View Article Online / Journal Homepage / Table of Contents for this issue

Journal of
Materials Chemistry

Dynamic Article Links <

Cite this: J. Mater. Chem., 2011, 21, 4209

PAPER

www.rsc.org/materials

Published on 09 February 2011. Downloaded by Jiwaji University on 03/06/2013 12:23:10.

Nanoscale ZnO/CdS heterostructures with engineered interfaces for high

photocatalytic activity under solar radiation
Paromita Kundu,a Parag A. Deshpande,b Giridhar Madrasb and N. Ravishankar*a
Received 16th September 2010, Accepted 9th December 2010
DOI: 10.1039/c0jm03116j
Semiconductor based nanoscale heterostructures are promising candidates for photocatalytic and
photovoltaic applications with the sensitization of a wide bandgap semiconductor with a narrow
bandgap material being the most viable strategy to maximize the utilization of the solar spectrum. Here,
we present a simple wet chemical route to obtain nanoscale heterostructures of ZnO/CdS without using
any molecular linker. Our method involves the nucleation of a Cd-precursor on ZnO nanorods with
a subsequent sulfidation step leading to the formation of the ZnO/CdS nanoscale heterostructures.
Excellent control over the loading of CdS and the microstructure is realized by merely changing the
initial concentration of the sulfiding agent. We show that the heterostructures with the lowest CdS
loading exhibit an exceptionally high activity for the degradation of methylene blue (MB) under solar
irradiation conditions; microstructural and surface analysis reveals that the higher activity in this case is
related to the dispersion of the CdS nanoparticles on the ZnO nanorod surface and to the higher
concentration of surface hydroxyl species. Detailed analysis of the mechanism of formation of the
nanoscale heterostructures reveals that it is possible to obtain deterministic control over the nature of
the interfaces. Our synthesis method is general and applicable for other heterostructures where the
interfaces need to be engineered for optimal properties. In particular, the absence of any molecular
linker at the interface makes our method appealing for photovoltaic applications where faster rates of
electron transfer at the heterojunctions are highly desirable.

1. Introduction
Clean energy and pollutant-free water/air are among the most
important challenges that we currently face; a common solution
for these seemingly different challenges lies in designing new
materials for the maximal harvesting of solar radiation. In terms
of high activity and chemical stability, TiO2 is an excellent
photocatalyst that can remove a wide range of organic pollutants
from air and water, as summarized in various reviews.13
However, owing to its large bandgap (3.2 eV), its activity is

a
Materials Research Centre, Indian Institute of Science, Bangalore,
560012, India. E-mail: nravi@mrc.iisc.ernet.in; Fax: +91-80-2360 7316;
Tel: +91-80-2293 2566
b
Department of Chemical Engineering, Indian Institute of Science,
Bangalore, 560012, India
Electronic supplementary information (ESI) available: Figure S1:
Variation of the average solar intensity during the different periods of
the day from April 01 to April 12, 2010. Figure S2: XRD pattern
obtained from the Cd-precursor coated ZnO before sulfidation.
Figure S3. XPS core-level spectra of Zn2p in ZnO nanorods.
Figure S4. Bright Field TEM image of ZnOCdS nanohybrid, ZC-3,
showing fine particle clusters of CdS attached to ZnO nanorods.
Figure S5. (A) XRD pattern of the CdS synthesized by precipitation
method using same conditions as ZC-3 and heated to 150  C for 30
mins and (B) degradation rate of methylene blue (MB) under solar
radiation in presence of CdS (cubic phase + hexagonal phase) is similar
to ZC-3. See DOI: 10.1039/c0jm03116j

This journal is The Royal Society of Chemistry 2011

largely restricted to the ultraviolet (UV) region, which only

contributes to about 4% of the entire solar spectrum. Several
strategies have been adopted to extend the spectral response of
TiO2 to develop viable visible light-driven photocatalysts.413 The
doping of transition metals in TiO2 has had limited success.46
The addition of a second metal oxide like SiO2, ZrO2 or Al2O3
has also been used to enhance the photocatalytic activity of
TiO2.14 However, both these strategies have had only limited
success and, therefore the development of new materials is of
paramount importance.
Sensitization of a wide bandgap semiconducting material with
dyes has been used both in dye-sensitized solar cells46,1416 as well
as in the visible-light photocatalytic degradation of organics.710
Sensitization using visible-light active materials (narrow
bandgap semiconductors, for instance) is another very common
strategy.1720 Besides being photocatalysts that are both UV and
visible-light active, these hybrid materials are good candidates
for photovoltaic applications.17,21 Sensitized ZnO and TiO2 have
been extensively investigated in this regard. The sensitization
process is primarily limited by the relative positions of the
conduction bands (CB) of the wide and narrow bandgap semiconductors and also by the nature of the interfaces in the
system.22,23 While the first factor could be controlled by the
appropriate choice of material and by tuning the bandgap of the
sensitizer,24 the second factor still remains a challenge in terms of
J. Mater. Chem., 2011, 21, 42094216 | 4209

Published on 09 February 2011. Downloaded by Jiwaji University on 03/06/2013 12:23:10.

View Article Online

creating a favorable interface/heterojunction in order to facilitate

facile electron transfer. Recently, many groups have attempted
creating efficient heterojunctions for CdSTiO2 and CdS
ZnO,2529 using different methods but high photocatalytic
activity was obtained in only a few cases.13 For the ZnO/CdS
system, studies have been limited to the synthesis of hybrid
nanostructures by low temperature solution routes while the
photocatalytic behavior of the resultant composite has not been
investigated in detail.19
Here, we demonstrate a novel method for tethering narrow
bandgap semiconductor nanoparticles (CdS, in particular) to
ZnO nanorods to produce nanoscale heterostructures with
controlled interfaces that exhibit an exceptionally high photocatalytic activity under solar light. We follow the strategy of
attaching a precursor phase on the substrate of interest and its
subsequent conversion to the desired nanostructure, as has been
recently demonstrated for attaching molecular scale Au nanowires to different nanoscale substrates.30 ZnO nanorods
(500 nm length and 50100 nm diameter) have been decorated
with CdS nanoparticles using a simple wet chemical route where
heterogeneous nucleation of a Cd2+ containing precursor phase
on ZnO is the primary step. The conversion of this precursor
phase to CdS was accomplished by controlled sulfidation using
an aqueous solution of Na2S. The control over loading, size,
chemistry and distribution of the product sulfide was achieved by
merely changing the concentration of Na2S. We propose
a mechanism to describe the differences in microstructure under
the different sulfiding conditions by comparing the solubility
product (Ksp) of the possible metal sulfides in the system. The
resulting nanoscale heterostructures exhibit high activity for the
degradation of methylene blue (MB) dye under solar irradiation.
Our method for synthesis is general and is applicable for other
nanoscale heterostructures for possible applications ranging
from sensing, catalysis and photovoltaics.

2. Experimental
2.1 ZnO nanorods synthesis
29.5 g of zinc acetate dihydrate (0.13 mol) was dissolved in
125 mL methanol at 60  C. A solution of potassium hydroxide
(14.8 g, 0.23 mol) in methanol (65 mL) was added to the above
mixture under vigorous stirring. The mixture was heated to
reduce the volume to half and was transferred to a Teflon vessel.
This was sealed in a stainless steel autoclave and heated at 120  C
for 6 h. The product was washed several times with methanol and
water and dried before using.31 The surface area of ZnO rods was
found to be 15 m2 g1.
2.2 ZnO/CdS nanohybrid synthesis
45 mg of Cd(NO3)2$2H2O was dissolved in 23 mL of methanol
followed by the addition of 300 mg of ZnO nanorods. The
mixture was sonicated until the ZnO was well dispersed in the
medium. The milky white slurry thus obtained was heated to
about 100  C until the dry solid was obtained. This was then
added to a freshly prepared 10 ml aqueous solution of Na2S of
a required strength under vigorous stirring to prevent aggregation. The pH of the solution was measured in each case, before
adding the solid, using a standard pH-meter. All sulfidation
4210 | J. Mater. Chem., 2011, 21, 42094216

reactions were carried out at room temperature for typically 5

10 min. The yellow solid formed was then separated by centrifugation and washed several times with distilled water until the
pH became 7. A yellow colored colloidal solution was obtained
which contained excess CdS nanoparticles and was thus decanted
out. The yellow solid was then dried in a hot air oven maintained
at 150  C for 30 min in each case. The final yellow powder thus
obtained was subjected to all characterization and catalysis
experiments.
2.3 Characterization
Transmission electron microscopy (TEM) images were recorded
using a Tecnai T20 microscope operated at 200 kV. Energy
dispersive spectra (EDS) analysis had also been done in the same
microscope. Powder X-ray diffraction (XRD) patterns were
obtained using a Bruker D8-Advance X-ray diffractometer with

monochromatized Cu Ka radiation (l 1.5418 A)).
X-ray
photoelectron spectra (XPS) of the powder samples were recorded on a MultiLab 2000, XPS ESCA Thermo Fisher Scientific
using monochromatic Al Ka radiation. UV-Visible absorption
spectra of the nanohybrids and ZnO were recorded using a Perkin Elmer Lambda 35 operating between 1000 nm to 200 nm and
photoluminescence (PL) spectra were recorded for the solid
samples using a Perkin Elmer LS 55 luminescence spectrophotometer. BrunauerEmmettTeller (BET) surface areas of the
ZnO nanorods were determined using a Smart Sorb 92/93 surface
area analyzer by the N2 adsorptiondesorption method.
2.4 Solar photocatalytic experiments
The photocatalytic activity of ZnO nanorods and ZnO/CdS
nanohybrids was tested under solar radiation. The degradation
of methylene blue (S.D. Fine Chem, India) was used as the model
to test the solar photocatalytic activity of the compounds. 50 mL
of 50 ppm methylene blue solution in water was placed into a 100
mL beaker. 50 mg of the compound was added to the dye solution to maintain a catalyst concentration of 1 g L1. The dye
solution with the catalyst was stirred in the dark for 2 h until the
equilibrium adsorption was attained. No significant decrease in
the concentration of the dye was observed after 2 h for all of the
experiments. Experiments without solar radiation in the presence
of the catalyst, and experiments with solar radiation in the
absence of catalyst, showed no degradation of dye. The solution
was then exposed to sunlight with a perfectly transparent glass
plate to cover the top. The beaker was kept in an ice-cold water
bath to ensure that the temperature was maintained and the
degradation was the result of photocatalytic activity only.
Samples were taken at regular intervals and centrifuged to
remove the catalyst particles. The clear supernatant solution was
taken and analyzed in a UV-vis spectrophotometer (Shimadzu,
1700 UV-VIS). The concentration of the samples was determined
from a linear calibration plot obtained from the absorbance of
standard dye solutions at 663 nm. Experiments in the presence of
the commercial Degussa P-25 catalyst were also carried out. The
intensity of the solar radiation was measured on all experimental
days, to ensure that the reactions were carried out at similar
conditions. The average solar intensity for twelve consecutive
days can be found in the supplementary data Fig. S1.
This journal is The Royal Society of Chemistry 2011

View Article Online

Published on 09 February 2011. Downloaded by Jiwaji University on 03/06/2013 12:23:10.

3. Results and discussion

A schematic illustration of the method of synthesis is shown in
Fig. 1 where the first step, viz., the coating of the precursor phase
on the ZnO nanorod substrate, is achieved by rapidly evaporating a cadmium nitrate solution containing the ZnO nanorods.
Powder XRD patterns indicate that the precursor phase could be
a mixture of Cd(NO3)2 (JCPDS # 70-0155) and Cd(OH)2
(JCPDS # 84-1767) along with ZnO (supplementary data,
Fig. S2). This precursor phase is then converted into the sulfide
phase by controlled sulfidation using an aqueous solution of
Na2S. We demonstrate here that the initial concentration of the
aqueous sulfiding agent plays a key role in determining the size,
morphology and the composition of the final product on the ZnO
substrate. This control is a result of two competing processes viz.,
dissolution of the precursor phase in the aqueous solution and
the conversion of the precursor phase to the final sulfide. The
relative rates of these two processes are altered by changing the
concentration of the sulfiding agent. At low initial concentrations
of Na2S, the rate of sulfiding is much slower than the rate of
dissolution and, therefore, most of the precursor film dissolves
before it is converted into the sulfide phase. At the other extreme
of a very high concentration of Na2S, most of the film is converted into the sulfide phase as the rate of conversion is much
faster than the rate of dissolution in this case; this primarily
occurs as the sulfide phase that forms is insoluble in the aqueous
medium. At an optimum concentration of Na2S, we obtain fine
CdS nanoparticles on the substrate. The net loading of the sulfide
phase on the nanorod can thus be changed with the initial
concentration of Na2S, as summarized in Table 1. The estimation
of Cd content in the resultant hybrids was carried out using the
atomic absorption spectroscopy (AAS) method, where a standard solution of cadmium sulfate was used as a reference. All the
hybrids were heated in a hot air oven, maintained at 150  C for
30 min to improve the crystallinity of the CdS and the quality of
the ZnO/CdS interfaces. The minimum loading that we obtained
using this strategy was 4.1% (in the sample designated ZC-1). We
note that lowering the concentration of the sulfiding agent
accelerates the dissolution of the precursor and thus there is
a minimum loading that can be obtained using this method.

Fig. 1 Schematic illustrating a general strategy for synthesizing nanoscale heterostructures consisting of metal sulfide nanoparticles attached
to a wide bandgap semiconductor matrix involving the attachment of
a precursor phase to the matrix followed by controlled sulfidation.

This journal is The Royal Society of Chemistry 2011

Table 1 Different loading of CdS (wt.% of ZnO) obtained for different

Na2S concentrations
Sample

Na2S concentration

CdS loading (wt.%)

ZC-1
ZC-2
ZC-3
ZC-4

0.025 M
0.05 M
0.1 M
1M

4.1
5.4
8.5
11.7

3.1 Microstructural characterization

Fig. 2 shows bright-field TEM images of the hybrid nanostructures with different loadings of CdS particles. At lower
loading, the particles are well separated (Fig. 2A and 2B) while
a nearly continuous film-like morphology is seen at higher
loading (Fig. 2C). The powder XRD pattern from the sample
(after heat treatment) with higher loading (Fig. 2D) clearly shows
the presence of the wurtzite phase of CdS. While the cubic phase
is more stable at room temperature, the hexagonal phase in this
case presumably forms due to the influence of the ZnO nanorod
surface on which it forms. Fig. 3A is a bright-field image from the
sample with a Cd loading of 8.5% (denoted ZC-3 in Table 1)
showing the presence of fine nanoparticles on the nanorods. The
corresponding dark field image (Fig. 3B) from a diffraction ring
corresponding to CdS reveals the particles and the distribution
more clearly and shows that the size of the particles is less than 5
nm. EDS analysis (from sample ZC-2, Fig. 3C) indicates the
presence of Cd and S on the ZnO and the atomic percentage
indicates a near 1 : 1 stoichiometry of CdS. As the S/Cd ratio is
less than 1 and the Zn/O ratio is also less than 1, it indicates the
formation of a fraction of CdO. This could be due to the

Fig. 2 Bright-field TEM images of nanoscale heterostructures synthesized at room temperature using different initial concentrations of the
sulfiding agent, Na2S. (A) ZC-1, 0.025 M, (B) ZC-3, 0.1 M and (C) ZC-4,
1 M of Na2S. (D) XRD pattern from the ZC-4 sample showing the
presence of hexagonal CdS in the nanoscale composite.

J. Mater. Chem., 2011, 21, 42094216 | 4211

Published on 09 February 2011. Downloaded by Jiwaji University on 03/06/2013 12:23:10.

View Article Online

Fig. 3 (A) Bright-field and (B) corresponding dark-field TEM images of

CdS nanoparticles on ZnO nanorods. The CdS particles are 5 nm in size
with a uniform distribution as seen clearly in the dark-field image. (C)
EDS spectrum from the ZC-2 nanoscale hybrid showing the relative
atomic ratio of the components. (D) High resolution TEM image of the
heterostructure with the FFT (inset) illustrating the d-spacings corresponding to the 101 plane of CdS and the 002 plane spacing of ZnO.

Fig. 4 XPS analysis of the composite showing (A) Cd3d and (B) S2p
core-level spectra; (C) the O1s spectrum obtained from ZnO before
attaching CdS indicating the presence of two types of oxygen, OI (lattice
oxygen) and OII (OH species due to defects); (D) O1s spectrum from the
composite clearly showing the change in intensity of the OII and OI ratio
in the ZC-1 compared to the observed ratio in ZnO.

occupancy of the Zn vacancies by Cd at the interface. However,

optical properties, XRD, and XPS did not indicate the formation
of any CdO, which meant that even if it had formed, its
concentration is negligible in comparison to CdS.
A high resolution TEM image (and the corresponding FFT in
 that
the inset) shows the presence of lattice spacing of 3.16 A
matches with the 101 planes of the wurtzite phase of CdS
(Fig. 3D).
Core level X-ray photoelectron spectra of the converted
samples show the presence of peaks at 404.5 eV and 411.2 eV
corresponding to Cd2+ (Fig. 4A) and at 161.25 eV corresponding
to S2 (Fig. 4B), further confirming the formation of CdS. The
absence of an N1s peak at 406 eV indicated the complete
conversion of the cadmium nitrate precursor to sulfide. The
binding energies for Zn2p in the nanoscale hybrids at 1021 eV
and 1044 eV match with Zn2p binding energies from pristine
nanorods (supplementary data, Fig. S3). Fig. 4C shows the
spectrum obtained for O1s in the case of ZnO, which could be
deconvoluted into two individual peaks; the peak at 529.6 eV is
due to the O2-type oxygen species (OI) present in the ZnO
nanocrystal lattice, while the other peak at 531.8 eV is due to the
hydroxyl-type oxygen species (OII) primarily occupying the
surface of the nanorods and contributing to the defect states.32
These two types of oxygen peaks are present in the nanoscale
hybrids as well; however the intensity ratio of these two peaks
was found to vary with the CdS loading in the samples. The O1s
core-level spectrum in Fig. 4D corresponds to the sample with
the lowest loading of CdS and shows a significant increase in the
relative OII : OI intensity ratio in comparison to pristine ZnO
nanorods. The intensity ratio of the OII : OI peaks for all the

samples are presented in Table 2, which shows that this ratio is

higher in all the nanoscale hybrids as compared to ZnO but
decreases with an increase in CdS loading. This variation of
OII/OI is possible during the sulfidation step and is illustrated
schematically in Fig. 5. The difference in the OII : OI ratio
influences the catalytic activity of the nanohybrids, as discussed
later.

4212 | J. Mater. Chem., 2011, 21, 42094216

3.2 Optical property

Fig. 6 illustrates the UV-Visible absorption spectra for ZnO and
all the nanoscale hybrids with different loading of CdS. For ZC-1
and ZC-2 (where the loadings were 4.1% and 5.4%, respectively),
the band edges at 523 nm (2.36 eV) and 520 nm (2.38 eV),
respectively, are consistent with the bandgap of the bulk
hexagonal phase of CdS,33 and indicate the ability of these
hybrids to harvest the visible component of solar radiation. We
note that, in spite of the presence of CdS nanoparticles of size
5 nm, the quantization effect is not prominent in these samples,
presumably due to the presence of closely linked particles or
clusters of CdS (supplementary data, Fig. S4).34 A considerable
blue-shift in the CdS band edge was observed in the case of the
nanohybrids with higher loadings. For ZC-3 and ZC-4 (where
the loadings were 8.5% and 11.7% respectively), the band edges
were observed at 496 nm (2.49 eV) and 489 nm (2.53 eV),
respectively, as shown in Fig. 6. The size of the CdS particles is
certainly not smaller in ZC-3 and ZC-4 compared to the samples
with lower loading, and thus the observed increase in the
bandgap cannot be attributed to quantum confinement effects.
We attribute the increase in bandgap to the formation of
This journal is The Royal Society of Chemistry 2011

View Article Online

Table 2 XPS peak positions of O1s and intensity ratios of the two types of oxygen species in ZnO and all composites synthesized
Sample

O1s peak position (binding energy in eV)

Intensity ratio (OII : OI)

ZnO

529.6 (OI)
531.5 (OII)
530 (OI)
531.7 (OII)
530 (OI)
531.9 (OII)
530.4 (OI)
531 (OII)
530.2 (OI)
530.9 (OII)

0.72

ZC-1
ZC-2

Published on 09 February 2011. Downloaded by Jiwaji University on 03/06/2013 12:23:10.

ZC-3
ZC-4

Fig. 5 Schematic illustration showing the variation of concentration of

two types of oxygen species (OI and OII) in the nanoscale hybrids with
lower and higher CdS loading.

a Zn1xCdxS shell over the ZnO nanorods;35,36 a higher

concentration of Na2S for the synthesis of these samples favors
the formation of this phase, as described in the following section.
Apart from the optical bandgap of a photocatalyst, photoluminescence studies of composites based on solid semiconductors have been widely used to understand the surface/
interface phenomena in which electronhole pairs are generated
by absorption of a photon, i.e., a primary process in photocatalysis.37,38 The photoluminescence spectra of the hybrids were
recorded at room temperature with an excitation wavelength of
325 nm and 500 nm as shown in Fig. 7 to further investigate its
optical characteristics. Fig. 7A shows the PL spectra of the ZnO
nanorods and the hybrids with increasing CdS loading. We
studied the green emission zone for all samples, which is
important for understanding the photocatalytic behavior of the
nanohybrids. It was observed that the PL intensity of the broad
peak between 500550 nm, which corresponds to the green
yellow emission, was highest for ZC-1 and lowest for the ZnO
nanorods. The hybrids showed a decrease in intensity with an
increase in the loading of CdS. This could be attributed to the
decrease in the hydroxyl species on the ZnO surface (OII-type
oxygen) due to an increase in the content of CdS in the hybrids
This journal is The Royal Society of Chemistry 2011

2.07
1.46
0.91
0.96

Fig. 6 UV-visible absorption spectra obtained for all the composites

and ZnO used to compute the bandgap of the materials. The band edge of
CdS varies for different loading of CdS in the composites as 523 nm for
ZC-1 (C), 520 nm for ZC-2 (O), 496 nm for ZC-3 (+) and 489 nm for
ZC-4 (;). The band edge of ZnO was obtained at 383 nm (dotted line)
which shifted to 380 nm in case of ZC-1, ZC-2 and ZC-3 whereas no shift
was observed for ZC-4. All the band edges were obtained from the
differential plot of the corresponding absorption spectra.

Fig. 7 (A) Photoluminescence (PL) spectra of the bare ZnO nanorods

and the ZnO/CdS hybrids showing a broader greenyellow emission
spectra for ZC-1, ZC-2 and ZC-3 compared to the ZnO nanorods. The
PL intensity of the hybrids is higher for all the composites compared to
ZnO, and decreases in the order of increasing CdS loading. The spectra
were recorded at room temperature with a 325 nm excitation wavelength.
(B) PL spectra of ZC-1, ZC-2, ZC-3 and ZC-4 with excitation of the
hybrids at 500 nm at room temperature. The PL intensity of CdS attached
to ZnO nanorods is increased from ZC-1 to ZC-2.

J. Mater. Chem., 2011, 21, 42094216 | 4213

Published on 09 February 2011. Downloaded by Jiwaji University on 03/06/2013 12:23:10.

View Article Online

and is consistent with other studies on the effect of OH species on

the ZnO surface and incorporation of surface oxygen vacancies.39,40 These results are also consistent with the O1s core level
XPS spectra obtained for the hybrids where a decrease in peak
intensity, due to OII-type oxygen, and an increase in CdS loading
is observed. Previously, Ayyub et al. had reported a detailed
study on the enhancement of PL intensity in the green emission
zone for ZnOCdS thin film composites where the particle size
was in the order of 23 nm, compared to the individual ZnO and
CdS of the same particle size.41 In this study, we also observed
a similar trend in the photoluminescence quenching for the
hybrids which is indicative of the interaction between the ZnO
and the CdS nanoparticles in the hybrids. The emission spectra
recorded for CdS in the nanohybrid ZC-1 and ZC-2, as given in
Fig. 7B, clearly shows the effect of interface modification due to
an increase in sulfide concentration. It was observed that the PL
intensity of the broad peak between 650700 nm, which corresponds to the yellow emission from CdS nanoparticles, increases
from ZC-1 to ZC-2 indicating that the quenching of the emission
is higher at lower sulfide concentrations. However, this trend is
not followed for ZC-3 and ZC-4. The quenching effect shows the
order ZC-4 > ZC-3 > ZC-2 which could be rationalized by the
fact that the formation of Zn1xCdxS enhances the electron
transport from CdS to the transition layer. However, due to the
lower position of CB of Zn1xCdxS (assumed to be close to ZnS)
compared to ZnO, the electron gets trapped in the transition
layer instead of being transported to the ZnO.42,43 This means
that the interaction between CdS and ZnO is reduced when the
sulfide concentration is increased for making the hybrids, and
thus the process of sensitization is hindered due to the formation
of a Zn1xCdxS buffer layer, as discussed in detail later.
3.3 Mechanism of formation of ZnO/CdS nanohybrids
Based on the above observations, we propose a mechanism for
the formation of the nanohybrids under the different synthesis
conditions employed here (Fig. 8). The Cd-precursor coating is
formed on the ZnO surface by means of a forced heterogeneous
nucleation process. From the XPS observations, we conclude
that Cd2+ occupies the surface Zn vacancies,44 being loosely
anchored to the oxygen (OI). When this solid is treated with an
aqueous Na2S solution under highly basic conditions, the final
product formed depends upon the pH and the S2 ion

Fig. 8 Schematic representation of the mechanism of formation of ZnO/

CdS nanoscale structures by wet chemical route under different condition
of sulfidation. Possible interfacial reactions are indicated here.

4214 | J. Mater. Chem., 2011, 21, 42094216

concentration. At a lower concentration of Na2S the dissolution

of Cd-precursor is higher, resulting in a lower loading of CdS on
ZnO nanorods, as shown in Fig. 1. Apart from the competition
between dissolution and sulfidation of the precursor phase, other
competitive side reactions influence the final composition of the
hybrids. The various possible reactions at the interface are listed
in Fig. 8, which shows that both CdS and ZnS might form, but
considering the solubility product (Ksp) of CdS and ZnS
(sphalerite), which are 8  107 and 2  104 respectively, the
formation of ZnS would require a much higher concentration of
sulfide ions in the solution phase than CdS. Therefore, CdS
formation will be more favored than ZnS formation.
The critical concentrations of sulfide ion i.e, [S2] required for
the formation of CdS and ZnS were found to be 0.89 mM and
14.14 mM, respectively. In our case, [S2] during the sulfidation
step changes with different initial concentrations of Na2S, and
can be computed from the following chemical equations.
Na2S being a strong electrolyte, completely ionizes in an
aqueous medium as follows,
Na2S(aq) / 2Na+ + S2
Hence, y M Na2S will result in y M S2. Assuming that all the
sulfide remains in aqueous medium and is not liberated as H2S, it
accepts a proton from H2O, as shown in the following reversible
chemical reaction;
S2 + H2O 5 SH + OH
Thus, we may consider y M solution of S2 produces z M of SH
and z M of OH. Therefore, the effective [S2] in the medium will
be (y  z) M. Here, y is known from the Na2S concentration and
z is known from the pH measured for the solution at the sulfidation step. Thus the concentration of S2 for the synthesis of
different ZnO/CdS nanoscale hybrids can be determined. The
different pH and [S2] for the synthesis of the composites are
given in Table 3.
The formation of CdS or Zn1xCdxS can be explained based
on the reaction scheme given in Fig. 8 that shows the possible
reactions that can occur at the interface. At highly basic conditions (pH > 12) reactions 1, 3 and 4 are favorable while reaction 2
is hindered under basic conditions due to the effect of the OH
ion present in the system. Reactions 5 and 6 are dominant when
the [S2] is sufficiently high.
In the case of ZC-1 and ZC-2, the concentration of OH and
S2 are comparable (Table 3). Therefore, reactions 1, 3 and 4
occur with reaction 1 dominating due to presence of sulfide, while
reaction 2 attains equilibrium. This results in a ZnOCdS interface with a lower loading of CdS. Reactions 5 and 6 do not occur
due to the low [S2] and in these cases, no formation of ZnS or
Zn1xCdxS is observed.
In the case of ZC-3 and ZC-4, the concentration of OH is
much higher than that of S2 (Table 3), where reaction 1 is
favored, reaction 2 is unlikely to occur, and reactions 3 and 4
have a dominant effect resulting in Cd2+ and Zn2+ near the ZnO
surface. As the concentration of S2 is also much higher in these
cases, which is sufficient to exceed the solubility product of ZnS,
reactions 5 and 6 are highly favored. Thus, a combination of
reactions 3, 4, 5 and 6 results in the formation of a Zn1xCdxS
This journal is The Royal Society of Chemistry 2011

View Article Online

Published on 09 February 2011. Downloaded by Jiwaji University on 03/06/2013 12:23:10.

Table 3 Showing different conditions of reaction (pH and [S2]) for sulfidation and the obtained metal sulfide on ZnO
Sample

Na2S(aq) concentration

pH

[OH]

[S2]

Metal sulfide on ZnO

ZC-1
ZC-2
ZC-3
ZC-4

0.025 M
0.050 M
0.1 M
1M

12.2
12.6
12.8
13.9

15 mM
36 mM
69 mM
794 mM

10 mM
14 mM
31 mM
206 mM

CdS
CdS
CdS + Zn1xCdxS mixture
CdS on Zn1xCdxS shell

shell around ZnO nanorods as shown in the schematic in Fig. 8.

CdS thus formed on the shell, is unable to interact with the ZnO.
This also results in coverage of the free OI-type species, thus
limiting its conversion to OH (OII). Therefore, a decreasing trend
is observed in the OII : OI intensity ratio of two types of oxygen
species in the XPS data, as the loading of cadmium was
increased. The optical behavior of the hybrids, as discussed in an
earlier section, also confirms the presence of Zn1xCdxS in ZC-3
and ZC-4 and thus supports the reaction scheme that explains the
Zn1xCdxS formation.
3.4 Photocatalytic activity
The degradation of methylene blue (MB) was used for assessing
the solar photocatalytic activity of the nanoscale heterostructures. Fig. 9 shows the variation of MB concentration with
time for all the hybrids; with the performance of pristine ZnO
and Degussa P-25 also shown for comparison. It is seen that the
hybrids, particularly ZC-1 and ZC-2, show exceptionally high
photocatalytic activity under solar radiation. The ZC-3 and ZC-4
hybrids exhibit lower activity than ZC-1 and ZC-2. However, the
activity of ZC-3 is higher than that of a ZnO and P-25 TiO2
catalyst, while ZC-4 exhibits activity comparable to P-25 TiO2.
The study reveals that the activity is the highest for the
composites at a lower loading of CdS (4.1%) and decreases with
an increase in the loading. Hence the photoactivity of the hybrids
follow the order, ZC-1 > ZC-2 > ZC-3 > ZC-4. This is summarized in Table 4 according to the rate constants. Although,
a higher activity can possibly be obtained by further reducing
the loading of CdS, our synthesis method is limited and the
minimum loading obtained is 4.1%, as explained earlier. The
above order of activity can be rationalized based on the following
factors:
(a) bandgap of the sensitizer (CdS) increases from ZC-1 to
ZC-4,
(b) concentration of OH species (OII-type oxygen) decreases
from ZC-1 to ZC-4 and
(c) formation of Zn1xCdxS on the ZnO surface due to an
increase in pH and [S2] from ZC-1 to ZC-4.
It is to be noted (Fig. 6) that the order of the bandgaps of the
hybrids is ZC-4 > ZC-3 > ZC-2 > ZC-1, which suggests that CdS
in ZC-1 and ZC-2 can maximally harvest the visible component
of the solar spectrum and hence exhibit higher activities under
solar radiation. Recently, the role of hydroxyl species in the
photocatalytic activity of titania has been established. In this
study, we observe a similar effect due to the presence of hydroxyl
species in the ZnO/CdS hybrids.1,4547 The concentration of OH
species (OII-type oxygen) is at a maximum for the ZC-1
composite and decreases from ZC-1 to ZC-4 as is evident from
the XPS analysis leading to a decrease in photoactivity from
This journal is The Royal Society of Chemistry 2011

Fig. 9 The degradation rate of methylene blue (MB) under solar radiation in the presence of ZC composites with different loadings of CdS has
been compared to that of ZnO and Degussa-P25. As seen here, ZC-1
exhibits the highest photocatalytic activity, while ZC-4 has the lowest
activity.

Table 4 Photocatalytic activity (1st order rate constants, k) under solar

irradiation of all composites with different loadings compared to ZnO
and Degussa-P25
Sample

CdS loading (wt.%)

First order Rate Constant (I)

ZC-1
ZC_2
ZC-3
ZC-4
ZnO
Degussa-P25

4.1
5.4
8.5
11.7

28.3  103
20.9  103
6.7  103
3.2  103
6.2  103
3.6  103

ZC-1 to ZC-4. This was also evident from the variation in the
intensity of the greenyellow emission in the PL spectra obtained
for the ZnO nanorods and the hybrids (Fig. 7A). A higher
intensity of emission in the greenyellow zone is indicative of
a higher photoactivity and is consistent with our experimental
observations as shown schematically in Fig. 5. In addition to the
above two factors, the microstructure of the hybrids play
a significant role in controlling the photoresponse of the catalyst.
For enhanced photoresponse, a faster rate of electron transport
from the conduction band of the low bandgap semiconductor to
that of the higher bandgap semiconductor is required that
depends critically on the nature of the interface in the composite.
The PL spectra of the ZC-1 and ZC-2 samples as given in Fig. 7B
show that the emission of CdS at around 650700 nm is
quenched more in the case of ZC-1 than ZC-2. This indicates
better electron transport from the conduction band of CdS to
J. Mater. Chem., 2011, 21, 42094216 | 4215

Published on 09 February 2011. Downloaded by Jiwaji University on 03/06/2013 12:23:10.

View Article Online

that of ZnO in ZC-1 than ZC-2. This could be due to the

formation of Zn1xCdxS at the interface along with CdS as
evidenced in the previous sections also. The formation of
a Zn1xCdxS shell on the ZnO nanorods decreases the effective
ZnO/CdS interface area, thereby affecting the rate of electron
transport and the activity of the catalyst. This trend is expected
as the formation of Zn1xCdxS is higher for the ZC-3 and ZC-4
hybrids. Thus, in spite of the presence of a higher concentration
of OH species compared to ZnO, the activity of ZC-4 is lower
than that of ZnO. The photocatalytic behavior of only CdS
synthesized using the same conditions as ZC-3 was observed to
be very similar to ZC-3 (supplementary data, Fig. S5), which is
less active than ZC-1 and ZC-2. As the CdS obtained by
precipitation have mixed hexagonal and cubic phases, which was
different from the hybrids and also shows different optical
properties,33,34 it is difficult to compare their photoactivity
directly.

4. Conclusions
In summary, we have demonstrated a simple route to obtain
ZnO/CdS nanoscale heterostructures for highly active photocatalysis under solar irradiation conditions. Our protocol
involved anchoring the metal precursor phase to the ZnO and its
conversion to the metal sulfide using an appropriate sulfiding
agent. The morphology and composition of the nanohybrid at
the interface had been controlled by changing conditions of
sulfidation. Our study shows that the photoactivity of the
material can be tuned by manipulating the interface of the heterostructure. The method is general and applicable to a wide
range of similar hybrids for photocatalytic and photovoltaic
application.

Acknowledgements
We thank Department of Science and Technology (DST) for
funding through NSTI scheme. The Tecnai T20 TEM is a part of
the Advanced Facility for Microscopy and Microanalysis at IISc
and the XPS ESCA system is a part of the Institute Surface
Science Facility.

References
1 I. K. Konstantinou and T. A. Albanis, Appl. Catal., B, 2004, 49, 114.
2 S. D. Bhatkhande, G. V. Pangarkar and A. A. C. M. Beenackers, J.
Chem. Technol. Biotechnol., 2002, 77, 102116.
3 M. A. Rauf and S. S. Ashraf, Chem. Eng. J., 2009, 151, 1018.
4 N. Ye, J. Qi, Z. Qi, X. Zhang, Y. Yang, J. Liu and Y. Zhang, J. Power
Sources, 2010, 195, 58065809.
5 J. A. Mikroyannidis, M. S. Roy and G. D. Sharma, J. Power Sources,
2010, 195, 53915398.
6 V. Ganapathy, B. Karunakaran and S.-W. Rhee, J. Power Sources,
2010, 195, 51385143.
7 Q. Sun and Y. Xu, J. Phys. Chem. C, 2010, 113, 1238712394.
8 Y. Park, S.-H. Lee, S. O. Kang and W. Choi, Chem. Commun., 2010,
46, 24772479.
9 Y. Li, H. Zhang, X. Hu, X. Zhao and M. Han, J. Phys. Chem. C,
2008, 112, 1497314979.
10 K. S. Yao, D. Y. Wang, C. Y. Chang, W. Y. Ho and L. Y. Yang,
J. Nanosci. Nanotechnol., 2008, 8, 26992702.
11 H. Zhang, X. Lv, Y. Li, Y. Wang and J. Li, ACS Nano, 2009, 4, 380
386.
12 S. Rehman, R. Ullah, A. M. Butt and N. D. Gohar, J. Hazard.
Mater., 2009, 170, 560569.

4216 | J. Mater. Chem., 2011, 21, 42094216

13 L. Wu, J. C. Yu and X. Fu, J. Mol. Catal. A: Chem., 2006, 244, 25

32.
14 M. Wang, S.-J. Moon, D. Zhou, F. Le Formal, N.-L. Cevey-Ha,
R. Humphry-Baker, C. Gratzel and M. Gratzel, Adv. Funct. Mater.,
2010, 20, 18211826.
15 C. Xu, P. Shin, L. Cao and D. Gao, J. Phys. Chem. C, 2010, 114, 125
129.
16 P. Docampo, S. Guldin, M. Stefik, P. Tiwana, M. C. Orilall,
S. H
uttner, H. Sai and H. J. Snaith, Adv. Funct. Mater., 2010, 20,
17871796.
17 P. V. Kamat, J. Phys. Chem. C, 2008, 112, 1873718753.
18 X. Y. Lang, W. T. Zheng and Q. Jiang, IEEE Trans. Nanotechnol.,
2008, 7, 59.
19 S. Wei, Z. Shao, X. Lu, Y. Liu, L. Cao and Y. He, J. Environ. Sci.,
2009, 21, 991996.
20 J. Nayak, S. N. Sahu, J. Kasuya and S. Nozaki, Appl. Surf. Sci., 2008,
254, 72157218.
21 K. S. Leschkies, R. Divakar, J. Basu, E. Enache-Pommer,
J. E. Boercker, C. B. Carter, U. R. Kortshagen, D. J. Norris and
E. S. Aydil, Nano Lett., 2007, 7, 17931798.
22 K.-W. Kwon, B. H. Lee and M. Shim, Chem. Mater., 2006, 18, 6357
6363.
23 W. Shi, H. Zeng, Y. Sahoo, T. Y. Ohulchanskyy, Y. Ding,
Z. L. Wang, M. Swihart and P. N. Prasad, Nano Lett., 2006, 6,
875881.
24 N. M. Huang, C. S. Kan, P. S. Khiew and S. Radiman, J. Mater. Sci.,
2004, 39, 24112415.
25 J. Zhu, D. Yang, J. Geng, D. Chen and Z. Jiang, J. Nanopart. Res.,
2008, 10, 729736.
26 G.-S. Li, D.-Q. Zhang and J. C. Yu, Environ. Sci. Technol., 2009, 43,
70797085.
27 L. ChenSha, T. YaPing, K. BoNan, W. BinSong, Z. Feng, M. Qiang,
X. Ji, W. DaZhi and L. Ji, Sci. China, Ser. E: Technol. Sci., 2007, 50,
279289.
28 Y. Tak, S. J. Hong, J. S. Lee and K. Yong, Cryst. Growth Des., 2009,
9, 26272632.
29 P. Wu, H. Zhang, N. Du, L. Ruan and D. Yang, J. Phys. Chem. C,
2009, 113, 81478151.
30 P. Kundu, A. Halder, B. Viswanath, D. Kundu, G. Ramanath and
N. Ravishankar, J. Am. Chem. Soc., 132, 2021.
31 C. Pacholski, A. Kornowski and H. Weller, Angew. Chem., Int. Ed.,
2004, 43, 47744777.
32 J. Xu, Y. Chang, Y. Zhang, S. Ma, Y. Qu and C. Xu, Appl. Surf. Sci.,
2008, 255, 19961999.
33 D. R. T. Zahn, G. Kudlek, U. Rossow, A. Hoffmann, I. Broser and
W. Richter, Adv. Mater. Opt. Electron., 1994, 3, 1114.
34 M. Ichimura, F. Goto and E. Arai, J. Appl. Phys., 1999, 85, 7411
7417.
35 O. Raymond, H. Villavicencio, V. Petranovskii and J. M. Siqueiros,
Mater. Sci. Eng., A, 2003, 360, 202206.
36 Y. F. Nicolau and J. C. Menard, J. Appl. Electrochem., 1990, 20,
10631066.
37 K. Nagaveni, M. S. Hegde and G. Madras, J. Phys. Chem. B, 2004,
108, 2020420212.
38 R. Vinu and G. Madras, Appl. Catal., A, 2009, 366, 130140.
39 A. B. Djurisic, Y. H. Leung, K. H. Tam, Y. F. Hsu, L. Ding,
W. K. Ge, Y. C. Zhong, K. S. Wong, W. K. Chan, H. L. Tam,
K. W. Cheah, W. M. Kwok and D. L. Phillips, Nanotechnology,
2007, 18, 095702.
40 Y. Y. Tay, T. T. Tan, M. H. Liang, F. Boey and S. Li, Phys. Chem.
Chem. Phys., 2010, 12, 60086013.
41 P. Ayyub, P. Vasa, P. Taneja, R. Banerjee and B. P. Singh, J. Appl.
Phys., 2005, 97, 104310104314.
42 Y. Xu and M. A. A. Schoonen, Am. Mineral., 2000, 85, 543
556.
43 L. Wang, H. Wei, Y. Fan, X. Liu and J. Zhan, Nanoscale Res. Lett.,
2009, 4, 558564.
44 W. Zhong Lin, D. Yong and Y. Rusen, Solid State Commun., 2006,
138, 390394.
45 K. Nagaveni, M. S. Hegde, N. Ravishankar, G. N. Subbanna and
G. Madras, Langmuir, 2004, 20, 29002907.
46 K. Nagaveni, G. Sivalingam, M. S. Hegde and G. Madras, Environ.
Sci. Technol., 2004, 38, 16001604.
47 M. R. Hoffmann, S. T. Martin, W. Choi and D. W. Bahnemann,
Chem. Rev., 1995, 95, 6996.

This journal is The Royal Society of Chemistry 2011