Inhibition of hydrogen and oxygen recombination over amide–functionalized graphene and the enhancement of photocatalytic hydrogen generation in dye–sensitized AF–RGO/Pt photocatalyst dispersion
Graphical abstract
Introduction
The dye–sensitized photocatalytic hydrogen evolution (PHE) from water reduction is an important route for solar light store and conversion because these systems absorb longer wavelength visible light that comprises a majority of the solar spectrum [[1], [2], [3], [4], [5], [6], [7], [8], [9], [10], [11], [12], [13], [14], [15], [16], [17]]. Commonly, a dye–sensitized catalyst for PHE is composed of a light–absorbing dye, sensitized matrix and co–catalyst. An excellent co–catalyst can work as highly active sites to reduce the overpotential of hydrogen evolution [13,15]. Noble metal (Pt and Pd) co–catalysts show the high activity for the generation of hydrogen [[18], [19], [20], [21], [22], [23], [24]], however, they can also catalyze hydrogen and oxygen recombination back to water [[25], [26], [27], [28], [29], [30], [31], [32]]. Actually, hydrogen and oxygen recombination can easy occur over noble metal loaded catalysts at room temperature and has been widely used in hydrogen removal passive autocatalytic recombiner to avoid the explosion risk in water–cooled nuclear reactors and submarines under normal and emergency operating conditions [27,28].
In many cases, hydrogen and oxygen coexist and dissolve in the liquid water. It is known that more oxygen can be dissolved in water then hydrogen under same conditions [25,26]. Therefore, the nascent formed hydrogen will be consumed severely by dissolved oxygen due to spontaneous inverse hydrogen and oxygen recombination on the co–catalyst surface, which will seriously impede the over–all hydrogen generation rate during photocatalytic water splitting. Several strategies have been reported to inhibit this recombination in the PHE [25,26,[29], [30], [31], [32]]. For example, the higher oxidation state Pt can markedly inhibit the recombination of hydrogen and oxygen, while its hydrogen evolution capacity is still remained compared to that of conventional metallic Pt cocatalyst [29]. Zhu et al. [30] also found that the formaldehyde oxidation activities over halogen–adsorbed Pt–TiO2 catalyst could be significantly reduced due to the stronger Pt–X bond, which decreased the adsorption and activation of oxygen on the Pt surface. Recently, Kang et al. [31] reported that the ternary doping of F into SrTiO3:Cr/Ta could increase the hydrogen evolution rate under visible light irradiation due to F–SrTiO3 interaction. Our previous work reported that the backward reaction of hydrogen–oxygen recombination was successfully restrained by addition of oxygen transfer reagent hemin chloride in the Pt/TiO2 catalytic system [26]. We also found that the trace amount of fluorine ion could also remarkably inhibit hydrogen and oxygen recombination due to occupation of hydrogen and oxygen adsorption and activation sites by fluorine ion occupation on Pt sites [25]. Although H2 and O2 recombination reaction in PHE is well acknowledged, the controllable inhibition of oxygen adsorption and activation over catalyst is still challenging due to lacking of strategy and materials which exhibit tunable oxygen activation over catalyst.
Herein, the solvothermal preparation amide–functionalized graphene oxide (AF–RGO) matrix shows significant inhibition properties for H2 and O2 recombination reaction. It is confirmed that the amide functionalized groups over GO can effectively inhibit H2 and O2 recombination by decrease of adsorption dissolved oxygen by the enhancement of hydrophobicity properties of catalyst in water and variation of electrical properties of catalyst by the amide functionalized groups, as a result, the adsorption of O2 in the solution is remarkably reduced. The calculation results indicate adsorption energy changes significantly after the amide functionalization, which leads to the decrease of hydrogen and oxygen recombination. The AF–RGO/Pt nanohybrid catalyst displayed highly efficient dye–sensitized photocatalytic H2 evolution under the visible irradiation and the quantum efficiency achieved 36.4% at 430 nm. Results in this paper might light the possible route to build up active photocatalyst for dye–sensitized PHE from water.
Section snippets
Preparation of AF–RGO
GO was synthesized by the oxidation of high purity graphite powder according to the modified Hummers’ method [15,32]. The synthesized GO (100 mg) was dispersed in the DMF (100 mL) by ultrasound dissolving, which acted as a dispersing, reducing and stabilizing agent. Then, a small amount of de–ionized water was added into the resulting dispersion above. After being stirred for a few minutes, the obtained yellow brown dispersion of GO was placed into a two–necked, round–bottomed flask and heated
Structure and morphology characterizations
In order to study the variation of functional groups on the surface of samples after solvothermal treatment, the Fourier transform infrared spectra (FTIR) experiments are executed and analyzed. From the FTIR spectra of bare GO and T:AF–RGO in the Fig. 1a, it can be observed that the samples exhibit several characteristic absorption bands of oxygen containing groups. The peaks at 1076 and 1282 cm–1 are attributed to CO and COC stretching modes, respectively. The CC skeleton vibration peak can be
Conclusions
In this work, AF–RGO is prepared via a liquid phase solvothermal treatment using DMF as the dispersing and reducing agent. By controlling and adjusting the reaction temperature from 80 to 140 °C, incorporating amide groups onto graphene has been achieved. The hydrophobic property was enhanced in some extend by this treatment due to replacement oxygen–containing functional group by amide group on GO. Amide group modification also improves the electrical properties of catalyst, such as
Acknowledgement
This work is supported by the National Natural Science Foundation of China (Grant Nos. 21673262 and 21433007), respectively.
References (64)
- et al.
Appl. Catal. B
(2017) - et al.
Appl. Catal. B
(2017) - et al.
Carbon
(2016) - et al.
J. Catal.
(2015) - et al.
Appl. Catal. B
(2017) - et al.
Appl. Catal. B
(2018) - et al.
J. Catal.
(2017) - et al.
Appl. Catal. B
(2017) - et al.
Appl. Catal. A
(2015) - et al.
Appl. Surf. Sci.
(2016)