A detailed kinetic study of the direct photooxidation of 2,4,6-trichlorophenol

Article abstract of DOI:10.1016/j.jphotochem.2016.07.025

The direct photodegradation of 2,4,6-trichlorophenol (TCP) by UV–vis light was studied in aqueous solution in order to analyze the mechanism of the photochemical process and to determine the kinetic parameters including the quantum yield. Based on initial rate studies at different overall volumes and illumination patterns, it was proved that the rate of the process is directly proportional to the intensity of irradiating light. A significant, but moderate acceleration of the reaction rate with increasing temperature was revealed between 5.0 and 35.0?°C, which could be interpreted readily by assuming that the excited state of TCP is involved in two competing processes. High pressure liquid chromatography and mass spectrometry provided us information on the nature of the intermediates and the products formed. 2,6-Dichloro-1,4-benzoquinone, 3,5-dichloro-2-hydroxy-1,4-benzoquinone and 2,6-dihclorohydroxyquinone were detected as products and/or intermediates, and there were also hints of the formation of 3,5-dichlorobenzene-1,2-diol and 3,5-dichloro-1,2-benzoquinone. A possible degradation mechanism is proposed to interpret the kinetic findings.

Full text of DOI:10.1016/j.jphotochem.2016.07.025

Journal of Photochemistry and Photobiology A: Chemistry 330 (2016) 71–78
Contents lists available at ScienceDirect
Journal of Photochemistry and Photobiology A:
Chemistry
journal homepage: www.elsevier.com/locate/jphotochem
A detailed kinetic study of the direct photooxidation of
2
,4,6-trichlorophenol
Jose Ángel Pino-Chamorro, Tamás Ditrói, Gábor Lente*, István Fábián
Department of Inorganic and Analytical Chemistry, University of Debrecen, P.O.B. 21, H-4010, Debrecen, Hungary
A R T I C L E I N F O
A B S T R A C T
Article history:
Received 26 January 2016
Received in revised form 26 July 2016
Accepted 26 July 2016
The direct photodegradation of 2,4,6-trichlorophenol (TCP) by UV–vis light was studied in aqueous
solution in order to analyze the mechanism of the photochemical process and to determine the kinetic
parameters including the quantum yield. Based on initial rate studies at different overall volumes and
illumination patterns, it was proved that the rate of the process is directly proportional to the intensity of
irradiating light. A significant, but moderate acceleration of the reaction rate with increasing temperature
Available online 27 July 2016
Keywords:
was revealed between 5.0 and 35.0 C, which could be interpreted readily by assuming that the excited
ꢀ
Quantitative photochemistry
Light intensity dependence
Chlorophenols
Temperature dependence
Photochemical mechanism
Kinetics
state of TCP is involved in two competing processes. High pressure liquid chromatography and mass
spectrometry provided us information on the nature of the intermediates and the products formed. 2,6-
Dichloro-1,4-benzoquinone, 3,5-dichloro-2-hydroxy-1,4-benzoquinone and 2,6-dihclorohydroxyqui-
none were detected as products and/or intermediates, and there were also hints of the formation of
3,5-dichlorobenzene-1,2-diol and 3,5-dichloro-1,2-benzoquinone. A possible degradation mechanism is
Photodegradation
proposed to interpret the kinetic findings.
2
,4,6–trichlorophenol
ã 2016 Elsevier B.V. All rights reserved.
Polychromatic light
Mechanism of reaction
Quantum yield
1
. Introduction
form colorless solutions), their direct photolysis was a lot less
intensely investigated [3–5]. The iodine-sensitized photooxidation
was also considered as a possible alternative [14]. An interesting
and unintuitive observation was also made during a kinetic study
of the chemical oxidation of chlorophenols: the rate of reaction is
influenced by light, even if it is in the visible range [8]. The reason
for this phenomenon was understood to be the photochemical
transformations of quinones, which form as intermediates in
chlorophenol oxidation [8,23,24]. Therefore, an adequate kinetic
and mechanistic study of the chemical oxidation of chlorophenols
must be done in the dark in order to separate the thermal and
photolytic components of the process [9,11,12].
The accelerating effect of light on chlorophenol oxidation also
implies that their photochemical properties must be thoroughly
investigated together with the known intermediates of the
process. Recently, experimental techniques were developed for
the online monitoring of photochemical processes using diode
array spectrophotometers alone or with external light sources
[9,25–29]. These techniques use online monitoring and can yield
kinetic data on photoreactions that have very high time resolution
and thus be evaluated by up-to-date kinetic analysis methods.
They facilitated detailed studies on direct photolytic processes
Chlorophenols are common and persistent environmental
pollutants produced in a number of industrial processes. They
are not decomposed by usual waste water treatment techniques,
and the development of suitable degradation methods has
attracted the attention of a large and diverse research community
[1–22]. The use a specifically designed iron complexes for the
activation of hydrogen peroxide seems to be a particularly
successful strategy [2,7,11]. In these cases, no dioxins are produced
among the oxidation products [7], which is a major concern with
other waste treatment methods, for example the direct incinera-
tion of solid chlorophenols. The use of ozone [5,6], Fenton’s reagent
[5], peroxidase enzyme [13], electrooxidation [21], or even simple
absorption processes [22] are also possible alternatives in the
abatement of chlorophenol-containing waste.
Illumination with UV or visible light can also assist such
processes. This is typically done using heterogeneous photo-
catalysts such as TiO
Because of the limited light absorption of chlorophenols (which all
2 3 4
[1,15,19], spinels [18,20] and g-C N [17].
[23,29] and photochemically initiated chain reactions [25]. The
*
Corresponding author.
E-mail address: lenteg@science.unideb.hu (G. Lente).
http://dx.doi.org/10.1016/j.jphotochem.2016.07.025
010-6030/ã 2016 Elsevier B.V. All rights reserved.
1

7
2
J.A. Pino-Chamorro et al. / Journal of Photochemistry and Photobiology A: Chemistry 330 (2016) 71–78
experience gained in a number of different studies was summa-
rized in a review article a few years ago [26].
TCP and its degradation products were separated and analyzed
by a Shimadzu LC-10AP HPLC system equipped with a Shimadzu
In this paper, we report a detailed kinetic study on the
photochemical reaction of 2,4,6-trichlorophenol, which has the
widest industrial applications and is the most important among
the chlorophenols [7,11,14]. A new, custom-built photoreactor was
used to carry out this quantitative study, which improves the
applicability of online monitoring of photochemical reactions
compared to the earlier instruments used for this purpose [26].
SPD-M10A diode array detector, a Luna C18 (2) (3 mm)
100 ꢁ 4.6 mm reverse phase column and using 7:3 (V/V) mixture
of methanol and water as the mobile phase. The flow rate was set at
1.0 mL/min. The intermediates formed from TCP degradation were
identified by comparing their retention time and spectrum with
the external standards purchased from commercial sources. Under
the conditions described, the retention times of the external
standards were: TCP: 10.65 min, 2,6-dichloroquinone (DCQ):
2.75 min.
2. Experimental section
ESI–MS measurements were carried out with a Bruker micrO-
2.1. Materials
TOF
Q
mass spectrometer equipped with a quadrupole and time-of-
flight (Q-TOF) analyzer in positive ion mode. The nebulizer
Solid 2,4,6-trichlorophenol (TCP) was purchased from Sigma-
pressure was set to 6 psi (0.41 bar), N
2
was used as drying gas at
ꢀ
Aldrich and used as received. The concentration of TCP in our
4 L/min, the temperature was maintained at 160 C, and the applied
capillary voltage was ꢂ3.0 kV. The typical error of mass
measurement was below 0.005 in the m/z range from 80 to 600.
Spectra were accumulated and recorded by a digitizer at a rate of
2 GHz.
ꢂ4
experiments was typically around 2.0 ꢁ10 M. All the experi-
ments were carried out in 0.010 M phosphate and borax buffers at
ꢀ
2
5.0 C (except some experiments to study the dependence on
temperature). The buffer stock solutions were prepared using
sodium phosphate dihydrate (Reanal) and titrated to pH 7.04 with
NaOH purchased from VWR, and using disodium tetraborate
3. Results and discussion
(
Reanal) at pH 9.2. All stock solutions and samples were prepared
with double deionized, ultrafiltered and distilled water from a
3.1. Preliminary observations
Millipore Q system.
In the pH range used in the present study (pH = 7.04–9.2), 2,4,6-
2.2. Instruments and software
trichlorophenol has two UV bands centered at 254 and 312 nm that
a
are characteristic of the dissociated anionic form (pK = 6.15) [9,12].
The excitation light source (AvaLight-LDXE), the detecting light
The initial observations confirm that the photoreaction can be
driven even by a commercial diode array spectrophotometer, the
photochemical use of which is now well documented [26]. The
photodegradation of TCP was followed in a custom built photo-
reactor as descibred in the Experimental section. A cuvette
containing a solution of TCP was illuminated and the UV–vis
absorption spectrum was scanned repeatedly in the range of 230–
700 nm at pH 7.05 and 9.25 using phosphate and borax buffers at
30 s intervals. The photooxidation of TCP at these two pH values
had similar rates. An earlier research report found a large
difference between the TCP oxidation rates at pH 7.4 and 9.0
[13]. However, that must have been caused by the properties of the
peroxidase oxidant used in that study [13] as TCP itself undergoes
source (AvaLight DHc), the detector (Avaspec-ULS2048L-TEC-RS)
and all other optical cables of the photoreactor used in this study
were purchased from Avantes, Netherlands. The light source
(
excitation lamp) has a maximum brightness of 385 mW optical
output in the 170–1100 nm range (Fig. S1 in the Supplementary
information). A FOS-1-Inline fiber optic switch was used, which
allowed us to switch on and off the focused polychromatic light
beam that provided the driving source of the photoreaction.
Moreover, the halogen and deuterium lamps of the detecting light
source can be independently switched on and off. The sample
holder used was equipped with a temperature controller Peltier
unit configured for absorbance measurement. The cell holder was
modified with vertically movable detection ports to avoid
scattered or fluorescent light entering the detector. The spectro-
photometer (detector) used was AvaSpec-ULS2048L-TEC-RS,
which is an internally cooled CCD detector with variable slit size.
Standard four–sided quartz cuvettes (optical path length:
a
no major change in this pH region (its pK is 6.15). Figs. S2 and S3
show experiments in the two different media and also confirm an
important point: the kinetic observations do not depend on the
concentration of the buffer used in any significant way, so the
buffers are most probably innocent in this case.
1.000 cm) were used in all experiments. The solution in the
cuvette was stirred using a 4 mm ꢁ 2 mm ꢁ 2 mm Teflon-coated
stirring rod with the built-in magnetic stirrer of the sample holder
to keep the solutions homogeneous at all times. The detailed
description and performance characterization of this novel
photoreactor was published in a recent technical note [30].
Ferrioxalate actinometry was performed following a published
method [28]. A flow of nitrogen gas was circulated through the cell
holder to avoid the condensation of water in the walls of the cell
during the temperature dependence study. In these experiments,
the cuvette was kept in the sample holder during 10 min before
starting the experiments to ensure that the solution temperature
was homogeneous. Kinetic traces were fitted and other least
squares analyses were carried out by the software Scientist [31].
0
Particularly, initial rates of absorbance change (v values) were
determined by using a polynomial probe function to fit the
experimental kinetic trace. The coefficient of the first order term of
the polynomial gives the initial rate in this process. Further details
of this method can be found in the literature of chemical kinetics
Fig. 1. Typical spectral changes in an irradiated aqueous solution of 2,4,6-TCP.
3
[
TCP] = 0.2 mM; [phosphate buffer] = 0.010 M; path length 1.000 cm; V = 3.00 cm ;
ꢀ
[32].
T = 25.0 C; overall reaction time: 300 min).

J.A. Pino-Chamorro et al. / Journal of Photochemistry and Photobiology A: Chemistry 330 (2016) 71–78
73
Typical time-resolved spectra measured during the photoreac-
tion of TCP are shown in Fig. 1. Only spectra obtained by 30 min
intervals are plotted in the figure for clarity, but the time resolution
of the original data set is much higher. Literature results indicate
that these bands correspond to the formation of DCQ [14,16], which
is one of the products of the process. Its subsequent transformation
into a mixture of 2,6-dichloro-hydroquinone (DCHQ) and DCHB
also detected. These two signals could probably be assigned to 2,6-
dichloro-hydroquinone (DCHQ) and 3,5-dichlorobenzene-1,2-diol.
Moreover, an interesting observation was the appearance of a new
peak at 1.62 min. This compound has a band centered at 260 nm,
which is identical with the observed band for the final product
detected in the photoreactor and could indicate the presence of
3,5-dichloro-1,2-benzoquinone. More explanation about these
compounds will be given in later sections.
[23,26] is a known photoreaction, but DCHB may in fact also form
as the immediate product of the photochemical oxidation of TCP.
The reactants and products could be monitored with considerable
selectivity because the absorption bands of 2,6-dichloroquinone
3.3. ESI–MS experiments
(
DCQ, yellow) and 3,5-dichloro-2-hydroxybenzoquinone (DCHB,
ESI–MS spectra were also used to identify the photodegradation
products of TCP. The ESI mass spectrum of TCP after 1 h irradiation
shows the presence of two signals, which can be identified based
on the peak m/z values and the characteristic isotopic patterns
(Fig. S6). The first is centered at m/z 176.9 and could be assigned to
2,6-dichloro-hydroquinone (DCHQ) and probably also to
3,5-dichlorobenzene-1,2-diol, which has the same m/z value.
These two compounds might correspond to peaks 2 and 3 in the
chromatogram shown in Fig. 3. The peak at m/z = 194.9 is attributed
to and TCP. Other compounds were not observed in the mass
spectrum after 5 h (Fig. S7). It is possible that the ESI technique is
not able to ionize quinones such as DCQ, 3,5-dichloro-2-hydroxy-
1,4-benzoquinone (DCHB) or 3,5-dichloro-1,2-benzoquinone,
which have been identified previously in this reaction as products
[23]. Similar ionization problems have already been reported with
studies involving 2,4-dichlorophenoxybutanoic acid and 2,4,5-
trichlorophenoxyacetic acid [33,34].
purple) were reasonably separated from TCP (colorless, only UV
absorption). The oxidation involves the appearance of two bands
centered at ca. 260 and 480 nm after 300 min. A very slow
disappearance of the band at 480 nm was observed after 24 h. Fig. 2
shows the kinetic traces at 260, 312 and 480 nm in a five-hour
experiment. The trace at 260 nm clearly displays a small initial
decrease in the absorbance and the curve recorded at 480 nm also
hints at an induction period. Therefore, it can be concluded that
there is at least one primary and at least one secondary
photoproduct.
In addition to the literature data already mentioned
[
14,16,23,26], other researchers reported the presence of some
different intermediates in the process of TCP photodegradation
17]. Thus, in order to verify the nature of the compounds in the
[
reaction mixture, HPLC and mass spectrometry was applied to
obtain qualitative information on this system.
3.2. HPLC experiments
3.4. Kinetic studies
The degradation products of TCP were monitored using HPLC
First, it is important to note that due to partial acid dissociation
of chlorophenols in aqueous solutions, the reaction schemes of the
TCP degradation need to account for the acid-base properties of
with UV–vis detection. The HPLC chromatograms and UV–vis
spectra of the solutions are shown in Fig. 3 (after 1 h of
illumination) and Fig. 4 (after 5 h of illumination). The final
products from TCP degradation were identified primarily by
comparing the UV–vis spectra with those of pure compounds
Figs. S4 and S5 in the Supplementary information). The Uv–vis
spectra of TCP [9,12,13], DCQ [9,12,13,23,24] and DCHB [23,24] are
all available from the literature.
After 1 h of irradiation, the chromatogram shows the appear-
ance of two peaks with retention times 2.29 and 3.93 min together
with the TCP peak with retention time 9.86 min. After 5 h,
practically complete degradation of TCP was observed. An increase
in the peak at 2.29 min and a decrease in the peak at 3.92 min were
a
this molecule. The pK of TCP is 6.15, and the Uv–vis spectra of the
protonated and dissociated forms are different. [12] So, in principle
the reaction pathways including the protonated (neutral) and
dissociated (anionic) forms may contribute to the overall reaction
rate. Since our aim was to use conditions that were close to the
possible applications in water treatment technologies (neutral pH),
we set pH = 7.04 with a phosphate buffer in the majority of the
experiments reported here. Under these conditions, about 88% of
TCP is in the dissociated (phenolate) form [6]. This form also
happens to display much more substantial absorption in the
wavelength range of illumination than the neutral TCP molecule.
Therefore, for all practical purposes, it was sufficient to assume
that our observations provide information on the photoreactions of
the phenolate ion. In the rest of the paper, the common convention
of coordination chemistry will be used: the abbreviation TCP will
refer collectively to the two different protonated forms of 2,4,6-
trichlorophenol with the understanding that the pH-dependent
speciation must be kept in mind at each specified pH.
(
In Fig. 2, the kinetic trace at 312 nm reflects primarily the
concentration change of TCP. However, the contributions of other
species to the absorbance at this wavelength are non-negligible as
evidenced by the non-zero final absorbance seen in Fig. 2. This
trace could be fitted to a pseudo-first order expression (exponen-
tial curve) with an acceptable precision and the observed rate
ꢂ2
ꢂ1
constant was determined to be k = (1.82 ꢃ 0.01) ꢁ 10 min . The
good fit and the value are both in reasonable agreement with the
observations reported by Shen et al. at the same pH [3], especially if
the different illumination intensities are taken into account.
However, it should be noted that a true pseudo-first order process
should give identical rate constants at all wavelengths of
monitoring [32]. Furthermore, a homogenous photochemical
process is actually never expected to have a simple first order
Fig. 2. Absorbance changes as a function of time in an irradiated aqueous solution
of TCP at selected wavelengths. The wavelengths chosen are the maximum of DCQ
(
black line, 260 nm), TCP (red line, 312 nm) and DCHB (green line, 480 nm).
Conditions are the same as in Fig. 1. (For interpretation of the references to colour in
this figure legend, the reader is referred to the web version of this article.).

74
J.A. Pino-Chamorro et al. / Journal of Photochemistry and Photobiology A: Chemistry 330 (2016) 71–78
Fig. 3. HPLC chromatogram (left) and UV–vis spectra (right) for TCP degradation after 1 h under UV–vis light. Peak 1: 1.18 (injection peak); peak 2: 2.29 min,
: 3.93 min, = 287 nm; peak 4: 9.87 min, = 287 and 293 nm. [TCP] = 0.20 mM, [phosphate buffer] = 0.010 M.
l= 285 nm; peak
3
l
l
0
Fig. 4. HPLC–chromatogram (left) and UV–vis spectra (right) for the TCP degradation after 5 h under UV–vis light. Peak 1: 1.18 (injection peak); peak 2: 1.62 min,
peak 3: 2.29 min, = 285 nm; peak 4: 3.92 min, = 287 nm; peak 5: 9.88 min, = 287 and 293 nm. [TCP] = 0.20 mM, [phosphate buffer] = 0.010 M.
l= 260 nm
l
l
l
0
rate equation. Even the simplest photochemical transformation
has the following rate expression [23,27,32]:
standard fluorometric cuvette (path length: 1.000 ꢁ 1.000 cm), i.e.,
the light intensity per volume can be varied by a factor of 3.5. This is
substantially larger than the ranges in previous studies [23,25,29]
and is an advantageous aspect of the photoreactor used in this
study.
Z
ꢀ
P
ꢄꢁ
d½Aꢄ
F
P;
l
F
l
e
A;
l
½Aꢄ
e
A;
l
l½Aꢄꢂ
e
i;
l
l½B
i
¼
ꢂ
1 ꢂ 10ꢂ
eA;l½Aꢄ þ ei;l½Biꢄd
P
l
dt
V
ð1Þ
0
The initial rate of absorbance change (v ) for TCP degradation
was measured using the data obtained during the first 30 min of
reaction at 312 nm and is shown as a function of the volume of the
sample in Fig. 5 [26]. The initial rate clearly decreases when the
sample volume increases, and the measured points fit to a
hyperbola very well, implying an inverse proportionality between
In this equation, A is the photoactive species, B
photoactive species that absorb light, is the spectral photon
flux of the illumination, is the differential quantum yield, V is
the volume of the reactor, l is the optical path length, whereas
and are the molar absorption coefficients of the respective
i
are various non-
F
P,l
F
l
e
A,l
e
i,l
species. The terminology used here follows the current recom-
mendations of IUPAC [35]. Given the complexity of Eq. (1), it must
be concluded that the fact that the kinetic traces are very close to
pseudo-first order at 310 nm is a consequence of coincidences
involving rate constants and absorption coefficients, thus, the
information content of the determined observed rate constant is
very complex. To avoid these problems, only initial rates were used
in the rest of this study as a source of quantitative information
about the rate equation of the photodegradation.
In order to explore how the intensity of light affects the kinetics
of the reaction, we have carried out different experiments, in which
the reaction volume was changed at constant irradiating light flux.
Through the volume change, the number of photons absorbed in
unit volume can be influenced and its controlled variation can be
used to study the dependence of the reaction rate on the
illumination intensity. This fact has been explained previously in
several recent publications from our group [23,25,26,29]. The
Fig. 5. Dependence of the initial rates the sample volume at 312 nm in the
photodegradation of TCP. [TCP] = 0.20 mM; [phosphate buffer] = 0.010 M; T = 25.0 C.
3
volume of the sample was changed from 1.0 to 3.5 cm in a
ꢀ

J.A. Pino-Chamorro et al. / Journal of Photochemistry and Photobiology A: Chemistry 330 (2016) 71–78
75
the rate and the volume. A logarithmic version of this plot is shown
in the Supplementary information as Fig. S8, where the points
define a straight line with a slope of ꢂ0.98 ꢃ 0.01. As known from
previous literature [23,25,26], these observations prove that the
reaction rate is directly proportional to the intensity of illumina-
tion.
To gain more information about this process, experiments with
interrupted illumination were performed where the illumination
ratio is defined as the time of illumination relative to the total time
of reaction. [25,26] The average reaction rate is expected to be
strictly proportional to the illumination ratio in
photochemical reaction. The measurements in this series differed
in the duration of illumination (t ) and dark periods (t ) by opening
and closing the shutter for controlled periods. In this way, the
duration of a full cycle, t , is given as t = t + t . The cycles of
illuminated and dark periods are repeated. Therefore, time values t
and t define an illumination pattern.
a purely
i
d
c
c
i
d
Fig. 7. Dependence of the average initial rates on the illumination ratio for the
i
photodegradation of TCP. [TCP] = 0.20 mM; [phosphate buffer] = 0.010 M; path
3
ꢀ
length 1.000 cm; V = 2.00 cm ; T = 25.0 C.
d
Kinetic traces recorded during the experiments using different
illumination patterns are shown in Fig. 6, where degradation of TCP
relative intensity of the exciting light at each wavelength. For this
purpose, we used an Analytik Jena SPECORD S600 diode array
spectrophotometer. The relative energy spectrum of the excitation
lamp was recorded and converted to an absolute scale by routine
ferrioxalate actinometry [28]. The intensity spectrum of the
excitation light is given in Fig. S1 of the Supplementary
information. The total photon flux of the illumination was
was followed by using continuous illumination (t
red trace) or a pattern with 10% illumination ratio (t
i
= 50 s, t
= 5 s, t
d
= 0 s;
d
= 45 s;
i
black trace). For the quantitative evaluation of the experimental
data, the average rate of absorbance change is used. In agreement
with our expectation, the average initial rate value decreases with
increasing the length the dark periods while keeping the cycle time
constant. In fact, the average initial rates of the two kinetic
experiments mentioned (red and black curves) differ by a factor of
determined to be
P
F = 62 nmol/s.
The photon count (N) in a photochemical experiment was
defined in one of earlier works as the part of the photon flux
absorbed by the photoactive species [23]. In the present case, the
initial photon count is especially easily calculated from the
intensity spectrum of the excitation light and the measured
absorbance values of the studied solution because TCP, the
photoactive species, is the sole absorbing species in the initial
solution. A plot of reaction rates as a function of photon count is
suitable for determining the quantum yield in a series of
experiments where the initial concentration of the photoactive
species is varied [23]. Fig. 8 shows that this plot is linear for the
process studied here, as expected based on earlier observations
about the direct proportionality between the rate and light
intensity.
10. Further experiments confirmed the direct proportionality
between the illumination ratio and the reaction rate, as shown in
Fig. 7. This finding indicates that the studied process is a relatively
simple, direct photochemical transformation without a consider-
able parallel thermal pathways (dark reactions) or the involvement
of photoinitiated chain processes. Should these latter factors
contribute to the observations, more complex dependencies would
be expected in the plot shown in Fig. 7 [25,26]. Fig. 6 also shows
that the reaction is not driven by only visible light because the
absorbance change is marginal when a cut-off filter is inserted into
the excitation light beam, which eliminates light below 395 nm
(
green trace). Moreover, it is notable that the absence of stirring
results in irreproducible, badly defined traces as inhomogeneity
develops in the solution (blue trace in Fig. 6). This effect and the
need for stirring in this type of photoreactor were emphasized in
an earlier publication [23].
In order to analyze the photoreaction driven by polychromatic
light in a quantitative way, it is very important to measure the
A straightforward method for the calculation of quantum yields
based on this plot was already published [23]. The slope (
a
) of the
fitted line in Fig. 8, (1.13 ꢃ 0.03) ꢁ 10 nmol , the volume of the
reactor (V) and the molar absorption coefficients ( ) of the
reactants and products need to be used in the following manner:
ꢂ4
ꢂ1
e
a
ꢂ
V
e
F
¼
ð2Þ
ð
e
TCP
DCQ Þl
Fig. 6. Kinetic traces measured in the photochemical oxidation of TCP. [TCP] =
3
0
.20 mM; [phosphate buffer] = 0.010 M; path length 1.000 cm; V = 2.00 cm ;
ꢀ
T = 25.0 C; only the visible light was used (filter cut-off at 395 nm) (green trace);
10% illumination ratio (black trace); 100% illumination ratio (red trace); and no
Fig. 8. Initial rate at 312 nm in the aqueous photoreaction of TCP as a function of
stirring (blue trace). (For interpretation of the references to colour in this figure
legend, the reader is referred to the web version of this article.).
photon counts. [TCP] = 0.20 mM; [phosphate buffer] = 0.010 M; path length = 1.000
cm; V = 2.00 cm ; T = 25.0 C.
3
ꢀ

7
6
J.A. Pino-Chamorro et al. / Journal of Photochemistry and Photobiology A: Chemistry 330 (2016) 71–78
ꢂ1 ꢂ1
The molar absorbance for TCP (
eTCP = 4202 ꢃ 87 M cm ) and
ꢂ1
ꢂ1
for DCQ (
eDCQ = 307 ꢃ 9 M cm ) were determined experimen-
tally at 312 nm by measuring the UV–vis spectra of TCP and DCQ
solutions in known concentration, respectively. These values agree
well with previously reported ones [10,13]. The quantum yield
F
= 0.057 was calculated for TCP degradation as a result. This value
is in agreement with those calculated in previous studies under
different conditions [4,5,14].
Another possible factor to be considered in a photochemical
process is the dependence on the temperature. In this manner, a
detailed study in the range of temperatures between 5.0 and
ꢀ
3
5.0 C was performed. Fig. 9 shows the absorbance-time traces at
different temperatures. The decomposition occurs faster with
increasing temperature, which may be somewhat unexpected as
the rates of simple photochemical reactions at constant reactant
compositions depend only on the illumination intensity but not on
temperature. The photon counts are unchanged as the absorption
spectra of TCP and DCQ are not influenced by this minor change in
temperature. This fact can be explained by an increase of the
quantum yield of a photochemical process when the temperature
is raised [5]. Actually, the low value of the quantum yield (0.057)
leaves the possibility of a temperature-dependent increase open.
The phenomenon is readily interpreted by assuming that the
excited form of TCP formed after light absorption is not only
involved in a chemical reaction, but an energy loss not resulting in
net chemical change (fluorescence, vibrational relaxation, an
internal conversion followed by other processes, etc.) is also
operative. This energy loss is most probably radiationless, as
independent experiments showed that TCP practically does not
show any fluorescence in aqueous solution (see Fig. S9 in the
Supplementary information). Therefore, the reaction is very likely
to obey the set of reactions:
Fig. 10. Dependence of the initial rate on temperature for the photodegradation of
TCP. [TCP] = 0.20 mM; [phosphate buffer] = 0.010 M; path length 1.000 cm;
3
ꢀ
V = 2.00 cm ; T = 5.0-35.0 C. The solid lines correspond to the fit of data using
Eq. (5).
2
and A and activation energies of Ea1 and Ea2. The temperature
dependence of the quantum yield is simply interpreted in this line
of thought as follows:
ꢂEa2=RT
k2
k1 þ k2
A e
2
F
¼
¼
ð3Þ
ꢂEa1=RT
ꢂEa2=RT
2
e
A
1
e
þ A
The temperature dependence of the initial reaction rate (r
0
, now
expressed as the time derivative of the amount of substance rather
than the v used earlier, the time derivative of absorbance) is then
described as:
0
N
r
0
¼
FN ¼
ð4Þ
ÞeðEa2
ꢂEa1Þ=RT
ðA
1
=A
2
þ 1
h
n
k
1
k
2
ð2Þ
1
represents radiationless energy
By introducing new parameter combinations, this formula can
be written in a much simplified form:
TCP! TCP ꢅ TCP ꢅ ! TCPTCP ꢅ ! ꢆ ꢆ ꢆ ! DCQ
In this scheme, rate constant k
r
max
loss, whereas k is the rate constant of the pathway that eventually
2
r0 ¼
ð5Þ
aeb=RT þ 1
leads to the formation of DCQ.
The dependence of the initial rates on the temperature is
illustrated in Fig. 10. In this plot, an important effect of the
temperature was observed. As the temperature increased from 5.0
to 35.0 C, the rate of TCP degradation increased to 2.14-fold.
Following some literature precedents [36–41], this dependence
Here, rmax is the maximum rate of the photoreaction that would be
detectable in the absence of the radiationless energy loss, a
represents the ratio of the two pre-exponential factors, whereas b
ꢀ
is the difference of the two activation. The values of r
max
could be
ꢀ
calculated from the experiments done at 25.0 C and was fixed in
the evaluation (1.412 nmol/s). The plot shows that the formula in
Eq. (5) gives an excellent interpretation of the observations. Non-
1 2
was interpreted by assuming that both k and k have Arrhenius
type temperature dependencies with pre-exponential factors A
1
linear least squares fitting was used to obtain the following
ꢂ1
parameter estimations: a = 0.0046 ꢃ 0.0021 and b = 20 ꢃ1 kJ mol
.
The positive value of b means that the pathway leading to the
chemical reaction has a higher activation energy than the
radiationless energy loss, whereas a is smaller than 1, which
1
implies that the pre-exponential factor of the k pathway is lower
than that of k . The excellent fit in Fig. 10 suggests that the simple
2
interpretation proposed in Eq. (2) works well on a quantitative
level as well. However, it should also be pointed out that this
explanation is quite possible but not inevitable. The observed
temperature effect on the photodegradation of TCP degradation
might also be due to secondary thermal reactions between the
primary active species (i.e. phenoxyl radicals, or phenyl cations)
and/or products formed during the overall course of reaction (i.e.
DCQ + TCP). These two alternative interpretations could probably
be distinguished based on quantitative measurements by deter-
mining the quantum yield of the primary photochemical process
by a technique with very high time resolution (such as laser flash
photolysis or possibly pulse radiolysis).
Fig. 9. Kinetic traces measured in the photochemical oxidation of TCP. [TCP] =
3
0
.20 mM; [phosphate buffer] = 0.010 M; path length 1.00 cm; V = 2.00 cm ;
ꢀ
ꢀ
ꢀ
ꢀ
T = 35.0 C (pink trace), 25.0 C (blue trace), 20.0 C (yellow trace), 15.0 C (green
ꢀ
ꢀ
trace), 10.0 C (red trace) and 5.0 C (black trace). (For interpretation of the
references to colour in this figure legend, the reader is referred to the web version of
this article.).
Based on the findings reported in this paper and previous
literature data [4,33,34,42–44], a detailed mechanism with the

J.A. Pino-Chamorro et al. / Journal of Photochemistry and Photobiology A: Chemistry 330 (2016) 71–78
77
Scheme 1. Proposed reaction scheme for the aqueous photodegradation of 2,4,6-trichlorophenol.
possible chemical structures is shown in Scheme 1. In this
mechanism, absorption of light produces the excited form of
TCP. As shown in the previous paragraphs, this excited state is
involved in two processes: one is the loss of excitation energy
without radiation and chemical reactions (its conversion to heat),
whereas the other yields new products. Presumably, internal
conversion to a triplet state precedes the chemical reactions and
benzoquinone, 3,5-dichloro-2-hydroxy-1,4-benzoquinone and
2,6-dihclorohydroxyquinone were detected as products of the
photodegradation of 2,4,6-trichlorophenol, the second and third
probably originating from the independently known aqueous
photoreaction of 2,6-dichloro-1,4-benzoquinone. In addition, the
formation of 3,5-dichlorobenzene-1,2-diol and 3,5-dichloro-1,2-
benzoquinone is also possible. The rate of the photochemical
process was systematically studied as a function of reactant
concentration, reaction volume, illumination ratio and tempera-
ture. These observations allowed us to postulate a mechanism, in
which the excited form of TCP undergoes radiationless energy loss
and a series of chemical reactions in a competitive manner.
Because of the competition, the quantum yield of reactant loss
shows a modest dependence on temperature, proposing an
Arrhenius-type equation interpreted these data reasonably well.
the k
internal conversion. Again, based on the previous knowledge
4,13,23,33,34,42–44], it seems likely that the first direct product of
the photoreaction is hydroquinone, so the process is
2
may very well represent the unimolecular rate constant of
[
a
a
substitution of a chlorine atom by a hydroxyl group. This
substitution may occur at two positions. Therefore, two products
are possible. Each of these hydroquinone products can easily be
oxidized to the corresponding quinone. This process is known to be
reasonably fast under neutral conditions in the presence of oxygen
[9,24]. Finally, the 2,6-dichloro-1,4-benzoquinone has a well-
Acknowledgements
known photoreaction, which produces a hydroxyquinone and the
hydroquinone [23].
The authors wish to thank the Hungarian Science Foundation
(NK 105156) for financial support. The research was supported by
the EU and co-financed by the European Social Fund under the
project ENVIKUT (TÁMOP-4.2.2.A-11/1/KONV-2012-0043). An
anonymous reviewer is also acknowledged for insightful remarks.
In addition, Scheme 1 also indicates the possible formation of
3,5-dichlorobenzene-1,2-diol and 3,5-dichloro-1,2-benzoquinone,
which could be minor by-products and could explain some of the
fine details of experimental observations. For example, 3,5-
dichloro-1,2-benzoquinone is an isomer of DCQ and might be
responsible for the appearance of the peak at retention time
Appendix A. Supplementary data
1.62 min in the chromatogram shown in Fig. 4 and may also cause
the initial decrease in absorbance shown in Fig. 2 on the kinetic
trace detected at 260 nm.
Supplementary data associated with this article can be found, in
the
online
version,
at
http://dx.doi.org/10.1016/j.
jphotochem.2016.07.025.
4. Conclusion
References
2
,4,6-Trichlorophenol was almost completely oxidized by a UV–
[
1] J.C. D'Oliveira, C. Minero, E. Pelizzetti, P. Pichat, Photodegradation of
vis radiation after 5 h. However, oxidation did not occur when only
visible light was used. The products were identified by the UV–vis
absorption properties, HPLC and ESI–MS. 2,6-Dichloro-1,4-
dichlorophenols and trichlorophenols in TiO aqueous suspensions: kinetic
2
effects of the positions of the Cl atoms and identification of the intermediates,
J. Photochem. Photobiol. A Chem. 72 (1993) 261–267.

7
8
J.A. Pino-Chamorro et al. / Journal of Photochemistry and Photobiology A: Chemistry 330 (2016) 71–78
[
[
[
2] A. Sorokin, B. Meunier, J.L. Séris, Efficient oxidative dechlorination and [23] G. Lente, J.H. Espenson, Photoreduction of 2,6-dichloroquinone in aqueous
aromatic ring cleavage of chlorinated phenols catalyzed by iron
sulfophthalocyanine, Science 268 (1995) 1163–1166.
3] Y.S. Shen, Y. Ku, K.C. Lee, The effect of light absorbance on the decomposition of
solution. Use of a diode array spectrophotometer concurrently to drive and
detect a photochemical reaction, J. Photochem. Photobiol. A Chem. 163 (2004)
249–258.
chlorophenols by ultraviolet radiation and u.v./H
2
O
2
processes, Water Res. 29
[24] É. Józsa, M. Purgel, M. Bihari, P.P. Fehér, G. Sustyák, B. Várnagy, V. Kiss, E. Ladó,
K. Äsz, Kinetic studies of hydroxyquinone formation from water soluble
benzoquinones, New J. Chem. 38 (2014) 588–597.
[25] I. Kerezsi, G. Lente, I. Fábián, Highly efficient photoinitiation in the cerium(III)-
catalyzed aqueous autoxidation of sulfur(IV). an example of comprehensive
evaluation of photoinduced chain reactions, J. Am. Chem. Soc. 127 (2005)
4785–4793.
(
1995) 907–914.
4] Y.I. Skurlatov, L.S. Ernestova, E.V. Vichutinskaya, D.P. Samsonov, I.V. Semenova,
I.Y. Rod’ko, V.O. Shvidky, R.I. Pervunina, T.J. Kemp, Photochemical
transformation of polychlorinated phenols, J. Photochem. Photobiol. A 107
(
1997) 207–213.
5] F.J. Benítez, J. Beltrán-Heredia, J.L. Acero, F.J. Rubio, Chemical decomposition of
,4,6-trichlorophenol by ozone, fenton’s reagent, and UV radiation, Ind. Eng.
[
[
[
2
[26] I. Fábián, G. Lente, Light induced multistep redox reactions: the diode array
spectrophotometer as a photoreactor, Pure Appl. Chem. 82 (2010) 1957–1973.
[27] M. Gombár, É. Józsa, M. Braun, K. Äsz, Construction of a photochemical reactor
combining a CCD spectrophotometer and a LED radiation source, Photochem.
Photobiol. Sci. 11 (2012) 1592–1595.
Chem. Res. 38 (1999) 1341–1349.
6] F.J. Benitez, J. Beltrán-Heredia, J.L. Acero, F.J. Rubio, Rate constants for the
reactions of ozone with chlorophenols in aqueous solutions, J. Hazard. Mater.
7
9 (2000) 271–285.
7] S. Sen Gupta, M. Stadler, C.A. Noser, A. Ghosh, B. Steinhoff, D. Lenoir, C. Horwitz,
K.W. Schramm, T.J. Collins, Rapid total destruction of chlorophenols by
activated hydrogen peroxide, Science 296 (2002) 326–328.
[28] T. Lehóczki, É. Józsa, K. Äsz, Ferrioxalate actinometry with online
spectrophotometric detection, J. Photochem. Photobiol. A Chem. 251 (2013)
63–68.
[
8] G. Lente, J.H. Espenson, Photoaccelerated oxidation of chlorinated phenols,
Chem. Comm. (2003) 1162–1163.
9] G. Lente, J.H. Espenson, A kinetic study of the early steps in the oxidation of
chlorophenols by hydrogen peroxide catalyzed by a water-soluble Iron(III)
porphyrin, New J. Chem. 28 (2004) 847–852.
[29] J. Kalmár, É. Dóka, G. Lente, I. Fábián, Aqueous photochemical reactions of
chloride, bromide, and iodide ions in a diode-array spectrophotometer.
Autoinhibition in the photolysis of iodide ion, Dalton Trans. 43 (2014) 4862–
4870.
[30] T. Ditrói, J. Kalmár, J.A. Pino-Chamorro, Z. Erdei, G. Lente, I. Fábián,
Construction of a multipurpose photochemical reactor with on-line
spectrophotometric detection, Photochem. Photobiol. Sci. 15 (2016) 589–594.
[31] SCIENTIST, Version 2.0; Micromath Software: Salt Lake City, UT, USA (1995).
[32] G. Lente, Deterministic Kinetics in Chemistry and Systems Biology, Springer,
2015.
[
[
10] M. Siam, G. Reiter, R. Hunziker, B. Escher, A. Karpfen, A. Simperler, D. Baurecht,
U.P. Fringeli, Evidence for heterodimers of 2,4,5-trichlorophenol on planer
lipid layers. A FTIR-ATR investigation, Biochim. Biophys. Acta 1664 (2004)
8
8–99.
[
11] G. Lente, J.H. Espenson, Oxidation of 2,4,6-trichlorophenol by hydrogen
peroxide. comparison of different iron-based catalysts, Green Chem. 7 (2005)
[33] M.P. Yurkova, I.P. Pozdnyakov, V.F. Plyusnin, V.P. Grivin, N.M. Bazhin, A.I.
Kruppa, T.A. Maksimova, A mechanistic study of the photodegradation of
herbicide 2,4,5-trichlorophenoxyacetic acid in aqueous solution, Photochem.
Photobiol. Sci. 12 (2013) 684–689.
[34] I.P. Pozdnyakov, P. Sherin, V. Grivin, V. Plyusnin, Degradation of herbicide 2,4-
dichlorophenoxybutanoic acid in the photolysis of [FeOH]2+ and [Fe(Ox)3]3-
complexes: a mechanistic study, Chemosphere 146 (2016) 280–288.
[35] S.E. Braslavsky, Glossary of terms used in photochemistry 3rd edition, Pure
Appl. Chem. 79 (2007) 293–465.
2
8–34.
[
[
[
[
[
12] A. Simon, C. Ballai, G. Lente, I. Fábián, Structure-reactivity relationships and
substituent effect additivity in the aqueous oxidation of chlorophenols by
cerium(IV), New J. Chem. 35 (2011) 235–241.
13] B.E. Sturgeon, B.J. Battenburg, B.J. Lyon, S. Franzen, Revisiting the peroxidase
oxidation of 2,4,6-trihalophenols: ESR detection of radical intermediates,
Chem. Res. Toxicol. 24 (2011) 1862–1868.
14] M. Hu, Y. Wang, Z. Xiong, D. Bi, Y. Zhang, Y. Xu, Iodine-sensitized degradation of
2
,4,6-Trichlorophenol under visible light, environ, Sci. Technol. 46 (2012)
[36] C. Gonzalez, J. Pincock, Temperature effects, Arrhenius activation parameters
and rate constants for the photochemical reactivity of cyano, boronato,
trifluoromethyl and methoxy substituted toluenes in the excited singlet state,
Photochem. Photobiol. 82 (2006) 301–309.
[37] N.G.C. Astrath, F.B.G. Astrath, J. Shen, J. Zhou, K.H. Michaelian, C. Fairbridge, L.C.
Malacarne, P.R.B. Pedreira, P.A. Santoro, M.L. Baesso, Arrhenius behavior of
hydrocarbon fuel photochemical reaction rates by thermal lens spectroscopy,
Appl. Phys. Lett. 95 (2009) 191902/1–191902/3.
9
005–9011.
15] E. Pino, M.V. Encinas, Photocatalytic degradation of chlorophenols on TiO
2
-
3
25 mesh and TiO -P25. An extended kinetic study of photodegradation under
2
competitive conditions, J. Photochem. Photobiol. A Chem. 242 (2012) 20–27.
16] S. Franzen, K. Sasan, B.E. Sturgeon, B.J. Lyon, B.J. Battenburg, H. Gracz, R.
Dumariah, R. Ghiladi, Nonphotochemical base-catalyzed hydroxylation of 2,6-
dichloroquinone by H
2012) 1666–1676.
17] H. Ji, F. Chang, X. Hu, Y. Wang, W. Qin, J. Shen, Photocatalytic degradation of
,4,6-trichlorophenol over g-C under visible light irradiation, Chem. Eng. J.
18 (2013) 183–190.
18] M. Kurian, D.S. Nair, A.M. Rahnamol, Influence of the synthesis conditions on
the catalytic efficiency of NiFe and ZnFe nanoparticles towards the wet
peroxide oxidation of 4-chlorophenol, Reac. Kinet. Mech. Cat. 111 (2014)
91–604.
19] J.A. Rengifo-Herrera, R.A. Frenzel, M.N. Blanco, L.R. Pizzio, Visible-light-
absorbing mesoporous TiO modified with tungstosilicic acid as photocatalyst
in the photodegradation of 4-chlorophenol, J. Photochem. Photobiol. A Chem.
89 (2014) 22–30.
20] J. Rashid, M.A. Barakat, R.M. Mohamed, I.A. Ibrahim, Enhancement of
photocatalytic activity of zinc/cobalt spinel oxides by doping with ZrO for
2 2
O occurs by a radical mechanism, J. Phys. Chem. B 116
(
[38] M. Yamaji, C. Paris, M.T. Miranda, Steady-state and laser flash photolysis
[
studies on photochemical formation of 4-tert-butyl-4'-
2
2
3
N
4
methoxydibenzoylmethane from its derivative via the Norrish Type II reaction
in solution, J. Photochem. Photobiol. A: Chem. 209 (2010) 153–157.
[39] T. Asaka, N. Akai, A. Kawai, K. Shibuya, Photochromism of 3-butyl-1-methyl-2-
phenylazoimidazolium in room temperature ionic liquids, J. Photochem.
Photobiol. A: Chem. 209 (2010) 12–18.
[40] S. Rantamäki, E. Tyystjärvi, Analysis of S 2QA ꢂ charge recombination with the
arrhenius, eyring and marcus theories, J. Photochem. Photobiol. B: Biol. 104
(2011) 292–300.
[
[
2
O
4
2 4
O
5
2
[41] A. Francois-Heude, E. Richaud, E. Desnoux, X. Colin, A general kinetic model for
the photothermal oxidation of polypropylene, J. Photochem. Photobiol. A:
Chem. 296 (2015) 48–65.
2
[
[
2
[42] F. Bonnichon, C. Richard, G. Grabner, Formation of an a-ketocarbene by
visible light photocatalytic degradation of 2-chlorophenol in wastewater, J.
Photochem. Photobiol. A Chem. 284 (2014) 1–7.
photolysis of aqueous 2-bromophenol, Chem. Commun. 7 (2001) 3–7 (4).
[43] J.P. Da Silva, S. Jockusch, N.J. Turro, Probing the photoreactivity of aryl chlorides
with oxygen, Photochem. Photobiol. Sci. 8 (2009) 210–216.
[44] G. Grabner, C. Richard, The Handbook of Environmental Chemistry, in: P. Boule,
D.W. Bahnemann, P.K.J. Robertson (Eds.), Environmental Photochemistry, 2,
Springer-Verlag Heidelberg, Berlin, 2005, pp. 161–192 (Part M).
21] S. Biniak, M. Pakula, A. Swiatkowski, K. Kusmierek, G. Trykowski, Electro-
oxidation of chlorophenols on powdered carbon electrodes of different
porosity, React. Kinet. Catal. Lett. 114 (2015) 369–383.
[
22] K. Kusmierek, A. Swiatkowski, The influence of different agitation techniques
on the adsorption kinetics of 4-chlorophenol on granular activated carbon,
React. Kinet. Mech. Catal. 116 (2015) 261–271.