using an external DC bias source.Figure 2(b) shows the spectral characteristics of thegreen LEDs used as measured using Ocean Optics SD2000 spectrometer. Optical power output shown infigure 2(c) was measured (Newport Optical 1918-C)against various input currents (HP-6634B DC powersource). This plot was used to calculate a power outputvalue of 80 μW (per LED) when powered with6–7 mWusingNFC from a smartphone.

3. LFIA assay fabrication

The LFIA test kits were provided by MeridianBioscience Inc. (Cincinnati, USA) and designed fordetection of rotavirus. The operating principle of theLFIA is illustrated in figure 3. The test strip containsmouse anti-rotavirus conjugated toAuNPs as detectorparticles in the conjugate pad. The kit also contains a

nitrocellulose analytical membrane with capture anti-bodies (rabbit anti-rotavirus) embedded in the test linethat bind to an epitope of the rotavirus molecule. Acontrol line containing goat anti-mouse Ab is presentto validate the test, by capturing the conjugate mouseAbs even in the absence of rotavirus analyte. QD 655(Life Technologies) QDs conjugated to donkey anti-mouse Ab that target the mouse antibodies in theconjugate pad were introduced with the samplesolution. An antibody complex is formed as shown infigure 3(c) which then migrates toward the test lineand is captured in the presence of analyte. Hence, bothQDs and Au-NPs are present in the test and controllines, as can be seen infigures 3(a), (b).

The QDs possess wide absorption spectra in theUV-green region, but a very narrow emission line at655 nm. This large Stokes shift makes it easy to choosesimple colored plastic light filters to eliminate the exci-tation light signal. For the output light filter, amediumred (Rosco Labs) filter was used. Though the LEDhas afairly narrow emission spectrum, an input light filterwas used to ensure that there is no competing red spec-tral component from the excitation source. A Chromagreen light filter (Rosco Labs) was used for this

Figure 2.NFCpower harvesting for LEDoperation: (a) demonstration of smartphone powering LEDs (insert—photo showing LEDssurfacemounted on plastic sheet); (b) emission spectrumof the green LED; (c) LEDoutput optical power versus input current,separatelymeasured.

3

Flex. Print. Electron 1 (2016) 044001 VVenkatraman et al

purpose. The spectral characteristics of the QDs andlight filters have also been discussed in detail else-where [30].

4. Results & discussion

The rotavirus test kit was run with a standard positivesolution at a low concentration volume of 10 μl toyield a faint looking test line. Such a concentrationwould represent the LOD of this commercial test kitusing Au-NP and considered as the baseline. 3 μl ofQD solution at 1 μM concentration was added along

with the negative control solution (100 μl) and the testwas run. After subject to a uniform drying process, themembrane with captured test lines (QD & Au) wasseparated from the test kit and integratedwith the lightsource. The input and output lightfilters were attachedto either side of the analytical membrane usingadhesive. The setup consisted ofNFC-powered LED asthe light source and a spacer. The spacer was a 5 mmthick cardboardwith a circular opening. A digital CCDcamera was used as the sensor. The integrationcapability of the CCD by adjusting the shutter speedsettings was used to obtain themaximum signal for the

Figure 3. LFIAworking principle: (a) captured Au-NPs reflection under room light; (b) capturedQDparticles fluorescence excitedunderUV light; (c) immunochemistry of the LFIA showing capture of bothAu-NP andQDat test and control lines.

Figure 4. LFIA capture test line results: (a)fluorescence ofQDs excited usingNFCpowered LEDs; (b) room light reflectance of AuNPs; (c) gray value contrast comparison betweenQD andAu-NP as baseline.

4

Flex. Print. Electron 1 (2016) 044001 VVenkatraman et al

QD-based LFIA. This helped to mitigate the loss inintensity due to spacers.

Figures 4(a), (b) show the test line contrast underoptimum conditions for each case: room light condi-tions for Au NP case, dark fluorescence for QD case.As can be seen from the images, the QDs excited withlight powered by the NFC power source exhibit super-ior performance over Au NPs of the sameconcentration.

The image was converted to its equivalent grays-cale image using ImageJ [40] software and a line con-trast plot was measured. For the best results, redchannel conversion was selected for QD devices. Ascan be seen in figure 4(c) the contrast plot is consistentwith the images. The signal intensity of the test linewasquantified by calculating the area under each curvebetween 200–350 pixel distance. A ~10× increase inintensity for QD devices compared to conventionalAu-NPbased diagnostic kit was observed.

It should be noted that the signal contrast of thetest line is still limited by the amount of non-specificbinding around the test line. Hence, in an effort tobring it closer to an ideal case, a wash run was also per-formed to remove the non-bound conjugate. How-ever, in the eventual optimized case (i.e. no non-specific binding), the contrast can be amplified byintegration to the saturation level and thereby evenhigher sensitivity can be achieved.

5. Summary

In summary, a lab-on-chip vehicle powered with NFCRF power from smartphones was demonstrated.Fluorescent QD-based paper LFIAs were excited withNFC-powered LEDs fabricated by hybridmanufactur-ing processes on plastic substrates. A CCD camera wasused as detector that integrated the signal to providemaximum sensitivity. In the future the separatecamera may be replaced with the camera of thesmartphone that is also acting as a power source. Thedevices achieved nearly 10× increase in sensitivity incomparison to conventional commercially optimizedAu-NPbased LFIAs.

Such a high sensitivity device was achieved with aminimal increase in device complexity by using aprinted and hybrid electronics approach for fabri-cating a low-cost, disposable LoC device. The addedcosts for the printed electronics backbone (plasticsubstrate, conductive ink, antenna, RFID chip,micro-LEDs) is currently in the range of tens of cents(<1 $US) for mid-volume manufacturing (<1 mil-lion devices/year) and has the potential for furthercost decrease at higher volumes and continued pro-gress in printed electronics research. These resultsindicate the potential path for future work, such asintegrating a photodiode with the device to obtain anelectrical output that can then be transferred to thesmartphone via the same NFC technology. The

optoelectronic part may also be replaced with itsorganic counterparts (OLEDs and OPDs) for largerarea detection, as well as biodegradability and costeffectiveness. Such applications can also bring quan-titative detection along with increased sensitivitythat are some of the desirable characteristics of Lab-on-Chip diagnostic devices.

Acknowledgments

Two of the authors gratefully (VV and AJS) acknowl-edge support directly from an NSF grant (#1236987)and from NSF through the Center for AdvancedDesign andManufacturing of IntegratedMicrofluidics(IIP-1362048).

References

[1] WhitesidesGM2006The origins and the future ofmicrofluidicsNature 442 368–73

[2] Squires TMandQuake SR 2005Microfluidics:fluid physics atthe nanoliter scaleRev.Mod. Phys. 77 977

[3] RomanGT andKennedy RT2007 Fully integratedmicrofluidic separations systems for biochemical analysisJ. Chromatogr.A 1168 170–88

[4] YiC, Li C-W, Ji S andYangM2006Microfluidics technologyformanipulation and analysis of biological cellsAnal. Chim.Acta 560 1–23

[5] Haeberle S andZengerle R 2007Microfluidic platforms forlab-on-a-chip applications Lab on aChip 7 1094–110

[6] Yager P et al 2006Microfluidic diagnostic technologies forglobal public healthNature 442 412–8

[7] Xia Y andWhitesides GM1998 Soft lithographyAnn. Rev.Mater. Sci. 28 153–84

[8] YetisenAK,AkramMS and LoweCR2013 Paper-basedmicrofluidic point-of-care diagnostic devices Lab on aChip 132210–51

[9] NgomB,GuoY,WangX andBiD 2010Development andapplication of lateralflow test strip technology for detection ofinfectious agents and chemical contaminants: a reviewAnalyt.Bioanalyt. Chem. 397 1113–35

[10] Posthuma-TrumpieGA,Korf J and vanAmerongenA 2009Lateralflow (immuno) assay: its strengths, weaknesses,opportunities and threats. A literature surveyAnalyt.Bioanalyt. Chem. 393 569–82

[11] Phan J C et al 2016 Lateralflow immunoassays for ebola virusdisease detection in Liberia J. Infect. Diseases :jiw251(doi:10.1093/infdis/jiw251)

[12] Rivas L et al 2015Triple lines gold nanoparticle-based lateralflow assay for enhanced and simultaneous detection ofleishmaniaDNAand endogenous controlNanoRes. 8 3704–14

[13] Singh J, Sharma S andNara S 2015 Evaluation of goldnanoparticle based lateralflow assays for diagnosis ofenterobacteriaceaemembers in food andwater FoodChem.170 470–83

[14] TanakaR et al 2006Anovel enhancement assay forimmunochromatographic test strips using gold nanoparticlesAnalyt. Bioanalyt. Chem. 385 1414–20

[15] Steckl A J 2013Circuits on cellulose IEEE Spectrum 50 48–61[16] TobjörkD andÖsterbacka R 2011 Paper electronicsAdv.

Mater. 23 1935–61[17] ChinCD, LinderV and Sia SK 2012Commercialization of

microfluidic point-of-care diagnostic devices Lab on aChip 122118–34

[18] Bamrungsap S, Apiwat C, ChantimaW,Dharakul T andWiriyachaipornN 2014Rapid and sensitive lateralflowimmunoassay for influenza antigen using fluorescently-dopedsilica nanoparticlesMicrochim. Acta 181 223–30

5

Flex. Print. Electron 1 (2016) 044001 VVenkatraman et al

[19] Corstjens P L et al 2008Up-converting phosphor technology-based lateralflow assay for detection of Schistosomacirculating anodic antigen in serum J. Clin.Microbiol. 46171–6

[20] GuiC,WangK, Li C, Dai X andCuiD 2014ACCD-basedreader combinedwithCdS quantumdot-labeled lateralflowstrips for ultrasensitive quantitative detection of CagANanoscale Res. Lett. 9 1–8

[21] XiaX, XuY, ZhaoX and LiQ 2009 Lateralflow immunoassayusing europium chelate-loaded silica nanoparticles as labelsClin. Chem. 55 179–82

[22] CuiD et al 2008 Self-assembly of quantumdots and carbonnanotubes for ultrasensitive DNA and antigen detectionAnal.Chem. 80 7996–8001

[23] HanM,GaoX, Su J Z andNie S 2001Quantum-dot-taggedmicrobeads formultiplexed optical coding of biomoleculesNat. Biotechnol. 19 631–5

[24] Li Z,WangY,Wang J, TangZ, Pounds J G and Lin Y 2010Rapid and sensitive detection of protein biomarker using aportablefluorescence biosensor based on quantumdots and alateralflow test stripAnal. Chem. 82 7008–14

[25] HeR, YouX, Shao J, Gao F, Pan B andCuiD 2007Core/shellfluorescentmagnetic silica-coated composite nanoparticles forbioconjugationNanotechnology 18 315601

[26] LuW, Ji Z, Pfeiffer L,West K andRimbergA 2003Real-timedetection of electron tunnelling in a quantumdotNature 423422–5

[27] YangH et al 2010Anovel quantumdots–based point of caretest for syphilisNanoscale Res. Lett. 5 875–81

[28] Berlina AN, TaranovaNA, Zherdev AV,Vengerov YY andDzantiev B B 2013Quantumdot-based lateralflowimmunoassay for detection of chloramphenicol inmilkAnalyt. Bioanalyt. Chem. 405 4997–5000

[29] QuH et al 2016Rapid lateral-flow immunoassay for thequantumdot-based detection of puerarinBiosensorsBioelectron. 81 358–62

[30] VenkatramanV and Steckl A J 2015 IntegratedOLED asexcitation light source influorescent lateralflowimmunoassaysBiosensors Bioelectron. 74 150–5

[31] KooM et al 2012 Bendable inorganic thin-film battery for fullyflexible electronic systemsNano Lett. 12 4810–6

[32] Yin L et al 2014Materials, designs, and operationalcharacteristics for fully biodegradable primary batteriesAdv.Mater. 26 3879–84

[33] Choi S 2016 Powering point-of-care diagnostic devicesBiotechnol. Adv. 34 321–30

[34] QiaoW,ChoG and LoY-H2011Wirelessly poweredmicrofluidic dielectrophoresis devices using printable RFcircuits Lab on aChip 11 1074–80

[35] García A et al 2011Mobile phone platform as portablechemical analyzer Sensors ActuatorsB 156 350–9

[36] Martinez AW, Phillips S T, Carrilho E, Thomas SW,SindiH andWhitesides GM2008 Simple telemedicine fordeveloping regions: camera phones and paper-basedmicrofluidic devices for real-time, off-site diagnosisAnal.Chem. 80 3699–707

[37] Wang S et al 2011 Integration of cell phone imagingwithmicrochip ELISA to detect ovarian cancerHE4 biomarker inurine at the point-of-care Lab on aChip 11 3411–8

[38] Shen L,Hagen J A and Papautsky I 2012 Point-of-carecolorimetric detectionwith a smartphone Lab on aChip 124240–3

[39] Lee S et al 2016 Flexible opto-electronics enabledmicrofluidicssystemswith cloud connectivity for point-of-caremicronutrient analysisBiosensors Bioelectron. 78 290–9

[40] ImageJ software https://imagej.nih.gov/ij/

6

Flex. Print. Electron 1 (2016) 044001 VVenkatraman et al