This content has been downloaded from IOPscience. Please scroll down to see the full text.

Download details:

IP Address: 134.83.1.241

This content was downloaded on 14/03/2017 at 10:35

Please note that terms and conditions apply.

Graphene films printable on flexible substrates for sensor applications

View the table of contents for this issue, or go to the journal homepage for more

2017 2D Mater. 4 015036

(http://iopscience.iop.org/2053-1583/4/1/015036)

Home Search Collections Journals About Contact us My IOPscience

You may also be interested in:

Enhanced sheet conductivity of Langmuir-Blodgett assembled graphene thin films by chemical doping

Aleksandar Matkovi, Ivana Miloševi, Marijana Milievi et al.

Engineering electrical properties of graphene: chemical approaches

Yong-Jin Kim, Yuna Kim, Konstantin Novoselov et al.

Photonics and optoelectronics of two-dimensional materials beyond graphene

Joice Sophia Ponraj, Zai-Quan Xu, Sathish Chander Dhanabalan et al.

Combining graphene with silicon carbide: synthesis and properties – a review

Ivan Shtepliuk, Volodymyr Khranovskyy and Rositsa Yakimova

Electronic properties of graphene: a perspective from scanning tunneling microscopy and

magnetotransport

Eva Y Andrei, Guohong Li and Xu Du

Advances in graphene–based optoelectronics, plasmonics and photonics

Bich Ha Nguyen and Van Hieu Nguyen

Graphene-based wearable temperature sensor and infrared photodetector on a flexible polyimide

substrate

Parikshit Sahatiya, Sampath Kumar Puttapati, Vadali V S S Srikanth et al.

2DMater. 4 (2017) 015036 doi:10.1088/2053-1583/aa50f0

PAPER

Graphene films printable on flexible substrates for sensorapplications

Indrani Banerjee1,2, Tsegie Faris2,3, Zlatka Stoeva3, PaulGHarris2, J Chen2, AshwaniK Sharma4 andAsimKRay2

1 TheDepartment of Physics, Birla Institute of Technology,Mesra, Ranchi-835215, India2 Institute ofMaterials andManufacturing, BrunelUniversity London,Uxbridge,MiddlesexUB8 3PH,UK3 DZPTechnologies Limited, Future Business Centre, KingsHedges Road, Cambridge CB4 2HY,UK4 United States Air Force Research Laboratory, Space Vehicles Directorate, SEKirtlandAFB,NM87117USA

E-mail: asim.ray@brunel.ac.uk

Keywords: graphene ink, positive temperature coefficient of resistivity, photo-thermoelectric effect, Kohlrauch function

AbstractFifteen-layered graphene films have been successfully deposited ontoflexible substrates using acommercial ink consisting of graphene particles dispersed in an acrylic polymer binder. A value of

´ -74.9 10 cm5 2 was obtained for the density of defects, primarily located at the flake edges, from theratio of theD andGRaman peaks located at -1345 cm 1 and -1575 cm 1 respectively. m0.5 m thickdrop-cast films on interdigitated silver electrodes exhibitedOhmic conductionwith a small activationenergy of 12meVover the temperature range from 260 to 330 K.The photo-thermoelectric effect isbelieved to be responsible for photoconduction through graphene films under illumination intensityof 10mWm−2 at 270 nm, corresponding to theUV absorption peak. The photo-transient decay atthe bias of1 V involves two relaxation processes when the illumination is switched off and values of

´8.9 103 and ´4.3 10 s4 are found for the relaxation time constant using theKohlrauch stretchedexponential function analysis.

1. Introduction

The mechanical, electrical, and optical properties ofgraphene are outstandingly promising for exploitationin the wide range of possible applications such as fieldeffect transistors, sensors, for flexible and wearableelectronics [1–3]. Successful mechanical exfoliation ofhighly oriented pyrolytic graphite sample to producethin layers of graphene has been reported in 2004. Aone atom thick single layer graphene consists of sp2-hybridised carbon atoms arranged in a hexagonallattice. The overlap between the conduction andvalence bands is found to occur in ambipolar, mono-crystalline thin films showing a large carrier concen-tration of 1017 cm−2 and high mobility up to10 000 cm2 V−1 s−1 [4]. Mechanical exfoliation andchemical vapour deposition (CVD) are commonlyused to synthesise high quality graphene films, often inconjunction with high temperature annealing toreduce the number of defects [5]. Few-layer graphenewas grown on a catalyst copper foil substrate bycontrolled thermal evaporation of polystyrene carbon

source under atmospheric pressure [6]. The semi-metallic properties of zero band-gap and zero localdensity of states at the Fermi level limit the significantapplications of pristine graphene in nanoelectronics.Boron and nitrogen atoms are extensively used asdoping agents for graphene because of their similaritywith carbon atoms in atomic size. Nitrogen doping inhigh concentration contributes electrons to deloca-lised states of graphene, causing the shift of the Diracenergy [7]. The choice of doping strategies is veryimportant to achieve stable performance of the deviceover time. Stable, high transparency has been achievedfor CVD multilayer graphene doped with ferricchloride and tin (II) chloride [8]. Transverse electricfields are applied to the sample for tuning band gapwithout involving any chemical doping. For example,values of 250 meV and - -1000 cm V s2 1 1 wereobtained for tuneable band gap and mobility for anexfoliated graphene bilayer used in a dual gate field-effect transistor configuration between gold top andplatinumbottom electrodes [9].

OPEN ACCESS

RECEIVED

11 July 2016

REVISED

25November 2016

ACCEPTED FOR PUBLICATION

30November 2016

PUBLISHED

16December 2016

Original content from thisworkmay be used underthe terms of the CreativeCommonsAttribution 3.0licence.

Any further distribution ofthis workmustmaintainattribution to theauthor(s) and the title ofthework, journal citationandDOI.

© 2016 IOPPublishing Ltd

Printing is regarded being as an alternative, eco-nomically viable technique for producing micro-patternable conductive graphene films at roomtemperature over large area on flexible substratesunder ambient conditions without using vacuum orinert atmospheres. Graphene electrodes having lowsheet resistance of W 0.3M and high optical trans-parency in the order of 86% has been successfully fab-ricated by photothermal reduction of inkjet printedgraphene oxide sheets using the IR heat lamp in ambi-ent environment for about 10 min [10]. The challengewith printing graphene is in formulating the ink sincethe various rheological properties such as density, sur-face tension and viscosity have a strong effect on theprinting process. Low boiling point and environmen-tally-friendly solvents, such as ethanol andwater in thevolume ratio of 1:1 gives a surface tension of~ -31 m Nm 1 which meet the requirement of produ-cing graphene- and few-layer graphene-based inks[11]. If the graphene loading is too low, below the per-colation threshold of the system, then there will be noconductivity. A key advantage of graphene is its highaspect-ratio, flake-like, structure which will result in amuch lower threshold than if it were low aspect-ratiomorphology. Above the percolation threshold theconductivity will improve slightly as the loading isincreased, but at the expense of the coating’s mechan-ical properties, such as flexibility and extensibility.Poor mechanical stability can lead to cracks in thefilms and a loss of conductivity. The nature of the bin-der material is also critical in determining the flex-ibility of the deposits [12]. The liquid-phaseexfoliation methods suffer from poor controllabilityin size and thickness, poor production efficiency anddispersion instability. The shear exfoliation techniqueis recently reported to be faster than liquid-phase exfo-liation methods and therefore is regarded as beingscalable for mass production using commerciallyavailable inkjet printers ranging from desktop to roll-to-roll. The graphene ink (GI) is formulated with thegraphene dispersion through tuning the viscosity byadding ethylene glycol using amild sonication process.During the process, ethyl cellulose is added as stabi-liser in order to keep the shear exfoliated grapheneflakes stable inside the ink. The resulting few layer gra-phene films are highly conductive and stable over sev-eral months [13]. Graphene flake size should be fiftytimes smaller than the printing nozzle diameter. Theflake size can be controlled by fluorination of multi-layered graphene flakes in suspension of 3% aqueoussolution of hydrofluoric acid [14]. The aerogel is fabri-cated by combining drop-on-demand 3Dprinting andfreeze casting into continuous, boundary free micro-structure for 3D architectures [15].

We have recently reported a positive value of´ - -1.5 10 K3 1 for the temperature coefficient of

resistance for printed m0.5 m thick graphene films onflexible plastic substrates, over the temperature rangefrom 260 to 330 K [16]. In light of this preliminary

observation, further work has been undertaken toinvestigate the scope of fabricating low cost graphenethermistors and UV photosensors, using this material.A number of earlier studies have focused on the use ofgraphene in hybrid architectures for photodetectordevices [17]. This article presents the results of aninvestigation into the photocurrent transient relaxa-tion behaviour, of simple graphene films, that can beprinted over large area and on flexible substrates.Raman Spectroscopy, Scanning Electron Microscopy(SEM) andUV–Visible spectroscopy studies have beencarried out on graphene films, in an effort to get a bet-ter understanding of the material’s structure andmorphology, and hence shed light on its interestingelectrical properties.

2. Experimental procedure

The commercially available ink from DZP technolo-gies Ltd (product number G087) was formulated withfew-layer graphene particles of lateral dimensions

m<40 m, obtained by liquid phase exfoliation. Thegraphene particles were then dispersed in an acrylicpolymer binder system to obtain the ink suitable fordeposition on polyethylene terephthalate (PET) sub-strates (Melinex DuPont Teijin). Organic bindingadditives are not normally used in graphene filmsdeposited for example byCVDprocesses. The ink usedin this work is specifically formulated for industrial,large area printing and coating methods such as slot-die, gravure and flexography which are high-speed,volume manufacturing processes. However, the sam-ples for electrical measurements consisted of drop castGI films, m ( )t0.5 m in thickness , on predesignedinterdigitated silver electrode structures, on PETsubstrates. The dropcast method was chosen here forthe small scale, rapid deposition and the inkwasmixedusing high-speed mixer immediately before drop-casting using amicro-pipette with a view to ensuring ahigh degree of the film uniformity and reproducibility.This interdigitated electrode structure has been chosenin order to facilitate direct illumination on an enlargedphotodetection area. A Keithley 617 electrometer inthe microprocessor controlled measuring system wasemployed to record both dark current and photocur-rent characteristics as a function of bias voltage V forthe drop cast samples, which were held under vacuumof -10 mbar5 in anOxford Instrument liquid nitrogenconstant bath cryostat. The mechanical stability of thegraphene coating was investigated by measuring theroom temperature resistance of drop cast films, on0.1 mm thick PET substrates between two planar silverelectrodes at the ends of the film GI under bothextension and compression conditions of bending.The PET substrate was much longer than the GI filmin order to apply uniform stress on theGI film.

A monochromator (HR320 Jobin-Yvon, HOR-IBA) was used as a light source in the spectral range of

2

2DMater. 4 (2017) 015036 I Banerjee et al

300–700 nm for photoconduction experiments. UV–visible absorption spectra of the printed film on quartzsubstrates were recorded using a Perkin-ElmerLAMBDA 650 spectrophotometer between 190 and900 nm scanning at the rate 654.8 nmmin−1. The sur-face microstructure of the GI drop-cast on an alumi-nium foil was investigated using an S3400N scanningelectron microscope. The samples were characterisedbymeans of x-ray diffraction using Bruker D8 advancescanning instrument. Cu Kα radiation of wavelength0.15406 nm was used with scattering angle variedbetween 10° and 90° at scanning rate of 0.02° per secfor crystallographic analysis of the samples. TheRaman spectra were obtained by Horiba Jobin YvonLab RAM HR800 with incident laser beam of wave-length 514.5 nm (equivalent to the excitation laserenergy = )E 2.41 eVL of 2 μmspot size.

3. Results and discussions

Experimental results are analysed in order to derivephysically meaningful information regarding latticestructure, defect type and density, and electrical andoptical device parameters. Values of material para-meters have been compared with relevant publisheddata for identifying new knowledge.

3.1.Microstructural, crystallographic andcompositional studiesFigure 1 shows a scanning electron micrograph of thegraphene deposit. It is apparent that the lateraldimensions of the flakes are in the micrometre range,although of course their thickness is three orders ofmagnitude smaller. Such materials pose challenges forthis form of microscopy because the thickness of the

flakes is massively smaller than the penetration depthof the electron beam used, and is even small comparedto the range of secondary electrons used to form theimage. As a result the three-dimensional structure ofthe whole near surface contributes to the image, ratherthan the normal situation when the image is domi-nated by the topography of the outermost surface.Additionally, the backscattered electron and second-ary yields of graphene are both extremely low [18], andso ultimate resolution must be compromised in orderto achieve sufficient signal to form an image.

The XRD spectrum in figure 2 of the sample showsa strong and sharp [002] peak at 26.19° correspondingto an interlayer distance of =d 0.34 nm002 determinedusing Bragg’s law,where the subscript (002) refers to thediffraction plane. These results are in good agreementwith those obtained for exfoliated graphite oxide pre-cursor after electrochemically reduction at 1.5 V, andare close to those for pristine graphite [19]. A value of5.57 nm is estimated for the average crystallite size,D ,002 from theDebye–Scherrer formula in the form:

= lb q

( )D , 10020.94

cos

where the value of the full width at half maximum(FWHM) b = 1.5 is found from figure 2.

The number of graphene layers NGL is estimatedto be 15 from the expression = -( )D N d1002 GL 002

assuming D002 represents the thickness as measuredfrom the centre of the sample. The stacking of theselayers is responsible for the intensity and sharpness ofthe [002] peak [20]. The additional weak peak at2θ=13° due to the reflection from [010] planeimplies the presence of residual oxygenated functionalgroups with interlayer spacing of 0.68 nm in the crys-tallinematrix of GI [21].

Figure 1. SEMpicture of graphene ink sample drop cast on an aluminium foil.

3

2DMater. 4 (2017) 015036 I Banerjee et al

The Raman spectra in figure 3 for the GI samplewere recorded for the wavenumbers ranging from1000 to 3000 cm−1. The characteristics peaks ofD andG have been observed at 1345 cm−1 and 1575 cm−1

respectively in figure 3(a). These peak positions are inagreement with those observed for the annealed gra-phene layer ink-jet printed on a Si/SiO2 substrate [22].The D band is due to the breathing modes of sp2 ringsof six-atom rings. The intense G band arises from in-plane vibration of sp2 carbon atoms present in the

sample [23]. The intensity ratio of ID/IG is found to be0.3, reflecting the defect concentration of carbonac-eousmaterials [24]. A value of 20.6 nm is estimated foran average inter-defect distance LD in nm using therelation [25]:

= ´ -( )⎡⎣ ⎤⎦L , 2

E

I

ID2 4.3 10 1

D

G

3

L4

where =E 2.41 eVL is the laser excitation energy.Although the spot size is nominally 2 μm, the laserbeam is expected to penetrate into the depth of

Figure 2.X-ray diffraction spectrumof the drop-cast ink on an aluminium foil.

Figure 3.Raman spectrum for the graphene ink sample drop-cast on aluminium foil for (a) 1000 cm−1�λ−1�2000 cm−1 and (b)2500 cm−1�λ−1�3000 cm−1

fittedwith three Lorentzian components.

4

2DMater. 4 (2017) 015036 I Banerjee et al

>50 nm encountering a large number of grapheneflakes. Therefore, the estimated value of LD is consid-ered to be reasonable in the context of the scatteringwithin a large volume [26]. The value of

´ -74.9 10 cm5 2 is found for the defect density nDfrom the knowledge that =

p( )n .

LD2 1

D2 These defects

are believed to be located at the flake edges responsiblefor activation by double resonance giving rise to theDpeak. Figure 3(b) shows that a broad 2D peak due toabsorptions by second order zone-boundary phononsappears at 2690 cm−1, representing a blue-shift by∼30 cm−1 in relation tomonolayer samples [27]. Also,a value of 80 cm−1 for the FWHM of the 2D peak isapproximately three times larger than that reportedfrom single layer graphene. These observations areconsistent with earlier XRD studies that the sample ismulti-layered. This 2D peak, usually referred to as asecond-order overtone of the D peak, is not assignedto the defects for their activation since its origin isattributed to the momentum conservation by twophonons with opposite wave vectors. However, thepositions of both D and 2D peak are dispersivedependent on the laser excitation energy. Figure 3(b)also illustrates an analysis of the 2D peak envelope,assuming a Lorentzian peak shape. Three componentsare found to fit the envelope with their peaks to belocated at 2675, 2700 and 2725 cm−1.

3.2. Steady state electricalmeasurementsAs shown in figure 4(a), in-plane dc conductivitymeasurements were made using the interdigitatedelectrode systems with dimensions of =d 0.5 mm,

=W 0.82 cm and =N 5, where d,W and N are thedistance between the fingers, the overlapping widthand the number of fingers in the electrode system,respectively. Figure 4(b) shows the current I versusbias voltage V plots for the sample at =T 268 K and

=T 330 K. These I(V ) characteristics are linearwithin the bias voltage regime, implying the Ohmicnature of conduction. The dark Ohmic conductivitysD of the GI sample is estimated to be

-1642 123 S m 1 at 268 K from the relation:

s = - ( )( ) . 3R

d N

t wD1 1

.D

Values of dark Ohmic resistance RD were deter-mined from the slope of individual ( )I V plot infigure 4(a). This value of s = -1642 S mD

1 com-pares well with one observed for a printed m0.22 mthickmulti-layered graphene film which was annealedat 400 °C. The ink was produced from ultrasonicatedozonemodification of aqueous dispersions containingexfoliated graphite [28]. Similarly a value of

-1231 117 S m 1 was obtained for sD at 330 K. A

Figure 4. (a)Electrode configurations for in-plane conductivitymeasurements and (b) dark I(V ) characteristics for I(V )characteristics of graphene ink sample for =T 268 K (broken line) and =T 330 K (solid line).

5

2DMater. 4 (2017) 015036 I Banerjee et al

comparable value of -1100 S m 1 for the room temper-ature conductivity is reported for a ∼25 nm thick filmspin-coated with dispersions of few layered graphenein N,N-dimethylformamide (0.7 mgml−1) solvent[29]. However, the resistivity of thin printed samplesof UV-driven water-based graphene oxide/acrylicpoly(ethylene glycol) diacrylate nanocomposite isreported to vary between 104 and W10 m6 with theincrease in the film thickness from 4 to 12 μm [30].This shows the drop cast film of the present invest-igation is considerably more conducting than thesenanocomposites.

Four further measurements were carried out attemperatures between 268 and 330 K and thedependence of the dark resistivity rD on T is shown infigure 5(a). This behaviour of r ( )TD can be describedby a simplewell-known relation in the form:

r r= + µ[ ] ( )T1 , 4D D0

where rD0 and α denote the resistance at 0 K and thetemperature coefficient of resistivity of the GI sample.The ratio of the slope to the intercept at =T 0 K gives

´ - -( )3.07 0.12 10 K3 1 for the temperature coef-ficient α. This value is larger than one reported forgraphene nanowalls by an order of magnitude [31]. Itis, however, within the same order of magnitude asother highly conducting materials such as copper andsilver [32]. The existence of positive temperaturecoefficient may be attributed to an increase in the

interlayer distance with the rise of temperature,thereby reducing the tunnelling current between thelayers [33]. The contributions of both optical phononsand intervalley scattering by transverse and long-itudinal modes become increasingly significant withrising temperature [34]. The value of the activationenergyΔE is estimated to be 12 meV from the slope ofArrenhius plot in figure 5(b) of s( )ln D as the inverse oftemperature -T .1 D = E 12 1.5 meV represents anon-zero band gap between the conduction andvalence band and it is smaller than room temperaturethermal energy. The small value of D =E 12 meV inthe present work implies the possible break up oflattice symmetry at the Dirac points (K) of thegraphene films due to the presence of defects [35]. Thisobservation is consistent with our earlier observationsfrom Raman and SEM studies. ΔE is generallysensitive to the measurement environment such as thepresence of moisture, local humidity and the hydro-phobicity of the binder resin. A value ofD =E 29 meV which is almost equal to thermalenergy is reported for CVD deposited graphene filmsfrom Van Der Pauw conductivity measurement invacuum for the comparable temperature range [36].Both temperature coefficient and activation energyhave been measured for graphene flake- and solarexfoliated reduced graphene within a similar temper-ature range with a view to developing wearabletemperature sensors. The temperature coefficient isnegative with the value of the same ordermagnitude as

Figure 5. (a)Dependence of dark resistivity rD onT and (b)Arrhenius plot of s( )ln D as the inverse of temperature -T .1

6

2DMater. 4 (2017) 015036 I Banerjee et al

obtained in the present investigation, implying thepotential applications of the present GI in the develop-ment wearable temperature sensors. Also, values of24.19 and 90.7 meV for activation energies of gra-phene flake- and solar exfoliated reduced graphenerespectively are higher than one obtained in thepresent drop-cast graphene film, indicating non-metallic characteristics of these graphene [37].

3.3. Electrical properties under bending stressThe bending is commonly applied as strain foruniaxial tensile stress. As shown in figure 6(a), thestrain was applied parallel to the length of the film sothat the electrodes were not subjected to any strain.The strain is estimated in percentage from the

knowledge of the substrate thickness and radius ofcurvature. Three radii of curvature of 2, 7 and 12 nmwere used. Figure 6(b) shows the results of the bendingtest in terms of percentage change of resistances,D( )R

RBS

B0due to stretching. RB0 is the fresh film

resistance prior to application of any stretching andcompression stress and D = -R R RBS BS B0 where

RBS is the resistance of the strained film. DR

RBS

B0is found

to increase with increasing strain, implying an increasein the resistance (RBS) on stretching. Similar behaviourwas reported for CVD grown graphene layers [38].Phenomenologically, the stretching causes theincrease in length of the film with simultaneousdecrease in its thickness and both these dimensional

Figure 6. (a)A schematic illustration of the electricalmeasurement for theflexible GL/PETdevice under strain. (b)Resistance changerelative to fresh film resistance RB0 for GIfilmunder stretching (triangle symbols) and compression (rectangle symbol)s, (c) recoveryof the graphene film onwithdrawal of (A) stretching and (B) compression strains.

7

2DMater. 4 (2017) 015036 I Banerjee et al

changes contribute to the increase in Ohmic resistanceon stretching. The Dirac points in the strainedgraphene are believed to have undergone displacementfrom the K points, possibly opening up the energy gap.TheGI film surfacemay be not completely flat becauseof imperfect adhesion to the PET substrate. Thescattering of charges due to the non-flat nature of thestrained GI surface may also contribute to the increasein resistance [39]. The measurements were repeatedon the same film under compression over the samestrain range and the results are also illustrated in

figure 6(b) in the terms of the dependence of D( )R

RBC

B0

on the compressing strain whereD = -R R RBC BC B0

in which RBC is the resistance of the compressed film.As expected, the resistance is found to decrease withthe increase in the strain. The opposite dimensionalchanges of the films occur on compression, reducingthe Ohmic film resistances. In order to investigate therepeatability of the gauge performance, the resistanceof the film (RBw)was measured after the withdrawal ofbending stresses and the magnitude of recovery was

estimated as the ratio ( )R

RBw

B0and it is evident from

figure 6(c), values of this ratio lie between 97% and92% as the strain is reduced from 0.71% to 0.41% andthe recovery is consistent between both stretched andcompressed films. These results indicate a very gooddegree of the flexibility of the GI films, leading topossible application of these flexible GI films inresistance strain gauges for measuring strain withinthe range from0.41% to 0.71%.

3.4. Photoresponse studiesThe UV–vis absorption spectrum in figure 7(a) for thegraphene film is due to for these interband transitionsshowing a pronounced and asymmetric peak at270 nm [40]. This peak position is in close agreementwith that observed for films deposited from n-methyl-2-pyrrolidone and ethanol/water based dispersions[11]. The peak is due to a van Hove singularity in thedensity of states, which occurs close to the hoppingenergy [41, 42]. The considerable red-shift withrespect to the absorption peak of graphene oxide at226 nm implies reduced disruption of the π-conjuga-tion in the multi-layered graphene film [43]. The flatabsorption band over the visible region results fromthe linear dispersion of Dirac electrons in graphene.The weak broad absorption band (shoulder) around353 nm corresponds to residual graphitic sp2 domainsin the samples. However, other structural parameterslike the number of graphitic layers, curvature ofaromatic layers, crystallite size and so on, also affectthe shifting of (π–π*) bond position. The actionspectrum in figure 7(b) shows the dependence ofphotocurrent on the incident wavelength, at a fixedbias of 50 mV. As expected, it shows close similaritiesto the UV–vis absorption spectrum, with a sharp peakoccurring at 270 nm. The minor differences may beattributed to scattering and recombination of chargesin theGIfilm.

Figure 8(a) shows I(V ) characteristics from a GIsample recorded between 10 mV were recorded at

=T 268 K and =T 330 K under illumination of a

Figure 7. (a)Room temperatureUV–vis spectrum for the used graphene ink and (b) action spectra.

8

2DMater. 4 (2017) 015036 I Banerjee et al

constant light intensity =P 10 m Wmi2 corresp-

onding to the absorption peak at 270 nm. The result-ing plots were found to be linear for bothtemperatures. Values of photoconductivity sph areestimated to be -8.05 0.67 S m 1 and

-2.8 0.16 S m 1 at 268 K and 330 K, respectivelyusing the following relation:

s = - - ( )( ) , 5R R

R R

d N

t wph1

.D il

D il

where Ril is the resistance for the GI sample underillumination.

This decrease in sph by a factor of nearly 3 with anincrease in temperature of 62 K from268 to 330 Kmaybe interpreted by the photo-thermoelectric effect dueto a temperature gradient that is possibly producedbetween the graphene layers under illumination. Thephotocurrent in ordinary semiconductors is primarilydue to separation of excited electron–hole pairs by a

Figure 8. (a) I(V ) characteristics of graphene ink sample for =T 268 K (broken line) and =T 330 K (soild line) under illuminationof l = 270 nm. (b)Current versus time as the light was turned on and off is under illumination at 270 nm. (c)Kohlrauch equationfitting (solid line) to the experimental decay of current curve (solid symbols)when the illuminated light was turned off between370 and 490 s.

9

2DMater. 4 (2017) 015036 I Banerjee et al

built-in electric field. A photoexcited e–h pair leads toultrafast heating of the carriers in graphene because ofstrong e–e interactions [44, 45].

The photoresponsivity, S, is defined as the ratio ofthe photocurrent to the incident optical power Pi.Therefore S can explicitly be written for the inter-digitated electrode system in the form:

=s

-( )

( )S . 6t w

d N N

V

P

.

12

ph

i

The value of photosensitivity S at 268 K is esti-mated to be -161 A W 1 from equation (6). As thetemperature is raised to 330 K, S is found to decreaseto -56 A W .1 This value is of the same order of magni-tude as one obtained for the hybrid films containinggraphene quantum dots in the reduced grapheneoxide matrix [46]. The specific detectivity D, can beestimated from the knowledge of S using the expres-sion

=s -

( )( ( ( )))

D , 7S

q d N2 tw 1D

where = ´ -q 1.6 10 C19 is the absolute value ofelectronic charge. Using the expression above, valuesof ´4.9 1012 Jones and ´1.9 1012 Jones were esti-mated from equation (7) at 268 K and 330 K, respec-tively. These values of the specific detectivity representthe capability of the devices to detect a small photosignal.

Values of the short circuit current Isc at zero bias,V=0, are found to be m2.0 A and m0.4 A at 268 Kand 300 K, respectively, indicating that the number ofphoto-generated charges are smaller at higher tem-peratures for a given illumination strength [47]. Simi-larly, values of 6.7 mV and 1.6 mV are estimated foropen circuit voltage Voc using I=0 for 268 K and300 K, respectively. Figure 8(b) shows a typical currentversus time cycle for a fixed bias of 1 V as the light wasturned on and off. Two conducting states, high andlow, of the GI films are found to exist under illumina-tion and the recovery to nearly low conducting occurswhen the illumination is switched off. The responsetime tR is estimated to be 42 s from fitting the expo-

nential function = - -t( )⎡

⎣⎢⎤⎦⎥i i 1 exp t

0R

to exper-

imental rise curve in figure 8(b), where i0 is thesaturation photocurrent for =t 0 at which instantlight illumination was switched on at 150 s. This valueis found to be comparable with those obtained for theUV sensors based upon reduced graphene oxide deco-rated ZnO nanostructures [48]. Figure 8(c) presentsthe results of the analysis of the time-dependent decayof the photocurrent current in terms of the Kohlrauchequation in the form:

= -t

g( )( )( ) ( )i t i exp , 8t0

D

where i0 is the saturation photocurrent for =t 0 atwhich instant light illumination is switched off. Thestretching parameter g normally varies betweenvalues of 0 and 1. For g = 1, the function in

equation (6) approaches classical single-exponentialbehaviour and differences in energy transfer processesbecome indistinguishable under this condition [49]. Itis obvious that the decay curve in the present work ischaracterised by two regimes corresponding to (i) fastand (ii) slow relaxation processes. Values of therelaxation time constant, t ,D were found to be

´8.9 10 s3 and ´4.3 10 s4 for the time scale regime(i) from 371 s to 463 s and (ii) from 463 s to 482 srespectively by fitting equation (8) to the experimentaldata. The present g = 0.82 may be ascribed to thecombined effect of both defects and potential barriersbetween the edges of the single sheets within theoverall assemblies [50].

4. Concluding remarks

A low cost, easy fabrication method has been devel-oped for depositing graphene films on flexible sub-strates and their optoelectronic properties have beeninvestigated for practical applications. The filmsstudied in this article could be deposited using forexample rotary screen-printing, gravure or even off-set lithography which are high-speed, volume manu-facturing processes. The graphene film in the presentinvestigation is composed of multiple, randomlydistributed graphene nano-sheets, in contrast to CVDproduced graphene which is typically continuouslayers grown directly onto a substrate. It should also benoted that a polymer binder is added to the ink. Avalue of ´ - -3.07 10 K3 1 for the positive temperaturecoefficient of resistivity of the graphene film iscomparable to that of commonmetals like copper andsilver and these graphene films may therefore beemployed in the manufacture of thermistors for theiruses as electronic components in resettable controlcircuits. The graphene films show a high specificdetectivity, of the order of 1012 at its maximumabsorbance wavelength of 270 nm under a constantlight intensity of 10 mWm−2. Two distinct high andlow resistive states have been observed. The observa-tion that the properties of our films are similar to ‘real’graphene is important because these results may leadto potential exploitation for low cost, large-scaleindustrially development of highly efficient, large areaUV sensors using printed graphene films. Future workwill include optimisation of the deposition (printing)procedure to obtain uniform films and investigationinto the effect of different polymer binders with a viewto improving the mechanical and chemical stability ofthe film. The graphene content relative to that ofpolymer binder and the nature of the binder will beinteresting features of further investigations intodesign and fabrication of resistance strain gauges. Itmay then be possible to improve the recovery perfor-mance by using a binder with better elasticity for straingauge transducers.

10

2DMater. 4 (2017) 015036 I Banerjee et al

Acknowledgments

Dr Indrani Banerjee is grateful to CommonwealthAssociation, UK for funding the present research workunder the fellowship placement scheme (Grant refer-ence INCF-2014-66). The studentship of Ms Faris ispartially sponsored by the Air ForceOffice of ScientificResearch, Air ForceMaterial Command, USAF, underGrant No. FA9550-15-1-0123. We are also thankful toMiss VMTorrejon of Brunel University for support incomputer graphics.

References

[1] JangH, Park Y J, ChenX,Das T, KimMS andAhn JH2016Graphene-basedflexible and stretchable electronicsAdv.Mater. 28 4184–202

[2] TrungTQand LeeNE 2016 Flexible and stretchable physicalsensor integrated platforms for wearable human-activitymonitoring and personal healthcareAdv.Mater. 28 4338–72

[3] NguyenBHandNguyenVH2016 Promising applications ofgraphene and graphene-based nanostructuresAdv.Nat. Sci.-Nanosci. Nanotechnol. 7 023002

[4] NovoselovK S, GeimAK,Morozov SV, JiangD, Zhang Y,Dubonos SV,Grigorieva I V and FirsovAA 2004 Electricfieldeffect in atomically thin carbon films Science 306 666–9

[5] WuB, TuncerHM,Katsounaros A,WuWP,MatthewTC,YingK, Zhang LH,MilneW I andHaoY 2011Microwaveabsorption and radiation from large-areamultilayer CVDgrapheneCarbon 77 814–22

[6] Ahmadi S andAfzalzadeh R 2016 Few-layer graphene growthfrompolystyrene as solid carbon source utilizing simpleAPCVDmethod PhysicaE 81 302–7

[7] Joucken F et al 2015Charge transfer and electronic doping innitrogen-doped graphene Sci. Rep. 5 14564

[8] KangMH,MilneW I andColeMT2015Doping stability andopto-electronic performance of chemical vapour depositedgraphene on transparent flexible substrates IETCircuits DevicesSyst. 9 39–45

[9] ZhangYB, TangTT,Girit C,HaoZ,MartinMC, Zettl A,CrommieMF, Shen YR andWang F 2009Direct observationof awidely tunable bandgap in bilayer grapheneNature 459820–3

[10] KongD, Le LT, Li Y, Zunino J L and LeeW2012Temperature-dependent electrical properties of graphene inkjet-printed onflexiblematerials Langmuir 28 13467–72

[11] CapassoA, Castillo A ED, SunH,AnsaldoA, Pellegrini V andBonaccorso F 2015 Ink-jet printing of graphene for flexibleelectronics: an environmentally-friendly approach Solid StateCommun. 224 53–63

[12] ChenGH,WuD J,WengWUandWuCL 2003 Exfoliation ofgraphiteflake and its nanocompositesCarbon 41 619–21

[13] Majee S, SongM, Zhang S L andZhang ZB 2016 Scalable inkjetprinting of shear-exfoliated graphene transparent conductivefilmsCarbon 102 51–7

[14] NebogatikovaNA, Antonova IV, Kurkina I I, Soots RA,VdovinV I, Timofeev VB, Smagulova SA and PrinzVY 2016Fluorinated graphene suspension for inkjet printedtechnologiesNanotechnology 27 205601

[15] ZhangQQ, Zhang F,Medarametla S P, LiH, ZhouC and LinD2016 3Dprinting of graphene aerogels Small 12 1702–8

[16] Banerjee I, Faris T, Stoeva Z, SharmaAK andRayAK2016Printed graphene filmswith positive temperature coefficient ofresistivityMater. Today: Proc. 3 4035–39

[17] DangVQ, TrungTQ,KimD I,Duy LT,Hwang BU, LeeDW,KimBY, Toan LD andLeeNE 2015Ultrahigh responsivity ingraphene-ZnOnanorod hybridUVphotodetector Small 113054–65

[18] Luo J, Tian P, PanCT, RobertsonAW,Warner JH,Hill EWandBriggsGAD2011Ultralow secondary electronemission of grapheneACSNano 5 1047–55

[19] GuoHL,WangXF,QianQY,Wang FB andXia XH2009Agreen approach to the synthesis of graphene nanosheetsACSNano 3 2653–9

[20] Huh SH2014X-ray diffraction ofmulti-layer graphenes:instantmeasurement and determination of the number oflayersCarbon 78 617–21

[21] Toh SY, LohK S, Kamarudin SK andDaudWRW2014Graphene production via electrochemical reduction ofgraphene oxide: synthesis and characterisationChem. Eng. J.251 422–34

[22] Torrisi F et al 2012 Inkjet-printed graphene electronicsACSNano 6 2992–3006

[23] Ferrari AC 2007Raman spectroscopy of graphene andgraphite: disorder, electron–phonon coupling, doping andnonadiabatic effects Solid State Commun. 143 47–57

[24] EckmannA, FeltenA,Verzhbitskiy I, Davey R andCasiraghi C2013Raman study on defective graphene: effect of theexcitation energy, type, and amount of defects Phys. Rev.B 88035426

[25] Ferrari AC andBaskoDM2013Raman spectroscopy as aversatile tool for studying the properties of grapheneNat.Nanotechnol. 8 235–46

[26] NiZH,WangYY, YuT, Yu and ShenZX2008Ramanspectroscopy and imaging of grapheneNanoRes. 1 273–91

[27] Ferrari AC et al 2006Raman spectrumof graphene andgraphene layers Phys. Rev. Lett. 97 187401

[28] Rider AN,AnQ, Thostenson ET andBrackN2014Ultrasonicated-ozonemodification of exfoliated graphite forstable aqueous graphitic nanoplatelet dispersionsNanotechnology 25 495607

[29] WuYP,Wang B,MaY F,HuangY, LiN, Zhang F andChenY S 2010 Efficient and large-scale synthesis of few-layeredgraphene using an arc-dischargemethod and conductivitystudies of the resulting films’NanoRes. 3 661–9

[30] Giardi R, Porro S,ChiolerioA,Celasco and SangermanoM2013‘Inkjet printed acrylic formulationsbased onUV-reducedgraphene oxide nanocomposites’ J.Mater. Sci.481249–55

[31] Yang J,WeiDP, Tang L L, SongXF, LuoW,Chu J, Gao TP,ShiH F andDuCL 2015Wearable temperature sensor basedon graphene nanowallsRSCAdv. 5 25609–15

[32] Giancoli DC 1995Physics 4th edn (EnglewoodCliffs, NJ:Prentice-Hall)

[33] TianM,HuangY,WangWH, Li RQ, Liu P, LiuCX andZhang YG2014Temperature-dependent electrical propertiesof graphene nanoplatelets film dropped on flexible substratesJ.Mater. Res. 29 1288–94

[34] BorysenkoKM,Mullen J T, Barry EA, Paul S, Semenov YG,Zavada JM,NardelliMB andKimKW2010 First-principlesanalysis of electron–phonon interactions in graphene Phys.Rev.B 81 121412

[35] Yavari F, Kritzinger C, Gaire C, Song L,Gullapalli H,Borca-Tasciuc T, Ajayan PMandKoratkarN 2010Tunablebandgap in graphene by the controlled adsorption ofwatermolecules Small 6 2535–8

[36] DvorakM,OswaldWandWuZG2013 Bandgap opening bypatterning graphene Sci. Rep. 3 2289

[37] Sahatiya P, Puttapati S K, SrikanthVVS S andBadhulika S2016Graphene-basedwearable temperature sensor andinfrared photodetector on a flexible polyimide substrate Flex.Print. Electron. 1 025006

[38] Lee YHandKimY J 2012 Electrical and lattice vibrationalbehaviors of graphene devices onflexible substrate under smallmechanical strainAppl. Phys. Lett. 101 083102

[39] HaniffMASM,Hafiz SM,WahidKAA, Endut Z, LeeHW,BienDCS, Azid I A, AbdullahMZ,HuangNMandRahman SA2015 Piezoresistive effects in controllabledefectiveHFTCVDgraphene-based flexible pressure sensorSci. Rep. 5 14751

11

2DMater. 4 (2017) 015036 I Banerjee et al

[40] LeeC, LeconteN, Kim J, ChoD, Lyo IW andChoi E J 2016Optical spectroscopy study on the effect of hydrogenadsorption on grapheneCarbon 103 109–14

[41] MakKF, Ju L,Wang F andHeinz T F 2012Opticalspectroscopy of graphene: from the far infrared to theultraviolet Solid State Commun. 152 1341–9

[42] Kravets VG,GrigorenkoAN,Nair RR, Blake P, Anissimova S,NovoselovK S andGeimAK2010 Spectroscopic ellipsometryof graphene and an exciton-shifted vanHove peak inabsorptionPhys. Rev.B 81 155413

[43] SurUK, SahaA,Datta A, Ankamwar B, Surti F, Roy SD andRoyD2016 Synthesis and characterization of stable aqueousdispersions of grapheneBull.Mater. Sci. 39 159–65

[44] XuXD,GaborNM,Alden J S, van der ZandeAMandMcEuenP L 2010 Photo-thermoelectric effect at a grapheneinterface junctionNano Lett. 10 562–6

[45] Koppens FHL,Mueller T, Avouris P, Ferrari AC,VitielloMS and PoliniM2014 Photodetectors based on

graphene, other two-dimensionalmaterials and hybridsystemsNat. Nanotechnol. 9 780–93

[46] TamTV,Hur SH,Chung J S andChoiWM2015Ultravioletlight sensor based on graphene quantumdots/reducedgraphene oxide hybrid film Sensors ActuatorsA 233 368–73

[47] SunR J, Zhang Y, Li K,HuiC,HeK,MaXC and Liu F 2013Tunable photoresponse of epitaxial graphene on SiCAppl.Phys. Lett. 103 013106

[48] WangZX, ZhanXY,WangY J,Muhammad S,HuangY andHe J 2012AflexibleUVnanosensor based on reducedgraphene oxide decorated ZnOnanostructuresNanoscale 42678–84

[49] Covington LR andMoore J C 2013 Photoconductivity andtransient response of Al:ZnO: Al planar structures fabricatedvia a thermal oxidation processThin Solid Films 540 106–11

[50] Loomis J and Panchapakesan B 2012 Large photocurrents insingle layer graphene thin films: effects of diffusion and driftNanotechnology 23 265203

12

2DMater. 4 (2017) 015036 I Banerjee et al