article13

See discussions, stats, and author profiles for this publication at: https://www.researchgate.net/publication/236844382
A simple route for the preparation of Pmodified TiO2: Effect of phosphorus on
thermal stability and photocatalytic activity
ArticleinJournal of the Taiwan Institute of Chemical Engineers · January 2012
DOI: 10.1016/j.jtice.2011.06.011
CITATIONS
24
READS
35
4 authors, including:
Some of the authors of this publication are also working on these related projects:
Wet air oxidation of acetic acid over platinum and ruthenium catalysts supported on cerium based materials View project
Geopolymers View project
Kais Elghniji
Faculty of Science of Gafsa
23 PUBLICATIONS290 CITATIONS
SEE PROFILE
Sylvie Rossignol
Center european ceramic
227 PUBLICATIONS2,991 CITATIONS
SEE PROFILE
Mohamed Ksibi
University of Sfax/High Institute of Biotechnology
109 PUBLICATIONS4,005 CITATIONS
SEE PROFILE
All content following this page was uploaded by Kais Elghniji on 24 February 2019.
The user has requested enhancement of the downloaded file.

A simple route for the preparation of P-modified TiO
2
: Effect of phosphorus
on
thermal stability and photocatalytic activity
Kais Elghniji
a,
*,
Julien Soro
b
, Sylvie Rossignol
b
, Mohamed Ksibi
a
a
Laboratoire Eau, Energie et Environnement (LR3E), Ecole Nationale d’Inge
´nieurs
de Sfax, BP W 3038, Sfax, Tunisia
b
Groupe d’Etude des Mate
´riaux
He
´te ´roge `nes
(GEMH-ENSCI) Ecole Nationale Supe
´rieure
de Ce
´ramiques
Industrielles de Limoges, 47-73 Avenue Albert Thomas, 87065 Limoges, France
1. Introduction
TiO
2
is considered to be the most promising heterogeneous
photocatalyst,
because of its high photocatalytic activity, non-
toxicity,
inexpensiveness, chemical stability and favourable
optoelectronic
properties. Titania exists in three different forms:
anatase,
rutile, and brookite, of which anatase generally show the
best
photocatalytic activity performance [1,2]. Without any
modification,
transformation of anatase to rutile usually occurs
at
500–700 8C, which may result in the decrease in photocatalytic
ability
of TiO
2
[3–5].
In
addition to the crystallite transformation obtained by high-
temperature
calcination, crystallite growing and serious sintering
are
observed with increasing the temperature of calcinations leading
to
the drastic decrease in surface area and photocatalytic activity.
Hence,
the inhibition of the growth of the anatase crystals could
maintain
the anatase metastability at high temperature and control
crystallite
size. To improve the thermal stability and photocatalytic
activity,
titania has been modified with La
2
O
3
[6], ZrO
2
, SiO
2
[7], or
doped
by inorganic nonmetal such as sulphur [8] and fluor [9,10]. Lv
et
al. [10] developed a simple and novel synthetic method for the
fabrication
of TiOF
2
via a microwave-assisted hydrothermal route
using
tetrabutyl titanate and hydrofluoric acid as raw materials. The
prepared
anatase TiO
2
from TiOF
2
shows very high thermal stability
and
the phase transformation temperature from anatase to rutile is
up
to 1000 8C. The high thermal stability and the photocatalytic
enhancement
of catalysts were attributed to the adsorbed Fluoride
ion
on the surface of anatase after calcination at 700 8C. Among the
various
methods, modification and doping of TiO
2
by phosphorus
seems
to be a another promising approach [11–16]. Ko
˜ro ¨si
and
Dekany
[13] prepared a series of phosphate modified-TiO
2
samples
by
the sol–gel method. They argued that the surface bound
phosphate
have delayed the formation of the anatase phase,
crystallite
growth and inhibited the anatase–rutile phase transfor-
mation.
Lin et al. [14] demonstrated that the doping of phosphorus
could
efficiently inhibit the grain growth and enhance the surface
area
of TiO
2
nanoparticles. The above researches in literature seem
discrepant
and complicated. This is probably due to the variety of
synthetic
methods adopted to prepare the solid and the different
phosphorus
contents. The present work aims at complementing
such
investigations with P-modified TiO
2
nanoparticles which are
synthesized
through an easy procedure. The thermal stability, the
control
of crystalline structure and the effect of phosphorous content
on
the photocatalytic degradation of 4-chlorophenol were exten-
sively
investigated. To the best of our knowledge, there is no report
on
the impact of the 4-CP solution before and after photocatalytic
treatment
to environment using P-modified TiO
2
. In fact, chlor-
ophenols
constitute a group of serious environmental pollutants
that
must be eliminated [17–21]. As a result of their widespread use
in
mothproofing, miticides, pesticides, herbicides, germicides and
Journal of the Taiwan Institute of Chemical Engineers 43 (2012) 132–139
A R T I C L E I N F O
Article history:
Received 28 February 2011
Received in revised form 30 April 2011
Accepted 5 June 2011
Available online 15 August 2011
Keywords:
TiO
2
Phosphorus content
Particle
growth
TiP
2
O
7
Thermal stability
A B S T R A C T
Phosphorus-modified dioxide nanoparticles were prepared by sol–gel method. The effect of phosphorus
precursor
and calcination temperatures on phase transformation, grain growth and surface area were
investigated
using various spectroscopic and basic techniques (ICP-AES, XRD, BET,
31
P MAS NMR, FT-IR,
and
UV–vis methods). It was found that the phosphorus existed as amorphous titanium phosphate in
TiO
2
framework after calcination at temperature of 500–700 8C. As results, slows down the particle
growth
of anatase and increases the anatase-to-rutile phase transformation. The average crystallite size
of
P-modified TiO
2
increased dramatically from 8 to 59 nm when the temperature increased from 500 to
900
8C. This change was associated with the formation of pyrophosphate TiP
2
O
7
species through
condensation
of the concentrated phosphate species. The BET surface area of modified samples was 3.4-
fold
higher than that of unmodified TiO
2
and was 70% higher than that of commercial Degussa P-25. The
photocatalytic
activity of P-modified TiO
2
was 1.5-fold higher than that of commercial Degussa P-25 and
was
49% higher than that of unmodified TiO
2
under UV irradiation. Phytotoxicity was assessed before
and
after irradiation against seed germination of tomato (Lycopersicon esculentum).
ß
2011 Taiwan Institute of Chemical Engineers. Published by Elsevier B.V. All rights reserved.
* Corresponding author. Tel.: [phone]; fax: [phone].
E-mail
address: [email protected] (K. Elghniji).
Contents lists available at ScienceDirect
Journal of the Taiwan Institute of Chemical Engineers
jou r nal h o mep age: w ww.els evier .co m/lo c ate/jtic e
1876-1070/$ – see front matter ß 2011 Taiwan Institute of Chemical Engineers. Published by Elsevier B.V. All rights reserved.
doi:10.1016/j.jtice.2011.06.011

fungicides, chlorophenols pose a serious threat to the environment.
The
US Environmental Protection Agency [22] and the European
Union
directive [23] have labelled chlorophenols as ‘‘priority
pollutants’’,
which means that they need to be constantly monitored
in
the aquatic environment. Considering the potential effect of this
compound
on the ecosystem, a phytotoxicity test using tomato
(Lycopersicon
esculentum) seeds was successfully applied in
ecological
risk assessment of 4-CP phototreated solutions.
2.
Experimental
2.1.
Chemical and catalyst preparation
Titanium
(IV) isopropoxide (Ti(OC
2
H
5
)
4
, 97%, Aldrich) and
isopropylic
alcohol were obtained from Fluka ((CH
3
)
2
CHOH, 99.8%,
Riedel
de Haei#n), phosphoric acid (H
3
PO
4
, 85%, Aldrich,), the
commercially
available TiO
2
(Degussa P25) was obtained from
Degussa
Chemical and UHQ (Milli-Q 18.2 M V ) water.
The
catalyst P-modified TiO
2
, were prepared by the sol–gel
method.
In a typical preparation procedure, titanium (IV)
isopropoxide
was diluted in isopropylic alcohol (molar ratio
1:4).
Deionized water was added dropwise to Ti(OPri)
4
solution
during
stirring (molar ratio between water and alcohol 1:5). White
precipitate
starts appearing indicating the hydrolysis process.
After
being aged for an hour, 100 ml of a phosphoric acid aqueous
solution
with concentration 34.5 mM was added to the resulting
titania
suspension to prepare TiO
2
sample of phosphorus content.
For
comparison, unmodified TiO
2
was also prepared by the same
procedure
without the addition of H
3
PO
4
. The dried materials were
calcined
at 500, 700, 900 and 1000 8C in air for 3 h and with heating
rate
108/min and the calcined powders are labelled according to its
phosphorus
content and calcination temperature. TP
t
, and T
t
,
where
t means the calcination temperature.
2.2.
Structural characterization
The
powder X-ray diffraction patterns were recorded at room
temperature
on a (advanced D8, Bruker, Germany). The experimental
conditions
for refinement of XRD data of TP
1000
are the following: X-
ray
tube operating at 40 kV and 40 mA, 0.6 mm fixed divergence slits,
diffracted
beam curved graphite monochromator (Cu K
a
+1
radiation,
l
1
= 1.540600 A
˚,
l
2
= 1.544390 A
˚)
and 0.1 mm fixed slit in front of the
scintillation
detector. The data were collected in the 2 u range 2–708
with
a step size of 0.028 and a counting time of 5 s/step. Infrared
absorption
spectra were measured on a (Nicolet 380 ATR/FT-IR,
International
Equipment Trading Ltd., USA) spectrometer by the
transmission
method using the KBr pellet technique with 4 cm
1
resolution. The Brunauer–Emmett–Teller (BET) surface area was
measured
by a fully automated surface area analyzer (ASAP 2020
Accelerated
Surface Area and Porosimetry, Micromeritics, USA). The
samples
were degassed in vacuum overnight at 180 8C prior to
adsorption
measurements. The UV–Vis diffuse reflectance (DR UV–
Vis)
spectra were recorded by a Varian Cary 5/UV–Vis-N.I.R.spect-
rometer.
The
31
P and (coupling polarization)
1
H–
31
P CP NMR/MAS
spectra
of the prepared TiO
2
solids were recorded on a (300 ultra
shield,
Bruker, USA) spectrometer at 300 MHz, resonance frequency
operating
at 121.5 MHz and the external magnetic field was 9.4 T. The
pulse
repetition is 10 s, pulse width of 7 m s and a spinning speed of
8.0
kHz. Chemical shifts were indicated using an external H
3
PO
4
(85%) reference (0 ppm).
2.3.
Determination of phosphorus content
2.3.1.
First method
The
Ti and P content of the samples was determined by all argon
sequential
(Thermo Jarrell ASH, USA) inductively coupled plasma
atomic
emission (ICP-AES) spectrometry. The intensity of the
spectral
lines of 213.6 and 336.1 nm were measured.
2.3.2.
Second method
The
concentration of phosphate in the above samples was
measured
using UV-Vis Double Beam PC, Scanning spectropho-
tometer
(UVD-2950) following the ascorbic acid reduction method
[24].
2.4.
Surface acidity measurement
Surface
acidity was determined volumetrically by the adsorp-
tion
of sodium hydroxide NaOH from solutions of different
concentrations.
0.2 g of the solid oxide was mixed with 30 ml of
the
base solutions for 10 h, with continuous stirring then left 24 h.
For
each sample, a blank run was carried out first on the base alone
and
the difference between the blank run and that with the sample
gave
the amount adsorbed.
2.5.
Photocatalytic reaction experiment
Photodegradation
of 4-chlorophenol (4-CP) (98%) was con-
ducted
in a laboratory-scale photoreactor. It is a 120 cm
3
cylindrical photoreactor, operating in a closed recirculating circuit
driven
by a centrifugal pump and with a stirred reservoir tank
equipped
with a device for withdrawal of samples. Illumination
was
carried out using a lamp (11 W low-pressure mercury lamp,
Philips,
Holland) with a wavelength ( l
max
254 nm). At given time
intervals,
about 4 ml aliquots were sampled, centrifuged and
filtered
with a cellulose acetate membrane filter membrane
(0.45
m m pore size, 25 mm diameter) to remove all solid particles.
The
4-CP concentration was estimated by measuring their
absorbance
at 225 and 280 nm using a UV-Vis Double Beam PC,
Scanning
spectrophotometer UVD-2950.
Free
chloride ions are quantified directly after each irradiation
period
with an ion chromatograph (HIC-6A Shimadzu, Japan)
equipped
with a conductivity detector and a Shim-pack column.
The
separation was achieved using an isocratic elution at a flow
rate
of 1.5 ml/min. A mobile phase of 1 mM of tris(hydroxymethy-
l)aminomethane
and 1 g/l of sodium chloride was used as
standards
solution. Chemical oxygen demand (COD) was measured
according
to standard methods described in the Japanese
International
Standard handbook [25].
2.6.
Phytotoxicity (germination tests)
4-Chlorphenol
phytotoxicity was assessed before and after
irradiation
against seed germination of tomato (L. esculentum).
Phytotoxicity
was determined using a modified Zucconi test [26] by
measuring
seed germination. Twenty seeds were placed on filter
papers
in 9 cm Petri dishes and 6 ml of treated solution was then
uniformly
added to each dish. Dishes were incubated in the dark at
26
2 8C for 5 days. Distilled water was used as control. All samples,
including
controls, were triplicated. A germination index (GI) was
calculated
by counting the number of germinated seeds and the average
root
length observed in each sample compared to control treatments
[27].
Results finally expressed according to the following formula:
GI ¼number
of germinated seeds in sample
number
of germinated seeds in control
average
of root lengths in sample
average
of root lengths in control 
100
A
seed was considered germinated when the root length
exceeded
5 mm. For root lengths below 5 mm, it was considered
K. Elghniji et al. / Journal of the Taiwan Institute of Chemical Engineers 43 (2012) 132–139 133

equal to 0 and the seed was not considered germinated. The
average
sum of root lengths comprised the sum of the lengths of all
germinated
seeds in a Petri dish.
3.
Results and discussion
3.1.
Determination of phosphate content
Table
1 summarizes the P content of P-modified TiO
2
determined by ICP-AES and the ascorbic acid reduction method.
As
shown, the Phosphorus content is identical with this of
calculated
one, indicating the added phosphoric acid reacts
completely
with titania at the applied preparation conditions.
3.2.
XRD studies
XRD
was used to investigate the changes of phase structure of
the
as-prepared catalysts after heat treatment at different
temperatures.
Fig. 1 shows the XRD patterns of unmodified and
P-modified
TiO
2
calcined at different temperatures. With increas-
ing
calcination temperature, the peak intensity of anatase
increases,
and the width of the (1 0 1) plane diffraction peak of
anatase
becomes narrower, indicating the enhancement of
crystallization.
For unmodified TiO
2
, the rutile phase starts to
appear
at 700 8C, and it becomes the only phase at 900 8C (JPCDS
83-2242).
In contrast, P-modified TiO
2
retains a pure anatase phase
(JPCDS
84-1286) in the calcination temperature range of 500–
900
8C, and no obvious XRD peaks corresponded to titanium
phosphate
are observed after phosphorus modification. The
titanium
phosphate should be amorphous also in nature, mixed
to
anatase TiO
2
. When calcined at 900 8C, two peaks at 2 u = 22.55
and
27.708 appeared in spectrum of TP
900
correspond to the (6 0 0)
and
(7 2 1) XRD diffraction peak of crystalline TiP
2
O
7
(JPCDS 38-
1468)
[28,29]. This phase with cubic lattice phase comes into being
at
the same time with anatase phase. At temperature extremely
higher,
the anatase TP
900
transformed to the rutile TP
1000
structure
and
the TiP
2
O
7
phase still remained in the solid composition.
3.3.
Lattice parameters
From
Rietveld analysis of P-modified and unmodified TiO
2
,
lattice
parameters and anatase molar fraction were obtained and
summarized
in Table 2. As clearly shown, calcination at tempera-
ture
in the range from 500 to 900 8C produces a progressive
increasing
on the tetragonality (c/a) of anatase structure of P-
modified
TiO
2
catalyst (TP). This is similar to the value reported in
the
(JPCDS 84-1286) for well-crystallized anatase (c/a = 2.510).
Hence,
the observed expansion is not likely related to a crystal size
effect.
Even though small changes are expected in the lattice
parameters
(c/a) upon changing the titania anatase crystal size.
Regarding
the unmodified TiO
2
(T), the increase of particle size is in
inverse
proportion to the tetragonality. That means 47 nm is
critical
size to undergo anatase-to-rutile transformation at 700 8C
with
(c/a = 0.644) for well-crystallized rutile.
Fig.
2(a) shows a typical XRD data of TP
1000
analyzed using the
program
FULLPROF [30]. The quality of the agreement between
observed
and calculated patterns for each phase is measured by a set
of
factors given by the FULLPROF program. These results confirmed
the
presence of rutile TiO
2
and TiP
2
O
7
phases at 1000 8C. Fig. 2(b)
shows
the
31
P MAS NMR spectrum of the TP
1000
spun at 8 kHz and at
room
temperature. The existence of lines at 38.6, 40.5, 42.9,
44.5,
46.15, 49.48 and 52.35 ppm should be associated with
the
presence of several crystallographic sites for P atoms in a
3
3 3 crystal structure of TiP
2
O
7
with space group (Pa-3) [31,32].
3.4.
Average crystalline sizes and surface area
The
average crystalline size anatase and rutile can be calculated
from
the broadening of the (1 0 1) peak and the (1 1 0) peak, as
shown
in Fig. 3. For the T
500
, the average crystalline size of anatase
p [ 630]
p [ 721]
p [ 600]
r [110]
r [111]
r [101]
T900
TP900
TP1000
T700
TP700
T500
a [101]
r [220]
r [211]
r [210] r [210]
r [200]
r [220]
r [211]
r [111]
r [200]
r [101]
r [110]
a [101]
605040302010
TP500
a [211]
a [105]
a [200]
a [004]
a [211]
a [105]
a [200]
a [004]
a [112]
a [103]
a [101]
2 θ
Fig. 1. The XRD patterns of T and TP calcined at various temperatures. a, r and p
denote
anatase, rutile and TiP
2
O
7
respectively.
Table 2
Crystallographic
Parameters and phase contents of unmodified and P-modified
TiO
2
.
Sample
Phase contents
a
Lattice parameters V (A
˚
3
)
A
(%) R (%) a (A
˚)
b (A
˚)
c (A
˚)
T
500
100 – 3.783 3.783 9.472 135.77
T
700
26 74 4.589 4.589 2.960 62.38
T
900
– 100 4.598 4.598 2.963 62.37
TP
500
100 – 3.788 3.788 9.480 135.98
TP
700
100 – 3.789 3.789 9.507 136.30
TP
900
100 + P – 3.784 3.784 9.531 136.48
TP
1000
– 100 + P 4.594 4.594 2.958 62.44
a
The phase content of TiO
2
was obtained from the following formulas: W
R
= 1/
[1
+ 0.884(A
A
/A
R
)], W
A
= 1 W
R
, where W
A
and W
R
are the content of anatase and
rutile,
respectively, A
A
and A
R
are the diffraction intensity of anatase (1 0 1) and
rutile
(1 1 0). A, R and P denote anatase, rutile and TiP
2
O
7
respectively. The cell
volume
and lattice parameters obtained from the Rietveld refinement of X-ray
diffraction
data.
Table 1
Phosphorus
content in P-modified TiO
2
samples obtained from ICP-AES and
ascorbic
acid reduction method.
Sample
Weight calculated
(ppm)
ICP-AES weight
(ppm)
A.R. method
weight
(ppm)
TP
500
3.97 3.84 4.04
K.
Elghniji et al. / Journal of the Taiwan Institute of Chemical Engineers 43 (2012) 132–139 134

(21 nm) quickly increases at temperatures above 500 8C
,
indicating
the
sintering of TiO
2
particles and the rutile phase becomes the
major
phase at 700 8C. At this temperature, the mass fraction of
rutile
is 74%, and its crystallite size is 45 nm. At 900 8C, anatase
completely
turns into rutile phase and the crystallite size of rutile
rapidly
increases to more than 100 nm. The similar result is also
reported
by Yu and Wang [33]. They prepared TiO
2
nanotube
arrays
by electrochemical anodization of titanium foil. With the aid
of
XRD and SEM morphology of the 600–700 8C-calcined samples,
they
found a stable tubular structure with anatase/rutile particles
size
of 35–40 nm. At 800 8C, the nanotube arrays are destroyed and
only
dense rutile crystallites with size of over 180 nm are observed.
They
suggested that the high temperature and phase-transforma-
tion
heat cause the growth of rutile crystallites, on the other hand
the
diffusion of oxygen in air into the nanotube–support interface
region
oxidize the titanium in that region and directly transform
titanium
into rutile phase.
However,
for TP
500
, the crystallite size of anatase is 8.5 nm at
500
8C. That means the P species markedly slows down the crystal-
growth
rate of anatase and maintains the anatase metastability at
high
temperature. Beyond this temperature an increase in anatase
particle
size is observed, becoming significant at 900 8C, reflecting
increased
sintering. Reidy et al. [34] reported that the critical size
for
25% anatase-to-rutile conversion was 45.1 nm, respectively. In
the
current work, the particle size of TP
800
catalyst is 34 nm, i.e., it
is
small to undergo anatase-to-rutile transformation.
The
BET surface areas of the samples calcined at various
temperatures
are shown in Fig. 3 inset. With increase of calcination
temperature,
the measured specific surface areas oppose to the
calculated
crystallite size, indicating the crystallite growth of
anatase
during calcination. Upon calcination at 500–700 8C, the
surface
area of T
700
is quickly decreased to 5 m
2
/g, reflecting the
loss
of coordinated water (as revealed by FT-IR analysis) and
sintering
of rutile T
700
particles [35,36]. In contrast, the surface
area
of TP
500
is larger than that of the anatase T
500
one. This finding
indicates
that P-modified TiO
2
surface has more hydroxyl groups
than
the pure TiO
2
which may be attributed to the less loss of water
during
calcination temperature. Above 700 8C, a decrease in
surface
area is observed for modified TiO
2
, becoming significant
at
900 8C. This is a reflection of sintering of anatase particles at
higher
temperatures (as revealed by FT-IR and XRD analysis).
3.5.
FT-IR studies
FT-IR
spectra of the unmodified and P-modified TiO
2
samples
calcined
at various temperatures are shown in Fig. 4. For all
samples,
the broad peak at 1630 cm
1
and the peak at 3420 cm
1
corresponds to bending and asymmetric stretching modes of
molecular
water, respectively. Obviously, the P-modified TiO
2
has
more
surface-adsorbed water and hydroxyl groups than the
unmodified
TiO
2
, and these absorption bands gradually decrease
as
the temperature increases.
In
addition, three absorption bands at 1100, 1135 and
1035
cm
1
are observed in the IR spectra of TP
500
but absent for
T
500
(Fig. 4(a) inset). The shifts of these bands suggest that different
chemical
environments existed around the phosphorus. The
former
band is possibly the 1082 cm
1
band, characteristic of
the
n
3
vibration of the phosphate ions coordinated to TiO
2
, shifted
to
lower wavelength as a result of hydrogen bond interactions in
the
host metal oxide lattice (hydrogen phosphate states) [37]. The
shoulder
peak at 1135 cm
1
appears gradually, characteristic of
the
n
2
vibration of the phosphate in a bidentate state (bridging
bidentate
surface) [38]. The band at 1036 cm
1
is assigned peak to
the
vibrations of Ti–O–P bond [39–42]. Therefore, the phosphorus
may
exist as the surface (1100 and 1135 cm
1
) and the form Ti–O–
P
(1036 cm
1
) bulk of TiO
2
anatase, with P replacing part of Ti
4+
ions.
Furthermore,
the TP
700
catalyst shows a gradual appearance of
bands
at 1047 cm
1
, 968 cm
1
and a shoulder broad band at
1184
cm
1
(Fig. 4(b) inset). These bands can be assigned to the
phosphate
probably in a bidentate state [43].
Fig. 2. (a) Rietveld plot for the TP calcined 1000 8C, r and p are denote rutile TiO
2
and
TiP
2
O
7
, respectively. (b)
31
P MAS-NMR spectra of TiP
2
O
7
obtained at 1000 8C.
Chemical
shifts are referenced to that of phosphoric acid.
900800700600500
0
20
40
60
80
100
120
140
1000900800700600500
0
40
80
120
160
200
TP
T
Crystallite size (nm)
Temperature (°C)
Temperature(°C)
SBET(m2/g)
TP
T
Fig. 3. Average crystallite sizes and specific surface areas (inserted) of the samples
calcined
at various temperatures. The average crystallite sizes are calculated from
the
broadening degree of the (1 0 1) XRD peak of anatase phase.
K.
Elghniji et al. / Journal of the Taiwan Institute of Chemical Engineers 43 (2012) 132–139 135