A review of graphene based transparent conducting

Renewable and Sustainable Energy Reviews 99 (2019) 83–99

Contents lists available at ScienceDirect

Renewable and Sustainable Energy Reviews journal homepage: www.elsevier.com/locate/rser

A review of graphene based transparent conducting films for use in solar photovoltaic applications

T

⁎

Nurul Nazli Rosli, Mohd Adib Ibrahim , Norasikin Ahmad Ludin, Mohd Asri Mat Teridi, Kamaruzzaman Sopian Solar Energy Research Institute (SERI), Universiti Kebangsaan Malaysia, 43600 Bangi, Selangor, Malaysia

A R T I C LE I N FO

A B S T R A C T

Keywords: Figure of merit Graphene Nanostructure Solar cell Transparent conducting oxides

Graphene has been recognised for its various excellent properties and its potential to be applied in various applications such as transparent conducting films (TCF), optoelectronic devices and energy storage. Graphene has been successfully synthesised by various methods including chemical vapour deposition (CVD), exfoliation and solution-processed. As-synthesised graphene has been used in various applications and its potentials have been realised. However, graphene performance has been compromised due to various factors. In this review, we present the methods to improve graphene performance by doping, enhancement of transfer methods and hybrid films of graphene with other materials. In addition, graphene as TCF for photovoltaic applications has also been described. We also include a discussion on various Figure of Merit (FOM) determinations, to evaluate the performance of graphene as transparent conducting films.

1. Introduction Transparent conducting films (TCF) are highly conductive and highly transparent materials used in various applications such as solar cells, touch screens, light emitting devices, etc. Indium tin oxide (ITO) as TCF is a commonly used material in various applications as it exhibits excellent properties of low sheet resistance (Rs) in the range of 10 – 25 Ω/sq and optical transmittance (T) greater than 90%. However, the shortcomings of ITO such as the high cost of raw material (scarcity of indium), brittleness when applied in flexible devices and expensive fabrication process (high temperature processing), have led to the study

of alternative materials. In addition, the problem of instability in acids or bases, vulnerability to ion diffusion into the polymer layer and poor electrical contact with organic materials, have reduced the possibility of ITO for further use in organic devices. On the other hand, the other widely used TCF for its comparable performance with ITO, fluorine tin oxide (FTO), is associated with problems such as degradable performance at high temperatures due to an increase in sheet resistance and defects of FTO, causing current leakage. As TCF, the materials must fulfil the requirements of high electrical conductivity and high optical transparency for use in various device applications. On the other hand, other factors to be considered in

Abbreviation: a-C, amorphous Carbon; a-Si, amorphous silicon; a-Si:H, hydrogenated amorphous silicon; AgNC, silver nanocomposite; AgNW, silver nanowires; APTES, 3-Aminoproplytriethoxysilane; AZO, aluminium zinc oxide; BLG, bilayer graphene; BIPV, building integrated photovoltaics; C-CVD, Catalytic-Chemical Vapour Deposition; CdTe, cadmium telluride; CFG, crack-filled graphene; CIGS, copper indium gallium selenide; CNT, carbon nanotubes; CO, carbon monoxide; CuNF, copper nanofibers; CuNW, copper nanowires; CuNT, copper nanotrough; CVD, chemical vapour deposition;; CZTS, copper zinc tin sulfide; DC, direct current; DSSC, dye-sensitised solar cells; EPDM, ethylene-propylene-diene monomer; ETL, electron transport layer; FLG, few layer graphene; FOM, Figure of Merit; FTO, fluorine tin oxide; GO, graphene oxide; hBN, hexagonal boron nitride; HFTCVD, hot filament thermal chemical vapour deposition;; HJSC, heterojunction solar cells; HOPG, highly ordered pyrolytic graphite;; HTL, hole transport layer; IPA, isopropyl alcohol; ITO, indium tin oxide; LB, Langmuir-Blodgett; LbL, layer-by-layer; LPCVD, low-pressure chemical vapour deposition;; LT-PECVD, low-temperature plasma enhanced chemical vapour deposition; MIBK, methyl isobutyl ketone; MLG, multilayer graphene; MNWT, multi-walled nanotubes; OPV, polymer solar cells; OSC, organic solar cells; PAni, polyaniline; PCE, power conversion efficiency; PDMS, polydimethylsiloxane; PEDOT:PSS, poly(3,4-ethylenedioxythiophene:poly(styrene-sulfonate); PEI, polyethyleneimine; PET, polyethylene terephthalate; PMMA, polymethylmethacrylate; PSC, perovskite solar cells; PVA, polyvinyl alcohol; P3HT, poly (3,4-ethylenedioxythiophene; QD, quantum dots; rGO, reduced graphene oxide; SAM, self-assembled monolayers; SLG, single layer graphene; SWNT, single-walled nanotubes; TCF, transparent conducting films; TCL, transparent conducting layer; TCNQ, tetracyanoquinodimethane; TCO, transparent conducting oxides; TE, transparent electrode; TEM, transmission electron microscopy; TFSA, trifluoromethanesulfonyl-amide; TFSC, thin film solar cells; 0D, zero-dimensional; 2D, two-dimensional; 3D, three-dimensional; 6P, para-hexapheny ⁎ Corresponding author. E-mail address: [email protected] (M.A. Ibrahim). https://doi.org/10.1016/j.rser.2018.09.011 Received 30 May 2017; Received in revised form 30 August 2018; Accepted 5 September 2018 1364-0321/ © 2018 Elsevier Ltd. All rights reserved.


N.N. Rosli et al.

σDC,B n tmin R ε0 c n m* μ λ e σDC σOP nsub Zo εc Gsh A K Vo Jsc

Nomenclature Ar KMnO4 CH3NH3 Co Cu C10H16O HAuCl4 HI H2 H2SO4 NaNO3 Ni SOCl2 Rs T σ ρ t α

argon potassium permanganate methylammonium cobalt copper botanical derivative camphor tetrachloroauric acid hydriodic acid hydrogen sulphuric acid sodium nitrate nickel thionyl chloride sheet resistance optical transmittance conductivity resistivity film thickness optical absorption

bulk DC conductivity percolation exponent critical thickness reflectance permittivity of free space speed of light in a vacuum refractive index of the film effective mass of the conduction electrons mobility visible wavelength of light electronic charge DC conductivity optical conductivity refractive index of the substrate impedance in free space critical strain sheet conductance absorption coefficient of proportionality open circuit voltage short circuit current density

Due to these excellent properties, graphene has been applied in various applications such as solar cells, sensors, flexible electronics and others. In application, graphene may be applied as charge carrier or as a transparent conductive electrode. As graphene has been the centre of attention recently, many studies are conducted on the synthesis and application of graphene; as well as numerous reviews on graphene [15–17]. This review aims to give an update to the research progress in graphene, which includes graphene synthesis, the method to improve the performance of graphene as well as the performance of photovoltaic applications incorporating graphene-based transparent conducting films. We also include the performance evaluation of graphene-based TCF by using the Figure of Merit (FOM).

developing TCF include low-cost materials, low-temperature processing methods for lower fabrication costs, and flexibility for flexible applications. Numerous materials for TCF alternatives have been studied, including transparent conducting oxides (TCO) [1], conductive polymer [2], metal nanowires [3–5], carbon nanotubes [6] and also graphene. The semiconductor material of TCO was extensively applied as TCF owing to its properties of electrical conductivity (~ 10−4 Ω cm) and high transmittance over the visible light range (> 90%). Various metal oxides have been reported, including binary and multicomponent TCO as well as improvements to the TCO by proposing multilayers of TCO/ metal/TCO [7,8]. Reviews on the status of TCO have been described in detail by Minami et al. [9,10], Stadler [11] and Ellmer [12]. Among TCO, aluminium zinc oxide (AZO) is known to have comparable performance to ITO and FTO including other advantages of less toxicity, inexpensive cost and simple fabrication methods. However, grain boundaries and electron scattering have reduced the carrier mobility of TCO, consequently affecting the performance of these materials. Metal nanostructures consisting of nanowires, nanogrid, nanofibres etc. are also excellent candidate materials for TCF due to their exceptional properties including high conductivity and being highly transparent in visible wavelengths. The conductivity of these materials is dependent on the percolation network, of which the high aspect ratio of longer and thinner nanomaterial is preferable in order to obtain the best performance of TCF with low sheet resistance and high transmittance [13]. In addition, the mechanical flexibility of these materials makes them relevant for flexible applications. However, the shortcomings of metal nanostructures, i.e. the high junction resistance and poor adhesion to the substrate, require further improvement to these materials. Carbon nanotubes (CNT) are one-dimensional carbon-based materials with high conductivity and high transparency, applicable as a substitute for ITO TCF. CNT have high mobility despite having much lower carrier concentration compared to ITO, i.e. ~ 1017/cm3 and 1020/cm3, respectively. The properties of CNT differ according to the wall numbers such as single-walled nanotubes (SWNT) and multiwalled nanotubes (MNWT). Previously, CNT were applied as transparent electrodes for organic solar cells, showing a comparable performance to the solar cell with ITO as a transparent electrode [6,14]. However, the high sheet resistance of CNT compared to ITO is still the main concern. Recently, graphene has been studied due to its exceptional properties such as high carrier mobility of ~ 200,000 cm2/Vs, high transparency in visible light and near infrared, and its being highly flexible.

2. TCF characteristics properties and performance evaluation 2.1. Fundamentals of electrical and optical properties The best way to describe the property of a material is by using the conductivity, σ (S/cm) and resistivity, ρ (Ω cm); the conductivity is a measurement of the ability for material to conduct electricity, whereas the resistivity is a measurement of the ability of material to oppose electricity flow. For a thin film, its resistance is measured by the sheet resistance, i.e. a resistance in regard to the thickness. The sheet resistance can be described by

Rs =

ρ 1 = t σt

(1)

where t is the film thickness (nm), ρ = 1/σ is the resistivity and σ is the direct current (DC) conductivity. High conductivities (i.e. lower resistivity) can be obtained with a thicker film. However, the transparency of the thicker film may be affected due to an increase in the optical absorption of the film, as given by

T = e−αt

(2)

where α is the optical absorption [18]. The sheet resistance of bulk film may be obtained by combining Eqs. (1) and (2) to obtain

T = e−α / σDC, B Rs

(3)

where σDC,B is the bulk DC conductivity. The sheet resistance of nanostructure materials described by De et al. [19] is given by 84


N.N. Rosli et al.

Rs =

1 σDC , B (t / tmin)n t

Rs =

1

Rs =

vacuum, n is the refractive index of the film, m* is the effective mass of the conduction electrons, μ is the mobility, λ is the visible wavelength of light, and e is the electronic charge. By using this equation, the material with the highest mobility and effective mass will have higher FOM (σ/α). Next, the FOM defined by the ratio of DC conductivity (σDC ) and optical conductivity (σOP ) can be obtained by the following equation

n σDC , B t n +1 / tmin n +1 tmin

tmin σDC , B t n +1

(4)

where n is the percolation exponent and tmin is the critical thickness, below which the nanostructure materials no longer have DC bulk conductivity. The conductivity of nanostructure materials depends on the percolation network formed by conducting paths which will increase with increasing conducting links.

σDC 188.5 T = σOP Rs (1− T )

where σDC is conductivity by charge transport due to constant applied fields, whereas σOP is due to the motion of electrons in response to optical fields [24]. The T of 550 nm is often used for the FOM of Eq. (11), in which the solar conversion is a maximum at this wavelength. The (σDC /σOP ) ratio is independent of thickness and can be used to evaluate all films with different thickness, synthesised from different methods and materials. According to De et al. [25], the thin film is required to have thickness (t > 20 nm), below this thickness, the DC conductivity decreases significantly as observed for nanostructured films (viz. nanowires, nanotubes, graphene, nanogrids, etc.) of thinner films (t < 20 nm). Thus, Eq. (11) is suitable for bulk film and does not necessarily fit nanostructure films. In addition, Barnes et al. [26] reported that the FOM based on Eq. (2) is only suitable for freestanding films. Foreseeing using TCF in the application, the films are required to be deposited on the substrate; thus, the calculation of FOM for films coated on the substrate is required. The FOM (σDC / σOP ) for the film coated on the substrate may be obtained using the following equation

2.2. Figure of merit (FOM) determination The material requires low sheet resistance and high optical transparency to be applied as TCF. For most materials, particularly graphene, the best performance with high conductivity and low sheet resistance is often obtained at lower transparency. Thus, an indicator to determine and evaluate the performance of TCF is required, and FOM has been used to serve this purpose. Many researchers have reported the FOM for TCF; each of them has a different approach to determine the FOM using different FOM equations. Conversely, none has been declared as the sole method to be used for FOM determination. Nonetheless, we discuss the equations that have been used to determine the FOM for TCF. In 1972, Fraser and Cook [20] introduced a simple calculation of FOM (F) by dividing the transmittance (500 nm) by the sheet resistance of the film, of which the obtained value was used in a dimensionless form.

F = T / Rs

(5)

T=

A few years later, Haacke [21] proposed FOM (ϕTC ) with increased T to the power of 10, of which the maximum ϕTC occured at 90% transmittance. Haacke redefined the new FOM as the previous FOM (Eq. (5)) had too much influence on the sheet resistance and thickness-dependent for which the best FOM was obtained at certain thickness for films having a low transparency of 37%. The thickness-dependent FOM by Haacke is represented as follows:

ϕTC = T10Rs−1

(6)

σDC , B / σOP ⎤1/(n +1) Π=2 ⎡ ⎢ ⎣ Z0 tmin σOP ⎥ ⎦

(7)

(8)

ΠTC = σεc / α

Gordon et al. [23] listed a few FOM equations for evaluating the performance of TCF by using the ratio of electrical conductivity to the visible coefficient (σ/α), represented by the following equation

σ = −[Rs ln(T + R)]−1 α

(12)

(13)

where Π is dimensionless, and Zo is the impedance in free space (377 Ω). For flexible applications, Shim et al. [27] proposed the FOM (ΠTC) for use in evaluating the performance of flexible materials as represented by the following equation:

A modification was made by Schropp and Madan [22], where ρ was used instead of Rs, as the literature does not always report the thickness at which the resistivity was obtained. The FOM proposed by Schropp and Madan is as follows:

F = −1/(ρlnT )

2 16nsub 1 2 σ (1+nsub) 4 1+ Zo 1 ⎡ R (1 + n ) σOP ⎤ s sub DC ⎦ ⎣

where nsub is the refractive index of the substrate. As the nanostructure materials form the percolation network that generates the conductivity where longer connecting bundles or networks produce higher conductivity, the conductivity of nanostructure materials does not depend on the thickness of the films. Therefore, we may consider that Eq. (11) is not valid for nanostructure materials, where at the thickness < 20 nm, the nanostructure material may have conductivity due to the percolation network; but at greater thickness, we may consider that nanostructure materials have bulk-like film conductivity. The derivation for percolative FOM (Π) was demonstrated by De et al. [19] as:

The FOM obtained by Haacke is film thickness–dependent and is often used to evaluate the performance of TCF, as it is much easier to be calculated. Jain and Kulshreshtha [18] proposed a thickness–independent FOM incorporating σ and α of the film only, which can be represented as follows

F = − Rs lnT

(11)

(14)

where εc is the critical strain. Next, a different approach using σ/α was demonstrated by Rowell and McGehee [28], which is independent of film thickness and device construction. The FOM by σ/α can be obtained using the following equation,

(9)

where R is reflectance. In the same review, the authors discussed the FOM based on the transport theory of an electron,

σ 1/R t G = −ln (1sh−A) ≈ sh α A

m*μ ⎞2 σ = 4π 2ε0 c 3n ⎛ α ⎝ λe ⎠

where Gsh is the sheet conductance and A is the absorption. Kaskela et al. [29] reported the FOM for characterizing the quality of SWCNT networks using the coefficient of proportionality (K), as shown in the following equations:

t

(10)

where ε0 is the permittivity of free space, c is the speed of light in a 85

(15)


N.N. Rosli et al.

1 ⎡ −1⎤ Ω ⎥ Rs A ⎢ ⎦ ⎣

K=

2D carbon nanotubes and stacked graphene will form three-dimensional (3D) graphite. The structures of graphene, fullerene, CNT and graphite are illustrated in Fig. 1. Graphite oxide or graphene oxide (GO) has many functional oxygen groups, produced from the oxidation of graphite. Typically, reduced graphene oxide (rGO) is formed by reduction of graphene oxide, which alters its chemical structures by removing most functional groups, consequently causing some defects and lowering the electronic performance of rGO. Thus, the electronic properties of rGO have yet to surpass those of pristine graphene synthesised by mechanical exfoliation (tape-peeling method). The reduction of graphene oxide is illustrated in Fig. 2.

(16)

or

RE =

1 [Ω/sq] K log(10/9)

(17)

To evaluate the performance of TCF, the higher FOM is preferable, regardless of the FOM equations used. The description of FOM is summarised in Table 1. 2.3. TCF performance evaluation The best performance of TCF may be described by the films with low Rs at high optical transmittance, where the performance of these films can be evaluated by FOM, as higher FOM shows better TCF performance. We listed a few equations in Section 2.2; however, we would prefer to evaluate the performance of TCF based on the literature using Eqs. (6) and (11). These equations were chosen as they are commonly used in various studies and are much easier to calculate and evaluate the performance of TCF. For industrial purposes, the TCF requires Rs of ~ 100 Ω/sq at > 90% transmittance. Meanwhile, for device applications (particularly solar cells), Rs = 10 Ω/sq at > 90% transmittance is required. Thus, in order to set up a benchmark for TCF, the FOM is obtained using Eq. (2) with Rs = 10 Ω/sq and T = 90% transmittance, where the benchmark FOM using Eqs. (6) and (11) are ≈ 3.487 × 10−2 Ω−1 and ≈ 350, respectively. We reviewed 30 studies and compiled their data (Rs and T) to calculate the FOM for their TCF. Our calculated FOM are summarised in Tables 2–4; however, these calculations are only for comparison purposes and may not directly reflect the actual result, as the data are purely obtained from the literature.

4. Synthesis of graphene and optoelectronic properties of assynthesised graphene Various synthesis methods have been employed to obtain graphene for use in various applications. The synthesis methods of graphene are highly important since they have a significant effect on the yield and structure of graphene. The structure of graphene will determine its electrical performance. The most common methods of graphene synthesis are summarised in Fig. 3. The end of products of graphene and rGO are used in applications due to their excellent properties. Generally, graphene is produced by mechanical exfoliation, CVD and solution-processing. We report on the progress of these methods in the following sections. 4.1. Mechanical exfoliation The first graphene was discovered in 2004 by Novoselov and Geim by mechanical exfoliation of highly ordered pyrolytic graphite (HOPG) [34]. The high quality and large area graphene with the high mobility of ~10000 cm2/Vs at room temperature was successfully obtained; however, the yield of graphene was too small and the method was not feasible for large-scale production. Ever since, various methods and modifications of the compounds were done to increase the yield of graphene. The mechanisms and types of mechanical exfoliation techniques to obtain high-quality graphene were discussed in great detail by Yi and Shen [35]. In their review, types of mechanical exfoliation include micromechanical cleavage, sonication, ball milling, fluid dynamics, etc.; these techniques involve a mechanism of generating shear force or nominal force to overcome the Van der Waals attraction

3. Structure of graphite, graphene and reduced graphene oxide Carbon forms various allotropes such as fullerenes, graphite, graphene, carbon nanotubes, etc.; these materials exhibit different properties due to the atomic orientations for each material. A single sheet of carbon atoms in a two-dimensional (2D) structure with a honeycomb lattice is known as graphene. Graphene may be constructed to form the structure of different dimensions, where wrapped graphene will form zero-dimensional (0D) bulkyball or fullerene, rolled graphene will form Table 1 Summary of FOM equations for TCF evaluation. FOM Type

Equations

FOM 1 FOM 2 FOM 3

Eq. (5) Eq. (6) Eq. (7)

FOM 4 FOM 5

Eq. (8) Eq. (9)

FOM 6

Eq. (10)

FOM 7

Eq. (11)

FOM 8

Eq. (12)

FOM 9 FOM 10 FOM 11

Eq. (13) Eq. (14) Eq. (15)

FOM 12

Eq. (16)

a

Requirements/conditions dependent • Thickness dependent • Thickness optical absorption and electrical conductivity of the film • Involves independent • Thickness dependent • Thickness independent • Thickness dependent • Wavelength and effective mass dependent • Mobility not involve free-electron concentration • Does independent • Thickness conductivity for bulk thick films • DC not fit for thinner films (t < 20 nm) • Does losses at the interfaces of air, film and glass • Consider film coated on substrate • For follows percolation-like thickness • Conductivity include mechanical parameters (tensile strength, toughness, etc.) • May independent • Thickness not depend on the device architecture • Does • Thickness independent

Applicationa

References

Thin films Thin films TCO thin films

Fraser and Cook [20] Haacke [21] Jain and Kulshreshtha [18]

TCO thin films TCO thin films

Schropp and Madan [22] Gordon [23] Exarhous and Gordon [30] Gordon [23] Golobostanfard et al. [31] Barnes et al. [26] Ellmer [12]

TCO thin films Thin films

Thin films Nanostructure materials Flexible films Thin films

Barnes et al. [26] Ellmer [12] De et al. [19] Shim et al. [27] Rowell and McGehee [28]

Quality evaluation of SWNT networks

Kaskela et al. [29]

Note: The thin film indicated that the FOM suitable to thin film of materials whereas TCO thin films indicated that the FOM only suitable for TCO only. 86


N.N. Rosli et al.

Table 2 Evaluation of FOM on graphene based TCF by various synthesis methods. Materials

Synthesis method

Rs (Ω/sq)

T (%)

a FOM 2 (Ω−1)

rGO

Solution-processed Reduction of GO in hydrazine Solution-processed Solution-processed CVD CVD CVD CVD CVD CVD CVD APCVD CVD CVD APCVD CVD

70.0k 100k ~ 4 × 106 1 × 106 1.645k 860.0 ~ 1 × 103 4000 770–1000 730.0 650.0 1150 725.0 950.0 220.0 374.0

65.0 95.0 N.A. 88.0 81.0 91.6 85.0 98.0 90.0 97.7 97.0 97.0 97.6 97.0 84.0 84.2

1.920 × 10−7 5.987 × 10−6 N.A. 2.785 × 10−6 7.391 × 10−5 4.836 × 10−4 1.969 × 10−3 2.042 × 10−4 4.529 × 10−4 −3.487 × 10−4 1.085 × 10−3 1.134 × 10−3 6.412 × 10−4 1.082 × 10−3 7.762 × 10−4 7.950 × 10−4 4.789 × 10−4

Graphene

Monolayer graphene 4-layer graphene 7-layer graphene Multilayer graphene

b

FOM 7

0.011 0.073 N.A. 0.029 1.031 4.888 2.227 4.642 4.526–3.485 22.066 18.897 10.681 21.276 12.93 9.406 5.613

References

Eda et al. [58] Wu et al. [53] Gilje et al. [55] Ishikawa et al. [65] Kalita et al. [38] Kalita et al. [39] Ma and Zhang [66] Ho et al. [59] Reina et al. [67] Jang et al. [60] Cha et al. [61] Bi et al. [62] Gunes et al. [63] Kwon et al. [64] Bi et al. [62] Choi et al. [41]

*Note: N.A. – Not Available; APCVD – Atmospheric Pressure Chemical Vapour Deposition. a FOM calculated using Eq. (6). b FOM calculated using Eq. (11).

includes oxygen plasma cleaning to remove the ambient adsorbates from the substrate; and an additional heat treatment to maximise the uniform contact area at the interface between the source crystal and the substrates. These modifications aim to enhance and homogenise the adhesion force of the outermost sheet in contact with the substrates.

holding the graphene layer from the bulk precursor. We will not discuss the mechanism of mechanical exfoliation as it has already been discussed previously [35]. Despite the easy and simple procedure of mechanical exfoliation, this method has the disadvantages of producing graphene with small yield and a non-uniform nature. Therefore, an improvement to this method is highly needed; we now discuss the improvements to mechanical exfoliation that have been made so far. Instead of using the conventional Scotch tape method, the transfer printing technique using a polydimethylsiloxane (PDMS) stamp was explored by Jayasena et al., and a larger area of graphene (area of 12 × 12 mm2) was obtained; however, the film thickness was limited to tens of nanometre [36]. The modification to mechanical exfoliation by annealing the graphiteloaded tape before being transferred to the substrate was able to obtain 20–60 times greater total graphene area compared to the conventional mechanical exfoliation method [37]. Huang et al. demonstrated the modified exfoliation method which was able to produce a high quality monolayer and a few layers of graphene with a yield 20–60 times higher than the established exfoliation method [37]. The modification to mechanical exfoliation

4.2. Chemical vapour deposition (CVD) CVD is the best method to obtain graphene thin films, as it can produce large-area, uniform and high-quality graphene. Generally, CVD is a method that involves the decomposition of carbon precursor at high temperature (~ 1000 °C), the growth of graphene on the metal or insulating substrate upon cooling, and the transfer of graphene films to arbitrary substrates such as glass for use in the device applications. The growth mechanisms of graphene by CVD depend on various factors such as the type of carbon precursors, the type and flow rate of buffer gasses, and the type of metal substrates; these factors contribute to the graphene structure of monolayers, bilayers or multilayers. In CVD, a common carbon precursor used is methane (CH4) with a mixture of buffer gasses such as hydrogen (H2) and argon (Ar). Kalita

Table 3 Evaluation of FOM on graphene based TCF with doping materials. Type of graphene

Dopant

Doping type

Rs (Ω/sq)

T (%)

Graphene Graphene Graphene 4-layers graphene FLG FLG FLG FLG FLG 8-layers graphene rGO MLG rGO 4-layers graphene rGO MLG 4-layers graphene

AuCl3 AuCl3 AuCl3 AuCl3 AuCl3 AuCl3 Au(OH)3 Au2S AuBr3 HNO3 HNO3 HNO3 SOCl2 HAuCl4 TCNQ RhCl3 Hydrazine

p p p p p p p p p p p p P p p p n

150.0 445.0 220.0 54.00 300.0 130.0 820.0 600.0 530.0 90.00 1100 500.0 40.0k 79.00 1 × 104 350.0 56.00

87.0 87.0 91.0 85.0 96.0 N.A. 93.0 86.0 95.0 80.0 91.0 96.0 64.0 N.A. 88.0 93.0 94.0

*Note: N.A. – Not Available. a FOM calculated using Eq. (6). b FOM calculated using Eq. (11). 87

a

b

1.656 × 10−3 5.582 × 10−4 1.770 × 10−3 3.646 × 10−3 2.216 × 10−3 N.A. 5.902 × 10−4 1.107 × 10−3 1.130 × 10−3 1.193 × 10−3 3.540 × 10−4 1.330 × 10−3 2.880 × 10−7 N.A. 2.785 × 10−5 1.383 × 10−3 9.618 × 10−3

17.43 5.874 17.75 41.24 30.47 N.A. 6.221 12.03 13.69 17.74 3.549 18.28 0.018 N.A. 0.286 14.57 107.1

FOM 2 (Ω−1)

FOM 7

References

Kim et al. [72] Kim et al. [72] Jang et al. [60] Gunes et al. [63] Kwon et al. [64] Abdullah et al. [79] Kwon et al. [64] Kwon et al. [64] Kwon et al. [64] Kasry et al. [74] Lin et al. [75] Jang et al. [60] Eda et al. [58] Krajewska et al. [76] Ishikawa et al. [65] Jang et al. [60] Bult et al. [77]


N.N. Rosli et al.

Table 4 Evaluation of FOM on graphene hybrid films. Type of graphene

Hybrid film

Rs (Ω/sq)

T (%)

GO

AgNW AgNW AgNW AgNW AgNW AgNW AgNW AgNW AgNW AgNW AgNW CuNT CuNW CuNW AgNC AuNP Cu grids Ag grids Ag grids Ag grids Ag grids Ag grids Ag grids Au grids SWNT SWNT MWNT MWNT CNT CNT CNT CNT CNT CNT

150.0 40.00 200.0 13.30 64.00 33.00 34.40 19.90 22.00 10.00 74.00 8.000 21.70 53.80 26.00 0.370 10.00 14.00 56.00 12.00 26.00 0.600 5.000 18.00 400.0 73.00 151k 8.00k 636.0 240.0 44.00 600.0 735.0 395.0

86.0 86.0 89.0 82.8 93.6 94.0 92.8 88.6 90.0 84.0 89.0 94.0 84.0 89.3 80.5 N.A. 90.0 85.0 85.0 73.0 92.1 94.0 97.0 80.0 84.0 90.0 93.0 81.0 88.0 86.0 55.0 87.0 90.0 80.0

Graphene

rGO Graphene rGO Graphene Graphene Graphene Graphene Graphene

rGO GO rGO rGO Graphene

a

b

1.475 × 10−3 5.532 × 10−3 1.559 × 10−3 1.139 × 10−2 8.065 × 10−3 1.632 × 10−2 1.377 × 10−2 1.497 × 10−2 1.585 × 10−2 1.749 × 10−2 4.214 × 10−3 6.733 × 10−2 8.060 × 10−3 5.994 × 10−3 4.395 × 10−3 N.A. 3.487 × 10−2 1.406 × 10−2 3.516 × 10−3 3.581 × 10−3 1.689 × 10−2 8.977 × 10−1 1.475 × 10−1 5.965 × 10−3 4.373 × 10−4 4.776 × 10−3 3.205 × 10−6 1.520 × 10−5 4.379 × 10−4 9.221 × 10−4 5.757 × 10−5 4.140 × 10−4 4.744 × 10−4 2.718 × 10−4

16.04 60.16 15.71 143.2 87.59 181.8 143.9 151.8 158.3 206.9 42.45 749.9 95.36 60.18 63.28 N.A. 348.5 159.1 39.76 92.18 172.6 9998 2456 88.72 5.173 47.73 0.033 0.212 4.490 10.03 12.30 4.357 4.741 4.043

FOM 2 (Ω−1)

FOM 7

References

Yun et al. [94] Yun et al. [94] Yun et al. [94] Xu et al. [108] Kholmanov et al. [95] Lee et al. [96] Lee et al. [98] Lee et al. [99] Deng et al. [100] Deng et al. [100] Domingues et al. [109] Deng et al. [100] Zhu et al. [110] Ahn et al. [50] Sun et al. [101] Ho et al. [111] Ho et al. [59] Cha et al. [61] Cha et al. [61] Kahng et al. [112] Kang et al. [113] Gao et al. [114] Gao et al. [114] Qiu et al. [115] Zheng et al. [116] Gorkina et al. [107] Kim and Min [104] Hong et al. [105] Tung et al. [117] Tung et al. [117] Tung et al. [117] Tung et al. [117] Li et al. [106] Gan et al. [118]

*Note: N.A. – Not Available. a FOM calculated using Eq. (6). b FOM calculated using Eq. (11).

in CVD-grown graphene such as copper (Cu), nickel (Ni), cobalt (Co), etc. The growth of graphene by CVD on different substrates of metal and insulator was reviewed by Chen et al.: the mechanism of graphene growth depends on the carbon solubility of metal substrates. High carbon solubility prefers catalytic growth, whereas metal substrates with low carbon solubility prefers surface adsorption [40]. Low carbon solubility metal substrate such as copper tends to produce monolayer graphene due to their self-limiting factor; however, bilayer or multilayer graphene could be produced on Cu substrates by controlling a few factors such as the pressure at which the penetration growth mechanism occurs instead of surface adsorption. On the other hand, multilayer graphene (MLG) on Cu substrates was obtained by Choi et al. by repeatedly transferring the CVD-grown single layer graphene on Cu substrates onto the glass substrates [41]. While the mechanism of graphene growth on different substrates has been discussed by Chen et al. [40], there are other factors affecting the mechanism of graphene growth in CVD. The role of hydrogen for graphene growth in CVD has been investigated in many studies; particularly its parameters of pressure, flow rate and usage conditions, as they provide a different growth mechanism for graphene. The role of graphene includes: i) an etching agent and ii) a growth promoter, depending on the controllable parameter. Zhang et al. reported the etching of CVD-grown graphene on Cu substrate by the exposure of this film to hydrogen flow at high temperature, where the etching process is considered as the reverse reaction to the growth of graphene, can provide a hexagonal opening to the graphene and crystal orientation of graphene could be determined [42]. The role of hydrogen in CVD for graphene synthesis was further investigated by Zhang et al., as it has the

Fig. 1. 2D graphene-constructed the other dimensional, (a) wrapped 0D bulkyball (fullerene); (b) rolled 1D nanotubes and (c) stacked 3D graphite. Reproduced with the permission of Nature Publishing Group [32].

et al. reported on the use of botanical derivative camphor (C10H16O) as a carbon precursor which has a very good influence on the structure of the graphene formed [38,39]. Various metal substrates have been used 88


N.N. Rosli et al.

Fig. 2. The schematic diagram for reduction of graphene oxide to produce reduced graphene oxide [33].

the residual oxide that contributes to the uncontrollable nucleation and growth. In this study, carbon dioxide was used in the pre-treatment of copper substrates to remove carbon residues and enhance graphene growth using carbon dioxide/methane, producing high quality graphene films with a high mobility of 1975 cm2/Vs. Graphene with good quality, uniform growth and having Rs of 0.92 ± 0.07 kΩ/sq was obtained for carbon dioxide/methane growth in CVD. Low-pressure chemical vapour deposition (LPCVD) on Cu foils was reported by Yin et al. to produce high-quality graphene with transmittance > 95% [49]. The sheet resistance of the as-synthesised graphene was not mentioned but by doping with AuCl3 and the layer-bylayer (LbL) transfer method, the sheet resistance was reduced to as low as 50–100 Ω/sq. Low-temperature plasma enhanced chemical vapour deposition (LT-PECVD) at a relatively low temperature of 400 °C, firstly reported by Ahn et al. [50]. The transfer method of CVD-grown graphene is required; commonly, this method includes depositing a supporting layer of polymethylmethacrylate (PMMA) on the graphene, etching away the metal substrates using an appropriate chemical etchant, transferring the PMMA/graphene to a transparent substrate such as glass, and washing away the PMMA layer by sequential acetone wash. The growth of graphene by CVD and transfer process to the substrates produces structural defects of wrinkles, ripples and fold promoting electron scattering, thus weakening electron ballistic transport and carrier mobility. The wrinkles or defects results from the thermal stress due to the rapid cooling during the growth of graphene; these wrinkles may be released or preserved depending on the transfer method [51]. In addition, the yield of graphene is also affected during the transfer process. Therefore, modification of CVD parameters, especially the temperature setup; and the transfer process were necessary in order to overcome the drawbacks of this synthesis method. The enhancement of the transfer method will be discussed in Section 5.2.

function of blocking the decomposition of methane, regulating the diffusion of carbon on the surface of copper substrate, catalysing the methane adsorption, and etching the edges of graphene by controlling the flow rate of hydrogen [43]. As single layer graphene (SLG) is mostly obtained by CVD graphene on the Cu substrate due to the surface adsorption that leads to selflimiting growth [44], the hydrogen effect on the growth of bilayer graphene (BLG) or few layer graphene (FLG) on Cu substrates is investigated. The pressure of hydrogen has different effects on the growth of graphene. At the low pressures, single-layer graphene growth is favoured due to the passivation of the graphene edges by Cu surface, allowing C adsorption for sustaining the nucleation and growth of adlayer graphene; whereas at high hydrogen pressures, bilayer or fewlayer graphene is favoured as the graphene edges are terminated by H atoms allowing C diffusion to the area below the graphene top layer [45]. In another study by Hu et al., the H2 pre-dissolution could suppress the carbon dissolution in the Cu film and enhance the diffusionout of dissolved carbon atoms [46]. The effect of deposition time and hydrogen flow rate on the hot filament thermal chemical vapour deposition (HFTCVD) for graphene synthesis was observed by Hafiz et al. [47]. The Cu2O nano-dots formed from the reaction between copper vapour in equilibrium and oxygen and acted as the nucleation sites for graphene growth. At low hydrogen flow rate (i.e. high growth rates), graphene growth started at the borders of nucleation sites, where an exposed Cu2O nano-dot surface was left for the next graphene layer deposition and stopped as the Cu2O nano-dot surface was covered completely by the graphene layer. For a slow growth rate, the stable graphene cap floated over the Cu2O nanodot surface and the chemically active graphene lattice border became the incorporation sites for carbon atoms, hence aggregating a graphene layer. The etching rate of graphene depends on the hydrogen flow rate which it will increase with increasing hydrogen flow rate. On the contrary, a carbon atmosphere for CVD-grown on Cu substrate was demonstrated by Strudwick et al. [48], as hydrogen use in pre-treatment or during graphene growth does not completely remove

Fig. 3. Summary on common methods for synthesis of graphene. 89


N.N. Rosli et al.

4.3. Solution processing

5. Recent approaches in graphene deposition

Solution processing is a method of synthesising graphene oxide or graphite oxide and reducing GO to produce rGO. The first graphitic oxide was obtained by Hummers and Oppeman [52], and later some modifications were made. Graphene oxide was obtained from chemical oxidation of graphite by using the modified Hummer's method [53,54]. In this method, the graphite flake is suspended in a solution of potassium permanganate (KMnO4), sodium nitrate (NaNO3) and sulphuric acid (H2SO4). The GO films obtained from the modified Hummer's method are deposited onto transparent substrates before undergoing a reduction process. In addition, graphite oxide can also be formed by the suspension of graphite in water and further sonication [55]. The reduction of graphene oxide was discussed in detail by Pei and Cheng [56]. The reduction of the GO film is necessary to make the insulating GO films become conductive due to the restoration of the sp2 bond of graphene. In general, reduction of graphene includes chemical reduction, annealing at high temperature, or both. Various methods of reduction have been successfully performed such as exposing GO films to hydrazine vapour [54], thermal treatment with Ar and/or H2 flow [57], vacuum annealing [53], and a combination of hydrazine vapour and Ar annealing [53]. Prior to the reduction process, graphene oxide requires a transfer process to the transparent substrates such as glass and polyethylene terephthalate (PET). Various methods of transfer have been reported, such as vacuum filtration, dip coating, spraying etc. Vacuum filtration is used to deposit the suspended GO from chemical exfoliation using cellulose ester membrane before being transferred to the transparent substrate [58]. The transferred GO incorporated into cellulose ester membrane is then sequentially washed in acetone to etch the cellulose ester membrane, leaving the GO films on the glass substrates. The thickness of GO may be varied by controlling by filtration volume [58]. Gilje et al. [55] use the spray method of GO onto the preheated substrate to freeze the GO platelets upon reaching the substrate after evaporation of the solvent in order to reduce the agglomeration of GO platelets. The GO platelets agglomeration is not good for film formation, as it causes non-uniformity of the film and affects the electrical properties of the films. The spraying method is a very quick, as it is able to produce 100% yield of graphene sheets in a few hours; and the scaleup process to deposit a graphene sheet is applicable to any number and sizes of substrate, including pre-patterned electrodes [55]. The performance of graphene-based TCF is highly influenced by the method of graphene synthesis. The sheet resistance and transparency of graphene synthesised by various methods with FOM to evaluate the performance of graphene-based TCF are summarised in Table 2. The FOM for graphene-based TCF by various methods was lower compared to the benchmark FOM of 3.487 × 10−2 Ω−1 and ≈ 350, obtained by Eqs. (6) and (11), respectively. The lower FOM was obtained due to the high Rs and low T of the graphene films caused by some defects that formed during the synthesis, particularly the graphene synthesised by solution-processing which showed the lowest FOM compared to the CVD-synthesised one. Among the literature reviewed, the lowest FOM is 1.920 × 10−7 Ω−1 and 0.011 obtained for solution-processed rGO film by Eda et al. [58] which was due to the high Rs of 70 kΩ/sq and low T of 65%. The graphene films demonstrated by Wu et al. [53], showed a slightly higher FOM of 5.987 × 10−6 Ω−1 (FOM 2) and 0.073 (FOM 7) compared to graphene films by Eda et al. [58] despite having higher Rs. The highest FOM was obtained due to the graphene film of Wu et al. having a higher T of 95%. Most CVD-synthesised graphene films [39,59–64] showed high T of > 90%; however, due to the high Rs, these films do not exceed the benchmark FOM.

Graphene is known for its excellent properties and being suitable for various applications. Nonetheless, various factors have degraded the performance of graphene, such as the synthesis method, transfer process etc. Hence, solutions to overcome these drawbacks are highly recommended. In the past few years, various modifications to improve graphene (particularly to reduce the defects caused by graphene synthesis and the transfer method) have been made. These modifications include the doping of graphene, enhancement of the graphene transfer method and hybrid film of graphene with other materials. 5.1. Doping Graphene is a zero band gap material; its properties may be improved through the opening of the band gap of graphene and tuning the conducting type by substituting carbon atoms by foreign atoms. By introducing or doping impurities to the graphene, the type and concentration of charge carriers in graphene (namely electrons and holes), can be controlled. Over the years, various doping methods have been applied such as chemical doping, surface doping, etc. The mechanism of surface doping for graphene through electronic and electrochemical doping has been discussed previously by Pinto and Markevich [68]. In a review by Guo et al., the doping of graphene was classified into hetero atom doping, the chemical modification strategy and the method of electrostatic field tuning [69]. Doping of graphene by wet and chemical deposition was previously reviewed by Oh et al. [70]. However, apart from the improved carrier mobility of graphene by doping, the graphene films are still unstable for a long period and do not directly refine the defects of graphene; thus, the problem remains. Kang et al. demonstrated the doping of CVD graphene using the common dopants of FeCl3, SnCl2, IrCl3, RhCl3 and AuCl3, which successfully reduced Rs of the graphene film except for IrCl3 doping [71]. The transmittance of the graphene films was increased after exposure to the air for 200 h, particularly graphene doped-AuCl3 which had a 1.9% increment. The increased transmittance is attributed to the time-dependent desorption of physisorbed dopants, which is activated by ambient thermal excitation. The p-type doping for graphene TCF using AuCl3 in nitromethane was demonstrated by Kim et al., which was able to reduce Rs of the graphene films by up to 77% depending on the doping concentration [72]. The graphene film was synthesised by CVD on Ni substrate producing one- or two-layer graphene, transferred by the PMMA method and subsequently doped with AuCl3 in solvent of nitromethane. The lowest Rs of 150 Ω/sq at a transmittance of 87% was obtained for 10 mM AuCl3 doping concentration. However, the removal of PMMA by acetone washing has increased the Rs of 846 Ω/sq at 89% transparency and 445 Ω/sq at 87%. In other research, the trilayer CVD-grown graphene showed a ~ 66% improvement in Rs, which reduced from 466.1 Ω/sq to 158.5 Ω/sq at a transmittance of 65% after doping with AuCl3 [73]. LbL doping of AuCl3 doping to CVD-grown graphene films was proposed by Gunes et al. [63]. By using this approach, a uniform and environmentally stable graphene films with the low sheet resistance of 54 Ω/sq and high transmittance of 85% and a distinguished bending stability were achieved. Kwon et al. [64] observed the effect of Au anions for graphene doping using several dopants of Au(OH)3, Au2S, AuBr3 and AuCl3. The graphene films were prepared by CVD, producing few layer graphene (FLG) and spin-coated with dopants Au complex powders of Au(OH)3, Au2S, AuBr3 and AuCl3. The doping of AuCl3 showed the greatest reduction in the Rs, from 950 Ω/sq to 300 Ω/sq at a high transparency of 96% due to the higher electronegativity between C and the anions. Thermal annealing increased Rs and reduced the transmittance of these films, particularly when doping with AuCl3, which suffered the most degradation in the performance when annealed at 400 °C due to the Cl anions. 90


N.N. Rosli et al.

graphene film was obtained by Bult et al. [77] with an FOM value of 9.618 × 10−3 Ω −1 (FOM 2) and 107.1 (FOM 7) as the graphene film had quite a low Rs of 56 Ω/sq at high T of 94%, yet it was still below the benchmark FOM.

The doping of graphene by nitric acid (HNO3) also exhibits p-type characteristics as the electron is transferred from graphene to nitric acid. Kasry et al. demonstrated HNO3 doping for eight-layer stacked graphene; each layer of the graphene was doped with HNO3 [74]. The graphene films could be doped by two approaches; doping of each layer after stacking (interlayer-doped) or doping the last layer after stacking (last-layer-doped). The interlayer-doped graphene showed better performance, with a low Rs of 90 Ω/sq at 80% transparency, as the doping effectively reduced Rs by a factor of 3. The interlayer-doped and lastlayer-doped graphene exhibited similar Rs due to the possibilities of HNO3 intercalating into the graphene stacks or dopants evaporating before the addition of the other layer. rGO prepared by solution processing, deposited on flexible substrate by Langmuir-Blodgett (LB) assembly, reduced by hydriodic acid (HI) and chemically doped with HNO3 and thionyl chloride (SOCl2) was demonstrated by Lin et al. [75]. After reduction by HI acid, the rGO exhibited Rs in the range of 2.4–14.4 kΩ/sq and was further decreased to 1.1–6.2 kΩ/sq due to chemical doping. Krajewska et al. explored the p-doping of CVD-grown graphene using tetrachloroauric acid (HAuCl4) as the precursor to AuNP and found that the higher concentration of HAuCl4 caused an increase in the carrier concentration and decrease in the Rs [76]. By doping with 30% concentration HAuCl4 for graphene film on the SiO2/Si substrate, Rs reduced by about 75%, the carrier concentration increased up to 5.8 × 1013 cm−2 and the carrier mobility decreased to 409 cm2/V. The lowest Rs of 79 Ω/sq was obtained for four-layer graphene on the PET substrate after doping with HAuCl4 at 30% concentration. P-type doping of tetracyanoquinodimethane (TCNQ) for graphene showed an improved resistivity without degradation of optical transparency [65]. The rGO films without doping exhibited high Rs of 1 × 106 Ω/sq at 88% transmittance, which was further decreased by two orders of magnitude without changing the transmittance of the films. In other research, Jang et al. demonstrated the doping of graphene using three graphene dopants of HNO3, AuCl3, and RhCl3, where the graphene with RhCl3 showed the most stable Rs compared to the other doped graphene due to the formation of metal nanoparticles as a protective barrier [60]. Rs of the graphene film were reduced from 730 Ω/sq to 350 Ω/sq; however, the transmittance decreased from 97.7% to 93% after doping with RhCl3. In addition, there was a slight decreased in the hole mobility, from 3850 cm2 V−1 s−1 to 2884 cm2 V−1 s−1 after doping with RhCl3. N-type graphene could be achieved by doping with polyethyleneimine (PEI) or hydrazine, as demonstrated by Bult et al. [77]. The graphene films were synthesised by CVD, producing a monolayer and four-layer graphene having Rs of ~ 400 Ω/sq and 96 Ω/sq, respectively. Rs of four-layer graphene reduced to ~ 56 Ω/sq after doping with hydrazine. The doping with hydrazine produced n-type graphene as the hydrazine provided sufficient density of electrons into the graphene films. The PEI-doped graphene had a similar performance with hydrazine-doped graphene; however, the PEI-doped graphene was more stable, as it acted as the thin overcoat. The other common material for graphene doping is nitrogen. In a study by Sandoval et al., rGO was doped by annealing the graphene oxide in a flow of nitrogen, which results in rGO with high thermal stability against oxidation by air [78]. The performance of graphene-based TCF after doping is evaluated by FOM and was summarised in Table 3. The performance of graphene TCF was evaluated by using a benchmark FOM of 3.487 × 10−2 Ω−1 and ≈ 350, obtained by Eqs. (6) and (11), respectively. Even though, the doping method has improved the Rs and T of the graphene films, its performance has yet to achieve the required TCF performance of Rs ~ 10 Ω/sq and T > 90%, which is the benchmark FOM of TCF. Based on the literature, there are only a few studies that reported on high T (> 90%) obtained for graphene films with the doping of AuCl3 [64], Au(OH)3 [64], AuBr3 [64], HNO3 [60,75] and RhCl3 [60]; however, due to the lower Rs, the FOM of these films did not exceed the benchmark FOM. The highest FOM for doped-

5.2. Enhancement to the transfer method CVD-grown graphene is known to produce large-area graphene with better performance compared to other synthesis method; however, the as-grown graphene requires post-transfer method from the metal substrate layer before being deposited to transparent insulating substrates. The post-transfer method involves spin-coating of the supporting layer (PMMA or PDMS, thermal release tape, etc.) on the graphene/metal substrate, removing the metal substrate by etching, depositing the graphene/supporting layer onto the substrate (glass or plastic) and finally, removing the catalyst layer by annealing or acetone dissolution. The post-transfer method may have some effects on the graphene films such as defects (crack, wrinkles, etc.), unintentional doping effect and residues, which significantly affect the properties of the graphene films. Thus, enhancements of the post-transfer of the graphene film are highly required. Barin et al. [80] demonstrated improvements to the conventional PMMA method by applying a double layer of PMMA to the CVD-grown graphene. The second layer of PMMA was coated using a lower concentration of PMMA diluted solution compared to the first PMMA layer, resulting in a better quality of transferred graphene with fewer wrinkles, cracks and polymer residue. In addition, the longer time of baking process improved the adhesion of the graphene film to the substrate as well as reduced of the generated wrinkles due to the transfer films. In Kratzer et al. [81] the growth of the organic molecule, namely 6P (parahexapheny), on MLG was observed with respect to the residues of PMMA. This transfer method provides an additional step of H2 annealing prior to the deposition of 6P in order to remove the residual PMMA. The residual PMMA limited the length of the growing 6P needle structure, where by H2 annealing the PMMA residues can be reduced, consequently recovering the growth of 6P on the MLG. Another modification to the conventional PMMA wet chemical transfer was demonstrated by Deoker et al., which included the steps of spin-casting the PMMA, removing the trapped water and removing PMMA from the graphene films [82]. The PMMA was prepared by diluting the PMMA in anisole, spin casting on the one side of graphene/ Cu and removing the unprotected side of graphene/Cu by the baking process. The slow Cu etchant of (NH4)2S2O8 was used to etch away the Cu substrates which took a long time of 15 h; then, the graphene/ PMMA was rinsed with DI water. As for the removal of trapped water, the water was naturally dried for 10 min, capillary action by the clean wiper room removed the trapped interfacial water as it was placed at the graphene edges, and the remaining trapped water droplets were removed by the baking process. Finally, the PMMA was removed in sequence steps of exposure in deep UV light, dissolving in analytical reagent of methyl isobutyl ketone (MIBK) followed by warm acetone, rinsing in isopropyl alcohol and drying with a mild nitrogen blow. No structural defects were detected and high quality graphene transfer was successfully carried out by using the modified wet chemical transfer process. Instead of PMMA, another polymer can also be used in graphene transfer. A comparison of polymers was made by Antonova using PMMA, PDMS, thermoscotch and polycarbonate [83]. Among these polymers, polycarbonate showed the best promising material for use as the supporting layer in the graphene transfer as it provided the cleanest transferred graphene with minimum resistance in the range of 250–900 Ω/sq and maximum carrier mobility in the range of 900–2500 cm2/V cm. In a different study, graphene films were successfully transferred to the flexible surface by using PDMS and an etching-free method to remove the Cu substrate from graphene [84]. The peeling of graphene from the Cu substrate was possible due to the 91


N.N. Rosli et al.

holder [93]. The transferred graphene by this method was free from organic residue which led to improved electronic properties having lower Rs compared to polymer-transferred graphene.

stronger adhesion force between PDMS and graphene. In addition, this etching-free method by peeling off can avoid damage to the graphene films. Ma et al. [85] demonstrated the etching-free method by intercalation of carbon monoxide (CO) under ambient pressure which can weaken the interaction between graphene and metal substrates, and the transfer of graphene through stamping using PDMS or peeling-off graphene using water as a peeling agent. Marta et al. reported an improvement to the transfer method by dry transfer utilising a thick polyvinyl alcohol (PVA) to remove graphene films from Cu substrates without reducing the quality of the CVD-grown graphene films [86]. The delamination approach using hexagonal boron nitride (hBN) to transfer CVD-grown graphene from Cu foils to the arbitrary substrate was proposed Banszerus et al., by considering the van der Waals interaction between hBN and graphene, the transfer of graphene was successfully carried out [87]. The hybrid film of graphene and polymer was reported by Jung et al. to improve the conductivity of graphene films through a modification to the substrate by self-assembled monolayers (SAM) [88]. In that study, CVD-synthesised graphene was combined with SAM of 3-Aminoproplytriethoxysilane (APTES) polymers on the PET flexible substrate, producing defect-free graphene/polymer hybrid films. The self-supporting method (SSM) of layer-stacked graphene was proposed by Ning et al., where the smoother surface morphologies was achieved by five-layer-stacked-SSM graphene films with a lower Rs of 20.16 Ω/sq and transparency of ~ 88% [89]. The method included the steps of spin-coating PMMA on the graphene; etching away the Cu substrates; transferring the PMMA/graphene onto the second layer of graphene/Cu; repeating the etching and transferring of a new layer of graphene to obtain few layers-stacked graphene; and finally, removing the PMMA. Graphene with a smooth surface and lower Rs as the number of layers incr eased, it can be achieved by this transfer method. The coupling effect of the SSM-stacked graphene can be enhanced by acid treatment. A novel method for simultaneously transferring and doping CVDgrown graphene using a flouropolymer was first introduced by Lee et al. [90]. The flouropolymer was used as an alternative for PMMA and still needed to be removed; the residue of this polymer acted as a dopant to the graphene. The doping of the fluoropolymer to graphene occured due to the arrangement of fluorine atoms on the graphene basal plane, which was activated after thermal annealing or soaking in solvent. As a result, a monolayer graphene film with Rs of ~ 320 Ω/sq was obtained for CVD-grown graphene with the simultaneous transfer and doping of fluoropolymer. Castro et al. proposed a PMMA-free transfer method by using a polymer blend of ethylene-propylene-diene monomer (EPDM) and polyaniline (PAni) for both the transfer and doping of graphene without any removal process, and can be used directly in organic electronic devices as a transparent electrode of graphene/EPDM-PAni [91]. Graphene/EDPM-PAni exhibited improved electronic properties compared to the graphene transferred by the conventional PMMAmethod with a high transparency of 95% and having comparable performance to ITO/glass-based devices when incorporating this transparent electrode into organic devices. The polymer-free transfer method or direct transfer was demonstrated by Regan et al. using an amorphous Carbon (a-C) transmission electron microscopy (TEM) grid, which is a cleaner method compared to the conventional transfer method using a polymer [92]. In addition, the mechanical damage can be avoided as the a-C grid film was used to support graphene: in the early step, both graphene and the grid's a-C were put in close contact as the isopropyl alcohol (IPA) evaporates and the films adhered to each other. By using the a-C TEM grid films, complete grid coverage, a highly bonded graphene-grid support film, and a low contamination of graphene surface could be achieved. Later, Lin et al. made some modifications to the a-C TEM grid transfer method by mixing the IPA solution with water to reduce the surface tension of the solution and reduce the external force as well using a graphite

5.3. Hybrid films of graphene-based TCF Graphene films and other materials were combined to form a hybrid film in order to improve the performance of graphene as TCF. The common materials used together as hybrid films with graphene are 1D metal nanostructures and 2D carbon nanotubes. Hybrid films aim to improve the electrical and optical performance of the films by (i) covering the defects of the graphene film due to synthesis methods and transfer process; (ii) reducing the junction resistance between nanowires; (iii) providing a conductive pathway for charge transport for both films; and (iv) occupying the voids or empty spaces in the structure of the other films. Flexible and stable graphene oxide and silver nanowires (AgNW) hybrid film having a low Rs of 150 Ω/sq at 86% transmittance were achieved by Yun et al. [94]. The performance of these films was improved due to the annealing of GO and AgNW improving the adhesion of both films as well as connecting the junction between the wires, which significantly reduced the junction resistance between the wires of AgNW. Kholmanov et al. demonstrated the use of nonconductive AgNW to provide conductive pathways for charge transfer of graphene by acting as a bridge to the line defects and grain boundaries of graphene films resulting in a low Rs of 64 ± 6.1 Ω/sq at a high transparency of 93.6% [95]. The performance of these films was solely attributed to the conductive graphene as the density of AgNW was below the percolation threshold, which made it non-conductive. The lowest Rs of 24 ± 3.6 Ω/sq was obtained by adding one more layer of graphene film, but the transmittance was slightly reduced to 91%. In another study by Lee et al., hybrid films with a low sheet resistance of 33 Ω/sq, high transmittance of 94% in the visible range, high flexibility of 27% bending strain and highly stretchable with 100% tensile strain were achieved by integrating graphene with high densities AgNW [96]. The high conductivity of these films was due to the charge carriers contributed by both films simultaneously being transported through conductive pathways provided by AgNW bridges over graphene defects and the empty spaces of AgNW being filled with graphene. Choi et al. showed an enhancement of 30% electrical conductance due to the integration of AgNW and thermal annealing, which bridged the graphene domain and reduced its contact resistance, respectively [97]. Graphene films also act as the encapsulation layer to reduce the oxidation of AgNW since graphene is impermeable to gases, which may improve the conductivity as well as providing greater stability to the hybrid films. As the stability of AgNW was improved by graphene layers, graphene/AgNW hybrid films and sandwich structure of graphene/AgNW/graphene having a sheet resistance of 34.4 ± 1.5 Ω/sq at 92.8% transmittance and 19.9 ± 1.2 Ω/sq at transparency of 88.6% were obtained, respectively [98,99]. The best performance of graphene/AgNW hybrids film was obtained by Deng et al. with the lowest Rs of ~ 8 Ω/sq at 94% transmittance with other outstanding properties of high mechanical flexibility and chemical stability [100]. The FOM for graphene/AgNW hybrid films also exceeded the benchmark FOM which was 6.733 × 10−1 Ω−1 and 749.9 obtained by Eqs. (6) and (11), respectively; this indicated the high performance of the TCF. In addition, the fabrication method of roll-to-roll production is feasible for mass production as proposed in the research. Sun et al. used the Ag nanocomposite of nanowires and nanoparticles in order to minimise the defects of graphene films [101]. By using the silver nanocomposite (AgNC), the conductivity of hybrid films was improved by the enhancement to the interfacial contact with graphene and nanocomposites. The sheet resistance of the hybrid films of graphene and AgNC obtained was 26 Ω/sq, which was same as with the perfect graphene having sheet resistance 30 Ω/sq. On the other hand, 92


N.N. Rosli et al.

increasing concentration of MWNT. The author concluded that MWNT posed a conductive bridge at high surface densities of 60 μg/mL, which contributed to the better performance of the hybrid films. The hybrid films of rGO nanosheets with MWNT proposed by Hong et al. underwent heat treatment, which was able to increase the conductivity as the sp2 bond of graphene was restored [105]. In addition, the conductivity of the hybrid films also depends on the number of layers for graphene nanosheets, as the sheet resistance decreases with the increasing number of graphene layers, attributed to the increasing of graphene connectivity with CNT networks. The method of solidphase layer stacking, proposed by Li et al., was able to obtain the hybrid films of graphene and CNT with a comparable performance of the graphene/CNT film with additional doping processes [106]. The void of the CNT network was occupied by graphene sheets and formed a homogenous film, consequently improving the performance of graphene/CNT films. Currently, the best performance for rGO/CNT hybrid films with the lowest Rs of 73 Ω/sq at 90% transmittance (including other advantages of the method preparation being simple and applicable for mass-production), was demonstrated by Gorkina et al. [107]. In summary, the integration of graphene with other materials has a significant effect on the optoelectronic properties; most hybrids films have an improvement to their properties and are then comparable to the ITO TCF. In addition, the hybrid film also offers other advantages of being flexible and stable over a longer period. Nonetheless, the properties of the hybrid films may vary due to some factors such as the synthesis methods of graphene, the preparation of the other materials, the method to prepare the hybrid films etc. The optoelectronic properties of graphene hybrid films are summarised in Table 4. The performance of graphene hybrid films was evaluated by FOM as the high performance TCF film should surpassed 3.487 × 10−2 Ω−1 and ≈ 350 for FOM 2 and FOM 7, respectively. Gao et al. demonstrated the high performance hybrid film of graphene and Ag grids having low Rs of 0.6 and 5 Ω/sq obtained at 94% and 97% diffusive transmission, respectively [114]. The integration of Ag grids improved the conductivity; initially graphene had a higher Rs of 1000 Ω/sq while the

Yin et al. [49] demonstrated the use of AgNW to improve the contact layer of graphene films as the transparent electrode in a CIGS solar cell due to poor contact between graphene with CdS layer. The sheet resistance of PMMA/graphene/AgNW has reduced to < 30 Ω/sq from 50 to 10 Ω/sq, consequently, improving the performance of CIGS solar cells with a power conversion efficiency of 9.65%. Copper nanowires (CuNW) are an alternative material to AgNW TCF due to their excellent properties of high conductivity, cheap cost and abundance of material source compared to silver. However, copper suffers from oxidation at ambient atmosphere, which may degrade the performance of CuNW as TCF. Previous studies reported the methods to reduce the oxidation of copper such as the passivation coating of AZO for copper nanofibers (CuNF) and coating of the Ni shell for CuNW [102,103]. Having the same purpose to increase the chemical and thermal stability of CuNW, hybrid films of graphene/CuNW were proposed by Ahn et al. [50]. CuNW/graphene TCF with a slight increase of less than 9% for Rs were obtained after 30 days of thermal oxidation test under ambient condition; initially the films had a Rs of 53.8 Ω/sq at 89.3% transmittance [50]. CNT is being one of the alternatives for TCF that exhibit excellent properties, especially conductivity. However, the high Rs of CNT is a drawback for this material to be applied as TCF. CNT may consist of SWNT and MWNT; both structures exhibit different properties. The double layer of the large area of graphene films and carbon nanotubes was successfully fabricated by Kim and Min [104]. In their research, reduced graphene oxide was synthesised by the Hummer's method and as-prepared rGO-coated substrates were immersed in suspensions of aqueous aminated MWNT and subsequently reduced with hydrazine to produce GO and rGO/MWNT thin films. By varying the concentration of hydrazine, the obtained sheet resistances for GO and rGO films, were in the range of 30 × 106–11.5 × 106 Ω/sq. The conductivity of the double layer of GO and rGO/MWNT was further improved by annealing in Ar atmosphere; Rs significantly reduced to 990–280 and 265–103 kΩ/sq, respectively. Finally, the lowest Rs of 151 kΩ/sq at 93% transmittance was obtained as Rs significantly reduced with the

Fig. 4. Graphene-based transparent electrode for solar cell applications; (a) inorganic solar cell (TFSC, HJSC and PSC); (b) organic solar cell (OSC and OPV); and (c) DSSC. Reproduced with the permission of Nature Publishing Group [122]. 93


N.N. Rosli et al.

The general structure of thin film solar cell (inorganic solar cell) is shown in Fig. 4(a) with an intrinsic layer sandwiched between the pand n- layer. Copper indium gallium selenide (CIGS), cadmium telluride (CdTe) and amorphous silicon (a-Si) are the types of leading thin film technologies which have been commercialised in building integrated photovoltaics (BIPV) and consumer product (a-Si only). In recent years, there are many emerging thin film solar cell technologies such as copper zinc tin sulfide (CZTS) solar cell, PSC and quantum dots (QD) solar cell. CIGS solar cell consists of contact (Ni/Al), window layer (i.e. TCO, mostly AZO), n-type layer (i-ZnO), buffer layer (n-type CDS), absorber layer (p-type CIGS), back contact (molybdenum) and substrate (i.e. glass, metal foil or plastic). The use of graphene as a transparent electrode to replace AZO for CIGS solar cells was evaluated by Yin et al. [49]. The synthesised method of LPCVD and further doping with AuCl3 doping and LBL transfer method, resulted in a four-layer graphene with low Rs in the range of 50–100 Ω/sq. The PMMA/graphene was applied to CIGS solar cell as the front electrode, obtaining the efficiency of 7.74%, however, after the removal of PMMA, the PCE dropped to 0.3%. AgNW was applied to the CIGS solar cells as the interface layer between graphene and i-ZnO, a higher PCE of 9.65% was obtained. The modification by AgNW has improved Rs from 50–100 Ω/sq to < 30 Ω/sq which contributed to the good performance of CIGS solar cells. High PCE of 13.5% was obtained for large area CIGS solar cell (~ 45 mm2) by 4 layered AuCl3-doped graphene transparent electrode with a thin layer of PMMA [124]. Instead of removing PMMA during the CVD-synthesised graphene transfer process, the PMMA was made into a very thin layer by spin coating and drying in air at room temperature, resulting in better surface contact between PMMA/graphene and i-ZnO layer. By incorporating PMMA/graphene as transparent electrode, the

integration of graphene improved the thermal stability and flexibility of Ag grids. Due to low Rs (0.6 Ω/sq) and high T (94%), graphene/Ag grids hybrid films showed higher FOM compared to the benchmark FOM. The FOM calculated for the hybrid graphene/Ag grids film were 8.977 × 10−1 Ω−1 and 9998, obtained by Eqs. (6) and (11), respectively. The other hybrid films that surpassed the benchmark FOM were the graphene-copper nanotrough (CuNT) with low Rs of ~ 8 Ω/sq and T of 94%, as demonstrated by Deng et al. [100]. The FOM calculated by Eqs. (6) and (11) for the graphene/CuNT hybrid films were 6.733 × 10−2 Ω−1 and 749.9, respectively. The improved performance of the graphene/CuNT hybrid films was attributed to the reduction of CuNT junction resistance and graphene grain boundaries. Due to the combination of graphene and CuNT films, the other properties of corrosion resistance and mechanical flexibility also improved. On the other hand, the graphene/Cu grids hybrid films by Ho et al. [59] showed equal FOM to the benchmark as the films exhibited Rs of 10 Ω/sq at T of 90%, which was also comparable to the performance of ITO. The thermal stability of the hybrid films improved due to the graphene acting as the oxidative barrier for the Cu grids. A uniform conducting film was formed as the graphene filled the void of the Cu grids film, which improved the electrical properties of the hybrid films. In addition, the uniform film is highly needed, as it can reduce the series resistance in the solar cell application. 6. Graphene-based transparent conductive films for photovoltaic application Graphene TCF or graphene transparent electrode is a promising replacement for conventional material to various devices including solar cells, touch screens [119], LED (organic and inorganic) [120,121], etc. The requirement for each device is differs, where the low Rs and T are necessary for solar cells whereas for LED application, high conductivity is more crucial than optical transmittance. Since we have emphasized on low Rs and high T graphene TCF in the earlier parts of this paper, we will only discuss the application of graphene TCF as the front contact of solar cells as both properties (Rs and T) are essential for them. For solar cells, graphene may be applied as a transparent electrode or as a charge carrier in the active layer. As the transparent electrode or also known as front window of the solar cell, TCF require high optical transparency of the layer to allow light to pass through to the active layer without unwanted absorption of the solar spectrum and low resistivity to facilitate the electron transfer process and reduces energy losses. Graphene-based TCF is applied as front contacts in various types of solar cells; namely, thin film solar cells (TFSC), heterojunction solar cells (HJSC), organic solar cells (OSC), organic photovoltaic (OPV), dye-sensitised solar cells (DSSC) and perovskite solar cells (PSC). Graphene-based transparent electrode for solar cell applications are shown in Fig. 4; representing the general structure of (a) inorganic solar cell (TFSC, HJSC and PSC); organic solar cell (OSC and OPV); and (c) DSSC. The performance of solar cell incorporating graphene as TCF is summarised in Table 5. The properties of Rs and T for TCF differ according to the applications; the lowest Rs and high transparency are highly desired. Rs may vary from 10 Ω/sq to 106 Ω/sq, but typically, transparency must be ≥ 90%. For solar cells, a low sheet resistance of ≤ 10 Ω/sq is required, which is as low as possible to reduce power loss in the solar cell. The performance of graphene-based TCF is highly influenced by the method of graphene synthesis. For high transparency, the sheet resistance of graphene is too high and may degrade the performance of solar cells. Conventional thin film solar cell comprises of transparent electrode, window layer (n-type semiconductor), absorber layer (p-type semiconductor), back contact (mostly, Al) and substrate, in which the contact between n-type and p-type semiconductor forms the main junction.

Table 5 Performance of graphene-based TCF for solar cells application. Type of solar cell

Type of graphene

Voc (V)

Jsc (mA/cm2)

FF (%)

η (%)

References

TFSC TFSC

Graphene Graphene

0.41 0.82

20.06 9.30

34.00 0.31

2.81 2.81

TFSC HJSC

Graphene Graphene/ AgNW CNT/ Graphene Graphene/ CNT Graphene/ AuNP MLG Graphene

0.60 0.61

32.40 9.94

69.10 54.40

13.5 3.30

Bi et al. [62] Veronese et al. [123] Yin et al. [124] Lee et al. [98]

0.44

24.80

24.60

2.70

Li et al. [106]

N.A.

N.A.

N.A.

8.50

Gan et al. [118]

0.56

27.70

79.00

12.30

Ho et al. [111]

0.52 0.54

6.91 4.82

32.60 26.00

1.17 0.68

0.48 0.58

2.10 3.47

0.34 42.10

0.40 0.85

0.58 0.56 0.60 0.66 0.74

9.20 3.00 28.40 1.43 6.90

38.00 30.00 78.40 0.33 0.41

2.00 0.33 13.30 0.31 2.10

DSSC

rGO Graphene/ CNT Graphene rGO GO/AgNW rGO Graphene/Ag grid rGO

0.70

1.01

0.36

0.26

DSSC

Graphene

N.A.

N.A.

N.A.

2.00

PSC PSC PSC PSC PSC

Graphene Graphene Graphene Graphene Graphene

0.95 N.A. 1.01 1.08 0.96

17.75 N.A. 19.65 21.00 21.20

71.72 N.A. 67.27 76.60 0.70

12.03 11.50 13.35 17.40 14.20

Choi et al. [41] Kalita et al. [38] Wu et al. [53] Tung et al. [117] Cha et al. [61] Yun et al. [125] Xu et al. [108] Yin et al. [126] Kahng et al. [112] Wang et al. [57] Selopal et al. [127] You et al. [128] Liu et al. [129] Kim et al. [130] Heo et al. [131] Jeon et al. [132]

HJSC HJSC HJSC OSC OSC OSC OSC OSC OSC Hybrid Hybrid OPV

*Note: N.A. – Not Available. 94


N.N. Rosli et al.

being low cost, simple and requiring low-temperature processing. Kalita et al. synthesised a large-area CVD graphene film using botanical camphor as a carbon precursor on Ni and Cu substrates, and used it for the transparent electrode in OSC [38]. The graphene films grown on Ni and Cu substrate showed Rs of 1.645 kΩ/sq and 324 Ω/sq, respectively. The graphene-based OSC devices showed a lower performance compared to the ITO-based OSC due to the high Rs of the graphene electrode and the poor contact between organic materials and the hydrophobic graphene electrode. In another research, Choi et al. demonstrated the CVD-grown MLG electrode with Rs of 374 Ω/sq and optical transparency of 84.2% integrated into the OSC [41]. The PMMA-based multi-transfer process left some cracks or residual PMMA, which increased the Rs of the MLG films. The MLG-based OSC resulted in Voc (0.52 V), Jsc (6.90 mA/cm2), FF (32.6%) and PCE (1.17%) and the performance of these OSC was not as good as ITO-based OSC. The high Rs of the graphene films and shunt resistance (Rsh) in the OSC were the factors that caused the lower PCE of the solar cell. The HJSCs with hybrid graphene-CNT films as TCF were able to obtain high a PCE of 8.50% despite the low optoelectronic properties of the hybrid graphene-CNT TCF [118]. The high PCE of graphene-CNT based HJSCs was attributed to many factors, particularly the bridging of CNT networks over defects of graphene films, subsequently improving the conductance of graphene and enhancing the charge transport of the materials. P-type doping of the solar cell by HNO3 vapour increased the fill factor and open circuit voltage due to the optimisation of charge separation and transfer in the solar cell. A hybrid solar cell of n-Si/poly(3,4-ethylenedioxythiophene: poly (styrene-sulfonate)) (PEDOT: PSS) heterojunction with the GO-welded AgNW was fabricated by Xu et al. [108]. The incorporation of GO improved the properties of AgNW as it provided conductive pathways for charge transport and covered the spacing between overlapping AgNW, consequently lowering the Rs of the AgNW networks. The Rs of the AgNW with GO coating obtained was 13.3 Ω/sq, comparable to the ITO. The AgNW/GO was applied to the hybrid solar cell and an excellent performance (PCE of 13.3%) was obtained. Furthermore, Yin et al. reported on a hybrid solar cell with a layered structure of quartz/rGO/ZnO NR/poly (3-hexythiophene)/poly (3,4ethylenedioxythiophene): poly (styrenesulfonate) (P3HT/PEDOT:PSS)/ Au [133]. The GO was reduced by a two-step reduction process consisting of hydrazine vapour reduction and thermal annealing of the rGO films resulting in an rGO film with Rs of 0.42 ± 0.01 kΩ/sq, which was better than the rGO film reduced only with a one-step reduction process of hydrazine vapour reduction. The PCE of the hybrid solar cells with the rGO (two-step reduction) was better than rGO with only one-step reduction, which was 0.33% and 0.31%, respectively. However, the performance of the rGO-based hybrid solar cell was lower than other solar cell, which may be caused by other factors such as defects, lower conductivity and poor transmittance of rGO. In general, DSSC comprises the photoelectrode of mesoporous metal oxide semiconductor (i.e. TiO2) coated on transparent conducting substrates (i.e. ITO or FTO), photosensitiser (dye), electrolyte (i.e. iodide) and counter electrode (i.e. Pt-coated conductive glass substrates). Due to the exceptional properties of graphene, it can be applied in a few parts of DSSC such as transparent graphene electrode; TiO2/graphene hybrid layer acts as a bridge for photoelectrons to improve the charge transport and avoid charge recombination while graphene counter electrode (Pt-free) to enhance the reduction of the electrolyte; as shown in Fig. 4(c). Nonetheless, we are focusing on the implementation graphene as the front contact of DSSC. DSSC with rGO as TCF were demonstrated by Wang et al. [57]. The rGO TCF prepared by the Hummers method and reduction with thermal treatment under treatment of Ar and/or H2 flow, had the Rs of 0.46 ± 0.03 kΩ/sq. The PCE of 0.26% was obtained for graphenebased DSSC, which was lower than FTO-based DSSC with a PCE of 0.84%. The lower performance of DSSC was due to the series resistance

high PCE can be obtained as well as improving the stability of the CIGS solar cell due to the graphene protection from moisture. The structure of CdTE solar cell includes glass substrate, TCO layer (low resistivity TCO layer, ITO; and high resistivity TCO layer, SnO2), window layer (CdS), absorber layer (CdTe) and metal contact (Au or Ni/Al). LPCVD-synthesised graphene exhibited high mobility of more than 600 cm2 V−1 s−1, Rs in the range of 220–1150 Ω/sq and optical transparency in the range of 83.7–97.1% was used as the front electrode for a CdTE solar cell [62]. The number of graphene layers was tunable by controlling the hydrogen flow rate, with a single layer having higher transparency of 97%, whereas the seven-layer exhibited a much lower transparency of 84% with Rs of 220 Ω/sq. The low efficiency of 2.81% obtained for the CdTE solar cells with configuration of glass/graphene/ CdS/CdTE/ (graphite paste) was attributed to the poor contact of CdS and the graphene films, causing a current leakage. To overcome this problem, ZnO film barrier layer for the new configuration solar cell of glass/graphene/ZnO/CdS/CdTe/ (graphite paste) was applied and higher power conversion efficiency (PCE) of 4.17% was obtained. Third-generation solar cells of silicon nanodots requires a high–temperature fabrication process, which enables graphene to be used as a transparent conducting layer (TCL) due its high thermal stability properties. Graphene membrane was synthesised by Catalytic-Chemical Vapour Deposition (C-CVD), annealed with capping layer of hydrogenated amorphous silicon (a-Si:H) at temperature up to 1100 °C in inert atmosphere (N2) or in a high vacuum, removed the capping layer by TMAH wet etching and incorporated into p-i-n photovoltaic as TCL layer, was demonstrated by Veronese et al. [123]. Despite the higher the open circuit voltage (Voc) of p-i-n solar cell compared to ITO-based solar cells due to the larger work function of graphene, the lower fill factor (FF) of the solar cell was also observed. The lower FF was obtained due to the high Rs of the graphene film, which also contributed to the high shunt resistance of 4.5 × 102 Ω cm2. Nonetheless, the high temperature fabrication processing was possible due to the high thermal stability of graphene film, including the capping layer of a-Si: H. The highest PCE of 12.3% for graphene/Si Schottky junction solar cell by using the large area crack-filled graphene (CFG) as TCF was reported by Ho et al. [111]. The cracked graphene due to the growth condition was filled by Au nanoparticles, resulting in lower Rs of CFG films. The high PCE of graphene/Si Schottky junction solar cell was attributed to the low Rs of 0.37 Ω/sq due to trifluoromethanesulfonylamide (TFSA) doping to the device. The high FF was obtained due to the low series resistance between graphene and Si. The OSC with solution processed graphene transparent electrode were fabricated by Wu et al., where the graphene film was prepared by the modified Hummers method and reduced by vacuum annealing at 1100 °C [53]. The graphene film exhibits high Rs in the range of 100–500 kΩ/sq due to the incomplete reduction of functionalized graphene. The lower short circuit current density (Jsc) and FF of the solar cell were obtained due to the high Rs of graphene thin films, consequently affecting the PCE of the solar cell. In addition, the smaller shunt resistance also was also observed, which also contributed to the lower performance of the OSC. The rGO synthesised by the modified Hummers method and reduced by p-toluenesulfonyl hydrazide applied to the OSC as transparent electrode was demonstrated by Yun et al. [125]. The Rs of the rGO films was improved, decreasing from ~ 103 to 10 kΩ/sq with a transparency of 57–87% by a combination of annealing temperature and coating cycles. The PCE of the OSC improved from 0.1% to 0.3% when the rGO was applied as the transparent electrode (TE); however, it was lower than ITO-based OSC due to the higher series resistance and the smaller shunt resistance. The improved PCE for the as-fabricated OSC with rGO electrodes was obtained by annealing temperature of 200 °C and sevencycles coating, which determined the Voc of 0.56 V, Jsc of 3 mA/cm2, FF of 30% and the PCE of 0.33%. The PCE of these OSC is similar to other reported rGO-based OSC, but it has the advantages of fabrication for 95


N.N. Rosli et al.

enhancement to the transfer method and the hybrids of graphene with other materials. Some of the doping of graphene have improved its properties, such as being able to reduce the sheet resistance of graphene and improve its thermal stability. However, the transmittance of the graphene film suffers some degradation after the doping which may affect the performance of device application, particularly solar cells, when low transmittance graphene is applied. The improvement of graphene transfer method to the substrate can reduce the wrinkles, defects and PMMA residues, and will produce graphene with better quality and high performance. The defects of graphene caused by synthesis and the transfer method may be further reduced by using hybrid films of graphene with other materials. Most materials used for hybrid films with graphene have good electrical performance, which also contributes to the better electrical conductivity of the hybrid films. Evaluating the performance of TCF using FOM is a simple approach; however, there are too many FOM approaches (equations), and the best FOM has not yet been defined. However, Haacke's FOM (Eq. (2)) [21] and the bulk-film FOM (Eq. (7)) [12,26] are suitable to be used as these equations are much simpler and many studies have reported the property values (Rs and T) required in these calculations.

of the solar cells. In a related study by Selopal et al., the CVD-grown polycrystalline FLG was used as the front contact for DSSC and a PCE of 2% was obtained [127]. The Rs of the graphene film for active area of 1 cm2 was in the range of 0.5–1.0 kΩ/sq. PSCs represent an emerging solar cell technology, which incorporates perovskite material with various excellent properties such as high light absorption coefficients, direct band gap, high carrier mobility and long diffusion length. In general, PSC consists of substrate (glass or flexible substrate) coated with TCO layer (ITO or FTO), electron transport layer (ETL, mostly TiO2), absorber layer (perovskite material), hole transport layer (HTL, i.e. spiroOMeTAD, PEDOT:PSS) and metal back contact (Au or Ag). Perovskite material is based on organicinorganic materials of ABX3, where methylammonium (CH3NH3) is mostly used for A; B can be Sn or Pb; and X is a halide atom (I, Cl and Br). PSCs can be constructed into a few architectural structures such as normal structure (n-i-p), inverted (p-i-n), mesoscopic (with porous scaffold) and planar. Mesoscopic and planar architectures may exist in both normal and inverted structure. You et al. fabricated a semi-transparent PSC with laminated CVD stacked graphene as a transparent electrode (top layer) and FTO (bottom layer), having a PCE of 11.65% when illuminated from the top layer and 12.02% when illuminated from the bottom layer [128]. The PEDOT: PSS was used as a dopant and an adhesion layer to the active layer of PSCs. The Rs range of the CVD-grown graphene was obtained at about 1050 ± 150–260 ± 40 Ω/sq and the two-stacked graphene layer used for device fabrication had a Rs of 140 ± 35 Ω/sq and transmittance of > 90% in the visible region. Flexible perovskite solar was fabricated by Liu et al. with PET as flexible substrate, graphene as transparent conducting layer and P3HT as HTL [129]. During the transfer process of CVD-synthesised graphene, P3HT was used as a supporting layer instead of PMMA which obtained better quality graphene without polymer residue. In addition, the P3HT also promoted p-type doping to graphene, thus, improving its electrical properties. In another research, Kim et al. also demonstrated flexible PSC with 2 layered graphene transparent electrode and a high PCE of 13.35% was obtained [130]. The stability of PSCs with PEDOT:PSS as HTL often deteriorate due to the degradation of ITO caused by the acidic nature of PEDOT:PSS, thus, Heo et al. replaced the conventional transparent electrode with AuCl3-doped graphene p-type transparent conducting electrode in p-i-n type CH3NH3PBI3 PSC and a high PCE of 17.4–17.9% was obtained [131]. The doping of AuCl3 has improved the Rs of graphene from ~ 890 Ω/sq to ~ 70 Ω/sq. Comparison of carbon-based transparent electrode for inverted PSC was made by Jeon et al. using SWNT and graphene transparent electrode with doping of MoO3 [132]. Higher efficiency was obtained by graphene-based PSC with PCE of 14.8% whereas SWNT had a lower efficiency of 12.8%. Better performance (higher Voc and Jsc) of graphene-based PSC was attributed to the high transmittance and better morphology of graphene compared to SWNT.

Acknowledgement We would like to acknowledge the Ministry of Education Malaysia for funding this study (Grant: FRGS/2/2014/SG06/UKM/03/1) and the research facilities supported by Universiti Kebangsaan Malaysia (UKM). References [1] Nomoto J, Hirano T, Miyata T, Minami T. Preparation of Al-doped ZnO transparent electrodes suitable for thin-film solar cell applications by various types of magnetron depositions. Thin Solid Films 2011;520:1400–6. https://doi.org/10.1016/j. tsf.2011.10.003. [2] McCarthy JE, Hanley CA, Lambertini VG, Gun’ko YK. Fabrication of highly transparent and conductive PEDOT:PSS thin films for flexible electrode applications. Nanocon.Eu, Czech Republic; 2013. [3] De S, Higgins TM, Lyons PE, Doherty EM, Nirmalraj PN, Blau WJ, et al. Silver nanowire networks as flexible, transparent conducting films: extremely high DC to optical conductivity ratios. ACS Nano 2009;3:1767–74. [4] Hu L, Kim HS, Lee J, Peumans P, Cui Y. Scalable coating and properties of transparent, flexible, silver nanowire electrode. ACS Nano 2010;4:2955–63. [5] Langley DP, Giusti G, Lagrange M, Collins R, Jiménez C, Bréchet Y, et al. Silver nanowire networks: physical properties and potential integration in solar cells. Sol Energy Mater Sol Cells 2014;125:318–24. https://doi.org/10.1016/j.solmat.2013. 09.015. [6] Du Pasquier A, Unalan HE, Kanwal A, Miller S, Chhowalla M. Conducting and transparent single-wall carbon nanotube electrodes for polymer-fullerene solar cells. Appl Phys Lett 2005;87:203511. https://doi.org/10.1063/1.2132065. [7] Wang YP, Lu JG, Bie X, Ye ZZ, Li X, Song D, et al. Transparent conductive and nearinfrared reflective Cu-based Al-doped ZnO multilayer films grown by magnetron sputtering at room temperature. Appl Surf Sci 2011;257:5966–71. https://doi.org/ 10.1016/j.apsusc.2011.01.068. [8] Jung YS, Park YS, Kim KH, Lee W. Properties of AZO/Ag/AZO multilayer thin film deposited on polyethersulfone substrate. Trans Electr Electron Mater 2013;14:9–11. [9] Minami T. Substitution of transparent conducting oxide thin films for indium tin oxide transparent electrode applications. Thin Solid Films 2008;516:1314–21. https://doi.org/10.1016/j.tsf.2007.03.082. [10] Minami T. Present status of transparent conducting oxide thin-film development for Indium-Tin-Oxide (ITO) substitutes. Thin Solid Films 2008;516:5822–8. https://doi.org/10.1016/j.tsf.2007.10.063. [11] Stadler A. Transparent conducting oxides — an up-to-date overview. Materials 2012;5:661–83. https://doi.org/10.3390/ma5040661. [12] Ellmer K. Past achievements and future challenges in the development of optically transparent electrodes. Nat Photonics 2012;6:809–17. https://doi.org/10.1038/ NPHOTON.2012.282. [13] Hu L, Wu H, Cui Y. Metal nanogrids, nanowires and nanofibers for transparent electrodes. MRS Bull 2011;36:760–5. https://doi.org/10.1557/mrs.2011.234. [14] Van De Lagemaat J, Barnes TM, Rumbles G, Shaheen SE, Coutts TJ, Weeks C, et al. Organic solar cells with carbon nanotubes replacing In2O3: Sn as the transparent electrode. Appl Phys Lett 2006;88:233503. https://doi.org/10.1063/1.2210081. [15] Song Y, Fang W, Brenes R, Kong J. Challenges and opportunities for graphene as transparent conductors in optoelectronics. Nano Today 2015;10:681–700. https:// doi.org/10.1016/j.nantod.2015.11.005. [16] Wang X, Shi G. Flexible graphene devices related to energy conversion and

7. Conclusions and future challenges Graphene is indeed a unique material: it has excellent properties applicable to various applications. However, the best electrical performance of graphene often occurs at low transparency in the range 70–80% of visible light and is still unable to compete with the performance of ITO as a transparent electrode for various applications. In addition, the method to synthesis graphene to achieve the best performance is still complicated and cumbersome. The synthesis of graphene that involves chemical treatments is still having a significant effect on the structure of graphene, thus affecting its performance. Therefore, an improvement to the methods or a new synthesis method is highly desired. To date, a few methods have been proposed to improve the performance of graphene, which include doping with impurities, 96


N.N. Rosli et al.

Atmospheres. ACS Nano 2014;9:31–42. https://doi.org/10.1021/nn504822m. [49] Yin L, Yang C, Zhang K, Luo H, Zhang X, Liu Z, et al. Application of CVD graphene as transparent front electrode in Cu(In,Ga)Se2 solar cell. Photovolt Spec Conf 2014:1740–4. https://doi.org/10.1109/PVSC.2014.6925257. [50] Ahn Y, Jeong Y, Lee D, Lee Y. Copper nanowire-graphene core-shell nanostructure for highly stable transparent conducting electrodes. ACS Nano 2015;9:3125–33. https://doi.org/10.1021/acsnano.5b00053. [51] Liu N, Pan Z, Fu L, Zhang C, Dai B, Liu Z. The origin of wrinkles on transferred graphene. Nano Res 2011;4:996–1004. https://doi.org/10.1007/s12274-0110156-3. [52] Hummers WSJ, Offeman RE. Preparation of graphitic oxide. J Am Chem Soc 1958;80:1339. [53] Wu J, Becerril HA, Bao Z, Liu Z, Chen Y, Peumans P. Organic solar cells with solution-processed graphene transparent electrodes. Appl Phys Lett 2008;92:263302. https://doi.org/10.1063/1.2924771. [54] Becerril HA, Mao J, Liu Z, Stoltenberg RM, Bao Z, Chen Y. Evaluation of solutionprocessed reduced graphene oxide films as transparent conductors. ACS Nano 2008;2:463–70. [55] Gilje S, Han S, Wang M, Wang KL, Kaner RB. A chemical route to graphene for device applications. Nano Lett 2007;7:3394–8. https://doi.org/10.1021/ nl0717715. [56] Pei S, Cheng HM. The reduction of graphene oxide. Carbon N Y 2012;50:3210–28. https://doi.org/10.1016/j.carbon.2011.11.010. [57] Wang X, Zhi L, Müllen K. Transparent, conductive graphene electrodes for dyesensitized solar cells. Nano Lett 2008;8:323–7. https://doi.org/10.1021/ nl072838r. [58] Eda G, Lin Y-Y, Miller S, Chen C-W, Su W-F, Chhowalla M. Transparent and conducting electrodes for organic electronics from reduced graphene oxide. Appl Phys Lett 2008;92:233305. https://doi.org/10.1063/1.2937846. [59] Ho X, Lu H, Liu W, Tey JN, Cheng CK, Kok E, et al. Electrical and optical properties of hybrid transparent electrodes that use metal grids and graphene films. J Mater Res 2013;28:620–6. https://doi.org/10.1557/jmr.2012.399. [60] Jang CW, Kim JM, Kim JH, Shin DH, Kim S, Choi S. Degradation reduction and stability enhancement of p-type graphene by RhCl3 doping. J Alloy Compd 2015;621:1–6. https://doi.org/10.1016/j.jallcom.2014.09.182. [61] Cha MJ, Kim SM, Kang SJ, Seo JH, Walker B. Improved performance in flexible organic solar cells via optimization of highly transparent silver grid/graphene electrodes. RSC Adv 2015;5:65646–50. https://doi.org/10.1039/C5RA10838A. [62] Bi H, Huang F, Liang J, Xie X, Jiang M. Transparent conductive graphene films synthesized by ambient pressure chemical vapor deposition used as the front electrode of CdTe solar cells. Adv Mater 2011;23:3202–6. https://doi.org/10. 1002/adma.201100645. [63] Gunes F, Shin H, Biswas C, Han GH, Kim ES, Chae SJ, et al. Layer-by-layer doping of few layer graphene film. ACS Nano 2010;4:4595–600. [64] Kwon KC, Kim BJ, Lee J, Kim SY. Effect of anions in Au complexes on doping and degradation of graphene†. J Mater Chem C 2013;1:2463–9. https://doi.org/10. 1039/c3tc00046j. [65] Ishikawa R, Bando M, Morimoto Y, Sandhu A. Doping graphene films via chemically mediated charge transfer. Nanoscale Res Lett 2011:6. https://doi.org/10. 1186/1556-276X-6-111. [66] Ma X, Zhang H. Fabrication of graphene films with high transparent conducting characteristics. Nanoscale Res Lett 2013;8:440. https://doi.org/10.1186/1556276X-8-440. [67] Reina A, Jia X, Ho J, Nezich D, Son H, Bulovic V, et al. Large area, few-layer graphene films on arbitrary substrates by chemical vapor deposition. Nano Lett 2009;9:30–5. [68] Pinto H, Markevich A. Electronic and electrochemical doping of graphene by surface adsorbates. Beilstein J Nanotechnol 2014;5:1842–8. https://doi.org/10. 3762/bjnano.5.195. [69] Guo B, Fang L, Zhang B, Gong JR. Graphene doping: a review. Insci J 2011;1:80–9. https://doi.org/10.5640/insc.010280. [70] Oh JS, Kim KN, Yeom GY. Graphene doping methods and device applications. J Nanosci Nanotechnol 2014;14:1120–33. https://doi.org/10.1166/jnn.2014.9118. [71] Kang MH, Milne WI, Cole MT. Doping stability and opto-electronic performance of chemical vapour deposited graphene for plastic electronic applications. Act Flatpanel Disp Devices 2000:229–32. [72] Kim KK, Reina A, Shi Y, Park H, Li L, Lee YH, et al. Enhancing the conductivity of transparent graphene films via doping. Nanotechnology 2010;21:285205. https:// doi.org/10.1088/0957-4484/21/28/285205. [73] Kim B, Yang G, Kim H, Baik KH, Michael A, Hite JK, et al. GaN-based ultraviolet light-emitting diodes with AuCl3 -doped graphene electrodes. Opt Express 2013;21:29025–30. https://doi.org/10.1364/OE.21.029025. [74] Kasry A, Kuroda MA, Martyna GJ, Tulevski GS, Bol AA. Chemical doping of largearea stacked graphene films for use as transparent, conducting electrodes. ACS Nano 2010;4:3839–44. https://doi.org/10.1021/nn100508g. [75] Lin X, Jia J, Yousefi N, Shen X, Kim J-K. Excellent optoelectrical properties of graphene oxide thin films deposited on a flexible substrate by Langmuir-Blodgett assembly. J Mater Chem C 2013;1:6869–77. https://doi.org/10.1039/c3tc31497a. [76] Krajewska A, Oberda K, Azpeitia J, Gutierrez A, Mierczyk Z, Munuera C, et al. Influence of Au doping on electrical properties of CVD graphene. Carbon N Y 2016;100:625–31. https://doi.org/10.1016/j.carbon.2016.01.066. [77] Bult JB, Crisp R, Perkins CL, Blackburn JL. Role of dopants in long-range charge carrier transport for p-Type and n-Type graphene transparent conducting thin films. ACS Nano 2013;7:7251–61. [78] Sandoval S, Kumar N, Sundaresan A, Rao CNR, Fuertes A. Enhanced thermal oxidation stability of reduced graphene oxide by nitrogen doping. Chem Eur J

storage. Energy Environ Sci 2015;8:790–823. https://doi.org/10.1039/b000000x. [17] Lee Y, Ahn J. Graphene-based transparent conductive films. Nano 2013;8:133001. https://doi.org/10.1142/S1793292013300016. [18] Jain VK, Kulshreshtha AP. Indium-tin-oxide transparent conducting coatings on silicon solar cells and their “figure of merit.”. Sol Energy Mater 1981;4:151–8. [19] De S, King PJ, Lyons PE, Khan U, Coleman JN. Size effects and the problem with percolation in nanostructured transparent conductors. ACS Nano 2010;4:7064–72. [20] Fraser DB, Cook HD. Highly conductive, transparent films of sputtered In2xSnxO3-y. J Electrochem Soc 1972;119:1368–74. https://doi.org/10.1149/1. 2403999. [21] Haacke G. New figure of merit for transparent conductors. J Appl Phys 1976;47:4086–9. https://doi.org/10.1063/1.323240. [22] Schropp REI, Madan A. Properties of conductive zinc oxide films for transparent electrode applications prepared by rf magnetron sputtering. J Appl Phys 1989;66:2027–31. https://doi.org/10.1063/1.344341. [23] Gordon RG. Criteria for choosing transparent conductors. MRS Bull 2000;25:52–7. [24] King PJ, Khan U, Lotya M, De S, Coleman JN. Improvement of transparent conducting nanotube films by addition of small quantities of graphene. ACS Nano 2010;4:4238–46. [25] De S, King PJ, Lotya M, Neill AO, Doherty EM, Hernandez Y, et al. Flexible, transparent, conducting films of randomly stacked graphene from surfactant-stabilized, oxide-free graphene dispersions. Small 2010;6:458–64. https://doi.org/ 10.1002/smll.200901162. [26] Barnes TM, Reese MO, Bergeson JD, Larsen BA, Blackburn JL, Beard MC, et al. Comparing the fundamental physics and device performance of transparent, conductive nanostructured networks with conventional transparent conducting oxides. Adv Energy Mater 2012;2:353–60. https://doi.org/10.1002/aenm. 201100608. [27] Shim BS, Zhu J, Jan E, Critchley K, Kotov NA. Transparent conductors from layerby-layer assembled swnt films: importance of mechanical properties and a new figure of merit. ACS Nano 2010;4:3725–34. [28] Rowell MW, McGehee MD. Transparent electrode requirements for thin film solar cell modules. Energy Environ Sci 2011;4:131–4. https://doi.org/10.1039/ C0EE00373E. [29] Kaskela A, Nasibulin AG, Timmermans MY, Aitchison B, Papadimitratos A, Tian Y, et al. Aerosol-synthesized SWCNT networks with tunable conductivity and transparency by a dry transfer technique. Nano Lett 2010;10:4349–55. https://doi.org/ 10.1021/nl101680s. [30] Exarhos GJ, Zhou X. Discovery-based design of transparent conducting oxide films. Thin Solid Films 2007;515:7025–52. https://doi.org/10.1016/j.tsf.2007.03.014. [31] Golobostanfard MR, Abdizadeh H, Mohammadi S, Baghchesara MA. Carbon nanotube/indium tin oxide hybrid transparent conductive film: effect of nanotube diameter. Sol Energy Mater Sol Cells 2015;132:418–24. https://doi.org/10.1016/j. solmat.2014.09.036. [32] Geim AK, Novoselov KS. The rise of graphene. Nat Mater 2007;6:183–91. [33] Reduction of graphene n.d. 〈http://www.graphene-uses.com/graphene-oxide/〉. [34] Novoselov KS, Geim AK, Morozov SV, Jiang D, Zhang Y, Dubonos SV, et al. Electric field effect in atomically thin carbon films. Science (80- ) 2004;306:666–9. [35] Yi M, Shen Z. A review on mechanical exfoliation for the scalable production of graphene. J Mater Chem A 2015;3:11700–15. https://doi.org/10.1039/ C5TA00252D. [36] Jayasena B, Melkote SN. An investigation of PDMS stamp assisted mechanical exfoliation of large area graphene. Procedia Manuf 2015;1:840–53. https://doi. org/10.1016/j.promfg.2015.09.073. [37] Huang Y, Sutter E, Shi NN, Zheng J, Yang T, Englund D, et al. Reliable exfoliation of large-area high-quality flakes of graphene and other two-dimensional materials. ACS Nano 2015;9:10612–20. https://doi.org/10.1021/acsnano.5b04258. [38] Kalita G, Wakita K, Umeno M, Hayashi Y, Tanemura M, Large-Area CVD. Graphene as transparent electrode for efficient organic solar cells. Photovolt Spec Conf 2012:3137–41. https://doi.org/10.1109/PVSC.2012.6318244. [39] Kalita G, Wakita K, Umeno M, Hayashi Y, Tanemura M. Synthesis of continuous graphene on metal foil for flexible transparent electrode application. Nanoelectron Conf 2013:281–4. https://doi.org/10.1109/INEC.2013.6466023. [40] Chen X, Zhang L, Chen S. Large area CVD growth of graphene. Synth Met 2015;210:95–108. https://doi.org/10.1016/j.synthmet.2015.07.005. [41] Choi Y-Y, Kang SJ, Kim H-K, Choi WM, Na S-I. Multilayer graphene films as transparent electrodes for organic photovoltaic devices. Sol Energy Mater Sol Cells 2012;96:281–5. https://doi.org/10.1016/j.solmat.2011.09.031. [42] Zhang Y, Li Z, Kim P, Zhang L, Zhou C. Anisotropic hydrogen etching of chemical vapor deposited graphene. ACS Nano 2012;6:126–32. [43] Zhang X, Ning J, Li X, Wang B, Hao L, Liang M, et al. Hydrogen-induced effects on the CVD growth on high quality graphene structures. Nanoscale 2013;5:8363–6. [44] Li X, Cai W, An J, Kim S, Nah J, Yang D, et al. Large-area synthesis of high-quality and uniform graphene films on copper foils. Science (80- ) 2009;324:1312–4. https://doi.org/10.1126/science.1171245. [45] Zhang X, Wang L, Xin J, Yakobson BI, Ding F. Role of hydrogen in graphene chemical vapor deposition growth on a copper surface. J Am Chem Soc 2014;136:3040–7. [46] Hu B, Jin Y, Guan D, Luo Z, Shang B, Fang Q, et al. H-dependent carbon dissolution and diffusion-out in graphene CVD growth. J Phys Chem C 2015;119:24124–31. https://doi.org/10.1021/acs.jpcc.5b07127. [47] Syed MH, Chong SK, Huang NM, Abdul Rahman S. Fabrication of high-quality graphene by hot-filament thermal chemical vapor deposition. Carbon N Y 2015;86:1–11. https://doi.org/10.1016/j.carbon.2015.01.018. [48] Strudwick AJ, Weber NE, Schwab MG, Weitz RT, Wünsch JR, Müllen K, et al. Chemical Vapor Deposition of High Quality Graphene Films from Carbon Dioxide

97


N.N. Rosli et al.

2014;20:11999–2003. https://doi.org/10.1002/chem.201403833. [79] Abdullah-al-galib M, Hou B, Shahriad T, Zivanovic S, Radadia AD. Stability of few layer graphene films doped with gold (III) chloride. Appl Surf Sci 2016;366:78–84. https://doi.org/10.1016/j.apsusc.2016.01.065. [80] Barin GB, Song Y, Gimenez IDF, Gomes A, Filho S, Barreto LS, et al. Optimized graphene transfer: influence of polymethylmethacrylate (PMMA) layer concentration and baking time on graphene final performance. Carbon N Y 2015;84:82–90. https://doi.org/10.1016/j.carbon.2014.11.040. [81] Kratzer M, Bayer BC, Kidambi PR, Matković A, Gajić R, Cabrero A, et al. Effects of polymethylmethacrylate-transfer residues on the growth of organic semiconductor molecules on chemical vapor deposited graphene. Appl Phys Lett 2015;106:103101. https://doi.org/10.1063/1.4913948. [82] Deokar G, Avila J, Codron J, Boyaval C, Galopin E, Asensio M, et al. Towards high quality CVD graphene growth and transfer. Carbon N Y 2015;89:82–92. https:// doi.org/10.1016/j.carbon.2015.03.017. [83] Antonova IV, Golod SV, Soots RA, Komonov AI, Seleznev VA, Sergeev MA, et al. Comparison of various methods for transferring graphene and few layer graphene grown by chemical vapor deposition to an insulating SiO2/Si substrate. Semiconductors 2014;48:804–8. https://doi.org/10.1134/S1063782614060049. [84] Ko PJ, Takahashi H, Koide S, Sakai H, Thu TV, Okada H, et al. Simple method to transfer graphene from metallic catalytic substrates to flexible surfaces without chemical etching. J Phys Conf Ser 2013;433:12002. https://doi.org/10.1088/ 1742-6596/433/1/012002. [85] Ma D, Zhang Y, Liu M, Ji Q, Gao T, Zhang Y, et al. Clean transfer of graphene on Pt foils mediated by a carbon monoxide intercalation process. Nano Res 2013;6:671–8. https://doi.org/10.1007/s12274-013-0342-6. [86] Marta B, Leordean C, Istvan T, Botiz I, Astilean S. Efficient etching-free transfer of high quality, large-area CVD grown graphene onto polyvinyl alcohol films. Appl Surf Sci 2015;363:613–8. https://doi.org/10.1016/j.apsusc.2015.11.265. [87] Banszerus L, Schmitz M, Engels S, Dauber J, Oellers M, Haupt F, et al. Ultrahighmobility graphene devices from chemical vapor deposition on reusable copper. Mater Sci 2015:1. [88] Jung D, Ko Y-H, Cho J, Adhikari PD, Lee S Il, Kim Y, et al. Transparent and flexible conducting hybrid film combined with 3-Aminopropyltriethoxysilane-coated polymer and graphene. Appl Surf Sci 2015;357:287–92. https://doi.org/10.1016/ j.apsusc.2015.08.139. [89] Ning J, Wang D, Zhang C, Wang Z, Tang S, Chen D, et al. Electrical and optical properties of layer-stacked graphene transparent electrodes using self-supporting transfer method. Synth Met 2015;203:215–20. https://doi.org/10.1016/j. synthmet.2015.02.007. [90] Lee WH, Suk JW, Lee J, Hao Y, Park J, Yang JW, et al. Simultaneous transfer and doping of CVD-grown graphene by fluoropolymer for transparent conductive films on plastic. ACS Nano 2012;6:1284–90. [91] Castro RK De, Araujo JR, Valaski R, Costa LOO, Archanjo BS, Fragneaud B, et al. New transfer method of CVD-grown graphene using a flexible, transparent and conductive polyaniline-rubber thin film for organic electronic applications. Chem Eng J 2015. https://doi.org/10.1016/j.cej.2015.03.092. [92] Regan W, Alem N, Alemán B, Geng B, Girit Ç, Maserati L, et al. A direct transfer of layer-area graphene. Appl Phys Lett 2014;113102:113102. https://doi.org/10. 1063/1.3337091. [93] Lin W-H, Chen T-H, Chang J, Taur J-I, Lo Y-Y, Lee W-L, et al. A direct and polymerfree method for transferring graphene grown by chemical vapor deposition to any substrate. ACS Nano 2014;8:1784–91. [94] Yun YS, Kim DH, Kim B, Park HH, Jin H-J. Transparent conducting films based on graphene oxide/silver nanowire hybrids with high flexibility. Synth Met 2012;162:1364–8. https://doi.org/10.1016/j.synthmet.2012.05.026. [95] Kholmanov IN, Magnuson CW, Aliev AE, Li H, Zhang B, Suk JW, et al. Improved electrical conductivity of graphene films integrated with metal nanowires. Nano Lett 2012;12:5679–83. https://doi.org/10.1021/nl302870x. [96] Lee M, Lee K, Kim S, Lee H, Park J, Choi K. et al. high-performance, transparent, and stretchable electrodes using graphene − metal nanowire hybrid structures; 2013. [97] Choi HO, Kim DW, Kim SJ, Yang SB, Jung H. Role of 1D metallic nanowires in polydomain graphene for highly transparent conducting films. Adv Mater 2014;26:4575–81. https://doi.org/10.1002/adma.201306234. [98] Lee D, Lee H, Ahn Y, Jeong Y, Lee D, Lee Y. Highly stable and flexible silver nanowire–graphene hybrid transparent conducting electrodes for emerging optoelectronic devices†. Nanoscale 2013;5:7750–5. https://doi.org/10.1039/ c3nr02320f. [99] Lee D, Lee H, Ahn Y, Lee Y. High-performance flexible transparent conductive film based on graphene/AgNW/graphene sandwich structure. Carbon N Y 2014;81:439–46. https://doi.org/10.1016/j.carbon.2014.09.076. [100] Deng B, Hsu P-C, Chen G, Chandrashekar BN, Liao L, Ayitimuda Z, et al. Roll-toroll encapsulation of metal nanowires between graphene and plastic substrate for high-performance flexible transparent electrodes. Nano Lett 2015;15:4206–13. https://doi.org/10.1021/acs.nanolett.5b01531. [101] Sun H, Ge G, Zhu J, Yan H, Lu Y, Wu Y. High electrical conductivity of graphenebased transparent conductive fi lms with silver nanocomposites. RSC Adv 2015;5:108044–9. https://doi.org/10.1039/C5RA24650D. [102] Hsu PC, Wu H, Carney TJ, McDowell MT, Yang Y, Garnett EC, et al. Passivation coating on electrospun copper nanofibers for stable transparent electrodes. ACS Nano 2012;6:5150–6. https://doi.org/10.1021/nn300844g. [103] Guo H, Lin N, Chen Y, Wang Z, Xie Q, Zheng T, et al. Copper nanowires as fully transparent conductive electrodes. Sci Rep 2013;3:2323. https://doi.org/10.1038/ srep02323. [104] Kim Y, Min D. Durable large-area thin films of graphene/carbon nanotube double

[105]

[106]

[107]

[108]

[109]

[110]

[111]

[112]

[113]

[114]

[115]

[116]

[117]

[118]

[119]

[120]

[121]

[122] [123]

[124]

[125]

[126]

[127]

[128]

[129]

98

layers as a transparent electrode. Langmuir 2009;25:11302–6. https://doi.org/10. 1021/la9029744. Hong T, Lee DW, Choi HJ, Shin HS, Kim B. Transparent, flexible conducting hybrid multilayer thin films of multiwalled carbon nanotubes with graphene nanosheets. ACS Nano 2010;4:3861–8. Li C, Li Z, Zhu H, Wang K, Wei J, Li X, et al. Graphene nano-“patches” on a carbon nanotube network for highly transparent/conductive thin film applications. J Phys Chem C 2010;114:14008–12. Gorkina A, Tsapenko A, Gilshteyn E. Transparent and conductive hybrid graphene/carbon nanotube films. Carbon N Y 2016;100:501–7. https://doi.org/10. 1016/j.str.2014.12.012. Xu Q, Song T, Cui W, Liu Y, Xu W, Lee ST, et al. Solution-processed highly conductive PEDOT:PSS/AgNW/GO transparent film for efficient Organic-Si hybrid solar cells. ACS Appl Mater Interfaces 2015;7:3272–9. https://doi.org/10.1021/ am508006q. Domingues SH, Kholmanov IN, Kim T, Kim J, Tan C, Chou H, et al. Reduction of graphene oxide films on Al foil for hybrid transparent conductive film applications. Carbon N Y 2013;63:454–9. https://doi.org/10.1016/j.carbon.2013.07.007. Zhu Z, Mankowski T, Balakrishnan K. Ultrahigh aspect ratio copper-nanowirebased hybrid transparent conductive electrodes with PEDOT: PSS and reduced Graphene Oxide exhibiting reduced surface roughness and improved stability. ACS Appl Mater Interfaces 2015;7:16223–30. https://doi.org/10.1021/acsami. 5b01379. Ho P-H, Liou Y-T, Chuang C-H, Lin S-W, Tseng C-Y, Wang D-Y, et al. Self-crackfilled graphene films by metallic nanoparticles for high-performance graphene heterojunction solar cells. Adv Mater 2015;27:1724–9. https://doi.org/10.1002/ adma.201404843. Kahng YH, Kim MK, Lee JH, Kim YJ, Kim N, Park DW, et al. Highly conductive flexible transparent electrodes fabricated by combining graphene films and inkjetprinted silver grids. Sol Energy Mater Sol Cells 2014;124:86–91. https://doi.org/ 10.1016/j.solmat.2014.01.040. Kang J, Jang Y, Kim Y, Cho S-H, Suhr J, Hong BH, et al. An Ag-grid/graphene hybrid structure for large-scale, transparent, flexible heaters. Nanoscale 2015;7:6567–73. https://doi.org/10.1039/c4nr06984f. Gao T, Li Z, Huang P, Shenoy GJ, Parobek D, Tan S, et al. Hierarchical graphene/ metal grid structures for stable, flexible transparent conductors. ACS Nano 2015;9:5440–6. https://doi.org/10.1021/acsnano.5b01243. Qiu T, Luo B, Liang M, Ning J, Wang B, Li X, et al. Hydrogen reduced graphene oxide/metal grid hybrid film: towards high performance transparent conductive electrode for flexible electrochromic devices. Carbon N Y 2014;81:232–8. https:// doi.org/10.1016/j.carbon.2014.09.054. Zheng Q, Zhang B, Lin X, Shen X, Yousefi N, Huang Z-D, et al. Highly transparent and conducting ultralarge graphene oxide/single-walled carbon nanotube hybrid films produced by Langmuir–Blodgett assembly. J Mater Chem 2012;22:25072–82. https://doi.org/10.1039/c2jm34870e. Tung VC, Chen L, Allen MJ, Wassei JK, Nelson K, Kaner RB, et al. Low-temperature solution processing of graphene - carbon nanotube hybrid materials for highperformance transparent conductors. Nano Lett 2009;9:1949–55. Gan X, Lv R, Bai J, Zhang Z, Wei J, Huang Z, et al. Efficient photovoltaic conversion of graphene – carbon nanotube hybrid films grown from solid precursors. 2D Mater 2015;2:34003. https://doi.org/10.1088/2053-1583/2/3/034003. Bae S, Kim H, Lee Y, Xu X, Park JS, Zheng Y, et al. Roll-to-roll production of 30inch graphene films for transparent electrodes. Nat Nanotechnol 2010;5:574–8. https://doi.org/10.1038/nnano.2010.132. Kim RH, Bae MH, Kim DG, Cheng H, Kim BH, Kim DH, et al. Stretchable, transparent graphene interconnects for arrays of microscale inorganic light emitting diodes on rubber substrates. Nano Lett 2011;11:3881–6. https://doi.org/10.1021/ nl202000u. Lima AH, Mendonça JP, Duarte M, Stavale F, Legnani C, Carvalho GSG De, et al. Reduced graphene oxide prepared at low temperature thermal treatment as transparent conductors for organic electronic applications. Org Electron 2017. https://doi.org/10.1016/j.orgel.2017.05.054. Bonaccorso F, Sun Z, Hasan T, Ferrari AC. Graphene photonics and optoelectronics. Nat Photon 2010;4:611–22. https://doi.org/10.1038/nphoton.2010.186. Veronese GP, Allegrezza M, Canino M, Centurioni E, Ortolani L, Rizzoli R, et al. Graphene as transparent conducting layer for high temperature thin film device applications. Sol Energy Mater Sol Cells 2015;138:35–40. https://doi.org/10. 1016/j.solmat.2015.02.026. Yin L, Zhang K, Luo H, Cheng G, Ma X, Xiong Z, et al. Highly efficient graphenebased Cu(In, Ga)Se2 solar cells with large active area. Nanoscale 2014;6:10879–86. Yun J-M, Noh Y-J, Jeon Y-J, Kim S-S, Kim D-Y, Na S. Morphological, optical, and electrical investigations of solution-processed reduced graphene oxide and its application to transparent electrodes in organic solar cells. J Ind Eng Chem 2014;21:877–83. https://doi.org/10.1016/j.jiec.2014.04.026. Yin Z, Wu S, Zhou X, Huang X, Zhang Q, Boey F, et al. Electrochemical deposition of ZnO nanorods on transparent reduced graphene oxide electrodes for hybrid solar cells. Small 2010;6:307–12. https://doi.org/10.1002/smll.200901968. Selopal GS, Milan R, Ortolani L, Morandi V, Rizzoli R, Sberveglieri G, et al. Graphene as transparent front contact for dye sensitized solar cells. Sol Energy Mater Sol Cells 2015;135:99–105. https://doi.org/10.1016/j.solmat.2014.10.016. You P, Liu Z, Tai Q, Liu S, Yan F. Efficient semitransparent perovskite solar cells with graphene electrodes. Adv Mater 2015;27:3632–8. https://doi.org/10.1002/ adma.201501145. Liu Z, You P, Xie C, Tang G, Yan F. Ultrathin and flexible perovskite solar cells with graphene transparent electrodes. Nano Energy 2016;28:151–7. https://doi.


N.N. Rosli et al.

[132] Jeon I, Yoon J, Ahn N, Atwa M, Delacou C, Anisimov A, et al. Carbon nanotubes versus graphene as flexible transparent electrodes in inverted perovskite solar cells. J Phys Chem Lett 2017;8:5395–401. https://doi.org/10.1021/acs.jpclett. 7b02229. [133] Wu S, Yin Z, He Q, Huang X, Zhou X, Zhang H. Electrochemical deposition of semiconductor oxides on reduced graphene oxide-based flexible, transparent, and conductive electrodes. J Phys Chem C 2010;114:11816–21. https://doi.org/10. 1021/jp103696u.

org/10.1016/j.nanoen.2016.08.038. [130] Kim S, Lee HS, Kim JM, Seo SW, Kim JH, Jang CW, et al. Effect of layer number on flexible perovskite solar cells employing multiple layers of graphene as transparent conductive electrodes. J Alloy Compd 2018;744:404–11. https://doi.org/10. 1016/j.jallcom.2018.02.136. [131] Heo JH, Shin DH, Kim S, Jang MH, Lee MH, Seo SW, et al. Highly efficient CH3NH3PbI3 perovskite solar cells prepared by AuCl3-doped graphene transparent conducting electrodes. Chem Eng J 2017;323:153–9. https://doi.org/10. 1016/j.cej.2017.04.097.

99