Page 92 - Multifunctional Photocatalytic Materials for Energy
P. 92

Graphene photocatalysts                                            81

           route toward the large-scale production of graphene for different applications [22]
           (Fig. 5.1C). Furthermore, rGO offers an important advantage, namely the possibility
           to obtain a tailored hydrophilic surface of graphene decorated with oxygenated func-
           tionalities, which results in low production costs [19,20]. These surface groups can be
           used to facilitate the anchoring of semiconductors and metal nanoparticles and even
           for the assembly of macroscopic structures, which are relevant to developing highly
           efficient photocatalysts [23,24].
              Heteroatom doping of graphene is a rising research approach toward enhancing
           the performance of graphene-based materials for a wide range of applications [25],
           which presents an opportunity to further extend the role of graphene in photocatalysis
           [26]. In heteroatom-doped graphene materials, a certain percentage of carbon atoms
           (typically below 10 wt.%) is replaced by other elements, such as nitrogen (N) [27–29],
           boron (B) [30], phosphorus (P) [31], and sulfur (S) [32,33] (Fig. 5.1D). Several pos-
           sibilities exist for preparing doped-graphene materials, such as CVD; arc-discharge
           between two graphite electrodes in the presence of a suitable reagent containing the
           dopant element, for example, NH 3  and H 2 S; ball-milling; and pyrolysis under inert at-
           mosphere of a natural biopolymer [34]. It is worth noting that the presence of external
           atoms and defects in graphene is a critical point in catalysis applications. Indeed, the
           main graphene materials that have been applied in photocatalysis are GO, rGO, and
           doped-graphene derivatives [25,26].

           5.2.1   General properties of graphene-based materials

           Graphene can be described as a zero-energy band gap semiconductor; this means
                    ⁎
           that the  π -state conduction band (CB) and the  π-state valence band (VB) of
           graphene touch each other at the Dirac point [35], as shown in Fig. 5.2 [35,36]. This
           unique band structure causes graphene to display amazingly high conductivity and




















           Fig. 5.2  (A) 3D band structure of graphene; (B) approximation of the low energy band
           structure as two cones touching at the Dirac point. The position of the Fermi level determines
           the nature of the doping and the transport carrier.
           Adapted with permission from P. Avouris, Graphene: electronic and photonic properties and
           devices, Nano Lett. 10 (11) (2010) 4285–4294. Copyright 2010, American Chemical Society.
   87   88   89   90   91   92   93   94   95   96   97