Low temperature synthesis and characterization of substitutional Na-modified K2Ti6O13 nanobelts with improved photocatalytic activity under UV irradiation

Article abstract of DOI:10.1039/c5ra10640k

The substitutional Na-modified K₂Ti₆O₁₃ (NKTO) nanobelts were synthesized by a facile one-step method from NaCl-KCl melts at a relatively low temperature below 700°C. The obtained samples were fully characterized by X-ray diffraction (XRD), energy dispersive spectrometer (EDS), scanning electron microscopy (SEM), transmission electron microscopy (TEM), high-resolution transmission electron microscopy (HRTEM), selected area electron diffraction (SAED), thermogravimetry and differential scanning calorimetry (TG-DSC), N₂ adsorption-desorption analysis, UV-Vis diffuse reflectance spectroscopy and photoluminescence (PL) spectroscopy. The results revealed the highly crystalline NKTO nanobelts obtained above 680°C exhibited not only a narrow distribution of particle size but good dispersivity. The replacement of K⁺ by Na⁺ ions reduced the interplanar spacing of K₂Ti₆O₁₃ but did not destroy its oriented growth along [010]. From these results, the formation mechanism of the NKTO nanobelts was proposed. The specific surface area of the nanobelts showed a decrease with the increase of the synthetic temperature. Different titania phases were found to have a significant effect on the formation of NKTO nanobelts and rutile was the optimum titanium precursor. Substitutional Na-modification was demonstrated to not only red-shift the absorption edge and increase UV absorption but also to efficiently suppress the recombination of photogenerated electron-hole pairs. The photocatalytic tests showed the substitutional Na modification and lower synthetic temperature greatly promoted the activity for photodegradation of methyl orange under UV irradiation and the improved photocatalytic mechanism was also discussed.

Full text of DOI:10.1039/c5ra10640k

RSC Advances
View Article Online
View Journal | View Issue
PAPER
Low temperature synthesis and characterization of
substitutional Na-modified K Ti O nanobelts with
2
6 13
Cite this: RSC Adv., 2015, 5, 66086
improved photocatalytic activity under UV
irradiation†
Qiang Wang, Qingjun Guo and Bing Li*
The substitutional Na-modified K
2
Ti
6
O13 (NKTO) nanobelts were synthesized by a facile one-step method
ꢀ
from NaCl–KCl melts at a relatively low temperature below 700 C. The obtained samples were fully
characterized by X-ray diffraction (XRD), energy dispersive spectrometer (EDS), scanning electron
microscopy (SEM), transmission electron microscopy (TEM), high-resolution transmission electron
microscopy (HRTEM), selected area electron diffraction (SAED), thermogravimetry and differential
scanning calorimetry (TG-DSC),
N
2
adsorption–desorption analysis, UV-Vis diffuse reflectance
spectroscopy and photoluminescence (PL) spectroscopy. The results revealed the highly crystalline
ꢀ
NKTO nanobelts obtained above 680 C exhibited not only a narrow distribution of particle size but
+
+
2 6
good dispersivity. The replacement of K by Na ions reduced the interplanar spacing of K Ti O13 but did
not destroy its oriented growth along [010]. From these results, the formation mechanism of the NKTO
nanobelts was proposed. The specific surface area of the nanobelts showed a decrease with the increase
of the synthetic temperature. Different titania phases were found to have a significant effect on the
formation of NKTO nanobelts and rutile was the optimum titanium precursor. Substitutional Na-
modification was demonstrated to not only red-shift the absorption edge and increase UV absorption
but also to efficiently suppress the recombination of photogenerated electron–hole pairs. The
photocatalytic tests showed the substitutional Na modification and lower synthetic temperature greatly
promoted the activity for photodegradation of methyl orange under UV irradiation and the improved
photocatalytic mechanism was also discussed.
Received 4th June 2015
Accepted 28th July 2015
DOI: 10.1039/c5ra10640k
www.rsc.org/advances
crystalline, chemically puried, single-phase and non-agglom-
erated powders at lower calcination temperature than the
1
. Introduction
Alkali metal hexatitanates, M Ti O (M ¼ Li, Na, K, Rb, Cs), conventional solid-state reaction and always in overall shorter
2
6 13
3
1,32
with unique tunnel or layer crystalline structures have attracted reaction time as compared with hydrothermal synthesis.
increasing interest due to their excellent thermal durability,
ꢀ
However, a heating temperature above 900 C is usually
chemical resistivity, mechanical performance, ion-exchange needed to obtain highly crystalline one-dimensional alkali-
1
–11
properties and photocatalytic activity.
and Na Ti
13 have been studied to be adequate for the from molten chloride salts. Sometimes, in order to favor
degradation of toxic substances,
2 6
Especially, K Ti O
2 6 2 6
13 metal hexatitanates (such as K Ti O13 and Na Ti O13) crystals
2
6
O
10–14
decomposition of pure synthesizing reaction and improve uniformity of the brous
as well as for oxygen electrodes in potentiometric products, a nonionic surfactant like NP-9 (nonylphenol poly-
1
5–17
water
sensors owing to their better photocatalytic properties under ethylene glycol ether) will be introduced into the molten
18
2
8,33,34
UV irradiation. They are traditionally prepared by calcination salts,
which increases the production cost. Therefore,
19–21
22–25
14,26
ꢀ
method,
and molten salt synthesis (MSS).
methods, MSS has been proved to be one of the simplest, most further. Moreover, the obtained single-phase K
hydrothermal synthesis,
sol–gel technique
developing a facile and low-temperature (<900 C) MSS of well-
27–30
Among these synthesis crystallized brous hexatitanates is still worth considering
Ti 13 products
O
6
2
versatile, and cost-effective approaches available for obtaining can't perform as well as expected in photocatalytic reactions due
to its wide band gap, high recombination rate of photo-
generated electron–hole pairs and small specic surface area.
School of Mechanical and Power Engineering, East China University of Science and
Technology, 130 Meilong Road, Shanghai 200237, China. E-mail: drlibing@163. As a result, a great effort has been devoted to improve the
com; Tel: +86-21-64252601
2 6
photocatalytic activity of K Ti O13. For example, Tretyachenko
et al. reported nanostructured potassium titanates modied
†
Electronic supplementary information (ESI) available. See DOI:
0.1039/c5ra10640k
1
66086 | RSC Adv., 2015, 5, 66086–66095
This journal is © The Royal Society of Chemistry 2015

View Article Online
Paper
RSC Advances
using transition metals exhibit a higher photocatalytic activities 2.2 Sample preparation
35
in the ultraviolet and visible ranges. Amano et al. prepared
In a typical procedure, K CO and rutile TiO were mixed with
2
3
2
Pt-loaded nanowire-structured potassium titanate to improve
NaCl–KCl eutectic salts (50 mol% NaCl + 50 mol% KCl) at a
molar ratio of 1 : 5.5 : 110, and ground homogeneously in an
agate mortar under ethyl alcohol for 30 min. The obtained
mixture were dried and transferred into a corundum crucible.
36
photocatalytic activity. Several researchers investigated the
photocatalytic properties of titanates containing different
9,37
amounts of alkaline ions.
Park also reported that the
hydroxyapatite precipitation could enhance the photocatalytic
ꢀ
ꢀ
ꢀ
Aer that, the crucible was heated to 660 C (680 C, 700 C)
10
activity of K
2
Ti
6
O
13 whiskers. These modication methods
ꢀ
ꢁ1
with a heating rate of 5 C min and held for 3 h at the required
temperature and then cooled naturally to room temperature.
The resulted products were washed with deionized water
repeatedly to remove the residual NaCl–KCl and nally dried
above oen include two steps, i.e., preparation of pure titanates
and the following modied treatments.
Until now, however, no reports on one-step synthesis of Na-
modied K Ti O (referred hereaer as NKTO) nanostructures
2
6
13
ꢀ
overnight in air at 100 C. The obtained products were named as
with improved photocatalytic activity via MSS method below
NKTO-660, NKTO-680 and NKTO-700, respectively.
ꢀ
7
00 C have been published. The NaCl–KCl eutectic salts with
In order to study the effect of different titania phases on the
ꢀ
ꢀ
lower melting point than KCl (657 C vs. 771 C) can be used as
solvent, which contributes to the growth of smaller particle size,
i.e. larger surface area. Such a large surface area is favorable to
the enhanced photocatalytic properties, because it can promote
adsorption, desorption and diffusion of reactants and products.
It was reported that the activity of hexatitanates photocatalytic
water-splitting was associated with their efficiency to produce
photoexcited charge carriers and at the same time limited by the
formation of products, anatase TiO
2
was selected as titanium
ꢀ
ꢀ
precursor to synthesize the samples at 660 C (not given), 680 C
ꢀ ꢀ ꢀ
(
not given), 700 C, 800 C (NKTO-800), 900 C (NKTO-900) for
3
h by the same procedure mentioned above. The samples
ꢀ
synthesized at 900 C from KCl melt (KTO-900) and from NaCl
melt (NTO-900) were also obtained as the referential materials.
17
2.3 Characterizations
recombination rate of photoexcited electron–hole pairs.
However, during the growth of K Ti 13 crystals in the NaCl–
KCl molten salts, K ions in the tunnel structures can be
2
6
O
The crystalline phase and crystal lattice parameters of the as-
synthesized samples were determined by XRD (D/max 2550 V)
+
+
partially replaced by smaller Na ions from the melts to product
˚
with Cu-Ka radiation (l ¼ 1.54 A). Morphology, chemical
38
NKTO nanobelts, which may change pristine crystal and band
structures and so provides a great potential driving force to
suppress the recombination of photogenerated electron–hole
pairs, thus resulting in the enhancement of photocatalytic
activity.
compositions and microstructure of the samples were charac-
terized by SEM (JEOL JSM-6360LV), TEM (JEM-2100, 200 kV)
equipped with EDS (VARIO ELIII). HRTEM images and SAED
patterns were also obtained on the JEM-2100 microscope
operated at an acceleration voltage of 200 kV to investigate the
crystallinity and growth direction of samples. TG-DSC analysis
was measured by an STA 449F3 thermal analyzer (Netzsch,
Germany) to study the formation mechanism of NKTO nano-
belts. The optical properties of the samples are studied by UV-
Vis diffuse reectance spectra using an ultraviolet-obviously-
near-infrared spectrophotometer (UV-VIS-NIR, American Varian
Cary 500) and PL spectrum measured at room temperature with
a uorescence spectrouorometer (Jobin Yvon Inc. Fluorolog-3).
The specic surface area of the samples were deduced by the
BET method using a Macromeritics ASAP2460 apparatus.
Here, we reported for the rst time, that the NKTO nano-
ꢀ
belts were synthesized by one-step MSS method below 700 C
increasing the specic surface area of the products and
+
successfully incorporating Na into K Ti O . The crystal
2
6 13
structure, chemical composition, morphology and optical
properties of the NKTO nanobelts were systematically char-
acterized by XRD, EDS, SEM, TEM, SAED, HRTEM, TG-DSC, N
2
adsorption–desorption analysis and UV-Vis diffuse reectance
spectroscopy combined with PL spectra. Based on the results,
the formation mechanism of NKTO nanobelts was proposed.
The effect of different titania phases on the formation of
NKTO nanobelts was also investigated. The improved photo-
catalytic activity of the synthesized samples was evaluated by
2.4 Photocatalytic activity tests
photodegradation of MO solution under UV irradiation and The photocatalytic activity of the as-synthesized samples was
correspondingly the photocatalytic mechanism was also evaluated by photodegradation of MO solution (300 mL, 10
discussed.
ppm) placed in a glass in the presence of 0.3 g as-synthesized
samples under irradiation of a TUV8W/G8/T5 UV lamp at room
temperature. Prior to UV irradiation, the suspension is
magnetically stirred in dark for 1 h to reach absorption–
desorption equilibrium. At certain time intervals, a 10 mL
aqueous suspension was taken with syringe lter unit (Milli-
2. Experimental
2.1 Materials
The starting materials used in this study included rutile and pore, pore size 0.22 mm) and centrifuged to remove the photo-
anatase TiO nanoparticles (50 nm), K CO , Na CO (AR, Xiya catalyst particles which could be recycled. The variations of
2
2
3
2
3
Reagent, Ltd), KCl and NaCl (AR, Sinopharm Chemical Reagent characteristic absorbance intensity in the UV-Vis absorption
Co., Ltd). All the materials were used without any further spectra of MO at l ¼ 465 nm was monitored on a ultraviolet-
purications.
visible-near infrared spectrophotometer (American Varian Cary
This journal is © The Royal Society of Chemistry 2015
RSC Adv., 2015, 5, 66086–66095 | 66087

View Article Online
RSC Advances
Paper
500) to evaluate the photocatalytic degradation efficiency (h).
Herein, h was calculated by the following formula:
ꢀ
ꢁ
ꢀ
ꢁ
C
0
ꢁ C
t
A
0
ꢁ A
t
h ¼
ꢂ 100% ¼
ꢂ 100%
(1)
C
0
A
0
where C is the concentration of MO solution at absorption–
0
desorption equilibrium, C
tion; A and A are the corresponding absorbance values,
respectively.
t
is the time dependent concentra-
0
t
3. Results and discussion
3.1 Structural and morphological characterizations
Fig. 1 (combined with Fig. S1†) shows the XRD patterns of the
samples synthesized from NaCl–KCl molten salts at different
synthetic temperature, which clearly reveals they can be
indexed to K
C2/m(12), JCPDS Card no. 74-0275), without any noticeable
peaks belonging to the unreacted TiO or Na Ti 13 (Fig. S2†),
indicating the starting materials (TiO , K CO
almost transformed into K Ti
2 6
Ti O13 phase (monoclinic structure, space group
2
2
6
O
2
2
3
) have been
13 products even at 660 C and
Fig. 1 XRD patterns of the as-synthesized samples: (a) NKTO-660, (b)
NKTO-680 and (c) NKTO-700 and the bar chart shows the standard
ꢀ
2
6
O
TiO does not react with NaCl melt yielding Na Ti O phase XRD pattern of K
2
Ti
6 2 6 13
O13 (JCPDS Card no. 74-0275) and Na Ti O
2
2
6 13
(
JCPDS Card no. 73-1398).
under current synthetic conditions. The sharp and intense
peaks indicate that the samples are highly crystalline. It is
worth noting that the diffraction peaks of these samples
present a slight shi to higher 2q position compared to the
The crystal structure of the obtained nanobelts were further
standard XRD pattern of K
the fact that the potassium ions in tunnel crystalline structure sample NKTO-680 showing the products consist of uniform
of K Ti
13 have been locally replaced by smaller sodium ions nanobelts with a at and smooth surface. Fig. 4b shows a high
from NaCl–KCl melts, in other words, the decrease of inter- magnication TEM image of an individual nanobelt with the
2 6
Ti O
13, which can be attributed to characterized with TEM. Fig. 4a is a typical TEM image of the
2
6
O
+
+
planar spacing results from the ion replacement of K by Na . width of about 110 nm and the SAED pattern (Fig. 4c) taken on it
This is also conrmed by the EDS analysis. As shown in Fig. 2, indicates the NKTO nanobelt has a preferential growth along
the representative EDS spectra of NKTO-680 reveals that a [010] crystallographic direction, which is in good agreement
certain amount of sodium element is detected. In Table 1, the with K
chemical composition of the as-synthesized samples show the modifying do not destroy the oriented growth of K
2
Ti
6
O
13 and also further suggests the substitutional Na
Ti
The streaking between the diffraction spots
13 aver- shown in SEAD pattern may be attributed to a small amount o+f
agely, i.e., the Na ions incorporated into the crystal lattice of crystal defect in the nanobelts result from the replacement of K
K Ti O are not interstitial but substitutional, which is
2
6 13
O
5,29
molar ratio of (Na + K)/Ti for these samples are all close to nanobelts.
: 3, dening the chemical formula of Na0.88
1
6
K1.12Ti O
+
2
6 13
consistent with the result of XRD analysis. We can conclude
that, therefore, highly crystalline substitutional Na-modied
ꢀ
K
2
Ti
6
O
13 crystals have been synthesized directly below 700 C
from NaCl–KCl melts.
The morphology and dimensions of as-synthesized NKTO
products were characterized by SEM. It can be seen in Fig. 3a
combined with Fig. 3b that the sample NKTO-660 mainly
consists of nonuniform and agglomerated belt-like products
mixed with some particles, which suggests the heating
temperature is so low that NaCl–KCl melts can not provide an
enough homogeneous liquid phase for crystal growth. As the
ꢀ
temperature increasing to 680 C, a large quantity of well-
formed nanobelts with a typical thickness of 100–200 nm and a
length in excess of 5 mm are observed in Fig. 3c. The brous
morphology of NKTO can be attributed to its unique rectangular
tunnel crystal structure. Similar to Fig. 3c, Fig. 3d shows that the
non-agglomerated NKTO nanobelts are obtained at 700 C and
the size distribution is relatively narrow.
Fig. 2 EDS spectra of the sample NKTO-680. The presence of Al peak
likely originates from the contamination of alumina crucibles during
the synthesis.
ꢀ
66088 | RSC Adv., 2015, 5, 66086–66095
This journal is © The Royal Society of Chemistry 2015

View Article Online
Paper
RSC Advances
Table 1 Chemical composition of the as-synthesized NKTO nanobelts samples
Samples
Na (at%)
K (at%)
Ti (at%)
O (at%)
Al (at%)
C (at%)
(Na + K)/Ti
NKTO-660
NKTO-680
NKTO-700
4.54
4.96
4.38
5.48
5.64
6.08
30.79
32.04
31.07
54.78
53.28
54.31
0.28
0.29
0.31
4.13
3.79
3.85
0.33
0.32
0.33
+
ions by smaller Na ions. The corresponding HRTEM image in 3.2 The formation mechanism of NKTO nanobelts
Fig. 4d shows that the lattice fringe of this nanobelt is examined
Based on the above characterization results and previous
related studies, a possible formation mechanism of NKTO
nanobelts was proposed as illustrated in Fig. 6. Firstly, the
to be 0.186 nm, which is smaller than the standard interplanar
spacing value (0.191 nm) of (020) planes of K Ti O , consistent
39
2
6 13
with the result of XRD. Therefore, the crystal structure of the as-
synthesized NKTO nanobelts can be illustrated as Fig. 5.
starting materials (rutile TiO
evenly so that the reactants including nanosized TiO
CO particles are well dispersed in the NaCl–KCl salts used as
2
, K
2
CO
3
, KCl–NaCl) are mixed
2
and
K
2
3
reaction medium (Fig. 6a). Next, in Fig. 6b, when the calcination
temperature rises higher than the eutectic point of NaCl–KCl
ꢀ
(
657 C), NaCl–KCl salts become into the homogeneous liquid
melts and simultaneously K CO begin to be dissolved in it,
2
3
while TiO
2
can agglomerate with each other and grow to form
has no solubility in molten chlo-
rides. Then as shown in Fig. 6c, the molten K CO diffusing
ions forms will react
bulks at the solid–liquid interface to yield K Ti
bigger rod-like bulks as TiO
2
34
2
3
+
2ꢁ
3
randomly in the melts as K and CO
with TiO
2
2
6 13
O
nuclei (as the templates for further grow) and CO gas, which
2
can be described by the following formula:
K
2
CO
3
+ 6TiO
2
/ K
2
6
Ti O
13 + CO
2
[
(2)
40
According to the dissolution–precipitation mechanism,
with the continuous precipitation and consumption of K CO3
2
on the templates, the reaction (2) proceeds continually and
Fig. 3 SEM images of the as-synthesized samples: (a and b) NKTO-
naturally the K
2 6
Ti O13 nuclei grow further along the preferential
6
60, (c) NKTO-680, (d) NKTO-700.
crystallization direction gradually to form one-dimensional
+
nanocrystals. During this process, the K ions in K
2
Ti
6 13
O
+
nanostructures will be replaced by the smaller Na ions from
Fig. 4 (a) A typical TEM image of the sample NKTO-680. (b) A high
magnification TEM image of an individual NKTO-680 nanobelt. (c) The Fig. 5 Crystal structure of the as-synthesized NKTO nanobelts along
24
+
+
selected area electron diffraction (SAED) pattern of the individual the [010] direction.
K and the substitutional Na ions are located
nanobelt (shown in (b)) recorded along the [001] zone axis. (d) The into the rectangular tunnel structures consisting of adjacent [TiO
6
]
HRTEM image corresponding to SEAD pattern.
octahedra.
This journal is © The Royal Society of Chemistry 2015
RSC Adv., 2015, 5, 66086–66095 | 66089

View Article Online
RSC Advances
Paper
Fig. 6 (a–d) Schematic illustrations of the formation mechanism of NKTO nanobelts in NaCl–KCl melts.
the NaCl–KCl melts. In such way, a large quantity of well-crys- suggesting the presence of mesoporous structures. The hyster-
tallized and uniform NKTO nanobelts are obtained nally esis loop in the low pressure range (0.5 < P/P < 0.9) is associated
Fig. 6d). It should be mentioned that the large amount of NaCl– with the intra-aggregated pores made by the agglomerate of
0
(
KCl melts provided a desired environment for the growth of nanobelts and particles observed in SEM images. The high-
NKTO nanocrystals and at the same time prevented the pressure hysteresis loop (0.9 < P/P < 1) is related to the larger
0
43
agglomeration of the formed NKTO nanobelts.
pores formed between secondary particles.
While the
Then as shown in Fig. 7, the TG-DSC curves for NKTO isotherms of both NKTO-680 and NKTO-700 can be attributed to
precursor was used to support the formation mechanism type II, which indicates the presence of macroporous structures,
proposed above. The TG results suggest the sample loses weight i.e., the nanobelts are well dispersive. The BET specic surface
ꢀ
about 1.8% slowly from 60 to 90 C corresponding to the area (SBET) of samples (NKTO-660/680/700) are calculated to be
ꢀ
2
ꢁ1
2
ꢁ1
2
ꢁ1
endothermic peak at 84.7 C in the DSC curve, which may be 16.53 m
g
, 13.87 m
g
and 13.02 m
g
respectively,
due to volatilization of physically absorbed water. As the highlighting the decrease of surface area with the increasing of
ꢀ
temperature increases from 650 to 1000 C, the sample exhibits synthetic temperature. This decreased SBET is due to the fact
further and sharper weight loss about 67.1%, which is mainly
due to the reaction (2) releasing CO
2
and the evaporation of
NaCl–KCl molten salts supported by the endothermic peak at
ꢀ
41
9
29.2 C in the DSC curve. Note that the sharp endothermic
ꢀ
peak of DSC at 664.5 C corresponds to not only the melting of
ꢀ
NaCl–KCl salts (657 C) and the occurring of reaction (2) but
+
also the ions exchange of K in the tunnel crystal structure of
+
K Ti O by Na ions from the molten salts.
2
6 13
3.3
2
N adsorption–desorption isotherm analysis
In order to characterize the specic surface areas and the
porosity of the as-synthesized NKTO nanobelts, the nitrogen
adsorption–desorption measurements were carried out and the
results were shown in Fig. 8. According to the Brunauer–
42
Deming–Deming–Teller (BDDT) classication, the isotherm of
NKTO-660 can be nearly categorized as type IV with small H
type hysteresis loop observed in the range (P/P
) of 0.5–1.0, Fig. 7 (a–d) TG-DSC curves for the NKTO precursor.
3
0
66090 | RSC Adv., 2015, 5, 66086–66095
This journal is © The Royal Society of Chemistry 2015

View Article Online
Paper
RSC Advances
Fig. 10 (a) UV-Vis diffuse reflection spectra of the as-synthesized
NKTO nanobelts; (b) UV-Vis diffuse reflection spectra for KTO-900,
NTO-900, NKTO-900, NKTO-800. The inset shows the enlarged UV-
Vis spectra.
Fig. 8 The nitrogen adsorption–desorption isotherms of the as-
synthesized samples.
that the higher synthetic temperature promotes the crystal
growth of NKTO leading to the formation of bigger particles,
and compared with 1-dimensional nanobelts, 0-dimensional
nanoparticles (in NKTO-660) have higher surface-to-volume
ratio, which means larger surface area.
NKTO, specially, the diffraction intensity from anatase phase is
higher compared with that of NKTO phase, indicating the
amount of unreacted anatase is relatively large. With the
increasing of synthetic temperature, the anatase phase can
ꢀ
change into NKTO completely until 800 C and the resulted
ꢀ
single-phase NKTO is stable at 900 C. These observations
3
.4 Effects of different titania phases on the formation of
clearly suggests that anatase TiO
2
isn't favorable to low
ꢀ
NKTO from NaCl–KCl melts
temperature (<700 C) synthesis of NKTO nanobelts from NaCl–
KCl melts, in other words, rutile is the optimum titanium
precursor.
Aiming to identify the effect of different titania phases on the
formation of NKTO, the XRD patterns of the samples synthe-
sized using anatase as Ti precursor are shown in Fig. 9 (and
Fig. S3†). The results reveal that the products obtained at 700 C
2
consist of two kinds of crystalline phases, i.e., anatase TiO and
Besides, it can be seen that the synthesis preformed in KCl
ꢀ
ꢀ
melt at 900 C have yielded a pure K
2
6
Ti O13 (JCPDS Card no.
74-0275) phase and single-phase Na
2
Ti
6
O
13 (JCPDS Card no.
ꢀ
73-1398) has been also obtained from NaCl melt at 900 C. It is
worth noting that the diffraction peaks for NKTO-900 become
sharper compared with KTO-900, suggesting an effective
enhancement of crystallization and grain growth in NaCl–KCl
41
molten salts.
3
.5 Optical properties
As shown in Fig. 10a, the optical properties were investigated
rstly through UV-Vis diffuse reection spectra. From these
spectra, the band gap energy (E ) of the as-synthesized samples
can be calculated using the equation:

g
1
)
/2
ahn ¼ A(hn ꢁ E
g
(3)
where a is the absorption coefficient, hn is the photon energy
44
and A is a constant. The simplication of the above equation
into the formula:
1
240:82
E
g
¼
(4)
l
45
where l is the absorption band edge wavelength.
Fig. 9 XRD patterns of the samples synthesized using anatase as Ti
precursor at different synthetic temperature: (a) 700, (b) NKTO-800,
The absorption intensity in ultraviolet region exhibited an
enhancement with the increasing of synthetic temperature, as
(
c) KTO-900, (d) NKTO-900, (e) NTO-900.
This journal is © The Royal Society of Chemistry 2015
RSC Adv., 2015, 5, 66086–66095 | 66091

View Article Online
RSC Advances
Paper
observed in Fig. 10a, indicating that NKTO-700 can absorb more increasing of UV irradiation time, the suspension in the present
light and induce more photogenerated electrons and holes of as-synthesized photocatalysts faded rapidly from orange to
available for photocatalytic reaction than both NKTO-660 and approximately colorless, while the blank experiment indicates
46
NKTO-680, and the absorption edge of these samples are all the photodegradation of MO is negligible. Fig. 12a presents
located at around 370 nm that corresponds to the E
of ꢃ3.35 clearly the temporal evolution of UV-Vis spectra and color
eV. As shown in Fig. 10b, the E of KTO-900 and NTO-900 can be (inset) of MO solution mediated by NKTO-660 nanobelts. It can
calculated to be about 3.42 eV and 3.48 eV, respectively, which is be observed from Fig. 12b that the photocatalytic activity of the
g
g
34,47
ꢀ
consistent with the previous reports.
Combined with the samples obtained below 700 C increase in the order of NKTO-
inset, a red shi is found for NKTO-900 compared with KTO-900 660 > NKTO-680 > NKTO-700, i.e., the larger SBET of the sample
and NTO-900, and the absorption intensity of the spectrum for is, the higher activity of the sample is, which is just contrary to
NKTO-900 is enhanced obviously in ultraviolet region. NKTO- the result from UV-Vis spectra (Fig. 10a), highlighting the SBET is
9
00 also has the same E with NKTO-800 of ꢃ3.35 eV, while the a more primary factor in determining the photocatalytic activity
g
latter obviously has stronger optical absorption than the of NKTO nanobelts than optical adsorption. This is just why P25
former.
In order to investigate the effect of Na modifying on the report ) exhibits the best activity, considering that high SBET
transfer and lifetime of photogenerated charge carriers in means strong adsorption ability of a photocatalyst and then
photocatalyst particles, the room temperature PL spectra for
2
ꢁ1
with high SBET of 45.71 m g (Fig. S4,† close to the previous
48
NTO-900, KTO-900 and NKTO-900 under an excitation wave-
length of 325 nm were carried out. The results presented in
Fig. 11 show that they can exhibit strong and broad emission in
the wavelength range of about 400–700 nm, whereas the peak
intensities of NKTO-900 are signicantly lower than that of
KTO-900 and NTO-900. Considering the PL emission originates
from the recombination of photogenerated electron–hole pairs,
the decreased peak intensity suggest that NKTO-900 exhibits
lower electrons–holes recombination rate compared with both
of KTO-900 and NTO-900. When the recombination rate
decreases, more photogenerated charge carriers can participate
in the photochemical transformation, resulting in the
enhanced photocatalytic activity.
3.6 Photocatalytic evaluation
The photocatalytic activity of the NKTO samples was evaluated
by photodegradation of MO in the catalyst suspension under UV
light, in comparison with those of the pure K
2 6 2 6 13
Ti O13, Na Ti O
samples and Degussa P25 powders. In the tests, with the
Fig. 12 (a) Temporal evolution of the UV-Vis spectra and color (inset)
of MO solution in the presence of 0.3 g as-synthesized NKTO-660
nanobelts under UV irradiation. (b) Photocatalytic degradation of
Fig. 11 The room temperature PL spectra for NTO-900, KTO-900 methyl orange by the as-synthesized samples under UV irradiation.
and NKTO-900. The blank experiment is also shown in the figure.
66092 | RSC Adv., 2015, 5, 66086–66095
This journal is © The Royal Society of Chemistry 2015

View Article Online
Paper
RSC Advances
enhance its photocatalytic efficiency. The absorption of MO in arithmetic mean of the atomic electron affinity and the rst
dark for 1 h is 10.46%, 5.34%, 4.20% and 3.81% for P25, NKTO- ionization energy; E0 is the energy of free electrons on the
5
2
6
60, NKTO-680 and NKTO-700, respectively, which accords with hydrogen scale (ꢃ4.5 eV). The calculated ECB and EVB for
their SBET K Ti O , Na Ti O , and Na0.88K1.12Ti O were shown in
Combined with the fact that NKTO-800 preform higher Fig. 13, respectively.
activity than NKTO-900, it could be concluded that the lower The ECB of NKTO is negative to the redox potential of O
.
2
6
13
2
6
13
6 13
2
2ꢁ
/cO
5
3
synthetic temperature resulted in a better performance in (ꢁ0.13 eV vs. NHE), implying that the photogenerated elec-
photocatalysis. Besides, the comparative studies between trons can react with O
NKTO-900 and KTO-900 as well as NTO-900 revealed that cO
2
dissolved in the solution to produce
ꢁ
2
radicals that are the important active species involved in
substitutional Na-modifying greatly contributed to an increase photocatalysis. Moreover, the EVB of NKTO is more positive than
ꢁ
54
in the photocatalytic activity, which could be ascribed to the the standard redox potential of OH /cOH (1.99 eV vs. NHE),
efficient separation of photogenerated electron–hole pairs, suggesting that the holes on the surface of NKTO can oxidize
ꢁ
smaller E
g
as well as the enhancement of optical absorption.
OH into active cOH radicals. It is noted that the ECB/EVB of
NKTO move downwards about 0.90 eV/0.78 eV and 0.08 eV/0.02
3.7 Possible mechanism of improved photocatalytic activity
2 6 2 6
eV with respect to those of Na Ti O13 and K Ti O13, respectively,
which may contribute to produce a high amount of active
radicals for photocatalytic reactions from the harvested
photons by the following reaction:
Based on the results above, we can conclude that photocatalytic
activity of the samples is governed by some crucial factors, such
as phase composition, SBET, adsorption ability, and separation
efficiency of photogenerated electrons and holes. Except the low
temperature synthesis resulting in high SBET, the substitutional
Na-modifying changes the phase composition and band struc-
ture, which is deemed to play a vital role in improving the
50
ꢁ
+
NKTO + hn / e + h
ꢁ 2ꢁ
2
O + e / cO
49–51
ꢁ ꢁ
cO2 + 2H + e / H O
2 2
+
photocatalytic activity.
From this angle, a possible mechanism for improved pho-
tocatalytic activity of NKTO nanobelts was tentatively proposed
and illustrated in Fig. 13. The conduction band and valence
band potentials (named as ECB and EVB) of the semiconductors
could be calculated by the following equation:
+
ꢁ
2 2 2
H O + H + e / cOH + H O
ꢁ
ꢁ
H O + cO / cOH + OH + O2
2
2
2
H
2
O
2
+ hn / 2cOH
E
VB ¼ c ꢁ E
0 g
+ 1/2E
2ꢁ
+
2
, H
2
cO , cOH, h + MO / degraded products (CO
O, etc.)
2ꢁ
2
2
ꢁ
Apparently, cO , cOH and H
O
act as the efficient e traps,
E
CB ¼ c ꢁ E
0
ꢁ 1/2E
g
(5)
suppressing the recombination of the photogenerated electrons
in which c is the absolute electronegativity of the semi- and holes to improve the photocatalytic activity. Besides, the
conductor, determined by the geometric mean of the absolute improved photocatalytic activity could be also attributed to the
electronegativity of constituent atoms, which is dened as the dipole moment due to the distortion of TiO octahedra induced
6
Fig. 13 Schematic diagram for the band structure of K
2 6
Ti O
13 and Na Ti
2 6
O13 and the possible photocatalytic mechanism of NKTO nanobelts.
This journal is © The Royal Society of Chemistry 2015
RSC Adv., 2015, 5, 66086–66095 | 66093

View Article Online
RSC Advances
Paper
+
+
by the replacement of K by Na ions promoting photogenerated 13 V. ˇS tengl, S. Bakardjieva, J. ˇS ubrt, E. Vecernıkova,
charges separation.
´
´
´
55
L. Szatmary, M. Klementov ´a and V. Balek, Appl. Catal., B,
006, 63, 20–30.
2
1
1
1
1
1
1
4 M. A. Siddiqui, V. S. Chandel and A. Azam, Appl. Surf. Sci.,
2012, 258, 7357.
5 Y. Inoue, T. Kubokawa and K. Sato, J. Chem. Soc., Chem.
Commun., 1990, 1298.
6 S. Ogura, M. Kohno, K. Sato and Y. Inoue, Phys. Chem. Chem.
Phys., 1999, 1, 179–183.
4
. Conclusions
In summary, high-quality NKTO nanobelts with enhanced
ultraviolet photocatalytic activity have been synthesized via a
facile one-step method from NaCl–KCl melts at a relatively low
temperature below 700 C. The substitutional Na ions were
incorporated into K Ti O without destroying its tunnel-like
structure and oriented growth along [010] direction. Based on
these results, the formation mechanism of NKTO nanobelts was
proposed and rutile was found to be the optimum titanium
precursor. The substitutional Na modifying and low-tempera-
ture synthesis not only red-shied the edge, enhanced the
absorption intensity and enlarged surface area but also effi-
ciently suppressed the recombination of photogenerated elec-
tron–hole pairs, which was studied to improve the
photocatalytic activity of NKTO nanobelts. And, the possible
photocatalytic mechanism was also discussed. Therefore, the
ꢀ
7 S. Ogura, M. Kohno, K. Sato and Y. Inoue, Appl. Surf. Sci.,
2
6 13
1997, 121/122, 521–524.
8 J. Ram ´ı rez-Salgado, E. Djurado and P. Fabry, J. Eur. Ceram.
Soc., 2004, 24, 2477.
9 C. Y. Liu, H. B. Yin, Y. M. Liu, M. Ren, A. L. Wang, C. Ge,
H. P. Yao, H. Feng, J. Chen and T. S. Jiang, Mater. Res.
Bull., 2009, 44, 1173–1178.
2
2
2
2
2
2
2
0 N. Bao, L. Shen, X. Feng and X. Lu, J. Am. Ceram. Soc., 2004,
8
7, 326.
1 Q. Wang, Z. Guo and J. S. Chung, Mater. Res. Bull., 2009, 44,
973–1977.
2 Y. Hakuta, H. Hayashi and K. Arai, J. Mater. Sci., 2004, 39,
977–4980.
3 T. Zhang, Q. Chen and L. M. Peng, Adv. Funct. Mater., 2008,
8, 3018–3025.
1
ꢀ
low temperature (<700 C) one-step synthesis of NKTO nano-
4
belts from NaCl–KCl melts is not only facile and cost-effective
but also greatly benecial to improve the photocatalytic prop-
erties and this study will be inspiring for efficiently preparing
and modifying other functional materials.
1
4 B. L. Wang, Q. Chen, R. H. Wang and L. M. Peng, Chem. Phys.
Lett., 2003, 376, 726.
5 X. D. Meng, D. Z. Wang, J. H. Liu, B. X. Lin and Z. X. Fu, Solid
State Commun., 2006, 137, 146.
6 L. M. Torres-Mart ´ı nez, I. Ju ´a rez-Ram ´ı rez, K. D. Angel-
S ´a nchez, L. Garza-Tovar, A. Cruz-L ´o pez and G. Del Angel, J.
Sol-Gel Sci. Technol., 2008, 47, 158–164.
Acknowledgements
´
´
This work was supported by National Natural Science Founda-
tion of China (No. 51274102).
2
7 K. Teshima, K. Yubuta, T. Shimodaira, T. Suzuki, M. Endo,
T. Shishido and S. Oishi, Cryst. Growth Des., 2008, 8, 465–
4
69.
References
2
8 C. Y. Xu, Q. Zhang, H. Zhang, L. Zhen, J. Tang and L. C. Qin,
1
2
3
4
D. M. Yu, J. S. Wu, L. M. Zhou, D. R. Xie and S. Z. Wu,
Compos. Sci. Technol., 2000, 60, 499–508.
J. Am. Chem. Soc., 2005, 127, 11584–11585.
29 L. Q. Xu and L. Cheng, Mater. Charact., 2010, 61, 245–248.
X. K. Zhang, S. L. Tang, L. Zhai, J. Y. Yu, Y. G. Shi and 30 S. Suzuki, K. Teshima, M. Kiyohara, H. Kamikawa,
Y. W. Du, Mater. Lett., 2009, 63, 887–889.
S. O. Kang, H. S. Jang, K. I. Kim, K. B. Kim and M. J. Jung,
Mater. Lett., 2007, 61, 473–477.
K. Yubuta, T. Shishido and S. Oishi, CrystEngComm, 2012,
14, 4176–4180.
31 Y. B. Mao, T. Park, F. Zhang, H. J. Zhou and S. Wong, Small,
Y. C. Kim, M. H. Cho, S. J. Kim and H. Jang, Wear, 2008, 264,
2007, 3, 1122.
2
04–210.
32 Q. Wang, Q. J. Guo, H. Wang and B. Li, Mater. Lett., 2015,
155, 38.
S. Q. Wu, Z. S. Wei and S. C. Tjong, Compos. Sci. Technol., 33 C. Y. Xu, Y. Z. Liu, L. Zhen and Z. L. Wang, J. Phys. Chem. C,
000, 60, 2873–2880. 2008, 112, 7547–7551.
H. X. Zhang, X. D. He and F. He, J. Alloys Compd., 2009, 472, 34 C. Y. Xu, J. Wu, P. Zhang, S. P. Hu, J. X. Cui, Z. Q. Wang,
5
6
J. Z. L u¨ and X. H. Lu, J. Appl. Polym. Sci., 2001, 82, 368–374.
2
7
8
9
1
94.
D. V. Bavykin and F. C. Walsh, J. Phys. Chem. C, 2007, 111,
4644.
Y. D. Huang and L. Zhen, CrystEngComm, 2013, 15, 3448–
3453.
1
35 E. V. Tretyachenko, A. V. Gorokhovsky, G. Y. Yurkov,
F. S. Fedorov, M. A. Vikulova, D. S. Kovaleva and
E. E. Orozaliev, Particuology, 2014, 17, 22–28.
36 F. Amano, T. Yasumoto, T. Shibayama, S. Uchida and
B. Ohtani, Appl. Catal., B, 2009, 89, 583.
T. W. Kim, I. Y. Kim, J. H. Im, H.-W. Ha and S.-J. Hwang, J.
Photochem. Photobiol., A, 2009, 205, 173–178.
0 J. Park, J. Alloys Compd., 2010, 492, 57–60.
1 X. K. Zhang, J. J. Yuan, H. J. Yu, X. R. Zhu, Z. Yin and H. Shen,
J. Alloys Compd., 2015, 631, 171–177.
1
1
37 S. Tawkaew, M. Chareonpanich and S. Supothina, Mater.
Chem. Phys., 2008, 111, 232.
1
2 L. Zhen, C. Y. Xu, W. S. Wang, C. S. Lao and Q. Kuang, Appl.
Surf. Sci., 2009, 255, 4149.
38 T. Zaremba, Mater. Sci.–Pol., 2012, 30, 185.
66094 | RSC Adv., 2015, 5, 66086–66095
This journal is © The Royal Society of Chemistry 2015

View Article Online
Paper
RSC Advances
3
4
4
4
9 B. R. Li, W. Shang, Z. L. Hu and N. Q. Zhang, Ceram. Int., 47 G. H. Du, Q. Chen, P. D. Han, Y. Yuan and L.-M. Peng, Phys.
014, 40, 78. Rev. B: Condens. Matter Mater. Phys., 2003, 67, 035323.
0 P. M. Rørvik, T. Lyngdal and R. Sæterli, Inorg. Chem., 2008, 48 K. Sridharan, E. Jang and T. J. Park, Appl. Catal., B, 2013,
7, 3173. 142–143, 718–728.
1 H. B. He, W. J. Yao, C. S. Wang, X. Feng and X. H. Lu, Ind. 49 H. Kato, H. Kobayashi and A. Kudo, J. Phys. Chem. B, 2002,
Eng. Chem. Res., 2013, 52, 15037. 106, 12443.
2 K. S. W. Sing, D. H. Everett, R. A. W. Haul, L. Moscou, 50 F. Z. Ren, J. H. Zhang and Y. X. Wang, RSC Adv., 2015, 5,
2
4
R. A. Pierotti, J. Rouquerol and T. Siemieniewska, Pure
Appl. Chem., 1985, 57, 603–619.
3 J. Zhang, S. Z. Hu and Y. J. Wang, RSC Adv., 2014, 4, 62914.
29058.
51 Z. Zou, J. Ye and H. Arakawa, Chem. Phys. Lett., 2000, 332,
271–277.
4
4
4 F. Chang, Y. C. Xie, J. Zhang, J. Chen, C. L. Li, J. Wang, 52 X. J. Guan and L. J. Guo, ACS Catal., 2014, 9, 3025.
J. R. Luo, B. Q. Deng and X. F. Hu, RSC Adv., 2014, 4, 53 W. Y. Teoh, J. A. Scott and R. Amal, J. Phys. Chem. Lett., 2012,
2
8519–28528.
5 S. D. Delekar, H. M. Yadav, S. N. Achary, S. S. Meena and 54 L. Kong, Z. Jiang, T. Xiao, L. Lu, M. O. Jones and
S. H. Pawar, Appl. Surf. Sci., 2012, 263, 539. P. P. Edwards, Chem. Commun., 2011, 47, 5512–5514.
6 B. Yue, Q. Y. Li, H. Iwai, T. Kako and J. H. Ye, Sci. Technol. 55 A. Kudo, H. Kato and S. Nakagawa, J. Phys. Chem. B, 2000,
Adv. Mater., 2011, 12, 034401. 104, 571–575.
3, 629–639.
4
4
This journal is © The Royal Society of Chemistry 2015
RSC Adv., 2015, 5, 66086–66095 | 66095