Different performances of Ni3(PO4)2 in TiO2 photocatalysis under aerobic and anaerobic conditions

Article abstract of DOI:10.1039/c9cy02350j

It has been reported that cobalt phosphate (CoP) deposited on TiO₂ is beneficial to the photoelectrochemical (PEC) oxidation of water in a phosphate solution, but detrimental to the photocatalytic (PC) oxidation of phenol even in a phosphate solution. Herein, we report amorphous nickel phosphate (NiP) deposited on TiO₂ being more active than TiO₂ in a phosphate-free solution. This was observed not only from the PEC oxidation of water, but also from the PC oxidation of phenol under air, and from the PC reduction of water to H₂ under N₂ in the presence of methanol, respectively. During the PC reactions under air and N₂, however, the suspension of NiP/TiO₂ changed its colour from white to pale yellow and dark grey, respectively. Such colour changes were also observed from CoP/TiO₂ under similar conditions. Through a spectral and thermodynamic analysis, it is proposed that under air, NiP is oxidized to a high valence Ni species, and then regenerated through water and phenol oxidation, while NiP under N₂ is reduced to metallic Ni, followed by proton reduction.

Full text of DOI:10.1039/c9cy02350j

Catalysis
Science &
Technology
View Article Online
View Journal
PAPER
Different performances of Ni3IJPO4)2 in TiO2
photocatalysis under aerobic and anaerobic
Cite this: DOI: 10.1039/c9cy02350j
conditions†
*
Yaru Wang, Jianjun Zhao,
Chen Chen and Yiming Xu
It has been reported that cobalt phosphate (CoP) deposited on TiO2 is beneficial to the
photoelectrochemical (PEC) oxidation of water in phosphate solution, but detrimental to the
a
photocatalytic (PC) oxidation of phenol even in a phosphate solution. Herein, we report amorphous nickel
phosphate (NiP) deposited on TiO2 being more active than TiO2 in a phosphate-free solution. This was
observed not only from the PEC oxidation of water, but also from the PC oxidation of phenol under air,
and from the PC reduction of water to H2 under N2 in the presence of methanol, respectively. During the
PC reactions under air and N2, however, the suspension of NiP/TiO2 changed its colour from white to pale
yellow and dark grey, respectively. Such colour changes were also observed from CoP/TiO2 under similar
conditions. Through a spectral and thermodynamic analysis, it is proposed that under air, NiP is oxidized to
a high valence Ni species, and then regenerated through water and phenol oxidation, while NiP under N2 is
reduced to metallic Ni, followed by proton reduction.
Received 20th November 2019,
Accepted 19th January 2020
DOI: 10.1039/c9cy02350j
rsc.li/catalysis
promoting the counterpart hole oxidation of phenol.8,9
Besides that, Pt/TiO2 is also more active than TiO2 for H2
1. Introduction
Replacement of fossil fuels with solar energy and remediation
of environmental pollution via semiconductor photocatalysis
have attracted great attention since the last century.1,2 The
primary process occurring in an irradiated semiconductor
(SC) is generation of conduction band electrons (ecb) and
valence band holes (hvb), respectively. These charge carriers
then migrate onto the surface reacting with appropriate
sorbents. Among many SCs, TiO2 is the most studied because
of its low cost, good activity, and high stability.3 More
importantly, ecb of anatase TiO2 can reduce O2 to HO2˙ or
evolution in the presence of sacrificial reagents.10,11 However,
a catalyst that acts as an hvb trap and facilitates oxidation
reactions is rare. Cobalt phosphate (CoP), after being partially
oxidized to CoPi, can promote hole oxidation of water on a
TiO2 photoanode.12,13 This is due to the hole oxidation of
CoII/III to form highly reactive CoIV. Unfortunately, CoPi
retains its activity only in a phosphate or alkaline solution
through a repair mechanism.12–14 Interestingly, CoPi can
promote H2 evolution on C3N4 or CdS in a phosphate-free
neutral aqueous suspension containing CH3OH or lactic
acid.15,16 This is due to the CoPi-mediated hole oxidation of
organics making ecb live longer for proton reduction. In
contrast, phenol oxidation on irradiated TiO2 is inhibited by
CoPi even in a phosphate solution.17 This is due to the
surface deposited CoPi inhibiting O2 adsorption and slowing
down O2 reduction and the surface bound and short-lived
CoIV inefficiently oxidizing the poorly adsorbed phenol on
TiO2.
−
O2 ˙, while its hvb can oxidize water and various organics. Like
other SCs, however, TiO2 also suffers from fast recombination
of ecb and hvb, and hence it shows a low quantum yield for
redox reactions.4
To suppress the charge recombination, one of the
strategies is to accelerate the interfacial charge transfer with
a cocatalyst.5–7 For example, Pt nanoparticles are often used
as ecb traps of TiO2 and a catalyst for O2 reduction, hence
Herein, we report amorphous Ni3IJPO4) (NiP) efficiently
promoting TiO2 photocatalyzed reactions in a phosphate-free
aqueous solution, as observed not only from H2 evolution
under N2, but also from water and phenol oxidation under
air. There have been several reports examining the effect of
Ni2+, but the mechanism is controversial. On an Fe2O3
photoanode, NiP helps glycerol oxidation due to NiIII/II
recycling,18 whereas NiP after being partially oxidized to NiPi
helps water oxidation, due to NiPi reducing the number of
State Key Laboratory of Silicon Materials and Department of Chemistry, Zhejiang
University, Hangzhou, China. E-mail: xuym@zju.edu.cn; Fax: +86 57187951895
† Electronic supplementary information (ESI) available: Tables for XRD analysis,
solid physical parameters, phenol degradation, EDS spectrum and elemental
content of NiP, XRD patterns, DRS, Tauc plots, N2 adsorption isotherms, H2O2
decomposition, IC, phenol degradation on parallel samples, XPS of P 2p and O
1s, photographs and XPS of Ni 2P and Co 2P before and after the photoreaction,
O2 reduction, pH-dependent potentials, and current–voltage curves. See DOI:
10.1039/c9cy02350j
This journal is © The Royal Society of Chemistry 2020
Catal. Sci. Technol.

View Article Online
Paper
Catalysis Science & Technology
surface states and hence increasing the efficiency of hole
injection.19 In an irradiated methanol aqueous suspension of
TiO2 under N2, surface deposited NiIJOH)2 can help H2
production, due to the formation of metallic Ni0.20,21
Meanwhile, in an irradiated aqueous suspension of TiO2
under air, surface adsorbed Ni2+ can also help phenol
degradation due to the formation of NiOOH, which is
reactive to phenol, but not to water.22 It seems that Ni2+ ions
play different roles in different systems, which is worthy of
being clarified. For practical applications, moreover, Ni2+ ions
need to be fixed without loss of catalytic activity. In this
respect, NiP is better than NiIJOH)2, because the former is
much less soluble than the latter (Ksp5 × 10−31 versus 7.9 ×
10−17).23
Micromeritics ASAP 2020. X-ray photoelectron spectra (XPS)
were recorded on a Kratos AXIS Ultra DLD, and they were
calibrated with C 1s at 284.6 eV. Elemental analysis was
conducted with ICP-OES (inductively coupled plasma optical
emission spectroscopy) on an Optima 2100 DV.
Photoluminescence (PL) spectra were recorded on
Shimadzu F-2500 spectrophotometer.
a
2.3. Photocatalysis and analysis
Reactions were carried out at 25 °C on photochemical
apparatus (Nanjing Xujiang Power Plant, China). Unless
stated otherwise, conditions were 1.0 g L−1 catalyst, 0.43 mM
phenol, and an initial pH of 8.0. An aqueous suspension
(50.0 mL) containing necessary components was stirred in
the dark for 30 min, and then irradiated with UV light at
wavelengths longer than 320 nm. At given intervals, 2.0 mL
of the suspension was taken and filtered through a 0.22 μm
membrane. Organics were analyzed by HPLC (high
performance liquid chromatography) on a Dionex UltiMate
3000 (50% CH3OH as an eluent). Phosphate anions were
analyzed by IC (ion chromatography) on a Dionex ISC90
(AS14 column and 3.5 mM Na2CO3/1.0 mM NaHCO3 mobile
phase). H2O2 was measured through POD catalyzed oxidation
of DPD.24 Solution absorbance at 553 nm was recorded on an
Agilent 8453 UV-visible spectrophotometer.
In this work, composite materials were prepared by in situ
precipitation of NiP onto TiO2. As
a reference, cobalt
phosphate or calcium phosphate was also deposited onto
TiO2. Solids were characterized with techniques such as X-ray
photoelectron spectroscopy. Reactions were carried out under
UV light in an aqueous suspension for phenol degradation
and in a pre-degassed methanol aqueous suspension for H2
evolution. In both cases, NiP/TiO2 was more active than TiO2.
To understand the role of NiP, several experiments were
performed, including NiP loading effect analysis, analysis of
independent reduction and oxidation reactions on a film
electrode, and charge transfer behaviour analysis via
photoluminescence
spectroscopy. Furthermore,
responsible for the observed effect of NiP in different systems
is discussed, in terms of both thermodynamics and kinetics.
and
electrochemical
possible
impedance
mechanism
Experiment for H2 evolution was carried out with a
suspension (50.0 mL) containing 5.0 g L−1 catalyst and 10%
CH3OH in volume. Prior to light irradiation, the suspension
was evacuated for 30 min so as to remove the dissolved O2.
Then, the reactor was closed and irradiated with four 3 W
LED lamps (365 nm). The amount of H2 in gas phase was
determined by GC (gas chromatography) on an Agilent 7820A
equipped with a thermal conductivity detector.
a
2. Experimental
2.1. Materials
Anatase TiO2, metal perchlorate, horseradish peroxidase
(POD), and N,N-diethyl-p-phenylenediamine (DPD) were
purchased from Sigma-Aldrich. Other chemicals including
sodium phosphate and phenol were from Shanghai
Chemicals. All of the reagents were used as received. NiP/
TiO2 samples were prepared by a precipitation method.
Typically, 0.2 g of TiO2 was mixed with 50 mL of 1 M Na3PO4.
Then, 0.18, 0.36, 1.08, 1.8 and 3.6 mL of 50 mM NiIJClO4)2
was added, respectively. The precipitate was filtered, washed
with water, and dried at 80 °C. The samples were denoted as
xNiP/TiO2 (x = 0.5, 1.0, 3.0, 5.0, and 10, respectively), where x
represents the calculated mass percent of Ni3IJPO4)2. By
following similar procedures, bare NiP, 3.0 CoPi/TiO2, and
3.0 CaP/TiO2 were also prepared.
2.4. (Photo)electrochemical measurement
A working electrode of TiO2 or NiP/TiO2 was prepared by a
doctor blade method. First, an indium-doped tin oxide (ITO)
substrate was cleaned with ethanol and water, and dried in
air. Then, the ITO electrode was coated with a gel containing
0.8 wt% catalyst and 2.9 wt% polyIJvinyl alcohol). The film
glass was annealed in air at 500 °C for 1 h, so as to remove
organics. Finally, the glass was cut into several pieces, and
each one had an exposed film area of
Measurements were carried out on
electrochemical station (Chenhua, Shanghai) using
1
a
×
1
cm2.
CHI660E
a
platinum gauze counter electrode and an Ag/AgCl reference
electrode. The working electrode was illuminated with a 500
W Xe lamp through a quartz window. The light intensity
reaching the external window surface was 17.3 mW cm−2,
measured with an irradiance meter (Beijing Normal
University). Electrochemical impedance spectroscopy (EIS)
was conducted under N2 at a frequency from 1 Hz to 100
kHz. The electrolyte was 0.5 M NaClO4 (pH 8.0) and the
2.2. Characterization
X-ray diffraction (XRD) patterns were recorded on a D/Max-
2550 PC (Rigaku). Transmission electron microscopy (TEM)
was carried out on a HT7700 field-emission instrument.
Diffuse reflectance spectra (DRS) were recorded on a Varian
Cary 500 with BaSO4 as
a reference. N2 adsorption–
desorption isotherms of solids were measured at 77 K on a
scanning rate was 10 mV s−1
.
Catal. Sci. Technol.
This journal is © The Royal Society of Chemistry 2020

View Article Online
Catalysis Science & Technology
Paper
3. Results and discussion
3.1. Solids characterization
a little higher than 1.5 for Ni3IJPO4)2. It means that NiP
contains some NiIJOH)2, which is 1.0–9.6 wt%, depending on
the sample. The measured content of NiP in 0.5, 1.0, 3.0, 5.0,
and 10 NiP/TiO2 was 0.46, 1.02, 2.83, 4.70, and 11.2 wt%,
respectively, close to those used in the sample synthesis.
Moreover, the survey XPS spectrum of 3.0 NiP/TiO2 shows the
presence of all the elements (Ti, Ni, P and O) without
noticeable impurities (Fig. S3B†).
The XRD patterns of the solids are shown in Fig. S1 in the
ESI.† With TiO2, there was a series of peaks well matching
with anatase TiO2 (PDF # 65-5714). With NiP, the diffraction
was rather weak. The diffraction pattern of NiP/TiO2 was
similar to that of TiO2, but the peak intensity or area was
reduced (Table S1†), due to NiP present on TiO2. From the
Scherrer equation, the average crystallite diameter of (101)
anatase was calculated, which was 14.4 0.6 nm for all the
samples (Table S1†). The TEM images show that TiO2 was
well crystallized, with distinct edges and faces, while NiP was
amorphous (Fig. 1). Moreover, in the image of 3.0 NiP/TiO2,
there were many tiny particles (NiP) on the surface of
crystalline nanoparticles (TiO2, 30 nm).
The absorption spectra of the solids are shown in Fig.
S2A.† Bare TiO2 had an intensive band at wavelengths shorter
than 400 nm, due to the band-to-band transition. Through a
Tauc plot, the band gap energy of TiO2 for indirect transition
was estimated to be 3.20 eV, very close to the spectral onset
at 380 nm. Bare NiP had three weak peaks at 398, 541, and
684 nm, due to the forbidden d–d transition of NiIJII).
However, the 541 and 684 nm peaks of NiPi in NiP/TiO2
shifted to 570 and 750 nm, respectively, as NiP loading was
increased from 0 to 10 wt% (Fig. S2B†). This is indicative of
an interaction between TiO2 and NiP, changing the
coordination spheres of Ni(II) probably from octahedra to
distorted octahedra.25,26
3.2. Phenol degradation
Fig. 2A shows the result of phenol degradation, measured
under UV light in an air-exposed aqueous suspension. With
bare TiO2, the phenol concentration decreased with time. With
bare NiP, the phenol concentration remained unchanged with
time. Interestingly, 3.0 NiP/TiO2 was more active than TiO2,
while 3.0 CoP/TiO2 was less active than TiO2. The latter is due
to the Co2+ ion acting as a recombination centre of TiO2
photocatalysis.17 Moreover, the time profiles of phenol
degradation fit well with the pseudo-first-order rate equation
(Fig. S4A†). The resulting rate constants (kobs), in a unit of 10−3
min−1, with TiO2, 3.0 NiP/TiO2, and 3.0 CoP/TiO2 were 4.65,
12.36 and 0.74, respectively. Such kinetics is often observed in
TiO2 photocatalysis and ascribed to the constant formation of
a reactive species under fixed conditions.27,28 Then, according
to the values of kobs, it follows that 3.0 NiP/TiO2 is
approximately 2.7 and 16.7 times more active than TiO2 and
3.0 CoP/TiO2, respectively, in generation of a reactive species
for phenol degradation.
The porosity of the solid was studied by N2 adsorption–
desorption isotherm measurements (Fig. S3A†). All of the
isotherms were type IV and had a hysteresis loop, due to
mesopores present in the sample, but NiP/TiO2 showed a
surface area not only smaller than that of TiO2, but also
smaller than the sum of the individual surface areas
calculated from TiO2 and NiP (Table S1†). As NiP loading
increased, moreover, the pore volume of NiP/TiO2 decreased
initially, and then increased. These observations indicate that
the external surface of TiO2 is occupied by NiP fine particles,
which block the pores and construct a new porous structure.
The chemical composition of NiP/TiO2 was analysed by
ICP-OES (Table S2†). The mole ratio of Ni to P was 1.52–1.71,
Meanwhile, hydroquinone (HQ) and benzoquinone (BQ)
were produced as the major intermediates of phenol
oxidation (Fig. 2B). Interestingly, HQ formed from TiO2 was
much more than BQ, while HQ formed from NiP/TiO2 was
much less than BQ. Since BQ is the product of HQ oxidation,
more oxidative species are produced from NiP/TiO2 than
from TiO2. Meanwhile, in both cases, the amount of the
intermediates formed was much less than the amount of
phenol disappeared (Table S3†). It means that the
intermediates once formed also degrade in situ, for which
NiP/TiO2 is also more active than TiO2.
At the same time, H2O2 was produced (open symbols,
Fig. 2B), but in this case, there was
a maximum
Fig. 1 TEM images of (a) TiO2, (b) NiP, and (c) 3.0 NiP/TiO2.
This journal is © The Royal Society of Chemistry 2020
Catal. Sci. Technol.

View Article Online
Paper
Catalysis Science & Technology
Fig. 2 (A) Phenol degradation on (a) TiO2, (b) 3.0 NiP/TiO2, (c) 3.0 CoP/TiO2, and (d) NiP. (B) The corresponding formation of hydroquinone (circle
symbols), benzoquinone (triangle symbols), and H2O2 (open symbols). (C) Effect of NiP loading in NiP/TiO2, where kobs is the apparent rate
constant of phenol degradation and Asp is the surface area of the solid.
concentration of H2O2, which was 2.9 μM with 3.0 NiP/TiO2
and 15.9 μM with TiO2. This reminded us of H2O2 adsorption
and decomposition occurring on TiO2. It has been reported
that the initial rate of H2O2 decomposition on TiO2 under UV
light is proportional to the initial amount of adsorbed
H2O2.29 To verify this, a test with 100 μM H2O2 was
conducted (Fig. S4B†). In the dark, the amount of adsorbed
H2O2 on 3.0 NiPi/TiO2 (58 μmol g−1) was less than that on
TiO2 (67 μmol g−1). Under UV light, the initial rate of H2O2
decomposition on 3.0 NiP/TiO2 (8.32 μM min−1) was higher
than that on TiO2 (3.06 μM min−1). Then, the slow formation
of H2O2 on 3.0 NiP/TiO2 is ascribed to the fast decomposition
of H2O2, as compared with that on TiO2. However, evidence
is scarce about the effect of NiP on the TiO2-photocatalyzed
reduction of O2, which will be discussed in the (Photo)
electrochemical measurement section.
It is worth noting that after the light reaction, the
dissolved Ni2+ ions in the aqueous suspension of 3.0 NiP/
TiO2 were below the detection limit (less than 10 μM). Such a
trace amount of Ni2+ did not have an influence on the rate of
phenol degradation on the irradiated TiO2.22 Meanwhile, the
dissolved phosphate ions were 14.4 μM (Fig. S4C†). It has
been reported that in an air-exposed acidic aqueous
suspension of TiO2, phosphate anions are beneficial to
phenol degradation with a complicated pathway.30,31 Then,
experiments with TiO2, TiO2 plus 15 μM Na3PO4, and 3.0
CaP/TiO2 for phenol degradation under UV light were
performed (Fig. S4D†), but these samples had similar
activities, and they were also much less active than 3.0 NiP/
TiO2. Therefore, the observed positive effect of NiP on the
TiO2 photocatalyzed degradation of phenol is surely due to
nickel cations, rather than phosphate anions in NiP.
NiP loading is probably due to excess NiP, inhibiting O2
adsorption, and hence slowing down phenol oxidation. In
the system, NiP is also the light absorbing species, but its
absorbance at wavelengths shorter than 400 nm is much less
than that of TiO2 (Fig. S2A†).
3.3. Hydrogen evolution
Fig. 3A shows the time profiles of H2 evolution, measured in
a pre-degassed aqueous suspension. In this case, methanol
was used as the hole scavenger, so as to speed up the
reduction of protons. In each case, the amount of H2
produced in gas phase increased with the irradiation time,
but such a time profile was different from one catalyst to
another. With a highly active catalyst, the rate of H2
production decreased with time. This is because the CH3OH
concentration decreases with time. With a weakly active
catalyst, the rate of H2 production increased with time. This
is due to H2 dissolution in an aqueous suspension during
the initial period of time. Then, to compare the catalyst
activity, the amount of H2 produced at 5 h was used. The
maximum production of H2 was observed with 0.5 NiP/TiO2,
no matter what the surface area of the solids was.
However, there were some differences between phenol
degradation and H2 production. First, 3.0 NiP/TiO2 was much
more active than 3.0 CoPi/TiO2 for phenol degradation
(Fig. 2A), but they showed similar activities for H2 evolution
(Fig. 3A). Second, the most active photocatalyst was 3.0 NiP/
TiO2 for phenol degradation, and 0.5 NiP/TiO2 for H2
evolution. Third, a suspension of NiP/TiO2 plus phenol, after
being irradiated for 2 h, changed its color from white to
yellow (Fig. S5A†). This is due to the yellowish BQ formed
from phenol oxidation, but a degassed suspension of NiP/
TiO2 plus CH3OH, after being irradiated for 5 h, changed its
color from white to dark gray (Fig. S5B†). To interpret this,
several supplementary experiments were carried out. Under
light for 5 h, a degassed suspension of TiO2 plus CH3OH
changed its color from white to blue (Fig. S5C†). This is due
to CH3OH oxidation being much faster than proton
reduction, so the electrons (Ti3+) of TiO2 remain for a long
time.32,33 Under light for 5 h, an aerated suspension of NiP/
The positive effect of NiP was further confirmed with
other NiP/TiO2 samples (Fig. 2C). As NiP loading increased,
the kobs value of phenol degradation increased initially, and
then decreased after reaching the maximum with 3.0 NiP/
TiO2. Since the solids have different surface areas (Asp, Table
S1†), kobs needs to be normalized with Asp. However, the
trend in kobs/Asp among the samples was quite similar to that
in kobs. Then, the increased reaction rate is surely due to the
positive effect of NiP. The decreased reaction rate at a high
Catal. Sci. Technol.
This journal is © The Royal Society of Chemistry 2020

View Article Online
Catalysis Science & Technology
Paper
Fig. 3 (A) H2 evolution in a methanol aqueous suspension of (a) TiO2, (b) 0.5 NiP/TiO2, (c) 1.0 NiP/TiO2, (d) 3.0 NiP/TiO2, (e) 5.0 NiP/TiO2, (f) 10
NiP/TiO2, and (g) 3.0 CoP/TiO2. (B) The amount of H2 formed at 5 h. (C) Ni 2p XPS spectrum of the sample (e) collected at 5 h.
TiO2 plus CH3OH did not change its color (Fig. S5D†). This is
because the electrons of TiO2 are consumed by O2. Under a
H2 balloon in the dark for 15 min, a degassed suspension of
NiP/TiO2 did not change its color as well (Fig. S5E†). It means
that Ni2+ of NiP is not reducible by H2 at room temperature.
Then, the above observed dark gray color is probably due to
the formation of Ni0 fine particles from the Ni2+ reduction on
the irradiated NiP/TiO2 in the absence of O2.
but a pre-degassed suspension of CoPi/TiO2 plus CH3OH
changed its color from white to dark gray (Fig. S8B†). Then,
the solid for H2 production was collected, dried, and analyzed
by XPS. In the XPS spectrum of Co 2p, the peaks at 780.6 and
796.6 eV are assigned to Co 2p3/2 and 2p1/2 of Co2+,
respectively (Fig. S8C†).37,38 Since the Co 2p3/2 peak was
asymmetrical, it was fitted with two peaks at 780.6 and 782.2
eV, well matching those for CoIJOH)2.36,39 After light
irradiation, unfortunately, the Co 2p3/2 peak due to metallic
Co at 778.1 eV was not observed, but the peak area at 780.6
and 782.2 eV increased and decreased, respectively (Fig.
S8D†). Since the increased peak area exists at a binding
energy close to that of Co0, it follows that Co2+ is also
reduced to Co0 by the electrons of TiO2 under anaerobic
conditions.
To verify the chemical states of Ni, a pre-degassed
suspension of NiP/TiO2 plus CH3OH was irradiated for 5 h.
Then, the solid was collected, dried, and analyzed by XPS. In
the spectrum of Ni 2p (Fig. 3C), the peaks at 855.4 eV (a
satellite at 861.3 eV) and 873.3 eV (a satellite at 879.5 eV) are
assigned to Ni 2p3/2 and 2p1/2 of Ni2+, respectively.34,35
Another peak at 853.1 eV is assigned to Ni 2p3/2 for metallic
Ni (852.6 eV).36 Meanwhile, no change was observed from the
XPS spectrum of either P 2p or O 1s (Fig. S6†). By contrast,
the sample after phenol degradation did not show an XPS
signal of Ni0 (Fig. S7†). These observations confirm that Ni2+
reduction occurs only under anaerobic conditions. Moreover,
the average binding energy of Ni 2p3/2 for NiP was 855.8 eV,
higher than that for NiO (853.7 eV) and NiIJOH)2 (854.9 eV),
respectively.36 This is probably indicative of the interaction of
3.4. (Photo)electrochemical measurement
Fig. 4A shows the dark current–voltage curves of TiO2 and 3.0
NiP/TiO2 film electrodes in 0.5 M NaClO4. As the potential
shifted cathodically, each electrode had an increased current,
due to an improved reduction process, but the cathode
current under air was much higher than under N2. It means
that O2 reduction predominates over proton reduction under
air. Moreover, the current of 3.0 NiP/TiO2 was much lower
than that of TiO2, either under air or under N2. Through a
log plot, the onset potentials for O2 reduction (Eon,OR) and for
proton reduction were estimated (Fig. S9†). The values of
Ni2+ with PO4 being stronger than those with OH− and O2−,
3−
respectively.
Furthermore, an experiment was carried out with 3.0
CoPi/TiO2. After light irradiation, an aerated suspension of
CoPi/TiO2 plus phenol did not change its color (Fig. S8A†),
Fig. 4 (A) Current–voltage curves under N2 (dotted lines) and air (solid lines). (B) Photocurrent–voltage curves under N2 in the absence (dotted
lines) and presence (solid lines) of 0.43 mM phenol. (C) Photocurrent–time curves without phenol, recorded at 0.6 V vs. AgCl/Ag. Experiments
were carried out in 0.5 M NaClO4 with (a) TiO2 and (b) 3.0 NiP/TiO2.
This journal is © The Royal Society of Chemistry 2020
Catal. Sci. Technol.

View Article Online
Paper
Catalysis Science & Technology
Eon,OR for NiPi/TiO2 and TiO2 were 50 and 61 mV vs. AgCl/Ag,
respectively, while the values of Eon,HR for NiPi/TiO2 and TiO2
were 34 and 53 mV vs. AgCl/Ag, respectively. That is, NiP/TiO2
is less active than TiO2, either for O2 reduction or for proton
reduction. Since NiP is present on the surface of TiO2, it
would inhibit the adsorption of O2 or protons. As a result,
the negative effect of excess NiP was observed either from
phenol degradation or from H2 production (Fig. 2A and 3B).
Fig. 4B shows the photocurrent–voltage curves of the TiO2
and 3.0 NiP/TiO2 film electrodes in 0.5 M NaClO4 under N2,
measured with or without 0.43 mM phenol. As the potential
shifted anodically, the photocurrent of the TiO2 electrode
increased, due to an improved charge separation. At a given
potential, the photoanode current increased on the addition
of phenol. This is indicative of phenol oxidation being easier
than water oxidation. Meanwhile, the onset potentials for
phenol and water oxidation were −0.214 and −0.104 V vs.
AgCl/Ag, respectively. However, the photocurrent and onset
potential of NiP/TiO2 were higher and more negative than
those of TiO2, respectively. The onset potentials for phenol
and water oxidation were −0.292 and −0.231 V vs. AgCl/Ag,
respectively. These observations indicate that 3.0 NiP/TiO2 is
more active than TiO2 for the hole oxidation of either phenol
or water. Since NiOOH is not reactive to water,22 the
increased water oxidation on NiP/TiO2 is ascribed to the
formation of high valence Ni4+, as proposed for a NiIJOH)2/
Fe2O3 electrode.40,41
Fig.
5
(A) Photoluminescence spectra of the NiP/TiO2 samples
containing 0–10 wt% NiP, as indicated by the legends. (B) Nyquist plots
of (a) TiO2 and (b) 3.0 NiPi/TiO2, measured under N2 in 0.5 M NaClO4.
The inset is an equivalent circuit model of the simulated curves, where
Rs is the series resistance, Rct is the charge transfer resistance, and Csc
is the bulk capacitance.
Fig. 4C shows the result of water oxidation at a potential
of 0.6 V vs. AgCl/Ag. During five repeated experiments, the
photocurrent of 3.0 NiP/TiO2 was always higher than that of
TiO2. Besides that, there were some differences between the
two electrodes. When the light was turned on, the
photocurrents of TiO2 and 3.0 NiP/TiO2 slowly and quickly
increased, respectively. Since photocurrent originates from
electrons and holes, such a spike is indicative of fast Ni2+
oxidation, which improves the efficiency of charge
separation. Then, equilibrium is reached between the charge
generation, recombination and consumption, resulting in a
steady photocurrent. When the light was turned off, the
current quickly decreased with time. This is due to the
remaining charge carriers recombining and/or reacting with
sorbates. Meanwhile, NiP/TiO2 showed a faster decay of
current with time, as compared with TiO2. It implies that the
reduction of Ni3+/4+ to Ni2+ by the remaining electrons is
faster than the reduction of protons to H2. Moreover, at the
first and fifth run, the steady photocurrents of TiO2 were 19.7
and 19.2 A cm−2, respectively, while the first and fifth
325 nm light under air, several emission peaks were observed
(Fig. 5A). The peak at 3.14 eV (395 nm) well matches that of
the band-to-band transition of anatase TiO2 (Eg = 3.22 eV).
Other peaks at 2.53–2.76 eV are due to electron
recombination with the trapped holes within the band gap.42
With NiP/TiO2, there were also several peaks. Although the
peak positions were nearly the same as those of TiO2, the
peak intensity, especially at 3.14 eV, decreased as the NiP
loading was increased. The reduced PL intensity due to the
increased electron transfer to O2 is less likely, because O2
adsorption on TiO2 is inhibited by NiP (Fig. 4A). Then, the
reduced PL intensity is indicative of the increased hole
transfer from TiO2 to NiPs, retarding the hole recombination
with electrons, and hence weakening all emissions. In the
Nyquist plots of EIS, there were semicircles (Fig. 5B), but NiP/
TiO2 showed a radius smaller than TiO2. By using an
equivalent circuit model, the charge transfer resistance was
estimated, which was 55.3 Ω for TiO2 and 3.4 Ω for NiP/TiO2.
It means that the charge transfer from 3.0 NiP/TiO2 to species
in the electrolyte is more efficient than that from TiO2.41,43,44
In aqueous solutions at pH 0 and pH > 5, the one-
electron reduction potentials of O2 are −0.05 and −0.33 V,
respectively, versus NHE (normal hydrogen electrode).45 These
potentials are more positive than the conduction band edge
potential for anatase TiO2, Ecb (V vs. NHE) = −0.12–0.059
pH.46 Then, the one-electron reduction of O2 on TiO2 can
occur at any pH, as reviewed by Nosaka.47 Meanwhile, the
currents of 3.0 NiP/TiO2 were 32.2 and 30.2
A
cm−2,
respectively. These observations indicate that NiP/TiO2 is
more photoactive than TiO2 for water oxidation, all of which
are all quite stable against corrosion.
3.5. Possible mechanism
To understand the role of NiP, the solid was further
characterized with PL and EIS. After TiO2 was excited with
Catal. Sci. Technol.
This journal is © The Royal Society of Chemistry 2020

View Article Online
Catalysis Science & Technology
Paper
and the rate of the surface reaction is increased. This
work provides
semiconductor
a
facile strategy for improvement of
photocatalysis, as well as in
photoelectrochemistry.
Conflicts of interest
The authors declare no competing financial interest.
Scheme 1 Possible mechanisms for the positive effect of NiP on TiO2
photocatalysis under aerobic (pink region) and anaerobic (blue region)
conditions.
Acknowledgements
This work was supported by the Funds for Creative Research
Group of NSFC (No. 21621005).
reduction of protons to H2 on TiO2 is also allowed at any pH,
but the driving force for H2 reduction is smaller than that for
O2 reduction (Fig. S10†).
Notes and references
1 T. Kawai and T. Sakata, Nature, 1980, 286, 474–476.
2 R. W. Matthews, J. Phys. Chem., 1987, 91, 3328–3333.
3 J. Schneider, M. Matsuoka, M. Takeuchi, J. Zhang, Y.
Horiuchi, M. Anpo and D. W. Bahnemann, Chem. Rev.,
2014, 114, 9919–9986.
4 A. V. Emeline, X. Zhang, M. Jin, T. Murakami and A.
Fujishima, J. Phys. Chem. B, 2006, 110, 7409–7413.
5 A. Meng, L. Zhang, B. Cheng and J. Yu, Adv. Mater.,
2019, e1807660.
6 M. Wang, J. Han, Y. Hu, R. Guo and Y. Yin, ACS Appl. Mater.
Interfaces, 2016, 8, 29511–29521.
7 Y. Zhang, M. Gong, X. Liu, L. Ji, Z. Yang and X. Zhu,
J. Mater. Sci., 2019, 54, 2975–2989.
In aqueous solutions at pH values of 0 and 14, the Ni2+/Ni
couple has standard redox potentials of −0.25 and −0.72 V vs.
NHE, respectively. Then, the reduction of Ni2+ to Ni0 on TiO2
is not allowed at pH 0, but it is possible at pH > 2.2 (Fig.
S10†). In the experiment, the formation of metallic Ni was
observed from an irradiated pre-degassed aqueous
suspension of NiP/TiO2 (pH 8). Under air, however, Ni2+
reduction is not competitive with O2 reduction at pH > 7.4.
As a result, no Ni0 was observed from the irradiated aqueous
suspension of NiP/TiO2 (pH 8) under air. Similarly, Co2+
reduction occurs only in the absence of O2 at pH > 2.8 (Fig.
S10†). Since CoP and NiP have similar water solubility
products, and their metal ions have similar reduction
potentials as well, they show similar activities for H2
production.
8 J. S. Lee and W. Y. Choi, J. Phys. Chem. B, 2005, 109,
7399–7406.
9 X. Xiong and Y. Xu, J. Phys. Chem. C, 2016, 120, 3906–3912.
10 J. Xing, Y. Li, H. Jiang, Y. Wang and H. Yang, Int. J. Hydrogen
Energy, 2014, 39, 1237–1242.
Under anaerobic conditions, metallic Ni0 and Co0 can
catalyze proton reduction, respectively, as observed from Ni0
or Co0 pre-deposited on TiO2.48–50 Under aerobic conditions,
the electrons of TiO2 prefer to be captured by O2. Then, the
counterpart holes are used to oxidize Ni2+ in NiP/TiO2, or to
oxidize Co2+ in CoP/TiO2. The resulting Ni3+/4+ species then
oxidize phenol to phenolic radicals, followed by hydrolysis to
HQ and BQ. Meanwhile, Ni2+ or NiP is regenerated for
subsequent use. That is, in TiO2 photocatalysis, NiP plays
different roles under anaerobic and aerobic conditions
(Scheme 1).
11 L. Qi, J. Yu and M. Jaroniec, Phys. Chem. Chem. Phys.,
2011, 13, 8915–8923.
12 R. S. Khnayzer, M. W. Mara, J. Huang, M. L. Shelby, L. Chen
and F. N. Castellano, ACS Catal., 2012, 2, 2150–2160.
13 G. Ai, R. Mo, H. Li and J. Zhong, Nanoscale, 2015, 7, 6722–6728.
14 K. J. McDonald and K. S. Choi, Chem. Mater., 2011, 23,
1686–1693.
15 L. Ge, C. Han, X. Xiao and L. Guo, Appl. Catal., B, 2013, 142,
414–422.
16 T. Di, B. Zhu, J. Zhang, B. Cheng and J. Yu, Appl. Surf. Sci.,
2016, 389, 775–782.
17 X. Zhang, X. Xiong, L. Wan and Y. Xu, Langmuir, 2017, 33,
8165–8173.
18 R. Chong, B. Wang, D. Li, Z. Chang and L. Zhang, Sol.
Energy Mater. Sol. Cells, 2017, 160, 287–293.
19 J. Wang, J. Yang, Z. Zheng, T. Lu and W. Gao, Appl. Catal., B,
2017, 218, 277–286.
20 J. Yu, Y. Hai and B. Cheng, J. Phys. Chem. C, 2011, 115,
4953–4958.
4. Conclusion
In this work, the surface of TiO2 is modified with
amorphous NiP. This results in a greatly improved activity
for organic degradation, proton reduction and water
oxidation, respectively. However, the roles of NiP in these
reactions are quite different. Under anaerobic conditions,
Ni2+ ions in NiP are reduced by the electrons of TiO2 to
form metallic Ni0, while under aerobic conditions, Ni2+
ions in NiP are oxidized by the holes of TiO2 to form a
Ni3+/4+ species. These pathways are also applicable to CoP
deposited on TiO2. Due to the NiP-mediated charge
transfer, the efficiency of charge separation is improved,
21 W. Wang, S. Liu, L. Nie, B. Cheng and J. Yu, Phys. Chem.
Chem. Phys., 2013, 15, 12033–12039.
22 Y. Wang, J. Zhao, X. Xiong, S. Liu and Y. Xu, Appl. Catal., B,
2019, DOI: 10.1016/j.apcatb.2019.117903.
This journal is © The Royal Society of Chemistry 2020
Catal. Sci. Technol.

View Article Online
Paper
Catalysis Science & Technology
23 S. V. Mattigod, D. Rai, A. R. Felmy and L. F. Rao, J. Solution
Chem., 1997, 26, 391–403.
37 H. Dang, X. Dong, Y. Dong, H. Fan and Y. Qiu, Mater. Lett.,
2015, 138, 56–59.
24 H. Bader, V. Sturzenegger and J. Hoigne, Water Res.,
1988, 22, 1109–1115.
38 H. Xu, S. Li, L. Ge, C. Han, Y. Gao and D. Dai, Int. J.
Hydrogen Energy, 2017, 42, 22877–22886.
25 C. Lepetit and M. Che, J. Phys. Chem., 1996, 100, 3137–3143.
26 J. Yu, A. Wang, J. Tan, X. Li, J. A. van Bokhoven and Y. Hu,
J. Mater. Chem., 2008, 18, 3601–3607.
27 D. F. Ollis, J. Phys. Chem. B, 2005, 109, 2439–2444.
28 A. V. Emeline, V. K. Ryabchuk and N. Serpone, J. Phys. Chem.
B, 2005, 109, 18515–18521.
39 J. Yang, H. W. Liu, W. N. Martens and R. L. Frost, J. Phys.
Chem. C, 2010, 114, 111–119.
40 G. Wang, Y. Ling, X. Lu, T. Zhai, F. Qian, Y. Tong and Y. Li,
Nanoscale, 2013, 5, 4129–4133.
41 Q. Liu, F. Cao, F. Wu, H. Lu and L. Li, Adv. Mater. Interfaces,
2016, 3, 1600256.
29 K. Sahel, L. Elsellami, I. Mirali, F. Dappozze, M. Bouhent
and C. Guillard, Appl. Catal., B, 2016, 188, 106–122.
30 H. Sheng, Q. Li, W. Ma, H. Ji, C. Chen and J. Zhao, Appl.
Catal., B, 2013, 138–139, 212–218.
42 D. K. Pallotti, L. Passoni, P. Maddalena, F. Di Fonzo and S.
Lettieri, J. Phys. Chem. C, 2017, 121, 9011–9021.
43 J. Ângelo, P. Magalhães, L. Andrade and A. Mendes, Appl.
Surf. Sci., 2016, 387, 183–189.
31 J. Kim and W. Choi, Appl. Catal., B, 2011, 106, 39–45.
32 H. H. Mohamed, C. B. Mendive, R. Dillert and D. W.
Bahnemann, J. Phys. Chem. A, 2011, 115, 2139–2147.
33 R. Qian, H. Zong, J. Schneider, G. Zhou, T. Zhao, Y. Li, J.
Yang, D. W. Bahnemann and J. Pan, Catal. Today, 2019, 335,
78–90.
34 S. Yu, Z. Li, X. Fan, J. X. Li, F. Zhan, X. Li, Y. Tao, C. Tung
and L. Wu, ChemSusChem, 2015, 8, 642–649.
35 M. C. Biesinger, B. P. Payne, A. P. Grosvenor, L. W. M. Lau,
A. R. Gerson and R. S. Smart, Appl. Surf. Sci., 2011, 257,
2717–2730.
44 S. Singh and N. Khare, Sci. Rep., 2018, 8, 6522.
45 D. T. Sawyer and J. S. Valentine, Acc. Chem. Res., 1981, 14,
393–400.
46 D. Duonghong, J. Ramsden and M. Gratzel, J. Am. Chem.
Soc., 1982, 104, 2977–2985.
47 Y. Nosaka and A. Y. Nosaka, Chem. Rev., 2017, 117,
11302–11336.
48 P. Wang, S. Xu, F. Chen and H. Yu, Chin. J. Catal., 2019, 40,
343–351.
49 H. Tian, S.-Z. Kang, X. Li, L. Qin, M. Ji and J. Mu, Sol. Energy
Mater. Sol. Cells, 2015, 134, 309–317.
36 M. C. Biesinger, B. P. Payne, L. W. M. Lau, A. Gerson and
R. S. C. Smart, Surf. Interface Anal., 2009, 41, 324–332.
50 P. D. Tran, L. F. Xi, S. K. Batabyal, L. H. Wong, J. Barber and
J. S. C. Loo, Phys. Chem. Chem. Phys., 2012, 14, 11596–11599.
Catal. Sci. Technol.
This journal is © The Royal Society of Chemistry 2020