Low temperature synthesis of graphene hybridized surface defective hierarchical core–shell structured ZnO hollow microspheres with long-term stable and enhanced photoelectrochemical activity

Abstract

The present work reports on successful in situ synthesis of chemically converted graphene (CCG) hybridized, surface defective core–shell structured ZnO hollow microspheres (ZG-CSHM) from a surfactant/template free precursor by adopting a low temperature solution method. This special architecture has been synthesized as an intermediate product between solid and hollow microspheres via Ostwald ripening process by optimizing the reaction time, as observed by field emission scanning and transmission electron microscopic studies. The samples have also been characterized by X-ray photoelectron, FTIR and Raman spectral as well as X-ray diffraction analyses. From textural property measurement by BET N₂ adsorption–desorption isotherms, it is seen that the ZG-CSHM possesses an enhanced specific surface area with narrow distribution of mesopores. Relatively higher photoelectrochemical activity with long term stability of ZG-CSHM is found compare to pristine core–shell structured ZnO hollow microspheres. The synergic effect of graphene hybridization and the presence of surface defects of ZnO nanoparticles in the mesoporous sample can play the key roles in advancing its photoelectrochemical activity. The surface defects can prolong the recombination rate of photogenerated charge carriers and the high surface area with narrow sized mesopore distribution can provide large number of active sites, make electrolyte diffusion and mass transportation easier. The ZG-CSHM sample also shows an improved photocurrent density compare to solid and hollow microspheres. Moreover, the existence of chemically interacted CCG with ZnO inhibits the photocorrosion, resulted long-term stable photoelectrochemical activity of ZG-CSHM. This facile process can create an avenue for synthesis of core–shell structured microspheres from different metal oxide semiconductors for improving their photoelectrochemical activity.

---

1. Introduction

Micro/nano-structures with an interior space always draw special attention due to their unique structure and advanced functional properties.^1–4 In this respect, hollow micro/nanostructures of metal oxide semiconductors are fascinating owing to their high surface-to-volume ratio, surface permeability and light harvesting properties^3,4 that can offer a wide range of applications including photoelectrochemical (PEC) water splitting, photocatalysis etc.^3–8 However, the properties especially textural properties (like surface area, pore size, pore volume) as well as light harvesting capabilities can further be improved in core–shell structured hollow microspheres (CSHM) due to the presence of inner solid core structure.^9,10 This property of CSHM can make them advanced photo anode materials for efficient PEC water splitting.

Various hollow or multi-shell/core–shell structured metal oxide semiconductors (MOS)^9–12 in nano dimension can improve the PEC properties by enhancement of electrolyte permeability and light harvesting capability, have been studied mainly based on their textural properties. However, the presence of surface defects in MOS nanoparticles can take part strongly in photogenerated charge separation and improve its light harvesting (absorption) capability.^13,14 But, the long term photo stability of the material is highly desirable for the PEC application. Hence, there is significant challenges still remained to improve the PEC property of the MOS.

It is known that different types of extrinsic/intrinsic surface defects can be created in metal oxide semiconductor (MOS).^13–17 These surface defects can trap photogenerated charges and can help to enhance photocatalytic activity towards degradation organic dyes^15–17 as well as promoting PEC water oxidation.^13,14 This is because the decrease of relative concentration of bulk to surface defects in MOS, can significantly improve the charge separation efficiency of photogenerated electrons and holes.¹⁶ On the other hand, graphene (graphene oxide/chemically converted graphene, reduced graphene) can easily form nanocomposites with several materials including MOS^16,18–20 and it can also develop a p–n junction or a sink of electron that enhances in photogenerated charge separation.^18,20 Moreover, the graphene can also diminish photocorrosion of MOS.²¹ Therefore, to enhance the rate of charge transport, chemically stable graphene can be hybridized/coated (chemically interacted) with MOS nanoparticles.^16,21 In this respect, Kim et al.²¹ recently reported graphene coated ZnO quantum dots which showed ultra-fast photogenerated charge transfer with enhanced photocatalytic activity as well as long term stability of the sample. In brief, an effective charge separation is possible in a graphene–MOS nanocomposite due to fast electron transport property of graphene and the presence of surface defects of MOS.

Generally, hydrothermal process can be implemented for synthesis of core–shell/multi-shell structures of MOS^1–3,22 (like ZnO, a wide band gap MOS) but the process requires relatively high pressure and post annealing temperature for improving the crystallization of the material. In this respect, Dong et al.¹¹ prepared multi shell ZnO hollow microspheres using carbonaceous spheres template and they controlled the microstructure through a heating process. However, in many cases, the hydrothermal process has some limitations towards understanding of mechanistic principles for a rational synthesis strategy.²³ In this respect, it is generally agreed that low temperature solution method is advantageous over the others for the synthesis of surface defective hierarchical nano structures in several applications.^24,25 However, to the best of our knowledge, low temperature solution based graphene hybridized surface defective core–shell structured ZnO hollow microspheres with improved PEC property is not available in the literature.

The present report is devoted on the successful synthesis of graphene hybridized surface defective core–shell structured ZnO hollow microspheres (ZG-CSHM) adopting low temperature (95 °C) solution method in air atmosphere from surfactant/template free precursor. This special architecture of the composite has been synthesized in situ as an intermediate product between solid and hollow microspheres of ZnO–graphene nanocomposite by prolonging the reaction time only. Structural (crystal phase, morphology and microstructure), textural (multi point BET surface area, porosity), chemical bonding, surface defects and light harvesting properties have been characterized systematically. An enhanced photoelectrochemical and photocatalytic properties are found in ZG-CSHM compare to other structures. Finally, a co-relation has been drawn between structures and properties of materials.

2. Experimentals

2.1 Synthesis of ZG nanocomposites

Graphene oxide was synthesized from modified Hummers' method (see ESI for details†). The synthesis of in situ ZnO-chemically converted graphene (ZG) was performed by adopting a facile low temperature (95 °C) soft-chemical process using as-synthesized graphene oxide (30 mg, GO) and zinc acetate dihydrate (0.95 g, Zn(CH₃COO)₂·2H₂O, ZA; Sigma-Aldrich, ≥98%) in dimethyl formamide (DMF, 240 mL) medium. In a typical synthesis, GO was uniformly dispersed in DMF (40 mL) by ultrasonication for about 60 min and the dispersed GO in DMF was mixed with ZA in DMF during stirring condition followed by heating the aliquot at 95 °C in an air oven. It was observed that at 8 h reaction time, solid microspheres (ZG-SM) were found to form but the sample collected at 9 h, the solid microspheres transformed into core–shell structured hollow microspheres (ZG-CSHM) and finally, after 12 h, the ZG-CSHM converted into the hollow microspheres (ZG-HM) (discussed in details under results and discussion). The as-synthesized samples were washed by centrifugation using distilled water and ethanol, sequentially. Then, the samples were dried in an air oven at ∼55 °C. Without using GO, pristine core–shell structured hollow zinc oxide (Z-CSHM) was also prepared under the identical reaction condition as maintained for the synthesis of ZG-CSHM sample.

3. Characterizations

3.1 Materials properties

Surface feature including clustered size of ZG microspheres present in the samples was analysed by field emission scanning electron microscope (FESEM, ZEISS, SUPRA™ 35VP). In addition, to clearly visualize the formation of solid, core–shell and hollow microspheres, transmission electron microscope (TEM) along with high resolution TEM (HRTEM) images were taken using an ultrahigh-resolution field emission gun (UHR-FEG) TEM instrument (model JEM 2100 F, JEOL) at 200 kV electron source. For the measurement, the specimen were prepared separately by dropping the dispersed samples in methanol onto a carbon coated 300 mesh Cu grid and finally, dried the grid in an air atmosphere. The crystalline phase was determined using an X-ray diffractometer (Philips PW 1730 X-ray diffraction unit employed with nickel-filtered CuK_α radiation source with 1.5418 Å wavelength). The diffraction angle (2θ) was chosen from 10° to 80°. The surface area and pore diameter of the samples were determined by Brunauer–Emmett–Teller (BET) nitrogen adsorption and desorption isotherm measurements at liquid nitrogen temperature (77 K) by employing a Quantachrome (Autosorb 1) instrument. The samples were out gassed in vacuum condition at suitable temperature for ∼3 h prior to the textural property measurement. FTIR vibrational spectral study was performed by adopting a FTIR spectrometer (Nicolet 5700, Thermo Electron Corporation, USA). The number of scans for each experiment was fixed to 100 numbers and the wavenumber resolution of the instrument was 4 cm⁻¹. Using micro-Raman (Renishaw inVia Raman microscope), the Raman spectra of the samples were measured at room temperature. In this study, an argon ion laser with an incident wavelength of 514 nm was used as the excitation source. The X-ray photoelectron spectra (XPS) of GO and the representative ZG sample (ZG-CSHM) was measured by employing PHI Versaprobe II Scanning XPS microprobe surface analysis system using Al-K_α X-rays (hν, 1486.6 eV; ΔE, 0.7 eV at room temperature). The energy scale of the spectrometer was calibrated with pure (Ag) sample. The pressure in the XPS analysis chamber was better than 5 × 10⁻¹⁰ mbar and the position of C1s peak with the binding energy of 284.5 eV was taken as standard. Absorption spectra of the samples in the UV-visible regions were recorded by diffused reflectance method using an UV-Vis-NIR spectrophotometer (UV3600, Shimadzu, Japan) with ISR 3600 attachment. Room temperature photoluminescence (PL) property of the samples was measured by Perkin-Elmer (LS55) spectrofluorimeter, fixing the excitation wavelength at 340 nm. To understand the faster charge transfer in graphene based ZG-CSHM sample compare to pristine core–shell structured hollow ZnO (Z-CSHM), time-resolved photoluminescence (TRPL) measurements were performed with the help of Horiba-Jobin-Yvon FluoroCube fluorescence life time measurement system using NanoLED (IBH UK, 300 nm excitation source) and the TBX photon detection module was used as the detector. For the set-up, the instrumental response time was approximately 1 ns. The photoluminescence decays of pristine ZnO (Z-CSHM) and ZG (ZG-CSHM) nanocomposite were measured for the PL emission at 485 nm. On the other hand, to analyze the decays, the IBH DAS-6 decay analysis software was used. In this respect, the TRPL curve fitting was made from χ² criterion of the data. From the decay times (τ₁, τ₂, and τ₃) and the normalized pre-exponential factors (a₁, a₂ and a₃), average (mean) fluorescence lifetimes (τ_avg) for triexponential iterative fittings were calculated using the relation as given in eqn (1).²⁶

τ_avg = a₁τ₁ + a₂τ₂ + a₃τ₃ (1)

3.2 Photoelectrochemical activity

Photoelectrochemical measurement of pristine ZnO microspheres (Z-CSHM) and ZG samples was carried out with the help of Metrohm, Autolab AUt 85930 instrument using standard three-electrode cell under dark as well as visible-light irradiation. A 300 W xenon lamp with a water filter of 1 M NaNO₂ solution was used as the visible light (≥400 nm) source.²⁷ In this measurement, a Pt wire and an Ag/AgCl/3 M KCl electrode were taken as the counter electrode and reference electrode, respectively. The individual working electrode was prepared on cleaned ITO coated glass from paste-like viscous material from the samples adopting doctor-blading technique. The sample was thoroughly grinded and mixed with DMF solvent for making the paste. Finally, the coated glass was dried at 90 °C for a long time (∼5 h) in an air oven. It is noted that an aqueous solution of 0.1 M Na₂SO₄ was used as an electrolyte for the measurement of photoelectrochemical properties. The chronoamperometric study (photocurrent density vs. time, I–t) of pristine ZnO (Z-CSHM) and ZG samples was carried out at zero bias potential vs. Ag/AgCl/3 M KCl.

3.3 Photocatalytic activity

In this measurement, the pristine ZnO (Z-CSHM) and ZG nanocomposites were dispersed separately in 50 mL 10⁻⁵ M (0.00479 g L⁻¹) rhodamine B (RhB) aqueous solution. The amount of samples in the dye solution was kept fixed (1.0 mg mL⁻¹). Initially, the dispersed solutions were stirred in a dark condition at room temperature for 60 min to achieve adsorption–desorption equilibrium. In this study, C₀ and C represent the initial concentration after adsorption–desorption equilibrium in the dark and actual concentration of RhB dye solution at a specific time of the light irradiation, respectively. It is worthy to note that the same visible light source as demonstrated in photoelectrochemical study was also used in this experiment. The photocatalytic activity (PA) of samples under the visible-light exposure towards degradation of RhB (10⁻⁵ M solution) was monitored by measuring the concentration change of RhB dye at ∼554 nm. At different time intervals, 5 mL of the suspensions were taken out and measured their UV-Vis absorption spectra. The performance of PA of the samples was obtained from the plot of C/C₀ against the time of visible light exposure. It is important to note that for successive recycling PA test of the samples, the photocatalysts were washed about 5 times by ethanol and deionized water and measured their PA.

4. Results and discussion

4.1 Material properties

In the reaction medium, a major content of zinc acetate dihydrate (ZA) precursor could form ZnO nanoparticles (NPs) and the rest of ZA would produce Zn²⁺ ions, chemisorb onto the surface of the ZnO nanoparticles forming ZnO/Zn²⁺ moiety.²⁸ Also, during the reaction, the ZnO/Zn²⁺ moiety would interact with the available oxygen functional groups of graphene oxide (GO), transforming GO into chemically converted graphene (CCG) as revealed from XPS, FTIR and Raman spectral analyses (Fig. S1, ESI†). Further, it is expected that the ZnO NPs could start to aggregate preferentially and self-assembled into metastable microspheres by “oriented attachment” to minimize the total surface energy.²⁹ In the present work, on prolonging the reaction time, the solid microspheres (reaction time, 8 h; ZG-SM) converted into core–shell structured hollow microspheres (reaction time, 9 h; ZG-CSHM) and finally, to hollow microspheres (reaction time, 12 h; ZG-HM) by the loss of smaller NPs into larger NPs via Ostwald ripening process.³⁰ A probable mechanism for the formation of solid, core–shell and hollow microspheres of ZG samples is displayed in Scheme 1. The structural aspect (Fig. 1 and 2) of the ZG microspheres was systematically studied by using electron microscopes and discussed this issue in the next section.

Scheme 1 Schematic illustration for the formation of solid, core–shell and hollow structured microspheres of ZnO-chemically converted graphene nanocomposite.

Fig. 1 FESEM (a–d), TEM images (e–h) with different magnifications and HRTEM images (i and j) of ZG-CSHM showing the formation of core–shell structured hollow microspheres from CCG hybridized ZnO nanoparticles.

Fig. 2 (a) X-ray diffraction patterns of ZG samples prepared by varying reaction time (8 to 12 h). TEM images of the ZG samples collected at different reaction time: (b) 8 h, (c) 9 h, (d) 10 h and (e) 12 h.

Morphology and microstructure analyses of ZG microspheres were done by FESEM and TEM measurements (Fig. 1). The FESEM images (Fig. 1a–d) of ZG-CSHM (the sample collected from 9 h reaction time) show the formation of core–shell structured hollow microspheres (size range, 1.5–2.5 μm). The presence of broken pores with a core is clearly visible in the microspheres from the magnified FESEM images (Fig. 1c and d) of the sample. Moreover, the TEM images (Fig. 1e–h) of the specimen also confirm the existence of core–shell features of the hollow microspheres and support the FESEM results (Fig. 1a–d). In the microspheres, the core and shell are found to be constructed by nanoparticles (average diameter, 12 nm) as evident from TEM study (Fig. 1g–j). The TEM images clearly visualize nanoparticles (NPs) that are closely interacted by wavy strips of CCG layers (Fig. 1i) whereas the HRTEM images (Fig. 1j) of the NPs show distinct lattice fringes having inter-planar distance of 0.26 nm, fully matched with (002) plane of hexagonal ZnO (h-ZnO).

A systematic investigation on the change of morphology of ZnO–CCG (ZG) nanocomposite towards formation of microspheres having different types of porous networks was performed by time dependent experiments with the help of TEM and XRD studies (Fig. 2). It is also noted that the ZG nanocomposites, synthesized at different reaction times were collected and observed their structural changes. In this respect, it was seen that at ≥7 h reaction time, h-ZnO was found to form as confirmed from XRD pattern of the sample (Fig. 2a). However, the characteristic XRD peak (2θ) of GO at 10.9° was completely disappeared at ≥7 h reaction times without the appearance of obvious XRD peaks for the existence of CCG. This observation could be explained on the basis of chemical interaction that happened between the oxygen functional groups of the organic and inorganic moieties (ZnO/Zn²⁺) as well as due to chemical exfoliation of graphene layers/decrease in size of sp² domain of graphene. Further, it was observed that at 8 h reaction time, uniform solid microspheres were formed (Fig. 2b) but the sample collected at 9 h, the solid microspheres transformed into core–shell structured hollow microspheres (Fig. 2c). This would form via subsequent Ostwald ripening process.³⁰ It is also observed that the core of the microspheres became thinner at 10 h reaction time as obtained from the TEM image of the sample (Fig. 2d) but on further prolonging the reaction time to 12 h, the core–shell structures converted into hollow interiors (Fig. 2e). In this respect, a gradual change of relative intensity of XRD peaks, along (002) and (101) planes of h-ZnO would convey an information for the change of morphology ZG nanocomposite depending upon the reaction time for the synthesis of materials. In this process, by prolonging the reaction time, the process of hollowing of microsphere would start at the region underneath an immediate surface layer in the microspheres by further Ostwald ripening.^10,30 It is important to note that without addition of GO in the precursor, the solid and hollow architectures of ZnO were also formed in the similar synthesis condition as adopted for the preparation of ZG microspheres (Fig. S2, ESI†). This result implies that the presence of CCG has no role on the formation of solid, core–shell hollow or hollow microspheres.

To understand the surface defects in ZnO nanoparticles of ZG samples, Raman spectral study was performed (Fig. 3a). Two high intensity Raman peaks at observed 99 and 437 cm⁻¹ could assign to low and high E₂ non-polar vibration modes of hexagonal ZnO.^15,17 In addition, the Raman vibrations appeared at 203 and 331 cm⁻¹ could attribute to the secondary Raman scattering arising from zero-boundary phonons of ZnO.¹⁵ In addition, a relatively weak intensity peak located at ∼580 cm⁻¹ could suggest the presence of oxygen deficiency of the samples.^15,31 It is worthy to note that the conversion of graphene oxide to chemically converted graphene is well established from FTIR and Raman spectral analyses. In this respect, the details are given under the Fig. S1b of ESI.†

Fig. 3 (a) Raman spectra of pristine ZnO and ZG samples. Typical XPS data for O1s signals of pristine ZnO (Z-CSHM) (b) and ZG-CSHM (c).

The presence of surface oxygen defects was further confirmed from XPS study (Fig. 3b and c). The O1s spectra of pristine ZnO (Z-CSHM) and ZG-CSHM could be resolved into three Gaussian fitted peaks. Among these, two peaks are observed at 530.1 and 531.7 eV, could correspond to the binding energies related to lattice oxygen (O_lattice) and oxygen deficient region (O_defect), respectively in hexagonal ZnO crystal.^31,32 On the other hand, the higher binding energies located at ∼532 eV in ZnO and ∼533.1 eV in ZG-CSHM could be due to the presence of hydroxyl groups (O_hydroxyl) and oxygen functional groups (hydroxyl/COOH/epoxy) of CCG (O_{hydroxyl/carboxyl/epoxy}),²⁸ respectively. It is further noted that the OH-related XPS peak intensity is seemed to be decreased in ZG-CSHM compare Z-CSHM as observed in the O1s XPS signals (Fig. 3b and c) of the samples. Therefore, during the transformation of GO to CCG, suitable oxygen functional groups of graphene oxide could chemically react with the surface OH-groups of ZnO/Zn²⁺.²⁸ This process could reduce the OH content in the graphene hybridized ZG-CSHM nanocomposite. It is worthy to note that the in situ hybridization of graphene with ZnO nanoparticles, forming ZG nanocomposites also preserve the surface defects especially oxygen vacancies which would play a crucial role in light harvesting as well as photogenerated charge separation in photoelectrochemical process.^13,14 It should mention that the conversion of graphene oxide to CCG in the ZG nanocomposite is confirmed by C1s XPS analyses of ZG-CSHM (given in details under Fig. S1c and d of ESI†).

The optical property of pristine ZnO and ZG samples were measured using UV-Vis diffused reflectance spectral study. It is found (Fig. 4a) that the incorporation of GO in the reaction medium, produced ZG nanocomposites, significantly enhances the visible light absorption compare to pristine ZnO (Z-CSHM). However, all the ZG samples show high visible absorption compare to Z-CSHM indicating that the CCG in the nanocomposite would play a crucial role for the effect.¹⁶ We calculated the direct band gap energy (Fig. 4b) of pristine ZnO (Z-CSHM) of ∼3.24 eV whereas the band gap energy of ZG nanocomposites was found to decrease to certain extent (3.0 ± 0.04 eV). This narrowing of band gap would result from the synergic effect of hybridization of graphene and the presence of surface defects.^16,33

Fig. 4 (a) UV-Vis diffused reflectance spectra along with (b) determination of direct band gap energy of pristine ZnO (Z-CSHM) and ZG samples.

Textural property of pristine ZnO (Z-CSHM) and ZG nanocomposites were investigated by BET nitrogen adsorption–desorption isotherms (Fig. 5). The isotherms of all the ZG samples could highlight an IUPAC type IV with H3 hysteresis loop that imply an existence of slit-like mesopores.³⁴ However, the calculated BET specific surface area of the nanocomposites was always higher than that of the pristine ZnO (Table 1). In this respect, the presence of chemically interacted CCG layers with high surface area in the ZG nanocomposites could be one of the reasons. It is worthy to mention that in situ hybridization of graphene with the ZnO nanoparticles (NPs) in ZG-CSHM could create mesopores with narrow pore size (average pore size, ∼15.0 nm) distribution due to mutual interaction of ZnO NPs with the CCG.³⁵ Moreover, among the ZG nanocomposites, ZG-CSHM possessed high specific surface area with a narrow pore size distribution compare to other samples including Z-CSHM (Table 1). It is expected that the generation of large surface area with narrow pore size distribution of a material could provide accessible diffusion and mass transport of an electrolyte in photoelectrochemical process (discussed later).

Fig. 5 Multipoint BET nitrogen adsorption–desorption isotherms of pristine ZnO (Z-CSHM) and ZG samples (insets show the individual pore size distribution curve).

Table 1 Textural properties of pristine ZnO (Z-CSHM) and ZG nanocomposites

Sample	SBET (m^2 g^−1)	Pore volume (cm^3 g^−1)	Average pore size (nm)
Z-CSHM	29.4	0.19	26.9
ZG-SM	30.7	0.20	27.2
ZG-CSHM	40.5	0.15	15.3
ZG-HM	37.6	0.12	12.2

---

4.2 Photoelectrochemical and photocatalytic activities

The advanced structural, textural and optical properties (Fig. 1–5) of hierarchical core–shell structured hollow ZnO–graphene nanocomposites could expect to show an improved photoelectrochemical and photocatalytic activities. In this respect, to demonstrate the superiority of the material towards visible light water splitting, the ZG nanocomposites with different morphologies were coated on ITO glass and were used as working electrodes. The photoelectrochemical activity was investigated using three electrode systems in 0.1 M Na₂SO₄ solution. Fig. 6a shows the linear sweep voltammograms (LSV) from −0.4 V to 1 V vs. Ag/AgCl/3 M KCl reference electrode of pristine ZnO (Z-CSHM) and ZG nanocomposites under illumination of visible light from the xenon lamp source. The photocurrent response at zero volt vs. Ag/AgCl/3 M KCl electrode of the respective samples is displayed in Fig. 6b. It is found (Fig. 6b) that the photocurrent density of ZG-CSHM is higher than that of pristine ZnO (Z-CSHM) as well as other ZG nanocomposites (ZG-SM and ZG-HM). However, we obtained the maximum photocurrent density, 0.15 mA cm⁻² of ZG-CSHM at 1 volt vs. Ag/AgCl/3 M KCl electrode as measured from LSV curve (Fig. 6a). To further enhance the photocurrent of the material, metal chalcogenide quantum dots (such as CdS, CdSe) having low band gap energies or noble nano metal would be incorporated^36–39 into the nanocomposite. The high current density of core–shell structured microspheres in comparison to solid and hollow microspheres of ZG nanocomposites could originate from its superior light scattering ability, high surface area and more active sites for electron conduction.^9,10,13 On the other hand, core–shell structured microspheres of pristine ZnO (Z-CSHM) showed poor photocurrent density compare to core–shell structured ZnO–graphene microspheres (ZG-CSHM). In this respect, the intimate contact of graphene layers with the ZnO nanoparticles is evidenced from TEM study (Fig. 1). This strategy could suppress the recombination of photogenerated charge carriers.^16,21 It is known that an important criterion for the photoelectrochemical water splitting is the photo stability of materials under prolonged light illumination.²¹ It is observed that various surface defects including oxygen defects still existed even after hybridization of CCG with ZnO (Fig. 3c and 7a discussed later). Therefore, the synergic effect of graphene hybridization and the surface defects especially oxygen vacancies in the ZG nanocomposites could led to enhance the visible absorption.^16,33 As mentioned before, the photocurrent density (Fig. 6c) of pristine ZnO (Z-CSHM), ZG-SM, ZG-CSHM and ZG-HM were investigated at zero volt versus Ag/AgCl/3 M KCl electrode. For pristine ZnO, the poor stability of photocurrent was found with continuous decay on prolonging the irradiation time. This could be considered due to the consumption of photogenerated holes as a result of self-oxidation of ZnO with existing oxygen atom on the ZnO surface whereas the existence of graphene layers on ZnO NPs could offer a sufficient transport pathway of photogenerated electrons in the system.²¹ As a result, the photo stability of ZG-CSHM is enhanced. This enhanced photoelectrochemical activity of the ZG nanocomposites compare to pristine ZnO could be due to the direct chemical contact and charge transfer phenomenon that happened between ZnO and graphene.

Fig. 6 Electrochemical properties of samples: (a) linear sweep voltammetry (LSV) curves, (b) photocurrent response through on–off cycles at zero volt vs. Ag/AgCl/3 M KCl and (c) change of current density with time (I–t curves) recorded under visible light irradiation of pristine ZnO (Z-CSM), ZG-SM, ZG-CSHM and ZG-HM samples.

Fig. 7 (a) Photoluminescence spectra of ZG samples with pristine ZnO (Z-CSHM). (b) Schematic presentation on the charge separation of photoelectrochemical process for ZG nanocomposites under the visible light irradiation.

The charge transfer mechanism between ZnO and graphene could be revealed from the decrease of photoluminescence (PL) band edge emission intensity at ∼385 nm of ZnO.¹⁶ The PL spectra (Fig. 7a) showed a prominent decrease of the emission peak intensity in ZG nanocomposites compare to pristine ZnO. The quenching of the PL emission could be accounted for the electron transfer from excited ZnO NPs through an interfacial charge transfer mechanism. A possible charge transfer mechanism is schematically shown in Fig. 7b. It is noted that except the UV emission at ∼385 nm, several visible emissions were observed at ∼420 nm, ∼460 nm, ∼485 nm and ∼530 nm in the PL spectra of the sample. These PL emissions could suggest the presence of various intrinsic/extrinsic surface defects in ZnO nanoparticles as well as in the ZG samples. It is known that different surface defects such as oxygen and zinc vacancies, oxygen and zinc interstitials, antisite oxygen^27,31 would form within the band gap of ZnO semiconductor. The formation and the density of the defects would also depend upon several factors like preparative methods,^15–17 doping³¹ etc. It is noted that these defects could able to generate PL emissions on excitation with photons of suitable energy. In the present work (Fig. 7a), the appearance of a PL emission at 460 nm could imply the presence of single ionized oxygen vacancy.³¹ At the same time, the presence of zinc interstitial could also be supported from the observation of PL emission at 420 nm.^17,31 In addition, the existence of neutral oxygen vacancy could be formed in the materials because a prominent emission peak is observed at 530 nm.^27,31 It is further noted that the combination of electrons between the antisite oxygen and the conduction band would be the origin for the emission appeared at 485 nm. The UVPL emission peak appeared at ∼385 nm would correspond to the radiative recombination (band edge emission) of the electrons in the conduction band with the holes in the valence band of ZnO semiconductor. To understand the faster charge transfer that took place between ZnO and graphene in ZG-CSHM nanocomposite, the decay of the PL emission at 485 nm was measured for pristine ZnO (Z-CSHM) and ZG-CSHM nanocomposite. It is found that the PL decay of ZG-CSHM was faster (average life time τ_avg, 1.25 ns) than that of the pristine ZnO (τ_avg, 2.24 ns), implying the presence of an additional high-efficiency relaxation channel in the nanocomposite²¹ (details are given in Fig. S3 and Table S1 of ESI†). This result also confirms the faster charge transfer that took place²¹ between the ZnO and graphene in the ZG-CSHM nanocomposite compare to Z-CSHM.

The presence of surface defects (such as oxygen deficiencies) present in the sample is also confirmed from Raman (Fig. 3a) and XPS (Fig. 3b) spectral analyses. These surface oxygen defect would play an important role for enhancing visible light absorption. In addition, the defects would act as charge carrier traps that could prevent electron–hole recombination process¹³ as schematically shown in Fig. 7b. Thus, synergetic effect of graphene hybridization with ZnO and the presence of surface defects in ZnO in the sample could play the crucial role for the enhancement of photocurrent density in ZG nanocomposites compare to Z-CSHM.

It is known that photoelectrochemical activity of a material could depend on several factors such as microstructure, crystallinity, surface area, porosity.^5,40 For a photoelectrochemical reaction, the hierarchical porous structure having large surface area and interconnected mesopores with narrow pore size distribution (Fig. 5 and Table 1) could provide large number of active sites for ionic diffusion and mass transportation.⁴⁰ Hence, ZG-CSHM having large specific surface area with narrow pore size distribution as well as high crystalline framework could be responsible for generating high photocurrent density compare to other ZG microspheres even Z-CSHM. In this respect, the narrow pore size distribution (Fig. 5) of ∼15.0 nm interconnected to the inner and outer wall of hierarchical core–shell structured microspheres (ZG-CSHM) could create more pore opening for internal surface.⁴⁰

To investigate the long term photo stability of ZG nanocomposites, we also measured photocatalytic activity (PA, Fig. 8a) of the samples under the visible light irradiation using RhB dye solution (10⁻⁵ M) adopting successive recycles of the PA for the photocatalysts. It was seen that the ZG nanocomposites show higher photocatalytic activity than pristine ZnO (Z-CSHM). However, the calculated photodegradation rate constant (k) value (considering pseudo-first order chemical reaction kinetics, Fig. 8b) of ZG-CSHM shows higher k value than ZG-SM and ZG-HM nanocomposites. This difference could be related to the change of surface area of the samples (Table 1). On the other hand, the long term stable photocatalytic activity towards degradation of RhB dye of the photocatalyst by the recycling test using pristine ZnO (Z-CSHM) and ZG-CSHM samples is shown in Fig. 8c. It could be seen that ZG-CSHM shows nearly similar photocatalytic stability than pristine ZnO after third successive recycling. This prevention of photocorrosion of ZG-CSHM sample could be due to the presence of chemically interacted graphene layers onto the ZnO NPs (Fig. 1i and j and S1 of ESI†).

Fig. 8 (a) Photocatalytic degradation (remnant dye concentration versus irradiation time) of aqueous RhB dye solution in presence of Z-CSHM, ZG-SM, ZG-CSHM and ZG-HM samples as well as in absence of photocatalyst (blank) under visible light irradiation. (b) Determination of pseudo-first order reaction rate constants for the degradation of RhB of the samples (the rate constants of the dye degradation reaction for the different photocatalysts are embedded in the figures). (c) Performance of Z-CSHM and ZG-CSHM as photocatalysts after 1^st, 2^nd and 3^rd runs (successive recycles).

5. Conclusion

A successful in situ synthesis of graphene hybridized surface defective core–shell structured ZnO hollow microspheres has been reported adopting low temperature solution method in air atmosphere from surfactant/template free precursor. This particular core–shell structure is found to be developed via Ostwald ripening, has been achieved from other morphologies (solid and hollow microstructures) by optimizing the reaction time only. In situ hybridization of graphene with ZnO nanoparticles enhances specific surface area and generates narrow mesopore size distribution via chemically interacting with ZnO nanoparticles. A high photoelectrochemical activity with long term photo stability of graphene hybridized core–shell structured hollow ZnO (ZG-CSHM) microspheres is observed than other microspheres including pristine ZnO core–shell structured hollow microspheres. The synergic effect of graphene hybridization and the presence of surface defects of the ZnO nanoparticles can be caused for advancing the property. In ZG-CSHM, the graphene hybridization can also improve the photocurrent density and inhibits the photocorrosion by intimate contact of graphene with the pristine ZnO nanoparticles as well as for the generation of narrow mesopore size distribution which can provide large number of active sites for easy electrolyte diffusion and mass transportation. In addition, the presence of surface defects can prolong the recombination rate of photogenerated charge carriers. This facile process can create an avenue for synthesis of graphene hybridized core–shell hollow microspheres in other systems for improving the photoelectrochemical activity of the materials.

Acknowledgements

Authors wish to acknowledge the Director, CSIR-CGCRI for his kind permission to publish this work. The authors, SB, AN and MP thank Council of Scientific and Industrial Research (CSIR) and UGC, Govt. of India for providing their research fellowships. Authors also acknowledge the help rendered by Nanostructured Materials Division and Electron Microscopy Section for Raman and microstructural characterizations, respectively. The authors are also very much thankful to Prof. Nitin Chattopadhyay, Department of Chemistry, Jadavpur University, Kolkata, India for measurement of PL lifetimes of the samples in his lab. The work has been done as an associated research work of 12^th Five Year Plan project of CSIR (No. ESC0202).

References

1. X. W. Lou, L. A. Archer and Z. Yang, Adv. Mater., 2008, 20, 3987–4019.
2. X. Lai, J. E. Halpert and D. Wang, Energy Environ. Sci., 2012, 5, 5604–5618.
3. M. Chen, C. Ye, S. Zhou and L. Wu, Adv. Mater., 2013, 25, 5343–5351.
4. J. Liu, T. Luo, S. T. Mouli, F. Meng, B. Sun, M. Li and J. Liu, Chem. Commun., 2010, 46, 472–474.
5. S. S. Shin, D. W. Kim, J. H. Park, D. H. Kim, J. S. Kim, K. S. Hong and I. S. Cho, Langmuir, 2014, 30, 15531–15539.
6. R. Gao, L. Hu, M. Chen and L. Wu, Small, 2014, 10, 3038–3044.
7. C. K. Ngaw, V. B. Wang, Z. Liu, Y. Zhou, S. Kjelleberg, Q. Zhang, T. T. Y. Tan and S. C. J. Loo, npj Biofilms and Microbiomes, 2015, 1, 15020.
8. S. Liu, J. Yu and M. Jaroniec, J. Am. Chem. Soc., 2010, 132, 11914–11916.
9. J. Feng, Y. Hong, J. Zhang, P. Wang, Z. Hu, Q. Wang, L. Hanb and Y. Zhu, J. Mater. Chem. A, 2014, 2, 1502–1508.
10. Z.-Q. Li, W.-C. Chen, F.-L. Guo, L.-E. Mo, L.-H. Hu and S.-Y. Dai, Sci. Rep., 2015, 5, 14178.
11. Z. Dong, X. Lai, J. E. Halpert, N. Yang, L. Yi, J. Zhai, D. Wang, Z. Tang and L. Jiang, Adv. Mater., 2012, 24, 1046–1049.
12. Z. Dong, H. Ren, C. M. Hessel, J. Wang, R. Yu, Q. Jin, M. Yang, Z. Hu, Y. Chen, Z. Tang, H. Zhao and D. Wang, Adv. Mater., 2014, 26, 905–909.
13. M. A. Rahman, S. Bazargan, S. Srivastava, X. Wang, M. A. Ellah, J. P. Thomas, N. F. Heinig, D. Pradhan and K. T. Leung, Energy Environ. Sci., 2015, 8, 3363–3373.
14. C. Zhu, C. Li, M. Zheng and J.-J. Delaunay, ACS Appl. Mater. Interfaces, 2015, 7, 22355–22363.
15. D. Liu, Y. Lv, M. Zhang, Y. Liu, Y. Zhu, R. Zong and Y. Zhu, J. Mater. Chem. A, 2014, 2, 15377–15388.
16. X. Bai, L. Wang, R. Zong, Y. Lv, Y. Sun and Y. Zhu, Langmuir, 2013, 29, 3097–3105.
17. Z. Pei, L. Ding, M. Lu, Z. Fan, S. Weng, J. Hu and P. Liu, J. Phys. Chem. C, 2014, 118, 9570–9577.
18. C. Chen, W. Cai, M. Long, B. Zhou, Y. Wu, D. Wu and Y. Feng, ACS Nano, 2010, 4, 6425–6432.
19. S. Bera, N. Das, M. Pal, S. Mahanty and S. Jana, J. Sol-Gel Sci. Technol., 2015, 76, 402–413.
20. T.-F. Yeh, J. Cihlar, C.-Y. Chang, C. Cheng and H. Teng, Mater. Today, 2013, 16, 78–84.
21. J. K. Kim, S. Bae, W. Kima, M. J. Jeong, S. H. Lee, C.-L. Leee, W. K. Choi, J. Y. Hwang, J. H. Park and D. I. Son, Nano Energy, 2015, 13, 258–266.
22. J. Qi, X. Lai, J. Wang, H. Tang, H. Ren, Y. Yang, Q. Jin, L. Zhang, R. Yu, G. Ma, Z. Su, H. Zhaod and D. Wang, Chem. Soc. Rev., 2015, 44, 6749–6773.
23. M. Niederberger, Acc. Chem. Res., 2007, 40, 793–800.
24. P. Biswas, S.-D. Baek, S. H. Lee, J.-H. Park, S. J. Lee, T. I. Lee and J.-M. Myoung, Sci. Rep., 2015, 5, 17961.
25. R. T. Ginting, H. B. Lee, S. T. Tan, C. H. Tan, M. H. H. Jumali, C. C. Yap, J.-W. Kang and M. Yahaya, J. Phys. Chem. C, 2016, 120, 771–780.
26. S. Ghosh, S. Chakraborty, D. Bhowmik, G. S. Kumar and N. Chattopadhyay, J. Phys. Chem. B, 2015, 119, 2090–2102.
27. S. Martha, K. H. Reddy and K. M. Parida, J. Mater. Chem. A, 2014, 2, 3621–3631.
28. S. Bera, M. Ghosh, M. Pal, N. Das, S. Saha, S. K. Dutta and S. Jana, RSC Adv., 2014, 4, 37479–37490.
29. D. Li, M. H. Nielsen, J. R. I. Lee, C. Frandsen, J. F. Banfield and J. J. D. Yoreo, Science, 2012, 336, 1014–1018.
30. B. Liu and H. C. Zeng, Small, 2005, 1, 566–571.
31. M. Pal, S. Bera, S. Sarkar and S. Jana, RSC Adv., 2014, 4, 11552–11563.
32. W. Liu, W. Li, Z. Hu, Z. Tang and X. Tang, J. Appl. Phys., 2011, 110, 013901.
33. Y. Feng, N. Feng, Y. Wei and G. Zhang, RSC Adv., 2014, 4, 7933–7943.
34. C. Sangwichien, G. L. Aranovich and M. D. Donohue, Colloids Surf., A, 2002, 206, 313–320.
35. S.-M. Paek, E. J. Yoo and I. Honma, Nano Lett., 2009, 9, 72–75.
36. P.-Y. Kuang, Y.-Z. Su, K. Xiao, Z.-Q. Liu, N. Li, H.-J. Wang and J. Zhang, ACS Appl. Mater. Interfaces, 2015, 7, 16387–16394.
37. H. Kim and K. Yong, ACS Appl. Mater. Interfaces, 2013, 5, 13258–13264.
38. C. Zhang, M. Shaon, F. Ning, S. Xu, Z. Li, M. Wein, D. G. Evans and X. Duan, Nano Energy, 2015, 12, 231–239.
39. X. Zhang, Y. Liu and Z. Kang, ACS Appl. Mater. Interfaces, 2014, 6, 4480–4489.
40. R. Zhang, D. Shen, M. Xu, D. Feng, W. Li, G. Zheng, R. Che, A. A. Elzatahry and D. Zhao, Adv. Energy Mater., 2014, 4, 1301725.

---

Footnote

† Electronic supplementary information (ESI) available. See DOI: 10.1039/c6ra03410a

---