NATURE MATERIALS DOL:10.1038/NMAT4170 PROGRESS ARTICLE 00 .000 lonic radius 2M+0,+20M,0 Voltage M=u of gr s.a,G nd graphene(RGO) EESDs e.G of graphe ch a5 GO)on pset es defor mations and presence of surface rous)in metal-ar batteries capacitance the in RGO (Fig.3e).Va t the nano Qaic whicn tuance by increasing electrical conductivity and ir ele t of discha 厂he xam 000 the graphene sheets Lithium-air batteri s.The growing dem and for energy has led to ed r om.athouh achieving the theoretical y(LAB) the rechargeability of this 2006 ries over by Bruce and colleagues Although different LABs may employ NATURE MATERIALS|ADVANCE ONLINE PUBLICATIONw.ure.atureate 5NATURE MATERIALS | ADVANCE ONLINE PUBLICATION | www.nature.com/naturematerials 5 and cycle life62. In this regard, graphene-based electrodes could be viable candidates for improving the performance of pseudoca￾pacitors62. Despite its lower electrical conductivity, GO, owing to its large number of oxygen-containing groups, has a higher pseudo￾capacitance than RGO65. However, as previously discussed, these groups may negatively affect the electrochemical behaviour of the electrode by reducing the cycling stability and reversibility62,65 (Fig.  3e). Various graphene-conducting polymer and graphene– metal-oxide composites have also been developed and investi￾gated for use as pseudocapacitors6,62. In these composites, graphene provides a support matrix for the growth of the electroactive species at the nanoscale, which results in a larger SSA and thus enhances the electrochemical performance by increasing electrical conductivity and mechanical stability62,66. It seems the key to exploiting the full potential of graphene in pseudocapacitors relies on the development of composite materials that offer the synergistic effect of the graphene substrate and the elec￾troactive component, along with an optimized spatial orientation of the graphene sheets12,39,66. Lithium–air batteries. The growing demand for energy has led to the development of new EESDs with higher energy densities than metal-ion batteries. In this regard, the lithium–air battery (LAB), which offers a theoretical energy density of 5,200 Wh kg–1 (ref. 67), represents one of the best candidates. Although lithium–air chem￾istry was introduced in 1976, the rechargeability of this system was brought to the attention of the scientific community only in 2006 by Bruce and colleagues68. Although different LABs may employ different typologies of electrolyte, they are generally composed of metallic lithium and oxygen (or air) as, respectively, the anode and cathode. The rechargeability of the system relies on the conversion of reduction products (LiO2 and, mainly, Li2O2) formed during dis￾charge (oxygen-reduction reaction), back to the original reagents during charge (oxygen-evolution reaction)39. Unfortunately, the entire system suffers from a low energy efficiency, short lifetime and low rate capability (discharge capacity of 400 mAh g–1 after 50 cycles at a specific current of 100 mA g–1)68–70. Reports indicate a maximum of only 100 capacity-limited (1,000  mAh  g–1) cycles71. Among the various factors that influence the performance of LABs, the mor￾phology of the air electrode (cathode) is particularly important for obtaining high discharge capacity. In fact, the SSA and porosity of the air electrode determine the morphology and amount of discharge products. It was demonstrated that RGO, with its large SSA, could deliver higher capacities than other carbon substrates (for example, 8,700 mAh g–1 with respect to 1,000–2,000 mAh g–1 in the first cycle). Defects and functional groups can also play a catalytic role for the formation of discharge products69 (Fig. 3f). So far, the use of RGO as a bare material or substrate for other catalyst72 in LAB cathodes has improved performance, although achieving the theoretical energy density is still far away. Different aspects are still unclear and further studies are needed to demonstrate an effective role of graphene in LABs. Further investigations of graphene with stable electrolytes are needed before we can assess its effective role in such batteries69. Sodium–air batteries. Over the past five years, sodium–air bat￾teries (SABs), despite having an energy density half that of LABs, Figure 3 | Features and limitations of graphene as an active material in different EESDs. a, Graphite and graphene in LIB anodes. Correlation of characteristics in terms of defect amount, SSA and ratio between reversible (Crev) and irreversible (Cirr) capacity during the first charge/discharge cycle. b, Typical voltage profiles of graphite and graphene (RGO) during constant current Li+ insertion/de-insertion. c, Li+ and Na+ insertion mechanisms in graphene and graphite. d, Layers re-stacking in graphene during electrode manufacturing and electrochemical cycling. Re-stacking is a serious issue that affects the performances of all graphene-based EESDs. e, Generic voltammetric behaviour of graphene-based electrochemical capacitors over prolonged cycling. Top: Effect of graphene layers re-stacking (such as in RGO) on the double-layer capacitance. Bottom: Effect of surface group degradation (such as in GO) on pseudocapacitance. f, Catalytic effect of graphene defects (vacancies, deformations and presence of surface groups) in metal–air batteries. Capacitance Voltage Graphene Graphite Ionic radius Electrode manufacturing Electrochemical cycling Surface group degradation (for example in GO) Decrease of pseudocapacitance Hysteresis Cirr 0 1,000 1 2 3 2,000 a d f b c e Graphite Graphene Graphite Graphene Layers re-stacking (for example in RGO) Decrease of double-layer capacitance Specific gravimetric capacity (mAh g–1) Defects M = Li, Na Interlayer spacing Potential versus Li/Li+ (V) SSA Crev /Cirr 2M+ + O2 + 2e– M2O2 Li+ Na+ Na+ Li+ NATURE MATERIALS DOI: 10.1038/NMAT4170 PROGRESS ARTICLE © 2014 Macmillan Publishers Limited. All rights reserved