Cu₂S-incorporated ZnS nanocomposites for photocatalytic hydrogen evolution

Abstract

Sodium Lauryl Sulphate (SLS) surfactant and propylene diamine (PD), assisted wet chemical synthesis was used to make ZnS–Cu₂S nanoflakes wherein the added copper(II) transforms into copper(I) sulphide and is found to be deposited on the surface of ZnS. With a variation of copper concentration from 0 to 5% there is a morphological transformation from ZnS nanorods eventually to ZnS–Cu₂S nanoflakes through a transition morphology of nanocactus leaves. During the addition of copper there is an incorporation of Cu₂S into the ZnS phase (100) as is clearly evidenced by various characterization methods. Visible light photocatalytic hydrogen production activities using these nanoflakes of ZnS–Cu₂S are reported with good results. The influence of Cu₂S shifts the band gap of ZnS from the UV to the visible region, reducing the need for an expensive co-catalyst like platinum for photolysis of water. Though hydrogen production is not as high as that depicted by other earlier works the material that we have created is a relatively cheap, simple two component heterostructure with no expensive third component. It is also free from toxic materials such as CdS. However, our results are better than for most other copper loaded ZnS systems in the literature. Furthermore, the morphological evolution to nanoflakes from nanorods, the concentration and dispersion of Cu₂S over ZnS and the interface between Cu₂S and ZnS semiconductors play a vital role in hydrogen production. 5% Cu₂S on ZnS seems to be the optimum concentration for maximum evolution of hydrogen.

---

Introduction

With the depletion of fossil fuel sources, hydrogen is considered to be one of the most likely alternative fuels for future energy needs. It is also a clean fuel source since it minimizes CO₂ input into the environment. Hydrogen can also be commercially produced by steam reforming of hydrocarbons which is neither ecofriendly nor economically viable.^1,2 Photocatalytic hydrogen evolution, in which the energy from a visible light source is trapped with feasible quantum efficiency, is quite challenging. It is also equally challenging to tune the photocatalyst without precious metals as co-catalysts in order to cut the cost. The strategy is to dope the semiconductors with suitable non-precious metals as co-catalysts in order to make it visible light active and yet to maintain quantum efficiency. The morphology and particle size of the semiconductor also play significant roles in the photocatalytic activity of the semiconductor. A CdS semiconductor photocatalyst with a narrow band gap of 2.4 eV exhibits visible light activity; along with this an added advantage of a large negative flat band potential has been used for hydrogen evolution from aqueous solutions containing sacrificial reagents of Na₂S/Na₂SO₃.³ Yet another advantage of CdS is the migration of generated electrons and holes inside the CdS to the surface, which in turn decrease the bulk e⁻/h⁺ recombination.⁴ Past reports reveal that the CdS NC's were synthesized by many methods using complexing agents like thiourea.⁵ However, CdS has serious limitation due to its photo corrosion and toxicity. Among the various metal sulfide semiconductor photo catalysts, zinc sulfide (ZnS) has certain advantages over others, like nontoxicity, ability to produce electron and hole pair for photocatalysis, and low cost. Apart from advantages, the disadvantage is its activity only in the higher energy edge of the visible region. In order to circumvent this disadvantage, much efforts have been done in past like band engineering by doping with metals and addition of co-catalyst.^6–8 Transition metal ion doping has attracted much attention due to its capability to introduce impurity levels in the forbidden bands. Among the various metal ions like Ni and Sn, copper has shown significant activity without the need for the additional co catalyst.^9,10 High quantum yields can only be achieved by addition of precious metal like platinum, leading to significant increase in cost. Zinc sulfide photocatalyst due to its high conduction band potential, even after doping shows good photocatalytic property without the need for additional co-catalyst.^11,12 ZnS photocatalyst with two other components graphene and MoS₂ as cocatalysts has been used as reported in a more recent paper.¹³ In this report we have tried to overcome the foresaid challenges by adapting surfactant and amine assisted synthesis¹⁴ of copper loaded zinc sulfide to achieve proper morphology, visible light activity and fine dispersion of the copper over zinc sulfide. The present paper also investigates the mechanism for improvement of visible light active photocatalytic hydrogen production due to the one dimensional growth of the nanoparticles of ZnS on copper addition as well as the role of copper in the photocatalytic activity. Unlike the other researchers¹⁵ they used hydrothermal method to synthesize nano sheets of the copper sulfide loaded zinc sulfide which is energy intensive process whereas we could do the synthesis by simple soft chemical route which is equally efficient in photocatalytic hydrogen evolution. Also the copper loaded on the ZnS surface is of the form Cu₂S rather CuS as in the earlier work.¹⁵ In the soft chemical synthesis of ZnS–Cu we use thiourea for sulphur source which is well known for its complexation properties. The basic nature of our work is to probe the creation of a photocatalyst by a simple procedure with nontoxic materials and with no expensive co-catalyst metals like Pt.

Experimental

Preparation

The preparation involves a slight modification of our previously reported procedure,¹⁶ 3 g SLS (sodium lauryl sulphide) was dissolved in appropriate amount of ultra-pure distilled water. To this, 0.8924 g of zinc acetate and 10 ml of 1,3-diaminopropane (propylene diamine; PD) were added and the resultant solution kept for refluxing with constant stirring. During the refluxing, 0.684 g of thiourea, dissolved in minimum amount of distilled water, was added to the reaction mixture and the refluxing continued for 20 h at 120 °C. The appearance of white solids indicated the formation ZnS nanocrystals. The precipitate was centrifuged, washed several times with distilled water and methanol. Finally the product was thoroughly dried in vacuum oven at 80 degrees for 3 h. The copper doped ZnS nanocrystals have been synthesized with appropriate amount of copper acetate as 0.1, 0.5, 1 and 5 mol% mixed with zinc acetate employing the same procedure.

Characterization

The X-ray diffraction (XRD) patterns are recorded using Bruker Discover D8 diffractometer with Cu-K_α radiation in the 2θ range of 10–80°. The morphology of the sample is scanned using FEI Quanta FEG 200-High Resolution Scanning Electron Microscope (SEM) equipped with energy dispersive X-ray spectrometry (EDX). The optical properties of the samples were analysed by UV-visible Diffuse Reflectance Spectroscopy (UV-vis DRS) using UV2250, Shimadzu, in which BaSO₄ was used as the background. The metal compositions were determined using inductively coupled plasma optical emission spectroscopy (ICP-OES) on a Perkin Elmer ICP Optima 5300DV machine. The chemical states of the samples were determined by X-ray photoelectron spectroscopy (XPS) in Omicron with a monochromatic Al-K_α source and charge neutralizer. Photoluminescence measurement was done in the range of 180–1550 nm with PMT detector for UV-visible region (180 to 850 nm) with a resolution of 0.2 nm, using Jobin Yvon Flurolog-3-11 Spectrofluorometer. Radiation source is 450 W xenon lamp. The lattice fringes and selected area diffraction of the samples were recorded using High Resolution Transmission Electron Microscopy (HRTEM) JEOL 3010 working at 300 kV. BET (Brunauer, Emmett and Teller) surface area analysis were measured at 77 K on a Quadrasorb SI, Quantachrome Instruments using liquid N₂ adsorption desorption isotherm method. Prior to analysis, the samples were degassed at 473 K for 2 hours.

Results and discussion

The two main objectives of this work are the photocurrent measurement and subsequently the photocatalytic evolution of hydrogen from water. Details on photo-electrochemical characterization and photocatalytic hydrogen evolution are given in ESI.† It is necessary to bring into the making of the photocatalyst and its thorough characterization especially the introduction of cuprous sulphide, Cu₂S into ZnS facilitated by a change of phase on increasing the concentration of cupric ions transforming into cuprous ions. The phase change from nanorods to nanoflakes to form Cu₂S incorporated ZnS nanocomposites is the main tenet of this work. While XPS and EPR characterizes the oxidation state of cuprous ions the XRD, optical spectroscopy via both UVDRS and PL measurements probe the transformation of ZnS from nanorods to nanoflakes on adding about 0.5% copper. Further support for this comes from microscopy measurements. The role of morphology in the photocatalytic activity is well proved in many recent reports^17–20 supporting our claim that both morphology change and chemical composition influence the photocatalytic hydrogen evolution.

Identification of morphology and growth phase by XRD

The morphology and the influence of incorporation of second component towards photo catalytic hydrogen evolution are well proven by many previously reported studies.²¹ The cubic zinc blende and hexagonal wurtzite structures are the two major structures of ZnS in which the hexagonal structure has a band gap of 3.91 eV whereas the band gap in cubic morphology is 3.54 eV. The photocatalytic activity of wurtzite ZnS nanorods are known for its superior photocatalytic activity for degradation of dye.²¹ Fig. 1 shows the X-ray diffraction (XRD) patterns of pristine ZnS, and various percentages of Cu₂S incorporated (0.5% to 5%) zinc sulphide. The XRD peaks of 1% and 5% Cu₂S incorporated zinc sulfide matches with JCPDS card number (79-2204) which is hexagonal in nature. The one and five percent Cu₂S incorporated ZnS nanocomposites how differences in the intensities of two theta values of 27.0° (100) and 47.5° (110) with “d” spacing values of 3.3 Å and 1.9 Å when compared to other samples. These two peaks are broad for lesser concentration of copper but become sharper in 1% and 5% Cu²⁺ addition. The increase in the amplitude and sharpness of this particular phase is attributed to the growth in phases of (100) and (110). Further changes on increasing the copper concentration to 10% can be witnessed in the relative heights and widths of peaks in the 2θ region of 27–30°. These observations are supported by the morphologic evolution in Scanning Electron Microscope images (SEM) since there is formation of flake like structure with 1–5 nanometer thickness 5% concentration which decreases at 10% (vide infra). The addition of Cu₂S influences the morphology and phase change in the structure of zinc sulfide. Low concentration of 0.1 and 0.5% Cu₂S incorporated zinc sulfide does not show any significant change in the width of the XRD peaks of phases (100) and (110), essentially being broad. Seemingly, the pristine ZnS formed during the addition of thiourea to zinc salt in presence of amine is more like rods even though they are curvaceous in appearance. This morphology is not disturbed by addition of smaller amounts (0.1 and 0.5%) of copper ions. However, on further addition of Cu²⁺ ions sulphur is possibly facilitating the growth along (100) phase by using additional copper ions to form Cu₂S the phase of which is termed as (111) with the 2θ value being 27.4°. The Cu₂S corresponds to the JCPDS number 53-0502 which is in cubic phase. Even in the 0.5% addition of copper there is a slight change in the morphology from nanorods to nanoflakes which is clearly visible in the SEM images of ZnS–Cu₂S(0.5). This indicates that in 1% and 5% Cu₂S incorporated samples, the ZnS rod-like morphology, formed from zinc precursor and 1,3-diaminopropane (PD) complex, is being used to direct sulphur to grow in the (100) phase or the formation of Cu₂S over zinc sulfide nanorods. It is, however, very clear from XRD and SEM that copper sulfide grows in the (100) phase of the zinc sulfide. The propylenediamine (PD) acts as a structure directing agent and SLS surfactant acts as a capping agent in which the nucleation and growth are controlled. The dual advantage of structural directive and size control is achieved by this method of synthesis. The presence of Cu₂S in 5% samples is also proved by X-ray photoelectron spectroscopy (XPS) study which reveals the oxidation state of copper to be +1 (vide infra). It is true for the 1% sample also but the XPS data in that region is quite noisy due to low levels of cuprous ions. In terms of photocatalytic application one dimensional growth in (100) and (110) planes which are predominant in 1 and 5% cuprous sulphide loaded samples shows better activity. This may be due to the fact that the (100) and (110) planes are more exposed to the incoming light source, along with another growth plane at 2θ = 56.4° also identified as (112) plane. A good exposure of these three planes (100), (110) and (112) to light makes the sample more active towards photocatalytic hydrogen evolution.

Fig. 1 XRD patterns of ZnS–Cu₂S with different copper loadings.

Revelation of phase change and band gap change by UV-visible diffuse reflectance spectroscopy (UV-vis-DRS)

Fig. 2 shows the UV-vis-DRS spectroscopic images of SLS assisted Cu₂S incorporated ZnS systems. The UV-vis-DRS of SLS assisted copper loaded zinc sulfide show shift in the wavelength to the visible region with increase in the copper loading. The visible light shift is due to the formation of Cu₂S which does not have clear absorption edge when it is loaded in the ZnS lattice and it covers from UV to the visible region but the band gap of Cu₂S is 1.2 eV and it is a p-type semiconductor which is ideal for absorption of solar radiation.²² The band gap of the zinc sulfide is altered by Cu₂S incorporation; (Table 1) this is proved by the absorption shift from lower to the higher wavelength. The presence of Cu₂S in ZnS system introduces impurity level band structures which shifts the absorption towards visible region. The pure zinc sulfide sample also shows some deflection from the bulk zinc sulfide due to the particle size reduction leading to quantum confinement. Commercial or bulk ZnS shows band gap in the UV region and this limits the material for photocatalytic applications whereas in the present Cu₂S incorporated zinc sulfide systems the bandgap shift from UV to visible region promotes visible light photocatalytic hydrogen evolution. A look at the relative intensities of UV-vis spectra shows a continuous reduction on the incorporation of Cu₂S to ZnS samples. This could be attributed to possible change in morphology. The emergence of flake structure appears from 0.5% copper addition, transforming to a mixture of rods + flakes at 1% and then to pure flakes at 5%. However, an additional reason may be due to the nature of oxidation state of copper ions as a function of added copper. The hydrogen evolution from the visible light source clearly indicates the visible light activity of the samples. Moreover, the band pass filter and water circulated over the reactor which acts opaque towards light from the UV region and also water cuts off the IR region.

Fig. 2 UV-vis-DRS images of SLS assisted (a) 0 (b) 0.1 (c) 0.5 (d) 1 (e) 5 percent Cu₂S loaded ZnS-systems with estimated band gaps and (f) is the combined plot.

Table 1 Band gap, BET surface area, hydrogen production and Cu₂S concentration from ICP of ZnS–Cu₂S(0–5) samples

Sample	Band gap (eV)	Surface area^a (m^2 g^−1)	H2 production (μmol h^−1 g^−1)	Copper (%) from ICP
a Calculated using BET equation measured from adsorption of nitrogen at liquid nitrogen temperature.
ZnS–Cu2S(0)	3.4	21	20	0
ZnS–Cu2S(0.1)	3.2	24	1488	0.05
ZnS–Cu2S(0.5)	3.1	26	1785	0.35
ZnS–Cu2S(1)	2.9	28	2083	0.63
ZnS–Cu2S(5)	2.1	25	2232	3.07

---

Oxidation state of copper from X-ray photoelectron spectroscopy (XPS)

The elemental composition and oxidation state of copper of the prepared catalyst are probed by XPS, an elegant tool to determine the surface composition of the samples. The XPS survey spectrum of Cu₂S incorporated zinc sulfide, ZnS–Cu₂S(5) as shown in the Fig. 3a. The survey XPS spectra of the 5% Cu₂S incorporated zinc sulfide show the presence of zinc, sulphur, copper and oxygen in the system. Further high resolution scan of Cu 2p, S 2p and Zn 2p in the respective regions are shown in the Fig. 3b–d. The Fig. 3b represents the XPS high resolution scan of Cu 2p in ZnS–Cu₂S(5) with peaks positions at 951.7 eV (Cu 2p_1/2) and 931.8 eV (Cu 2p_3/2), the binding energies of the samples reveal that copper is present as cuprous sulfide (Cu₂S) form which is in agreement with earlier reports.^23,24 Weak shake-up lines of copper are also seen with ZnS–Cu₂S(5) samples Fig. 3c. The shake-up line shows minute quantities of paramagnetic species but EPR spectrum of the samples does not show signal for the presence of paramagnetic copper. This is in concurrence with the earlier reports on the absence of EPR signal for Cu₂S species. Wang et al.,²³ observed a lack of EPR signal for such samples calcined in oxygen atmosphere upto 300 °C but the EPR signals due to Cu²⁺ appeared for samples calcined at higher temperatures. Accordingly, the high resolution scan of the XPS spectrum of our samples of copper in ZnS–Cu₂S(5) resembles that of Wang et al.²³ measured for their samples calcined below a temperature of 300 °C. The XPS spectra confirm the presence of copper as Cu(I) from its binding energy. The binding energy value of zinc 2p of ZnS–Cu₂S(5) at 1022.3 eV which shows little shift in the binding energy value of 0.2 eV indicates the presence of few minute quantities of copper also in the interstitial position,²⁴ but the UV-DRS spectra clearly shows that there is the formation of ZnS–Cu₂S(0–5) nanocomposites.

Fig. 3 X-ray photoelectron spectroscopy (XPS) of ZnS–Cu₂S (a) survey spectrum; high resolution scan of (b) copper 2p (c) sulphur 2p and (d) zinc 2p.

Photoluminescence and photocatalytic activity of cuprous sulphide incorporated ZnS system

The photoluminescence (PL) spectroscopy is a useful tool to find the band position or band gap if it is an unknown semiconductor, also the nature of the sample like defects present in the semiconductor. The photoluminescence intensity with the defects present in the sample can be correlated with the photocatalytic activity.²⁵ PL spectra of vcopper (0 to 5%) doped zinc sulfide show a broad emission from 340 nm to 500 nm when it is excited from 325 nm. The PL peak position at ∼380 nm is due to band gap emission or band edge emission, which generally depends on the particle size which can be explained by quantum confinement.^25,26 The general characteristics of PL emission follow those of UV DRS results; there is a continuous reduction in the intensities on increasing copper content. In addition the shape of UV DRS at 1 and 5% copper addition is quite different from those of 0, 0.1, and 0.5% copper revealing the importance of Cu₂S incorporated into ZnS leading to morphological changes. Similarly the PL spectra of 0, 0.1, and 0.5% copper samples lose their intensity on the increase of copper but on further addition to 1.0 and 5.0% it drastically gets reduced with a little shift to the blue. These changes are essentially due to the incorporation of cuprous sulphide, also influenced by change in morphology and the deposition of Cu₂S on the surface. Furthermore, the addition of copper increases the efficiency of the photocatalyst since the PL intensity of the materials goes down; copper being present as Cu₂S traps the electrons and holes which are induced by photons from the light. The trapping of the generated electrons and holes leads to reduced recombination of electron and holes thereby improving the efficiency of the photocatalyst.²⁵ The incorporation of Cu₂S and its contribution to two dimensional sheet formation result in improved migratory capability of the charge carriers and creation of donor level (Fig. 4).²¹

Fig. 4 Photoluminescence spectra of ZnS–Cu₂S with different Cu₂S concentration.

Morphology and lattice evaluation from SEM and HRTEM measurements

The SEM EDAX images of the ZnS–Cu₂S(1) and ZnS–Cu₂S(5) are shown in the Fig. 6a–c. The role of morphology in photocatalytic activity has been indicated by many other researchers,¹ like nanospheres of Zn_xCd_1−xS–Cu₂S which shows a quantum yield of 15.7% at 420 nm and 19.1% with platinum^27,28 whereas in our present study the morphology of the samples clearly indicates the formation of the nanoflake (Fig. 5) like structure with the addition of copper in ZnS. The formation of cuprous sulphide incorporated into the zinc sulphide is also complimented by XRD measurements. The presence of Cu₂S in 1% incorporated sample can be seen as dots over the edges of the flakes and it creates a thorny structure to flakes which appears to be controlled growth of the Cu₂S over zinc sulphide due to the addition of surfactant SLS which controls the nucleation and growth. When the concentration of copper was further increased to 5% the nano flake like structure gets smoothened over the edges. The reason for the edges getting smoothened is due to the difference in the concentration of the copper to the surfactant ratio which impairs the nucleation and growth rate.

Fig. 5 SEM images of (a) pure ZnS (b) ZnS–Cu₂S(0.5) (c) ZnS–Cu₂S(1) (d) ZnS–Cu₂S(5) (e) ZnS–Cu₂S(7.5) and (f) ZnS–Cu₂S(10).

Fig. 6 SEM-EDAX images of (a) pure ZnS (b) ZnS–Cu₂S(1) and (c) ZnS–Cu₂S(5) showing copper concentrations.

The HRTEM measurements of the pristine and 5% incorporated zinc sulphide are shown in Fig. 7. The high resolution TEM image of 5 nm also indicates the single strand (Fig. 7d) of the nanoflakes with lattice fringes measuring 3.2 Å which corresponds to the zinc sulphide plane of (002) width. The phase of ZnS (110) could not be observed in the present scanning position of the sample. In general, HRTEM shows the zinc sulphide system with cuprous sulphide as flakes-like morphology; the incorporation of Cu₂S leads to the growth of ZnS nano crystals along some orientations like (100), (110) and (112) planes. The Selected Area Electron Diffraction pattern (SAED) of the 1% copper sulphide incorporated samples show ring patterns which is typical of polycrystalline material, whereas in 5% Cu₂S loaded samples we observe both dots and ring pattern which indicate it is a mixture of polycrystalline and crystalline samples. The increased crystallinity of the 5% Cu₂S incorporated samples as proved by SAED pattern is the reason for the higher photocatalytic activity of the sample.²⁹

Fig. 7 HRTEM images of pure ZnS nanorods shown as lattice fringes in (a) and as rods in (b) and of ZnS–Cu₂S(5) in lattice fringes (d) and flakes in (e); the figures (c) and (f) represent the respective SAED patterns for pure ZnS and for ZnS–Cu₂S(5).

Cu₂S concentration enhances the photocurrent revealing the potential of ZnS–Cu₂S material for photocatalysis

Photoelectrochemical characterizations are carried out for all five samples pure ZnS to ZnS–Cu₂S(0.1–5) as shown in Fig. 8. The measure of photocurrent intensity is a direct evidence for enhanced charge separation and increased migratory capability of charge carriers.^30,31 This in turn gets reflected in the enhanced photocatalytic hydrogen evolution. There is quite an appreciable shift in the intensity of the photocurrent with application of visible light. When the light is switched on the intensity increases drastically and stabilizes until the light is turned off (blocking the light source). The intensity goes down immediately in the absence of light. The first increase in the intensity of photocurrent upon light switching on is attributed to the effective separation of electrons and holes in the ZnS surface and copper presence as Cu₂S influences the charge mobility which does not allow electron and hole recombination. This leads to the increase in the photocurrent intensity (see Fig. 8a) as a function of copper concentration. The ZnS–Cu₂S(5) shows the highest intensity and also the maximum photocatalytic hydrogen evolution (vide infra). The enhanced photocatalytic activity, viz. the photocurrent is due to effective prevention of electron and hole recombination in the ZnS–Cu photocatalyst which is complimented by the PL data,²⁵ i.e. the increase in photocurrent is directly related to the decrease in the PL intensity. Again in accordance with the optical and microscopy results the effect of the transition from nanorods to nanoflakes is felt by a noisy photocurrent behavior at 0.5%. Sudden increase in photo current at and above 1.0% clearly represents a case of increased photon uptake due to the 2-dimensional sheet structure of the latter.

Fig. 8 (a) Photocurrent of pure ZnS, and ZnS–Cu₂S(0.1 to 5) showing the influence of Cu₂S concentration on the enhancement of photocurrent with visible light irradiation. (b) Nyquist plots of pure ZnS and ZnS–Cu₂S(0.1 to 5) and (c) amplified version which indicates the photocatalytic efficiency with copper concentration increase.

In addition to photocurrent measurements the electrochemical impedance spectroscopy and Nyquist plots are done for pure ZnS and ZnS–Cu₂S (0.1 to 5%) samples. The resistances measured are from the electrode used since the electrolyte and electrodes are alike then the high frequency corresponds to the charge transfer limiting process which is due to double layer capacitance with charge transfer resistance at the contact of the interface between the electrode and the electrolyte solution.^32–34 The introduction of copper(I) in ZnS decreases the arc indicating that the formation of Cu₂S influences the interfacial charge transfer which is complimented by photocurrent and the photoluminescence studies. The PL intensity reduction is the also the direct evidence for the reduced electron and hole recombination and the corresponding increase in photocatalytic activity.

Role of BET surface area and photocatalytic activity

The BET surface analysis of the samples ZnS–Cu₂S(0), ZnS–Cu₂S(1) and ZnS–Cu₂S(5) are measured and given in Table 1. The surface area of the samples increases with increase in the copper loading. The 1% copper incorporated ZnS sample shows the highest value of 28 m² g⁻¹ and 5% sample show the least in the surface area, since the formation of flakes due to deposition of Cu₂S on the surface of the zinc sulphide. The increase in loading of the Cu₂S utilises the surface area to some extent and this results in the dip in the surface area of the material. Though there is a dip in surface area of the material the photocatalytic hydrogen evolution of all the samples show a steady increase on increasing copper loading. The increase in copper loading leads to increased formation of Cu₂S as revealed by XPS and complimented by absence of signal in EPR indicates Cu₂S formed prevents the recombination of the e⁻/h⁺ pairs formed during the excitation of the photocatalyst upon shining the visible light from the lamp. The interfacial charge transfer kinetics and the maintenance of surface area even after loading of copper upto 5% are reasons for increased hydrogen production (Fig. 9).

Fig. 9 Comparison of visible light photocatalytic activity of ZnS with consequent evolution of hydrogen with different % of Cu₂S doping.

Morphological changes in ZnS–Cu₂S nanocomposites related to photocatalytic hydrogen evolution

There is no appreciable hydrogen evolution in the case of pure ZnS prepared by the same method whereas the cuprous ion incorporated ZnS (maximum of 5% copper) produce hydrogen to the peak of 2232 μmol h⁻¹ g⁻¹. The copper loading effects the hydrogen evolution rate, the 0.1% of copper loading produces 1488 μmol h⁻¹ g⁻¹ compared to pure ZnS which is 20 μmol h⁻¹ g⁻¹. There is a significant change in the hydrogen produced in higher 5% Cu₂S incorporation since the copper impurity doping shifts the absorption band to the visible region facilitating more absorption of light in that region. There could be several other reasons for increased hydrogen production by copper doping: the copper in its impurity level creates donor levels in the forbidden band structure of ZnS. A mechanism for H₂ evolution is given in Scheme 1 which is different from those proposed for multicomponent heterostructures.

Scheme 1 Proposed scheme of ZnS–Cu₂S nanoflakes for photocatalytic hydrogen evolution.

The high conduction band potential of ZnS makes it a very good photocatalyst without co-catalyst like platinum or other noble metals. The increased photocatalytic activity can also be attributed to the formation of cuprous sulfide (Cu₂S) as confirmed by XPS and complimented by the XRD where the Cu₂S acts as an electron transfer agent; also the nanoflake like structure possessing a larger surface area further promotes electron transfer.^12,15,35 Further addition of copper (0.5, 1 and 5%) increased the hydrogen evolution rate progressively from 1785 μmol h⁻¹ g⁻¹, 2083 μmol h⁻¹ g⁻¹ to 2232 μmol h⁻¹ g⁻¹ respectively. This again suggests that the presence of copper incorporation in the form of Cu₂S is responsible for the photo catalytic activity. The morphology of 1 and 5% copper loaded samples from HR-SEM reveals the presence of nanoflake like structures with a thickness of the order of 1–5 nm which can significantly contribute to the charge transfer in the interface as also proved by Y. B. Wang et al.³⁶ The surface composition of copper in the zinc sulfide sample is further confirmed by the ICP measured which are given in the Table 1. It is surprising, however, to note that the change in the quantity of hydrogen production on increasing the copper concentration from 0.5% to 5.0% is not commensurate with the corresponding increase photocurrent measurement for the same concentration change, i.e. the increase in the former is only ∼25% while in the latter it is ∼36%. It probably reflects on the reduced quantum yield during the production of hydrogen.

Very often structural integrity suffers at high concentration of incorporated component. So it may be necessary to check for structural continuity (and hence continued increase in hydrogen evolution) on higher loading with copper (Fig. 10). In order to observe the effect of increase in concentration above 5% we have synthesized the 10% Cu₂S loaded sample and checked for photocatalytic hydrogen evolution in which we could observe only 1750 μmol h⁻¹ g⁻¹. The amount of hydrogen evolved is less compared to the 0.5% copper loaded samples. This may be due to the fact that copper present on the surface is less (Fig. 10c) and also the morphology has changed to a large extent as can be seen in Fig. 10a and b from that of 1 and 5% samples. Though the XRD pattern (Fig. 10d) is similar to that of 5% samples with the peaks of the phases (100) and (110) with increased intensity but the morphology of the 10% sample shows less of flakes; also the surface composition of the copper is getting reduced so it makes the photocatalyst less active as compared to the other samples. So the optimum concentration of copper incorporation for photocatalytic hydrogen evolution is 5%. It is also evident from the fact that there is only small increase from 1 to 5% indicating that 5% is the optimum concentration.

Fig. 10 (a) TEM image of ZnS–Cu₂S(10) (b) show the SEM images of ZnS–Cu₂S(10) and (c) shows the corresponding EDAX showing copper peaks. (d) XRD pattern of ZnS–Cu₂S.

In this context it is necessary to compare our results with those of earlier work. Broadly speaking, majority of the three component heterostructures as photocatalyst with a few 2-component structures^37–39 containing the expensive Pt metal as one of its component give a hydrogen yield ranging from 4.2 mmol h⁻¹ g⁻¹ to 5.6 mmol h⁻¹ g⁻¹, largest by the use of CdS–graphene–Pt. In addition, all these structures contain toxic CdS or CdSe. Some other heterostructures without Pt^40,41 give an yield in the region of 1.8 mmol h⁻¹ g⁻¹ but even these have the toxic CdS. Ours is probably the only nontoxic photocatalyst material and that too with no expensive third component, namely, Pt metal though the yield at this moment is only 2.23 mmol h⁻¹ g⁻¹. A more recent work by Zhu et al.¹³ is worth mentioning because of its similarity to our system containing ZnS except for the fact that they have used a 3-component heterostructured material in the form of ZnS/graphene/MoS₂ nanocomposite which reached a H₂-evolution rate of 2258 μmol h⁻¹ g⁻¹ under a 300 W Xe lamp irradiation. Despite the use of 3-component system and higher power irradiation their success is limited when compared to our production of 2232 μmol h⁻¹ g⁻¹ using only a simple less expensive two component material under 150 Watts metal halide lamp. However, we have listed in Table 2 a comparative account of the photocatalysts using copper as a second or third component which clearly shows that our photocatalyst gives the best yield, the exception being that of Zhang and coworkers¹⁵ all without the use of expensive Pt metal as cocatalyst.

Table 2 A comparison of hydrogen production from copper containing heterostructures

Sample	Morphology	Mass (g)	Aqueous reaction solution	Activity/co-catalyst μmol h^−1 g^−1	Reference
ZnS–Cu2S	Change from nanorods to nanoflakes	0.03	Na2SO3, Na2S	2232	Present work
ZnxCd1−xS–CuS	Nanospheres	0.3	Na2SO3, Na2S	2080/Pt-2466.7	42
CuInS2	Microsphere	0.25	Na2SO3, Na2S	59.4/Pt	43
Zn1−xCuxS	—	1	Na2SO3	370	12
Cu-doped ZnS	Shell structure particles	0.2	Na2SO3, Na2S	210	44
(CuIn)xZn2(1−x)S2	Microspheres	0.2	Na2SO3, Na2S	990.45/Ru	45
(CuIn)xZn2(1−x)S2	Particles with hexagonal phase	0.3	K2SO3, Na2S	2280/Pt	46
CuS–ZnS	Porous nanosheets	0.05	Na2SO3, Na2S	4147	15

---

Recyclability of the photo catalyst

The recyclability of the photocatalyst (Fig. 11) is studied by using the ZnS–Cu₂S(5) sample which as the highest hydrogen evolution for three cycles of experiments with adding more amounts of the electron scavengers Na₂S and Na₂SO₃ in one hour interval with repeated evacuation and refilling of the inert gas argon. The recyclability test showed there is not much lose in the activity of the materials for at least three cycles. The little loss in activity can be attributed to the copper sulfide deposited over the zinc sulfide nanoflake morphology gets deteriorated by the stirring in the reaction conditions. The sodium sulfide present in the solution during the reaction compensates any depreciation in the activity of the photocatalyst, which may occur due to corrosion of the photocatalyst. The morphology change can affect the close contact of copper sulfide with the zinc sulfide lattice which in turn affects the electron transfer process and activity of the photocatalyst.

Fig. 11 Recyclability of hydrogen evolution from ZnS–Cu₂S(5) catalyst with respect to time.

Conclusions

Surfactant and amine assisted synthesis of zinc sulfide with cupric ions leads to the formation of ZnS–Cu₂S (0–5%) as well characterized by a variety of spectroscopic techniques with morphological changes as a function of incorporated Cu₂S. These acts as good nanocatalysts for photocatalytic water splitting and the performance is good despite the use of a low power lamp for a nontoxic system with no expensive third component. In fact the photocurrent measurement of these materials suggested their photocatalytic properties.

Acknowledgements

PTM acknowledges the DST, Government of India for a research scheme (SR/S1/IC/0053/2012) and the INSA for a Senior Scientistship. RJVM acknowledges the University Grants Commission (UGC), Govt. of India., for a Basic Scientific Research (BSR) Meritorious Senior Research Fellowship (SRF) and partial support from the above DST scheme.

References

1. A. Kudo and Y. Miseki, Chem. Soc. Rev., 2009, 38, 253–278.
2. A. Kudo, Pure Appl. Chem., 2007, 79, 1917–1927.
3. C. Xing, Y. Zhang, W. Yan and L. Guo, Int. J. Hydrogen Energy, 2006, 31(14), 2018–2024.
4. N. Bao, L. Shen, T. Takata and K. Domen, Chem. Mater., 2008, 20, 110–117.
5. N. Bao, L. Shen, T. Takata, K. Domen, A. Gupta, K. Yanagisawa and C. A. Grimes, J. Phys. Chem. C, 2007, 111, 17527–17534.
6. Y. Nosaka, Y. Ishizuka and H. Miyama, Ber. Bunsen-Ges. Phys. Chem., 1986, 90, 1199.
7. J. R. Darwent and G. Porter, J. Chem. Soc., Chem. Commun., 1981, 145–146.
8. X. Zong, H. J. Yan, G. P. Wu, G. J. Ma, F. Y. Wen, L. Wang and C. Li, J. Am. Chem. Soc., 2008, 130, 7176–7177.
9. M. Kirni, L. Yuliati and M. Shamsuddin, J. Photochem. Photobiol., A, 2012, 230, 15–22.
10. X. H. Zhang, D. W. Jing and L. J. Guo, Int. J. Hydrogen Energy, 2010, 35, 7051–7057.
11. K. Zhang and L. Guo, Catal. Sci. Technol., 2013, 3, 1672–1690.
12. A. Kudo and M. Sekizawa, Catal. Lett., 1999, 58, 241–243.
13. B. Zhu, B. Lin, Y. Zhou, P. Sun, Q. Yao, Y. Chen and B. Gao, J. Mater. Chem. A, 2014, 2, 3819–3827.
14. M. Niederberger and N. Pinna, Metal Oxide Nanoparticles in Organic Solvents Engineering Materials and Processes, Springer, 2009, pp. 19–52.
15. J. Zhang, J. Yu, Y. Zhang, Q. Li and J. R. Gong, Nano Lett., 2011, 11, 4774–4779.
16. B. Sambandam, N. Rajendran, M. Kanagaraj, S. Arumugam and P. T. Manoharan, J. Phys. Chem. C, 2011, 115(23), 11413–11419.
17. H. Pang, C. Wei, X. Li, G. Li, Y. Ma, S. Li, J. Chen and J. Zhang, Sci. Rep., 2014, 4.
18. L. Zhang, W. Wang, L. Zhou and H. Xu, Small, 2007, 3, 1618–1625.
19. J. Kundu and D. Pradhan, ACS Appl. Mater. Interfaces, 2014, 6, 1823–1834.
20. X. Yang, H. Xue, J. Xu, X. Huang, J. Zhang, Y.-B. Tang, T.-W. Ng, H.-L. Kwong, X.-M. Meng and C.-S. Lee, ACS Appl. Mater. Interfaces, 2014, 6, 9078–9084.
21. Y. Liu, J. Hu, T. Zhou, R. Che and J. Li, J. Mater. Chem., 2011, 21, 16621–16627.
22. C. H. Lai, M. Y. Lu and L. J. Chen, J. Mater. Chem., 2012, 22, 19–30.
23. Y. Wang, X. Ai, D. Miller, P. Rice, T. Topuria, L. Krupp, A. Kellock and Q. Song, CrystEngComm, 2012, 14, 7560–7562.
24. A. M. Diamond, L. Corbellini, K. R. Balasubramaniam, S. Chen, S. Wang, T. S. Matthews, L.-W. Wang, R. Ramesh and J. W. Ager, Phys. Status Solidi A, 2012, 209, 2101–2107.
25. K. Zhang, D. W. Jing, Q. Y. Chen and L. J. Guo, Int. J. Hydrogen Energy, 2010, 35, 2048–2057.
26. B. Sambandam, R. J. V. Michael, N. Rajendran, S. Arumugam and P. T. Manoharan, J. Nanopart. Res., 2012, 14, 1–15.
27. W. Zhang, Z. Zhong, Y. Wang and R. Xu, J. Phys. Chem. C, 2008, 112, 17635–17642.
28. K. Zhang and L. Guo, Catal. Sci. Technol., 2013, 3, 1672–1690.
29. H. Kato, K. Asakura and A. Kudo, J. Am. Chem. Soc., 2003, 125, 3082–3089.
30. J. Zhang, S. Z. Qiao, L. Qi and J. Yu, Phys. Chem. Chem. Phys., 2013, 15, 12088–12094.
31. Y. B. Xie, Adv. Funct. Mater., 2006, 16, 1823–1831.
32. J. W. Ondersma and T. W. Hamann, Langmuir, 2011, 27, 13361–13366.
33. D. Wang, D. Choi, J. Li, Z. Yang, Z. Nie, R. Kou, D. Hu, C. Wang, L. V. Saraf, J. Zhang, I. A. Aksay and J. Liu, ACS Nano, 2009, 3, 907–914.
34. S. Weng, B. Chen, L. Xie, Z. Zheng and P. Liu, J. Mater. Chem. A, 2013, 1, 3068–3075.
35. M. Muruganandham and Y. Kusumoto, J. Phys. Chem. C, 2009, 113, 16144–16150.
36. Y. B. Wang, Y. S. Wang and R. Xu, Int. J. Hydrogen Energy, 2010, 35, 5245–5253.
37. J. Yu, L. Qi and M. Jaroniec, J. Phys. Chem. C, 2010, 114(30), 13118–13125.
38. J. Yu, Y. Yu and B. Cheng, RSC Adv., 2012, 2, 11829–11835.
39. Y. Li, Y. Hu, S. Peng, G. Lu and S. Li, J. Phys. Chem. C, 2009, 113, 9352–9358.
40. J. Zhang, J. Yu, M. Jaroniec and J. R. Gong, Nano Lett., 2012, 12, 4584–4589.
41. Y. Wang, J. Wu, J. Zheng and R. Xu, Catal. Sci. Technol., 2011, 1, 940–947.
42. W. Zhang and R. Xu, Int. J. Hydrogen Energy, 2009, 34, 8495–8503.
43. L. Zheng, Y. Xu, Y. Song, C. Z. Wu, M. Zhang and Y. Xie, Inorg. Chem., 2009, 48, 4003–4009.
44. T. Arai, S. I. Senda, Y. Sato, H. Takahashi, K. Shinoda, B. Jeyadevan and K. Tohji, Chem. Mater., 2008, 20, 1997–2000.
45. Y. Q. Lei, G. H. Wang, S. Y. Song, W. Q. Fan, M. Pang, J. K. Tang and H. J. Zhang, Dalton Trans., 2010, 39, 3273–3278.
46. I. Tsuji, H. Kato, H. Kobayashi and A. Kudo, J. Phys. Chem. B, 2005, 109, 7323–7329.

---

Footnote

† Electronic supplementary information (ESI) available. See DOI: 10.1039/c5ra03621f

---