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

Download details:

IP Address: 80.250.180.203

This content was downloaded on 07/10/2013 at 10:03

Please note that terms and conditions apply.

Properties and applications of chemically functionalized graphene

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

2013 J. Phys.: Condens. Matter 25 423201

(http://iopscience.iop.org/0953-8984/25/42/423201)

Home Search Collections Journals About Contact us My IOPscience

IOP PUBLISHING JOURNAL OF PHYSICS: CONDENSED MATTER

J. Phys.: Condens. Matter 25 (2013) 423201 (22pp) doi:10.1088/0953-8984/25/42/423201

TOPICAL REVIEW

Properties and applications of chemicallyfunctionalized graphene

M F Craciun, I Khrapach, M D Barnes and S Russo

Centre for Graphene Science, College of Engineering, Mathematics, and Physical Sciences,University of Exeter, Exeter EX4 4QL, UK

E-mail: m.f.craciun@exeter.ac.uk and s.russo@exeter.ac.uk

Received 17 July 2013, in final form 26 July 2013Published 17 September 2013Online at stacks.iop.org/JPhysCM/25/423201

AbstractThe vast and yet largely unexplored family of graphene materials has great potential for futureelectronic devices with novel functionalities. The ability to engineer the electrical and opticalproperties in graphene by chemically functionalizing it with a molecule or adatom is wideningconsiderably the potential applications targeted by graphene. Indeed, functionalized graphenehas been found to be the best known transparent conductor or a wide gap semiconductor. Atthe same time, understanding the mechanisms driving the functionalization of graphene withhydrogen is proving to be of fundamental interest for energy storage devices. Here we discussrecent advances on the properties and applications of chemically functionalized graphene.

(Some figures may appear in colour only in the online journal)

Contents

1. Introduction 12. Engineering the physical properties of graphene by

covalent functionalization 22.1. Graphene oxide 22.2. Hydrogenated graphene 72.3. Fluorinated graphene 9

3. Patterning of graphene circuits by local functionaliza-tion 103.1. Electron-assisted defluorination 113.2. Laser-assisted fluorination 113.3. Electron-assisted hydrogenation 113.4. AFM assisted thermochemical reduction of GO 11

4. Two-dimensional intercalated graphitic materials 125. Intercalated quasi-free few-layer graphene 136. Making graphene the best transparent conductor 14

6.1. FeCl3-intercalated few-layer graphene 146.2. Engineering the optoelectronic properties of

graphene by functionalization with quantumdots 15

7. Future prospects 16Acknowledgments 16References 16

1. Introduction

Atomically thin materials are the thinnest materials whichcan be conceived and yet they have a unique combinationof physical properties. Graphene is certainly the mostcelebrated and studied representative of this new familyof materials [1–3]. This single layer of carbon atomsis the strongest known material, and the best electricaland thermal conductor which is mechanically flexible andtransparent [1, 3]. In its pristine form the electrical conductionof graphene cannot be switched off, limiting the range ofpotential applications which will exploit the properties ofgraphene. At the same time, though graphene is nearlytransparent (it absorbs only 2.3% of the light in the visiblewavelength [4–6]), its electrical conductivity is still high formaking pristine graphene suitable as a transparent electrode.However, the limitations of the physical properties ofgraphene can be overcome using chemical functionalization.

10953-8984/13/423201+22$33.00 c© 2013 IOP Publishing Ltd Printed in the UK & the USA

J. Phys.: Condens. Matter 25 (2013) 423201 Topical Review

Indeed, recent experiments have shown that the optical andelectrical properties of graphene can be controlled to fit therequirements of specific applications by chemical bonding ofa molecule or a chemical element to the pristine material.For example, the functionalization with FeCl3 of few-layergraphene results in a new graphene-based material which isthe best known transparent electrical conductor and whichoutperforms the indium tin oxide (ITO) commonly usedtransparent electrode by the display companies. On theother hand, the functionalization with fluorine, hydrogen andoxygen makes graphene semiconducting. Engineering theproperties of graphene with chemical functionalization is atthe focus of current research since it holds the promise toexpand significantly the potential range of novel applicationswhich will be based on graphene materials.

2. Engineering the physical properties of grapheneby covalent functionalization

The non-interacting first neighbour tight binding approx-imation predicts a gapless energy dispersion for single-and few-layer graphene (FLG) [7]. The lack of an energygap is the direct consequence of the energetic equivalencebetween the onsite energy of the sublattices constitutingthe unit cell of these graphene materials (see figures 1(a)and (b)). Breaking this energetic symmetry between thesublattices directly leads to the opening of an energy gap.This can easily be accomplished in doubly gated transistorstructures of FLGs with no mirror reflection symmetry [8](e.g. AB-stacked bilayer, ABC-stacked trilayer, etc). In thesedevices the FLG is sandwiched between a back and a topgate which allow the independent control of a perpendicularelectric field acting on the FLG and of the Fermi energy,see figures 2(a) and (b). More specifically, the voltagesapplied to the top and back gate originate a difference in theelectrostatic energy in the layers composing the FLG, leadingto the opening of a gate-tunable bandgap, see figures 2(c)and (d). Optical spectroscopy experiments in doubly gatedAB-stacked bilayer graphene reported for the first time thedirect observation of this gate-tunable energy gap whose valuesaturates of ∼200 mV [9]. These experimental results arein stark contrast to previous electrical transport experiments,where the characteristic energy scales for the non-linearityreported in measurements of current versus voltage bias wasat most 1 mV [10]. The contrast between the observationsreported by optical spectroscopy and electrical transportexperiments is yet unsolved. So far, the direct observation ofa gate-tunable energy gap in electrical transport experimentsremains elusive.

Although the doubly gated FLG devices are of greatinterest for both fundamental and applied research, themaximum values of bandgap which can be opened in thesematerials is limited to applications in the far-infrared rangeand not in the visible wavelength range. More specifically,while at low enough perpendicular electric fields (Eperp) thevalue of the energy gap increases linearly with increasingEperp, at large Eperp the value of the energy gap saturatesdue to screening caused by the finite density of states

Figure 1. Tilted view of the crystal structure (left) and schematicband structure (right) of monolayer graphene (a) and bilayergraphene (b). The unit cell contains two equivalent carbon atoms Aand B for monolayer graphene (a), four atoms A1, B1, A2 and B2for bilayer graphene (b) highlighted in the figure. For Bernal (orAB) stacked bilayer the top layer has its A1 atom on top of the B2atom of the bottom layer as indicated by the dashed line in (b).

in the FLGs. The chemical functionalization of grapheneoffers a unique way to extend the limited range of bandgapenergies achievable by doubly gated devices [8]. Indeed,a change in the hybridization of the graphene electronicorbitals from sp2 to sp3 can open large band-gaps (>4 eVin the case of fully fluorinated and hydrogenated graphene).This functionalization can be accomplished by chemicallybonding a molecule [11], a functional group containingoxygen (i.e. graphene oxide) and/or an element such asfluorine or hydrogen to the carbon atoms of graphene as wewill review in the following sections.

2.1. Graphene oxide

Graphene oxide (GO) is formed when oxygen containingfunctional groups are covalently bonded to the network ofsp2 hybridized carbon atoms of graphene. Several functionalgroups have been shown to exist in GO, such as carboxyl,hydroxyl and epoxy groups. GO has triggered researchinterest due to its wide range of physical and chemicalproperties [13–15], but also to its potential for the productionof graphene on the ton scale [16, 17] by using differentmethods to remove the oxygen and create the so-calledreduced GO. The chemistry of GO is still under debate, withseveral models that have been proposed for the chemicalstructure of GO and comprehensive reviews that have recentlyappeared on this topic [18, 14]. The most accepted modelwhich is supported by solid-state 13C NMR spectroscopy [19]is shown in figure 3(a). In this model there are epoxy andhydroxyl groups which lie above and below the layer ofsp2 carbon atoms, whereas carboxylic groups are thought to

2

J. Phys.: Condens. Matter 25 (2013) 423201 Topical Review

Figure 2. (a) Cross section view of a suspended and doubly gated few-layer graphene device. (b) False colour SEM image of a suspendedand doubly gated ABC-stacked trilayer graphene taken under a tilted angle. Reproduced from [12]. (c) Schematic band structure ofABC-stacked trilayer graphene with broken energetical symmetry between the outer planes, for example by means of an external voltage(VG) applied to the top and back gates as illustrated in (a). (d) The resistance versus back-gate voltage (Vbg) measured at T = 300 mK andfor different values of fixed top-gate voltage as indicated in the graph in a doubly gated ABC-stacked trilayer. Reproduced with permissionfrom [12]. Copyright 2012 American Institute of Physics.

Figure 3. (a) Schematic model for the structure of GO indicating the presence of different oxygen functional groups. Reproduced withpermission from [18]. Copyright 2009 Royal Society of Chemistry. (b) High-resolution transmission electron micrograph of GO, illustratingits amorphous character. The inset represents a selected area electron diffraction pattern in which diffraction dots are absent and onlydiffraction rings are visible, confirming the disordered, amorphous nature of GO. Reproduced with permission from [20]. Copyright 2008American Chemical Society. (c) Atomic resolution, aberration-corrected transmission electron micrograph of reduced GO. The defect-freecrystalline graphene area is displayed in the original light grey colour, whereas the different structures have been highlighted with differentcolours: dark grey, contaminated regions; blue, disordered single-layer carbon networks or extended topological defects; red, individualadatoms or substitutions; green, isolated topological defects; yellow, holes and their edge reconstructions. Reproduced with permissionfrom [50]. Copyright 2010 American Chemical Society. (d) Comparison of transmittance at 550 nm as a function of sheet resistance of GOfilms after undergoing different reduction treatments. Reproduced with permission from [15]. Copyright 2010 Wiley. (e) Photograph ofunreduced GO (leftmost) and a series of high-temperature reduced GO films of increasing thickness. Reproduced with permissionfrom [30]. Copyright 2008 American Chemical Society.

3

J. Phys.: Condens. Matter 25 (2013) 423201 Topical Review

populate the edges of the layers. Due to the presence of thesefunctional groups, which are randomly adsorbed, GO hasa disordered, amorphous structure which has been revealedby both x-ray diffraction and high-resolution transmissionelectron microscopy (HRTEM) studies [20] (see figure 3(b)).

There are several methods to produce GO which arebased either on wet chemistry or plasma oxidation. The mostcommon wet chemical techniques consist of oxidation ofgraphite in strong acids, followed by a liquid exfoliationprocess. Graphite oxide can be produced by either theBrodie [21], Staudenmaier [22] or Hummers method [23].The Brodie and Staudenmaier methods use a mixture ofpotassium chlorate with nitric acid, whereas the Hummersmethod is based on a solution of potassium permanganateand sulfuric acid. The liquid exfoliation of GO from graphiteoxide can be achieved by simple sonication of graphiteoxide in water and various organic solvents. These techniqueshave been extensively discussed in several reviews [18, 17].Although the wet techniques employed to synthesize GO aresuited for large-scale production, they inevitably introducecontamination in the final product. Therefore, alternativeplasma-based routes to produce GO have been recentlydeveloped. These methods are based on exposure ofgraphene sheets to low-power oxygen plasma [24–29]. Afterexposure to oxygen plasma, a transition from semimetallicto semiconducting behaviour in graphene has been reported,which was due to the functionalization of graphene withoxygen atoms [26, 27]. More controlled plasma-oxidationtechniques include the use of a water vapour plasma [28]or downstream oxygen plasma [29]. For instance, Liuused a water vapour plasma to oxidize graphene througha nanosphere lithography mask, which ensured that theoxidation occurs in a mild and controllable manner [28]. Onthe other hand, in the downstream oxygen plasma technique,the samples are placed downstream from the plasma sourceso that generated ions are relaxed upon arrival at the graphenesurface and functionalize it in a controllable way [29].

The carbon–oxygen bonds in GO are sp3 hybridized andthe substantial sp3 fraction in GO renders it a wide bandgapsemiconductor. Thus as-prepared GO is typically insulatingand exhibits sheet resistance values of around 1012 �/sq orhigher [30]. However, the conductivity of GO can be tunedby reducing the oxygen groups through various processesand changing the hybridization of carbon atoms from sp3 tosp2 bonding. A variety of methods for the reduction of GOhave been studied [18, 31–39], ranging from reaction withhydrazine [18, 31, 32], high-temperature pyrolysis [30, 40, 41]and plasma-assisted reduction [42–44] to bacterial treat-ment [34]. The reduction of GO results in the creation ofclusters of sp2 carbon atoms which are visible in HRTEMstudies (see figure 3(c)). The presence of sp2 clusters leadsto the decrease of GO sheet resistance by several orders ofmagnitude [45] as shown in figure 3(d) and in the increaseof its optical transparency (figure 3(e)). However, due to thecomplications arising from the structural disorder in GO,only a few studies have been devoted to the understandingof the electronic structure of GO. Thus, the fundamentalelectronic properties such as the bandgap values for GO

and reduced GO are yet not well understood. For instancetheoretical studies predict a bandgap in GO to be in the rangeof 1.7–2.4 eV depending on the degree of oxidation [46–48],but experimental studies have yielded lower values of around0.25 eV [49].

One of the attractive properties of GO is that its oxygenfunctional groups render it hydrophilic and make it solublein water [51] and many other solvents [52]. This solubilityallows GO to be uniformly deposited onto any substrateusing cost-effective methods such as drop-casting [18, 17],spin coating [30] or inkjet printing [53]. Thus, due to suchinexpensive methods of production and the availability oflarge quantities, GO and its reduced version are commerciallyavailable at reasonable prices, which makes these materials oftremendous interest for a wide range of applications. Belowwe give an overview of the application of GO and reducedGO which have been demonstrated in different fields.

The readily tunable electrical and optical propertiesover a wide range via chemical engineering, togetherwith the simple, low-temperature deposition onto a widevariety of substrates make GO attractive for electronics andoptoelectronics [15]. The major areas where GO can beexpected to have potential is in the production of transparentconductive electrodes, as well as in its use as an activepart of devices such as semiconducting channel or memorymaterial. Any type of substrate can be coated with solutionprocessed GO thin films, which can be subsequently convertedinto a conductor using different reduction treatments.Several groups have evaluated reduced GO films astransparent and conducting materials [54, 55, 30, 41].It has been shown that thin films of reduced GO can be madetransparent and conducting by optimizing the thickness andreducing at high temperatures. Various oxidation, exfoliation,dispersion, deposition, and reduction procedures have beenemployed, which resulted in a range of transmittance andsheet resistance values that have been reported in theliterature [15]. The highest degree of reduction was achievedusing high-temperature pyrolysis (1100 ◦C) which yieldingfilms with sheet resistances of a few k�/sq and 90%transmittance [30, 40, 41] (see figure 3(d)). Such coatingshave been used as electrodes for photovoltaic [41, 56–58]and light-emitting [59] devices. However, in all of theseapplications the devices employing electrodes based onreduced GO have lower performance than devices based oncommonly used transparent electrodes such as fluorine tinoxide (FTO) and indium tin oxide (ITO) (see figures 4(a)and (b)). For example, Wang et al compared the performanceof solar cells based on reduced GO electrode and on fluorinetin oxide (FTO) electrode [41]. The cell based on reducedGO electrode has a short-circuit photocurrent density (Isc) of1.01 mA cm−2 with an open-circuit voltage (Voc) of 0.7 V,filling factor (FF) of 0.36, and overall power conversionefficiency of 0.26%. The cell based on FTO electrode(which is commonly used in organic solar cells) has Isc of3.02 mA cm−2, Voc of 0.76 V, FF of 0.36, and an efficiencyof 0.84%. In this case the lower Isc and efficiency of thecell based on reduced GO film was attributed to the seriesresistance of the device, the relatively lower transmittance ofthe electrode, as well as the electronic interfacial change.

4

J. Phys.: Condens. Matter 25 (2013) 423201 Topical Review

Figure 4. (a) Comparison of performance of solar cells based on reduced GO electrode (black I–V curve) and on fluorine tin oxide (FTO)electrode (red I–V curve), illuminated under AM solar light (1 sun). The inset shows the schematic structure of dye-sensitized solar cell, thefour layers from bottom to top are Au, dye-sensitized heterojunction, compact TiO2, and reduced GO film. Reproduced with permissionfrom [41]. Copyright 2008 American Chemical Society. (b) Left panel: current density (filled symbols) and luminance (open symbols)versus applied forward bias for an (organic light-emitting diode) OLED on graphene (squares) and ITO (circles), with OLED devicestructure anode/PEDOT:PSS/NPD(50 nm)/Alq3(50 nm)/LiF/Al as shown in the inset. Right panel: external quantum efficiency (EQE)(filled symbols) and luminous power efficiency (LPE) (open symbols) for an OLED on graphene film (squares) and ITO glass (circles).Reproduced with permission from [59]. Copyright 2010 American Chemical Society. (c) Transfer characteristics of reduced GO withdifferent degrees of reduction measured at T = 300 and 78 K. The labels, 8 m, 15 m, 30 m, and 16 h (m = minutes, h = hours)correspond to total time of exposure to hydrazine. HG-A and HG-B are devices with 16 h of exposure which were further annealed in N2/H2

(90/10) atmosphere at 150 ◦C for 1 h. The inset shows an optical micrograph of a typical reduced GO device. Reproduced with permissionfrom [64]. Copyright 2009 American Chemical Society. (d) The top left panel shows a schematic illustration of a GO based flexible memorydevice. The top right panel shows a typical I–V curve of a device plotted on a semilogarithmic scale. The arrows indicate the voltage sweepdirection. The left inset is a photo of the device. Continuous bending effect of the device is shown in the bottom left panel, with the insetsshowing photographs of repeated two bending states. the bottom right panel shows the endurance performance of the device measuredduring 100 sweep cycles. Reproduced with permission from [68]. Copyright 2010 American Chemical Society.

Another area where GO has been employed is itsuse as channel material for field effect transistors (seefigure 4(c)). Thin films of reduced GO exhibit ambipolar fieldeffect [60], with on/off ratio of typically lower than 10 andmobility values ranging between 0.001 and 10 cm2 V−1 s−1,depending on film thickness and reduction conditions [60].Individual sheets of reduced GO have also been investigatedas semiconducting channel [61, 55, 62–66]. Their ambipolarcharacteristics and mobilities are similar to those of the thinfilms, suggesting that sheet junctions in films play a minorrole in carrier transport.

Perhaps one of the electronics areas where GO is mostsuited is in resistive switching memory applications [67–73](see figure 4(d)). For these applications a metal/GO/metalsandwich structure is fabricated and the conductivity of GOis electrically switched from its high resistance state to a low

resistance state by applying a voltage between the two metalelectrodes. Reversible and reproducible resistive switchingbehaviours have been observed in GO thin films with differenttype of metal electrodes, but the mechanisms driving theswitching process have not been un-ambiguously determined.Several mechanisms have been proposed such as creatingand breaking metal filaments [67], oxygen migration to/froman insulating interface layer between the electrode and GOfilm [68] or the formation of sp2 graphene clusters in ansp3 insulating graphene oxide layer [69]. Observations ofreversible resistive switching in GO films has been reportedby several groups [67–73]. In all these cases the switchingwas ambipolar and recent studies [73] have demonstratedreasonable on/off ratios (∼300), practical high and lowresistance values (∼300 and 1 k�), reasonable switchingvoltages (∼2 V) and some cyclability (up to 100 cycles).

5

J. Phys.: Condens. Matter 25 (2013) 423201 Topical Review

Figure 5. (a) Top: normalized photoluminescence (PL) spectra of the GO suspensions after different exposure times (0180 min) tophotothermal reduction treatment. Bottom: photographs of tunable PL emission from GO at reduction times of 0 min (yellow–red), 75 min(green) and 180 min (blue). Reproduced with permission from [75]. Copyright 2012 Wiley. (b) Digital camera image of graphene oxidepaper (left) and SEM side-view image (right). The bottom panel shows a comparison of tensile strength σ and modulus E for a set of thinpaper-like materials which demonstrates that GO paper outperforms many other paper-like materials in stiffness and strength. Reproducedwith permission from [16]. Copyright 2007 Nature. (c) Electrochemical characterizations of lithium-ion batteries using reduced GO whichshows that the capacity retention of Mn3O4/reduced GO is superior to that of free Mn3O4 nanoparticles without graphene. Reproduced withpermission from [89]. Copyright 2010 American Chemical Society.

The heterogeneous electronic structure of GO translatesinto unique optical properties which make GO a chemicallytunable platform for optical applications [74]. Thus, incontrast to pristine graphene, GO is fluorescent over a broadrange of wavelengths such as near-infrared (NIR), visibleand ultraviolet (UV), with a maximum intensity locatedbetween 500 and 800 nm [74–79]. The tunability of theemission spectrum of GO has also been demonstrated bymeans of progressive reduction of the oxygen content [76, 75](see figure 5(a)). Thus for reduced GO, blueshift [75, 80–82]and redshift [76] of the fluorescence from UV to NIR havebeen reported. Although considerable progress in the studiesof GO has been achieved, the origin of the fluorescence ofGO is still a controversial issue with several mechanismsproposed in the literature. In early studies, it was suggestedthat the fluorescence arises from the recombination ofelectron–hole pairs in localized electronic states originatingfrom the heterogeneous electronic structure of GO [74].For reduced GO it has been proposed that the fluorescenceis due to the disorder-induced states and the sp2 clustersformed during the reduction reaction [75, 80]. At the sametime, Luo et al [76] proposed that the bond distortionsalso give contributions to the fluorescence of GO andreduced GO. Studies of pH-dependent fluorescence of GOsuggested that the emission of quasi-molecular fluorophoresgive rise to the fluorescence [83]. Recent time-resolvedfluorescence measurements of GO in water, complementedby theoretical analysis, suggest that the GO fluorescenceis due to electron–hole recombination from the bottomof the conduction band and nearby localized states towide-range valance band [84]. In particular it has been

found that electronic transitions between the non-oxidizedcarbon regions and the boundary of oxidized carbon atomregions give the predominant emission from GO. Finally,similar photoluminescence features have been observed in GOproduced by oxygen plasma treatment of graphene, which wasassigned to correlated localized electronic states of oxidationsites [24]. Such unique optical properties of GO over awide range of wavelengths are promising for devices suchas light-emitting diodes, white light emission for solid-statelighting and flexible display applications.

Owing to its fluorescence, GO has found applicationsin biology and medicine, such as bio-sensing, early diseasedetection, and even assisting in carrying cures for cancer,as well as other drugs and gene delivery. Thus, GO hasbeen successfully used in fluorescent-based biosensors forthe detection of DNA and proteins, with a promise of betterdiagnostics of HIV. Furthermore, GO was tested as a drugcarrier and it has been shown that it has a unique abilityin the attachment and delivery of aromatic, water insolubledrugs [85]. It has also been shown that GO is superiorto many other anticancer drugs because it does not targethealthy cells, only tumours, and has a low toxicity [86].The intrinsic photoluminescence of GO has been used forlive cell imaging in the near-infrared [85] and it was alsodemonstrated that GO is a robust candidate for DNA andprotein analysis, and intracellular tracking [87]. GO haseven been used as antibacterial material as it was shownto have superior antibacterial effects. Thus the growth ofE. coli bacteria has been effectively inhibited by GO withminimal cytotoxicity [88]. This finding is of great potential fordeveloping antibacterial materials that may assist in healing

6

J. Phys.: Condens. Matter 25 (2013) 423201 Topical Review

Figure 6. (a) Schematic structure of graphene (top) and its characteristic flat structure with sp2 hybridized electronic orbitals (bottom left)which is opposed to the crumpled structure of graphane in the chair conformation with the hydrogen atoms alternating on both sides of theplane and characterized by sp3 hybridized electronic orbitals (bottom right). The carbon atoms are shown in blue and red while thehydrogen atoms are in grey. (b) Predicted band structure for the fully hydrogenated chair conformer. Reproduced with permissionfrom [104]. Copyright 2007 American Physical Society.

wounds by killing bacteria or that can be used for moresanitary packaging than current materials.

GO can mix readily with a wide range of organic andinorganic materials by interacting in non-covalent, covalentand/or ionic way to form functional hybrids, compositesand paper-like materials. Graphene-based composite ma-terials [90, 91] have been proposed as a possible routeto harness the properties of graphene such as thermalconductivity, mechanical stiffness, fracture strength andelectronic transport properties. Such composites have beensuggested as a radical alternative to conventional polymersin the fields of transportation and electronics. Incorporationof graphene sheets in composites requires the graphenesheets to be produced on a sufficient scale and to beincorporated, and homogeneously distributed, into variousmatrices. Due to its solubility, GO allows the preparation ofgraphene-polymer composites and molecular-level dispersionof individual reduced graphene oxide sheets within polymerhosts. Thus graphene composites with styrenic polymerssuch as polystyrene, acrylonitrile–butadiene–styrene andstyrene–butadiene rubbers have been shown to exhibitthe lowest reported value of room temperature electricalconductivity for any carbon-based composite except carbonnanotubes (at only 1 vol%, the polystyrene–graphenecomposite has a conductivity of 0.1 S m−1). In its solidform, GO platelets tend to stack one to another, formingthin and extremely stable paper-like structures that can befolded, wrinkled, and stretched [16, 92]. The GO paper is afree-standing carbon-based membrane material with a uniqueinterlocking-tile arrangement of the nanoscale GO sheets.This new material has been shown to have a combinationof macroscopic flexibility, stiffness and strength whichoutperforms many other paper-like materials (see figure 5(b)).Such free-standing GO films are considered for applicationsincluding protective layers, chemical filters, componentsof electrical batteries or supercapacitors, adhesive layers,electronic or optoelectronic components, ion conductors,nanofiltration membranes and molecular storage.

GO and its reduced forms have an extremely largesurface area and therefore have been considered aspotential materials in energy-related applications such aselectrodes in rechargeable lithium-ion batteries [93, 89] andsupercapacitors [94–96], as well as fuel cells [97, 98].For instance, it was demonstrated that nanocompositeanchored on conducting graphene as anode material forlithium-ion batteries displays superior battery performancewith large reversible capacity, excellent cyclic performance,and good rate capability [93, 89]. Several groups havedemonstrated both types of supercapacitors based on GO:(1) electrochemical double-layer capacitors that store energythrough the accumulation of charges at the interface of a highsurface area electrode and an electrolyte [94, 95] and (2)redox capacitors that store energy Faradically by battery-typeoxidation reduction reactions [99].

2.2. Hydrogenated graphene

Adsorbed chemical elements such as hydrogen or fluorine onthe surface of graphene can form covalent bonds with thecarbon atoms and create graphane (hydrogenated graphene)or fluorographene (fluorinated graphene). This process isknown as chemisorption and it is accompanied by a changein the hybridization of the electronic orbitals of graphenefrom sp2 to sp3. Consequently, the planar two-dimensionalcrystal structure of graphene planes characterized by sp2

bonds is transformed into a three-dimensional structure withsp3 bonds whereby the C atoms are pulled out of the grapheneplane [100–103] (see figure 6(a)). The first theoretical studypredicting that hydrogenated single-layer graphene is a stablematerial was presented by Sofo et al [104] using densityfunctional theory and presenting first-principles total-energycalculations. In contrast to the case of adsorbed oxygen whichis randomly adsorbed on graphene [16, 17, 15], the adsorptionof hydrogen [105–107] and similarly of fluorine [108–114]only takes place in a highly ordered way. More specifically,

7

J. Phys.: Condens. Matter 25 (2013) 423201 Topical Review

the hydrogen (or fluorine) atoms are covalently bonded to bothsides of the carbon sheet in an alternating manner, see figure 3.

Density functional theory calculations predict that thefunctionalization of graphene with hydrogen modifies theenergy dispersion of graphene in a radical way, leadingto the opening of a wide energy gap (3.5 eV forfully hydrogenated graphene [104, 115]), see figure 6(b).Uniquely to functionalized graphene, the value of the energygap is theoretically expected to scale with the hydrogencoverage [116]. Engineering an energy gap in graphenematerials is a first necessary step for the development ofgraphene-based digital electronics which would enable novelflexible electronic applications in the offing.

The change in hybridization of the electronic orbitalsfrom sp2 to sp3 is expected to induce a radical modificationalso in other physical properties of these new materialssuch as a large spin–orbit interaction [117–120], whichwas recently experimentally measured [121]. Furthermore,even magnetic properties are expected to originate from theinduced change in the hybridization of electronic orbitalswhich would widen considerably the range of potentialapplications of functionalized graphene. At the same time, theability to bind and reversibly release hydrogen from graphenesuggests the possibility to use this material for high-densityhydrogen storage [104], which is of interest for example forthe automotive industry [122].

To date, the successful hydrogenation of graphene hasbeen demonstrated by several groups using various techniquessuch as exposure to hydrogen plasma [106, 105, 123–125],atomic hydrogen beams [107, 126–129], electron-induceddissociation of hydrogen silsesquioxane [130, 121] andmore recently also with the Birch reduction process [131].Each method presents some advantages as well as somedisadvantages, making the resulting material suitable fordifferent specific applications.

2.2.1. Remote hydrogen plasma. In the remote plasmatechnique, molecular hydrogen is dissociated in the hydrogenplasma and free to bond to the carbon of pristine graphene.To minimize the creation of atomic defects due to thehighly energetic ions in the plasma, the graphene substrate istypically placed in a remote side of the chamber away fromthe discharge zone [106, 105, 76, 124]. In this method it isdifficult to control the degree of induced atomic defects aswell as the stoichiometry of the functionalization. The remoteplasma makes it possible to obtain a large hydrogen coveragealthough when the graphene flakes are laying on a substratethe hydrogenation process takes place only on the exposedside of graphene, limiting the coverage of adatoms. Alsoin this one-sided hydrogenated graphene, it is energeticallyfavourable for the hydrogen atoms to bond with differentsublattices [132, 133].

2.2.2. Atomic hydrogen beam. Another method ofhydrogenation relies on the use of an atomic hydrogenbeam [107, 126–129]. To reach hydrogen saturation usuallythe source is operated at high temperatures (2200 K) and ina very low hydrogen gas partial pressure (10−7 mbar) [126].

The opening of a bandgap in the energy dispersion ofgraphene through functionalization with an atomic hydrogenbeam was directly measured by angle-resolved photoemissionspectroscopy [107, 126–129]. These experiments also pointedout that to accomplish large hydrogen absorption on graphene,previously grown on oriented transition metal substrates,a reactive substrate is needed. Similarly to the plasmatechnology, also with this method it is difficult to control thecoverage of hydrogen atoms and the amount of atomic defectsinduced in graphene.

2.2.3. Electron-induced dissociation of hydrogensilsesquioxane. The future development of whole-grapheneelectronics is reliant on the ability to pattern conductiveand semiconductive graphene structures in a circuit, andthe local control of the functionalization is a cornerstonefor the full exploitation of the potential of functionalizedgraphene in electronic applications. The local hydrogenationof graphene by breaking Si–H bonds of spin-on-graphenehydrogen silsesquioxane (HSQ) upon irradiation with 30 keVelectron beam at various doses (0.5–8 mC cm−2) is an originalsolution to the spatial control of functionalization [130, 121].This method reliably delivers hydrogenated graphene whichis stable only up to moderate temperatures, since it reduces topristine graphene at temperatures in the range of 100–200 ◦C.However, the ability to pattern hydrogenated graphenestructures in a sheet of pristine graphene might enablethe birth of a new generation of flexible graphene-onlyelectronic circuits. The challenge posed by this method is theconsiderable disorder-induced states with sub-gap energieswhich make it difficult to directly measure the bandgap intransistor structures.

2.2.4. Birch reduction. All the aforementioned methodsfor hydrogenating graphene are of interest for engineeringan energy gap in the gapless energy dispersion of graphene,however they generally use specialized conditions of vacuumand low hydrogen pressure to ignite a plasma of hydrogen,reduce defects, and control locally the functionalization.Therefore, these methods are not suitable for applicationsin hydrogen storage where a low-temperature regenerationprocess is preferred [122]. Indeed, the most desirablehydrogen storage device should be light, compact and shouldhave an efficient regeneration cycle. This implies that thematerials forming this ideal device should have a highhydrogen volumetric and gravimetric density [122, 134].Graphane has a high density weight percentage of hydrogenstored as compared to the total weight of the system(i.e. high hydrogen gravimetric density) as well as a highstored hydrogen mass per unit volume of the system(i.e. high hydrogen volumetric density). However, whetherhydrogenated graphene will have a considerable impact onthe hydrogen storage industry—for instance for automotiveapplications—will be heavily determined by the regenerationprocess [122]. To date, the ability to regenerate largequantities of hydrogen storage by using low pressure andlow temperature represents one of the major challenges forhydrogenated graphene which is typically obtained by highly

8

J. Phys.: Condens. Matter 25 (2013) 423201 Topical Review

energetic processes [106, 105, 76, 124, 107, 126–129]. Therecently demonstrated hydrogenation of graphene based onthe Birch reduction process [135] represents a promisingsolution to this problem.

Birch reduction [135] is commonly used to changethe sp2 hybridization of carbon-based materials into sp3 byintroducing a covalent C–H bond and this is a method whichis of great interest for hydrogen storage devices for it iseasy to implement. In particular, a change of weight up to≈5 wt% has been demonstrated when conducting the Birchreduction of FLG with lithium in liquid ammonia [131].The functionalization of graphene obtained by this methodis stable to heating up to 500 ◦C, and the C–H bondcan decompose also by exposure to UV light or laserradiation, paving the way to whole-graphene electronicswhere conductive and semiconducting parts of a circuit aresimply drawn on a sheet of graphene by locally reducinggraphane to graphene. So far, little is known on the electricalproperties of hydrogenated graphene obtained by the Birchreduction, limiting the full exploitation of this method forelectronic applications.

While the low values of binding energy (0.7 eV) andlow chemisorption barrier (0.3 eV) characterizing atomic hy-drogen chemisorption on graphene [136, 103, 104, 137–139]make it easy to functionalize graphene with atomic hydrogen,it also causes the desorption of hydrogen at moderatetemperatures, limiting the range of potential applications. Thechemisorption with fluorine is a valid alternative when there isthe need for a graphene-based wide gap semiconductor stableto temperatures >500 ◦C [140].

2.2.5. Electrical transport versus spectroscopy measurements.Interestingly, the direct observation of a large energygap in hydrogenated graphene with photoemission spec-troscopy [107] is contrasted by the lack of a direct observationof this energy gap in electrical transport experiments,see figure 7. The opening of the energy gap due tothe hydrogenation process leads to a strong temperaturedependence of the electrical resistance in functionalizedgraphene. However, experiments consistently show that theresistance in these devices usually increases exponentiallywhen reducing the temperature [105] with a characteristicfunctional dependence intrinsic of hopping conductionmechanisms.

The discrepancy between optical spectroscopy andtransport experiments can be understood when consideringthe presence of disorder-induced states with sub-gap energies.In this case, when the Fermi level of graphane laysin the energy range corresponding to the energy gap,hopping of charge carriers between disorder-induced sub-gapenergy states dominates the electrical conduction. Thisprocess of conduction hinders the direct observation of theenergy gap in electrical transport measurements. However,optical spectroscopy is a technique mostly sensitive to thevalence-to-conduction band transitions rather than transitionsfrom the impurity states to the conduction band. The fullexploitation of the semiconducting properties of graphane fortransistor applications requires a clean energy gap with no

Figure 7. (a)–(c) The measured photoemission intensity along theAKA direction of the Brillouin zone (see inset) for (a) cleangraphene on Ir(111) and after exposing graphene to (b) 30 s and(c) 50 s to atomic hydrogen. A bandgap clearly develops uponincreasing the dose of atomic hydrogen. Reproduced withpermission from [107]. Copyright 2010 Nature. (d) The temperaturedependence of the maximum of resistivity for pristine (red circles),hydrogenated (blue squares), and annealed (green triangles)graphene measured in the multiterminal Hall bar geometry shown inthe inset. A pronounced temperature dependence of the resistivity isobserved for hydrogenated graphene, and the continuous line is a fitto the variable range hopping dependence which describes well themeasurements. Reproduced with permission from [105]. Copyright2009 The American Association for the Advancement of Science.

disorder-induced states. Obtaining a clean-gap hydrogenatedgraphene in which the energy gap is the only energydominating both the electrical and optical transport propertiesis currently an open quest.

2.3. Fluorinated graphene

Similarly to hydrogenated graphene, the chemisorption offluorine atoms onto graphene modifies the hybridization of theelectronic orbitals of graphene from sp2 to sp3. Consequently,fluorinated graphene is also a semiconducting material witha wide bandgap in the energy dispersion. When full fluorinecoverage is attained, this material is called fluorographene.Theoretical calculations of the fluorographene bandgap bythe generalized gradient approximation and density functionaltheory provide values around 3.1 eV [139, 141, 142] which aresimilar to the value for the bulk graphite fluoride [143, 144].Compared to hydrogenated graphene, fluorinated graphene ismore stable to higher temperatures (that is >400 ◦C). Thisis mainly due to a stronger binding energy between fluorineand carbon as compared to the binding energy between

9

J. Phys.: Condens. Matter 25 (2013) 423201 Topical Review

hydrogen and carbon. Therefore, fluorinated graphene is afew-atoms-thick semiconductor suitable for a different nicheof applications to those potentially targeted by hydrogenatedgraphene. This diversity in material properties and potentialapplications constitutes the true richness of chemicallyfunctionalized graphene.

The fluorination of graphene has been successfullydemonstrated with various techniques which often exploitpreviously developed methods for the fluorination ofcarbon nanotubes [145–148] or bulk graphite [149].More specifically, these methods include the reactionof graphite with fluorine gas at moderate temperatures(400–600 ◦C) [150, 112], exposure to F-based plasma[109, 151, 114], exposure to XeF2 [111, 110, 113] and thefluorination mediated by laser irradiation of a fluoropolymercovering graphene [152]. Different approaches have beenused to isolate a single layer of fluorinated graphene.More specifically, it is possible to conduct directly thefluorination of single layer pristine graphene [110] oralternatively it is possible to simply isolate a single layerof fluorinated graphene from bulk fluorinated graphite byexfoliation [153, 154]. In the latter process, the exfoliationof the bulk fluorinated graphite can take place by mechanicalcleavage [153, 154] or by dispersion and exfoliation in organicsolvents [113]. While the mechanical exfoliation is a reliablemethod to fabricate devices for fundamental research, theliquid phase exfoliation is perhaps one of the most promisingmethods for manufacturing large area fluorinated graphene atlow costs.

2.3.1. Exposure to F2 gas. The fluorination of graphite byexposure to F2 gas at moderate temperatures, in the range of400–600 ◦C, is a well-established process [149, 150] in whichcovalent C–F bonds are formed. This process of fluorinationis easy to control, for example by controlling the temperatureand the pressure of the reactant gases, leading reproduciblyto the same C/F stoichiometries [112]. Fluorinated graphiteobtained by this method is commonly used as a lubricantand in the cathodes of primary lithium batteries [155].Single- and few-layer fluorinated graphene were mechanicallyexfoliated from fluoro-graphite grains, and their electricalproperties characterized in transistor devices [153, 154]. Itwas found that the electrical resistance of fluorinated graphenevaries significantly with the temperature, as expected fora semiconducting material. However, similarly to the caseof graphane, also in fluorinated graphene the presence ofa finite disorder-induced density of states with sub-gapenergies hinders the direct observation of the true energygap (corresponding to the band-to-band transition fromvalence-to-conduction band). Indeed, in the presence ofdisorder-induced states the electrical conduction for sub-gapenergies is dominated by the signal due to hopping throughthese states. Despite the presence of disorder, fluorinatedgraphene remains a very attractive material system forapplications as well as fundamental science. For example,recent experiments demonstrated that the electronic transportproperties in fluorinated graphene can be tuned by adjustingthe fluorine coverage, so that different transport regimes

can be accessed, such as Mott variable range hopping(VRH) in two dimensions, Efros–Shklovskii VRH andnearest-neighbour hopping transport [112, 154].

2.3.2. Remote plasma of a fluorinating agent. The chemicalfunctionalization of graphene with fluorine can also canbe carried out by using a remote plasma. It was shownthat the plasma-assisted decomposition of CF4 employing aradiofrequency plasma source at room temperature can inducethe chemisorption of fluorine onto graphene [109, 151, 114].The formation of covalent C–F bonds was confirmed byinfrared spectroscopy. Furthermore, it is also possible tofluorinate graphene using the plasma-assisted decompositionof CHF3, however a comparative study of the Raman spectraof fluorinated graphene obtained from CF4 and CHF3 hasshown that the fluorination by CF4 plasma treatment inducesthe lowest magnitude of lattice defects although a highermagnitude of hole-doping to graphene was observed. Theelectrical transport properties of F-based plasma fluorinatedgraphene are also dramatically affected by the presenceof disorder-induced states with sub-gap energies which arealso at the origin of the observed colossal anisotropicmagnetoresistance [114].

2.3.3. Exposure to XeF2. An alternative method to fluorinategraphene which avoids ion bombardment consists in exposinggraphene to xenon difluoride (XeF2) at 30 ◦C [111] or at70 ◦C [110] and 350 ◦C [113] for a faster reaction. XeF2is a well known etchant for silicon and it was also usedto functionalize carbon nanotubes with fluorine [147, 148].This processing has been successfully transferred tographene where optical spectroscopy studies reported abandgap >3 eV [110]. Photoluminescence measurements offluorographene produced with this method and subsequentlydispersed in acetone have also identified an emission peak at3.8 eV as the band-to-band recombination of free electronsand holes [113]. However, this bandgap has not been directlymeasured in electrical transport experiments, since thisprocess of functionalization also introduces a considerabledensity of states with sub-gap energies responsible of hoppingconduction mechanisms which hinder the observation ofactivated transport over the true energy gap.

2.3.4. Laser irradiation of a fluoropolymer coating graphene.Finally, the laser-assisted fluorination of graphene wasdemonstrated by creating fluorine radicals upon laserirradiation of a fluoropolymer (Cytop, CTL-809 M fromAsahi Glass Co.) which coats graphene. This method offersa straightforward way to locally control the fluorinationof graphene, enabling a new concept of whole-grapheneelectronics whereby conductive and semiconductive parts of acircuit are all made from graphene but with different degreesof functionalization—i.e. different coverage of adatoms [152].

3. Patterning of graphene circuits by localfunctionalization

Engineering the electrical and optical properties of grapheneby chemical functionalization with molecules or adatoms

10

J. Phys.: Condens. Matter 25 (2013) 423201 Topical Review

widens considerably the potential range of applications forthis novel material. In particular, the possibility to controllocally the functionalization of graphene would enable awhole new concept of electronic applications whereby thecurrent electronic printed circuit boards would be replacedby whole graphene-based boards where devices are simplypatterned on a graphene matrix by functionalization. Thisidea was theoretically studied in [156–158] and tunableelectronic and magnetic properties depending on the edgegeometry (zigzag or armchair orientation) and edge adsorbedatoms have been predicted. Recent experimental advancesin the local control of fluorination have shown thatthese graphene circuit boards are a closer reality thaninitially anticipated [153, 152]. To date two complementaryapproaches to fabricate whole-graphene circuit boards withfunctionalization have been experimentally demonstrated:(1) start off from functionalized graphene sheet and reducethe functionalization in specific regions [153]; (2) start offfrom pristine graphene sheet and functionalize only specificregions [152].

3.1. Electron-assisted defluorination

In the first method, the starting material was a sheet offunctionalized graphene obtained by mechanical exfoliationfrom fluorinated graphite by exposure to F2 gas at 450 ◦C,resulting in a fluorine content of 28%. Subsequently, theirradiation of the functionalized graphene with an electronbeam with appropriate energy was used to dissociate theC–F bonds and reduce the fluorinated graphene. In this way,conductive channels as narrow as 40 nm have been drawn byelectron beam defluorination in the hosting insulating sheet offluorographene [153], see figure 8.

This innovative technique offers a simple and direct wayto change the coverage of fluorine adatoms, and it has enabledthe experimental demonstration that the energy dispersionsof fluorinated graphene has an energy gap with a valuedependent on the degree of functionalization—i.e. coverage ofadatoms [159]. More specifically, a semiconductor-to-metaltransition was experimentally observed upon changing insitu the F-coverage from 28% to <1% with electron beamirradiation. This transition is accompanied by a significantchange of the resistivity of the diluted fluorinated graphene(see figure 9(a)). At the same time, the energy gap betweenthe impurity band and the top of the valence band changes by500%, while the spreading in energy of the localized statesonly changes by 30% (see figures 9(b) and (c)).

3.2. Laser-assisted fluorination

An alternative route to control locally the covalentfunctionalization of graphene with fluorine consists in coatingthe pristine graphene with a fluoropolymer (Cytop, CTL-809M from Asahi Glass Co.). Subsequently, the CYTOP-polymeris decomposed by means of a laser (wavelength 488 nm andspot size 500 nm) and active fluorine radicals are producedboosting the formation of the fluorinated graphene. Due tothe small laser spot size, this process naturally offers a

Figure 8. (a) Schematic illustration of the device configurationunder irradiation with a beam of electrons. The fluorinated graphene(green) is reduced to graphene (grey) upon electron irradiation.(b) Illustrates that owing to the high resolution of the electron beam,ribbons of various widths (W) of pristine graphene can be directlywritten in the insulating sheet of fluorinated graphene. The graph ofpanel (c) shows the sample resistance plotted against inverse widthW for a device. The continuous line is a linear fit. Reproduced withpermission from [153]. Copyright 2011 American ChemicalSociety.

good control over the spatial resolution of the fluorinationprocess, and fluorinated circuits were indeed demonstrated ona pristine graphene sheet [152]. The possibility to tune in situthe optical and electrical properties of graphene from a fullymetallic state all the way to a wide gap semiconductor simplyby controlling the coverage of fluorine adatoms enables anew class of potential applications in flexible and wearableelectronics.

3.3. Electron-assisted hydrogenation

As discussed in the previous section, spatial control of hydro-genation in graphene has been demonstrated by electron beamirradiation of an HSQ layer deposited on graphene [130, 121].Furthermore, AFM lithography has also been used tolocally hydrogenate graphene [160]. However, as opposed topatterned fluorinated graphene, the hydrogenated grapheneproduced by patterning is not stable at above 100 ◦C andreduces to pristine graphene at temperatures in the range of100–200 ◦C.

3.4. AFM assisted thermochemical reduction of GO

Finally, nanoscale patterning of conductive reduced GOwithin an insulating GO sheet has also been reported usingan AFM tip-based thermochemical nanolithography methodto control the extent of reduction of GO [161]. In this method,reduced GO features can reliably be obtained by scanning

11

J. Phys.: Condens. Matter 25 (2013) 423201 Topical Review

Figure 9. (a) Graph of the measured zero-bias square resistance(Rsq) at room temperature after subsequent steps of electron-assisted defluorination conducted on the same sample. The insetsshow the process of electron-assisted defluorination wherefluorinated graphene (left structure with carbon atoms in grey andfluorine atoms in green) is reduced to graphene (right structure).(b) and (c) Graphs of the estimated energy gap ε1 between theimpurity states and the valence band edge and ε2 the impurity bandwidth respectively. The inset in (b) shows a sketch of the density ofstates versus energy. Reproduced from [159].

a heated AFM tip over isolated GO flakes deposited ona SiO2/Si substrate. In a related method, a metallic AFMtip was used to locally reduce GO through electrochemicalconversion, which generated micrometre-scale features withtunable conductance [162]. Similarly to hydrogenation, AFMlithography was employed to oxidize locally graphene andcreate pattern nanoscale GO [160].

4. Two-dimensional intercalated graphitic materials

Another way to functionalize FLG is by introducingmolecular and/or atomic species between the planes ofthe FLG. This intercalation process has been extensivelystudied in bulk graphite and has granted access to anoverwhelming variety of novel properties such as an intrinsicsuperconducting state as well as magnetic properties [163].To date, few experiments have been conducted on few-layer graphene systems on insulating, metallic or carbidicsubstrates [164]. These substrates initially serve as ahost material to produce graphene by means of chemicalvapour deposition (CVD) or epitaxial growth. In this casemonolayer or multilayer graphene used as a starting material

for intercalation is not free-standing but electronicallycoupled to the substrate. It is therefore traditionally calledtwo-dimensional graphitic film (2DGF) [165]. 2DGFs can beformed on many metals such as Pt, Ir, Rh, Ru, Ni, Pd, Re,silicon carbide SiC and metal carbides (for example, MoC,ZrC, NbC, TaC, TiC).

One of the first successful attempts to intercalate guestatoms between a monolayer 2DGF and Ir substrate wascarried out using Ni as intercalant [164]. More specifically,Ni atoms were deposited on top of graphene in UHVand the intercalation process was boosted by heating attemperatures higher than 1000 K. The phenomenon is similarto that of intercalation into bulk graphite (resulting ingraphite intercalation compounds, GIC) when guest atomsor molecules penetrate into the interlayer space whilegraphene planes remain unchanged and only move apart toaccommodate the intercalated molecules. This topic is stillthriving with more new molecules and/or atoms intercalatedrecently, for example hydrogen between monolayer andbilayer graphene and SiC [166], Eu between graphene andIr [167], Al between graphene and Ni [168], HF betweengraphene and Ir [169], oxygen between graphene and Ir [170],oxygen between graphene and Ru [171] and finally Csbetween graphene and Ni(111) substrate [172].

Intercalation of 2DGFs was a strategy to obtaingraphene free of electronic coupling to the substrate (quasi-free-standing graphene) before the method of mechanicalexfoliation was put forward, although in several cases theelectronic properties of pristine graphene are modified bythe presence of the intercalant specie. For example, asite-dependent periodic variation of the local effective spinpolarization in cobalt-intercalated graphene on Ir(111) wasobserved by means of spin-polarized scanning tunnellingmicroscopy [173]. Modelling based on density functionaltheory shows that the origin of this variation is a site-dependent magnetization of the graphene, where grapheneis coupled to the cobalt underneath either ferromagneticallyor antiferromagnetically. Such a surface may be utilizedin the future to inject spin currents with different spinsigns energy-selectively, by choosing appropriate adsorptionsites for spin-active adsorbates. For example, a giant (up to100 meV) spin–orbit splitting of π -states in Au-intercalatedgraphene was observed with the help of angle- andspin-resolved photoemission spectroscopy [174, 175], makingfunctionalized graphene a promising material for applicationsin spintronics.

Another intercalant of FLG which has been recentlyshown to decouple electrically graphene layers from thesubstrates and from each other is methane [176]. Uponintercalation, each graphene layer was found to behaveas quasi-free-standing high-quality monolayer graphene.Furthermore, it has been theoretically predicted that anexternal electric field applied perpendicular to the plane ofthis methane intercalated few-layer graphene should opena controllable bandgap [177], widening considerably therange of potential applications targeted by this functionalizedgraphene.

Finally, intercalated compounds are of great interest alsofor energy storage devices. For example, the intercalation of

12

J. Phys.: Condens. Matter 25 (2013) 423201 Topical Review

bilayer graphene on SiC by Li not only decouples electricallythe bilayer from the SiC substrate [178], but it also intercalatesin the bilayer leading to a well-ordered structure as in C6Li.Since lithium is traditionally widely used as an anode materialin rechargeable batteries, these findings are of potentialinterest for nanoscale Li-ion battery. Furthermore, theoreticalinvestigations show that bilayer graphene with structuraldefects and isolated C atom vacancies will store more Liions between two graphene sheets than bulk graphite [179].This defect-full FLG is also expected to have shortercharging/discharging times [178].

Apart from the use for energy storage devices, Liintercalation is expected to grant access to new statessuch as the superconducting state. First-principles densityfunctional theory calculations predict phonon-mediatedsuperconductivity of monolayer graphene with adjacentlayer of lithium [180]. Although later studies based onthe cluster expansion method and density functional theorycalculations [181] suggest that Li cannot reside on thesurface of defect-free graphene, the quest for an intrinsicsuperconducting state in intercalated few-layer graphene isstill open. Another attractive property of Li-FLG ariseswhen narrow strips of FLG (nanoribbons) are used. In thiscase the possibility of antiferromagnetic coupling betweenthe edge states of adjacent nanoribbon layers arises [182].These nanostructures with spin magnetic edge states are ofgreat interest for spintronics. Finally, a large enhancementof the charge carrier density can be achieved by Liintercalation into quasi-free bilayer graphene, as can be seenby means of density functional theory [183]. Therefore,Li intercalation could be used to improve significantlythe electrical conductivity of graphene and replace existenttransparent electrodes such as indium tin oxide. However,issues related to the air stability of Li-FLG will need to beaddressed.

The thrilling possibility to induce an intrinsic super-conducting state in atomically thin systems by doping isof great fundamental relevance [184]. It is well establishedthat graphite becomes superconducting after intercalationwith alkali metals and alkaline-earth metals, thereforeintercalation is probably one of the most promising fabricationtechniques to search for intrinsic superconductivity infew-layer graphene. For example, Ca-intercalated graphenebilayer was considered theoretically [185], and substantialsimilarities between Ca-intercalated bilayer and bulk graphitecompound CaC6 were predicted. In particular, the nearly freeelectron band in intercalated bilayer is expected to crossthe chemical potential, which is an essential ingredient forsuperconductivity in intercalated graphites. Furthermore, thecalculated properties of zone-centre phonons for intercalatedbilayer are very similar to those of bulk CaC6. Experimentally,Ca intercalation was recently successfully demonstratedin bilayer graphene grown on silicon carbide [186]. Thestructure and electronic states of this material was studiedby scanning tunnelling microscopy and angle-resolvedphotoemission spectroscopy. The free-electron-like interlayerband at the Brillouin-zone centre was observed. However,direct observation of the superconducting state has not yet

been reported. Similarly, a parabolic metallic band at theBrillouin-zone centre of Rb-intercalated bilayer graphene on6H-SiC(0001) was observed [187], making Rb-intercalationpotentially interesting for inducing a superconducting statein FLGs. Finally, potassium doped FLG is the only systemin which intrinsic superconductivity has been experimentallyobserved. In this case, the FLGs were prepared in a solutionwith K/Na alloy and 1,2-dimethoxyethane as the impregnantand subsequently a sheet containing mainly doped fourgraphene layers was found to have a critical temperature of4.5 K [188].

5. Intercalated quasi-free few-layer graphene

Unlike CVD graphene on metals or epitaxial graphene onSiC, graphene deposited on SiO2, Si3N4, glass or graphenein liquid suspension has pristine electronic spectrum and werefer to this as quasi-free single- and few-layer graphene.Various molecules and particles can be intercalated in thisquasi-free FLG, and it has also been demonstrated that it ispossible to wrap nanocrystals by graphene. More specifically,Fe2O3, Co3O4, and SnO2 nanocrystals were inserted intobilayer graphene through a mechanism analogous to colloidalcoagulation [189]. Similarly, a 3D nanostructure was achievedwith nano-sized TiO2 intercalated between graphene layersas pillars [190]. The obtained composites can be used asan anode material for Li-ion batteries with improved lithiumstorage capabilities.

Intercalated FLG is expected to grant access to a largevariety of unique properties. For example, bilayer grapheneintercalated with Br and Br2 has been predicted to becharacterized by a significant charge transfer from grapheneto Br/Br2 which can result in a significant enhancementof the electrical conductivity of graphene work [191]. Theintercalation of C, N or O atoms is expected to open a bandgapin bilayer graphene since these atoms are preferentiallyabsorbed by one of the graphene layers as opposed to the casewhen they are being placed in the middle between graphenelayers, leading to a potential difference between the outerlayers and the opening of a bandgap [192]. Furthermore,intercalation of C and N atoms is also expected to originatemagnetic properties of interest in spintronics [192]. At thesame time, the intercalation of Ni into bilayer graphene isexpected to make bilayer a nonmagnetic semiconductor withbandgap of 0.64 eV, whereas incorporation of Fe and Coshould result in ferromagnetic two-dimensional metals [193].Finally, the band structure and electron–phonon coupling ofquasi-free K-intercalated bilayer graphene calculated usingdensity functional theory also suggests that K-intercalatedbilayer graphene is a good candidate to search for intrinsicsuperconductivity [194].

The experimental realization of intercalation of molecularand atomic species into quasi-free FLG was recently reportedfor K, Rb [195], ICl and IBr [196] using the vapour-phasemethod. In particular, partial intercalation of K and Rbwas performed by means of a two-zone vapour-phasemethod at 210 and 160 ◦C, whereas the intercalation of ICland IBr requires relatively low temperatures (just 35 ◦C).

13

J. Phys.: Condens. Matter 25 (2013) 423201 Topical Review

Similarly, the intercalation of FeCl3 was demonstrated usingtemperatures of about 350 ◦C [197]. A large upshift of theG-peak in the Raman spectra is commonly observed uponintercalation, and it signifies that a large charge transfer occursfrom (to) the molecular or atomic specie to (from) graphene.Furthermore, a strong anisotropy of resistance manifestedin high c-axis (perpendicular to graphene planes) resistivityis expected [196] owing to a strong localization of theπ -electronic orbitals in the graphene layers separated by IClor IBr. The substantial reduction of the coupling betweenintercalated graphene layers makes these systems promisingfor studying excitonic condensates.

Finally, another possible route to intercalate FLGsexploits the reduction of intercalated GO [198]. Morespecifically, exfoliated GO in liquid ammonia and lithiummetal was reduced chemically and filtered through apolytetrafluoroethylene membrane filter of 0.2 m pore size.Partial intercalation of lithium ions in between graphenelayers of the resulting structure was demonstrated, making thismaterial a good candidate for electrochemical energy storageapplications.

6. Making graphene the best transparent conductor

Transparent materials able to conduct electricity are nowadaysembedded in many optoelectronic devices, such as displays,organic solar cells, etc. One of the most widely usedtransparent conductors is indium tin oxide (ITO). This is ahighly conductive material (just 10�/sq at room temperature)and highly transparent. However, ITO is a brittle mate-rial [199] which is therefore not suitable for future flexibleelectronic applications. Furthermore, organic photovoltaicdevices which implement ITO anodes are known to sufferof ion diffusion from ITO into the organic films [200, 201]which limits severely the lifetime of these devices. Therefore,there is a growing need to find materials which canreplace ITO and potentially display added functionalities fordeveloping novel optoelectronic applications.

Networks of carbon nanotubes and metallic nanostruc-tures have been at the focus of current research as potentialreplacement for ITO [202] and very recent advances in thisresearch area have demonstrated transparent electrodes basedon a metal nanotrough network with superior optoelectronicperformances to those of ITO, with a sheet resistance of2 �/sq at 90% transmission in the visible wavelengthrange [203]. Networks of conductive structures typicallyemploy opaque metallic nanostructures (with transversedimensions of ∼300 nm [203, 202]), and one of the majordrawbacks of these metallic nanostructures is hazing. Haze isan unwanted side-effect caused by light scattering [204, 202],which can be unacceptably high for displays and touch panels.Hazing can be counteracted by reducing the size of themetallic nanostructures, however lower conductivities andloss of film uniformity constitute a further limit to the sizeof these structures [204].

Graphene constitutes a very attractive alternative tometallic networks, owing to its unique mechanical flexibility,high optical transparency and ability to conduct electricity.

However, the high values of sheet resistance of graphene inits pristine form limits the possible range of applications.The chemical functionalization of graphene is usuallyaccompanied by the charge transfer to/from graphene and thiscan reduce dramatically the sheet resistance of graphene. Inparticular, the recent discovery that intercalation of FeCl3 inFLG gives an air stable compound with 8 �/sq and opticaltransmittance as high as 96% in the visible wavelength rangemakes graphene the best known transparent conductor [197].

6.1. FeCl3-intercalated few-layer graphene

Intercalated graphitic compounds have been extensivelystudied in the past decades. However, very little is knownon intercalated FLG compounds, and it is likely that theoverwhelming variety of electrical, optical and magneticproperties which can potentially be engineered in intercalatedFLG compounds will be at the focus of future fundamentaland applied research.

Since the discovery in 2012 of the best known transparentconductor based on FeCl3 intercalated FLG (FeCl3-FLG),a material whimsically called graphExeter, it has becomeclear that it is possible to conduct the intercalation of largemolecules between the planes of a few-layer graphene [197].This intercalation process with FeCl3 is typically performedin vacuum using a multi-zone furnace where anhydrous FeCl3and the FLG substrates are positioned in different zones insidea glass tube. At first, the tube is pumped down to 2×104 mbarat room temperature. Subsequently, the FLG and the powderare heated for 7.5 h at 360 ◦C and 310 ◦C, respectively. Itwas found that this intercalation process does not depend onthe substrate even when intercalating bilayer graphene. Theresulting material is a very good conductor of electricity witha sheet resistance of 8 �/sq and optical transmittance as highas 96% in the wavelength range from 400 to 700 nm.

The values of electrical conductivity and opticaltransmittance characterizing FeCl3-FLG are far better thanany other known transparent conductor, making this novelgraphene-based material a valuable replacement for thewidely used ITO in the display industry. Furthermore, therecord high charge densities attained in FeCl3-FLG as highas 9 × 1014 cm−2—together with typical charge carriermobility of 1000 cm2 V−1 s−1—gives a mean free path forthe charge carriers of almost 1 µm at room temperature. Incontrast to pristine graphene where the electrical propertiesdepend dramatically on the substrate, the large mean free pathfound in FeCl3-FLG does not depend at all on the substratesupporting this intercalated compound—for example identicalvalues of mean free path were demonstrated on SiO2/Sisubstrate as well as on microscope glass slides. Indeed, theextremely high values of charge densities present in theseintercalated compounds screen the charge defects of thesubstrate, making the electrical properties of these materialsextremely robust and not vulnerable to the quality of thesupporting substrates.

Interestingly, a comparative study of the Raman spectraand magneto-electrical measurements has shown that uponintercalation the charge carriers in each individual graphene

14

J. Phys.: Condens. Matter 25 (2013) 423201 Topical Review

Figure 10. (a) Summary chart of the square resistance versus transmittance at 550 nm for FeCl3-FLG, ITO, carbon-nanotube films anddoped graphene materials. FeCl3-FLG outperform the current limit of transparent conductors, which is indicated by the grey area.(b) Raman spectra collected at different positions on a FeCl3-5L sample, after keeping the samples for three months (top) and one year(bottom) in air. The spectra show no appreciable change, demonstrating the stability in air of FeCl3-5L. Reproduced with permissionfrom [197]. Copyright 2012 Wiley.

layer forming the FLG are decoupled. Indeed, the Ramanspectra of intercalated FLG shows a change of the 2D-bandfrom multi- to single-Lorentzian peak structure and this hasbeen reported using several kinds of intercalants such aspotassium [195], rubidium [195] and also FeCl3 [197, 205].At the same time, from the temperature dependenceof magneto-electric measurements it has been clearlydemonstrated that upon intercalation the charge carriers ineach graphene layer are still Dirac fermions [197].

The optoelectronic properties of intercalated graphenecompounds are also superior to the doped graphene. Forexample, doped graphene with AuCl3 can result in a77% decrease in the sheet resistance [206]. The highwork function and low sheet resistance which characterizesAuCl3 and FeCl3 intercalated graphene are key requirementsfor improving significantly the luminous efficiency oforganic light-emitting diodes. Air stability is anotherextremely important requirement for transparent electrodesand though AuCl3-FLG has an attractive combination of highelectrical conductivity and high optical transmittance in thevisible wavelength range, a study of the stability of thisfunctionalization to air as a function of time and/or humidity isstill an open question. It is also known that many intercalatedcompounds of graphite are air-unstable materials [163]. Incontrast, the intercalation with FeCl3 is gives air stablecompounds [197]. This was demonstrated experimentally bymeasuring an unchanged Raman spectra measured on 10different locations in a 5L-intercalated device after keepingthe samples for three months and one year in air, see figure 10.

6.2. Engineering the optoelectronic properties of grapheneby functionalization with quantum dots

The gapless linear energy dispersion of graphene allows it toabsorb ∼2.3% of the impinging light nearly independently

of the wavelength from UV to THz [4–6], with a slightincrease in UV absorption attributed to π? inter-bandelectronic transitions [207]. The unique absorption offar-infrared frequencies is at the core of the recentlydemonstrated room-temperature THz detectors [208] and thistruly is another uncommon property in most semiconductordevices [209]. The low optical absorption makes graphenealso an ideal material for use as flexible transparent electrodeas demonstrated in photodetectors [210], solar cells [41],and LEDs [211]. Furthermore, high-speed phototransistorsbased on gated FLG with speeds up to 500-GHz have alsobeen demonstrated [212] as a result of the high chargecarrier mobilities [213] in FLGs. While in some casesadvantageous, the lack of a well-defined bandgap in pristinegraphene hampers its use in frequency-specific detectionwhere a bandgap is required. At the same time, low lightabsorption can constitute a limit in applications where a highphoto-gain is required. Functionalized graphene can offerunique solutions to the aforementioned limitations extendingthe domain of potential optoelectronic applications targetedby graphene.

The functionalization of graphene with colloidal quantumdots is a promising novel way to engineer its properties. Forexample, the deposition of colloidal quantum dots (QD) ontographene has recently been shown to significantly increase thefrequency-dependent absorption of graphene. Furthermore,thanks to recent advances in low-dimensional semiconductortechnologies, colloidal solutions of quantum dots can bemanufactured from a variety of semiconductors (e.g. PbS,ZnO, CdSe, PbSe) by chemical synthesis. In quantum dots aneffective bandgap can also be tuned for specific applicationssimply by controlling the size of the dots [214], and theirelectrical transport properties can be further engineered byfunctionalizing the surface of the dots. Solution-processed

15

J. Phys.: Condens. Matter 25 (2013) 423201 Topical Review

QDs have been used as the photoactive layers in non-graphenedevices for light emission [215], photovoltaics [216] andphotodetection [217–219], and have recently appeared inconsumer products such as ultra-bright television screens.

The unique combination of the properties found ingraphene and QDs has led to IR [220, 221] and UVgraphene photodetectors [222] through functionalization withPbS and ZnO QDs respectively. These novel graphene-basedphotodetectors consist of a thin layer of colloidal QDs spin-coated onto a pristine graphene transistor. The photogeneratedcharges in the QDs are easily transferred to the grapheneand extracted from the drain electrode. These QD/graphenephotodetectors may also be tuned by field-effect leadingto the multiple recirculation of charge carriers for a singleabsorbed photon and yielding frequency-dependent gains ofup to ∼107 A W−1 for PbS QDs [221].

The application of a reverse bias back-gate pulsecauses the charge carriers in the graphene to recombine asexcitons in the QDs—effectively resetting the photodetector.To date most of these early examples of graphene–QDhybrid devices make use of QDs for detection of UV orIR-THz frequencies. Such devices show desirable effectssuch as charge carrier trapping in the QDs upon absorptionof photons [223]—introducing an intrinsic gain mechanismwhich may be controlled by source–drain bias. A secondintrinsic gain mechanism may be due to multiple exciton-generation in semiconductor QDs as reported by Sargentet al [224]. Graphene–QD hybrid structures have a greatpotential for future applications and their exploitation is stillin its infancy. For example, heterostructures obtained byseveral alternating layers of graphene and CdS QDs haverecently been used for photovoltaic applications with a recordhigh incident photon-to-charge-carrier conversion efficiency(16%) as compared to previously reported carbon/QD solarcells [225].

Since the functionalization of graphene with QDs iseasily scalable, fast and low-cost, it is a very promisingmethod for device production on an industrial scale. Thevariety of possible applications is also helped by the stunningproperties of both graphene and QDs. For example, solarpowered flexible smart windows might be realized throughgraphene–QD heterostructures able to harvest electricity fromthe UV light while remaining perfectly transparent to thevisible wavelength.

7. Future prospects

Intercalated FLG truly are two-dimensional systems justa few atoms thick, in which the physical propertiescan be engineered by chemical functionalization to fit aspecific requirement for both fundamental science and futureapplications. For example, the superconducting state is themanifestation of the electron interactions in a materialand it is expected to be suppressed in two-dimensionalsystems by quantum fluctuations at zero temperature. Todate, experimental attempts to study the superconductingstate in the extreme two-dimensional limit have beenrestricted to cases where the superconducting order parameter

behaves as a 2D wavefunction, but the underlying electronsare still three-dimensional (such as the recently observedsuperconducting state at the interface between oxides [226]).This limitation was largely due to the inability to isolatestable two-dimensional atomic systems. FLG is probably theideal material for investigating the superconducting electronicordering in the two-dimensional limit. The superconductingstate in graphene is expected to occur at high chargecarrier concentrations (>1015 electrons cm−2 [227])—notachievable with ordinary oxide gate dielectrics. Ionic liquidgating would in principle offer the possibility to inducethe required charge carrier concentration necessary forattaining intrinsic superconductivity [228], however theseliquids freeze at cryogenic temperatures posing serioustechnical challenges. Chemical functionalization is a validand yet largely unexplored way to explore the intrinsicsuperconducting state in graphene and study the evolutionof this fundamental phase transition when changing thedimensionality of the system from 2D to 3D simply bychanging the number of atomic layers composing the FLG.Adding superconductivity to the list of the unique propertiesof graphene material would stimulate the development ofnovel concepts in superconducting circuits where localelectric fields can dynamically control the superconductingstate and lead to unprecedented functionalities.

Similarly to the superconducting phase transition,magnetic ordering is a property characteristic of bulkmaterials which is expected to be suppressed by quantumfluctuations at zero temperature in systems with reduceddimensionality. Graphene would offer the unique opportunityto study the interplay between these collective quantumphases in the 2D limit. Magnetic ordering could be inducedby chemical functionalizing of graphene with magneticmolecules such as FeCl3 [197]. Making graphene magneticwould enable the development of novel flexible andtransparent spintronic devices which cannot be fabricatedusing the conventional opaque and rigid magnetic materials.

The ability to dial up a specific physical property ingraphene-based materials simply by selecting the intercalatingmolecule or chemical specie which functionalizes grapheneextends dramatically the horizons of future electronicapplications targeted by graphene. For example, the abilityto engineer magnetic properties in intercalated FLG mightenable a new class of graphene-based flexible and transparentmemories. Intercalated compounds might also enable thedevelopment of graphene-based flexible and transparentbatteries, as well as the development of all-graphenelight-emitting devices.

Acknowledgments

SR and MFC acknowledge financial support from EPSRC(Grant nos EP/G036101/1 and EP/J000396/1).

References

[1] Novoselov K S, Geim A K, Morozov S V, Jiang D, Zhang Y,Dubonos S V, Grigorieva I V and Firsov A A 2004Electric field effect in atomically thin carbon films Science306 666

16

J. Phys.: Condens. Matter 25 (2013) 423201 Topical Review

[2] Zhang Y, Tan Y W, Stormer H L and Kim P 2005Experimental observation of the quantum Hall effect andBerry’s phase in graphene Nature 438 201

[3] Geim A K 2009 Graphene: status and prospects Science324 1530

[4] Dawlaty J M 2008 Measurement of the optical absorptionspectra of epitaxial graphene from terahertz to visibleAppl. Phys. Lett. 93 131905

[5] Wright A R, Cao J C and Zhang C 2009 Enhanced opticalconductivity of bilayer graphene nanoribbons in theterahertz regime Phys. Rev. Lett. 103 207401

[6] Nair R R, Blake P, Grigorenko A N, Novoselov K S,Booth T J, Stauber T, Peres N M R and Geim A K 2008Fine structure constant defines visual transparency ofgraphene Science 320 1308

[7] Castro Neto A H, Guinea F, Peres N M R, Novoselov K Sand Geim A K 2009 The electronic properties of grapheneRev. Mod. Phys. 81 109–62

[8] Craciun M F, Russo S, Yamamoto M and Tarucha S 2011Gate tunability of electronic properties in graphenesystems Nano Today 6 42–60

[9] Zhang Y, Tang T T, Girit C, Hao Z, Martin M C, Zettl A,Crommie M F, Shen Y R and Wang F 2009 Directobservation of a widely tunable bandgap in bilayergraphene Nature 459 820–3

[10] Oostinga J B, Heersche H B, Liu X, Morpurgo A F andVandersypen L M K 2008 Gate-induced insulating state inbilayer graphene devices Nature Mater. 7 151–7

[11] Niyogi S et al 2010 Spectroscopy of covalentlyfunctionalized graphene Nano Lett. 10 4061–6

[12] Khodkov T, Withers F, Hudson D C, Craciun M F andRusso S 2012 Electrical transport in suspended and doublegated trilayer graphene Appl. Phys. Lett. 100 013114

[13] Zhu Y, Murali S, Cai W, Li X, Suk J W, Potts J R andRuoff R S 2010 Graphene and graphene oxide: synthesis,properties, and applications Adv. Mater. 22 3906–24

[14] Georgakilas V, Otyepka M, Bourlinos A B, Chandra V,Kim N, Kemp K C, Hobza P, Zboril R and Kim K S 2012Functionalization of graphene: covalent and non-covalentapproaches, derivatives and applications Chem. Rev.112 6156–214

[15] Eda G and Chhowalla M 2010 Chemically derived grapheneoxide: towards large area thin-film electronics andoptoelectronics Adv. Mater. 22 2392–415

[16] Dikin D A, Stankovich S, Zimney E J, Piner R D,Dommett G H B, Evmenenko G, Nguyen S T andRuoff R S 2007 Preparation and characterization ofgraphene oxide paper Nature 448 457–60

[17] Park S and Ruoff R S 2009 Chemical methods for theproduction of graphenes Nature Nanotechnol. 4 217–24

[18] Dreyer D R, Park S, Bielawski C W and Ruoff R S 2010 Thechemistry of graphene oxide Chem. Soc. Rev. 39 228–40

[19] Cai W et al 2008 Synthesis and solid-state NMR structuralcharacterization of 13C-labeled graphite oxide Science321 1815–7

[20] Wang G, Yang J, Park J, Gou X, Wang B, Liu H and Yao J2008 Facile synthesis and characterization of graphenenanosheets J. Phys. Chem. C 112 8192–5

[21] Brodie B C 1859 On the atomic weight of graphite Phil.Trans. R. Soc. 149 249

[22] Staudenmaier L 1898 Verfahren zur Darstellung derGraphitsaure Ber. Deut. Chem. Ges. 31 1481

[23] Hummers W S and Offeman R E 1958 Preparation ofgraphitic oxide J. Am. Chem. Soc. 80 1339

[24] Gokus T, Nair R R, Bonetti A, Bohmler M, Lombardo A,Novoselov K S, Geim A K, Ferrari A C and Hartschuh A2009 Making graphene luminescent by oxygen plasmatreatment ACS Nano 3 3963–8

[25] Baraket M, Walton S G, Lock E H, Robinson J T andPerkins F K 2010 The functionalization of graphene usingelectron–beam generated plasmas Appl. Phys. Lett.96 231501

[26] Nourbakhsh A, Cantoro M, Vosch T, Pourtois G, Clemente F,van der Veen M H, Hofkens J, Heyns M M, De Gendt Sand Sels B F 2010 Bandgap opening in oxygen plasma-treated graphene Nanotechnology 21 435203

[27] Nourbakhsh A, Cantoro M, Klekachev A V, Pourtois G,Vosch T, Hofkens J, van der Veen M H, Heyns M M,De Gendt S and Sels B F 2011 Single layer versus bilayergraphene: a comparative study of the effects of oxygenplasma treatment on their electronic and optical propertiesJ. Phys. Chem. C 115 16619–24

[28] Liu l, Xie D, Wu M, Yang X, Xu Z, Wang W, Bai X andWang E 2012 Controlled oxidative functionalization ofmonolayer graphene by water–vapor plasma etchingCarbon 50 3039–44

[29] McEvoy N, Nolan H, Kumar N A, Hallam T andDuesberg G S 2013 Functionalisation of graphene surfaceswith downstream plasma treatments Carbon 54 283–90

[30] Becerril H A, Jie M, Zunfeng L, Stoltenberg R M, Bao Z andChen Y 2008 Evaluation of solution-processed reducedgraphene oxide films as transparent conductors ACS Nano2 463–70

[31] Stankovich S, Dikin D A, Piner R D, Kohlhaas K A,Kleinhammes A, Jia Y, Wu Y, Nguyen S T and Ruoff R S2007 Synthesis of graphene-based nanosheets viachemical reduction of exfoliated graphite oxide Carbon45 1558–65

[32] Tung V C, Allen M J, Yang Y and Kaner R B 2009High-throughput solution processing of large-scalegraphene Nature Nanotechnol. 4 25–9

[33] Schniepp H C, Li J L, McAllister M J, Sai H,Herrera-Alonso M, Adamson D H, Prud’homme R K,Car R, Saville D and Aksay I A 2006 Functionalizedsingle graphene sheets derived from splitting graphiteoxide J. Phys. Chem. B 110 8535–9

[34] Salas E C, Sun Z, Luttge A and Tour J M 2010 Reduction ofgraphene oxide via bacterial respiration ACS Nano4 4852–6

[35] Zhu Y, Stoller M D, Cai W, Velamakanni A, Piner R D,Chen D and Ruoff R S 2010 Exfoliation of graphite oxidein propylene carbonate and thermal reduction of theresulting graphene oxide platelets ACS Nano 4 1227–33

[36] Moon I K, Lee J, Ruoff R S and Lee H 2010 Reducedgraphene oxide by chemical graphitization NatureCommun. 1 73

[37] Wang Z, Zhou X, Zhang J, Boey F and Zhang H 2009 Directelectrochemical reduction of single-layer graphene oxideand subsequent functionalization with glucose oxidaseJ. Phys. Chem. C 113 14071–5

[38] Dubin S, Gilje S, Wang K, Tung V C, Cha K, Hall A S,Farrar J, Varshneya R, Yang Y and Kaner R B 2010A one-step, solvothermal reduction method for producingreduced graphene oxide dispersions in organic solventsACS Nano 4 3845–52

[39] Gao W, Alemany L B, Ci L and Ajayan P M 2009 Newinsights into the structure and reduction of graphite oxideNature Chem. 1 403–8

[40] Mattevi C, Eda G, Agnoli S, Miller S, Mkhoyan K A,Celik O, Mastrogiovanni D, Granozzi G, Garfunkel E andChhowalla M 2009 Evolution of electrical, chemical, andstructural properties of transparent and conductingchemically derived graphene thin films Adv. Funct. Mater.19 2577–83

[41] Wang X, Zhi L and Mullen K 2008 Transparent, conductivegraphene electrodes for dye-sensitized solar cells NanoLett. 8 323–7

17

J. Phys.: Condens. Matter 25 (2013) 423201 Topical Review

[42] Xu Z, Bando Y, Liu L, Wang W, Bai X and Golberg D 2011Electrical conductivity, chemistry, and bondingalternations under graphene oxide to graphene transitionas revealed by in situ TEM ACS Nano 5 4401–6

[43] Rani J R, Lim J, Oh J, Kim J W, Shin H S, Kim J H, Lee Sand Jun S C 2012 Epoxy to carbonyl group conversion ingraphene oxide thin films: effect on structural andluminescent characteristics J. Phys. Chem. C 116 19010–7

[44] Chhowalla M, Lee S W, Mattevi C and Sankaran R M 2012Plasma-assisted reduction of graphene oxide at lowtemperature and atmospheric pressure for flexibleconductor applications J. Phys. Chem. Lett. 3 772–7

[45] Jung I, Dikin D A, Piner R D and Ruoff R S 2008 Tunableelectrical conductivity of individual graphene oxide sheetsreduced at ‘low’ temperatures Nano Lett. 8 4283–7

[46] Boukhvalov D W and Katsnelson M I 2008 Modeling ofgraphite oxide J. Am. Chem. Soc. 130 10697–701

[47] Lahaye R J W E, Jeong H K, Park C Y and Lee Y H 2009Density functional theory study of graphite oxide fordifferent oxidation levels Phys. Rev. B 79 125435

[48] Yan J A, Xian L and Chou M Y 2009 Structural andelectronic properties of oxidized graphene Phys. Rev. Lett.103 086802

[49] Pandey D, Reifenberger R and Piner R 2008 Scanning probemicroscopy study of exfoliated oxidized graphene sheetsSurf. Sci. 602 1607–13

[50] Gomez-Navarro C, Meyer J C, Sundaram R S, Chuvilin A,Kurasch S, Burghard M, Kern K and Kaiser U 2010Atomic structure of reduced graphene oxide Nano Lett.10 1144–8

[51] Li D, Muller M B, Gilje S, Kaner R B and Wallace G G 2008Processable aqueous dispersions of graphene nanosheetsNature Nanotechnol. 3 101–5

[52] Compton O C and Nguyen S T 2010 Graphene oxide, highlyreduced graphene oxide, and graphene: versatile buildingblocks for carbon-based materials Small 6 711–23

[53] Shin K Y, Hong J Y and Jang J 2011 Micropatterning ofgraphene sheets by inkjet printing and its widebanddipole-antenna application Adv. Mater. 23 2113–8

[54] Watcharotone S et al 2007 Graphene–silica composite thinfilms as transparent conductors Nano Lett. 7 1888–92

[55] Gilje S, Han S, Wang M, Wang K L and Kaner R B 2007A chemical route to graphene for device applications NanoLett. 7 3394–8

[56] Eda G, Lin Y Y, Miller S, Chen C W, Su W F andChhowalla M 2008 Transparent and conducting electrodesfor organic electronics from reduced graphene oxide Appl.Phys. Lett. 92 233305

[57] Wu J, Becerril H A, Bao Z, Liu Z, Chen Y and Peumans P2008 Organic solar cells with solution-processed graphenetransparent electrodes Appl. Phys. Lett. 92 263302

[58] Liu Q, Liu Z, Zhang X, Yang L, Zhang N, Pan G, Yin S,Chen Y and Wei J 2009 Polymer photovoltaic cells basedon solution-processable graphene and P3HT Adv. Funct.Mater. 19 894–904

[59] Wu J, Agrawal M, Becerril H A, Bao Z, Liu Z, Chen Y andPeumans P 2010 Organic light-emitting diodes onsolution-processed graphene transparent electrodes ACSNano 4 43–8

[60] Eda G, Fanchini G and Chhowalla M 2008 Large-areaultrathin films of reduced graphene oxide as a transparentand flexible electronic material Nature Nanotechnol.3 270–4

[61] Gomez-Navarro C, Weitz R T, Bittner A M, Scolari M,Mews A, Burghard M and Kern K 2007 Electronictransport properties of individual chemically reducedgraphene oxide sheets Nano Lett. 7 3499–503

[62] Gengler R Y N, Veligura A, Enotiadis A, Diamanti E K,Gournis D, Jozsa C, van Wees B J and Rudolf P 2010

Large-yield preparation of high-electronic-qualitygraphene by a Langmuir–Schaefer approach Small 6 35–9

[63] Wang S, Ang P K, Wang Z, Tang A L L, Thong J T L andLoh K P 2010 High mobility, printable, and solution-processed graphene electronics Nano Lett. 10 92–8

[64] Eda G, Mattevi C, Yamaguchi H, Kim H and Chhowalla M2009 Insulator to semimetal transition in graphene oxideJ. Phys. Chem. C 113 15768–71

[65] Wu X, Sprinkle M, Li X, Ming F, Berger C and de Heer W A2008 Epitaxial-graphene/graphene-oxide junction:an essential step towards epitaxial graphene electronicsPhys. Rev. Lett. 101 026801

[66] Luo Z, Lu Y, Somers L A and Johnson A T C 2009 Highyield preparation of macroscopic graphene oxidemembranes J. Am. Chem. Soc. 131 898–9

[67] Zhuge F, Hu B, He C, Zhou X, Liu Z and Li R W 2011Mechanism of nonvolatile resistive switching in grapheneoxide thin films Carbon 49 3796–802

[68] Jeong H Y et al 2010 Graphene oxide thin films for flexiblenonvolatile memory applications Nano Lett. 10 4381–6

[69] Panin G N, Kapitanova O O, Lee S W, Baranov A N andKang T W 2011 Resistive switching in Al/grapheneoxide/Al structure Japan. J. Appl. Phys. 50 070110

[70] He C L et al 2009 Nonvolatile resistive switching ingraphene oxide thin films Appl. Phys. Lett. 95 232101

[71] Hong S K, Kim J E, Kim S O, Choi S Y and Cho B J 2010Flexible resistive switching memory device based ongraphene oxide IEEE Electron Device Lett. 31 1005–7

[72] Hong S K, Kim J E, Kim S O and Cho B J 2011 Analysis onswitching mechanism of graphene oxide resistive memorydevice J. Appl. Phys. 110 044506

[73] Wang L H, Yang W, Sun Q Q, Zhou P, Lu H L, Ding S J andZhang D W 2012 The mechanism of the asymmetric SETand RESET speed of graphene oxide based flexibleresistive switching memories Appl. Phys. Lett. 100 063509

[74] Loh K P, Bao Q, Eda G and Chhowalla M 2010 Grapheneoxide as a chemically tunable platform for opticalapplications Nature Chem. 2 1015–24

[75] Yamaguchi H et al 2012 Tunable photoluminescence fromgraphene oxide Angew. Chem. Int. Edn 51 6662–6

[76] Luo Z, Vora P M, Mele E J, Johnson A T C and Kikkawa J M2009 Photoluminescence and band gap modulation ingraphene oxide Appl. Phys. Lett. 94 111909

[77] Sun X, Liu Z, Welsher K, Robinson J T, Goodwin A, Zaric Sand Dai H 2008 Nano-graphene oxide for cellular imagingand drug delivery Nano Res. 1 203–12

[78] Cuong T V, Pham V H, Tran Q T, Hahn S H, Chung J S,Shin E W and Kim E J 2010 Photoluminescence andRaman studies of graphene thin films prepared byreduction of graphene oxide Mater. Lett. 64 399–401

[79] Shukla S and Saxena S 2011 Spectroscopic investigation ofconfinement effects on optical properties of grapheneoxide Appl. Phys. Lett. 98 073104

[80] Eda G, Lin Y Y, Mattevi C, Yamaguchi H, Chen H A,Chen I S, Chen C W and Chhowalla M 2010 Bluephotoluminescence from chemically derived grapheneoxide Adv. Mater. 22 505–9

[81] Subrahmanyam K S, Kumar P, Nag A and Rao C N R 2010Blue light emitting graphene-based materials and their usein generating white light Solid State Commun. 150 1774–7

[82] Xin G, Meng Y, Ma Y, Ho D, Kim N, Cho S M and Chae H2012 Tunable photoluminescence of graphene oxide fromnear-ultraviolet to blue Mater. Lett. 74 71–3

[83] Galande C, Mohite A D, Naumov A V, Gao W, Ci L,Ajayan A, Gao H, Srivastava A, Weisman R B andAjayan P M 2011 Quasi-molecular fluorescence fromgraphene oxide Sci. Rep. 1 85

18

J. Phys.: Condens. Matter 25 (2013) 423201 Topical Review

[84] Shang J, Ma L, Li J, Ai W, Yu T and Gurzadyan G G 2012The origin of fluorescence from graphene oxide Sci. Rep.2 792

[85] Liu Z, Robinson J T, Sun X and Dai H 2008 PEGylatednanographene oxide for delivery of water-insoluble cancerdrugs J. Am. Chem. Soc. 130 10876–7

[86] Yang X, Wang Y, Huang X, Ma Y, Huang Y, Yang R,Duan H and Chen Y 2011 Multi-functionalized grapheneoxide based anticancer drug-carrier with dual-targetingfunction and pH-sensitivity J. Mater. Chem. 21 3448–54

[87] Wang Y, Li Z, Hu D, Lin C T, Li J and Lin Y 2010 Aptamer/graphene oxide nanocomplex for in situ molecular probingin living cells J. Am. Chem. Soc. 132 9274–6

[88] Hu W, Peng C, Luo W, Lv M, Li X, Li D, Huang Q andFan C 2010 Graphene-based antibacterial paper ACS Nano4 4317–23

[89] Wang H, Cui L F, Yang Y, Sanchez C H, Robinson J T,Liang Y, Cui Y and Dai H 2010 Mn3O4 graphene hybridas a high-capacity anode material for lithium ion batteriesJ. Am. Chem. Soc. 132 13978–80

[90] Stankovich S, Dikin D A, Dommett G H B, Kohlhaas K M,Zimney E J, Stach E A, Piner R D, Nguyen S T andRuoff R S 2006 Graphene-based composite materialsNature 442 282–6

[91] Bai H, Li C and Shi G 2011 Functional composite materialsbased on chemically converted graphene Adv. Mater.23 1089–115

[92] Chen S, Muller M B, Gilmore K J, Wallace G G and Li D2008 Mechanically strong, electrically conductive, andbiocompatible graphene paper Adv. Mater. 20 3557–61

[93] Wu Z S, Ren W, Wen L, Gao L, Zhao J, Chen Z, Zhou G,Li F and Cheng H M 2010 Graphene anchored withCo3O4 nanoparticles as anode of lithium ion batteries withenhanced reversible capacity and cyclic performance ACSNano 4 3187–94

[94] Chen S, Zhu J, Wu X, Han Q and Wang X 2010 Grapheneoxide MnO2 nanocomposites for supercapacitors ACSNano 4 2822–30

[95] Zhu Y et al 2011 Carbon-based supercapacitors produced byactivation of graphene Science 332 1537–41

[96] Huang Y, Liang J and Chen Y 2012 An overview of theapplications of graphene-based materials insupercapacitors Small 8 1805–34

[97] Seger B and Kamat P V 2009 Electrocatalytically activegraphene-platinum nanocomposites. Role of 2-d carbonsupport in PEM fuel cells J. Phys. Chem. C 113 7990–5

[98] Hou J, Shao Y, Ellis M W, Moore R B and Yi B 2011Graphene-based electrochemical energy conversion andstorage: fuel cells, supercapacitors and lithium ionbatteries Phys. Chem. Chem. Phys. 13 15384–402

[99] Zhao B, Liu P, Jiang Y, Pan D, Tao H, Song J, Fang T andXu W 2012 Supercapacitor performances of thermallyreduced graphene oxide J. Power Sources 198 423–7

[100] Jeloaica L and Sidis V 1999 DFT investigation of theadsorption of atomic hydrogen on a cluster-model graphitesurface Chem. Phys. Lett. 300 157–62

[101] Sha X and Jackson B 2002 First-principles study of thestructural and energetic properties of h atoms on a graphite(0 0 0 1) surface Surf. Sci. 496 318–30

[102] Sha X, Jackson B and Lemoine D 2002 Quantum studies ofEley–Rideal reactions between H atoms on a graphitesurface J. Chem. Phys. 116 7158–69

[103] Miura Y, Kasai H, Dino W A, Nakanishi H and Sugimoto T2003 Effective pathway for hydrogen atom adsorption ongraphene J. Phys. Soc. Japan 72 995–7

[104] Sofo J O, Chaudhari A S and Barber G D 2007 Graphane: atwo-dimensional hydrocarbon Phys. Rev. B 75 153401

[105] Elias D C et al 2009 Control of graphene’s properties byreversible hydrogenation: evidence for graphane Science323 610–3

[106] Han Y, Maultzsch J, Heinz T F, Kim P, Steigerwald M L andBrus L E 2008 Reversible basal plane hydrogenation ofgraphene Nano Lett. 8 4597–602

[107] Balog R et al 2010 Bandgap opening in graphene induced bypatterned hydrogen adsorption Nature Mater. 9 315–9

[108] Worsley K A, Ramesh P, Mandal S K, Niyogi S, Itkis M Eand Haddon R C 2007 Soluble graphene derived fromgraphite fluoride Chem. Phys. Lett. 445 51–6

[109] Bon S B, Valentini L, Verdejo R, Garcia Fierro J L, Peponi L,Lopez-Manchado M A and Kenny J M 2009 Plasmafluorination of chemically derived graphene sheets andsubsequent modification with butylamine Chem. Mater.21 3433–8

[110] Nair R R et al 2010 Fluorographene: a two-dimensionalcounterpart of teflon Small 6 2877–84

[111] Robinson J T et al 2010 Properties of fluorinated graphenefilms Nano Lett. 10 3001–5

[112] Cheng S H, Zou K, Okino F, Gutierrez H R, Gupta A,Shen N, Eklund P C, Sofo J O and Zhu J 2010 Reversiblefluorination of graphene: evidence of a two-dimensionalwide bandgap semiconductor Phys. Rev. B 81 205435

[113] Jeon K J et al 2011 Fluorographene: a wide bandgapsemiconductor with ultraviolet luminescence ACS Nano5 1042–6

[114] Hong X, Cheng S H, Herding C and Zhu J 2011 Colossalnegative magnetoresistance in dilute fluorinated graphenePhys. Rev. B 83 085410

[115] Chandrachud P, Pujari B S, Haldar S, Sanyal B andKanhere D G 2010 A systematic study of electronicstructure from graphene to graphane J. Phys.: Condens.Matter 22 465502

[116] Gao H, Zhao J, Ding F and Lu J 2011 Band gap tuning ofhydrogenated graphene: H coverage and configurationdependence J. Phys. Chem. C 115 3236

[117] Castro Neto A H and Guinea F 2009 Impurity-inducedspin–orbit coupling in graphene Phys. Rev. Lett.103 026804

[118] Schmidt M J and Loss D 2010 Edge states and enhancedspin–orbit interaction at graphene/graphane interfacesPhys. Rev. B 81 165439

[119] Conan W, Hu J, Alicea J, Franz M and Wu R 2011Engineering a robust quantum spin Hall state in graphenevia adatom deposition Phys. Rev. X 1 021001

[120] Zhou J, Liang Q F and Dong J M 2010 Enhanced spin–orbitcoupling in hydrogenated and fluorinated grapheneCarbon 48 1405

[121] Balakrishnan J, Koon G K W, Jaiswal M, Castro Neto A Hand Ozyilmaz B 2013 Colossal enhancement of spin–orbitcoupling in weakly hydrogenated graphene Nature Phys.9 284–7

[122] Graetz J 2009 New approaches to hydrogen storage Chem.Soc. Rev. 38 73–82

[123] Luo Zh, Yu T, Kim K, Ni Zh, You Y, Lim S, Shen Z,Wang Sh and Lin J 2009 Thickness-dependent reversiblehydrogenation of graphene layers ACS Nano 3 1781

[124] Luo Zh, Shang J, Lim S, Li D, Xiong A, Shen Z, Lin J andYu T 2010 Modulating the electronic structures ofgraphene by controllable hydrogenation Appl. Phys. Lett.97 233111

[125] Luo Z Q, Yu T, Ni Z H, Lim S H, Hu H L, Shang J Z, Liu L,Shen Z X and Lin J Y 2011 Electronic structures andstructural evolution of hydrogenated graphene probed byRaman spectroscopy J. Phys. Chem. C 115 1422–7

19

J. Phys.: Condens. Matter 25 (2013) 423201 Topical Review

[126] Ng M L, Balog R, Hornekær L, Preobrajenski A B,Vinogradov N A, Martensson N and Schulte K 2010Controlling hydrogenation of graphene on transitionmetals J. Phys. Chem. C 114 18559

[127] Haberer D et al 2010 Tunable band gap in hydrogenatedquasi-free-standing graphene Nano Lett. 10 3360

[128] Haberer D et al 2011 Direct observation of a dispersionlessimpurity band in hydrogenated graphene Phys. Rev. B83 165433

[129] Haberer D et al 2011 Electronic properties of hydrogenatedquasi-free-standing graphene Phys. Status Solidi b248 2639

[130] Ryu S, Han M Y, Maultzsch J, Heinz T F, Kim P,Steigerwald M L and Brus L E 2008 Reversible basalplane hydrogenation of graphene Nano Lett. 8 4597–602

[131] Subrahmanyama K S, Kumara P, Maitraa U, Govindaraja A,Hembramb K P S S, Waghmareb U V and Raoa C N R2011 Chemical storage of hydrogen in few-layer grapheneProc. Natl Acad. Sci. USA 108 2674–7

[132] Roman T, Dino W A, Nakanishi H, Kasai H, Sugimoto T andTange K 2006 Realizing a carbon-based hydrogen storagematerial Japan. J. Appl. Phys. 45 1765–7

[133] Roman T, Dino W A, Nakanishi H, Kasai H, Sugimoto T andTange K 2007 Hydrogen pairing on graphene Carbon45 218–20

[134] Tozzini V and Pellegrini V 2013 Prospects for hydrogenstorage in graphene Phys. Chem. Chem. Phys. 15 80

[135] Birch A J 1944 Reduction by dissolving metals. Part I.J. Chem. Soc. 1944 430–6

[136] Gao H, Zhou J, Lu M, Fa W and Chen Y 2010 First-principles study of the IVA group atoms adsorption ongraphene J. Appl. Phys. 107 114311

[137] Boukhvalov D W, Katsnelson M I and Lichtenstein A I 2008Hydrogen on graphene: electronic structure, total energy,structural distortions and magnetism from first-principlescalculations Phys. Rev. B 77 035427

[138] Boukhvalov D W and Katsnelson M I 2009 Chemicalfunctionalization of graphene J. Phys.: Condens. Matter21 344205

[139] Leenaerts O, Peelaers H, Hernandez-Nieves A D, Partoens Band Peeters F M 2010 First-principles investigation ofgraphene fluoride and graphane Phys. Rev. B 82 195436

[140] Nomura K, Ohta H, Takagi A, Kamiya T, Hirano M andHosono H 2004 Room-temperature fabrication oftransparent flexible thin-film transistors using amorphousoxide semiconductors Nature 432 488–92

[141] Zboril R et al 2010 Graphene fluoride: a stable stoichiometricgraphene derivative and its chemical conversion tographene Small 6 2885

[142] Samarakoon D K, Chen Z F, Nicolas C and Wang X Q 2011Structural and electronic properties of fluorographeneSmall 7 965

[143] Takagi Y and Kusakabe K 2002 Transition from direct bandgap to indirect band gap in fluorinated carbon Phys. Rev. B65 121103(R)

[144] Charlier J C, Gonze X and Michenaud J P 1993 First-principles study of graphite monofluoride (CF)n Phys.Rev. B 47 16162

[145] Khare B N, Wilhite P and Meyyappan M 2004The fluorination of single wall carbon nanotubes usingmicrowave plasma Nanotechnology 15 1650–4

[146] Felten A, Bittencourt C, Pireaux J J, Van Lier G andCharlier J C 2005 Radio-frequency plasmafunctionalization of carbon nanotubes surface O2,NH3,and CF4 treatments J. Appl. Phys. 98 074308

[147] Unger E, Liebau M, Duesberg G S, Graham A P, Kreupl F,Seidel R and Hoenlein W 2004 Fluorination of carbon

nanotubes with xenon difluoride Chem. Phys. Lett.399 280–3

[148] Dettlaff-Weglikowska U, Skakalova V, Meyer J, Cech J,Mueller B G and Roth S 2007 Effect of fluorination onelectrical properties of single walled carbon nanotubes andc-60 peapods in networks Curr. Appl. Phys. 7 42–6

[149] Palin D E and Wadsworth K D 1948 Structure of carbonmonofluoride Nature 162 925–6

[150] Touhara H and Okino F 2000 Property control of carbonmaterials by fluorination Carbon 38 241–67

[151] Chen M, Zhou H, Qiu C, Yang H, Yu F and Sun L 2012Layer-dependent fluorination and doping of graphene viaplasma treatment Nanotechnology 23 115706

[152] Lee W H, Suk J W, Chou H, Lee J, Hao Y, Wu Y, Piner R,Akinwande D, Kim K S and Ruoff R S 2012 Selective-area fluorination of graphene with fluoropolymer and laserirradiation Nano Lett. 12 2374–8

[153] Withers F, Bointon T H, Dubois M, Russo S andCraciun M F 2011 Nanopatterning of fluorinated grapheneby electron beam irradiation Nano Lett. 11 3912–6

[154] Withers F, Russo S, Dubois M and Craciun M F 2011 Tuningthe electronic transport properties of graphene throughfunctionalisation with fluorine Nanoscale Res. Lett. 6 526

[155] Watanabe N 1980 Two types of graphite fluorides, (CF)n and(C2F)n, and discharge characteristics and mechanisms ofelectrodes of (CF)n and (C2F)n in lithium batteries SolidState Ion. 1 87–110

[156] Singh A K and Yakobson B I 2009 Electronics andmagnetism of patterned graphene nanoroads Nano Lett.9 1540

[157] Shen N and Sofo J O 2011 Dispersion of edge states andquantum confinement of electrons in graphene channelsdrawn on graphene fluoride Phys. Rev. B 83 245424

[158] Huang H, Li Z and Wang W 2012 Electronic and magneticproperties of oxygen patterned graphene superlatticeJ. Appl. Phys. 112 114332

[159] Martins S E, Withers F, Dubois M, Craciun M F and Russo S2013 Tuning the transport gap of functionalized graphenevia electron beam irradiation New J. Phys. 15 033024

[160] Byun I-S et al 2011 Nanoscale lithography on monolayergraphene using hydrogenation and oxidation ACS Nano5 6417–24

[161] Wei Z et al 2010 Nanoscale tunable reduction of grapheneoxide for graphene electronics Science 328 1373–6

[162] Mativetsky J M, Treossi E, Orgiu E, Melucci M,Veronese G P, Samorı P and Palermo V 2010 Localcurrent mapping and patterning of reduced graphene oxideJ. Am. Chem. Soc. 132 14130–6

[163] Dresselhaus M S and Dresselhaus G 1981 Intercalationcompounds of graphite Adv. Phys. 30 139–326

[164] Gall N R, Rut’kov E V and Tontegode A Ya 2000Intercalation of nickel atoms under two-dimensionalgraphene film on (111)Ir Carbon 38 663–7

[165] Gall N R, Rut’kov E V and Tontegode A Ya 1997Two dimensional graphite films on metals and theirintercalation Int. J. Mod. Phys. B 11 1865–911

[166] Riedl C, Coletti C, Iwasaki T, Zakharov A A and Starke U2009 Quasi-free-standing epitaxial graphene on SiCobtained by hydrogen intercalation Phys. Rev. Lett.103 246804

[167] Schumacher S, Forster D F, Rosner M, Wehling T O andMichely T 2013 Strain in epitaxial graphene visualized byintercalation Phys. Rev. Lett. 110 086111

[168] Rybkina A A, Rybkin A G, Fedorov A V, Usachov D Y,Yachmenev M E, Marchenko D E, Vilkov O Y,Nelyubov A V, Adamchuk V K and Shikin A M 2013Interaction of graphene with intercalated Al: the process

20

J. Phys.: Condens. Matter 25 (2013) 423201 Topical Review

of intercalation and specific features of the electronicstructure of the system Surf. Sci. 609 7–17

[169] Li L, Wang Y, Meng L, Wu R T and Gao H J 2013 Hafniumintercalation between epitaxial graphene and Ir(111)substrate Appl. Phys. Lett. 102 093106

[170] Johannsen J S et al 2013 Electron–phonon coupling inquasi-free-standing graphene J. Phys.: Condens. Matter25 094001

[171] Sutter P, Sadowski J T and Sutter E A 2010 Chemistry undercover: tuning metal–graphene interaction by reactiveintercalation J. Am. Chem. Soc. 132 8175

[172] Cupolillo A, Ligato N and Caputi L S 2013 Plasmondispersion in quasi-freestanding graphene on Ni(111)Appl. Phys. Lett. 102 111609

[173] Decker R, Brede J, Atodiresei N, Caciuc V, Blugel S andWiesendanger R 2013 Atomic-scale magnetism ofcobalt-intercalated graphene Phys. Rev. B 87 041403

[174] Marchenko D, Varykhalov A, Scholz M R, Bihlmayer G,Rashba E I, Rybkin A, Shikin A M and Rader O 2012Giant Rashba splitting in graphene due to hybridizationwith gold Nature Commun. 3 1232

[175] Shikin A M, Rybkin A G, Marchenko D, Rybkina A A,Scholz M R, Rader O and Varykhalov A 2013 Inducedspin–orbit splitting in graphene: the role of atomic numberof the intercalated metal and π -d hybridization New J.Phys. 15 013016

[176] Huang Q, Chen X, Lin J, Li K, Jia Y, Liu J, Guo L, Wang Wand Wang G 2011 Preparation of quasi-free-standinggraphene with a super large interlayer distance by methaneintercalation J. Phys. Chem. C 115 20538–45

[177] Hargrove J, Shashikala H B M, Guerrido L, Ravi N andWang X Q 2012 Band gap opening in methaneintercalated graphene Nanoscale 4 4443–6

[178] Sugawara K, Kanetani K, Sato T and Takahashi T 2011Fabrication of Li-intercalated bilayer graphene AIP Adv.1 022103

[179] Zhou J J, Zhou W W, Guan C M, Shen J Q, Ouyang C Y,Lei M S, Shi S Q and Tang W H 2012 First-principlesstudy of lithium intercalated bilayer graphene Sci. ChinaPhys. Mech. Astron. 55 1376–82

[180] Profeta G, Calandra M and Mauri F 2012 Phonon-mediatedsuperconductivity in graphene by lithium depositionNature Phys. 8 131–4

[181] Lee E and Persson K A 2012 Li absorption and intercalationin single layer graphene and few layer graphene by firstprinciples Nano Lett. 12 4624–8

[182] Xu B, Lu Y H, Sha Z D, Feng Y P and Lin J Y 2011Coupling of magnetic edge states in Li-intercalated bilayerand multilayer zigzag graphene nanoribbons Europhys.Lett. 94 27007

[183] Kaloni T P, Kahaly M U, Cheng Y C andSchwingenschlogl U 2012 Charge carrier density inLi-intercalated graphene Chem. Phys. Lett. 534 29–33

[184] Uchoa B and Castro Neto A H 2007 Superconducting statesof pure and doped graphene Phys. Rev. Lett. 98 146801

[185] Mazin I I and Balatsky A V 2010 Superconductivity inCa-intercalated bilayer graphene Phil. Mag. Lett. 90 731–8

[186] Kanetani K, Sugawara K, Sato T, Shimizu R, Iwaya K,Hitosugi T and Takahashi T 2012 Ca intercalated bilayergraphene as a thinnest limit of superconducting C6CaProc. Natl Acad. Sci. USA 109 19610

[187] Kleeman J, Sugawara K, Sato T and Takahashi T 2013 Directevidence for a metallic interlayer band in Rb-intercalatedbilayer graphene Phys. Rev. B 87 195401

[188] Xue M, Chen G, Yang H, Zhu Y, Wang D, He J and Cao T2012 Superconductivity in potassium-doped few-layergraphene J. Am. Chem. Soc. 134 6536–9

[189] Wei D, Liang J, Zhu Y, Yuan Z, Li N and Qian Y 2013Formation of graphene-wrapped nanocrystals at roomtemperature through the colloidal coagulation effect Part.Part. Syst. Charact. 30 143–7

[190] Hu T, Sun X, Sun H, Yu M, Lu F, Liu C and Lian J 2013Flexible free-standing graphene–TiO2 hybrid paper for useas lithium ion battery anode materials Carbon 51 322–6

[191] Fan X, Liu L, Kuo J L and Shen Z 2010 Functionalizingsingle- and multi-layer graphene with Br and Br2 J. Phys.Chem. C 114 14939–45

[192] Gong S J, Sheng W, Yang Z Q and Chu J H 2010 First-principles investigation of bilayer graphene withintercalated C, N or O atoms J. Phys.: Condens. Matter22 245502

[193] Zhou J, Wang L, Qin R, Zheng J, Mei W N, Dowben P A,Nagase S, Gao Z and Lu J 2011 Structure and electronicand transport properties of transition metal intercalatedgraphene and graphene-hexagonal-boron-nitride bilayerJ. Phys. Chem. C 115 25273–80

[194] Kaloni T P, Kahaly M U, Cheng Y C andSchwingenschlogl U 2012 K-intercalated carbon systems:effects of dimensionality and substrate Europhys. Lett.98 67003

[195] Jung N, Kim B, Crowther A C, Kim N, Nuckolls C andBrus L 2011 Optical reflectivity and Raman scattering infew-layer-thick graphene highly doped by K and Rb ACSNano 5 5708–16

[196] Jadaun P, Movva H C P, Register L F and Banerjee S K 2013Theory and synthesis of bilayer graphene intercalated withICl and IBr for low power device applications,arXiv:1301.4265

[197] Khrapach I, Withers F, Bointon T H, Polyushkin D K,Barnes W L, Russo S and Craciun M F 2012 Novel highlyconductive and transparent graphene-based conductorsAdv. Mater. 24 2844–9

[198] Kumar A et al 2011 Direct synthesis of lithium-intercalatedgraphene for electrochemical energy storage applicationACS Nano 5 4345–9

[199] Chen Z, Cotterell B, Wang W, Guenther E and Chua S J 2001A mechanical assessment of flexible optoelectronicdevices Thin Solid Films 394 202–6

[200] Schlatmann A R, Floet D W, Hilberer A, Garten F,Smulders P J M, Klapwijk T M and Hadziioannou G 1996Indium contamination from the indium–tin–oxideelectrode in polymer light-emitting diodes Appl. Phys.Lett. 69 17646

[201] Forrest S R 2004 The path to ubiquitous and low-cost organicelectronic appliances on plastic Nature 428 911–8

[202] Hecht D S, Hu L and Irvin G 2011 Emerging transparentelectrodes based on thin films of carbon nanotubes,graphene, and metallic nanostructures Adv. Mater.23 1482–513

[203] Wu H, Kong D, Ruan Z, Hsu P-C, Wang S, Yu Z, Carney T J,Hu L, Fan S and Cui Y 2013 A transparent electrode basedon a metal nanotrough network Nature Nanotechnol.8 421–5

[204] De S, Boland C S, King P J, Sorel S, Lotya M, Patel U,Xiao Z L and Coleman J N 2011 Transparent conductingfilms from NbSe3 nanowires Nanotechnology 22 285202

[205] Zhao W, Tan P H, Liu J and Ferrari A C 2011 Intercalation offew-layer graphite flakes with FeCl3: Ramandetermination of fermi level, layer by layer decoupling,and stability J. Am. Chem. Soc. 133 5941

[206] Kim K K, Reina A, Shi Y, Park H, Li L, Lee Y H and Kong J2010 Enhancing the conductivity of transparent graphenefilms via doping Nanotechnology 21 285205

21

J. Phys.: Condens. Matter 25 (2013) 423201 Topical Review

[207] Kravets V G, Grigorenko A N, Nair R R, Blake P,Anissimova S, Novoselov K S and Geim A K 2010Spectroscopic ellipsometry of graphene and an exciton-shifted van Hove peak in absorption Phys. Rev. B81 155413

[208] Vicarelli L, Vitiello M S, Coquillat D, Lombardo A,Ferrari A C, Knap W, Polini M, Pellegrini V andTredicucci A 2012 Graphene field-effect transistors asroom-temperature terahertz detectors Nature Mater.11 865–71

[209] Saleh B E A and Teich M C 1991 Fundamentals of Photonics(New York: Wiley)

[210] Xia F, Mueller T, Golizadeh-Mojarad R, Freitag M,Lin Y M, Tsang J, Perebeinos V and Avouris P 2009Photocurrent imaging and efficient photon detection in agraphene transistor Nano Lett. 9 1039–44

[211] Han T H, Lee Y, Choi M R, Woo S H, Bae S H, Hong B H,Ahn J H and Lee T W 2012 Extremely efficient flexibleorganic light-emitting diodes with modified grapheneanode Nature Photon. 6 105–10

[212] Xia F, Mueller T, Lin Y M, Valdes-Garcia A and Avouris P2009 Ultrafast graphene photodetector NatureNanotechnol. 4 839–43

[213] Urich A, Unterrainer K and Mueller T 2011 Intrinsicresponse time of graphene photodetectors Nano Lett.11 2804–8

[214] Murray C B, Norris D J and Bawendi M G 1993 Synthesisand characterization of nearly monodisperse CdE(E = sulfur, selenium, tellurium) semiconductornanocrystallites J. Am. Chem. Soc. 115 8706–15

[215] Tsukazaki A et al 2005 Repeated temperature modulationepitaxy for p-type doping and light-emitting diode basedon ZnO Nature Mater. 4 42–6

[216] Li F, Cho S H, Son D I, Kim T W, Lee S K, Cho Y H andJin S 2009 UV photovoltaic cells based on conjugatedZnO quantum dot/multiwalled carbon nanotubeheterostructures Appl. Phys. Lett. 94 111906

[217] Ghosh S, Hoogland S, Sukhovatkin V, Levina L andSargent E H 2011 A tunable colloidal quantum dot photofield-effect transistor Appl. Phys. Lett. 99 101102

[218] Konstantatos G and Sargent E H 2007 PbS colloidal quantumdot photoconductive photodetectors: transport, traps, andgain Appl. Phys. Lett. 91 173505

[219] Konstantatos G, Howard I, Fischer A, Hoogland S, Clifford J,Klem E, Levina L and Sargent E H 2006 Ultrasensitivesolution-cast quantum dot photodetectors Nature442 180–3

[220] Sun Z, Liu Z, Li J, Tai G A, Lau S P and Yan F 2012 Infraredphotodetectors based on CVD-grown graphene and PbSquantum dots with ultrahigh responsivity Adv. Mater.24 5878–83

[221] Konstantatos G, Badioli M, Gaudreau L, Osmond J,Bernechea M, de Arquer F P G, Gatti F andKoppens F H L 2012 Hybrid graphene-quantum dotphototransistors with ultrahigh gain Nature Nanotechnol.7 363–8

[222] Son D I, Yang H Y, Kim T W and Park W I 2013Photoresponse mechanisms of ultraviolet photodetectorsbased on colloidal ZnO quantum dot-graphenenanocomposites Appl. Phys. Lett. 102 021105

[223] Konstantatos G, Levina L, Fischer A and Sargent E H 2008Engineering the temporal response of photoconductivephotodetectors via selective introduction of surface trapstates Nano Lett. 8 1446–50

[224] Sukhovatkin V, Hinds S, Brzozowski L and Sargent E H2009 Colloidal quantum-dot photodetectors exploitingmultiexciton generation Science 324 1542–4

[225] Guo C X, Yang H B, Sheng Z M, Lu Z S, Song Q L andLi C M 2010 Layered graphene/quantum dots forphotovoltaic devices Angew. Chem. Int. Edn 49 3014–7

[226] Reyren N et al 2007 Superconducting interfaces betweeninsulating oxides Science 31 1196–9

[227] Nandkishore R, Levitov L S and Chubukov A V 2012 Chiralsuperconductivity from repulsive interactions in dopedgraphene Nature Phys. 8 158–63

[228] Ye J, Craciun M F, Koshino M, Russo S, Inoue S, Yuan H,Shimotani H, Morpurgo A F and Iwasa Y 2011 Accessingthe transport properties of graphene and its multilayers athigh carrier density Proc. Natl Acad. Sci. USA108 13002–6

22