A. History

The world's first flexible SC was fabricated in 1990 by Kishi and co-workers,^{Reference Matsuyama, Wakisaka, Kameda, Tanaka, Matsuoka, Tsuda, Nakano, Kishi and Kuwano12} on a transparent plastic substrate. The thickness of the amorphous silicon SC was only 0.12 mm and its bending radius was 5 mm, the cell had the world's biggest calorific value at the time (275 mW/g).^{Reference Zou, Wang, Chu, Lv and Fan1}

One of the main disadvantages of using this kind of substrate is their weak tolerance to high temperatures, which make sintering of the photo electrode films unattainable. This is one of the reasons why carbon materials and conducting polymers have been presented as credible alternatives to use on the fabrication of flexible SC as substrates.^{Reference Toivola, Halme, Miettunen, Aitola and Lund13}

B. Types of flexible SCs

DSSCs on alternative substrates to glass (metal sheets or plastic substrates) are also gaining great interest. Metal substrates in flexible SC have the advantage of being cheaper than transparent conductive oxide glass (TCO), having lower sheet resistance and enduring high-temperature processing. Plastic substrates, namely indium tin oxide polyethylene terephthalate (ITO)-PET or ITO-polyethylene naphthalate (PEN), have the advantage of being lightweight, cheap, and suitable for roll-to roll type cells and for continuous production. Setbacks are little tolerance for high temperatures (150°) and are not as efficient as glass as a protective layer from infiltration of unwanted agents (water and oxygen). Despite the reasons presented above, plastic substrates may be a good way to go because of its low cost of manufacture.^{Reference Toivola, Halme, Miettunen, Aitola and Lund13}

Fiber-based DSSCs have the advantage of being flexible, lightweight, relatively great active surface area due to the absence of limits in terms of dimension and most of all the capability of being integrated on wearable fabrics. The measured open-circuit voltage (V _oc) of the photoactive fiber was very close to the reported voltage for a normal DSSC (0.6–0.8 V). The best measured results for plastic fiber and for glass fiber were 0.42 V and 0.50 V, respectively.^{Reference Toivola, Halme, Miettunen, Aitola and Lund13}

Kang et al.^{Reference Kang, Park, Ryu, Chang and Kim15} presented a DSSC flexible SC reporting an efficiency of 4.2%, with the following structure: ITO and SiO₂-coated stainless steel as a substrate with a deposited layer of TiO₂.

Flexible SCs are a good candidate for a group of highly interesting applications, such as portable and wearable electronics, synthetic skin, and conformal SC for building integration.^{Reference Schubert and Werner16,Reference Kaltenbrunner, White, Głowacki, Sekitani, Someya, Sariciftci and Bauer17} However, this cannot be achieved using traditional materials (e.g., ITO), and graphene has raised as a promising material for flexible transparent electrodes, due to its extraordinary carrier mobility, good transparency and exceptional mechanical flexibility.^{Reference Kim, Zhao, Jang, Lee, Kim, Kim, Ahn, Kim, Choi and Hong18–Reference Bae, Kim, Lee, Xu, Park, Zheng, Balakrishnan, Lei, Kim, Song, Kim, Kim, Özyilmaz, Ahn, Hong and Iijima22}

Liu et al.^{Reference Liu, Li and Yan23} reported a device with the following configuration: PI/Cu/P3HT:PCBM/PEDOT:PSS/Graphene/PMMA with the following results: current density (J _sc) = 10.48 mA/cm²; V _oc = 0.38 V, fill factor (FF) = 34.5% and efficiency (η) = 1.39%.

Table I describes a list of prominent publications on the development of flexible DSSCs. The maximum conversion efficiency obtained for fully flexible metal substrate-based DSSCs was 8.6%, whereas DSSCs with polyethylene naphthalate (PEN) plastic substrates has reached 8.1% efficiency.

TABLE I.

A list of proeminent publications on flexible DSSCs.

In comparison with Si-based SC, SCs employing organic semiconductors are as efficient and less expensive with an easier production, a high flexibility, and very light, they could be easily integrated into structures already in existence. As example, an efficiency of 5.21% was reported by Konarka Technologies,^{Reference Kim, Cook, Tuladhar, Choulis, Nelson, Durrant, Bradley, Giles, McCulloch, Ha and Ree30} with an active layer of 1.024 cm². There are several examples of companies already producing flexible DSSC-based solar module that have the ability to be built in several devices like camping lanterns, tents that generate their own electricity and so forth.^{Reference Pagliaro, Palmisano and Ciriminna11}

Graphene-based materials have emerged, recently, as a single class of carbon materials with a fairly good energy conversion capability allied to low-cost and high-output fabrication via a solution base exfoliation technique. Graphene has been debated to be suitable for flexible ECs for energy storage.^{Reference He, Chen, Gao, Zhou, Li and Xie14}

With some modification of graphene's electronic properties and the improvement of its production techniques, graphene films have the capability of substitute indium tin oxide (ITO) as conductive and flexible electrodes.^{Reference Li, Zhu, Wang, Cao, Wei, Li, Jia, Li, Li and Wu31}

A. Mechanical exfoliation

Mechanical exfoliation is a simple peeling process that can be summarized in the following steps: a highly oriented pyrolytic graphite sheet is dry etched in oxygen plasma; then it is fixed onto a photoresist and skinned off; afterward the thin flakes left on the photoresist are cleaned off in acetone and finally transferred to a silicon wafer. These thin flakes were composed of monolayer or even few layers of graphene, although it lead to several exciting discoveries,^{Reference Novoselov, Geim, Morozov, Jiang, Zhang, Dubonos, Grigorieva and Firsov34} it is limited by its low production capability.^{Reference Singh, Joung, Zhai, Das, Khondaker and Seal33}

B. Liquid-phase exfoliation (LPE)

LPE is yet another attractive method, because it is direct and simple, production of graphene happens just by treating of graphite powders with a solvent.^{Reference Bourlinos, Georgakilas, Zboril, Sterioti and Stubos35} This method typically involves the exposition of graphite or graphite oxide powders to specific solvents, which are then exposed to sonication. LPEs can be divided in three categories: exfoliations from graphite oxide, pristine graphite, expanded graphite, and sonication-free LPE.^{Reference Cui, Zhang, Hao and Hou36}

C. Chemical oxidation

In counterpart, the chemical oxidation of graphite and the succeeding exfoliation provide huge amount of graphite oxide monolayer generating structural defects due to the chemical treatment.^{Reference Stankovich, Dikin, Piner, Kohlhaas, Kleinhammes, Jia, Wu, Nguyen and Ruoff37,Reference Eda, Fanchini and Chhowalla38} These defects change graphene to semiconductive, changing the electronic structure. That is why physical approaches are desirable where it is require that graphene maintains its structure. The drawbacks are the high cost and boiling point of the solvent which makes the successful deposition of graphene very difficult.^{Reference Singh, Joung, Zhai, Das, Khondaker and Seal33} Other examples/processes of exfoliation techniques are available on the literature, such as: mechanical exfoliation intercalating small molecules.^{Reference Kang, Leng and Zhang39–Reference Li, Zhang, Bai, Sun, Wang, Wang and Dai42}

D. Chemical vapor deposition (CVD)

Additionally to the methods presented above, several promising approaches to produce graphene sheets have been presented, for example growth from SiC and CVD. CVD has been cataloged as one of the best techniques for large-scale production, Somani and co-workers used camphor as precursor on Ni foils, reported to be the first to synthesize few layer graphene by this method.^{Reference Somani, Somani and Umeno43} Several reports on the literature show the amazing progress that has been made using several different metals to obtain graphene.^{Reference Kim, Zhao, Jang, Lee, Kim, Kim, Ahn, Kim, Choi and Hong18,Reference Li, Cai, An, Kim, Nah, Yang, Piner, Velamakanni, Jung, Tutuc, Banerjee, Colombo and Ruoff21,Reference Stankovich, Dikin, Piner, Kohlhaas, Kleinhammes, Jia, Wu, Nguyen and Ruoff37,Reference Cao, Yu, Colby, Pandey, Park, Lian, Zemlyanov, Childres, Drachev, Stach, Hussain, Li, Pei and Chen44–Reference Wang, Li, Zhong, Zhong and Zhao48}

E. Plasma enhanced CVD (PECVD)

Yet another alternative to production of graphene at a lower temperature and very short deposition time (<5 min), is PECVD. Through the production of “nanostructured graphite-like carbon” using a DC discharge PECVD, thick graphite structures were reported. Since this first testimony of the production of monolayer and few-layer graphene by PECVD, a great effort has been made to comprehend the growth mechanism, and the subsequent optimization of the experimental conditions, to control graphene films' thickness.^{Reference Malesevic, Vitchev, Schouteden, Volodin, Zhang, Tendeloo, Vanhulsel and Haesendonck49–Reference Zhu, Wang, Holloway, Outlaw, Zhao, Hou, Shutthanandan and Manos51}

F. Other alternative methods

An alternative to all of these techniques above to obtain few-layer graphene is thermal decomposition on SiC and other substrates. An interesting approach for the semiconductor industry is the fabrication of graphite through ultrahigh vacuum (UHV) annealing of SiC surface, this is due to the absence of the need to transfer the deposited material before processing devices.^{Reference Forbeaux, Themlin and Debever52–Reference Varchon, Feng, Hass, Li, Nguyen, Naud, Mallet, Veuillen, Berger, Conrad and Magaud55} At the time of the temperature raise of the SiC substrate under UHV, silicon atoms sublimate leaving carbon atoms on the surface to be integrated in the existing graphene layers. The thickness obtained is determined by the annealing time and temperature used on the deposition process.^{Reference Penuelas, Ouerghi, Lucot, David, Gierak, Estrade-Szwarckopf and Andreazza-Vignolle56}

Although it is interesting, this process has several downsides such as: control of graphene's thickness, different epitaxial growth patterns, and finally, the uncertain reason for the variation of the electronic properties on the interface existing between the substrate and graphene.^{Reference Varchon, Feng, Hass, Li, Nguyen, Naud, Mallet, Veuillen, Berger, Conrad and Magaud55}

In an opposite side there is chemically derived graphene (derived from graphite), that has been pointed as an interesting method to obtain graphene-based single sheets in relevant amounts.^{Reference Stankovich, Dikin, Dommett, Kohlhaas, Zimney, Stach, Piner, Nguyen and Ruoff57–Reference Gómez-Navarro, Weitz, Bittner, Scolari, Mews, Burghard and Kern61} The result of this derivation is graphene oxide (GO), typically synthesized through the oxidation of graphite using oxidants. A number of oxidant agents can be numerated: concentrated sulfuric acid; nitric acid, and potassium permanganate.^{Reference Hummers and Offeman62}

Using pristine graphene as a reference, GO behaves in a different way in several aspects: it is heavily oxygenated bearing hydroxyl and epoxy groups on sp³ hybridized carbon; it is highly hydrophilic and readily exfoliated in water. Given that a stable dispersion is one of the requirements.^{Reference Singh, Joung, Zhai, Das, Khondaker and Seal33}

Chemical reduction of GO has been performed with several reducing agents including hydrazine^{Reference Eda, Fanchini and Chhowalla38,Reference Gómez-Navarro, Weitz, Bittner, Scolari, Mews, Burghard and Kern61,Reference Becerril, Mao, Liu, Stoltenberg, Bao and Chen63,Reference Lee, Park, Ruoff and Dodabalapur64} and sodium borohydrate.^{Reference Lee, Park, Ruoff and Dodabalapur64–Reference Shin, Kim, Benayad, Yoon, Park, Jung, Jin, Jeong, Kim, Choi and Lee66} Taking a different path to reduce GO to reduced GO (RGO), a thermal reduction can be performed, using the heat treatment to eliminate the oxide functional groups from GO surfaces.^{Reference Schniepp, Li, McAllister, Sai, Herrera-Alonson, Adamson, Robert, Car, Seville and Aksay59–Reference Hummers and Offeman62,Reference Lee, Park, Ruoff and Dodabalapur64–Reference Dubin, Gilje, Wang, Tung, Cha, Hall, Farrar, Varshneya, Yang and Kaner81}

Surface functionalization of GO is yet another important method to keep control of the behavior of GO and RGO under exfoliation. Various examples can be found in the literature.^{Reference Eda, Fanchini and Chhowalla38,Reference Stankovich, Piner, Nguyen and Ruoff82–Reference Lu, Yang, Zhu, Chen and Chen101} Other synthesis approaches include: total organic synthesis; unzipping CNTs.^{Reference Singh, Joung, Zhai, Das, Khondaker and Seal33}

To these days, the only alternative to pristine graphene used was RGO, to accomplish this the use of toxic hydrazine vapor, and high-temperature annealing is required to transform GO into RGO. An easier way to obtain a successful dispersion in solvents to acquire monolayer RGO in large quantities in comparison to GO, is using RGO. RGO provides a better solvent solubility and dispersion due to its chemically functionalized groups. Despite that, a large number of defects still remain after the reduction, degrading the electronic properties. Nonetheless a fairly good quality graphene can be obtained by LPE of graphite powders using organic solvents.^{Reference Kaltenbrunner, White, Głowacki, Sekitani, Someya, Sariciftci and Bauer17,Reference Park, Rowehl, Kim, Bulovic and Kong19,Reference Luo, Vora, Mele, Johnson and Kikkawa102}

LPE using organic solvents is a method that has several desirable characteristics that could turn it in an industrial solution, it is not very time consuming and cheap as well. Additionally, it has amazing high charge mobility and electronic conductivity, arousing interest in a group of specific applications like: PVs, photocatalyst, nanoelectronics, sensors, and so forth.^{Reference Cheng, Hu, Zhao and Qu8,Reference Rothberg and Lovinger103}

The ability of GO sheet to form liquid crystals, with lamellar ordering or long-range helical frustrations simultaneously has led to GO fibers by continuously spinning this material. The successful fabrication of GO fibers was obtained by injecting the GO dispersions into a solution of 5 wt% NaOH/methanol, being the GFs further collected by chemical reduction in hydroiodic acid.^{Reference Cheng, Hu, Zhao and Qu8}

CVD deposition can also be used to obtain GFs, these fibers are from graphene films grown and transferred to an organic solvent. Then the solvent was evaporated to obtain the fiber, the evaporation rate has a significant impact on the porosity of the fibers.^{Reference Cheng, Hu, Zhao and Qu8} The following table summarizes the different characteristics of GFs obtained by different methods (Table II).

TABLE II.

GFs obtained from different methods (adapted from^{Reference Cheng, Hu, Zhao and Qu8}).

Abbreviations: CNT — carbon nanotube; CVD — chemical vapor deposition; GF — graphene fiber; GO — graphene oxide; GONR — GO nanoribbon.

^a After chemical reduction.

^b After annealing at high temperature.

Of all methods mentioned above, the most promising one is self-exfoliation without sonication, with several strong points mentioned like its low-energy input, high-quality, and high-concentration output. Principal challenges are the uncontrollable size and layer number, limiting the applications of graphene. All of this must be achieved avoiding graphene aggregation, aiming for monolayers of graphene or graphene nanoribbons.^{Reference He, Lin, Liu, Zhu, Zhao, Chan and Yan20}

A. Characterization techniques

Graphene characterization can be achieved by a set of different techniques: optical imaging of graphene layers; Fluorescence quenching microscopy (FQM); atomic force microscopy (AFM); transmission electron microscopy (TEM) and Raman spectroscopy.^{Reference Singh, Joung, Zhai, Das, Khondaker and Seal33}

Often, to study the different layers of graphene, it is necessary to combine two or more techniques. Initially, the optical microscope was the most used to image the several existing layers since it was of low-cost and already accessible in laboratories worldwide. It was necessary to have a good contrast, to improve the visibility of thin sheets.^{Reference Jung, Pelton, Piner, Dikin, Stankovich, Watcharotone, Hausner and Ruoff104–Reference Ni, Wang, Kasim, Fan, Yu, Wu, Feng and Shen106}

This in fact was a limiting factor because of the substrate that had to be used (Silicon Oxide). Another limiting aspect that affects contrast was the wave length of the incident light. The following image (Fig. 3) shows the SEM and optical image of a graphene film.^{Reference Lotya, Hernandez, King, Smith, Ronan, Nicolosi, Karlsson, Blighe, De, Zhiming, McGovern, Duesberg and Coleman107}

FIG. 3.

(a) SEM and (b) optical images of the surface of a graphene film. Reproduced with permission.^{Reference Lotya, Hernandez, King, Smith, Ronan, Nicolosi, Karlsson, Blighe, De, Zhiming, McGovern, Duesberg and Coleman107}

FQM procedure can also be used to study graphene, RGO, and GO, for the improvement of the synthesis process. The mechanism involves quenching the emission from a dye-coated GO and RGO, the interaction between the dye and the GO causes quenching of fluorescence.^{Reference Treossi, Melucci, Liscio, Gazzano, Samorì and Palermo108} On the image presented bellow (Fig. 4), some results obtained from a FQM analysis are exhibited.

FIG. 4.

(a) AFM image screening G–O single layers deposited on a SiO₂/Si wafer before application of a 30 nm thick fluorescein/PVP layer for FQM. (b) FQM image of the area of the wafer, (c) after washing off the dye coating, (d) line scan data on a folded sheet show no significant deviation in thickness before and after FQM imaging (all scale bars ±10 µm). Reproduced with permission.^{Reference Kim, Cote, Kim and Huang150}

This technique has the drawback of being limited to a specific size of the layers (micron range) due to the limited diffraction of FQM's lateral resolution, also because of the dye addition the sample cannot be re-used.^{Reference Singh, Joung, Zhai, Das, Khondaker and Seal33}

To obtain a nanometer scale layer thickness determination, one can use AFM. AFM can only give topographic contrast and so the successful identification of the GO and the graphene cannot be achieved in normal operation mode. The impact of attractive mode of AFM to conclude on the thickness of the sheet was proved by Paredes et al.^{Reference Paredes, Villar-Rodil, Solís-Fernández, Martínez-Alonso and Tascón109} It was established that an error in height measurement was caused by the deformation induced by the repulsive mode. Besides imaging and thickness detection this technique has also permitted the study of a certain number of graphene properties: mechanical, frictional, electrical, magnetic, and even elastic.^{Reference Singh, Joung, Zhai, Das, Khondaker and Seal33}

In the case of graphene and RGO, that is an atom thick, there is only one tool that can be used to determine its properties, TEM. However, TEM also has its limitations, especially at low operating voltage (to prevent damages to the monolayer), having a poorer resolution. More recently, graphene's matrix illustrating every single carbon atom arranged in hexagonal structures has been showed by Mayer's et al.^{Reference Meyer, Kisielowski, Erni, Rossell, Crommie and Zettl110}

Yet another imaging technique was used by Gass et al., showing the atomic lattice using scanning TEM (STEM).^{Reference Gass, Bangert, Bleloch, Wang, Nair and Geim111} This method allows the detection of defects and contaminant atoms at atomic scale, the images presented on Fig. 5 are an example of a clean graphene monolayers surrounded by contaminants, individual contaminant atoms of higher atomic number can be seen in (b). Graphene can serve as support material in TEM since it is one atom thick and also due to its crystallinity and high conductivity.^{Reference Singh, Joung, Zhai, Das, Khondaker and Seal33}

FIG. 5.

High-resolution images of monolayer graphene. Bright-field (a) and high-angle annular dark-field imaging (HAADF) (b) images of the monolayer. Reproduced with permission.^{Reference Gass, Bangert, Bleloch, Wang, Nair and Geim111}

Like any other carbon allotrope, under Raman spectroscopy analysis, graphene can also be characterized, allowing the characterization of graphene layers evaluating several aspects, like: number, effect of strain, doping concentration, effect of temperature and presence of defects. The D band is usually associated with the presence of disorder in atomic arrangement of edge defect, wrinkles and charge pools. On the graphs presented bellow (Fig. 6), a comparison between graphite, single and few layers graphene is presented.

FIG. 6.

(a) Comparison of Raman spectra between graphite and single layer graphene. (b and c) Evolution in 2D band as a function of layers at 514 and 633 nm excitations. Reproduced with permission.^{Reference Ferrari, Meyer, Scardaci, Casiraghi, Lazzeri, Mauri, Piscanec, Jiang, Novoselov, Roth and Geim151}

The absence of the D peak confirms the absence of defects on graphene. The single sharp 2D peak corresponds to the monolayer that is about four times more intense than the G peak. Raman spectroscopy has the advantage of being a nondestructive tool.^{Reference Singh, Joung, Zhai, Das, Khondaker and Seal33}

B. Properties of graphene

Graphene is a single-layer arrangement of two-dimensional graphite, it is very transparent and absorbs just 2.3% of incident light.^{Reference Reina, Jia, Ho, Nezich, Son, Bulovic, Dresselhaus and Jing112} Additional to the amazing properties stated along this document, graphene has also the maximum hypothetical specific surface area ratio and half integer quantum hall effect even at ambient temperature, which has aroused great interest on the research community. This section will review graphene properties in a resumed way, because the debate of such properties and the amazing range of applications have the potential to become infinite, going beyond the purpose of this work.^{Reference Singh, Joung, Zhai, Das, Khondaker and Seal33}

The reasons behind single layer graphene's high electronic conductivity is its high quality and low defect density of its crystal lattice, this generates scattering sites and inhibits charge transfer. Kim and co-workers showed that the electron mobility is affected by impurities trapped in between the graphene and substrate.^{Reference Bolotin, Sikes, Jiang, Klima, Fudenberg, Hone, Kim and Stormer113}

Large area epitaxial graphene has an interesting feature, the possibility of a scale up for electronic devices, even few layers graphene on SiC showed great electronic properties but still with charge transport mobility smaller by one order compared to pristine graphene. De Heer discovered that graphene grown on carbon terminated face has a greater mobility than that that is grown on Si terminated face, and also that energy gap decreases with increasing thickness.^{Reference de Heer, Berger, Wu, First, Conrad, Li, Li, Sprinkle, Hass, Sadowski, Potemski and Martinez54,Reference Peng and Ahuja114–Reference Kim, Ihm, Choi and Son116}

Several studies^{Reference Kim, Zhao, Jang, Lee, Kim, Kim, Ahn, Kim, Choi and Hong18,Reference Cao, Yu, Colby, Pandey, Park, Lian, Zemlyanov, Childres, Drachev, Stach, Hussain, Li, Pei and Chen44,Reference Reina, Jia, Ho, Nezich, Son, Bulovic, Dresselhaus and Jing112,Reference De Arco and Zhang117,Reference Yu, Lian, Siriponglert, Li, Chen and Pei118} were performed, using nickel foil as the substrate with good reported results for large area, high quality mono, and few layer graphene. Nickel foil limits the graphene films to reduced sized grains, deposition of several layers at grain boundaries and carbon solubility. Good results in copper were obtained by Ruoff and co-workers^{Reference Li, Cai, Colombo and Ruoff119} due to carbon's low solubility in copper and its predominantly growth in single layer, achieving mobility of 4050 cm² V⁻¹ s.

Nonetheless, in other reported work^{Reference Castro, Novoselov, Morozov, Peres, Dos Santos, Lopes, Nilsson, Guinea, Geim and Neto120} in which the electronic band gap of the bi-layer graphene was tuned by applying an electric field, this provides a possibility for its usage for many electronic applications.^{Reference Tonouchi121–Reference San-Jose, Prada, McCann and Schomerus123}

A promising approach that can provide graphene in scalable amount is the use of chemical and thermal RGO.^{Reference Stankovich, Dikin, Piner, Kohlhaas, Kleinhammes, Jia, Wu, Nguyen and Ruoff37,Reference Li, Muller, Gilje, Kaner and Wallace68} Continuous effort has been made with the objective of improving the electrical transport properties of RGO. The conductivity and mobility of RGO compared to pure graphene, was stated to be 3 and 4 orders of magnitude respectively smaller.^{Reference Singh, Joung, Zhai, Das, Khondaker and Seal33}

C. Optical properties

As stated before, single-layer graphene absorbs only 2.3% of incident light over a wide range of wavelengths as one can see in Fig. 7, where it is exhibited the UV spectra of graphene films with and without HNO₃ doping.^{Reference Bae, Kim, Lee, Xu, Park, Zheng, Balakrishnan, Lei, Kim, Song, Kim, Kim, Özyilmaz, Ahn, Hong and Iijima22,Reference Nair, Grigorenko, Blake, Novoselov, Booth, Peres, Stauber and Geim124} The absorption of light increases linearly with the addition of layers, each layer absorption A = 1 − T = πα = 2.3%, where α ≈ 1/37 is the fine structure constant.^{Reference Singh, Joung, Zhai, Das, Khondaker and Seal33}

FIG. 7.

Transmittance of different graphene layers on quartz substrates. Reproduced with permission.^{Reference Bae, Kim, Lee, Xu, Park, Zheng, Balakrishnan, Lei, Kim, Song, Kim, Kim, Özyilmaz, Ahn, Hong and Iijima22}

From 300 nm to 2500 nm the absorption of monolayer graphene is constant, showing a peak at ≈250 nm because of the electronic transition of the inter band from the unoccupied π* states (Fig. 7).^{Reference Kravets, Grigorenko, Nair, Blake, Anissimova, Novoselov and Geim125} Graphene's optical transition is tunable by changing the Fermi energy, this adjustability provides the possibility for the use in IR detectors, modulators and emitters.^{Reference Wang, Zhang, Tian, Girit, Zettl, Crommie and Shen122}

Combination of electronic transport properties and optical properties aroused a lot of interest because of the possibility of using large area monolayer and few layer graphene as ultrafast photodetectors.^{Reference Xia, Mueller, Golizadeh-Mojarad, Freitage, Lin, Tsang, Perebeinos and Avouris126,Reference Rana, George, Strait, Dawlaty, Shivaraman, Chandrashekhar and Spencer127}

Photoluminescence (PL) is yet another optical property of graphene, by inducing a suitable band gap. This can be accomplished by two purposed methods: cutting graphene in nanoribbons and quantum dots, or using a physical or chemical treatment with different gases to reduce the conductivity of the π electron network.^{Reference Park and Ruoff128–Reference Bonaccorso, Sun, Hasan and Ferrari130} Several examples can be found in the literature,^{Reference Luo, Vora, Mele, Johnson and Kikkawa102,Reference Rothberg and Lovinger103,Reference Gokus, Nair, Bonetti, Bohmler, Lombardo, Novoselov, Geim, Ferrari and Hartschuh131,Reference Sheats, Antoniadis, Hueschen, Leonard, Miller, Moon, Roitman and Stocking132} opening new avenues for various applications from touchscreens to optical emitters (Fig. 8).

FIG. 8.

(a) Optical photo of the twisted structure; (b) and (c) SEM images of the same structure; (d) sectional view. Reproduced with permission from Ref. Reference Fan, Chu, Wang, Zhang, Chen, Tang and Zou144.

D. Mechanical properties

Following CNTs, graphene has been stated to have the highest elastic modulus and strength. This is an important factor since any outside stress on a crystalline material has the ability to alter interatomic distances, modifying electron transport. Table III summarizes the mechanical properties of single, bilayer, and multilayer of graphene.^{Reference Singh, Joung, Zhai, Das, Khondaker and Seal33}

TABLE III.

Thermal and mechanical properties of graphene.

E. Thermal properties

In electronic devices a considerable amount of heat needs to be dissipated, during the device operation period. Graphene, like any other carbon allotrope, shows a high thermal conductivity. The maximum room temperature thermal conductivity of 5000W/mK was obtained for pure defect free graphene.^{Reference Balandin, Ghosh, Bao, Calizo, Teweldebrhan, Miao and Lau133}Table III also shows the comparison of the thermal conductivity for graphene obtained by several methods. Thermal conductivity is stated to change by the influence of two specific factors: defects edge scattering^{Reference Nika, Pokatilov, Askerov and Balandin134} and isotopic doping.^{Reference Jiang, Lan, Wang and Li135}