Monometallic nanoparticles decorated and rare earth ions doped KTaO3/K2Ta2O6 photocatalysts with enhanced pollutant decomposition and improved H2 generation

Article abstract of DOI:10.1016/j.jcat.2018.05.013

New, monometallic nanoparticles (MNPs) decorated surface and rare earth (RE) ions doped lattice of perovskite-type (KTaO₃)/pyrochlore-type (K₂Ta₂O₆) photocatalysts were successfully prepared by facile hydrothermal incorporation of RE ions into KTaO₃/K₂Ta₂O₆ lattices followed by photodeposition of MNPs. The impact of noble metal type (MNPs = Au, Pt, Rh) and rare earth dopant type (RE = Er, Pr) on the physicochemical properties correlated with photoactivity of MNPs/RE-KTaO₃ and MNPs/RE-K₂Ta₂O₆ has been examined. All as-prepared photocatalysts were systematically characterized by UV–vis diffuse reflectance spectroscopy, photoluminescence emission spectroscopy, transmission electron microscopy, energy-dispersive X-ray spectroscopy, fast Fourier transform, powder X-ray diffraction, and X-ray photoelectron spectroscopy. The crystal structure was presented using VESTA software. Photocatalytic activity of samples under visible light irradiation was determined by pollutant decomposition, while UV–vis induced activity was estimated by H₂ generation efficiency. The experimental results suggested a synergistic effect between MNPs and RE-ions-modified KTaO₃/K₂Ta₂O₆, which resulted in greatly improved photocatalytic performance compared with that of pristine KTaO₃/K₂Ta₂O₆. Inclusion of RE ions into KTaO₃/K₂Ta₂O₆ lattices probably leads to formation of RE4f states below the conduction band levels to electron transition by lower photon excitation energy, while photoreduction of MNPs on the RE-KTaO₃/RE-K₂Ta₂O₆ surface causes localized surface plasmon resonance and activates surface by adsorption for both O₂ and H₂O molecules. The highest visible-light-induced photoactivity was observed for the Au/Er-KTaO₃ sample (16.48%) in phenol solution degradation and for the Pt/Pr-K₂Ta₂O₆ photocatalyst (36.13%) in gaseous toluene removal, respectively. Active species-trapping tests indicate that O₂^[rad]? radicals are meaningfully involved in phenol oxidation under visible light irradiation. The largest amount of H₂ evaluation was observed for the Pt/Pr-K₂Ta₂O₆ sample (93.16 μmol/min) under UV–vis irradiation, which also showed a relatively good lifetime in recycling. The present study describes noble- and rare-earth-metal-modified KTaO₃/K₂Ta₂O₆ as an advanced material in photocatalysis.

Full text of DOI:10.1016/j.jcat.2018.05.013

Journal of Catalysis 364 (2018) 371–381
Contents lists available at ScienceDirect
Journal of Catalysis
journal homepage: www.elsevier.com/locate/jcat
Monometallic nanoparticles decorated and rare earth ions doped
KTaO3/K2Ta2O6 photocatalysts with enhanced pollutant decomposition
and improved H2 generation
Anna Krukowska a, Grzegorz Trykowski b, Wojciech Lisowski c, Tomasz Klimczuk d,
Michal Jerzy Winiarski d, Adriana Zaleska-Medynska a,
⇑
a Department of Environmental Technology, Faculty of Chemistry, University of Gdansk, 80-308 Gdansk, Poland
b Instrumental Analyses Laboratory, Faculty of Chemistry, Nicolaus Copernicus University, 87-100 Torun, Poland
c Institute of Physical Chemistry, Polish Academy of Sciences, 01-224 Warsaw, Poland
d Department of Solid State Physics, Faculty of Applied Physics and Mathematics, Gdansk University of Technology, 80-233 Gdansk, Poland
a r t i c l e i n f o
a b s t r a c t
Article history:
New, monometallic nanoparticles (MNPs) decorated surface and rare earth (RE) ions doped lattice of
perovskite-type (KTaO3)/pyrochlore-type (K2Ta2O6) photocatalysts were successfully prepared by facile
hydrothermal incorporation of RE ions into KTaO3/K2Ta2O6 lattices followed by photodeposition of
MNPs. The impact of noble metal type (MNPs = Au, Pt, Rh) and rare earth dopant type (RE = Er, Pr) on
the physicochemical properties correlated with photoactivity of MNPs/RE-KTaO3 and MNPs/RE-
K2Ta2O6 has been examined. All as-prepared photocatalysts were systematically characterized by UV–
vis diffuse reflectance spectroscopy, photoluminescence emission spectroscopy, transmission electron
microscopy, energy-dispersive X-ray spectroscopy, fast Fourier transform, powder X-ray diffraction,
and X-ray photoelectron spectroscopy. The crystal structure was presented using VESTA software.
Photocatalytic activity of samples under visible light irradiation was determined by pollutant decompo-
sition, while UV–vis induced activity was estimated by H2 generation efficiency. The experimental results
suggested a synergistic effect between MNPs and RE-ions-modified KTaO3/K2Ta2O6, which resulted in
greatly improved photocatalytic performance compared with that of pristine KTaO3/K2Ta2O6. Inclusion
of RE ions into KTaO3/K2Ta2O6 lattices probably leads to formation of RE4f states below the conduction
band levels to electron transition by lower photon excitation energy, while photoreduction of MNPs on
the RE-KTaO3/RE-K2Ta2O6 surface causes localized surface plasmon resonance and activates surface by
adsorption for both O2 and H2O molecules. The highest visible-light-induced photoactivity was observed
for the Au/Er-KTaO3 sample (16.48%) in phenol solution degradation and for the Pt/Pr-K2Ta2O6 photocat-
alyst (36.13%) in gaseous toluene removal, respectively. Active species-trapping tests indicate that O2ÅÀ
radicals are meaningfully involved in phenol oxidation under visible light irradiation. The largest amount
of H2 evaluation was observed for the Pt/Pr-K2Ta2O6 sample (93.16 mmol/min) under UV–vis irradiation,
which also showed a relatively good lifetime in recycling. The present study describes noble- and rare-
earth-metal-modified KTaO3/K2Ta2O6 as an advanced material in photocatalysis.
Received 3 April 2018
Revised 10 May 2018
Accepted 11 May 2018
Keywords:
Perovskite KTaO3/Pyrochlore K2Ta2O6
Rare earth ions doping
Metal nanoparticles photodeposition
Vis/UV–vis photocatalytic activity
Pollutant degradation/H2 generation
Ó 2018 Elsevier Inc. All rights reserved.
1. Introduction
tantalate demonstrate photocatalytic performance due to their
special features: (i) more negative conduction band minima than
Heterogeneous photocatalysts, such as tantalate-based oxides
(mainly TaOÀ3 or Ta2O26À), are able to convert photon energy into
chemical energy through decomposing industrial pollutants as
well as through producing H2 as a clean energy carrier [1–3].
Perovskite (KTaO3) and pyrochlore (K2Ta2O6) phases of potassium
other transition metal oxides [4], (ii) relatively wide band gaps of
about 3.4 eV/4.5 eV for KTaO3/K2Ta2O6 [5,6], (iii) high ferroelectric
and dielectric properties [7], and (iv) a Ta–O–Ta angle of around
180°, improving transport and separation of electron–hole pairs
[8]. The hydrothermal/solvothermal synthesis at lower OHÀ
concentration induces pyrochlore phase formation, while higher
OHÀ concentration initiates perovskite phase preparation [9,10].
Pristine KTaO3/K2Ta2O6 are activated only by ultraviolet (UV) light
⇑
Corresponding author.
E-mail address: adriana.zaleska@ug.edu.pl (A. Zaleska-Medynska).
https://doi.org/10.1016/j.jcat.2018.05.013
0021-9517/Ó 2018 Elsevier Inc. All rights reserved.

372
A. Krukowska et al. / Journal of Catalysis 364 (2018) 371–381
irradiation (about 4% of total solar energy); therefore we have
modified potassium tantalate to enhance its photoactivity in the
visible (vis) range of the solar spectrum by doping with a rare earth
metal (RE = Y, Yb, Ho, Pr, Er) [11] or by surface photodeposition of a
noble metal (M = Au, Ag, Pd, Pt, Rh, Ru) in the form of mono-/
bimetallic nanoparticles (MNPs/BNPs) [12]. Incorporation of RE
ions at the A+ lattice site into perovskite-type (ABO3)/pyrochlore-
type (A2B2O6) structures probably causes formation of RE4f states
located on or below the conduction band edge (except La4f levels
situated over the conduction band edge). The interactions between
RE4f states and photoexcited electrons from the valence band can
result in enhanced visible-light-induced photoactivity [13]. Other-
wise, the mechanism of the visible-light-excited plasmonic metal
loaded at the semiconductor is likely based on localized surface
plasmon resonance (LSPR) relying on direct transfer of electrons
from the excited metal to the conduction band of the semiconduc-
tor due to interband transitions. The LSPR effect strongly depends
on the dielectric environment, the separation distance, and the size
and shape of metal nanoparticles (NPs) immobilized on the photo-
catalyst surface [14,15]. Noble metal nanoparticles, embedded on
the semiconductor, can also work as electron traps and active reac-
tion sites. Furthermore, modification by MNPs probably activates
surface semiconductors by adsorption of both O2 and H2O mole-
cules, resulting in more effective formation of active species such
(575 mol/gcat h and 280 mol/gcat h, respectively). In a two-step
excitation mechanism it was assumed that Pt provided an H2 evo-
lution site and KTa(Zr)O3 provided an O2 evolution site [25]. The
literature also states that perovskite-type sodium tantalate
(NaTaO3) was employed in MNPs/RE modification, such as Pt/RE:
NaTaO3 (RE = Y, La, Ce, Yb) [26], Au/NaTaO3:La [27], and RuO2/La:
NaTO3 [28] applied to photocatalytic H2O splitting. Although a
few studies have been performed on metal-modified KTaO3/
K2Ta2O6, detailed investigation of synergistic effect between MNPs
decorated surfaces and RE ions doped lattices of potassium tanta-
late has not been yet pursued. Thus, fabrication of visible-light-
induced tantalate-based photocatalysts reveals novel possibilities
for creating advanced materials in photocatalysis applications.
Here, as a continuation of previous work [11,12], two series of
new MNPs/RE-KTaO3 and MNPs/RE-K2Ta2O6 photocatalysts have
been synthesized via
a one-step hydrothermal route of RE-
KTaO3/RE-K2Ta2O6 followed by photoreduction of MNPs. For the
first time, in this study, the effect of noble metal type and rare
earth dopant type on physicochemical properties of MNPs/RE-
KTaO3 and MNPs/RE-K2Ta2O6 (e.g. absorbance and photolumines-
cence properties, morphology, crystalline structure, and surface
elemental composition) were explored and related to photoactiv-
ity. The photocatalytic performance of received samples were
examined in three model reactions i.e. decomposition of phenol
aqueous solution, removal of gaseous toluene, and amount of
evolved H2 in the presence of formic acid aqueous solution. Active
oxygen species were investigated in phenol degradation under vis-
ible light irradiation. The durability and stability of the most pho-
toactive sample for H2 generation was also assessed. According to
the results, possible photocatalytic mechanisms under visible and
UV–vis light irradiation were discussed as well. To the best of
our knowledge, complementary experimental studies including
comprehensive photocatalytic activity of selective MNPs/RE-
KTaO3 and MNPs/RE-K2Ta2O6 (MNPs = Au, Pt, Rh; RE = Er, Pr) have
not yet been reported in the literature.
Å
as OH or OÅ2À radicals [16]. Therefore, a synergistic effect between
MNPs and RE-ions-modified KTaO3/K2Ta2O6 may exhibit novel
physicochemical properties or a combination of the properties of
two distinct metals resulting in enhanced photocatalytic activity.
In recent years there has been considerable interest in potas-
sium tantalate modification by surface deposition of noble metals,
such as silver (Ag-KTaO3 [17], Ag-K2Ta2O6 [18], Ag@AgCl-K2Ta2O6
[19]) and rhodium (Rh2O3-loaded-K2Ta2O6) [20]; doping with
metals such as lanthanum (K1-xLaxTaO3) [21], calcium or barium
(Ca- or Ba-doped KTaO3) [22], copper or vanadium (KTaO3:Cu,V)
[23], or doping with other metals and semimetals (Zn-, Y-, Al-,
Ga-, In-, Ti-, Zr-, Hf-, Si-, Ge-, Nb-, Sb-, or W-doped KTaO3) [24],
as well as both noble metal surface reduction and inclusion of
metal such as platinum and zirconium (Pt/KTa(Zr)O3) [25]. For
instance, Xu et al. reported photocatalytic H2 generation in water
splitting with aqueous methanol solution under simulated sunlight
over Ag-KTaO3 nanocubes (185.60 mmol/h g) and presented the
highest photoactivity in the degradation of tetracycline under vis-
ible light irradiation by octahedral Ag-K2Ta2O6 nanocrystals
(49.59%) [17,18]. Pan et al. fabricated Ag@AgCl-K2Ta2O6 compos-
ites by a precipitation–photoreduction reaction, which exhibited
almost 100% decomposition of Rhodamine B under natural light.
Introduction of Ag@AgCl on the K2Ta2O6 surface increased separa-
tion of photogenerated electron–-hole pairs and produced active
2. Experimental
2.1. Chemicals and apparatus
Ta2O5 (>99% purity) from Sigma-Aldrich (Germany), KOH (pure
p.a.) from Chempur (Poland), C2H5OH (99.8%) and polyethylene
glycol 400 (PEG-400) from POCh S.A. (Poland) were purchased for
the synthesis of potassium tantalate. Er(NO3)3Á5H2O (99.9%) and
Pr(NO3)3Á6H2O (99.9%) from Sigma-Aldrich (Germany) were used
as rare earth sources, while KAuCl4 (98%), H2PtCl6 (99%), and RhCl3
(98%) from Sigma-Aldrich (Germany) were applied as noble metal
sources in the preparation process. C6H5OH (pure p.a.), C6H5CH3
(99.8%), and HCOOH (85%) from POCh S.A. (Poland) were used in
photocatalytic performance. (NH4)2C2O4 (99.5%), AgNO3 (99%),
C6H4O2 (98%), and (CH3)3COH (99.5%) from Sigma-Aldrich
(Germany) were applied as scavengers. N2 gas (>99.9%) was
obtained from Air Liquide. Deionized H2O was used for all reactions
and treatment processes. All the chemicals were applied as
received without further purification.
Diffuse reflectance spectra (DRS) were characterized by a
Thermo Scientific Evolution 220 UV–vis spectrophotometer
equipped with an ISA-220 integrating sphere accessory. BaSO4
was used as the reference. The absorption spectra were converted
from DRS UV–vis data by the Kubelka–Munk function to estimate
band gap. Photoluminescence (PL) emission spectra were recorded
by a Perkin-Elmer 50B luminescence spectrophotometer equipped
with a Xe lamp as excitation source and an R928 photomultiplier
as detector. All the as-prepared photocatalysts were excited
with 325 nm wavelength. The PL spectra were corrected for the
instrumental response. Microscopy analysis was received by FEI
Å
species such as OÅ2À, Cl0, and OH to improve the photocatalytic
activity [19]. Ishihara et al. indicated that Rh2O3 loading on
K2Ta2O6 demonstrated an efficient ability to split H2O into H2
(60 mmol/h g) under UV–vis light irradiation. The maximum H2 for-
mation was found for 0.25 wt% Rh2O3 [20]. Liu et al. incorporated
La3+ ions at the K+ lattice site in KTaO3 to reduce particle size,
which enhanced separation of photogenerated electron–hole pairs
[21]. Paulauskas et al. introduced Ca or Ba doping in KTaO3 to
improve solar photoactivity of KTaO3 [22]. Rossella et al. found
out that in KTaO3, copper ions were incorporated at the Ta5+ lattice
site in two different charge states, Cu2+ and Cu+, causing charge
compensation by oxygen vacancies [23]. Ishihara et al. compared
the photocatalytic performance of KTaO3 doped with different
metal ions toward H2 generation in water splitting. The reduction
in charge density showed that a trivalent metal ion dopant was
the most efficient [24]. Hagiwara et al. explored photocatalytic
H2O splitting on Pt/KTa(Zr)O3 modified with cyanocobalamin
(vitamin B12) as the most improving for H2 and O2 formation

A. Krukowska et al. / Journal of Catalysis 364 (2018) 371–381
373
Tecnai F20 X-Twin transmission electron microscopy (TEM) with
integrating energy-dispersive X-ray spectroscopy (EDS) and the
fast Fourier transform (FFT). The average value of MNPs deposited
on the RE-KTaO3/RE-K2Ta2O6 surface was calculated based on mea-
surements of 50 particles from TEM images. Powder X-ray diffrac-
tion (PXRD) patterns were collected by a Philips X’Pert Pro MPD
intervals of 30 min during 90 min of irradiation and filtered
through syringe filters (Ø = 0.2 m). Phenol concentration was
l
estimated by a colorimetric method (k = 480 nm) after derivatiza-
tion with diazo-p-nitroaniline in alkaline solution using a UV–vis
spectrophotometer. The photocatalytic degradation of phenol
was preceded by blank tests in the absence of photocatalyst or illu-
mination, which revealed no significant reduction of the pollutant.
Supplementary active species in phenol decomposition assisted
with the most photoactive sample under visible light irradiation
were detected using radical scavengers i.e. AgNO3 for electrons
(eÀ), (NH4)2C2O4 for holes (h+), C6H4O2 for superoxide radical spe-
cies (OÅ2À), and (CH3)3COH for hydroxyl radical species (ÅOH). The
experiment was examined under conditions similar to those for
phenol degradation under visible light irradiation with an addi-
tional amount of radical scavenger in the reaction system.
diffractometer with Cu K
a radiation (k = 1.5418 Å). PXRD data
were processed by LeBail analysis using High Score Plus software.
The crystal structure was presented using visualization for elec-
tronic and structural analysis (VESTA) software [29]. X-ray photo-
electron spectroscopy (XPS) measurements were performed using
a PHI 5000 VersaProbe (ULVAC-PHI) spectrometer with AlK
a radi-
ation (h = 1486.6 eV). High-resolution (HR) XPS spectra were col-
m
lected with the hemispherical analyzer at a pass energy of 23.5 eV,
an energy step size of 0.05–0.1 eV, and a photoelectron takeoff
angle of 45° with respect to the surface plane. The binding energy
(BE) scale of all detected spectra was referenced by setting the BE
of the aliphatic carbon peak (C–C) signal to 285.0 eV.
2.3.2. Gaseous toluene removal
Toluene as a model gaseous pollutant was applied in photocat-
alytic contaminant degradation under visible light irradiation. Pho-
tocatalyst (0.2 g) applied on a glass plate (20 Â 20 mm) was placed
in a flat stainless steel reactor (V = 30 cm3) equipped with septa,
valves, and a quartz window. The gaseous toluene (C = 200 ppm)
was passed through the reactor space for 1 min. Then the reactor
was closed and kept in the dark for 30 min to reach adsorption–
desorption equilibrium. The content of the reactor was irradiated
by array of 25 visible (kmax = 415 nm)-light-emitting diodes (LEDs).
Toluene concentration was estimated at regular time periods of 10
min during 60 min of irradiation using a Perkin-Elmer Clarus 500
gas chromatograph equipped with an Elite-5 capillary column
2.2. Synthesis of MNPs/RE-KTaO3 and MNPs/RE-K2Ta2O6
One-step hydrothermal preparation was used to obtain RE-
KTaO3/RE-K2Ta2O6 according to previously published potassium
tantalate synthesis [30–32]. In preparation, KOH (16 g) dissolved
in deionized H2O (24 ml), Ta2O5 (4.4 g), PEG-400 (1.2 ml), and rare
earth precursor dissolved in deionized H2O (2 ml), were magneti-
cally stirred for 1 h. The optimized molar ratio of rare earth metal
to KTaO3/K2Ta2O6 was 2 mol%. The homogenous mixture was
transferred into a dried Teflon-lined stainless steel autoclave
(capacity 200 ml). The autoclave was sealed and heated at 200 °C
for 24 h to obtain perovskite-type RE-KTaO3 or at 180 °C for 24 h
to prepare pyrochlore-type RE-K2Ta2O6. After being cooled in the
air to room temperature, the precipitates were washed several
times with distilled H2O and then dried in an oven at 60 °C for
10 h. The desired white RE-KTaO3/RE-K2Ta2O6 powders were
obtained. Pristine KTaO3/K2Ta2O6 as the reference sample was syn-
thesized under the same preparation conditions.
(30 m  0.25 mm  0.25 mm) and
a flame ionization detector
(FID) with He as a carrier gas. The measured samples were dosed
using a gas-tight syringe (V = 200 ml). The photocatalytic toluene
decomposition was performed by blank tests in the absence of a
photocatalyst or illumination, which observed no meaningful
contaminant loss.
2.4. Procedure of photoactivity under UV–vis light irradiation
The photodeposition of MNPs on the RE-KTaO3/RE-K2Ta2O6 sur-
face was applied to prepare MNPs/RE-KTaO3 and MNPs/RE-
K2Ta2O6 based on an already reported photoreduction procedure
[33–35]. The suspension containing RE-KTaO3/RE-K2Ta2O6 (2 g),
noble metal precursor (KAuCl4, H2PtCl6, RhCl3), and C2H5OH
(50 ml) was degassed with N2 for 30 min (6 dm3/h), ultrasonicated
for 10 min, and UV light-irradiated for 1 h by a 250 W Xe lamp
(Heraeus Noblelight GmbH). The measured light flux was
60 mW/cm2 (Hamamatsu UV power meter). The optimized weight
ratio of noble metal to KTaO3/K2Ta2O6 was 2 wt%. The precipitates
were washed several times with distilled H2O and then dried in an
oven at 60 °C for 10 h. The desired colorful MNPs/RE-KTaO3 and
MNPs/RE-K2Ta2O6 powders were prepared.
Formic acid solution as a scavenger for holes (h+) was used in
photocatalytic H2 production. The temperature of the liquid phase
over the process was maintained at 10 0.5 °C by a thermostati-
cally controlled water bath. The suspension containing the
photocatalyst (0.2 g) dispersed in an aqueous formic acid solution
(80 ml, C = 10%) was tightly closed in a quartz reactor (V = 110 ml)
and then was magnetically stirred and degassed with N2 (6 dm3/h)
for 30 min in the dark to reach adsorption–desorption equilibrium.
The content of the reactor was irradiated by a 250 W Xe lamp
(Heraeus Noblelight GmbH), which emitted UV–vis light irradia-
tion. The measured light flux was 60 mW/cm2. The amount of
evolved H2 was determined at regular time periods of 30 min
during 240 min of irradiation using a Perkin-Elmer Clarus 500
gas chromatograph equipped with a HayeSep Q (80/100) column
and a thermal conductivity detector (TCD) with N2 as a carrier
gas. The measured samples were injected using a gas-tight syringe
(V = 200 ml). Photocatalytic H2 evolution was established by a blank
test in the absence of a photocatalyst, which reported no signifi-
cant H2 production.
2.3. Measurements of photoactivity under vis light irradiation
2.3.1. Phenol aqueous solution degradation
Phenol as a model water contaminant was used in photocat-
alytic pollutant decomposition under visible light irradiation. The
temperature of the liquid phase was maintained at 10 0.5 °C by
a thermostatically controlled water bath. The suspension contain-
ing photocatalyst (125 mg) and phenol aqueous solution (25 ml,
C = 20 mg/l) was placed in a quartz photoreactor (V = 25 ml) and
then magnetically stirred (500 rpm) and aerated (6 dm3/h) for 30
min in the dark to reach adsorption–desorption equilibrium. The
content of the reactor was irradiated by a 1000 W Xe lamp (Oriel)
equipped with a water filter to separate IR radiation and a glass fil-
ter (k > 420 nm) to isolate UV radiation. The measured light flux
was 5.3 mW/cm2. The suspension (1 ml) was collected at regular
Additionally, repeatability and stability in three consecutive
cycles of the most photoactive sample in H2 generation were also
measured after 1 and 12 weeks from the moment of the sample’s
preparation.
3. Results and discussion
Table 1 describes the preparation conditions and selected
characteristics of perovskite-type KTaO3, MNPs/RE-KTaO3 and

374
A. Krukowska et al. / Journal of Catalysis 364 (2018) 371–381
Table 1
Description of preparation conditions and physicochemical properties of KTaO3, MNPs/RE-KTaO3 and K2Ta2O6, MNPs/RE-K2Ta2O6.
Sample label
Noble metal precursor
Rare earth metal precursor
Lattice parameter a (Å)
Sample color
Type
Content (wt%)
Type
Content (mol%)
KTaO3
K2Ta2O6
KTaO3
–
–
–
–
3.99162(2)
3.99021(3)
3.99014(3)
3.99532(3)
3.99301(4)
3.99335(3)
3.99225(2)
3.99378(2)
–
–
–
–
White
Au/Er-KTaO3
Pt/Er-KTaO3
Rh/Pr-KTaO3
K2Ta2O6
Au/Pr-K2Ta2O6
Pt/Pr-K2Ta2O6
Rh/Er-K2Ta2O6
KAuCl4
H2PtCl6
RhCl3
–
KAuCl4
H2PtCl6
RhCl3
2.0 Au
2.0 Pt
2.0 Rh
–
2.0 Au
2.0 Pt
2.0 Rh
Er(NO3)3Á5H2O
Er(NO3)3Á5H2O
Pr(NO3)3Á6H2O
–
Pr(NO3)3Á6H2O
Pr(NO3)3Á6H2O
Er(NO3)3Á5H2O
2.0 Er
2.0 Er
2.0 Pr
–
2.0 Pr
2.0 Pr
2.0 Er
Light violet
Dark gray
Light beige
White
Violet
Gray
10.61875(4)
10.61943(9)
10.61579(4)
10.62071(6)
Beige
pyrochlore-type K2Ta2O6, MNPs/RE-K2Ta2O6 photocatalysts. The
presented methods and techniques used in experiments confirmed
the existence of MNPs immobilized on the surface and RE inclusion
into the structure of KTaO3/K2Ta2O6. The investigation presents the
influence of noble metal type and rare earth dopant type on
physicochemical properties correlated with photocatalytic perfor-
mance of MNPs/RE-KTaO3 and MNPs/RE-K2Ta2O6 compared with
pristine KTaO3/K2Ta2O6 (reference sample).
KTaO3/K2Ta2O6 corresponds to about 3.55/4.50 eV, which is in good
agreement with the literature [5,6], while the estimated band gap
energy of MNPs/RE-KTaO3 and MNPs/RE-K2Ta2O6 is reduced to
3.37–3.52 eV and 4.40–4.47 eV, respectively. Therefore, it is con-
firmed that MNPs/RE modification of potassium tantalate causes
band gap narrowing and extended visible light response. All of
MNPs/RE-KTaO3 and MNPs/RE-K2Ta2O6 attain higher absorbance
in the visible light range than reference KTaO3/K2Ta2O6 due to
absorption of light by loaded and doped metals. It can be seen that
the Au/Er-KTaO3 and Au/Pr-K2Ta2O6 present maximum absorption
peaks at about 540 nm due to LSPR of Au NPs. Panda and Chat-
topadhyay reported a LSPR effect for spherical Au NPs with average
diameter of about 5–10 nm exposed at around 540 nm [38].
PL emission spectroscopy was applied to characterize the
electron-hole recombination behavior of photocatalysts. The PL
spectra of reference KTaO3/K2Ta2O6 and MNPs/RE-KTaO3 and
MNPs/RE-K2Ta2O6 are shown in Fig. 1c. All as-prepared photocata-
lysts present broadband PL emission with maximum peaks at
around 400/408 nm and 450 nm (blue luminescence) attributed
to intrinsic irradiative emission and a maximum peak in the region
at about 525 nm (green luminescence) ascribed to extrinsic defect
irradiation [39,40]. It can be seen that perovskite-type KTaO3,
MNPs/RE-KTaO3 exhibit lower PL intensity compared to
pyrochlore-type K2Ta2O6, MNPs/RE-K2Ta2O6, indicating limited
recombination of photogenerated electron–hole pairs (a trend line
in Fig. 1c). Furthermore, MNPs/RE-KTaO3 and MNPs/RE-K2Ta2O6
reveal lower PL intensity than reference KTaO3/K2Ta2O6,
caused by efficient electron scavenging by deposited and doped
metals. The greatest reduction in charge carrier recombination
is observed for Pt/Er-KTaO3 among MNPs/RE-KTaO3, but for
Au/Pr-K2Ta2O6 and Pt/Pr-K2Ta2O6 among MNPs/RE-K2Ta2O6 photo-
catalysts, respectively.
3.1. Optoelectronic properties
The DRS UV–vis spectra of KTaO3, MNPs/RE-KTaO3 and K2Ta2O6,
MNPs/RE-K2Ta2O6 are presented in Fig. 1a and b, respectively. Both
reference KTaO3 and K2Ta2O6 reveal absorption edges in the UV
light region. Perovskite-type KTaO3 shows a steep absorption edge
rising at about 340 nm, while pyrochlore-type K2Ta2O6 exhibits
abroad absorption edge between 340 and 250 nm. Described spec-
tral responses are assigned to a charge carrier transfer from the
valence band formed by 2p orbitals of O2À anions to the conduction
band formed by 5d orbitals of Ta5+ cations. The difference in band
gaps between KTaO3 and K2Ta2O6, results from different overlap-
ping of tantalum and oxygen orbitals. The Ta–O distance is longer
in K2Ta2O6, therefore the level of the conduction band is higher in
K2Ta2O6 than in KTaO3 [36]. The absorption edges of both MNPs/
RE-KTaO3 and MNPs/RE-K2Ta2O6 slightly shift toward longer wave-
lengths than pristine KTaO3/K2Ta2O6. The described red shift effect
indicates a reduction in the band gap energy of MNPs/RE-modified
KTaO3/K2Ta2O6 [37]. The insets of Fig. 1a and b reveal a trans-
formed Kubelka–Munk function versus photon energy used to
estimate the optical band gaps of photocatalysts. The linear portion
of the plots was extrapolated to the energy of the light axis to
assess the band gap. The evaluated band gap energy of reference
Fig. 1. DRS UV–vis spectra of (a) KTaO3, MNPs/RE-KTaO3 and (b) K2Ta2O6, MNPs/RE-K2Ta2O6. The insets show the corresponding optical band gap estimated from the
transformed Kubelka–Munk function versus photon energy. (c) PL emission spectra of KTaO3, MNPs/RE-KTaO3 and K2Ta2O6, MNPs/RE-K2Ta2O6. Prepared samples were
excited with a 325 nm wavelength by a Xe lamp.

A. Krukowska et al. / Journal of Catalysis 364 (2018) 371–381
375
The reference KTaO3/K2Ta2O6 powders are white, while all of
the MNPs/RE-modified KTaO3/K2Ta2O6 samples are colored in
different brightnesses of violet, gray, and beige, confirming a
change in absorbance and photoluminescence properties (Table 1
and Table S.1 in the Supplementary Material).
hydrothermal synthesis. The larger size probably results from the
addition of PEG-400 as a hydrophilic reaction medium in prepara-
tion. Corresponding EDS mapping indicates the presence of K, Ta,
and O on the surface of bare KTaO3/K2Ta2O6 without other impuri-
ties. Fig. 2b–d and f–h show TEM images of approximately spher-
ical Au, Pt, and Rh NPs deposited on RE-KTaO3/RE-K2Ta2O6
surface. The Au and Pt NPs tended to agglomerate into larger par-
ticles, while Rh NPs are rarely and separately deposited on KTaO3/
K2Ta2O6 surface. The average size of Au NPs is in the range of 22–
29 nm, while the size dimension of Pt and Rh NPs is smaller than
10 nm (Table S.2). The specific crystallographic planes for gold
and platinum on RE-KTaO3 and RE-K2Ta2O6 surfaces were identi-
fied by fast Fourier transform (FFT) analysis. However, Rh NPs
are too dispersed on the semiconductor surface to be detected by
FFT analysis. The interplanar distances of noble metals confirm
the presence of Au(3 1 1), Au(2 0 0), and Pt(3 1 1) crystal planes
[43] (Fig. 2b, c, f, and g). Corresponding EDS mapping reaffirms
the existence of both MNPs and RE ions on the KTaO3/K2Ta2O6 sur-
face. Morphological mapping shows that Er and Pr elements are
homogeneously distributed in the microcubes. Introduction of RE
ions into the crystal lattice does not affect then shape and smooth
surface of potassium tantalate significantly (Fig. 2b–d and f–h).
3.2. Surface morphology
TEM imaging, EDS mapping, and FFT analysis were performed to
investigate the impact of MNPs deposition and RE ions doping on
the morphology and structure of potassium tantalate. Fig. 2a–d
and e–h display microscopic analysis of KTaO3, MNPs/RE-KTaO3
and K2Ta2O6, MNPs/RE-K2Ta2O6, respectively. The overall TEM
views of reference KTaO3 and K2Ta2O6 particles (Fig. 2a and e) pre-
sent cubelike shapes with smooth surfaces and side lengths of
about 0.2–2 lm. Some of the smaller microcubes form larger
agglomerates. The average edge size of pristine KTaO3/K2Ta2O6 is
much larger than that of KTaO3 nanocubes (0.2–0.25 mm) [41]
and K2Ta2O6 cubic-like crystallites (0.1 mm) [42] prepared in
3.3. Crystalline structure
The PXRD measurements were carried out to verify sample pur-
ity and estimate a lattice constant of potassium tantalate. Fig. 3a–d
and e–h present PXRD patterns of KTaO3, MNPs/RE-KTaO3 and
K2Ta2O6, MNPs/RE-K2Ta2O6, respectively. To show low-intensity
peaks, experimental data points (open circles) are plotted on a
log scale. A red solid line is a profile fitting (LeBail refinement),
whereas red and black vertical bars represent expected Bragg
reflection positions for KTaO3 (Pm-3m, s.g. 221) and K2Ta2O6
(Fd-3m, s.g. 227), respectively. For MNPs/RE-KTaO3 (Fig. 3a–d),
the perovskite type of potassium tantalate is a majority phase with
the pyrochlore K2Ta2O6 phase estimated between 2% for a refer-
ence KTaO3 sample and around 7% for the Rh/Pr-KTaO3. For
MNPs/RE-K2Ta2O6 (Fig. 3e–h), the pyrochlore type of potassium
tantalate is a majority phase with the perovskite KTaO3 phase
found in between 5% for a pristine K2Ta2O6 sample and about
10% for Rh/Er-K2Ta2O6. The only PXRD reflection originating from
a noble metal is marked by a vertical arrow at around 38.2°
(Fig. 3b and f) and confirms the presence of elemental gold. The
strongest PXRD reflection for Pt metal is expected to be 2h ꢀ 39.
3° (Fig. 3c and g) and is likely overlapped by a reflection of a main
phase. The Rh NP diffraction peaks are not observed due to low
deposition of Rh particles on RE-KTaO3/RE-K2Ta2O6 surfaces
(Fig. 3d and h). KTO3 and K2Ta2O6 represent the same chemical
composition, but different electronic configurations are assigned
to geometrical effects of cation and anion arrangements in the
crystal lattice. Table 1 gathers estimated lattice parameters for
KTaO3, MNPs/RE-KTaO3 and K2Ta2O6, MNPs/RE-K2Ta2O6 series.
The LeBail refinement for pristine KTaO3 reveals a lattice constant
a = 3.99162(2) Å, which is in good agreement with reported a = 3.
9883 Å [44], while for bare K2Ta2O6 an estimated lattice parameter
a = 10.61875(4) Å is very close to a = 10.6189 Å reported for
K1.93Ta2O5.9 [39]. Obtained lattice parameters for MNPs/RE-KTaO3
and MNPs/RE-K2Ta2O6 samples change by less than 0.09% and
0.03%, respectively. Fig. 3i, 3j present unit cells of potassium tanta-
late of perovskite type (KTaO3) and pyrochlore type (K2Ta2O6),
respectively. These two structures differ in the connectivity
between vertex-sharing TaO6 octahedra: In perovskite structure,
TaO6 units are arranged in an octahedron (Fig. 3k), while in
pyrochlore structure, TaO6 units form
a tetragonal network
(Fig. 3l). The Ta ion is located in the center of an octahedron formed
Fig. 2. TEM images/EDS mapping/FFT analysis of (a–d) KTaO3, MNPs/RE-KTaO3 and
(e–h) K2Ta2O6, MNPs/RE-K2Ta2O6.
by O atoms (TaO26À) in both structures of potassium tantalate

376
A. Krukowska et al. / Journal of Catalysis 364 (2018) 371–381
Fig. 3. PXRD patterns with a LaBail fit for (a–d) KTaO3, MNPs/RE-KTaO3 and (e–h) K2Ta2O6, MNPs/RE-K2Ta2O6. Red and black vertical bars represent expected Bragg reflection
positions for perovskite and pyrochlore structure, respectively. Visualization of (i) unit cell of perovskite KTaO3 with TaO26À octahedron highlighted in the center, (j) unit cell of
pyrochlore K2Ta2O6, (k) octahedron as dominant Ta5+ coordination polyhedra in perovskite-type KTaO3, and (l) tetrahedron as dominant Ta5+ coordination polyhedra in
pyrochlore-type K2Ta2O6. Images were drawn by VESTA software [29].
(Fig. 3i–l). In our recent work [11], RE ions likely substituted potas-
sium in K2Ta2O6, what was confirmed by theoretical simulations of
band structure and partial density of states (PDOS). Hence, we sup-
pose that RE ions partially incorporate into the KTaO3/K2Ta2O6
structure.
forms of both metals in addition to metal states [45] (Fig. 4a, b,
and Table 3). The Au4f spectra recorded for the Au/Er-KTaO3 and
Au/Pr-K2Ta2O6 samples are well fitted by three doublet peaks at
BE of Au4f7/2 signals located at 84.9, 83.9, and 83.0 eV, representing
the positively charged state Au (d+), metal state Au(0), and nega-
tively charged state Au (d-), respectively [46–48] (Fig. 4a and b).
The last two states are dominant fractions of Au surface species
in both the Au/Er-KTaO3 and Au/Pr-K2Ta2O6 samples (Table 3).
The large contribution of negatively charged Au nanoparticles
indicates electron transfer from the RE (Er, Pr)-doped K2Ta2O6/
KTaO3 supports to the gold nanoparticles [46]. The Er3d and Pr3d
spectra recorded for Au/Er-KTaO3, Pt/Er-KTaO3, Rh/Er-K2Ta2O6,
Au/Pr-K2Ta2O6, Pt/Pr-K2Ta2O6, and Rh/Pr-KTaO3 specimens identify
only the oxidized states of Er and Pr (Fig. 4a, b and Table 3). The
inspection of K2p and O1s spectra recorded for MNPs/RE-
modified KTaO3/K2Ta2O6 reveal both spectra to be significantly
changed from those of pristine KTaO3/K2Ta2O6. Mostly the peaks
representing the surface states in the K2p spectra (K2p3/2
signal close to 293 eV) and in oxygen the O1s spectra (O1s signal
close to 531 eV) are evidently rearranged (Fig. 4a and b). It
should be also noted that the surface amount of Pt in Pt/Er-
KTaO3 (28.62 wt%) and Pt/Pr-K2Ta2O6 (8.50 wt%) photocatalysts
as well as of Au in the Au/Pr-K2Ta2O6 sample (5.72 wt%) signifi-
cantly exceed the nominal Pt and Au content determined by
preparation processing (2 wt%; Table 2).
3.4. Surface elemental composition
The elemental composition and chemical character of elements
detected in the surface regions of both pristine and MNPs/RE-
modified KTaO3/K2Ta2O6 were identified using XPS, and the result-
ing data are collected in Tables 2 and 3, respectively. The presence
of KTaO3, MNPs/RE-KTaO3 and K2Ta2O6, MNPs/RE-K2Ta2O6 is con-
firmed by corresponding HR XPS spectra shown in Fig. 4a and b,
respectively. The chemical character of K, Ta, O, and C originated
from reference KTaO3/K2Ta2O6 is similar to that reported in our last
published papers [11,12]. However, the K2p spectrum recorded for
pyrochlore-type K2Ta2O6 is formed by much broader peaks (FWHM
equal to 1.97 and 1.89 for lattice and surface state, respectively)
than the corresponding K2p spectrum recorded for perovskite-
type KTaO3 (FWHM equal to 0.91 and 1.71 for lattice and surface
state, respectively). That indicates the coexistence of pyrochlore
and perovskite phases in the pristine K2Ta2O6 reference material
(Fig. 4a and b). Deconvolution of Pt4f and Rh3d spectra recorded
for Pt/RE- and Rh/RE-modified KTaO3/K2Ta2O6 reveals oxidized
Table 2
Elemental composition in the surface layer of KTaO3, MNPs/RE-KTaO3 and K2Ta2O6, MNPs/RE-K2Ta2O6 evaluated from XPS analysis.
Sample label
Elemental composition (wt%)
Ta
K
O
C
Au
Pt
Rh
Er
Pr
KTaO3
67.08
66.65
47.34
60.88
70.27
65.48
64.58
64.63
15.36
9.99
6.99
16.33
10.81
8.62
16.66
17.76
12.31
17.77
18.33
18.06
16.49
19.02
0.90
0.86
2.70
1.01
0.58
0.72
1.12
1.37
–
2.22
–
–
–
5.72
–
–
–
–
–
–
–
2.40
–
–
–
2.36
–
–
–
–
1.62
–
1.43
0.92
–
Au/Er-KTaO3
Pt/Er-KTaO3
Rh/Pr-KTaO3
K2Ta2O6
Au/Pr-K2Ta2O6
Pt/Pr-K2Ta2O6
Rh/Er-K2Ta2O6
2.52
2.03
–
–
–
28.62
–
–
–
8.50
–
8.38
11.72
–
0.91

A. Krukowska et al. / Journal of Catalysis 364 (2018) 371–381
377
3.5. Photocatalytic activity
The potential environmental application of KTaO3, MNPs/RE-
KTaO3 and K2Ta2O6, MNPs/RE-K2Ta2O6 was examined in three
model reactions i.e. decomposition of phenol solution and removal
of gaseous toluene under visible light irradiation and H2 evolution
from formic acid solution under UV–vis light irradiation. Table 4
presents efficiency of experimental wastewater and air purification
as well as green energy production, while Fig. 5a–c display photo-
catalytic kinetics of phenol/toluene degradation and H2 generation
influenced by MNPs/RE-modified KTaO3 and K2Ta2O6, respectively.
The results of wastewater purification show that photolysis
(2.75%) does not meaningfully affect the phenol oxidation reaction
with the photocatalyst under visible light irradiation (Fig. 5a and
Table 4). It can be seen that all prepared KTaO3-based semiconduc-
tors are more photoactive in phenol decomposition than obtained
K2Ta2O6-based photocatalysts. The highest photocatalytic activity
is observed for Au/Er-KTaO3 (16.48%) among MNPs/RE-KTaO3 and
for Au/Pr-K2Ta2O6 (9.27%) among MNPs/RE-K2Ta2O6 samples,
respectively. The Au NPs deposited on Er-KTaO3 reveal the highest
wastewater purification efficiency compared with RE-K2Ta2O6 [11]
and MNPs/BNPs-KTaO3 [12] photocatalysts due to the synergistic
effect between Au NPs and Er-ion-modified potassium tantalate.
The lowest photoactivity is shown by Rh/Er-K2Ta2O6 (6.39%), while
reference KTaO3/K2Ta2O6 exhibit 5.45%/4.23% phenol degradation,
respectively. The toluene concentration in photolysis (7.27%)
slightly decreases, which does not significantly influences toluene
decomposition by MNPs/RE-modified KTaO3/K2Ta2O6 under visible
LEDs light irradiation (Fig. 5b and Table 4). It can be observed that
pristine KTaO3 (13.11%) shows more effective photoactivity than
bare K2Ta2O6 (9.90%); however most of MNPs/RE-K2Ta2O6 present
better toluene removal performance than MNPs/RE-KTaO3 photo-
catalysts. The highest air purification efficiency is obtained by
Pt/Pr-K2Ta2O6 (36.13%). The MNPs/RE-modified KTaO3/K2Ta2O6
with incorporated Pr ions reveal the most successful photoactivity
in gaseous pollutant degradation, that is, Au/Pr-K2Ta2O6, Rh/Pr-
KTaO3, and Pt/Pr-K2Ta2O6 among prepared photocatalysts. The
lowest toluene decomposition efficiency is exhibited by Au/Er-
KTaO3_seq (22.83%). Formic acid aqueous solution is used as an
efficient source of electron donors in H2 production with photocat-
alysts under UV–vis light irradiation (Fig. 5c and Table 4). It can be
seen that all of MNPs/RE-modified KTaO3/K2Ta2O6 show higher
activity of H2 evolution than reference KTaO3/K2Ta2O6. The highest
photocatalytic activity is observed for Pt/Pr-K2Ta2O6 (93.16 mmol/
min), which is almost 11 times higher than for pristine K2Ta2O6
(8.85 mmol/min). The Pt/Pr-K2Ta2O6 reveals the largest amount of
H2 generation compared with RE-K2Ta2O6 [11] and MNPs/BNPs-
KTaO3 [12] photocatalysts due to the synergistic impact of the
Pt-NPs-decorated surface and Pr-ion-doped lattice of K2Ta2O6,
which enhanced separation and transfer of photogenerated elec-
tron–hole pairs and increased visible light harvesting. The amount
of H2 production after 240 min of UV–vis light irradiation is in the
range 35.25–82.35 mmol/min for MNPs/RE-KTaO3 and 36.94–93.1
6 mmol/min for MNPs/RE-K2Ta2O6 samples, respectively. The
lowest green energy production is shown by Au/Er-KTaO3 (35.25
mmol/min). Moreover, MNPs/RE-modified KTaO3/K2Ta2O6 are easy
to recycle by filtration due to their powder form. Fig. 5d shows
the stability of Pt/Pr-K2Ta2O6 as the most photoactive sample after
three consecutive cycles in H2 generation. The photocatalytic activ-
ity decreases from 88.78 mmol/min (first cycle) to 74.88 mmol/min
(third cycle), presenting beneficial reusability of photocatalyst.
After about 12 weeks from the moment of the sample’s prepara-
tion, photoactivity change over three consecutive runs from
71.11 mmol/min (first cycle) to 61.98 mmol/min (third cycle), indi-
cating relatively good long-term stability of the photocatalyst in
practical applications.

378
A. Krukowska et al. / Journal of Catalysis 364 (2018) 371–381
Fig. 4. HR XPS spectra of K2p, Ta4d, O1s, Au4f, Pt4f, Rh3d, Er4d, and Pr3d recorded in the surface regions of (a) KTaO3, MNPs/RE-KTaO3 and (b) K2Ta2O6, MNPs/RE-K2Ta2O6.
To sum up, KTaO3/K2Ta2O6 were modified by MNPs (Au, Pt, Rh)
decorated surface and RE (Er, Pr) ions doped lattice to improve the
effect of photoactivity in organic pollutant decomposition under
visible light irradiation and H2 generation under UV–vis light irra-
diation. The synergistic effect between MNPs and RE-ions-modified
perovskite (KTaO3) and pyrochlore (K2Ta2O6) phases of potassium
tantalate results in enhanced photocatalytic performance. Based
on the results in our previous work [11,12], inclusion of RE ions
in the KTaO3/K2Ta2O6 lattice probably leads to formation of RE4f
levels below the conduction band within the band gap, while

A. Krukowska et al. / Journal of Catalysis 364 (2018) 371–381
379
Table 4
shape with a smooth surface and a side length of about 0.2–2 lm
Photocatalytic performance of KTaO3, MNPs/RE-KTaO3 and K2Ta2O6, MNPs/RE-
K2Ta2O6.
particles. The perovskite-type MNPs/RE-KTaO3 exhibits lower PL
intensity than the pyrochlore-type MNPs/RE-K2Ta2O6, indicating
efficient electron scavenging ability in semiconductors with a
perovskite structure. The red shift effect described in absorption
spectra of both MNPs/RE-KTaO3 and MNPs/RE-K2Ta2O6, demon-
strating band gap narrowing due to MNPs/RE modification. The
estimated band gap energy is reduced from 3.55 to 3.37 eV in the
Pt/Er-KTaO3 and from 4.50 to 4.40 eV in the Pt/Pr-K2Ta2O6, respec-
tively. Ishihara et al. reported that pyrochlore-type K2Ta2O6 (3.6
mmol/h g) exhibited higher activity of H2O splitting into H2 under
UV–vis light irradiation than perovskite KTaO3 (33.1 mmol/h g). It
was explained that the octahedron complex (TaO26-) in K2Ta2O6 is
much distorted; therefore charge separation can be easier in
K2Ta2O6 than in KTaO3 due to the local polarization effects [20].
Sample label
Phenol
Toluene
Amount of H2
generation under
UV–vis light
irradiation after
240 min (mmol/min)
decomposition
under visible
light irradiation
after 90 min (%)
removal under
visible light
irradiation after
60 min (%)
KTaO3
5.45
16.48
14.77
9.98
4.23
9.27
8.85
6.39
13.11
22.83
30.92
34.73
9.90
31.06
36.13
27.24
7.23
Au/Er-KTaO3
Pt/Er-KTaO3
Rh/Pr-KTaO3
K2Ta2O6
Au/Pr-K2Ta2O6
Pt/Pr-K2Ta2O6
Rh/Er-K2Ta2O6
35.25
82.35
66.10
8.85
75.28
93.16
36.94
photoreduction of MNPs on the RE-KTaO3/RE-K2Ta2O6 surface
causes a LSPR effect and activates the surface by adsorption for
both O2 and H2O molecules. The Au NPs/Er ions modification is
advantageous for decomposition of phenol aqueous solution
(16.48%), while Pt NPs/Pr ions modification is beneficial for both
removal of gaseous toluene (36.13%) and H2 evolution from formic
acid solution (93.16 mmol/min), respectively. Also, the Pt/Pr-
K2Ta2O6 sample shows a relatively good lifetime in H2 production
(from 88.78 to 61.98 mmol/min). The Au NPs (22–29 nm) and
Pt NPs (smaller than 10 nm) are regularly deposited on the surface
in the form of agglomerates, while Er and Pr ions are homoge-
neously distributed in KTaO3/K2Ta2O6 crystals. The Pt, Rh, Er, and
Pr surface species reveal oxidized states, while Au has the forms
of a positively charged state Au (d+), a metal state Au(0), and a neg-
atively charged state Au (d-). KTaO3/K2Ta2O6 presents a cubelike
3.6. Discussion of photocatalytic mechanism
Fig. 5e identifies active species formed during phenol degrada-
tion with Au/Er-KTaO3 as the most photoactive sample under
visible light irradiation using radical scavengers: AgNO3 for
electrons (eÀ), (NH4)2C2O4 for holes (h+), C6H4O2 for superoxide
radical species (OÅ2À), and (CH3)3COH for hydroxyl radical species
(ÅOH), compared with experiments without any quencher. It can
be seen that introduction of a C6H4O2 scavenger significantly sup-
pressed phenol oxidation (5.87%), indicating that OÅ2À radicals are
the main active species involved in pollutant decomposition under
visible light irradiation. The addition of AgNO3, (CH3)3COH, and
(NH4)2C2O4 quenchers also slightly inhibits phenol degradation,
causing 9.81, 11.24, and 12.12% purification efficiency, respec-
tively. The results show that in majority OÅ2À radicals and in
Fig. 5. Photocatalytic performance of KTaO3, MNPs/RE-KTaO3 and K2Ta2O6, MNPs/RE-K2Ta2O6: (a) kinetics of phenol decomposition (b) kinetics of toluene removal (c)
amount of H2 generation (d) recycling after three consecutive runs in H2 production after about 1 and 12 weeks from the moment of preparation (e) detection of active
species using scavengers (radical) i.e. AgNO3 (eÀ), (NH4)2C2O4 (h+), C6H4O2 (OÅ2À) and (CH3)3COH (ÅOH).

380
A. Krukowska et al. / Journal of Catalysis 364 (2018) 371–381
minority ÅOH radicals, photogenerated electrons and holes are
actively involved in visible-light-induced pollutant decomposition.
For example, Liu et al. reported that methyl orange oxidation
caused by La3+-doped KTaO3 under UV–vis light irradiation was
limited by C6H4O2 (OÅ2À scavenger) and (CH3)3COH (ÅOH scavenger)
as well as being increased in the presence of AgNO3 (eÀ scavenger),
concluding that OÅ2À and ÅOH radicals were the main species present
under UV–vis light irradiation [21].
the RE-KTaO3/RE-K2Ta2O6 conduction band with simultaneous for-
mation of M+ ions. Meanwhile, photogenerated electrons from the
RE-KTaO3/RE-K2Ta2O6 valence band are transferred to lower-
energy sub-band-gap levels of RE ions. All these electrons are
transferred on the semiconductor surface to react with dissolved
O2 in water and generate O2ÅÀ radicals. In consequence, OÅ2À radicals
and their further combination forms (HOOÅ, H2O2, and HOÅ radicals)
cause oxidation and mineralization pollution into CO2 and H2O.
The M+ ions can also oxidize contamination and further are
reduced to metal again [51]. The proposed photocatalytic H2 gen-
eration process induced by UV–vis light is shown in Fig. 6b. Under
UV–vis light irradiation, electrons are photoexcited from the
KTaO3/K2Ta2O6 valence to the conduction band and from the
KTaO3/K2Ta2O6 valence band to sub-band-gap states of RE ions,
leaving behind photogenerated holes in the valence band. These
photoexcited electrons are transferred to MNPs, probably due to
the higher electronegativity of Au (2.54), Pt (2.28), and Rh (2.28)
than of Ta (1.50) [52]. Therefore, MNPs deposited on RE-KTaO3/
RE-K2Ta2O6 work as electron traps, where photoexcited electrons
reduce H+ to H2 [53]. Meanwhile, photogenerated holes from the
RE-KTaO3/RE-K2Ta2O6 valence band react with formic acid solution
and competitive water [54]. H2O oxidation causes a split into ÀOH
and H+ ions, where ÀOH ions are oxidized to ÅOH radicals and sub-
The possible mechanisms of pollutant degradation under visible
light irradiation and H2 generation under UV–vis light irradiation
in the presence of MNPs/RE-KTaO3 and MNPs/RE-K2Ta2O6 are pro-
posed in Fig. 6a and b. The simultaneous inclusion of RE ions in the
lattice and photodeposition of MNPs on the surface of KTaO3/
K2Ta2O6 enhances photocatalytic performance. The presence of
both components (RE ions and MNPs) induces optical absorption
in the visible light range of electromagnetic irradiation and slightly
shifts absorption edges toward longer wavelengths than for pris-
tine samples. Moreover, optical spectra describe small band gap
narrowing compatible with direct valance band (VB)–to–conduc
tion band (CB) excitation by visible photons. Consequently, addi-
tional components induce enhanced photoactivity under visible
light irradiation. Therefore, the photocatalytic mechanism of pollu-
tant degradation under visible light irradiation is probably caused
by a complex mechanism in the majority to produce a LSPR effect
due to MNPs as well as in the minority to form of RE4f states below
the conduction band levels to electron transition by lower photon
excitation energy. In contrast, the photocatalytic mechanism of H2
generation under UV–vis light irradiation is probably mainly
directly photoexcited electrons from the valance to the conduction
band and from the valence band to RE4f states below the conduc-
tion band levels induced by UV light irradiation. The noble metal
NPs can facilitate interfacial electron transfer to semiconductors
or other acceptors due to the LSPR effect. Moreover, noble metals
can act as electron traps to promote electron–hole pair separation
and reduction charge carrier recombination by serving as electron
sinks [49]. Furthermore, loading noble metals activates surface
semiconductors by adsorption of both O2 and H2O molecules,
resulting in formation of active species such as ÅOH or O2ÅÀ radicals.
On the other hand, inclusion of RE ions into the KTaO3/K2Ta2O6 lat-
tice probably forms RE4f states below the conduction band, which
are called sub-band-gap levels of RE ions. The electron transition
from the KTaO3/K2Ta2O6 valence band to RE4f levels requires less
photon excitation energy than the KTaO3/K2Ta2O6 valance-to-
conduction-band transition, contributing to improved visible
light photoactivity [50]. The possible photocatalytic pollutant
decomposition process induced by visible light is presented in
Fig. 6a. Under visible light irradiation, MNPs are excited due to
the LSPR effect to transfer photogenerated electrons directly into
Å
sequently OH radicals oxidize formic acid molecules to CO2, 2H+,
and 2eÀ. The obtained electrons reduce H+, resulting in evolution
of H2. The presented oxidation reaction is known to proceed at a
high rate.
4. Conclusions
To conclude, for the first time, in this study, the influence of
noble metal type (MNPs = Au, Pt, Rh) and rare earth dopant type
(RE = Er, Pr) on physicochemical and photocatalytic properties of
MNPs/RE-KTaO3 and MNPs/RE-K2Ta2O6 compared with pristine
KTaO3/K2Ta2O6 is reported. It is found that improvement in pho-
toactivity of MNPs/RE-KTaO3 and MNPs/RE-K2Ta2O6 can be
achieved by combination of physicochemical properties of two dis-
tinct metals due to MNPs/RE modification. Hydrothermal incorpo-
ration of RE ions into the structure probably forms sub-band-gap
levels to electron transition by lower photon excitation energy,
while photoreduction of MNPs on the surface likely produces a
LSPR effect and activates the surface by adsorption for both O2
and H2O molecules. The highest photocatalytic activity was
observed for the Au/Er-KTaO3 sample in phenol decomposition,
while the Pt/Pr-K2Ta2O6 sample possessed the highest activity in
toluene removal and H2 generation. It could be stated that the
photocatalytic performance of as-prepared photocatalysts depends
Fig. 6. Schematic diagram of possible mechanism over MNPs/RE-KTaO3 and MNPs/RE-K2Ta2O6 in (a) pollutant decomposition under visible light irradiation (b) H2 production
under UV–vis light irradiation.

A. Krukowska et al. / Journal of Catalysis 364 (2018) 371–381
381
[17] D. Xu, S. Yang, Y. Jin, M. Chen, W. Fan, B. Luo, W. Shi, Langmuir 31 (2015)
9694–9699.
[18] D. Xu, K. Liu, W. Shi, M. Chen, B. Luo, L. Xiao, W. Gu, Ceram. Int. 41 (2015)
4444–4451.
[19] J. Pan, J. Cao, J. Mei, X. Zhang, S. Wang, Y. Zheng, C. Cui, C. Li, Mater. Lett. 184
(2016) 52–56.
[20] T. Ishihara, N. Baik, N. Ono, H. Nishiguchi, Y. Takita, J. Photochem. Photobiol. A
167 (2004) 149–157.
[21] X. Liu, J. Lv, S. Wang, X. Li, J. Lang, Y. Su, Z. Chai, X. Wang, J. Alloys Compd. 622
(2015) 894–901.
[22] I.E. Paulauskas, J.E. Katz, G.E. Jellison Jr, N.S. Lewis, L.A. Boatner, G.M. Brown, J.
Electrochem. Soc. 156 (2009) B580–B587.
[23] F. Rossella, L. Perucchini, P. Galinetto, G. Samoggia, M.C. Mozzati, C.B. Azzoni,
A.G. Badalyan, V.A. Trepakov, P.P. Syrnikov, Phys. Status Solidi. 4 (2007) 1101–
1104.
[24] T. Ishihara, H. Nishiguchi, K. Fukamachi, Y. Takita, J. Phys. Chem. B 103 (1999)
1–3.
on crystalline structure (perovskite and pyrochlore phases of
potassium tantalate), chemical state and type of metals used in
modification (noble and rare earth metals), and particle shape
and size (spherical MNPs at a size of a few nm, cubic-like KTaO3/
K2Ta2O6 with a dimension of a few mm), as well as absorbance
and photoluminescence properties (band gap narrowing,
electron-scavenging ability). Active-species-trapping tests indicate
that OÅ2À radicals are meaningfully involved in phenol oxidation
under visible light irradiation. This work may provide valuable
information for noble- and rare-earth-metal-modified KTaO3/
K2Ta2O6 as an advanced material in photocatalysis applications
such as wastewater and air purification as well as green energy
production.
[25] H. Hagiwara, T. Inoue, K. Kaneko, T. Ishihara, Chem. Eur. J. 15 (2009) 12862–
12870.
Acknowledgments
[26] P. Jana, C.M. Montero, P. Pizarro, J.M. Coronado, D.P. Serrano, V.A. de la Pena
O’Shea, Int. J. Hydrogen Energy 39 (2014) 5283–5290.
[27] A. Iwase, H. Kato, A. Kudo, Appl. Catal. B 136 (2013) 89–93.
[28] L.M. Torres-Martinez, R. Gomez, O. Vazquez-Cuchillo, I. Juarez-Ramirez, A.
Curoz-Lopez, F.J. Alejandro-Sandoval, Catal. Commun. 12 (2010) 268–272.
[29] K. Momma, F. Izumi, J. Appl. Crystallogr. 44 (2011) 1272–1276.
[30] M. Makarova, P. Bykov, J. Drahokoupil, M. Cernansky, Z. Dlabacek, A. Dejneka,
L. Jastrabik, V. Trepakov, Mater. Res. Bull. 47 (2012) 1768–1773.
[31] Y. Zou, Y. Hu, H. Gu, Y. Wang, Mater. Chem. Phys. 115 (2009) 151–153.
[32] Y. He, Y. Zhu, N. Wu, J. Solid State Chem. 177 (2004) 2985–2990.
[33] K. Wenderich, G. Mul, Chem. Rev. 116 (2016) 14587–14619.
[34] D. Xu, W. Shi, C. Song, M. Chen, S. Yang, W. Fan, B. Chen, Appl. Catal. B Environ.
191 (2016) 228–234.
This research was financially supported by University of Gdansk
(PROJECT BMN 538-8625-B050-16). A.K. gratefully acknowledges
Daniel Kaczor from the Instrumental Analyses Laboratory, Faculty
of Chemistry, Nicolaus Copernicus University for his support of this
research in TEM/EDS/FFT measurements.
Appendix A. Supplementary material
Supplementary data associated with this article can be found, in
the online version, at https://doi.org/10.1016/j.jcat.2018.05.013.
[35] D. Xu, M. Chen, S. Song, D. Jiang, W. Fan, W. Shi, Cryst. Eng. Commun. 16 (2014)
1384–1388.
[36] T. Takayama, H. Nakanishi, M. Matsui, A. Iwase, A. Kudo, J. Photochem.
Photobiol. A Chem. 358 (2018) 416–421.
[37] J.W. Liu, G. Chen, Z.H. Li, Z.G. Zhang, Int. J. Hydrogen Energy 32 (2007) 2269–
2271.
References
[38] B.R. Panda, A. Chattopadhyay, J. Nanosci. Nanotechnol. 7 (2007) 1911–1915.
[39] C.-C. Hu, T.-F. Yeh, H. Teng, Catal. Sci. Technol. 3 (2013) 1798–1804.
[40] V.S. Vikhnin, A.B. Kutsenko, S.E. Kapphan, Solid State Commun. 140 (2006)
439–442.
[41] X. Wang, S. Zheng, Y. Zhang, Mater. Lett. 62 (2008) 1212–1214.
[42] T. Ma, R. Zhou, X. Li, Key Eng. Mater. 680 (2015) 201–207.
[43] N. Meijas, A. Serra-Muns, R. Pleixats, A. Shafir, M. Tristany, Dalton Trans. 37
(2009) 7748–7755.
[1] P. Zhang, J. Zhang, J. Gong, Chem. Soc. Rev. 43 (2014) 4395–4422.
[2] M. Kitano, M. Hara, J. Mater. Chem. 20 (2010) 627–641.
[3] A. Kudo, Y. Miseki, Chem. Soc. Rev. 38 (2009) 253–278.
[4] G. Zhang, W. Jiang, S. Yu, Mater. Res. Bull. 45 (2010) 1741–1747.
[5] B. Modak, S.K. Ghosh, Sol. Energy Mater. Sol. Cells 159 (2017) 590–598.
[6] S. Zhu, H. Fu, S. Zhang, L. Zhang, Y. Zhu, J. Photochem. Photobiol. A 193 (2008)
33–41.
[7] M. Sepilarsky, S.R. Phillpot, D. Wolf, M.G. Stachiotti, R.L. Migoni, J. Appl. Phys.
90 (2001), 4519-4509.
[8] H. Kato, A. Kudo, Catal. Today 78 (2003) 561–569.
[9] G.K.L. Goh, S.M. Haile, C.G. Levi, F.F. Lange, J. Mater. Res. 17 (2002) 3168–3176.
[10] S. Zlotnik, S.K. Sahu, A. Navrotsky, P.M. Vilarinho, Chem. Eur. J. 21 (2015)
5231–5237.
[11] A. Krukowska, M.J. Winiarski, J. Strychalska-Nowak, T. Klimczuk, W. Lisowski,
A. Mikolajczyk, H.P. Pinto, T. Puzyn, T. Grzyb, A. Zaleska-Medynska, Appl. Catal.
B Environ. 224 (2018) 451–468.
[12] A. Krukowska, G. Trykowski, M.J. Winiarski, T. Klimczuk, W. Lisowski, A.
Mikolajczyk, H.P. Pinto, A. Zaleska-Medynska, Appl. Sur. Sci. 441 (2018) 993–
1011.
[13] M. Machida, S. Murakami, T. Kijima, S. Matsushima, M. Arai, J. Phys. Chem. B
105 (2001) 3289–3294.
[14] E. Petryayeva, U.J. Krull, Anal. Chim. Acta 706 (2011) 8–24.
[15] E. Kowalska, O.O.P. Mahaney, R. Abe, B. Ohtani, Phys. Chem. Chem. Phys. 12
(2010) 2344–2355.
[44] E.A. Zhurova, Y. Ivanov, V. Zavodnik, V. Tsirelson, Acta Cryst. B 56 (2000) 594–
600.
[45] A.V. Naumkin, A. Kraut-Vass, S.W. Gaarenstroom, C.J. Powell, NIST X-ray
Photoelectron Spectroscopy Database 20 Version 4.1, 2012.
[46] J. Radnik, C. Mohr, P. Claus, Phys. Chem. Chem. Phys. 5 (2003) 172–177.
[47] D.C. Lim, I. Lopez-Salido, R. Dietsche, M. Bubek, Y.D. Kim, J. Chem. Phys. 330
(2006) 441–448.
[48] S. Peters, S. Peredkov, M. Neeb, W. Eberhardt, M. Al-Hada, Surf. Sci. 608 (2013)
129–134.
[49] L. Ge, Mater. Chem. Phys. 107 (2008) 465–470.
[50] Y. Xie, C. Yuan, Appl. Surf. Sci. 221 (2004) 17–24.
[51] Q. Xiang, J. Yu, B. Cheng, H.C. Ong, Chem. Asian J. 5 (2010) 1466–1474.
[52] G. Brown, The Inaccessible Earth: An Integrated View to its Structure and
Composition, Springer Science & Business Media, 2012.
[53] B. Zielin´ ska, Bull. Mater. Sci. 37 (2014) 911–916.
[54] P. Bogdanoff, N. Alonso-Vante, J. Electroanal. Chem. 279 (1994) 415–421.
[16] X. Wei, C. Dong, Z. Chen, K. Xiao, X. Li, RSC Adv. 6 (2016) 56303–56312.