New photocatalysts based on Cd0.3Zn0.7S and Ni(OH)2 for hydrogen production from ethanol aqueous solutions under visible light

Article abstract of DOI:10.1016/j.apcata.2018.07.002

The Cd_0.3Zn_0.7S photocatalysts modified by nickel hydroxide were prepared and characterized by different experimental techniques including X-ray diffraction, UV–vis spectroscopy, transmission electron microscopy, and X-ray photoelect

Full text of DOI:10.1016/j.apcata.2018.07.002

AppliedCatalysisA,General563(2018)170–176
Contents lists available at ScienceDirect
Applied Catalysis A, General
journal homepage: www.elsevier.com/locate/apcata
New photocatalysts based on Cd0.3Zn0.7S and Ni(OH)2 for hydrogen
production from ethanol aqueous solutions under visible light
⁎
Dina V. Markovskayaa,b, Ekaterina A. Kozlovaa,b, , Eugeniy Yu. Gerasimova,b
,
Andrey V. Bukhtiyarova, Denis V. Kozlova,b
a
Boreskov Institute of Catalysis, Novosibirsk 630090, Russia
b
Novosibirsk State University, Novosibirsk 630090, Russia
A R T I C L E I N F O
A B S T R A C T
Keywords:
The Cd0.3Zn0.7S photocatalysts modified by nickel hydroxide were prepared and characterized by different ex-
perimental techniques including X-ray diffraction, UV–vis spectroscopy, transmission electron microscopy, and
X-ray photoelectron spectroscopy. The obtained photocatalysts were tested in the photocatalytic hydrogen
evolution from ethanol aqueous solutions under visible light irradiation (λ = 450 nm). The nickel-modified
photocatalysts were found to be more active than the zinc-modified ones. The reasons of high activity of Ni-
contained photocatalysts and their transformations during the photocatalytic hydrogen production were dis-
cussed. The dependences of hydrogen production rate on ethanol and NaOH concentrations were obtained, the
highest value of the photocatalytic activity was 27 mmol h−1 g−1 for 1%Pt/10% Ni(OH)2/Cd0.3Zn0.7S (50% vol.
ethanol, 1 M NaOH), apparent quantum efficiency exceeded 90%. A kinetic model describing the experimental
data was proposed.
Photocatalytic hydrogen production
Nickel hydroxide
Cd1-xZnxS
Ethanol
1. Introduction
aqueous methanol solutions was significantly higher than for pristine
titanium dioxide [13]. The deposition of 0.5 wt % of nickel hydroxide
Photocatalytic decomposition of aqueous solutions of organic sub-
stances is known to be an attractive way for solving ecological problems
[1]. This method makes it possible to purify waste water containing
organic impurities and to produce available chemical sources such as
hydrogen. Nowadays it has been shown that many types of organic
compounds may act as the sacrificial agents, e.g., primary, secondary,
and tertiary alcohols [2–4], aldehydes and ketones [4], carboxylic acids
[4–6], organophosphorus compounds [5,7], azo-dyes [5,8] etc. A par-
ticular interest in this sphere is sparkled by organic alcohols which are
the products of the biomass conversion. H. Bahruji et al. demonstrated
that the rate of hydrogen photoproduction normalized to alcohol
quantity is approximately constant for the most of organic alcohols [4],
so one can study the features of its photocatalytic decomposition using
the simplest model compounds.
Earlier it was shown that the photocatalysts based on the solid so-
lutions of CdS and ZnS and co-catalysts (e.g., ε-Zn(OH)2) demonstrated
relatively high photocatalytic activity in the hydrogen evolution under
visible light [9–11]. Moreover, it is well known that nickel hydroxide is
an effective co-catalyst for photocatalytic hydrogen production [12].
The photocatalytic activity of titania modified by nickel hydroxide in
on the g-C3N4 surface led to the increase in the photocatalytic activity;
the hydrogen production rate was approximately equal to the rate over
platinized photocatalyst [14]. Although extensive research was per-
formed regarding the use of photocatalysts modified by nickel hydro-
xide in hydrogen evolution from organic compounds, ethanol was
practically not used as a hydrogen source and sacrificial agent. Also,
issues of stability of the co-catalyst based on nickel hydroxide remain
unexplored. Therefore, the aim of the present paper is to study the
changes in the photocatalytic activity and photostability caused by the
Ni(OH)2 deposition on the surface of Cd0.3Zn0.7S and to propose a ki-
netic model which displays the dependence of reaction rate on the
concentrations of reagents (ethanol and sodium hydroxide).
In the present work the photocatalysts Cd0.3Zn0.7S modified by
nickel hydroxide were prepared and characterized by a wide range of
the experimental techniques. These photocatalysts demonstrated higher
activities than those earlier reported for Cd1-xZnxS-based samples
[9,11].
⁎
Corresponding author at: Pr. Lavrentieva, 5, Novosibirsk, 630090, Russia.
E-mail address: kozlova@catalysis.ru (E.A. Kozlova).
https://doi.org/10.1016/j.apcata.2018.07.002
Received 19 March 2018; Received in revised form 12 June 2018; Accepted 1 July 2018
Availableonline08July2018
0926-860X/©2018ElsevierB.V.Allrightsreserved.

D.V. Markovskaya et al.
AppliedCatalysisA,General563(2018)170–176
Table 1
Properties and catalytic activities of the 1% Pt/y-Ni(OH)2/Cd0.3Zn0.7S samples. Experimental conditions: 10% (vol.) ethanol, 0.1 M NaOH, 50 mg of the photo-
catalyst, T = 20 °C.
y
Pt content, %
Pt average particle size (TEM), nm
Na content, %
SBET, m2/g
V, cm3/g
W(H2), μmol min−1
W(H2), mmol h−1 g−1
W(H2), μmol h−1 m-2
0
–
–
5.9
1.6
4.4
3.3
0.1
0.1
0.1
0.1
126
164
97
0.25
0.23
0.21
0.25
0.5
4.1
2.9
2.7
0.67
4.96
3.49
3.18
5.3
10
30
40
1.17
0.86
0.86
0.05
0.04
0.04
1.9
1.6
1.5
0.4
0.3
0.3
30.2
36.0
29.7
107
2. Experimental
2.1. Photocatalyst preparation
2.3. Measurements of the photocatalytic activity
The photocatalytic evolution of hydrogen from the aqueous solution
of ethanol was studied by the method described earlier [11]. The water
suspension of the photocatalyst suspension in water-ethanol-NaOH so-
lution was placed into a sealed reactor and illuminated with a 450-nm
LED. Before the reaction, the suspension was bubbled with argon for
30 min to remove oxygen. The ethanol content varied from 0 to 70 vol.
%, the concentration of NaOH varied from 0 to 1 M (the maximum
concentration of sodium hydroxide is limited by its solubility in aqu-
eous ethanol solutions); the catalyst concentration was 0.5 g L−1; the
temperature was 20 °C. The concentration of hydrogen was measured
with a gas chromatograph ‘Chromos GCh-1000′ equipped with a
thermal conductivity detector, a zeolite column and argon as the carrier
gas. The apparent quantum efficiency (AQE) was calculated by the
following formula:
The Cd0.3Zn0.7S photocatalyst was synthesized by the method de-
scribed in detail elsewhere [11]. The nickel-modified photocatalysts (y-
Ni(OH)2/Cd0.3Zn0.7S, where y is a weight content of Ni(OH)2) were
prepared as follows. 0.4 g of Cd0.3Zn0.7S was suspended in 10 ml of
distilled water, a certain volume of 0.1 M NiNO3 was added under
vigorous stirring, and after 5 min the stoichiometric amount of 0.1 M
NaOH was added. The mixture was stirred for 30 min, after that the
sample was collected, washed with water several times and dried at
80 °C for 4 h. The photocatalysts modified by zinc hydroxide (y-Zn
(OH)2/Cd0.3Zn0.7S, where y is a weight content of Zn(OH)2) were pre-
pared by the technique suggested in our previous work [11].
Before measuring the photocatalytic activity, all prepared samples
were modified by platinum. The technique is based on the impregnation
of the photocatalyst by chloroplatinic acid with the following reduction
by NaBH4 (2.5-fold excess). The amount of the deposited platinum is
1 wt. % [9,11] (see Table 1).
2⋅W
0
AQE =
⋅100%,
Nph
(1)
where W0 is the rate of produced hydrogen in μmol/min, and Nph is the
incident photon flux. In our experimental conditions the photon flux
was measured by a portable spectroradiometer (ILT950, ILT) and was
equal to 49 μE/min (λmax = 450 nm).
2.2. Photocatalyst characterization
The elemental analysis was carried out on an X-ray fluorescence
(XRF) spectrometer ARL ADVANT'X equipped with an X-ray tube with a
rhodium anode operated at 50 kV. The data were processed using the
QuantAS software (see Table 1). Phase composition of the catalysts was
analyzed by X-ray diffraction. The XRD patterns were recorded on a D8
Advance diffractometer (Bruker, Germany) using Cu Kα radiation.
Scanning was performed in the 2Θ range from 15° to 65° with a step of
0.05° and the acquisition time of 10 s at each point. The specific surface
area SBET was calculated with the Brunauer–Emmett–Teller (BET)
method using nitrogen adsorption isotherms measured at the liquid-
nitrogen temperature with an automatic ASAP 2400 Sorptometer (Mi-
cromeritics, USA). The HRTEM images were obtained with a JEOL JEM-
2010 transmission electron microscope operated at the accelerating
voltage of 200 kV that provided a lattice resolution of 0.14 nm. The
diffuse reflectance spectra were recorded on a Cary 100 (Varian) UV/
VIS spectrometer with an integrating sphere Labsphere DRA-CA-30I
with a standard reference USRS-99-010 in a wavelength range from 200
to 800 nm. The XPS measurements were performed using a photoelec-
tron spectrometer (SPECS Surface Nano Analysis GmbH, Germany)
equipped with PHOIBOS-150-MCD-9 hemispherical energy analyser
and X-ray source XR50 with double Al/Mg anode. The samples were
supported onto the double-sided conducting copper scotch tape.
Binding energy (BE) scale was preliminarily calibrated by the positions
of the peaks of Au4f7/2 (BE = 84.0 eV) and Cu2p3/2 (BE = 932.67 eV)
core levels. The valence band and core-level spectra were obtained
using AlKα irradiation (hν = 1486.6 eV; 200 W). The binding energy of
peaks was calibrated by the position of the C1s peak at 284.8 eV cor-
responding to the hydrocarbons presented on the samples surface.
Spectra analysis and peak fitting were performed with XPS Peak 4.1
software [15]. For the quantitative analysis the integral intensities of
the spectra were corrected by their respective atomic sensitivity factors
[16].
3. Results and discussions
3.1. Characterization of the Ni(OH)2-modified photocatalysts
The phase composition of the prepared photocatalysts was in-
vestigated by the XRD technique. Fig. 1 shows the XRD patterns of the
NiOH-y samples. The XRD patterns of the pristine Cd0.3Zn0.7S and 10-Ni
(OH)2/Cd0.3Zn0.7S photocatalysts are the same and exhibit only three
broad peaks at 2ϴ = 28°, 44°, 53° that are typical of the solid solutions
of cadmium sulfide and zinc sulfide with disordered structure [17]. The
average crystallite size was determined using a calibration graph
showing the dependence of FWHM (full width at half maximum) of the
first diffraction peak on crystallite size (see Fig. S1). The graph was
Fig. 1. XRD patterns of the Ni(OH)2-modified photocatalysts. Orange vertical
line denotes the X-ray reflections of β-Ni(OH)2.
171

D.V. Markovskaya et al.
AppliedCatalysisA,General563(2018)170–176
plotted on the base of XRD patterns simulated for sulfide crystallites
with disordered structure using the software developed earlier [18].
These calculations were performed on the base of the model of 1D-
disordered crystal. Such a model can describe stacking faults and other
planar defects and also finite size of coherently scattering domains [18].
The average crystallite sizes of the Cd0.3Zn0.7S phase are 2.1 nm for
by 300–400 nm) that overlap one another. The lattice spacing of
0.27 nm observed in Fig. 3b is corresponding to the (100) plane of the
β-Ni(OH)2 phase. Fig. 3c shows the presence of the other phase with the
lattice spacing of 0.31-0.33 nm corresponding to the (100) plane of the
Cd0.3Zn0.7S photocatalyst. The obtained data confirm the phase com-
position identified by the XRD and DRS methods. Fig. S2 shows the
distribution of Pt nanoparticle sizes in the 1% Pt/y-Ni(OH)2/
Cd0.3Zn0.7S (y = 10–40 wt%) photocatalyst. One can see that the
average metal particle size is 1.5–1.9 nm for all samples (Table 1). Note
that for the photocatalyst 1% Pt/10%Zn(OH)2/Cd0.3Zn0.7S platinum
Cd0.3Zn0.7
S and 2.0–2.3 nm for the y-Ni(OH)2/Cd0.3Zn0.7S photo-
catalysts as shown in Fig. S1. When the content of nickel hydroxide
reaches 20%, the peaks of β-Ni(OH)2 could be identified (Fig. 1). For
nickel hydroxide there are both sharp and broad peaks in XRD pattern;
it is might be due to the structural disorder in the sample. It has been
shown that the broadening of the (001) reflection occurs because of the
insertion of water molecules between the layers; turbostatic disorder
arising from the random orientation of the layers about the c-axis leads
to the broadening of (101) and (102) reflections [19]. Due to the
structural disorder the average crystallite size should be calculated
along the c-axis and in the orthogonal direction [19]. The average
crystallite sizes calculated from the diffraction peaks (001) and (101)
are 2.3 nm and 9.8 nm for 30-Ni(OH)2/Cd0.3Zn0.7S, 2.8 nm and 10.3 nm
for 60-Ni(OH)2/Cd0.3Zn0.7S, respectively.
The textural properties of the synthesized photocatalysts are sum-
marized in Table 1. Table 1 shows that the specific surface area of the
pristine Cd0.3Zn0.7S is 126 m2/g, for the Ni(OH)2-modified samples
these values are in the range of 97-164 m2/g. The nickel hydroxide
formation on the Cd0.3Zn0.7S surface leads to the increase of the surface
area; however, when the Ni(OH)2 content was higher than 30 wt. % this
value falls due to the decrease of the Cd0.3Zn0.7S nanoparticle amount
and the growth of the Ni(OH)2 particles with the bigger average particle
sizes. The pore volumes of all the photocatalysts are 0.21 – 0.25 cm3/g
as shown in Table 1. The high surface area and pore volume are ben-
eficial to the effective photocatalytic hydrogen production.
The optical properties of the prepared samples were studied by the
DRS technique. Fig. 2 shows the comparison of UV–vis diffuse re-
flectance spectra of pristine Cd0.3Zn0.7S, y-Ni(OH)2/Cd0.3Zn0.7S, and Ni
(OH)2 samples. The spectrum of Cd0.3Zn0.7S is typical of the solid so-
lution of CdS and ZnS [20] and demonstrates the presence of only one
phase in the sample. Additionally, the absorption at the 500–800 nm
regions is observed in the spectra of all Ni-contained photocatalysts.
These features can be assigned to Ni(II) d – d transitions in nickel hy-
droxide [13]. Besides, the deposition of nickel hydroxide on the
Cd0.3Zn0.7S surface redshifts the absorption edge from 450 nm (for bare
Cd0.3Zn0.7S) to 480 nm for 40-Ni(OH)2/Cd0.3Zn0.7S due to the inter-
facial charge transfer from the conduction band of the sulfide photo-
catalyst to the Ni(OH)2 phase.
particle size is equal to 1.6
0.3 nm [11].
To analyze chemical composition of the samples and to investigate
the oxidation state of nickel on the photocatalyst surface the XPS
measurements of 1%Pt/10-Ni(OH)2/Cd0.3Zn0.7
S were carried out.
Analysis of the survey photoelectron spectra (see Fig. S3) shows pre-
sence of Cd, Zn, Ni, Pt, S, O, and C on the surface. It should be men-
tioned that other elements were not detected in the survey spectra as
shown in Figs. S3–S4. The Cd3d5/2, Zn2p3/2, and S2p3/2 binding en-
ergies are equal to 405.3, 1022.3, and 161.7 eV, respectively, which is
typical of cadmium and zinc sulfides [21]. The atomic ratio of Cd to Zn
is close to 3:7, which is in good agreement with the chemical compo-
sition of the Cd0.3Zn0.7S photocatalyst (Table 2). The peak observed at
72.2 eV corresponds to metallic platinum with high dispersity as was
demonstrated in the previous work [10]. Fig. 4 shows the XPS spectra of
Ni2p of the 1%Pt/10-Ni(OH)2/Cd0.3Zn0.7S sample. The peaks centered
at 856.1 and 853.0 eV correspond to Ni2+ in Ni(OH)2 [22,23] and
metallic nickel [24], respectively. Note that the [Ni°]/[Ni2+] ratio is
very low. The presence of metallic nickel in the sample can be attrib-
uted to partial reduction of nickel hydroxide when the excess of NaBH4
was added at the stage of the platinum deposition. Thus, the XPS data
are consistent with the data obtained by the XRD, DRS, and TEM
techniques.
3.2. Photocatalytic performance over the 1%Pt/y-Ni(OH)2/Cd0.3Zn0.7
samples
S
The photocatalysts modified by nickel hydroxide were tested in the
photocatalytic hydrogen production from ethanol aqueous solutions.
Fig. 5 shows the dependence of the reaction rate on the Ni(OH)2 con-
tent. Kinetic data for 1%Pt/y-Zn(OH)2/Cd0.3Zn0.7S samples were added
for comparison because these samples were tested previously at the
same conditions and demonstrated very good results [11]. The reaction
rate grows with the increase of the co-catalyst content, reaches a
maximum, and after that declines for both hydroxides. This volcano-
type trend is typical of the deposited photocatalysts [25]. The growth of
the photocatalytic activity is caused by the increase in the interactions
at the interfaces of Cd0.3Zn0.7S and metal hydroxides. According to the
UV–vis diffuse reflectance spectra, the transfer of the photoinduced
charge carriers is observed for the 1%Pt/y-Ni(OH)2/Cd0.3Zn0.7S sam-
ples; its role in the reaction mechanism is discussed below. For the 1%
Pt/y-Zn(OH)2/Cd0.3Zn0.7S photocatalysts the grow of the zinc hydro-
The 1% Pt/10-Ni(OH)2/Cd0.3Zn0.7
S photocatalyst was also in-
vestigated by TEM, HRTEM, and XPS techniques. Fig. 3a demonstrates
that nickel hydroxide forms the elongated plates (their length estimated
xide content leads to the increase in interphases between Cd0.3Zn0.7
S
and Zn(OH)2 and the amounts of the realized heterojunctions [11]. The
optimal amount of metal hydroxides is 10 wt. % for both co-catalysts.
The fall of the hydrogen production rate with the further grow of the co-
catalyst content is assigned to its excessive amount. The redundant
quantity of the co-catalyst can enhance the light scattering and hinder
light absorption, therefore, the photocatalytic activity declines. More-
over, excessive co-catalyst nanoparticles may act as recombination
centers for photogenerated electrons and holes therefore decreasing the
photocatalytic activity. Note that the nickel-modified photocatalysts are
more active than the zinc-containing samples, thus the further experi-
ments were carried out with the use of only Ni(OH)2 co-catalyst.
It is important to define how the reaction rate and the specific
photocatalytic activity per surface area depend on the Ni(OH)2 content.
Table 1 shows that these values have a volcano-type dependence, but
Fig. 2. Diffuse reflectance spectra of pure Cd0.3Zn0.7S, Ni(OH)2, and y-Ni(OH)2/
Cd0.3Zn0.7S samples.
172

D.V. Markovskaya et al.
AppliedCatalysisA,General563(2018)170–176
Fig. 3. a)–c) TEM and HRTEM images of 10-Ni(OH)2/Cd0.3Zn0.7S (d) HRTEM image of 10-Ni(OH)2/Cd0.3Zn0.7S after 6 runs of photocatalytic hydrogen production.
Table 2
Surface composition of the 1%Pt/10-Ni(OH)2/Cd0.3Zn0.7S sample before and
after photocatalytic reaction.
Sample
[Zn]/[Cd] [Zn]/
([Cd]
[Cd]/
([Cd]
[S]/
([Cd]
[Ni]/
([Cd]
[Ni°]/
[Ni2+
]
+[Zn]) +[Zn]) +[Zn]) +[Zn])
1%Pt/10-Ni(OH)2/ 1.99
0.67
0.64
0.33
0.36
0.75
0.67
0.14
0.25
0.07
0.03
Cd0.3Zn0.7
S
1%Pt/10-Ni(OH)2/ 1.75
Cd0.3Zn0.7Sa
a
After 4 runs of photocatalytic hydrogen production.
Fig. 5. Photocatalytic hydrogen production over 1%Pt/y-Ni(OH)2/Cd0.3Zn0.7
S
and 1%Pt/y-Zn(OH)2/Cd0.3Zn0.7S samples. Experimental conditions: 10% (vol.)
ethanol, 0.1 M NaOH, 50 mg of the photocatalyst, T = 20 °C.
for the catalytic activity the maximum is observed at y = 10, while for
the specific photocatalytic activity the highest value is achieved at
y = 30. The specific surface area plays an important role in the pho-
tocatalysis. However, for the studied samples the increase in the specific
surface area is not always accompanied by the improvement in the
photocatalytic activity. Table 3 shows that this trend is observed for
different earlier described nickel-modified photocatalysts [27,29–39].
Probably, this fact is caused by the complex nature of the realizing
photocatalytic reaction. Along with the reagent adsorption the transfer
of the photoinduced electrons is a crucial factor for the efficient pho-
tocatalytic hydrogen production. This parameter may be determined by
the area of the interface contacts in multiphase photocatalysts, the
nature of the photocatalyst and co-catalysts, the surface structure, and
the nature of sacrificial reagents [2,5].
The mechanism of the improvement in the photocatalytic activity of
the Ni(OH)2-modified photocatalysts is studied in detail. For this pur-
pose, the phase composition of fresh 1%Pt/10-Ni(OH)2/Cd0.3Zn0.7S and
1%Pt/10-Ni(OH)2/Cd0.3Zn0.7S after six and four runs of irradiation was
analyzed by the HRTEM and XPS techniques. Fig. 3d shows the for-
mation of high dispersed nickel oxide particles (their length is
Fig. 4. Ni2p core-level XPS spectra of 1%Pt/10-Ni(OH)2/Cd0.3Zn0.7S before and
after photocatalytic hydrogen production.
173

D.V. Markovskaya et al.
AppliedCatalysisA,General563(2018)170–176
Table 3
Photocatalytic activities and AQE in hydrogen evolution from aqueous solutions of organic and inorganic compounds.
№
Photocatalyst
Substrate
Light source
Cut-off filter
W(H2), mmol g−1 h−1 SBET, m2/g W(H2), μmol m−2 h-1 AQE, % Ref.
1
2
3
4
5
6
7
8
9
Ni/Ni(OH)2/CdS
NiS/CdS
NiOx/CdS
Ni(OH)2/CdS
Ni(OH)2/Cd0.2Zn0.8
Co(NO3)2/CdS
Pt/CdS/ZnO
NiS/carbon dots/CdS
CuS/NiS/TiO2
Methanol
Lactic acid
Methanol
TEOA
Xe lamp
Xe lamp
λ > 400 nm 3.74
266
–
14
–
–
[27]
[28]
[29]
[30]
[12]
[31]
[32]
[34]
[35]
[36]
[37]
[38]
[39]
[8]
Present study
λ > 420 nm 7.30
51a
8.6b
28a
29.5a
–
λ > 400 nm 5.90
λ > 420 nm 5.08
22
90
56
40
–
13
49
267
56
98
42
105
164
269
56
128
413
–
108
328
30
434
89
S
7.16
16.50
9.10
1.40
16.0
7.50
Lactic acid
–
–
–
–
–
–
Na2S, Na2SO3
Methanol
10 NiS/TiO2
12 NiS/CdS
13 ZnS:Ni/graphene
14 NiS/CdS
15 Pt/Cd1-xZnxS/Zn(OH)2
16 Pt/Ni(OH)2/Cd0.3Zn0.7
Lactic acid
Na2S, Na2SO3, NaCl
Lignin, lactic acid, KCl Xe lamp
Ethanol
Ethanol
λ > 420 nm 24.4
Hg lamp
8.69
λ > 400 nm 1.50
36
32
165
44.9b
12.4
91
LED (450 nm)
LED (450 nm)
3.40
27
S
a
AQE was measured at λ = 420 nm.
AQE was measured at λ = 400 nm.
b
approximately 5–10 nm) with lattice spacing of 0.22 and 0.24 nm cor-
responding to (012) and (101) crystal planes. Fig. 4 demonstrates that
after photocatalytic reaction nickel on the surface exists in two forms:
the metallic one and nickel hydroxide. Simultaneously the nickel con-
tent on the surface or atomic ratio [Ni]/([Cd]+[Zn]) increases from
0.14 to 0.25 as shown in Table 2. Probably, these observations may be
explained by reduction of nickel hydroxide to metallic nickel with the
simultaneous decrease in the particle size as depicted in Graphical
Abstract. The potential (−0.23 V vs. NHE) of Ni2+/Ni° is less negative
than the conduction band level of Cd0.3Zn0.7S (−0.62 V [26]), thus the
photoinduced electrons in the conduction band can transfer to Ni(OH)2
and then effectively reduce it to Ni° [27]. Highly dispersive metallic
nickel is known to have a high reactivity and oxidize in air at the stages
of the photocatalyst washing and drying to nickel hydroxide [22]
identified by the XPS method. The nanoparticles of nickel hydroxide
under conditions of the TEM experiments are likely to transform into
the oxide identified by the TEM method.
It is interesting to define the influence of the metallic nickel for-
mation on the photocatalytic hydrogen production. Fig. 6 demonstrates
the time course of hydrogen evolution during six 1.5-h photocatalytic
runs. Fig. 6 shows the increase in the reaction rate on 1%Pt/10-Ni
(OH)2/Cd0.3Zn0.7S during the first five runs. Simultaneously, nickel
hydroxide transforms to the metallic nickel, therefore, the Ni formation
on the photocatalyst surface leads to the growth of the hydrogen evo-
4. Kinetic model for the hydrogen photoproduction
Along with the co-catalyst deposition, the optimization of the ex-
perimental conditions allows reaching high efficiency of the photo-
catalytic hydrogen production. In the present work we measured the
reaction rate of hydrogen photoproduction varying the ethanol content
(0–70 vol. %) and the NaOH concentration (0–1 M).
Fig. 7 demonstrates the influence of the ethanol content on the
hydrogen production rate, where the solid line corresponds to the ap-
proximation of the experimental data by semi-empirical kinetic model
described in detail below. It could be seen that the rate of the photo-
catalytic hydrogen production grows with the increase in the ethanol
concentration and reaches a maximum at 50 vol. % of ethanol solution.
The highest values of the photocatalytic activity and the apparent
quantum efficiency are 9.6 mmol g−1 h−1 and 32.6%, respectively, and
exceeds similar values which were recently reported [10,12,28–39] as
shown in Table 3. Further increase in the ethanol concentration leads to
the decrease in the catalytic activity due to the almost complete occu-
pancy of the adsorption sites by the organic molecule as was observed
for the glycerol [40].
Earlier it was shown that the influence of the organic alcohol con-
tent on the hydrogen photoproduction could be described by Langmuir-
Hinshelwood model for the monomolecular processes (Eq. (2)) [40]:
k × KEtOH × CE0tOH
lution rate. In opposite, the photocatalytic activity of 1%Pt/Cd0.3Zn0.7
S
W =
1 + KEtOH × CE0tOH
(2)
decreases gradually as demonstrated in Fig. 6. This effect is likely
caused by the photocorrosion of Cd1-xZnxS, zinc leaching from the
structure, and the aggregation of Pt nanoparticles [10]. Thus, the Ni
(OH)2 deposition improves the photostability and the photocatalytic
activity of pristine solid solution of cadmium and zinc sulfide.
where W is the initial H2 production rate (μmol/min), k is the apparent
Fig. 7. Dependences of the rate of photocatalytic production from water-
ethanol solution on the ethanol content. Experimental conditions: photocatalyst
1%Pt/10-Ni(OH)2/Cd0.3Zn0.7S, [NaOH] = 0.1 M, catalyst amount in suspension
is 50 mg, volume of suspension is 100 mL, T = 20 °C.
Fig. 6. The concentration of hydrogen during 1.5 h irradiation runs on 1%Pt/
10-Ni(OH)2/Cd0.3Zn0.7S and 1% Pt/Cd0.3Zn0.7S.
174

D.V. Markovskaya et al.
AppliedCatalysisA,General563(2018)170–176
rate constant (μmol/min), KEtOH is the adsorption constant of ethanol
(M−1), and C°EtOH is the initial concentration of ethanol (M). The main
feature of Langmuir-Hinshelwood model is that it corresponds to ki-
netic curve with saturation-like character, i.e. the reaction rate reaches
a maximum at high reagents concentration levels.
However, the experimental data do not satisfy this rule. The de-
pendence of the reaction rate on the substrate concentration has a
maximum at 50 vol. %, and then begins to fall. This phenomenon was
observed earlier by N. Strataki et al. for ethanol [41] and T. P. Lyubina
with co-authors for glycerol [40]. We suppose that such behavior could
be explained by the assumption that the first stage of the ethanol de-
composition is a second-order reaction:
Z-CH3CH2OH + Z → Z-CH3CH2O% + Z-H%,
(I)
Fig. 8. Dependence of the reaction rate on sodium hydroxide concentration.
Experimental conditions: photocatalyst 1%Pt/10-Ni(OH)2/Cd0.3Zn0.7S, 50%
(vol.) ethanol, catalyst amount in suspension is 50 mg, volume of suspension is
100 mL, T = 20 °C.
where denotes the free surface site while Z−CH3CH2OH,
Z
Z−CH3CH2O∙, Z-H∙ denotes adsorbed CH3CH2OH, CH3CH2O∙, and H∙
species.
For this assumption to be fulfilled, it is necessary to have the ad-
sorbed ethanol molecule and free surface site which are close to each
other for the reaction (I) to be realized. Let us assume that the reaction
(I) is a rate-limiting stage; then the dependence of the hydrogen pho-
toproduction on the ethanol concentration could be approximated with
the Langmuir-Hinshelwood model for the bimolecular processes (Eq.
(3)):
grows as well as the rate of photocatalytic hydrogen production [45]. In
this way the empirical equation could be proposed for the dependence
of the hydrogen photocatalytic evolution rate on sodium hydroxide
concentration:
W = A + B × [OH−]
(4)
k × KEtOH × CE0tOH
where A and B are the apparent rate constants.
W =
,
(1 + KEtOH × CE0tOH
)
2
Fig. 9 shows that the proposed dependence allows one to describe
the kinetic dependences with high accuracy for organic alcohols such as
methanol [44] and ethanol [45]. The dependence of the hydrogen
production rate on the concentrations of sodium hydroxide and ethanol
can be described by the combined form of Eqs. (2) and (3):
(3)
Dashed and solid lines in Fig. 7 correspond to the approximation of
the experimental data by Eqs. (2) and (3), respectively. The approx-
imation parameters are given in Table 4. As shown in Fig. 7 and
Table 4, the kinetic data can be described by both models with a good
convergence but the approximation parameters have smaller calcula-
tion errors when Eq. (3) is used for the fitting. Moreover, the correlation
coefficient obtained by the fitting with the use of Langmuir-Hinshel-
wood model for bimolecular processes have higher values than when
the monomolecular approximation is used. These facts confirm that the
bimolecular approximation should be used for the mathematical de-
scription of the kinetic data.
(α + β × [OH−]) × CE0tOH
W =
(1 + KEtOH × CE0tOH
)
2
(5)
where α and β are the apparent rate constants, KEtOH is the adsorption
constant of ethanol. It should be noted that semi-empirical Eq. (5) in-
cludes such features of the hydrogen production as (a) the bimolecular
reactions occurring on the surface of the photocatalyst with the parti-
cipation of adsorbed ethanol and (b) the linear dependence of the re-
action rate on the sodium hydroxide content. The obtained equation
allows one to predict the catalytic activity values under different con-
ditions such as ethanol or NaOH concentration. This knowledge is a
crucial for the further mathematical description of the photocatalytic
processes and the design of the reactors for the photocatalytic hydrogen
production from aqueous solutions of alcohols. Besides, the obtained
kinetic model may be useful for the studying of the kinetic features of
the photocatalytic hydrogen production from other organic compounds.
Our experiments have shown that at C0(NaOH) = 1 M and C0
Fig. 8 demonstrates the influence of the NaOH concentration on the
hydrogen photoproduction rate. The solid line corresponds to the the-
oretical approximation of the experimental data with linear depen-
dence. The first point showing the reaction rate in the absence of NaOH,
which is not taken into account in the approximation due to the state of
the surface. The isoelectric points for ZnS and CdS are in the region of
3.0–8.5 [42]. For all points which are approximated by Eq. 4 pH >
12.6. Thus, the surface of the photocatalysts exists mainly in ≡Zn−OH
or ≡Zn-O− and ≡Cd−OH or ≡Cd-O− forms [43], whereas for the first
point (pH ∼ 7) the surface consists of S2− groups.
As shown in Fig. 8, the dependence of the hydrogen photoproduc-
tion rate on the NaOH concentration could be approximated by the
linear fitting with a high accuracy. One can see that the H2 production
rate increases with the increase of NaOH concentration. The amount of
hydroxyl radicals formed in the reaction between hydroxide anions and
photoinduced holes grows with the NaOH concentration. With the
growth of the hydroxide radical concentration its oxidizing ability
(ethanol) = 50 vol. % the activity of the 1%Pt/10-Ni(OH)2/Cd0.3Zn0.7
S
photocatalyst is equal to 27 mmol g−1 h−1 and the apparent quantum
efficiency is equal to 91%. These parameters exceed the recently re-
ported data as shown in Table 3. Thereby, extremely high photo-
catalytic activity can be achieved by both the directed modification of
Cd1-xZnxS structure and the optimization of experimental conditions.
Table 4
Parameters of the experimental data approximation with classical Eq. (2) and modified Eq. (3) based on Langmuir-Hinshelwood kinetic model.
Photocatalyst
Approximation by Eq. (2)
Approximation by Eq. (3)
k,
К, M−1
R2
k,
К, M−1
R2
μmol/min
μmol/min
1%Pt/10-Ni(OH)2/
12
4
0.3
0.1
0.883
34
5
0.08
0.01
0.955
Cd0.3Zn0.7
S
175

D.V. Markovskaya et al.
AppliedCatalysisA,General563(2018)170–176
[10] E.A. Kozlova, S.V. Cherepanova, D.V. Markovskaya, A.A. Saraev, E. Yu,
V.N. Gerasimov, Appl. Catal. B 183 (2016) 197–205, https://doi.org/10.1016/j.
apcatb.2015.10.042.
[11] D.V. Markovskaya, E.A. Kozlova, S.V. Cherepanova, A.A. Saraev, E. Yu,
V.N. Gerasimov, Top. Catal. 59 (2016) 1297–1304, https://doi.org/10.1007/
s11244-016-0656-1.
[12] J. Ran, J. Zhang, J. Yu, S.Z. Qiao, Chemsuschem 7 (2014) 3426–3434, https://doi.
org/10.1002/cssc.201402574.
[13] J. Yu, Y. Hai, B. Cheng, J. Phys. Chem. C 115 (2011) 4953–4958, https://doi.org/
10.1021/jp111562d.
[14] J. Yu, S. Wang, B. Cheng, Z. Lin, F. Huang, Catal. Sci. Technol. 3 (2013) 1782–1789,
https://doi.org/10.1039/C3CY20878H.
[15] http://xpspeak.software.informer.com/4.1/.
[16] J.H. Scofield, J. Electron Spectrosc. Relat. Phenom. 8 (1976) 129–137, https://doi.
org/10.1016/0368-2048(76)80015-1.
[17] S. Cherepanova, J. Struct. Chem 53 (2012) 109–132, https://doi.org/10.1134/
S0022476612070116.
[18] S.V. Cherepanova, S.V. Tsybulya, Mater. Sci. Forum 443-444 (2004) 87–90, https://
Fig. 9. Approximation of the early reported data by Eq. (4) for organic alcohols
doi.org/10.4028/www.scientific.net/MSF.443-444.87.
[19] T.N. Ramesh, P.V. Kamath, Mater. Res. Bull. 43 (2008) 3227–3233, https://doi.org/
10.1016/j.materresbull.2008.02.024.
such as methanol and ethanol.
[20] D.V. Markovskaya, S.V. Cherepanova, A.A. Saraev, E. Yu, E.A. Gerasimov, Chem.
Eng. J. 262 (2015) 146–155, https://doi.org/10.1016/j.cej.2014.09.090.
[21] C. Li, X. Yang, B. Yang, Y. Yan, Y. Qian, J. Crystal. Growth 291 (2006) 45–51,
https://doi.org/10.1016/j.jcrysgro.2006.02.050.
[22] V.V. Kaichev, D. Teschner, A.A. Saraev, S.S. Kosolobov, A.Yu. Gladky, I.P. Prosvirin,
N.A. Rudina, A.B. Ayupov, R. Blume, M. Hävecker, A. Knop-Gericke, R. Schlögl,
A.V. Latyshev, V.I. Bukhtiyarov, J. Catal. 334 (2016) 23–33, https://doi.org/10.
1016/j.jcat.2015.11.009.
5. Conclusions
The Cd0.3Zn0.7S photocatalysts modified by nickel hydroxide de-
monstrated high values of the photocatalytic activity in the reaction of
hydrogen production from aqueous ethanol solutions under visible light
irradiation (λ = 450 nm). It was shown that the nickel-modified pho-
tocatalysts were more active than the zinc-modified ones. During the
irradiation nickel hydroxide is likely to be reduced by photoinduced
electrons to the metallic form that can improve the charge separation
and the photocatalytic activity. The dependences of hydrogen produc-
tion rate on the ethanol and NaOH concentrations over 1%Pt/10% Ni
(OH)2/Cd0.3Zn0.7S were obtained; the highest value of the photo-
catalytic activity was equal to 27 mmol h−1 g−1 for (50% vol. ethanol,
1 M NaOH). A semi-empirical equation which describes the experi-
mental data with high accuracy was proposed based on the Langmuir-
Hinshelwood mechanism.
[23] T.K. Kim, K.J. Lee, J. Yuh, S.K. Kwakc, H.R. Moon, New J. Chem. 38 (2014)
1606–1610, https://doi.org/10.1039/C4NJ00067F.
[24] J.F. Moulder, W.F. Stickle, P.E. Sobol, K.D. Bomben, Handbook of X-Ray
Photoelectron Spectroscopy, Perkin-Elmer Corp., Eden Prairie (MN), 1992.
[25] K. Maeda, J. Photochem. Photobiol. C 12 (2011) 237–268, https://doi.org/10.
1016/j.jphotochemrev.2011.07.001.
[26] D. Jing, L. Guo, L. Zhao, X. Zhang, H. Liu, M. Li, Sh. Shen, G. Liu, X. Hu, X. Zhang,
K. Zhang, L. Ma, P. Guo, Int. J. Hydrogen Energy 35 (2010) 7087–7097, https://doi.
org/10.1016/j.ijhydene.2010.01.030.
[27] D.V. Markovskaya, E. Yu, E.A. Gerasimov, V.N. Kozlova, Mater. Today Proc. 4
(2017) 11331–11335, https://doi.org/10.1016/j.matpr.2017.09.003.
[28] X. Chen, S. Chen, C. Lin, Z. Jiang, W. Shangguan, Int. J. Hydrogen Energy 40 (2015)
998–1004, https://doi.org/10.1016/j.ijhydene.2014.11.058.
[29] C.-T. Dinh, M.-H. Pham, F. Kleitz, T.-O. Do, J. Mater. Chem. A 1 (2013)
13308–13313, https://doi.org/10.1039/C3TA12914D.
Acknowledgment
[30] X. Chen, W. Chen, H. Gao, Y. Yang, W. Shangguan, Appl. Catal. B 152–153 (2014)
68–72, https://doi.org/10.1016/j.apcatb.2014.01.022.
This work was conducted within the framework of the budget
project №АААА-А17-117041710087-3 for Boreskov Institute of
Catalysis.
[31] J. Ran, J. Yu, M. Jaroniec, Green Chem. 13 (2011) 2708–2713, https://doi.org/10.
1039/C1GC15465F.
[32] J. Zhou, J. Jin, X. Zhu, J. Huang, J. Yu, W.-Y. Wong, W.-K. Wong, J. Mater. Chem. A
4 (2016) 5282–5287, https://doi.org/10.1039/C6TA00325G.
[33] Y.G. Kim, W.-K. Jo, Int. J. Hydrogen Energy 42 (2017) 11356–11363, https://doi.
org/10.1016/j.ijhydene.2017.02.176.
[34] R.-B. Wei, Z.-L. Huang, G.-H. Gu, Z. Wang, L. Zeng, Y. Chen, Z.-Q. Liu, Appl. Catal. B
231 (2018) 101–107, https://doi.org/10.1016/j.apcatb.2018.03.014.
[35] Q. Wang, G. Yun, Y. Bai, N. An, Y. Chen, R. Wang, Z. Lei, W. Shangguan, Int. J.
Hydrogen Energy 39 (2014) 13421–13428, https://doi.org/10.1016/j.ijhydene.
2014.04.020.
Appendix A. Supplementary data
Supplementary material related to this article can be found, in the
online version, at doi:https://doi.org/10.1016/j.apcata.2018.07.002.
[36] J. Huang, Z. Shi, X. Dong, J. Energy Chemistry 25 (2016) 136–140, https://doi.org/
10.1016/j.jechem.2015.11.007.
References
[37] X. Zhou, H. Sun, H. Zhang, W. Tu, Int. J. Hydrogen Energy 42 (2017) 11199–11205,
https://doi.org/10.1016/j.ijhydene.2017.03.179.
[38] C.-J. Chang, K.-W. Chu, M.-H. Hsu, C.-Y. Chen, Int. J. Hydrogen Energy 40 (2015)
14498–14506, https://doi.org/10.1016/j.ijhydene.2015.05.141.
[39] C. Li, H. Wang, S.B. Naghaden, J.Z. Zhang, P. Fang, Appl. Catal. B 227 (2018)
229–239, https://doi.org/10.1016/j.apcatb.2018.01.038.
[40] T.P. Lyubina, D.V. Markovskaya, E.A. Kozlova, V.N. Parmon, Int. J. Hydrogen
Energy 38 (2013) 14172–14179, https://doi.org/10.1016/j.ijhydene.2013.08.031.
[41] N. Strataki, M. Antoniadou, V. Dracopolous, P. Kianos, Catal. Today 151 (2010)
53–57, https://doi.org/10.1016/j.cattod.2010.03.036.
[42] M. Kosmulski, Adv. Colloids Interface Sci. 152 (2009) 14–25, https://doi.org/10.
1016/j.cis.2009.08.003.
[1] K.C. Christoforidis, P. Fornasiero, ChemCatChem 9 (2017) 1523–1544, https://doi.
org/10.1002/cctc.201601659.
[2] A.V. Puga, Coordin. Chem. Rev. 315 (2016) 1–66, https://doi.org/10.1016/j.ccr.
2015.12.009.
[3] H. Bahruji, M. Bowker, P.R. Davies, L.S. Al-Mazroai, A. Dickinson, J. Greaves,
D. James, L. Millard, F. Pedrono, J. Photochem. Photobiol. A 216 (2010) 115–118,
https://doi.org/10.1016/j.jphotochem.2010.06.022.
[4] H. Bahruji, M. Bowker, P.R. Davies, F. Pedrono, Appl. Catal. B 107 (2011) 205–209,
https://doi.org/10.1016/j.apcatb.2011.07.015.
[5] E.A. Kozlova, V.N. Parmon, Russ. Chem. Rev. 86 (2017) 870–906, https://doi.org/
10.1070/RCR4739.
[43] Y. Li, D. Gao, Sh. Peng, G. Lu, Sh. Li, Int. J. Hydrogen Energy 36 (2011) 4291–4297,
https://doi.org/10.1016/j.ijhydene.2011.01.038.
[44] X. Meng, Q. Yu, T. Wang, G. Liu, K. Chang, P. Li, L. Liu, J. Ye, APL. Mater. 3 (2015),
[6] X.-J. Zheng, L.-F. Wei, Z.-H. Zhang, Q.-J. Jiang, Int. J. Hydrogen Energy 34 (2009)
9033–9041, https://doi.org/10.1016/j.ijhydene.2009.09.019.
[7] E.A. Kozlova, A.V. Vorontsov, Int. J. Hydrogen Energy 35 (2010) 7337–7343,
https://doi.org/10.1016/j.ijhydene.2010.04.122.
[8] Y.J. Wei, B. Xie, M.-B. Wei, A. Patsoura, D.I. Kondarides, X.E. Verykios, Appl. Catal.
B 64 (2006) 171–179, https://doi.org/10.1016/j.apcatb.2005.11.015.
[9] E.A. Kozlova, D.V. Markovskaya, S.V. Cherepanova, A.A. Saraev, E. Yu,
T.V. Gerasimov, V.V. Perevalov, V.N. Kaichev, Int. J. Hydrogen Energy 39 (2014)
18758–18769, https://doi.org/10.1016/j.ijhydene.2014.08.145.
https://doi.org/10.1063/1.4931763 104401-1– 104401-6.
[45] T. Simon, N. Bouchonville, M.J. Berr, A. Vaneski, A. Adrovic, D. Volbers,
R. Wyrwich, M. Doblinger, A.S. Susha, A.L. Rogach, F. Jackel, J.K. Stolarczyk,
J. Feldmann, Nat. Mater. 13 (2014) 1013–1018, https://doi.org/10.1038/
nmat4049.
176