Photocatalytic removal of pentachlorophenol by means of an enzyme-like molecular imprinted photocatalyst and inhibition of the generation of highly toxic intermediates

Article abstract of DOI:10.1039/b9nj00255c

Pentachlorophenol (PCP) is a typical highly-toxic pollutant, and its direct photolysis and conventional photocatalysis may produce more toxic by-products such as dibenzodioxins. It is urgently needed to develop a photocatalytic process able to remove PCP without the generation of highly toxic by-products. To achieve this, enzyme-like molecular-imprinted photocatalysts were prepared by using structural analogues of PCP as pseudo templates. It was found that 2,4-dinitrophenol (DNP) was the best template among the tested analogues. The molecular imprinted polymer (MIP) coated P25 TiO₂ photocatalyst DNP-P25 prepared with DNP as the template greatly accelerated the photocatalytic degradation of PCP and depressed the generation of toxic intermediates. It was confirmed that the amino groups at the footprint cavities provided a well-defined micro reaction environment, which made the benzene ring of the adsorbed PCP be better exposed to photo-generated reactive OH radicals, leading to easier cleavage of the benzene ring. Both the intermediate analysis and toxicity evaluation confirmed that the MIP-coated TiO₂ can make the photocatalytic degradation a safe and green approach of removing PCP.

Full text of DOI:10.1039/b9nj00255c

View Article Online / Journal Homepage / Table of Contents for this issue
PAPER
www.rsc.org/njc | New Journal of Chemistry
Photocatalytic removal of pentachlorophenol by means of an enzyme-like
molecular imprinted photocatalyst and inhibition of the generation of
highly toxic intermediatesw
Xiantao Shen,a Lihua Zhu,*a Guoxia Liu,a Heqing Tang,*a Shushen Liub and
Weiying Lib
Received (in Gainesville, FL, USA) 11th June 2009, Accepted 4th August 2009
First published as an Advance Article on the web 2nd September 2009
DOI: 10.1039/b9nj00255c
Pentachlorophenol (PCP) is a typical highly-toxic pollutant, and its direct photolysis and
conventional photocatalysis may produce more toxic by-products such as dibenzodioxins. It is
urgently needed to develop a photocatalytic process able to remove PCP without the generation
of highly toxic by-products. To achieve this, enzyme-like molecular-imprinted photocatalysts were
prepared by using structural analogues of PCP as pseudo templates. It was found that 2,4-
dinitrophenol (DNP) was the best template among the tested analogues. The molecular imprinted
polymer (MIP) coated P25 TiO2 photocatalyst DNP–P25 prepared with DNP as the template
greatly accelerated the photocatalytic degradation of PCP and depressed the generation of toxic
intermediates. It was confirmed that the amino groups at the footprint cavities provided a well-
defined micro reaction environment, which made the benzene ring of the adsorbed PCP be better
exposed to photo-generated reactive OH radicals, leading to easier cleavage of the benzene ring.
Both the intermediate analysis and toxicity evaluation confirmed that the MIP-coated TiO2 can
make the photocatalytic degradation a safe and green approach of removing PCP.
various organic pollutants is rather slow with the generation of
more harmful by-products.11 Therefore, it is possible to use an
1. Introduction
Enzyme mimics or artificial enzymes have been synthesized as
catalysts via molecular imprinting techniques.1–8 The earliest
MIP catalyst for decomposing persistent organic pollutants in
waste waters without producing more harmful by-products.
work in this area dates back to the preparation of molecularly
Recently, we have introduced MIPs into the selective
photocatalytic degradation of organic pollutants.12–15 Using
imprinted polymers (MIPs) for catalytic hydrolysis of the
p-nitrophenyl ester of an amino acid.1 Later, the catalytic
o-phenylenediamine (OPDA) as a monomer and the target
MIPs were introduced into elimination reactions,2 Michael
pollutants (2-chlorophenol, 4-chlorophenol, 2-nitrophenol or
addition reductions,3 hydride transfer reductions,4 oxidation
4-nitrophenol) as templates, we have prepared a series of MIP-
reactions5 and conformational reactions.6,7 The methods to
coated TiO2 photocatalysts. These catalysts markedly promote
obtain MIPs with catalytic activity can also use analogues of
the photocatalytic degradation of the target organic compounds,
the substrate, transition state, or products as templates.8 There
in comparison with the naked TiO2 catalyst.12,13 It has been
are two prerequisites for the preparation of MIPs showing
found that these photocatalysts also decrease the accumulation
enzyme-like catalytic activity. Firstly, a cavity has to be made
of degradation intermediates, indicating that the photo-
with a defined shape corresponding to the shape of the target
catalytic degradation of the intermediates is accelerated at
substrate or its transition state in the reaction. Secondly,
the same time.
functional groups in a defined stereochemistry must be introduced
As a more harmful phenolic pollutant, pentachlorophenol
as binding sites, coenzyme analogues, or catalytic sites within
(PCP) is present in the environment mainly due to its wide-
the cavity.9
spread applications in agriculture and industry as an impor-
The applications of MIPs as catalysts suggests that the main
tant component part of bactericides, insecticides, fungicides
and wood preservatives.16 As PCP can cause severe damage to
reaction can be accelerated and the generation of unfavorable
by-products will be inhibited.10 This is helpful for environmental
human health even at low concentrations, it is widely recognized
protection, because it is known that the chemical treatment of
as one of the most toxic pollutants and listed as a priority
pollutant published by the European Union in the last directive
a College of Chemistry and Chemical Engineering, Huazhong
University of Science and Technology, Wuhan 430074, P.R. China.
E-mail: lhzhu63@yahoo.com.cn, hqtang62@yahoo.com.cn;
about water policy (Directive, 2000/60/EC). Several methods
have been investigated for the removal of PCP from the
environment. One is the biodegradation under aerobic and
anaerobic conditions,17,18 but it is slow and incomplete
because of the toxicity of PCP to the organisms.19 Direct
photolysis of PCP is also slow.20–22 Advanced oxidation
processes (AOPs) such as radiolysis,23 and the ozone,24
Fax: +86-27-87543632; Tel: +86-27-87543432
b Key Laboratory of Yangtze River Water Environment, Ministry of
Education, College of Environmental Science and Engineering,
Tongji University, Shanghai 200092, P.R. China
w Electronic supplementary information (ESI) available: Supplemen-
tary figures showing spectra of results. See DOI: 10.1039/b9nj00255c
ꢀc
This journal is The Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2009
2278 | New J. Chem., 2009, 33, 2278–2285

View Article Online
ozone/ultrasound,25 Fenton,26 and photocatalytic processes27,28
have been investigated for treating PCP. However, the
degradation in these processes is inefficient due to the
accumulation of intermediates.29 Unfortunately, more toxic
intermediates are generated in the reported processes of
removing PCP. For example, more harmful polychlorinated
dibenzo-p-dioxins (PCDDs) and polychlorinated dibenzofurans
(PCDFs) are generated during the biodegradation of PCP,30
and more harmful intermediates such as p-chloranil, tetra-
chlorohydroquinone (TeCHQ),31 PCDDs and PCDFs32 were
produced in the treatment of PCP by AOPs. Therefore, a green
and highly efficient approach for the removal of PCP is
urgently necessary, by which the degradation of PCP should
be significantly enhanced with the inhibition of the formation
of more toxic intermediates.
were similar to that described in previous work.12,13 By
varying the templates, the obtained photocatalysts were
correspondingly referred to as PCP–P25, 4-NP–P25,
DNP–P25 and TNP–P25. As a control, the photocatalyst
NIP–P25 was synthesized when no template was used.
To verify the important role of the imprinted cavities on the
MIP layer in the photodegradation, the functional amino
groups on the MIP-coated TiO2 (DNP–P25) were acetylated
with acetic anhydride. Briefly, acetic anhydride (0.1 mL,
1.1 mmol) was slowly added to the catalysts (0.1 g) in
dichloromethane (35 mL) in the presence of triethylamine
(0.2 mL, 1.4 mmol, 1.25 equiv. of acetic anhydride). The
mixture was stirred under an N2 atmosphere for 12 h at room
temperature. Excess solvent and reagents were removed by
evaporation. After being washed with water and dried at
room temperature, particles were obtained as amino-group-
protected DNP–P25.
As observed previously, MIP-coated TiO2 photocatalysts
not only promote the photocatalytic degradation of the target
organic compounds such as 2-chlorophenol, 4-chlorophenol,
2-nitrophenol and 4-nitrophenol, but also greatly reduce the
accumulation of degradation intermediates.12,13 This suggests
that we may also prepare MIP-coated TiO2 by using PCP as
the template and then use it to photocatalytically degrade
PCP. However, we have found that PCP itself is not a
satisfactory template from preliminary experiments. We have
to find an appropriate template to prepare the MIP-coated
TiO2 for the photocatalytic removal of PCP. In the present
work, on the basis of studying the interactions between the
functional monomer and the template molecule under the
conditions for synthesizing the MIP, we chose appropriate
water-soluble structural analogues as a pseudo template and
thus prepared (pseudo) molecular imprinted photocatalysts.
The footprint cavities and the amino groups around the
cavities are anticipated to play an important role in the
2.3 Characterization
The morphology of the photocatalysts was observed using
high-resolution transmission electron microscopy (HRTEM)
on a JEM-2010FEF TEM. FTIR spectra were recorded on a
Bruker VERTEX 70 spectrophotometer, and UV-visible
solid-state reflection spectra were measured on a Shimadzu
UV-2550 spectrophotometer.
2.4 Photocatalytic experiments
Photocatalytic experiments were carried out in a cylindrical
glass reactor, where a Philips 9 W UV lamp (lmax 253.7 nm)
with a quartz tube as the protective shelter was located in the
axial position. After it was filled with 250 mL of an aqueous
solution of PCP, the catalyst (0.1 g Lꢁ1) was added, followed
by an adjustment of the solution pH to 11, at which most of
the PCP in solution was changed to its basic form (PCP–Na).
The suspension was stirred for 20 min to favor the organic
adsorption/desorption equilibrium on the catalyst surface, and
the concentration of the pollutant(s) was determined as the
initial concentration c0, and then the photo-irradiation started.
In order to maintain a well dispersed photocatalyst suspension,
the PCP solution was homogenized with a magnetic stirrer
during the UV irradiation. Aliquots (2 mL) of the solution
were sampled at appropriate time intervals. The solution
samples were centrifuged at 14 000 rpm (Hettich EBA21) for
15 min, and the supernatant was filtered through a 0.22 mm
filter to remove the catalyst particles. The resultant solutions
were used for analysis.
photodegradation, because they may provide
a special
microreaction environment, which can make the benzene ring
of the adsorbed PCP be better exposed to reactive OH radicals
and then to be more easily cleaved for a rapid mineralization.
2. Experimental section
2.1 Materials
Degussa P25TiO2 nanoparticles (ca. 80% anatase, 20% rutile;
BET area, ca. 50 m2
g
ꢁ1) were used. PCP, 4-nitrophenol
(4-NP), 2,4-dinitrophenol (DNP), 2,4,6-trinitrophenol (TNP)
and OPDA of analytical purity, and were purchased from the
Shanghai Chemical Reagent Company. Methanol (chromato-
graphic purity) was obtained from Tedia. All these chemicals
were used without further purification. Solution pH was
adjusted by using diluted solutions (1 mol Lꢁ1) of H3PO4,
HCl or NaOH.
2.5 Analytical methods
The concentrations of the pollutants and intermediates were
measured by high-performance liquid chromatography
(HPLC) on a PU-2089 HPLC (JASCO), equipped with a
C18 ODS column and an ultraviolet detector. The detection
wavelength was set at 218 nm for PCP. In some cases, the
concentrations of the pollutants and intermediates were also
determined by UV-visible absorption spectrophotometry on a
Cary 50 spectrophotometer (Varian). The concentrations of
chloride ions and main organic acids as intermediates were
monitored by ion-chromatography on a Dionex DX-500 IC
2.2 Preparation of catalyst
MIP-coated photocatalysts were prepared by coating an
MIP layer on the surface of P25 nanoparticles via in situ
polymerization of OPDA (as a functional monomer and
cross-linking agent) in the presence of P25 nanoparticles and
the template (the target PCP and its analogues including 4-NP,
DNP and TNP). The details of the preparation procedures
ꢀc
This journal is The Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2009 New J. Chem., 2009, 33, 2278–2285 | 2279

View Article Online
system with an ED-40 electrochemical detector operating in
conductivity mode and an IonPac ICE-AS6 column. Total
organic carbon (TOC) analysis was carried out by a TOC
analyzer micro N/C model (Analytika jena, Germany)
equipped with an automatic sample injector and a platinum
based catalyst. After being extracted with n-heptane and
esterified with acetic anhydride, major aromatic products
were identified with an Agilent Technologies model 6890 gas
chromatograph (GC) equipped with a 5973N mass-selective
detector (He carrier gas, HP-5MS crosslinked 5% PH ME
siloxane column, electron impact ionization at 70 eV). At least
triplicate runs were carried out for each test, and the standard
deviation was generally less than 5%.
with the functional monomer. In consideration of these
requirements, 4-NP, DNP and TNP were investigated as
possible pseudo templates, in comparison with PCP as the
real template.
The most important factor influencing the performances of
the MIPs is the interaction between the functional monomer
and the template molecules. Such interactions can be evaluated
by comparing the UV-visible absorption spectrum of the
mixture solution of the functional monomer and the template
with the sum of the spectra of the individual functional
monomer and template. Based on the measured spectra of
the functional monomer OPDA (36.0 mmol Lꢁ1), the template
(36.0 mmol Lꢁ1), and their mixture solution (Fig. S1, ESIw),
the absorbance of the solutions at 228 nm was selected for the
evaluation of the interaction between the monomer and
The toxicity of the solution samples was determined by
using flesh water luminescent bacterium, Vibrio qinghaiensis
sp. Q67 as a detecting organism, applying the VeritasTM
luminometer with a 96-well microplate (Turner BioSystems
Inc., USA) according to the method reported by Zhang and
coworkers.33 The final relative light unit (RLU) measurements
in various wells in the test microplate were determined after a
5 min exposure to the toxicants at 22 ꢂ 1 1C. The toxicity of
each sample was expressed as an inhibition ratio (E) and
calculated from E = 100% ꢃ (I0 ꢁ I)/I0, where I0 is
an average of the RLU of Q67 exposed to the controls
(12 parallels) and I is an average of the RLU to the sample
(three parallels) in one microplate. At least triplicate runs were
carried out for each test of the toxicity evaluation, and the
standard deviation was generally less than 10%.
template molecules by using the equation F = (1 ꢁ Amixture
/
(Amonomer + Atemplate)), where A represents the absorbance of
the compound (specified by the subscript) at 228 nm, and F is
defined as the interaction factor hereafter, which indicates the
extinction effect of the mixing due to the interaction between
the monomer and template molecules. A larger F value means
a stronger interaction between the monomer and template
molecules. The values of the F factor in solutions at pH 2.0
were determined as 0.041, 0.130, 0.215, and 0.082 for the
templates of PCP, 4-NP, DNP and TNP, respectively. Thus,
the interaction between the functional monomer and template
molecules increases in the template order of PCP o TNP o
4-NP o DNP. This demonstrates that the increased number
of nitro groups favors the interaction between OPDA and
template molecules, but the existence of three nitro groups and
one –OH group in the benzene ring of TNP produces an
unfavorable effect of steric hindrance. After the interaction
between the functional monomer and the template molecules
was clarified, a series of MIP-coated photocatalysts were
prepared by using PCP, 4-NP, DNP or TNP as templates,
and the resultant MIP-coated photocatalysts were referred
to as PCP–P25, 4-NP–P25, DNP–P25 and TNP–P25,
3. Results and discussion
3.1 Preparation of MIP-coated TiO2 for removing PCP
3.1.1 Selection of the functional monomer. Synthesis of an
MIP with molecular recognition ability requires the formation
of a complex between the template and functional monomer
molecules via chemical and/or non-chemical bonding before
the polymerization. An additional requirement for the
functional monomer should be considered so that the MIP
layer on MIP-coated photocatalysts will not block the inter-
actions between the photo-generated active species and the
adsorbed organic pollutants, which requires the MIP to be
fairly semiconducting. We have recently demonstrated that
OPDA is a good function monomer,13 because its polymerization
results in polyaniline-like polymer being favorable to the
separation of photo-generated electron hole pairs.34 Therefore,
OPDA is further used as the functional monomer in the
present work.
correspondingly.
A control photocatalyst NIP–P25 was
similarly obtained but in the absence of any template.
3.1.3 Optimization of pH and feed ratio of monomer to
template. The effects of solution pH on the F factor between
DNP and OPDA were studied (Fig. S2, ESIw), and the
influences of the molar ratio of OPDA to DNP on the catalytic
activity compared to DNP–P25 were investigated (Fig. S3, ESIw).
Based on these experimental results, two important synthetic
parameters were optimized at pH 2.0 and the molar ratio of
OPDA to DNP at 4 : 1.
3.1.2 Templates from structural analogues of PCP. As we
know, the enzyme-like activity comes from the special inter-
action between the MIP and the template molecules. Because
it was found that PCP itself is not a good template for
preparing the MIP-coated photocatalyst, we checked the
possible pseudo templates from the structural analogues of
PCP. In order to prepare such MIP-coated photocatalysts, the
structural analogues should have (i) good solubility (in water),
(ii) a similar structure to the target pollutant PCP to create
cavities with size and shape fitting the adsorption of the target
molecules, and (iii) functional groups capable of interacting
3.2 Characterization of MIP-coated photocatalysts
The morphology of the MIP-coated photocatalysts was
observed with HRTEM, which indicated that the TiO2 core
in DNP–P25 was partially coated with a layer of MIP, having
a thickness of B5 nm (Fig. S4, ESIw). Furthermore, the MIP
coating on the surface of TiO2 particles was shown by the
UV-vis solid-state reflection spectra of neat P25 and DNP–P25
particles. In comparison with the neat P25, DNP–P25
prolongs the absorption in the visible region (Fig. S5, ESIw). Main
characteristic peaks of poly-o-phenylenediamine (POPD) and
ꢀc
This journal is The Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2009
2280 | New J. Chem., 2009, 33, 2278–2285

View Article Online
TiO2 were observed in the FTIR spectra of the imprinted
photocatalysts (Fig. S6, ESIw). However, in comparison to the
main bands of pure POPD at 1523 cmꢁ1 (the bending mode of
the N–H bond), 1229 and 1346 cmꢁ1 (QC–N stretching on the
benzene ring), all the bands were shifted to higher wavenumbers
when the polymer layer was coated on TiO2. Furthermore, a
sharp peak at 1619 cmꢁ1 being associated with the CQN
stretching vibration was shifted to 1627 cmꢁ1 with increasing
MIP layer thickness. These results indicate the novel structure
of the imprinted photocatalysts may cause the interface inter-
action between the MIP layer and TiO2 core, and the chemical
bond interaction is useful to transfer charge carriers and
induce the enhancing effect on the photocatalytic activity.
preparing the MIP-coated photocatalysts to remove them as
pollutants in the environment, but it is possible to find an
appropriate structural analogue of the target pollutant for
preparing the required MIP-coated photocatalyst.
3.4 Inhibited generation of harmful intermediates
It is feared that the photocatalytic oxidation of PCP on TiO2
could produce more harmful intermediates and by-products.
Mills and Hoffmann35 reported that the photooxidation of
PCP over TiO2 proceeds via the attacking of a hydroxyl
radical at the para position on the benzene ring to form a
semiquinone radical which in turn disproportionates to yield
p-chloranil and TCHQ, being possibly followed by a further
degradation. The toxic dimeric intermediates such as PCDDs
and PCDFs were also found by other groups.36,37 It is
necessary to clarify if these more toxic intermediates are
inhibited or not when PCP is degraded over DNP–P25,
which was performed by UV-visible absorption and GC/MS
measurements.
3.3 Enhanced catalytic ability of DNP–P25 photocatalyst
Fig. 1 shows the kinetics of direct photolysis and photo-
catalytic degradation of PCP over different photocatalysts,
which indicates that all the degradation processes follow a
pseudo first-order reaction. It is clearly seen from Fig. 1 that
the direct photolysis of PCP is the slowest. The added TiO2
will adsorb UV light, resulting in the decreasing of the direct
photolysis. Thus, the photocatalytic process is the dominant
one in the photodegradation of PCP over DNP–P25. This was
confirmed by the much enhanced degradation of PCP over
DNP–P25. The apparent rate constant (k) for PCP degradation
has an order of DNP–P25 4 4-NP–P25 4 TNP–P25 4
P25 4 PCP–P25 4 NIP–P25 4 direct photolysis. The
enhancing or depressing effect can be compared by the value
of kphotocatalyst/kP25, which was determined as 0.62, 0.79, 1.31,
1.49 and 2.75 for NIP–P25, PCP–P25, TNP–P25, 4-NP–P25
and DNP–P25, respectively. These results confirm that DNP is
the best template.
Firstly, the UV-visible absorption spectra of the PCP solution
(5 mg Lꢁ1, pH 11) were monitored during the degradation as
shown in Fig. 2. In the case of P25, two new absorption peaks
were observed at about 280 nm and 310 nm, being consistent
with the report that the intermediates TCHQ and p-chloranil
had absorptions at 308 nm and 284 nm, respectively,36 and
another new absorption band in the region from 320 nm to
410 nm was similar to the absorptions of the dibenzodioxins/
furans.22 However, the PCP photodegradation over DNP–P25
was much faster than over P25, and the accumulation of the
aromatic intermediates over DNP–P25 was almost unobservable
(as seen in the region from 240 to 325 nm). This indicates that
DNP–P25 not only promotes the photodegradation of the
parent pollutant, but also inhibits the generation of the
highly toxic intermediates, being favorable to the complete
mineralization of PCP.
Because kPCP–P25/kP25 = 0.79 o 1, PCP is not an appropriate
template for preparing MIP-coated photocatalysts to remove
the target pollutant PCP. This is possibly due to either the
poor interaction between the functional monomer and PCP
molecules, or a very limited solubility of PCP in the solution of
synthesis. The above experimental results demonstrate that
some organic compounds cannot be used as the templates for
Secondly, the reaction products were identified with GC/MS
by using 1,2-dibutylphthalate (DBP) as the internal standard
after the solution was esterified with acetic anhydride and then
extracted with heptane. The GC/MS chromatograms for the
PCP solutions irradiated over P25 and DNP–P25 for 21 min
are shown in Fig. 3. Besides dichlorophenol (DCP), many
aromatic intermediates and by-products are found after
the degradation of PCP over P25, which include TeCHQ,
tetrachlorocatechol (TeCC), acetoxypentachlorophenyl (APCP),
tetrachlorodibenzo-p-dioxin (TeCDD), pentachlorodibenzo-
p-dioxin (PCDD), hexachlorodibenzo-p-dioxin (HexCDD),
heptachlorodibenzo-p-dioxin (HepCDD) and octachlorodibenzo-
p-dioxin (OCDD). These intermediates are more stable and
harmful than PCP to human beings.37 Thus, the PCP photo-
degradation on P25 may increase the toxicity of the solution,
which must be forbidden from the view point of environmental
protection. Excitingly, the toxic aromatic intermediates are
almost inhibited in the case of DNP–P25. Therefore, the
photodegradation by using DNP–P25 is a safe and green
approach of removing PCP.
Fig. 1 Kinetics of direct photolysis and photocatalytic degradation of
PCP (c0 30 mg Lꢁ1, pH 11) over different photocatalysts. The rate
constant of the direct photolysis is 0.0112 minꢁ1, and that over P25,
PCP–P25, NIP–P25, 4-NP–P25, DNP–P25 and TNP–P25 is 0.0254,
0.0198, 0.0158, 0.0378, 0.0699 and 0.0332 minꢁ1, respectively.
Thirdly, the accumulation of Clꢁ ions and main organic
acids was also monitored during the photodegradation of PCP
by using ion-chromatography. Fig. 4 shows the profile of
ꢀc
This journal is The Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2009 New J. Chem., 2009, 33, 2278–2285 | 2281

View Article Online
Fig. 4 Effects of illumination time on the degradation and dechlorina-
tion of PCP (c0 5 mg Lꢁ1) over P25 and MIP–P25 (i.e. DNP–P25).
Table 1 gives the distribution of the intermediates during the
PCP photodegradation. Here, the intermediates include
total inorganic carbon (TIC), organic acids, and aromatic
intermediates. The generated organic acids (such as formic,
acetic, and dichloromaleic acids) were determined by ionic
chromatography (Fig. S7, w), and the TOC of the solution was
also measured. TIC was simply obtained by subtracting
the initially added TOC from the measured TOC after the
degradation, and the amount of the newly generated aromatic
intermediates was obtained by subtracting the measured TOC
from the measured amount of simple organic acids detected.
It is seen from Table 1 that after an illumination of 21 min,
DNP–P25 converted 95.67% of the carbon in the initial PCP
solution into inorganic carbon (39.65%) and small organic
acids (52.58%) with a small production of aromatic inter-
mediates (3.44%, including DCP and TCP), whereas the neat
P25 converted only 75.31% of PCP into inorganic carbon
(18.07%) and small organic acids (29.03%) with a large
production of aromatic intermediates (28.21%). The much
depressed generation of more toxic aromatic intermediates by
using DNP–P25 is consistent with the above-discussed
GC/MS analysis. This strongly demonstrates that the ring of
PCP is cleaved much more rapidly and completely during the
degradation over DNP–P25 than over P25.
Fig. 2 UV-visible absorption spectra of the PCP solution (5 mg Lꢁ1
,
pH 11) recorded during the photodegradation over DNP–P25 (a) and
P25 (b).
As discussed above, a fairly large amount of more toxic
and stable aromatic intermediates are generated during the
photodegradation of PCP over P25, indicating that the
photocatalytic degradation is not a safe approach for removing
PCP. By using DNP–P25, the PCP degradation is drastically
accelerated and the generation of aromatic intermediates is
significantly inhibited. Hence, the photocatalytic degradation
can become a safe approach of removing PCP by using
DNP–P25. To confirm it, the toxicity of the PCP solution
during its photodegradation on P25 and DNP–P25 was
monitored by using the luminescent bacterium Q67, and the
inhibition ratio (E) can represent the toxicity of the solution.
As shown in Fig. 5, the toxicity of the solution decreased
slightly during the degradation of PCP over P25, which
Fig. 3 GC/MS chromatograms of the heptane extract of the reaction
mixtures degraded over P25 (1) and DNP–P25 (2) at pH 11 for 21 min
of irradiation. Esterification of the solution was performed with acetic
anhydride before the heptane extraction.
Clꢁ ions in the PCP solutions during the degradation, where
Clꢁ ions were generated from the dechlorination of PCP. The
releasing of Clꢁ ions is rapid in the initial stage and slow
in the later stage on P25, being similar to the observation of
Mills and Hoffmann.35 In comparison with P25, DNP–P25
yields much faster dechlorination of PCP, and gives an
almost complete dechlorination of PCP after 50 min of
illumination. This further confirms that DNP–P25 can
accelerate the degradation and dechlorination of PCP.
reflects
a combination of the toxicity-decreasing effect
from the removal of PCP and the toxicity-increasing effect
from the generation of aromatic intermediates. However,
the PCP solution was rapidly detoxified by DNP–P25, which
is attributed to the accelerated removal of PCP and the
inhibited formation of more harmful aromatic intermediates.
ꢀc
This journal is The Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2009
2282 | New J. Chem., 2009, 33, 2278–2285

View Article Online
Table 1 Carbon mass balance (%) in the 5 mg Lꢁ1 PCP solution system after irradiation of 6 min and 21 min over P25 and DNP–P25
Over P25
Over DNP–P25
After 6 min
After 6 min
After 21 min
After 21 min
State of carbon
TIC
Formic acid
7.55
1.12
18.07
1.55
8.06
1.33
39.65
2.21
Acetic acid
Oxalic acid
Dichloromaleic acid
Residual PCP
Total of detected carbon
New aromatic toxicant
13.54
2.43
3.12
68.87
95.07
4.93
13.72
6.20
7.56
24.69
71.79
28.21
33.34
3.76
15.08
38.25
99.82
0.08
21.82
7.95
19.86
4.33
96.56
3.44
Accordingly, when the concentration of DNP was increased
from 0 to 10 mg Lꢁ1, a significant decrease of nmax was found
from 1.74 minꢁ1 to 0.79 minꢁ1. However, the addition
of DNP did not affect Km, which confirms that DNP is
working through a traditional reversible and noncompetitive
(non-repulsive) mechanism of inhibition. This is very important;
because DNP–P25 is prepared by using DNP as the template,
it is reasonable that DNP–P25 can distinguish DNP from PCP
molecules in the solution by the recognition ability of the MIP
layer. If the molecular recognition ability is perfect, DNP–P25
cannot be used to degrade PCP. It is fortunate that the
footprint cavities in the MIP layer on the surface of DNP–P25
can be shared by both DNP and PCP for their adsorption and
then degradation. This indicates that PCP molecules are able
to be adsorbed on DNP–P25 almost as easily as the used
template DNP, which is confirmed further by the evaluated
affinity of PCP and DNP over DNP–P25.
Fig. 5 Variation of the inhibition ratio (E) of the PCP (c0 30 mg Lꢁ1
,
pH 11) solution to Q67 during its photocatalytic degradation over
P25 (1) and DNP–P25 (2).
3.5 Tentative explanation of the enhanced degradation of PCP
over DNP–P25
To clearly demonstrate the affinity of PCP and DNP over
DNP–P25, eqn (1) can be simply changed to eqn (2), being
named as the Dixon equation.40
In the suspension of DNP–P25, the interaction between the
target (PCP) and the MIP substrate results in the formation of
a moderately stable complex (PCP–MIP) between the target
and MIP. Under the photocatalytic action of TiO2, the
PCP–MIP complex is oxidized, leading to the degradation of
PCP and the recovery of MIP. This is similar to the process of
enzyme catalysis. In order to explain the enhanced degradation
of PCP over DNP–P25, the photocatalytic degradation of
PCP at concentrations ranging from 2 to 50 mg Lꢁ1 was
conducted over DNP–P25, and the initial rate (v, in the time
window from 0 to 20 min) for the degradation of PCP was
obtained. As a control, 5 or 10 mg Lꢁ1 of the used template
DNP was added as an inhibitor to the solutions of PCP, and
the initial rate of PCP degradation was obtained similarly. The
experimental data were fitted with Lineweaver–Burk double
reciprocal plots (vꢁ1 B c0ꢁ1), as shown in Fig. 6. It is found
from Fig. 6 that the slopes of the reciprocal plots of the
rate vs. substrate concentration exhibit the characteristics of
noncompetitive inhibition.38,39 The inhibition constants were
calculated from eqn (1),
ꢀ
ꢁ
ꢀ
ꢁ
1
1
Km
1
Km
¼
1 þ
½Iꢄ þ
1 þ
ð2Þ
n
n
maxKi
½cꢄ
nmax
½cꢄ
Ki for the photocatalysis can be calculated by using eqn (2).
Under the experimental conditions, the Ki value of the used
template (DNP) over DNP–P25 was obtained as 8.85, whereas
the Km value of the substrate (PCP) was evaluated as 19.88.
Thus, the affinity (1/Km) of PCP on DNP–P25 is only slightly
ꢀ
ꢁ
ꢀ
ꢁ
1
Km
½Iꢄ
1
1
½Iꢄ
¼
1 þ
þ
1 þ
ð1Þ
n
nmax
Ki ½cꢄ nmax
Ki
where v is the initial reaction rate of PCP degradation, nmax is
the maximal rate of PCP degradation, [c] is the concentration
of PCP, [I] is the inhibitor concentration, Km is the Michaelis–
Menten constant, and Ki represents the dissociation constants
for the enzyme–inhibitor (EI) complex.
Fig. 6 Lineweaver–Burk double reciprocal plots for the photo-
catalytic degradation of PCP at different concentrations in the
presence of DNP (as an inhibitor) at concentrations of (1) 10,
(2) 5 and (3) 0 mg Lꢁ1
.
ꢀc
This journal is The Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2009 New J. Chem., 2009, 33, 2278–2285 | 2283

View Article Online
weaker than that (1/Ki) of DNP on DNP–P25. This high
affinity leads to the enhanced photodegradation of PCP on
DNP–P25.
Like previously reported MIP-coated TiO2 photocatalysts,
DNP–P25 promoted the photocatalytic degradation of the
template molecules of DNP and reduced the accumulation of
degradation intermediates, but the types of the intermediates
generated in the DNP degradation on DNP–P25 were almost
the same as that on neat P25 (Fig. S8 and S9, ESIw). This is
very different from the degradation of PCP on DNP–P25,
because the numbers of the degradation intermediates produced
on DNP–P25 are much lower than that on neat P25 (Fig. 3).
This suggests that the mechanism of the PCP degradation on
DNP–P25 is different from that on neat P25. The unusual
behavior for the degradation of PCP over DNP–P25 should be
related to the interaction between PCP molecules and the
amino groups fixed in the footprint cavities of the MIP layer.
To understand it, the functional amino groups in the foot-
print cavities were acetylated via acetic anhydride, and the
acetylated DNP–P25 was used to photocatalytically degrade
PCP (5 mg Lꢁ1). It was found that the degradation rate of
PCP over the acetylated DNP–P25 was considerably slower
than that over DNP–P25 without the acetylation. This
indicates that the amino groups in the footprint cavities are
very important to binding PCP molecules. Moreover, by
observing the absorption of the solution from 235 nm to
435 nm during the photocatalytic degradation of PCP, which
mainly consisted of the toxic compounds in the solution (such
as remaining PCP and the generated aromatic intermediates),
it was found that the amounts of the toxic compounds in the
solution had an increasing order on the photocatalysts of
DNP–P25 o TNP–P25, 4-NP–P25, o NIP–P25 o P25. This
indicates that the functional groups around the cavities of
the MIP greatly affect the inhibition of the generation of the
toxicants, and only the cavities tailored with DNP as the
template are appropriate.
Scheme
1
Mechanisms of the photodegradation of PCP over
different photocatalysts.
PCP molecules. This cannot tightly fix the PCP molecules in
the cavities, resulting in the generation of various inter-
mediates like in the cases of P25 and 4-NP–P25.
However, in the case of DNP–P25, the PCP molecules can
be adsorbed over the imprinted cavities on the MIP layer via
the formation of a moderately stable complex (PCP–MIP)
between PCP and MIP. In this way, the PCP molecules are
fixed into a well-defined micro environment (Scheme 1d). As
the hydrogen-bonding interaction between DNP and MIP is
dissymmetric, the electron atmosphere of PCP is thus sloped
to the left side, meaning that the ring of PCP is easy to open.
On the other hand, only one Cl atom of the adsorbed PCP
molecule is not protected by amino groups; the oxidizing
species will attack the molecule via this Cl atom, making the
C–Cl and C–C bonds be cleaved more easily here. The
enhanced splitting of the benzene ring inhibits the formation
of more harmful aromatics, due to its preventing the dimer-
ization and/or coupling of the initial intermediates from the
degradation of PCP. In this way, DNP–P25 can select a green
and safe pathway to remove PCP.
Based on above discussion, mechanisms of the photo-
catalytic degradation of PCP over the different catalysts are
proposed in Scheme 1. In the case of neat P25 TiO2, PCP
molecules are simply adsorbed on the surface of P25, and
chemical attack is allowed from almost all directions, leading
to easy generation of complicated degradation intermediates
(Scheme 1a). The rather loosely adsorbed degradation
intermediate may diffuse into the solution before its further
degradation, and then convert into more harmful by-products
as in the report of Kim and coworkers.36 Similar situations
will occur when NIP–P25 is used instead of neat P25. When
4-NP is used as pseudo template, the footprint cavities of
4-NP–P25 are considered to be formed with three amino
groups (Scheme 1b). Once the PCP molecules are rebound to
the adsorption sites at these imprinted cavities, only the –OH
and two –Cl could bind with the MIP, the other directions of
PCP are free and can be attacked by the photo-generated
oxidizing species such as OH radicals. As the oxidizing species
can attack the ring of PCP bilaterally, the degradation of PCP
over 4-NP–P25 may be similar to that over neat P25 TiO2.
When TNP is applied as the template, the footprint cavities of
TNP–P25 are composed of seven amino groups (Scheme 1c),
making the size of the cavities much larger than the size of the
4. Conclusion
The photocatalytic degradation of PCP over neat TiO2
not only was slow, but also produced various highly toxic
intermediates. To achieve a fast and safe removal of PCP from
aqueous solutions with complete inhibition of the generation
of the toxic intermediates, DNP–P25 was synthesized as a
MIP-coated photocatalyst by using molecular imprinting
technique with DNP as a pseudo template. Relative to naked
P25 TiO2, the pseudo MIP-coated photocatalyst DNP–P25
significantly promoted the dechlorination and degradation of
PCP due to its enzyme-like catalytic effects. Due to the similar
chemical structures of PCP and DNP, the affinity of PCP over
the photocatalyst DNP–P25 was comparable with that of the
used template DNP. More interestingly, the generation of
highly toxic intermediates was greatly depressed during
the photocatalytic degradation of PCP over DNP–P25. The
unique behavior of PCP degradation over DNP–P25 was
related to the special interactions between the adsorbed PCP
ꢀc
This journal is The Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2009
2284 | New J. Chem., 2009, 33, 2278–2285

View Article Online
molecules and the amino groups at the footprint cavities of
the MIP layer. From the view point of removing highly-toxic
low-level organic pollutants using a photocatalytic degradation
technique, the findings in the present work are very suggestive
because it is possible to prepare a MIP-coated TiO2 photo-
catalyst by using an appropriate structural analogue of the
target pollutant as the template, and then use it to photo-
catalytically degrade the target pollutant without the generation
of highly toxic intermediates, being confirmed as a green and
safe approach.
14 X. Shen, L. Zhu, H. Yu, H. Tang, S. Liu and W. Li, New J. Chem.,
2009, 33, 1673.
15 X. Shen, L. Zhu, C. Huang, H. Tang, Z. Yu and F. Deng, J. Mater.
Chem., 2009, 19, 4843.
16 J. P. Wilcoxon, J. Phys. Chem. B, 2000, 104, 7334.
17 K. Hoshino, R. Watanabe and M. Takano, J. Biotechnol., 2008,
136, S704.
18 D. Boyle, J. Environ. Manage., 2006, 80, 380.
19 S. Singh, R. Chandra, D. K. Patel, M. M. K. Reddy and V. Rai,
Bioresour. Technol., 2008, 99, 5703.
20 E. L. Puplampu and D. K. Dodoo, J. Photochem. Photobiol., A,
2000, 135, 81.
21 J. Hong, D. Kim, C. Cheong, S. Jung, M. Yoo, K. Kim, T. Kim
and Y. Park, Anal. Sci., 2000, 16, 621.
22 S. Vollmuth, A. Zajc and R. Niessner, Environ. Sci. Technol., 1994,
28, 1145.
Acknowledgements
23 J. Xue and J. Wang, J. Environ. Sci., 2008, 20, 1153.
24 M. Sung, S. Z. Lee and C. P. Huang, J. Contam. Hydrol., 2008,
98, 75.
25 L. K. Weavers, N. Malmstadt and M. R. Hoffmann, Environ. Sci.
Technol., 2000, 34, 1280.
26 M. A. Oturan, N. Oturan, C. Lahitte and S. Trevin, J. Electroanal.
Chem., 2001, 507, 96.
27 Z. Gao, S. Yang, C. Sun and J. Hong, Sep. Purif. Technol., 2007,
58, 24.
This research was supported by the National Natural Science
Foundation of China (Grant Nos. 20677019 and 20877031)
and the State Key Laboratory of Pollution Control and
Resource Reuse Foundation (No. PCRRF08007). The
Analytical and Testing Center of Huazhong University
of Science and Technology is thanked for its help in the
characterization of the catalysts.
28 Y. Guo, X. Quan, N. Lu, H. Zhao and S. Chen, Environ. Sci.
Technol., 2007, 41, 4422.
29 J. Gunlazuardi and W. A. Lindu, J. Photochem. Photobiol., A,
2005, 173, 51.
References
¨
30 L. G. Oberg and C. Rappe, Chemosphere, 1992, 25, 49.
31 W. F. Jardim, S. G. Moraes and M. M. K. Takiyama, Water Res.,
1997, 31, 1728.
32 W. Liu, X. Quan, Q. Cui, M. Ma, S. Chen and Z. Wang,
Ecotoxicol. Environ. Saf., 2008, 71, 267.
33 Y. Zhang, S. Liu, X. Song and H. Ge, Ecotoxicol. Environ. Saf.,
2008, 71, 880.
1 A. Leonhardt and K. Mosbach, React. Polym., Ion Exch.,
Sorbents, 1987, 6, 285.
2 K. M. Shokat, C. J. Leumann, R. Sugasawara and P. G. Schultz,
Nature, 1989, 338, 269.
3 R. Weinstain, R. A. Lerner, C. F. Barbas, III and D. Shabat,
J. Am. Chem. Soc., 2005, 127, 13104.
4 K. Polborn and K. Severin, Chem. Eng. J., 2000, 6, 4604.
5 A. Jakubiak, B. N. Kolarz and J. Jezierska, Macromol. Symp.,
2006, 235, 127.
34 J. Li, L. Zhu, Y. Wu, H. Yutaka, A. Zhang and H. Tang, Polymer,
2006, 47, 7361.
6 M. Emgenbroich and G. Wulff, Chem.–Eur. J., 2003, 9, 4106.
7 D. Lavabre, J. Micheau, J. R. Islas and T. Buhse, J. Phys. Chem.
A, 2007, 111, 281.
35 G. Mills and M. R. Hoffmann, Environ. Sci. Technol., 1993, 27,
1681.
36 J. Kim, K. Choi, I. Cho, H. Son and K. Zoh, J. Hazard. Mater.,
2007, 148, 281.
37 A. Hirvonen, M. Trapido, J. Hentunen and J. Tarhanen, Chemo-
sphere, 2000, 41, 1211.
38 L. Casella, M. Gullotti, E. Monzani, L. Santagostini,
G. Zoppellaro and T. Sakurai, J. Inorg. Biochem., 2006, 100, 2127.
39 M. C. Pirrung, H. Liu and A. T. Morehead, Jr., J. Am. Chem. Soc.,
2002, 124, 1014.
8 O. Ramstrom and K. Mosbach, Curr. Opin. Chem. Biol., 1999, 3,
759.
¨
9 G. Wulff, Chem. Rev., 2002, 102, 1.
10 Y. Paz, C. R. Chim., 2006, 9, 774.
11 M. Fukushima and K. Tatsumi, Environ. Sci. Technol., 2005, 39,
9337.
12 X. Shen, L. Zhu, J. Li and H. Tang, Chem. Commun., 2007, 1163.
13 X. Shen, L. Zhu, G. Liu, H. Yu and H. Tang, Environ. Sci.
Technol., 2008, 42, 1687.
40 M. Dixon and E. C. Webb, Enzymes, Academic Press, New York,
1979.
ꢀc
This journal is The Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2009 New J. Chem., 2009, 33, 2278–2285 | 2285