Graphene is one of the hottest topics in the fields of materials science, chemistry and condensed matter physics [1, 2]. It represents a carbon monolayer packed into a two-dimensional (2D) honeycomb lattice with a hybridized sp2 bonding orbitals. Defect-free graphene is the basic form of 2D carbon family. Various graphene configurations have different properties, depending on factors such as morphology, sub-dimensions, the number of layers and defects [1, 3, 4]. Graphene offers unique functionalities and characteristics such as high mechanical properties (~1 TPa Young’s module and ~130 GPa intrinsic strength) , high surface area (up to 2,675 m2.g-1) , high electrical and thermal conductivity (0.5-100 S.m-1 and 3,080-5,150 W.m-1.k-1, respectively) and high electron mobility at room temperature (200,000 cm2.V-1.S-1) . This makes graphene an extraordinary material of choice that can be used in high performance energy storage devices. Graphene has been used as an active or inactive component in electrochemical energy-storage devices (EESDs). It can be actively involved in electrochemical reactions of EESDs. It can also be an inactive component of EESD by only providing electrical conductivity and without being actively involved in the electrochemical reaction. Graphene-based EESDs that provide high power density, high temperature resistance, long cycle life and high energy density have gained increasing attention recently. This is, in part also, due to increase in demand for portable electronic devices, hybrid vehicles, and clean energy storage, which has become a major concern in today’s world as population rises globally.
Although graphene is theoretically an excellent material for energy storage, it practically cannot serve this purpose efficiently. The main reason is that graphene nanosheets tend to stack up due to weak Van der Waals attractions between individual layers . This leads to the reduced specific surface area and as a result, hindering the ion exchange between the electrode and electrolyte. One solution is to synthesise three-dimensional (3D) structures of graphene to facilitate the access and diffusion of ions . For instance, intercalation of nano crystals into the interlayers of graphene or graphene oxide (GO) has been proven to be effective on reducing the re-stacking issue of graphene flakes (Figure 1) .
Graphene is able to host Li ions (Li+) two times higher than conventional graphite when used as an anode in lithium (Li) batteries [11, 12]. Theoretically, graphene can adopt Li+ on both sides. However, this does not occur in the real practical situation because of the repulsion forces between Li+ and graphene . One way to improve the storage capacity is to form graphene hybrid composites (GHCs) with other electroactive materials. GHCs have been demonstrated to provide considerably higher storage capacities than that of bare graphene (Table 1).
Similar to Li+ batteries, GHCs in sodium ion (Na+) batteries enable higher specific capacity, better rate capability and longer cycle life than bare graphene . Electrochemical capacitors (ECs) or supercapacitors are one type of EESDs that are being used in the portable electronic devices because of their faster charge/discharge, greater power density, and cyclability [36-39]. Typically, ECs are classified into two types: electrochemical double-layer capacitors (EDLCs) and pseudo-capacitors. EDLCs are based on the ion adsorption on electrodes whereas the operation of pseudo-capacitors is based on the phenomenon of fast surface redox reactions [40-43]. The preferable electrode materials of EDLCs should have high surface area to maximize ion adsorption. Graphene is considered as an excellent electrode material for EDLCs because of its high surface area, good corrosion resistance, excellent thermal stability, and high conductivity. On the other hand, a hybrid form of graphene-based composites could offer great promise for electrochemical pseudo-capacitors. For example, Zhang et al. investigated a 3D hybrid composite of nickel-aluminum layered hydroxide-graphene nanosheets (NiAl-LDH/GNS) as an advanced electrochemical pseudo-capacitor and a high specific capacitance, high rate capability and exceptional cycling ability was achieved.
There are large variety of techniques available for the preparation of graphene including thermal decomposition of GO followed by chemical reduction, CVD growth, unzipping carbon nanotubes (CNTs), mechanical cleavage of graphite, and electrochemical exfoliation of graphite . Different fabrication methods yield different graphene structures with varied layer number and type of defects, which indicates a strong correlation between graphene preparation methods and resultant properties.
Physical and chemical properties of graphene-based nanomaterials are also strongly influenced by the presence of defects on graphene. Hence, defects play an important role in graphene’s applications. It has been postulated that defects could limit the ultimate performance of graphene in energy storage devices. However, it has been observed that properly engineered defects can increase the performance of graphene for energy storage purposes .
The aim of this review paper is to discuss the role and importance of graphene in energy storage and conversion devices by introducing different types of 2D and 3D graphene-based hybrid composites (GHCs). In addition, it also aims to outline the recent works that have been done by various researchers to improve the performance of graphene in EEDs.
2D graphene nanosheets can be fabricated by thermal or chemical reduction of GO. Due to the strong π-π interaction, GO nanosheets has high tendency to re-stack and form agglomeration [47-52]. As a result, the intrinsic surface area of graphene diminishes, which leads to poor performances. To solve this problem, there have been attempts to fabricate GHCs by incorporation graphene with other functional materials. The new structures of GHCs could effectively inhibit the re-stacking of graphene.
Various GHCs have been prepared by introducing hydroxides, transition metal oxides (TMOs) and conducting polymers [16, 53, 54]. These composites support redox reactions by forming a conducting network. Thus, the electrochemical performance of GHCs is enhanced. Several structures from graphene can be assembled to be used as electrode materials in EESDs. Most recently, a significant body of researches have focused on manipulation of both 2D and 3D structures of graphene to enhance its electrochemical performance.
Graphene with 2D structure and fascinating physical, mechanical and chemical properties is the most suitable candidate for the next generation of flexible thin-film EESDs. In fact, graphene nanosheets can fulfil the requirement of a flexible thin-film EESD such as being lightweight and functioning well under bending or folding conditions. However, agglomeration is the most drawback of graphene nanosheets in such applications. Therefore, graphene should be processed in such a way that overcomes this problem. To that end, several attempts have been made through modification procedures such as template assisted growth, crumpling of graphene sheets, and additive intercalated graphene nanosheets [55-58]. Among them, intercalation of additive materials has been shown to be the most effective way to reduce re-stacking of graphene nanosheets. The additive materials that can be integrated into graphene nanosheets to make GHCs could be of an organic or inorganic type. Those additive materials can be incorporated to increase the performance of graphene based EESD. Furthermore, results indicate that if defects of graphene nanosheets are engineered properly, the performance of graphene-based energy storage devices can be improved even beyond the theoretical limits .
Hence, increasing the active surface and electrochemical reaction areas acquired by integration of intercalating organic/inorganic materials and/or defects in graphene can enhance the energy and power density of GHCs [59-61].
Basically, there are three ways to functionalize the surface of graphene and tune its chemical properties including non-covalent, intercalation and covalent functionalization (Figure 2) . Graphene-based organic hybrid composites are prepared by covalent or non-covalent cross-linking of graphene or GO nanosheets, using organic cross-linkers . Functionalization of redox active organic materials on the graphene sheet may lead to unique architectures (Figure 2). This reduces the diffusion path length, agglomeration and re-stacking of graphene nanosheets. Several covalent and non-covalent routes for functionalization of graphene nanosheets have been reported [63-65]. Most of the covalent functionalization lead to conductivity decrease of graphene . This is because covalent functionalization creates a strong bonding between graphene nanosheets and organic molecules. The carbon bonding site changes from sp2 to sp3 as soon as covalent functionalization arises (Figure 2c). This connection results in the removal of π electrons from the carbon atom in graphene and reducing the carrier density . In comparison, non-covalent functionalization does not alter much to the graphene π-π bonding system. Hence, graphene conductivity is sustained after non-covalent functionalization [36, 64, 67].
The organic cross-linkers used in the covalent linking method can be bifunctional such as 1,4-diisocyanato-benzene, 3,3′-diaminobenzidine, 4-Aminobenzene sulfonic acid or multifunctional such as rGO fibers and yarns, and poly(methyl methacrylate) (PMMA) molecules [70-74]. The present functional groups on bifunctional and multifunctional molecules react with carbon skeleton and oxygen containing groups of graphene or GO nanosheets. In contrast, there are a number of driving forces for non-covalent linking of GO and organic cross-linker, namely, electrostatic interaction, hydrophobic interaction, chemisorption, π-π stacking, and hydrogen bonding [75, 76].
The conducting polymers such as polyaniline (PANI), polystyrene (PS), poly(3,4-ethylenedioxythiophene) (PEDOT), polyethylene terephthalate (PET) and poly (methyl methacrylate) (PMMA) have mostly been explored as organic cross-linkers. The specific capacitance of these hybrid composites lies in the range of 300 to 1,500 F.g-1. Table 2 shows the electrochemical performance and main findings of some graphene-based polymer hybrid composites.
The fabrication method sometimes could affect the performance of the composite. For instance, graphene/PANI composite prepared by in situ anodic electropolymerization of aniline monomer on graphene paper exhibited 233 F.g-1 gravimetric capacitance and 135 F.cm-3 volumetric capacitance . Whereas graphene/PANI composite fabricated by coating PANI onto graphene sheets, exhibits a specific capacitance of 1,046 F.g-1. Even when the current density was increased from 10,000 to 100,000 mA.g-1, its capacitance value was maintained at 96% which indicates its superior retention ability [86, 88]. Electrochemical performance of graphene could also be improved by intercalating polypyrrole (PPy) spheres into graphene layers . This created a hollow structure and increased capacitance value to 500 F.g-1 with charging/discharging current density of 5,000 mA.g-1 .
PEDOT and its derivatives are well known for their high electrical conductivity, chemical stability, thermal stability . Electropolymerization can be used to combine PEDOT to GO and GO can be further reduced to rGO through electrochemical reduction. The pseudocapacitance of PEDOT and large accessible surface area of rGO led to the capacitance of 14 F.cm-2 of rGO/PEDOT devices .
Graphene/Metal-organic framework (MOF) 2D hybrid composites are one of the most promising electrode materials for a broad range of low-cost and highly efficient energy storage and conversion devices [91, 92]. Xia et al.  developed a novel N-doped graphene aerogel (NG-A) assisted method to prepare metal oxide nanoparticles (NPs) from bulk cobalt-based MOF [Co(mIM)2 (mIM = 2-methylimidazole)] (Figure 3). The synthesized structure served as a robust platinum free electrocatalyst with an excellent activity and stability for the oxygen reduction reaction in an alkaline electrolyte solution. Furthermore, when cobalt was eliminated from the structure, the N-rich porous carbon-graphene aerogel composite exhibited extraordinary energy storage with exceptional capacitance and rate capability in a supercapacitor. This novel method can provide a strategy for fabricating advanced nanostructure from MOFs. Moreover, it can provide a new pathway for fabricating high performance electrochemical materials that satisfy both energy storage and energy conversion devices from a single MOF/NG-A derived nanostructure.
Zhang et al.  reported that the cross-linking of GO sheets with organic diisocyanates led to the formation of graphite oxide hybrid porous materials with a lot more porosity compared to powder GO. Therefore, not only cross-linking can increase the porosity, it can also enhance the mechanical stability . The cross-linking of GO paper with glutaraldehyde (GA) by exposing GO paper in the vapors of GA shows better interlayer adhesions. This effect can increase the elastic modulus and strength of composite in comparison with GO paper, by 190% and 60%, respectively .
Conducting polymers have shown to be potential organic cross-linkers for graphene. However, they tend to swell during the charge/discharge process because of ion doping/un-doping . As a result of this, the structure of conductive polymers degrades and its life decreases [64, 96-98]. Therefore, for sustainable energy storage applications, other electrochemically active organic materials that have longer life span and higher energy density with fast reversible redox reactions are being investigated. For example, a variety of quinone structures were attached to carbon substrates, good cyclability and high rate performance were obtained [64, 99-101]. However, there is still a challenge to find the best quinone structure which shows significant redox capacitance, mechanical, thermal stability and adhesion to carbon.
2,5-dimethoxy-1,4-benzoquinone (DMQ) is an electrochemically active organic molecule. It has been reported that DMQ functionalized graphene, exhibited the highest capacitance, up to 650 F.g-1 (780 F.cm-3) at 5 mV.s-1 in aqueous 1 M H2SO4 . It has been shown that 99% of capacitance could be retained after 25,000 cycles at 50 mV.s-1 . Generally, important parameters for enhancing electrochemical performance of a graphene/organic 2D hybrid composite include electrode architecture and the selection of electrochemically active organic molecule.
Metals and metal oxides which fall under the category of inorganic materials have shown great properties for various energy storage and conversion applications when combined with graphene [102, 103]. These tremendous potentials in graphene/inorganic hybrid composites are because of the synergetic results that can be achieved from such composites. The synergetic effects in graphene/inorganic hybrid composites work as follows: (i) graphene can facilitate the uniform nucleation, grown up of metal oxides and their assembly into specific size, shape and crystallinity; (ii) intercalation of metal oxides into graphene layers can prevent the re-stacking of graphene nanosheets; (iii) graphene prevents not metal oxides from agglomeration during the charge/discharge process in energy storage devices; (iv) there will be a good interfacial interactions and electrical contacts between graphene and metal oxides due to oxygen-containing functional groups on graphene nanosheets . Many innovative synthesis techniques have been developed to fabricate graphene/inorganic 2D hybrid composites. Inorganic nanostructure could be incorporated between graphene sheets through chemisorption interaction . Metal sulfide, metal telluride, metal selenide, metal nitride, carbonaceous materials and multi-element compounds incorporated graphene/inorganic hybrid composites have demonstrated their great performance for energy storage and conversion devices [105-107]. Wang et al.  synthesized graphene/cobalt sulfide (CoS2) hybrid composites through a facile solvothermal process which was used for high performance supercapacitors. The graphene/CoS2 nanocomposites displayed a specific capacitance of 314 F.g-1 in 6 M potassium hydroxide (KOH) solution . Carbonaceous materials such as carbon black and CNTs that can be used as a spacer in graphene hybrid composites have been extensively investigated. For instance, Qiu et al . applied CNTs as a spacer between graphene nanosheets and specific capacitance of 140 F.g-1 was achieved at the current density of 100 mA.g-1 in 1 M H2SO4 solution. Carbonaceous materials led to an increase in specific capacitance of graphene when applied as spacers in graphene hybrid composite. However, these spacers are not pseudocapacitive materials where, for instance, charge storage mechanism is based on faradic, redox reactions . Hence, pseudo-capacitive materials with high pseudo-capacitance are required to be used as spacers in graphene/inorganic hybrid composites in order to achieve higher energy density . Ruthenium(IV) oxide (RuO2), Iron(II,III) oxide (Fe3O4), Copper(II) oxide (CuO), Nickel(II) hydroxide [Ni(OH)2], MnO2, Co3O4 are various inorganic pseudo-capacitive materials that can be used as spacers in graphene .
MnO2 is the most common pseudo-capacitive material that can be used as a spacer between graphene layers because of its low cost and low environmental impact . Graphene/MnO2 hybrid composites fabricated by filtration method provided a specific capacitance of 389 F.g-1 in 1 M NaSO4. Table 3 shows the electrochemical/electrical performance and the underlying mechanism of some graphene/inorganic 2D hybrid composites.
Graphene/transition metal oxides (TMOs) hybrid composites have great potentials in energy storage devices such as batteries, supercapacitors, solar cells and catalysis . This is because: (i) TMOs have natural capability to keep charged ions on their surfaces without intermixing ; (ii) graphene/TMOs nanostructures exhibit high conductivity by injecting electron from graphene into TMOs and increasing the hole concentration in graphene ; (iii) graphene/TMOs have extraordinarily high surface area which maximize the interaction with electrolyte and further involvement in electrochemical reactions and charge interactions ; (iv) TMOs nanosheets between graphene nanosheets preventing re-stacking of graphene layers. These are some factors that determine the overall power performance of supercapacitor electrodes, which are suitable in batteries and electrocatalysis as well. Based on these attributes, layered model of graphene/TMOs has been developed to be used in metal-ion battery anodes, cathodes, and supercapacitors . Although specific properties of graphene and TMOs have been revealed and investigated [131-133], we are still far from understanding how we can operate these kinds of materials effectively with high-performance in a heterostructured architecture. This is because, (i) it is important in batteries and supercapacitors that whether the layered TMOs change phase during operation and (ii) how their long- and short-term stability is affected by the intercalation process,besides, (iii) how properties of new TMOs can be boosted by using graphene substrates. To address these challenges, it is important to figure out physical, chemical, and structural characterizations in details. In the case of layered heterostructures (Figure 4), it is also significant to understand the role of interface structure on physical, chemical, and mechanical behavior of hybrid nanostructures. Therefore, the study of various 2D heterostructured graphene/TMOs hybrid composites and demonstration of their performances are expected to achieve fundamental dimensionality effects on the properties of 2D nanomaterials structures and facilitate their usage in energy storage society. In fact, assembly of graphene and TMOs nanosheets into heterostructured architectures can open an opportunity to manufacture high performance electrodes, especially for portable electronics and wearable devices, which have gained great attention recently and are growing fast.
Various efforts have been made to increase the energy density of graphene. Generally, this can be achieved either by increasing the active surface area or by the introducing pseudo-capacitive materials. A new approach is to create defects in graphene. Previously it was thought that defects reduce the graphene performance. Therefore, all endeavors were made to heal the defects in graphene. However, it has been understood that defects on graphene sheets can enhance its electrochemical performance beyond the predictable limits if they are properly engineered . Thus, this has turned the attention of researchers towards the creation and enlargement of defects to enhance the electrochemical performance. The creation of defects on graphene sheets can increase its energy density in energy storage devices. However, structural defects like holes, lattice defects, grain boundaries with broken hexagonal symmetry and merged graphene domains may decrease electrical conductivity and influence its reactivity [65, 135, 136].
Incorporation of pseudo-capacitive materials in graphene results in an increase of pseudocapacitance of GHCs. However, the presence of excessive functional groups on the graphene surface can decrease its electrical conductivity. Thus, distribution and density of these functional groups should be controlled in such a way that it doesn’t limit the performance of graphene as an electrochemical device, it should also increases the charge storage capacity. Commonly, defects in graphene are extrinsic or intrinsic in nature. Stone-Wales, vacancies, adatoms, armchair and zigzag edges are intrinsic defects, whereas doping of graphene with N, boron (B), sulfur (S) or phosphorus (P) atoms falls under the category of extrinsic defects .
N is the most doped heteroatom in carbon when comes to energy applications . N is a neighbor of carbon; thus, it is relatively easy to bring two types of atoms together. Doping of carbon materials with N generally leads to increase in pore size and pore volume and can also lead to the generation of additional pores [139, 140]. N atoms provide necessary pseudocapacitance. Thus, its usage in carbon materials for energy storage applications is promising . S, B and P can be co-doped with N in carbon for energy applications . Graphene co-doped with B and N shows higher specific capacitance than the one doped by B or N atoms alone . Doping of graphene with B (3.95 at%) and N (19.73 at%) exhibited up to 130.7 F.g-1 (200 mA.g-1, 1 M H2SO4) capacitance and stability up to 2,000 cycles . Figure 5 shows mechanisms for enhanced performance of doped graphene in energy storage applications.
To increase the accessible surface area of graphene, defects-induced pores are incorporated in graphene. Porous graphene exhibits capacitance of 200 F.g-1 at a current density of 700 mA.g-1 . However, optimization of the pore size and location is another great challenge. An increase in capacitance was observed by Luo et al.  when hole size of graphene varied from 4.2 Ao to 10 Ao because of easy penetration of diffused ions.
Formation of holes on and through the thickness of graphene sheets can be achieved by removing a large number of atoms from the graphitic plane that results in holey graphene (HG) . It should be noted that there is a fundamental difference between HG and porous graphene. The developed pores in HG are because of large structural defects whereas pores in porous graphene are empty spaces between graphene sheets. The pores in HG make the conduction path of ions easier and ion transport is highly facilitated (Figure 6a). Therefore, HG exhibits higher intrinsic capacitance compared to graphene (Figure 6b) . Figure 6c also shows cross-sectional SEM images for HG and graphene films at equivalent areal mass densities of 0.31 mg.cm-2. It can be observed that the thickness of HG is almost half of the graphene film.
It has been demonstrated that 2D GHCs can provide excellent efficiency in energy storage devices. However, mass transfer, ion access, conductivity and surface area of 2D graphene are limited due to easily agglomeration, re-stacking and blocking the active sites of graphene nanosheets, which leads to poor electrocatalytic properties . Thus, research is getting somewhat inclined towards developing 3D graphene-based network structures to address a way of tackling these issues [9, 38, 150]. In fact, 3D graphene-based network structures exhibit; (i) high electrical conductivity, (ii) low diffusion resistance and easy electrolyte penetration to cations or photons, (iii) high electroactive areas, and (iv) improved structural stability, due to high structural porosity and space between graphene sheets [41, 149, 151]. Therefore, this provides opportunities for 3D graphene-based hierarchical architectures to be served as high efficiency energy storage devices with high stability and power density . For example, Wang et al.  fabricated a 3D few layer graphene/CNTs foam architecture using ambient pressure chemical vapor deposition process (Figure 7a and 7b). The synthesized composite contained a homogenous and densely packed nanostructure with a large surface area (743 m2.g-1), low density (normally < 500 g.m-2), high electrochemical stability (~99% capacitance keep over 8,500 cycles), great specific capacitance (286 F.g-1), excellent energy density (39.72 Wh.kg-1) and a superior power density (~154.67 kW.kg-1). The obtained energy density and power density between the 3D few layer graphene/CNTs foam and some other synthesized composites, including graphene films (Figure 7c), indicated that this unique 3D architecture was promising for future energy storage devices .
Lin et al.  synthesized a robust flexible 3D graphene sponge/S NPs composite (G/SNPs), through a facile and environmentally friendly reduction method induced by a self-assembly technique (Figure 8a). They have demonstrated that the 3D G/SNPs structure offered a fast Li+ transport, excellent electrolyte absorbability and efficient electrochemical redox reactions. Therefore, 3D G/SNPs provided a stable capacity of 580 mA.h.g-1, after over 500 cycles at a high rate of 0.0015 mA.g-1, corresponding to a high capacity retention of 78.4% and to a low fading rate of 0.043% per cycle (Figure 8b) . These results are much better than those of rGO/S composite cathode. The flexible 3D G/SNPs design allows the increase of S mass loading over the whole electrode which can be applied in flexible Li-S batteries .
Progress has been made in the design of 3D GHCs for ultrahigh-rate energy storage at practical levels of mass loading (more than 10 milligrams per square centimeter). Sun et al.  fabricated a 3D HG/niobia (Nb2O5) nanostructure that provided excellent electron transport properties. Figure 8 is the demonstration of the two-step process to synthesise 3D hierarchically porous composite structure. The hierarchical porous structure also facilitated quick ion transport . The obtained results indicated that if the porosity in the holey graphene backbone was systematically tailored, then charge transport in the hierarchical structure could be optimized to deliver high areal capacity and high-rate capability at practical levels of mass loading (from 1 to 11 mg.cm-2). The 3D HG/Nb2O5 composite displayed at rates of as high as 10 C, there was a small difference in specific capacity for mass loadings in practical device applications. The high area capacity with high-rate capability achievements at wide mass loading indicates a vital step towards practical EESDs .
It has been reported that thermal conductivity of a single layer graphene is in the range of 1500-5800 W.m-1.k-1 [153, 154]. This makes graphene an efficient material for thermal management in microelectronic devices, where these electronic components are shrinking ever more and their usage are dramatically expanding. However, when the number of atomic planes increases from 2 to 4 in a few layers of graphene then thermal conductivity decreases from 2,800 to 1,300 W.m-1.k-1 . This phenomenon is due to the coupling of low energy phonons in cross planes and changes in phonon Umklapp scattering (i.e., scattering results in the transformation of a wave vector to another Brillouin zone) . On one hand, thermal energy storage is gaining good attention because of increasingly high demand for clean energy the depletion of fossil fuels . Thermal energy storage is affected by thermal conductivity as low thermal conductivity can prolong the charging/discharging periods. Thermal energy storage works on the principle of storing latent and sensible heats. Materials with high thermal energy storage capacity generally possess low thermal conductivity . Thus, thermal energy storage materials are composited with graphene to enhance its thermal energy storage efficiency.
Mehrali et al.  fabricated palmitic acid (C16H32O2)/GO (PA/GO) composite using vacuum impregnation method, and tested as a phase change material in thermal energy storage material. The prepared composite exhibited a thermal conductivity up to 1.24 W.m-1.k-1 at solid state which was three times higher than pure PA . Yavari’s group also synthesized a nanostructured graphene/1-octadecanol hybrid composite prepared by mixing of graphene with 1-octadecanol . Their results show that the thermal conductivity of the prepared composite increased up to 140% which motivates it to be used in thermal energy storage devices . The increase in thermal conductivity of graphene-based polymer composites is due to the network of graphene fillers which provides lower resistance for phonon to travel, as phonons are the main reason for thermal conduction in polymers. In contrast, the covalent bond between matrix and filler reduces phonon scattering at the interface of matrix and filler .
GHCs are also used in thermoelectric devices . In order to improve a thermoelectric conversion, low thermal conductivity is required in thermoelectric devices. Graphene has high intrinsic thermal conductivity . Thus, graphene is composited with various materials to reduce the thermal conductivity [163, 164]. The GHCs could decrease thermal conductivity and boost the performance of thermoelectric materials by boundary phonon scattering. For instance, Suh et al.  developed graphene/Bi0.5Sb1.5Te3 composite and thermal conductivity of that composite was observed to be 0.7 W.m-1.k-1, at room temperature, whilst Bi0.5Sb1.5Te3 ingot had a thermal conductivity of ~1 W.m-1.k-1 . Also, a decrease in thermal conductivity was observed in incorporated graphene with PbTe (graphene/PbTe, ~0.9 W.m-1.k-1 at room temperature), BiSbTe (graphene/BiSbTe, ~2.1 W.m-1.k-1 at room temperature) and CoSb3 (graphene/CoSb3, ~1.25 W.m-1.k-1 at room temperature) [163-165].
Electrical properties of GHCs play an important role in depicting its behavior in energy storage devices. An ideal energy storage device requires high power density . A single layer of graphene has a theoretical specific capacitance of 550 F.g-1 . However, its full potential cannot be realized due to re-stacking of graphene nanosheets . Therefore, graphene is generally composited with pseudo-capacitive materials to enhance its specific capacitance . Electrical conductivity of pseudo-capacitive materials is low and they usually exhibit slow electron transport. The electron transport can be accelerated when it is composited with graphene, which results in an enhancement of its electrical conductivity . High surface area and electrical conductivity of a material are two factors that contribute to the specific capacitance . Graphene has been composited with many metal/metal oxide NPs, conducting polymers to enhance its specific capacitance [170-173]. For example, the synthesized graphene/silver hybrid composite exhibit a capacitance of 220 F.g-1 in potassium nitrate (KNO3) solution at a scan rate of 10 mV.s-1, which is higher than that of the graphene electrode with 140 F.g-1 . This enhancement of capacitance in the synthesized graphene/silver hybrid composite is because of the large pseudo-capacitance (due to residual C-O and C=O functional groups), lower aggregation and relatively slow charge/discharge kinetic . Graphene/silver sulfide (Ag2S) hybrid composite fabricated by the solvothermal process also exhibited a pseudo-capacitance of 1,063 F.g-1 at a scanning rate of 100 mV.s-1 .This increment of capacitance in graphene/Ag2S hybrid composite is due to redox transitions of Ag2S between a semiconducting and a conducting state. In another work by Chen et al. , a graphene/Co(OH)2 hybrid composite was prepared using water isopropanol mixed solvent which exhibited a specific capacitance of 972 F.g-1 in KOH electrolyte. This specific capacitance value was more than their individual counterpart [Co(OH)2, 726.1 F.g-1 and graphene, 137.6 F.g-1], which was due to the effective prohibition of aggregation of graphene nanosheets and provided a much higher surface area, by Co(OH)2 NPs . The synergistic effect between components of GHCs also improves the capacitance of the composite system. For instance, graphene foam/Co(OH)2 composite exhibited an improved specific capacitance of 1139 F.g-1 at a current density of 10000 mA.g-1 in 1 M KOH electrolyte . This improvement is due to a synergistic effect between graphene and Co(OH)2 nanorods .
Some conducting polymers were composited with graphene mainly to leverage their pseudocapacitive effect . Polyacrylate, poly(phenylenediamine), PANI, PMMA and PPy are some well-known polymers that have been composited with graphene to increase specific capacitance of the system [176-179]. Some factors like; (i) dispersion state of graphene in the polymer matrix (i.e., how uniformly it is dispersed in the polymer matrix), (ii) interfacial interaction (i.e., weak or strong) and (iii) network structure in the matrix, affect the properties of graphene/polymer composites. A high specific capacitance of 1,126 F.g-1 was observed for graphene/PANI composite which was synthesized by in situ polymerization reduction/dedoping-redoping process . Also, the power density and energy density of graphene/PANI composite were ~140 W.kg-1 and ~35 Wh.kg-1, respectively,which washigher than pure graphene or PANI . [Note: Power density and energy density of pure graphene are ~15 W.kg-1 and ~5 Wh.kg-1, respectively . Also, power density and energy density of pure PANI are 42 W.kg-1 and 11.4 Wh.kg-1, respectively ]. The prepared GO/PANI composite by in situ polymerizations of aniline monomer in presence of GO also exhibited a specific capacitance of 480 F.g-1 at a current density of 100 mA.g-1 . Saswata et al.  employed in situ polymerization method to fabricate GNS/PPy composite. The prepared composite showed a specific capacitance of 267 F.g-1 at a scan rate of 100 mV.s-1 which was higher than its individual components (PPy, 137 F.g-1 and GNS, 34 F.g-1) . The higher specific capacitance of graphene/PPy composite even at very low scanning rate was attributed to the synergistic effect between PPy and graphene nanosheets.
It has been reported that graphene, despite being one atom thick, can be optically visualized and present remarkable optical properties . Visual transparency/transmittance of graphene can be defined through fine structure constant . Fine structure constant describes the coupling between relativistic electrons and light which is related to quantum electrodynamics rather than materials science. The linear dispersion of Dirac electrons in graphene causes Pauli