Supercritical fluid mediated synthesis of highly exfoliatedgraphene/ZnO composite for photocatalytic hydrogen production

Yuvaraj Haldorai a,b, Jae-Jin Shim a,n

a School of Chemical Engineering, Yeungnam University, Gyeongsan, Gyeongbuk 712-749, Republic of Koreab Department of Energy and Materials Engineering, Dongguk University-Seoul, Seoul, Republic of Korea

a r t i c l e i n f o

Article history:Received 8 May 2014Accepted 25 June 2014Available online 1 July 2014

Keywords:GraphemeZnONanocompositeSupercritical fluidHydrogen production

a b s t r a c t

Reduced graphene oxide (RGO)/zinc oxide (ZnO) composite was synthesized via a simple andenvironmentally-friendly route using supercritical carbon dioxide. The structure and morphology ofthe resulting composite was characterized by transmission electron microscopy, X-ray diffraction andRaman spectroscopy. The nanocomposite exhibited enhanced photocatalytic activity for hydrogen (H2)production from water photo-splitting because the ZnO nanoparticles on the RGO sheets can capturelight energy and facilitate excited electron transfer for photocatalytic H2 production via RGO, which actsas an efficient electron mediator.

& 2014 Elsevier B.V. All rights reserved.

1. Introduction

Hydrogen (H2) is considered an ideal fuel for the future becauseit can be produced from clean and renewable energy sources [1].Since the discovery of the photocatalytic splitting of water onTiO2, semiconductor-based photocatalysis for H2 production hasattracted considerable attention [2]. Zinc oxide (ZnO) is a promis-ing semiconductor photocatalyst on account of its wide band gapof 3.37 eV, low cost, non-toxicity, and good oxidizing power. Onthe other hand, electron-hole recombination is a major impedi-ment to the widespread applications of its photocatalytic activity.Recent studies have shown that carbon nanomaterials, particularlycarbon nanotubes (CNTs), are promising cost effective cocatalysts[3]. CNTs are believed to be capable of accepting, transporting andstoring electrons, and hence reducing the recombination of thephotoinduced electrons and holes [3]. Similar to CNTs, graphene-related materials can also act as an outstanding source of acceptorsand electron transport because of their conductivity, high electronmobility and large theoretical specific surface area [4,5]. In addi-tion, graphene-related compounds have great adsorption abilityand are expected to be a good choice for adsorbent materialswith a photocatalyst [6]. Owing to its remarkable electron storageand shuttling properties, graphene facilitates photoinduced chargeseparation and inhibits electron-hole recombination when semi-conductors are immobilized on its surface in photocatalyticprocesses [4]. Therefore, considerable effort has been made to

use graphene as a co-catalyst with semiconductor oxide-basedhybrid materials for superior photocatalytic activity [7,8].Although there are some reports of semiconductor/graphenecomposites for photocatalytic H2 production, there are no accountsof H2 production using reduced graphene oxide (RGO)/ZnO nano-composite as a photocatalyst.

In this article, RGO/ZnO nanocomposite was synthesized usingsupercritical carbon dioxide (scCO2) and evaluated as a photo-catalyst for H2 production. The advantages of this scCO2 chemicaldeposition technique lay in its flexibility, simplicity, green proper-ties, and efficiency in material science and chemical processing.The nanocomposite was characterized by Raman spectroscopy,X-ray diffraction (XRD) and transmission electron microscopy(TEM). The photocatalytic activity was evaluated by measuringthe level of H2 production from water photo-splitting under UV-irradiation.

2. Experimental

Synthesis of RGO/ZnO nanocomposite: All chemicals used in thisstudy were purchased from Sigma-Aldrich and used as received.Ultra high purity CO2 (99.999%) was obtained from DeokyangEnergy Co. GO was prepared from natural graphite using amodification of Hummers method [9]. The RGO/ZnO compositewas synthesized using a previously reported procedure [10]. In atypical experiment, GO was dispersed in ethanol (5 mg/mL) undersonication for 2 h at room temperature to form a homogeneoussuspension. A precursor solution was prepared by dissolving100 mg of Zn(NO3)2 �6H2O in 1 mL of ethanol. The precursor

Contents lists available at ScienceDirect

journal homepage: www.elsevier.com/locate/matlet

Materials Letters

http://dx.doi.org/10.1016/j.matlet.2014.06.1500167-577X/& 2014 Elsevier B.V. All rights reserved.

n Corresponding author. Fax: þ82 53 810 4631.E-mail address: jjshim@yu.ac.kr (J.-J. Shim).

Materials Letters 133 (2014) 24–27

solution was added to the above solution, and the resultingmixture was transferred to a 20 mL stainless steel reactor. CO2

was then charged into the reactor. The temperature and pressureof the reactor were adjusted to 300 1C and 9 MPa, respectively. ThescCO2 conditions were maintained for 6 h with magnetic stirring.The CO2 was then vented slowly, and the product was collected.Pure ZnO and RGO were also prepared in a similar manner inscCO2. The weight ratio of RGO to TiO2 in composite was found tobe 1:20.

Photocalytic H2 production: The photocatalytic H2 productionexperiment was performed in a 100 mL double walled quartz flaskwith a water inlet and outlet to maintain the temperature of thephotoreactor [11]. The photocatalytic reaction was carried out atroom temperature and atmospheric pressure. In a typical experi-ment, 0.1 g of the sample was dispersed in 60 mL of an aqueoussolution containing Na2S (0.1 mol/L) and Na2SO3 (0.05 mol/L) asthe sacrificial reagents. The solution containing the photocatalystwas ultrasonicated for 20 min, degassed for 30 min, and irradiatedwith UV-light with stirring to ensure uniform exposure of thesuspension throughout the process. To avoid the photo-corrosionof ZnO, the experiment was performed under a N2 atmosphere.The amount of H2 produced was analyzed by gas chromatography(Agilent Technologies: 6890N) using a thermal conductivity detec-tor (TCD), molecular sieve 5A and N2 carrier gas.

Characterization: TEM (Technai G2 F20) was performed at anaccelerating voltage of 200 kV. The phase and crystallinity wereexamined by XRD (Brooker) using Cu Kα radiation over the range,10 to 801 2θ. The Raman spectra were recorded on a confocalmicro-Raman spectrometer (LabRAM ARAMIS, HoribaJobin Yvon)with 532 nm laser excitation.

3. Results and discussion

The RGO/ZnO composite was synthesized via a facile approachusing scCO2. In this study, scCO2 played a key role in coating theRGO surface with ZnO nanoparticles. First, scCO2 is miscible withethanol under suitable conditions. When ethanol acts as a solventfor the precursor, the zero surface tension of scCO2 allows ethanolto wet the GO surface during the entire experimental process.Consequently, scCO2 helps the precursor adsorb easily on thesurface of GO and enhances the physical attraction of the twosubstances. When the temperature of the reactor reaches 300 1C,both thermal reduction of GO and thermal decomposition ofprecursor occurs simultaneously. Second, scCO2 might act as anantisolvent for the ZnO nanoparticles in the expanded ethanolsystem. As the dissolution of CO2 increased, the solvent power ofthe liquid phase on ZnO decreased, which led to phase separation,

and the ZnO nanoparticles precipitated from the supersaturatedsolution. In addition, scCO2 diffuses between the RGO layersbecause of its liquid like density, gas like diffusivity and zerosurface tension. The CO2-intercalated RGO is forced to delaminateby the expansion of scCO2. The inherent properties of GO as asubstrate are the key factors affecting the size and morphology ofthe ZnO formed [10].

Fig. 1a shows the Raman spectra of the RGO and RGO/ZnOcomposite. In the RGO spectrum, the two prominent peaks atapproximately 1358 cm�1 and 1600 cm�1 were assigned to the Dand G bands, respectively. The D band corresponds to the breath-ing modes of the rings or the K point phonons with A1g symmetry,whereas the G band represents the in-plane bond-stretchingmotion of pairs of C sp2 atoms (the E2g phonons) [12]. On theother hand, the D and G bands of RGO in the composite wereshifted. The D band in the composite was shifted to a lowerwavenumber by 8 cm�1, whereas the G band showed a blue shiftof 21 cm�1. In addition, the intensity ratio of the D and G bands(ID/IG) of the composite was much lower than that of RGO, whichwas attributed to interactions between the ZnO nanoparticles andRGO sheets. Furthermore, the peak at 2720 cm�1 was assigned toan overtone of the D band, indicating increased disorder in theRGO [10,13].

Fig. 1b shows a typical XRD pattern of the as-synthesized RGO/ZnO composite. The diffraction peaks were indexed to the (100),(002), (101), (102), (110), (103), (200), (112), (201), (004), and (202)planes of ZnO with a wurtzite structure (JCPDS no. 396-1451). Theadditional peak at 26.51 was assigned to the (002) plane ofgraphite. No other peaks for impurities were detected, indicatingthe high purity of the composite.

Fig. 2a shows a representative TEM image of the RGO/ZnOcomposite. An analysis of the image revealed a homogeneousdistribution of ZnO nanoparticles at the RGO surface with a meanparticle size of r8 nm. Nevertheless, few large aggregated parti-cles were also observed by TEM. According to these observations,the in-situ method is a simple and effective strategy for fabricatingRGO/ZnO composites. High-resolution TEM (HR-TEM, Fig. 2b)showed that the lattice spacing of ZnO was approximately0.28 nm, corresponding to the (100) plane. This confirmed thatthe black dots in the TEM image were ZnO nanoparticles. Thecorresponding selected area electron diffraction pattern (Fig. 2c)also confirmed the wurtzite crystal structure of ZnO, which isconsistent with the XRD pattern.

Fig. 3a presents the photocatalytic H2 production activities ofthe as-synthesized ZnO, RGO and composite under UV-light. Noappreciable H2 was detected by RGO under UV-irradiation. Incontrast, ZnO showed low levels of photocatalytic H2 productionbecause of the rapid recombination of the conduction band

Fig. 1. (a) Raman spectra and (b) XRD pattern of the RGO/ZnO composite.

Y. Haldorai, J.-J. Shim / Materials Letters 133 (2014) 24–27 25

(CB) electrons and valance band (VB) holes. The introduction ofRGO as a cocatalyst resulted in significant improvement in thephotocatalytic H2 production activity of ZnO. The level of H2

production of the composite increased to 28.9 μmol (289 μmol/g) within 2 h, which was 4.5 times higher than that of pure ZnO(6.15 μmol, 61.5 μmol/g). No H2 was detected in the absence ofeither UV-light or the photocatalyst, highlighting the role of RGO/ZnO as a photocatalyst in H2 production.

The stability of the composite was tested four times using thesame catalyst for photocatalytic H2 production (Fig. 3b). After fourrecycle steps, the catalyst did not exhibit any significant lossof activity, indicating its high stability during photocatalytic H2

production.Fig. 3c shows a schematic diagram of the efficient transfer of

electrons from ZnO to RGO to explain the mechanism for thephotocatalytic generation of H2 by the composite. Under UV-lightirradiation, the electrons of ZnO are excited from the VB (�7.25) tothe CB (�4.05 eV). The CB electrons [14] are transferred efficientlyto RGO (work function¼�4.42 eV) because of their high chargecarrier mobility. This leads to the migration of these electrons tothe RGO sheets, which reduces the recombination rate of

photoinduced electrons and holes. In addition, the unique featureof RGO allows the photocatalytic reaction to take place, not only onthe surface of the ZnO photocatalyst, but also on the RGO sheet,causing an enhanced reaction space. Eventually, these electronsreact with the adsorbed Hþ ions to produce H2. At the sametime, the remaining holes are scavenged by the sacrificial reagentspresent in the solution.

4. Conclusion

This paper reported the feasibility of H2 production from waterphoto-splitting using RGO/ZnO composite as a photocatalyst. Theformation of a composite was confirmed by Raman spectroscopy,XRD and TEM. ZnO nanoparticles with a mean diameter ofr8 nmwere dispersed homogeneously over the RGO surface. The ZnO/RGO composite exhibited high photocatalytic H2 production activ-ity of 28.9 mmol, which was 4.5 times higher than that of pure ZnO(6.15 mmol). The positive synergetic effect between ZnO and RGO isbelieved to efficiently suppress charge recombination, improveinterfacial charge transfer and provide a larger number of

Fig. 2. (a) TEM, (b) HR-TEM and (c) SAED pattern of the RGO/ZnO composite.

Y. Haldorai, J.-J. Shim / Materials Letters 133 (2014) 24–2726

photocatalytic reaction centers. Therefore, the RGO/ZnO compositeis a potential photocatalyst for the production of H2 from watersplitting.

Acknowledgment

This study was supported by Basic Science Research Programthrough the National Research Foundation of Korea funded bythe Korean Ministry of Education, Science and Technology (2012009529).

References

[1] Ni M, Leung MKH, Leung DYC, Sumathy K. Renew Sust Energ Rev 2007;11:401–25.[2] Jia TT, Kolpin A, Ma CS, Chan RCT, Kwok WM, Tsang SCE. Chem Commun

2014;50:1185–8.

[3] Leary R, Westwood A. Carbon 2011;49:741–72.[4] An X, Yu JC. RSC Adv 2011;1:1426–34.[5] Wang P, Zhai Y, Wang D, Dong S. Nanoscale 2011;3:1640–5.[6] Zhang X, Quan X, Chen S, Yu H. Appl Catal B 2011;105:237–42.[7] Zhang X-Y, Li H-P, Cui X-L, Lin Y. J Mater Chem 2010;20:2801–6.[8] Xiang Q, Yu J, Jaroniec M. J Am Chem Soc 2012;134:6575–8.[9] Hummers WS. J Am Chem Soc 1958;80:1339.[10] Haldorai Y, Voit W, Shim J-J. Electrochim Acta 2014;120:65–72.[11] Khan Z, Chetia TR, Vardhaman AK, Barpuzary D, Sastri CV, Qureshi M. RSC Adv

2012;2:12122–8.[12] Ni ZH, Wang HM, Kasim J, Fan HM, Yu T, Wu YH, et al. Nano Lett

2007;7:2758–63.[13] Ferrari C, Meyer JC, Scardaci V, Casiraghi C, Lazzeri M, Mauri F, et al. Phys Rev

Lett 2006;97:187401.[14] Kudo A, Miseki Y. Chem Soc Rev 2009;38:253–78.

Fig. 3. (a) Reaction time profile of H2 production in the presence of a photocatalyst under UV-irradiation, (b) four cycles of photocatalytic H2 production, and (c) schematicdiagram of the mechanism for the photocatalytic H2 production under UV-irradiation.

Y. Haldorai, J.-J. Shim / Materials Letters 133 (2014) 24–27 27