Materials Letters 314 (2022) 131864

Contents lists available at ScienceDirect

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

Cesium-decorated reduced graphene oxide for photocatalytic

A R T I C L E I N F O A B S T R A C T

Keywords: Cs-decorated reduced graphene oxide (CsRGO) was synthesized by addition of a CsOH solution to a GO sus­
Decorated GO pension. The resulting material is amorphous, while GO reduces upon Cs incorporation as Cs2O-like clusters.
Cs Structure modelling indicates that Cs atom replaces hydrogen in hydroxyl and carboxyl groups in the graphene
Hydrogen photogeneration
plane. CsRGO exhibits transitions associated to the π-to-π*, remanent C––O groups and Cs2O absorption. The
Work function
material is n-type semiconductor, with a direct optical band gap of 2.85 eV. Carrier injection from the Cs2O to
RGO enhances H2-evolution. Stable photocatalytic generation of hydrogen after three four-hour cycles, is
attributed to work function reduction, high conductivity, and high optical absorption upon Cs-incorporation.

1. Introduction 2. Methodology

Photogenerated hydrogen would play major role in a low carbon 2.1. Synthesis of Cs2O decorated graphene oxide
future [1]. Hydrophilic GO is suitable for photocatalytic hydrogen
generation due to its unique 2D carbon structure with tunable electronic GO was synthesized using a modified Hummer’s method reported
properties through functional group modification, decoration with elsewhere [10]. For CsRGO, 500 mg of GO were suspended in 300 mL of
nanoparticles, doping with heteroatoms or attaching functionalities to H2O. Another solution made of 2 mL of CsOH 50% in water and 48 mL of
the carbon skeleton [2,3]. Heteroatom doping enhances photocatalytic deionized water H2O was added drop-by-drop to GO until pH 12.34. The
hydrogen generation through widening GO light absorption in the resultant powder was washed with water and dried at 50 ◦ C.
visible range, enhancing carrier mobility and increasing the population
of active binding or catalytic sites [4–6]. On the other hand, cesium is 2.2. Characterization
known for catalyzing reactions [7,8]. J. Liu et al. reported Cs-based GO
derivative for bulk heterojunction solar cell, where H is replaced by Cs in The photocatalytic activity was determined by suspending CsRGO in
–COOH groups. The material performed remarkably in charge extraction a Pyrex reactor in 1:1 water with methanol as hole scavenger. The re­
properties due to reduction of work function [9]. Herein, for the first action system was illuminated with a pen lamp in a quartz tube at 254
time, Cs2O-decorated RGO was evaluated for photocatalytic H2- nm and 4400 µW cm− 2. Hydrogen was measured by online gas chro­
generation. matography. Electrochemical properties are described in Supplementary
Information. The model structure was made using ChemBio 3D.

* Corresponding author.
E-mail address: [email protected] (F. Caballero-Briones).

https://doi.org/10.1016/j.matlet.2022.131864
Received 19 October 2021; Received in revised form 25 January 2022; Accepted 1 February 2022
Available online 5 February 2022
0167-577X/© 2022 Elsevier B.V. All rights reserved.
N. Gnanaseelan et al. Materials Letters 314 (2022) 131864

Fig. 1. Characterization of CsRGO a) XRD, b) FTIR, c) Raman spectra, d) survey XPS, e) HRXPS O1s f. HRXPS O 1s, g) HRXPS Cs 3d, h) HRTEM and i) simulated
model structure.

3. Results and discussion group as well as the C– – C band, indicating the reduction of GO. In
Fig. 1c), the Raman spectra show the D and G bands at 1600 cm− 1 and
In Fig. 1a) the diffraction peak attributed to the (0 0 1) plane of GO 1300–1355 cm− 1 related with the E2g degenerative mode of graphite,
was observed around 11.7◦ . In contrast, CsRGO featureless diffracto­ and the defect-activated A1g mode [12]. The calculated crystallite size
gram indicates it is amorphous [11]. GO FTIR spectra in Fig. 1b) shows (La) [13] of the samples was 3.4 nm for GO and 2 nm for CsRGO and its
bands at 3290, 1720, 1640, 1569, 1380 and 1054 cm− 1 related with decrease reflects the disorder induced by the incorporation of Cs, that
hydroxyl/adsorbed H2O, carboxyl, carbonyl, C– – C stretching of sp2 hinders the increase of the sp2 dominion [14].
carbon atoms, hydroxyl and epoxy functional groups respectively [10]. CsRGO XPS spectra in Fig. 1d) confirms the presence of ca. 90 at. %
The FTIR spectra of CsRGO presents a small band related to carbonyl C, 7 at. % O and 3 at. % Cs [9]. The HR-XPS of C 1 s shown in Fig. 1e),

Fig. 2. a) UV–vis spectrum and Tauc plot, and b) PL spectrum.

2
N. Gnanaseelan et al. Materials Letters 314 (2022) 131864

Fig. 3. a) Cyclic voltammograms, b) Mott-Schottky plots and, c) photocatalytic H2-evolution of GO, CsRGO and blanks; d) electronic diagram for photocatalytic
H2– evolution.

shows peaks at 284.6 eV, 285.6 eV and 286.6 eV assigned to C– – C, C–O/ assess photocatalyst stability three four-hour cycles were done. After
C–OH and C– – O respectively [15]. HR-XPS of O 1s depicted in Fig. 1f) each cycle, the system was purged with N2 for two minutes. The H2-
shows peaks at 533.1 eV, 531.8 eV and 530.8 eV assigned to Cs2O, C– –O evolution for CsRGO reached 1701 µmolg− 1 in CsRGO in the first 4-hr
and C–O/C–OH. The HRXPS Cs 3d peaks at 726.4 eV (Cs 3d3/2) and 739 cycle and increased up to 2000 µmolg− 1 at the end of the fourth cycle.
eV (Cs 3d5/2) in Fig. 1g) confirm Cs in the form of Cs2O [16]. The TEM For comparison, reported photocatalytic H2-generation for N-doped
micrograph in Fig. 1h) inset depicts folded and wrinkled sheets [17]; the RGO, B-doped RGO and MoS2–TiO2-RGO are 64.6 µmolg− 1h− 1, 66.7
high resolution image shows both crystalline graphitic dominion (red) µmolg− 1h− 1 and 165.3 µmolg− 1h− 1 after one cycle [4 23]. The photo­
and nanoparticles about 5 nm and smaller (blue). We suggest that Cs is catalyst showed good stability without any degradation of the photo­
bonded to the GO sheet as shown in the model structure of CsRGO in catalytic activity, which indicates the activation of the reaction centers,
Fig. 1i), where COOH groups in the GO periphery, are replaced by although the nature of this activation will be subject of a future work.
COOCs giving rise to Cs2O-like clusters or small nanoparticles. The The enhanced photocatalytic hydrogen evolution rate in CsRGO is
epoxy ring near the hole in the graphene lattice is broken during GO attributed to an increased charge extraction due to reduction of work
reduction, and the C–O–C groups are replaced with C–O–Cs. Cs may also function upon Cs addition, and improved charge separation upon GO
replace hydrogen atoms in OH groups forming C–O–Cs [9,18]. The reduction. The proposed band diagram for photocatalytic H2-evolution
partial reduction of graphene oxide during Cs decoration is related to of CsRGO is shown in Fig. 3d). The presence of nanoparticles was
highly alkaline pH 12.34 [19] confirmed by TEM; then, the conduction band (CB = − 1.53 eV) and
valence band (VB = 0.44 eV) values of Cs2O taken from reference [24],
4. Optical, and electronic properties are compared with the work function of RGO, which is around –4.0 eV
and with the values from Fig. 3a and b); CsRGO donor level energy ED is
In the absorbance spectra of CsRGO at Fig. 2a) the peak at 230 nm is higher that the work function of GO and H+/H− reduction potential. The
due to π-π* transition, while the peaks at 290 nm and 350 nm are due to photogenerated electrons are then injected from Cs2O CB and ND to RGO
optical absorption of C═O and C− O moieties [20]. The Tauc plot (αhν)2 and to the solution, increasing the H2-evolution rate. Holes created in VB
vs (hv) at Fig. 2a) inset shows a direct bandgap of 2.85 eV. The extended are scavenged by methanol [25]. The effect of different hole scavengers
absorption at the lower energy is attributed to structural defects and on the H2-generation will be studied in a future work. Due to its high
provides evidence of charge injection from the Cs2O-like clusters to the conductivity, RGO assists in the fast charge separation and acts as
RGO sheet. The PL spectra with 350 nm excitation shown in Fig. 2b) cocatalyst to produce hydrogen during the process [26].
depicts an emission peak at 390 nm assigned to σ*→n transitions. The
emission at 461 nm was attributed to the π*→π transition of the sp2 5. Conclusions
carbon atoms but also to direct band-to-band recombination in the Cs2O-
like clusters. The groups associated with C– – O cause σ*→n electronic Cesium-decorated graphene oxide was synthesized using facile
transitions at 550 nm [21]. titration-based technique. Cs is present as Cs2O-like clusters, replacing
The cyclic voltammograms shown in Fig. 3a) present the potential the hydrogen atoms in carboxyl and hydroxyl groups. Cs would also
onset of hydrogen evolution E = -0.5 V and E = -0.95 V vs Pt, for GO and break epoxy ring and form the C-O-Cs groups. The material is n-type
CsRGO respectively, which correspond to an electronic energy of –4.0 semiconductor, amorphous, exhibits an electrical conductivity of 10− 4
and − 3.55 eV. Fig. 3b) presents the Mott-Schottky plots 1/C2 vs E where S.cm and a direct band gap of 2.85 eV. Cs-doping contributes with one
the flat band potential values, i.e. when 1/C2 = 0, were − 0.05 V and electron to graphene plane and the reduction of the work function of
− 0.5 V vs Pt for GO and CsRGO respectively; the slope indicate both RGO, resulting in an enhanced photocatalytic activity. A maximum H2
materials behave as n-type semiconductors [22]. photocatalytic generation of ca. 2000 µmolg− 1 was achieved after three
The hydrogen evolution from GO and CsRGO is plotted in Fig. 3c). To four-hour cycles of H2-generation, indicating the synthesized material

3
N. Gnanaseelan et al. Materials Letters 314 (2022) 131864

has a stable activity upon cycling. parameters, electrochemistry. Supplementary data to this article can be
found online at https://doi.org/10.1016/j.matlet.2022.131864.
CRediT authorship contribution statement
References
N. Gnanaseelan: Conceptualization, Methodology, Writing – orig­
inal draft. M. Latha: Validation, Resources. A. Mantilla: Validation, [1] I. Staffell, D. Scamman, A. Velazquez Abad, et al., Energy Environ. Sci. 12 (2019)
463.
Resources. S.K. Kamaraj: Validation, Resources. F.J. Espinosa-Faller: [2] K. Krishnamoorthy, R. Mohan, S.J. Kim, Appl. Phys. Lett. 98 (2011) 2013.
Validation, Resources. F. Caballero-Briones: Conceptualization, [3] T.F. Yeh, J. Cihlář, C.Y. Chang, et al., Mater. Today 16 (2013) 78.
Methodology, Validation, Resources, Writing – review & editing, [4] L.K. Putri, B.J. Ng, W.J. Ong, et al., ACS Appl. Mater. Interfaces 9 (2017) 4558.
[5] D.W. Chang, J.B. Baek, Chem. – An Asian J. 11 (2016) 1125.
Supervision. [6] D. Akyüz, R.M. Zunain Ayaz, S. Yılmaz, et al., Int. J. Hydrogen Energy 44 (2019)
18836.
Declaration of Competing Interest [7] S.L. Foster, S.I.P. Bakovic, R.D. Duda, et al, Nat. Catal. 2018 17 1 (2018) 490.
[8] N. Pasha, N. Lingaiah, N.S. Babu, et al., Catal. Commun. 10 (2008) 132.
[9] J. Liu, Y. Xue, Y. Gao, et al., Adv. Mater. 24 (2012) 2228.
The authors declare that they have no known competing financial [10] J. Guerrero-Contreras, F. Caballero-Briones, Mater. Chem. Phys. 153 (2015) 209.
interests or personal relationships that could have appeared to influence [11] K. Krishnamoorthy, M. Veerapandian, K. Yun, et al., Carbon N. Y. 53 (2013) 38.
[12] L.G. Caņado, K. Takai, T. Enoki, et al., Appl. Phys. Lett. 88 (2006) 12.
the work reported in this paper.
[13] L.G. Cançado, M.G. Da Silva, E.H. Martins Ferreira, et al., 2D Mater. 4 (2017).
[14] S. Azizighannad, S. Mitra, Sci. Rep. 8 (2018) 1.
Acknowledgements [15] B. Gupta, N. Kumar, K. Panda, et al., Sci. Rep. 7 (2017) 1.
[16] T. Chukeaw, A. Seubsai, P. Phon-In, et al., RSC Adv. 6 (2016) 56116.
[17] D. Long, W. Li, L. Ling, et al., Langmuir 26 (2010) 16096.
Financed by CONACYT 2019-40798, CONACyT CB-285711 and SIP- [18] P. Kaewmee, J. Manyam, P. Opaprakasit, et al., RSC Adv. 7 (2017) 38747.
20210702 projects. XPS measurements at LANNBIO-CINVESTAV- [19] X. Mei, H. Zheng, J. Ouyang, J. Mater. Chem. 22 (2012) 9109.
Merida, were financed thru FOMIX-CONACYT. A.C. Espindola-Flores [20] Q. Lai, S. Zhu, X. Luo, et al., AIP Adv. 2 (2012) 2.
[21] M. Li, S.K. Cushing, X. Zhou, et al., J. Mater. Chem. 22 (2012) 23374.
and S. Pacheco are acknowledged for FTIR and XRD measurements. J. [22] A. Palacios-Padrós, F. Caballero-Briones, F. Sanz, Electrochimica Acta 111 (2013)
E. Samaniego is acknowledged for H2-evolution measurements. LM was 837.
partially financed by CONACyT and UAQ. [23] X. Yang, H. Huang, M. Kubota, et al., Mater. Res. Bull. 76 (2016) 79–84.
[24] T.F. Yeh, J.M. Syu, C. Cheng, et al., Adv. Funct. Mater. 20 (2010) 2255.
[25] A. Hezam, K. Namratha, Q.A. Drmosh, et al., J. Mater. Chem. A 6 (2018) 21379.
Appendix A. Supplementary data [26] Q. Xiang, J. Yu, M. Jaroniec, J. Am. Chem. Soc. 134 (2012) 6575.

FESEM/TEM images, EDS mapping, electrical conductivity, Raman