Kinetics of photodegradation and ozonation of pentachlorophenol

Article abstract of DOI:10.1016/S0045-6535(03)00153-X

The oxidation of 2,3,4,5,6-pentachlorophenol (PCP) has been carried out by a photodecomposition process using a polychromatic UV irradiation, and by an ozonation process. In the photodegradation process, the pH accelerated the decomposition rate and the approximate first-order rate constants were evaluated, with values between 0.16 ± 0.005 min^-1 at pH = 3 and 0.26 ± 0.007 min^-1 at pH = 9. A more rigorous kinetic study led to the determination of the quantum yields of the reaction, with values of 200 ± 7 x 10^-3 mol/Eins for pH = 3 and 22 ± 1.1 x 10-3 mol/Eins for pH = 9. In the ozonation process, the rate constants for the reaction between ozone and PCP were determined by means of a competition kinetics, with values in the range from 0.67 x 10⁵ to 314 x 10⁵ l/mol s. The specific rate constants for the un-dissociated and dissociated forms of PCP were also calculated. Finally, in both processes, the intermediate reaction products were identified, the most important being tetrachlorocatechol, tetrachlorohydroquinone and tetra-p-chlorobenzoquinone. Free chloride ion released, which was favored at high pHs, was also followed in both processes.

Full text of DOI:10.1016/S0045-6535(03)00153-X

Chemosphere 51 (2003) 651–662
www.elsevier.com/locate/chemosphere
Kinetics of photodegradation and ozonation of
pentachlorophenol
*
ꢀ
F. Javier Benitez , Juan L. Acero, Francisco J. Real, Juan Garcıa
Departamento de Ingenierꢀıa Quꢀımica y Energeꢀtica, Universidad de Extremadura, Avda. de Elvas, s/n 06071Badajoz, Spain
Received 5 August 2002; received in revised form 20 January 2003; accepted 28 January 2003
Abstract
The oxidation of 2,3,4,5,6-pentachlorophenol (PCP) has been carried out by a photodecomposition process using a
polychromatic UV irradiation, and by an ozonation process. In the photodegradation process, the pH accelerated the
decomposition rate and the approximate first-order rate constants were evaluated, with values between 0.16 ꢀ 0.005
minꢁ1 at pH ¼ 3 and 0.26 ꢀ 0.007 minꢁ1 at pH ¼ 9. A more rigorous kinetic study led to the determination of the
quantum yields of the reaction, with values of 200 ꢀ 7 Â 10ꢁ3 mol/Eins for pH ¼ 3 and 22 ꢀ 1.1 Â 10ꢁ3 mol/Eins for
pH ¼ 9. In the ozonation process, the rate constants for the reaction between ozone and PCP were determined by means
of a competition kinetics, with values in the range from 0.67 Â 105 to 314 Â 105 l/mol s. The specific rate constants for
the un-dissociated and dissociated forms of PCP were also calculated. Finally, in both processes, the intermediate
reaction products were identified, the most important being tetrachlorocatechol, tetrachlorohydroquinone and tetra-
p-chlorobenzoquinone. Free chloride ion released, which was favored at high pHs, was also followed in both processes.
ꢀ 2003 Elsevier Science Ltd. All rights reserved.
Keywords: Pentachlorophenol; UV irradiation; Quantum yield; Ozone; Rate constants; Dissociated and un-dissociated species
1. Introduction
methods for their removal from waters by degrading
them, either to less harmful intermediates or to complete
mineralization.
The presence of halogenated phenolic compounds
like chlorophenols is known to cause severe pollution
problems in aquatic environments. These chlorophenols
are used widely in agriculture and related industries as
an ingredient in pesticides, insecticides, preharvest her-
bicides, etc., as well as intermediates of dyes, and other
industrial chemicals. However, they are highly toxic:
thus, several of them have been listed among the 65
priority pollutants by the US EPA (Keith and Telliard,
1979) because they are refractory and hard to remove
by conventional biological treatment processes. There-
fore, it is highly recommended to conduct very effective
Among the degradation methods to be followed,
some chemical oxidation processes have been proposed
as effective technologies for the removal of such as toxic
and hazardous pollutants. Thus, although at low rates,
UV irradiation decomposes organic molecules by bond
cleavage, and therefore it has shown significant in-
dustrial applications in the treatment of wastewaters
containing harmful pollutants like organo-chlorinated
compounds in general, and chlorophenols in particular
(Legrini et al., 1993; Yue, 1993). In addition, several
studies on the photochemical oxidation of chlorophe-
nols can be found in the literature (Benoit-Guyod et al.,
1994; Shen et al., 1995; Hugul et al., 2000).
* Corresponding author. Tel.: +34-924-289384; fax: +34-924-
271304.
E-mail address: javben@unex.es (F.J. Benitez).
Similarly, ozone has been found to be an effective
oxidizing agent for the removal of many organic pollu-
tants from water (Masten and Davies, 1994). It is often
0045-6535/03/$ - see front matter ꢀ 2003 Elsevier Science Ltd. All rights reserved.
doi:10.1016/S0045-6535(03)00153-X

652
F.J. Benitez et al. / Chemosphere 51(2003) 651–662
preferred over chlorination because it has a higher
oxidation potential than chlorine and also because
chlorination has been linked to the generation of triha-
lomethanes such as chloroform, methylene chloride
and carbon tetrachloride. Studies on the ozonation of
chlorophenols have also been reported by several
authors (Kuo and Huang, 1995; Trapido et al., 1997; Ku
and Wang, 2002).
rachlorophenol from Riedel-de-Haen, all of them with a
purity of 99% and were used without further purifica-
tion. Finally, 2,4,6-trichlorophenol (TCP), used in the
ozonation experiments simultaneously with PCP, was
also purchased from Sigma. All the solutions were pre-
pared using ultrapure water obtained from a Mill-Q
system (Millipore Co.).
Specifically 2,3,4,5,6-pentachlorophenol (PCP) has
a primary end-use as a wood preservative, and therefore
it can be found a contamination of PCP in surface and
ground waters and soils in the vicinities of past wood-
treating facilities. It is a needlelike white solid, with a
solubility in water in the range of 10–20 mg/l at room
temperature. Even a low concentration of 0.1 mg/l is
toxic to plants, animal and the human body, because it
can easily be absorbed thorough the skin. So, PCP has
been included in the list of priority pollutants published
by the European Union in the last directive about water
policy (Directive, 2000/60/EC). As the degradation rates
by some bacteria are very low and natural environ-
mental conditions are usually unfavorable for the bio-
logical process to be activated, PCP contamination often
requires chemical or physical methods for its removal or
degradation. However, kinetic studies of PCP photode-
gradation and ozonation have rarely been reported in
the literature.
2.2. Equipment
The photochemical experiments were carried out in a
500 cm3 cylindrical glass reactor which contained a ra-
diation lamp located in axial position and a quartz
sleeve which houses the lamp. This radiation source was
a Hanau TQ150 high pressure mercury vapor lamp
which emitted a polychromatic radiation in the range
from 185 to 436 nm. Fig. 1 shows the fraction emitted by
the lamp for each wavelength (Pi) according to the data
supplied by the manufacturer. As can be seen, within
this wavelength range, the strongest emission fractions
are in the wavelengths of 254 and 366 nm. The two main
wavelengths within this range at which PCP absorbs are
254 and 313 nm, while at 366 nm its absorbance is al-
most negligible as it is observed in Fig. 1, which also
shows the experimentally determined molar absorption
coefficients (e) of PCP solutions at different pHs. It can
be seen an increase in the absorbance of PCP when the
pH is increased, with great e values at pH ¼ 9.
Due to these considerations, the photodecomposition
by UV irradiation and the oxidation by ozone of PCP
was proposed to be investigated in the present research,
as a continuation of preceding studies were these de-
gradation processes where applied to other CPs like
4-CP, 2,4-DCP, 2,4,6-TCP and 2,3,4,6-TeCP (Benitez
et al., 2000; Benitez et al., 2001). Thus, the first objective
of this research was to report values of the degradation
levels obtained in each process. Secondly, to investigate
the kinetics of PCP elimination in both processes, with
the determination of the quantum yield in the photo-
degradation reaction and the rate constant of the re-
action between ozone and PCP. Finally, the specific
ozonation rate constants for the deprotonated and
protonated forms of PCP were also evaluated, and some
intermediates and final products were identified and
quantified in both processes.
On the other hand, the ozonation experiments were
performed in the same 500 cm3 cylindrical glass reactor.
In this case, the ozone was produced in an ozone gen-
erator (Yemar, mod. HPA) from commercial oxygen,
and the final mixture ozone-oxygen was introduced into
the reactor through a porous gas sparger located at
the bottom of the reactor. The reactor operated in
semi-batch mode, with continuous feeding of ozone. The
system had an inlet for the measurement of the tem-
perature and outlets for taking samples and venting the
effluent gas.
2.3. Procedures and analysis
For every experiment conducted in the photochemi-
cal degradation experiments, the reactor was filled with
350 cm3 of an aqueous solution of PCP with the desired
initial concentration. An external jacket surrounded the
reactor, and a water stream was pumped from a ther-
mostatic bath in order to maintain the temperature at
the selected value of 20 ꢁC within ꢀ0.5 ꢁC. The desired
pH for every experiment (3, 5, 7 or 9) was adjusted by
means of a phosphoric acid/NaOH buffer so that the
total ionic strength in the final solution was always 0.05
M. During each experiment, samples were periodically
withdrawn from the reactor for analysis of PCP and the
intermediate degradation products. In the later photo-
2. Experimental
2.1. Chemicals
99% PCP was obtained from Fluka Chemical
Co. Sodium hydroxide and phosphoric acid of ana-
lytical grade were obtained from Panreac. The in-
termediate products, tetrachlorohydroquinone and
tetrachloro-p-benzoquinone were from Sigma Chemical
Co., tetrachlorocatechol from Aldrich, and 2,3,5,6-tet-

F.J. Benitez et al. / Chemosphere 51(2003) 651–662
653
Fig. 1. Fraction of the total radiation emitted by the lamp at each wavelength (Pi) and molar absorption coefficients (e) at different
pHs.
chemical kinetic study, it will be necessary to know the
radiation emitted by the lamp into the reactor. For this
purpose, previous experiments based on the uranyl sul-
fate photodecomposition of oxalic acid (Alfano et al.,
1986) were conducted, and the value obtained was
1.76 Â 10ꢁ5 Eins/s. This procedure was described in de-
tail in a former publication (Benitez et al., 1995).
In the ozonation process, simultaneous ozonation
reactions of PCP and TCP were conducted in heter-
ogeneous conditions: in every experiment, a solution
containing a mixture with the desired initial concentra-
tions of both organics was previously introduced into
the reactor and the experiment started when the ozone-
oxygen mixture was fed. The pH for each experiment
and the temperature (always maintained constant at 20
ꢁC) were adjusted to the desired value in the same way as
in the photochemical process. Similarly, samples were
retired at regular intervals in order to analyse both CPs
concentrations. Further ozonation experiments of PCP
alone were carried out to determine its intermediate
degradation products from ozone attack.
with a 996 Photodiode array detector and a Nova-Pak
C18 column. The detection was performed at 300 nm,
and the mobile phase was a mixture 75:25 methanol and
1% aqueous acetic acid with a 1 ml/min flow. A different
mobile phase was used for the reaction samples where
the intermediate degradation products were analysed. In
this case, the initial conditions were 35% of methanol
and 65% of aqueous acetic acid (1%) with a linear gra-
dient to 77% of methanol within 21 min, and then 5 min
isocratic with 77% of methanol. After that, the initial
conditions were reached within 2min, and the system
was equilibrated for 6 min. The injection volume was
100 ll in all samples. Chloride ions were measured
by potentiometry using a chloride selective electrode
(Metler Toledo).
3. Results and discussion
3.1. Photodegradation of 2,3,4,5,6-pentachlorophenol
Specifically, the stoichiometric ratio of the reaction
between ozone and PCP was evaluated by conducting
several experiments in homogeneous conditions: aque-
ous solutions of ozone and PCP of known initial
concentrations were mixed in the reactor. The initial
concentration of PCP was around 10 times higher than
that of ozone, in order to assure immediately the total
consumption of ozone by PCP. At the end of the ex-
periment, the remaining PCP concentration was mea-
sured and the stoichiometric ratio was evaluated as is
described in the Section 3.2.1.
Experiments on the photodecomposition of PCP by
UV irradiation were carried out at 20 ꢁC and modifying
the pH (3, 5, 7, and 9). The concentration of PCP at pH
7 and 9 was 10 ppm (equivalent to 3.75 Â 10ꢁ5 M).
However, the low solubility of PCP at lower pHs, in-
duced to prepare its solution at pH 3 and 5 with an
initial concentration of 5 ppm (1.875 Â 10ꢁ5 M). This
fact has also been pointed out by Huang et al. (2000),
which reported data on the PCP solubility from pH ¼
1.74 to 13.2, significantly lowers than those for 2,4-DCP,
2,4,6-TCP and 2,3,4,6-TeCP. In the same way, Trapido
et al. (1997) worked with PCP solutions of 4 Â 10ꢁ5 M in
their study on the ozonation of several CPs, when the
The concentrations of PCP (and TCP in the ozona-
tion process) as well as the rest of intermediate products
were analysed by HPLC, in a Waters Chromatograph

654
F.J. Benitez et al. / Chemosphere 51(2003) 651–662
Table 1
Determination of the first-order rate constant and quantum
yield for the photochemical degradation of PCP
pH
kp, minꢁ1
/=V
/ Â 103, mol/E
3
5
7
9
0.16 ꢀ 0.005
0.19 ꢀ 0.004
0.21 ꢀ 0.005
0.26 ꢀ 0.007
0.57 ꢀ000.022
0.32 ꢀ 0.01
0.15 ꢀ 0.005
0.06 ꢀ 0.003
ꢀ 7.0
112 ꢀ 3.5
52.5 ꢀ 1.8
22 ꢀ 1.1
½PCPꢂ
0
ln
¼ kpt
ð2Þ
½PCPꢂ
where kp represents the first-order rate constant.
This simplified model has been used successfully by
several authors (Sundstrom et al., 1989; Shen et al.,
1995). Then, a plot of ln [PCP]0/[PCP] versus time leads
to a straight line whose slope is kp. The values deduced
for the first-order rate constants are depicted in Table 1,
all of them with correlation coefficients greater than
0.99. These kp values confirm the trend previously ob-
served for the PCP concentration curves: that is, the
highest kp corresponds to pH ¼ 9 and the lowest to
pH ¼ 3. These findings for PCP are in agreement with
the results of Shen et al. (1995), who studied the pho-
todegradation of several CPs (2-CP, 2,4-DCP and 2,4,6-
TCP) in the range of pH between 3 and 11. They also
reported increasing values for the first-order rate con-
stant when the pH was increased in every CP studied.
Thus, first-order rate constants between 7.1 Â 10ꢁ3 at
pH ¼ 3 and 101 Â 10ꢁ3 minꢁ1 at pH ¼ 11 for 2-CP;
3.6 Â 10ꢁ3 at pH ¼ 3 and 126 Â 10ꢁ3 minꢁ1 at pH ¼ 11
for 2,4-DCP, and 2.3 Â 10ꢁ3 pH ¼ 3 and 36.3 Â 10ꢁ3
minꢁ1 at pH ¼ 11 for 2,4,6-TCP were reported. Al-
though in the same range, those rate constants are lower
than the values obtained in this research and presented
in Table 1, probably due to the fact that they used a
monochromatic radiation source (at constant wave-
length of 254 nm) while the present study was conducted
with a polychromatic radiation source with higher input
energy.
Fig. 2. Evolution of the normalized concentration of PCP at
different pHs in the photochemical oxidation process.
concentrations for the remaining CPs were 1 Â 10ꢁ4 and
4 Â 10ꢁ4 M.
Fig. 2shows the PCP normalized concentration
curves with reaction time in these experiments: it can be
observed a slightly positive influence of the pH on the
process, with decreasing values of [PCP]/[PCP]0 and
subsequent increasing decomposition rates when the pH
is increased. These results can be explained by taking
into account that, for a given wavelength of the poly-
chromatic irradiation, the absorbance of PCP increases
when the pH also increases. This effect can be seen in
Fig. 1, which also represents the experimentally deter-
mined molar absorption coefficients e versus the wave-
lengths at different pHs. In conclusion, the pH effect on
the light absorbance and on the photodecomposition
rate is found to be quite similar.
3.1.1. Kinetic study
The reaction mechanism for the photodecomposition
of general pollutants in aqueous solutions by UV radi-
ation has been studied in previous works (Ollis et al.,
1991; Legrini et al., 1993). Applying these reaction
mechanisms to the photodegradation of PCP, a first
approach can be made by assuming that the photo-
chemical oxidation:
The first-order kinetics constitutes an approximation
for a more complex reaction mechanism as noted pre-
viously for other organics, like atrazine, chloroethanes
and CPs (Hugul et al., 2000). In this mechanism, the
overall reaction (1), between PCP and the UV radiation,
can be separated into the following partial reactions:
/iliIi
PCP þ hm ! PCPoxid
ð1Þ
•
Activation of a PCP molecule by the absorption of
one photon, leading to an excited molecule:
follows first-order kinetics. In this Eq. (1), /i, li and
Ii represent the quantum yield of the reaction, the ab-
sorbance of the solution and the radiation intensity
respectively, for every individual wavelength of the poly-
chromatic irradiation source used in this research.
In accordance with this model, the kinetic equation
for PCP will be:
/iliIi
PCP þ hm ! PCPꢃ
ð3Þ
•
Deactivation of this excited specie to its basic state:
k1
PCPꢃ ! PCP
ð4Þ

F.J. Benitez et al. / Chemosphere 51(2003) 651–662
655
•
Real photodecomposition of the excited PCP mole-
cule:
k2
PCPꢃ !PCPoxid
ð5Þ
This hypothesized reaction scheme can be analyzed
via steady state approximation, by setting the rate of
accumulation of unstable intermediates to zero. Then,
this rate equation is obtained for the photodecomposi-
tion of PCP
ꢀ
ꢁ
X
k2
k1 þ k2
rPCP
¼
/
liIi
ð6Þ
i
If the overall quantum yield / for all the wavelength
range of the polychromatic radiation is expressed by
ꢀ
ꢁ
X
k2
k1 þ k2
/ ¼
/
ð7Þ
i
Fig. 3. Determination of the quantum yield / in the photo-
chemical oxidation of PCP.
then Eq. (6) is transformed into
X
rPCP ¼ /
ðliIiÞ
ð8Þ
ous research (Benitez et al., 1995). Then, with Wabs
R
evaluated, each term
Wabs dt was calculated for every
If this Eq. (8), which represents the reaction rate in a
specific point inside the reactor, is integrated for the
whole reactor, the following expression is obtained for
the total photodecomposition rate
[PCP] by fitting the experimental data (Wabs; t) to a poly-
nomial expression by least squares regression analy-
sis, and finally, by integrating the resulting function
between the initial time and the selected time of photo-
reaction.
ꢀ
ꢁ
Z
Z
X
d½PCPꢂ
dt
1
/
ꢁ
¼
r
PCP dV ¼
liIi dV
ð9Þ
According to Eq. (12), values of [PCP] are plotted
against the corresponding term
V
V
v
V
R
Wabs dt; and from the
By considering the definition of the global radiation
flow rate absorbed by the solution when irradiated by a
polychromatic radiation
straight line obtained for every experiment, the intercept
[PCP]0 and the slope /=V can be deduced. Fig. 3 shows
this plot for the different PCP photodegradation experi-
ments carried out with the variation of pH. A regression
analysis provides the slope /=V . The overall quantum
yield / of the photoreaction can then be calculated,
being its values reported in Table 1. As it is observed, the
quantum yield decreases when the pH is increased; that
is, an opposite effect to that found for the first-order
rate constant kp, which increased with the increase in the
pH.
Z
X
X
Wabs
¼
Wabs
¼
liIi dV
ð10Þ
i
v
it is finally obtained for the overall photodecomposition
rate
P
d½PCPꢂ
dt
/
ðWabs
Þ
/Wabs
i
ꢁ
¼
¼
ð11Þ
V
V
This apparent contradiction can be explained by
taking into account Eq. (9): as it is seen, the total re-
action rate is dependent on the quantum yield / as well
as on absorbance of the solution li. At the same time,
the absorbance is related with the molar absorption
coefficient, e, by the expression
which can be integrated yielding to the following final
equation:
Z
t
/
½PCPꢂ ¼ ½PCPꢂ ꢁ
Wabs dt
ð12Þ
0
V
0
The integral term in Eq. (12) cannot be solved ana-
lytically, because the flow rate absorbed Wabs varies with
the reaction time, and therefore, it must be solved nu-
merically. For this evaluation, the following procedure
was conducted: firstly the values of Wabs at any reaction
time were determined by means of the Line Source
Spherical Emission Model (Alfano et al., 1986), ac-
cording to the procedure described in detail in a previ-
li ¼ ei½PCPꢂ
ð13Þ
As it has been previously discussed, ei increases with
the pH, and subsequently li also increases; then, / must
decrease for a specific value of the rate––ðd½PCPꢂ=dtÞ.
This fact confirms the results obtained for the overall
quantum yield at different pHs.

656
F.J. Benitez et al. / Chemosphere 51(2003) 651–662
3.1.2. Reaction products
In the photochemical degradation of PCP conducted
in this study, tetrachlorocatechol and tetrachlorohy-
droquinone have been identified, as well as important
amount of another un-identified compound X1. Both
compounds were also observed by Hong et al. (2000) in
the UV-irradiation of PCP, while tetrachlorohydroqui-
none was reported too by Mills and Hoffmann (1993) as
one of the principal intermediates formed in the photo-
catalytic destruction of PCP in the presence of TiO2.
Jardim et al. (1997) in their study on this process, con-
firmed the formation of tetrachlorohydroquinone; in
addition, they also detected tetrachlorobenzoquinone,
although this compound has not been detected in the
present study.
The first step of the PCP photodegradation mecha-
nism must be the PCP reductive dechlorination and the
formation of OH radicals, similarly to what occurs in
the UV-irradiation of phenol and benzoic acid in
aqueous solutions (Mills and Hoffmann, 1993). Then,
the OH radicals formed in the photolysis of PCP can
further attack to new molecules of PCP to give photo-
oxidation products. Thus, tetrachlorocatechol and tet-
rachlorohydroquinone are produced from OH radical
attack to ortho- and para-positions, respectively. In ad-
dition, free chloride ions are generated in these de-
chlorination steps. Besides the degradation products,
polychlorinated dioxins, polychlorinated hydroxy di-
phenylethers and polychlorinated diphenylethers have
been identified by GC-MS as minor degradation prod-
ucts (Hong et al., 2000).
The profiles of the species identified during the PCP
(10 ppm) photodecomposition in the present study at
pH 4, 5 and 7 are shown in Fig. 4. As can be observed,
the formation of tetrachlorocatechol was more rapid
than that of tetrachlorohydroquinone, which indicates
that the addition of OH radicals at ortho-position is
more favorable than at para-position. Also, it is seen
that at pH 7, a very low formation of tetrachlorohy-
droquinone takes place and higher concentrations
of tetrachlorocatechol are detected. Unfortunately, the
compound X1, which could not be identified, was the
main degradation product at pH 4 and 5. This com-
pound has been estimated using the response factor of
tetrachlorocatechol and tetrachlorohydroquinone in the
HPLC chromatogram. The photodegradation of tetra-
chlorocatechol and tetrachlorohydroquinone seems to
be a rather slow process at pH 4. However, at pH 5
and specially at pH 7, these two intermediate are easily
phototransformed according to the bell-shape of their
concentration curves. The mass balance (sum of PCP
and identified intermediate concentration) is poor,
which indicates the important existence of un-identified
primary degradation products such as X1.
Fig. 4. Profile of the identified intermediate products during the
photochemical oxidation of PCP at different pHs. [B] represents
any compound; TeCHqn and TeCCtc stand for tetrachloro-
hydroquinone and tetrachlorocatechol respectively. PCP con-
centration and the mass balance are represented in the left axis,
while the degradation products are in the right axis.
periments. The results obtained in experiments at pH 4
and 7 are shown in Fig. 5. Higher chloride ion concen-
tration is observed at pH 7 which indicates that a de-
chlorination pathway is preferred at higher pH to form
either dechlorinated aromatic compounds or dechlori-
nated aliphatic compounds. The low chloride concen-
tration at pH 4 also indicates that the primary reaction
intermediates are hardly photolyzed at acidic pH. In
order to corroborate these statements, the chloride for-
mation yield, defined as the mole of chloride formed per
The evolution of the chloride ion concentration in the
aqueous phase has also been measured during the ex-

F.J. Benitez et al. / Chemosphere 51(2003) 651–662
657
pH 9. So, the use of tert-butyl alcohol assured that
PCP was only degraded by a direct molecular ozone
attack in alkaline solutions and the radical attack can be
neglected. Some additional experiments of PCP ozona-
tion in absence of tert-butyl alcohol were carried out to
identify the intermediate degradation products. These
experiments were performed at pH 4, 5 and 7 with initial
PCP concentration of 10 ppm.
3.2.1. Kinetic study
The results of Trapido et al. (1997) in the ozonation
of some CPs, as well as our previous results (Benitez
et al., 2000) in the same field, revealed that the degra-
dation rate of the direct reaction between ozone and a
CP increased when the pH increased, and also when the
number of substituent chlorine atoms to the aromatic
ring was also increased. According to this trend, it can
be expected for the ozonation of PCP very high reaction
rates, which could be extremely fast at pH ¼ 9.
Fig. 5. Evolution of chloride ion concentration (continuous
lines) and the chloride formation yield (dashed lines) during the
photochemical oxidation of PCP.
In order to conduct the corresponding kinetic study
for the reaction between ozone and PCP, a competition
kinetics model proposed by Gurol and Nekouinaini
(1984) which is adequate for the evaluation of high rate
constants as in the present study, was applied. This
model, described in detail in a previous research (Benitez
et al., 2000), basically consists on the simultaneous de-
gradation of mixtures constituted by two organic com-
pounds, one of them act as a reference compound which
degradation rate constant at the pH of work is previ-
ously known. In the present study, the reference com-
pound was TCP, which rate constants at different pHs
were evaluated before (Benitez et al., 2000), while the
second organic substance is the target compound, PCP
in this case, which rate constants are unknown. The
competitive method also establishes that the reaction
between ozone and the organics must follow second-
order kinetics, and more specifically, first-order with
respect to each reactant. This condition is fulfilled in the
ozonation of CPs in general, as has been previously re-
ported (Yao and Haag, 1991; Trapido et al., 1997).
According to that model, in this specific case of PCP
ozonation, the disappearance rate equation will be
mole of PCP degraded, has been calculated and included
in Fig. 5. A constant chloride formation yield of 1.7 is
observed at pH 4, which demonstrates that chloride is
released in the first step of the PCP photolysis. However,
at pH 7, the chloride yield increases continuously during
the reaction time to reach a value close to 5, indicating
that most of the chlorine contained in the PCP has been
released as chloride ions.
In conclusion, it can be proposed that in a great ex-
tent, and specially at pH 7, the degradation preferen-
tially occurs with the detachment of the chlorine from
the benzene ring, and then most of the PCP is converted
into low molecular weight acids and chloride, as well as
mineralized into CO2 and water.
3.2. Ozonation of 2,3,4,5,6-pentachlorophenol
The oxidation of PCP by ozone (with a constant
ozone partial pressure in the oxygen–ozone mixture of
90 Pa) was conducted at 20 ꢁC and by varying the pH at
the same values than the photodegradation process (3, 5,
7 and 9) as well as at pH ¼ 2.5. Again, and due to the
low solubility of PCP in water, different initial con-
centrations were used: 2.85 ppm (1.03 Â 10ꢁ5 M) at
pH ¼ 2.5, 5 ppm (1.875 Â 10ꢁ5 M) at pH ¼ 3, and 10
ppm (3.75 Â 10ꢁ5 M) at pH 5, 7 and 9.
d½PCPꢂ
dt
kO3i
zi
ꢁ
¼ ki½O3ꢂ½PCPꢂ ¼
½O3ꢂ½PCPꢂ
ð14Þ
while the rate equation for the disappearance of the
reference compound TCP is:
These experiments were carried out in the presence of
tert-butyl alcohol, a well-known scavenger of free radi-
cals (Staehelin and Hoigne, 1985). In the present case,
these radicals can be generated from the ozone decom-
position, which is accelerated rapidly by the increase in
the pH, and therefore, the reactions promoted by hy-
droxyl radical would present an important contribution
at basic solutions like those of the present research at
d½TCPꢂ
dt
kO3R
zR
ꢁ
¼ kR½O3ꢂ½TCPꢂ ¼
½O3ꢂ½TCPꢂ
ð15Þ
where kO3R and kO3i represent the overall ozone disap-
pearance rate constants for the reference (TCP) and
target (PCP) compounds, respectively; kR and ki repre-
sent the overall CP disappearance rate constants, and zR

658
F.J. Benitez et al. / Chemosphere 51(2003) 651–662
Table 2
Determination of the rate constant for the ozonation of PCP
and zi the stoichiometric ratios for the direct reaction
between ozone and both CPs.
Regarding to the stoichiometric ratios, the value for
zR was determined in a previous research (Benitez et al.,
2000), and it resulted 2 moles of ozone consumed per
mol of TCP reacted. For the evaluation of zi, previous
experiments in homogeneous conditions were performed
as described in Section 2. The concentration of ozone
and PCP in the initial aqueous solutions and the re-
maining PCP concentrations in the final solutions were
measured. By means of the following equation:
pH
kO3R Â 10ꢁ5
,
M
kO3i  10ꢁ5
,
l/mol s
l/mol s
2.5
3
0.20
0.50
41.3
497
4.48 ꢀ 0.21
3.80 ꢀ 0.121.44
1.03 ꢀ 0.01
0.74 ꢀ780.022
0.75 ꢀ 0.01
0.67
32.0
314
5
7
9
559
in a similar way as it has been reported for the ozonation
of other CPs (Trapido et al., 1997; Benitez et al., 2000).
Unfortunately, no data has been found in the literature
for this rate constant for the PCP ozonation, and
therefore, no comparison can be made. Only Hoigne and
Bader (1983) have proposed a value bigger than 3 Â 105
l/mol s for this rate constant at pH 2.
½O3ꢂ
0
zi ¼
ð16Þ
½PCPꢂ ꢁ ½PCPꢂ
0
the experimental stoichiometric ratio zi can be calcu-
lated. Following this procedure, an experimental zi of
1.5 moles of ozone consumed per mol of PCP reacted
was obtained.
3.2.2. Rate constants for the un-dissociated and dissoci-
ated species
From Eqs. (14) and (15), and integrating between
t ¼ 0 and t ¼ t, it is obtained:
The observed increase in the rate constants, and
subsequently in the reaction rate of ozone and PCP, with
the increase of the pH can be explained by the acidic
nature of CPs in general and PCP in particular, which
pKa is 4.7. According to this acidic nature, the pH of the
solution leads to different concentration of both species,
un-dissociated and dissociated. The reactivity of both
species towards ozone is quite different, being the dis-
sociated forms generally more reactive towards oxidants
than the un-dissociated forms (Hoigne and Bader, 1983).
Therefore, at alkaline pHs, when the dissociated species
are predominant, it can be expected a higher reaction
rate; on the contrary, at acidic pHs, when the un-dis-
sociated species predominate, it can be expected lower
reaction rates. These considerations are confirmed by
the results obtained previously for the ozonation of
other CPs (Benitez et al., 2000), and also in this research
by the rate constants obtained for the reaction between
ozone and PCP.
½PCPꢂ
zRkO3i
zikO3R
½TCPꢂ
½TCPꢂ
0
0
0
ln
¼
ln
¼ M ln
ð17Þ
½PCPꢂ
½TCPꢂ
½TCPꢂ
According to Eq. (17), Fig. 6 shows the plot of ln
([PCP]0/[PCP]) against ln ([TCP]0/[TCP]) corresponding
to the experiments conducted at different pHs. After
regression analysis, the slopes M provide the ratio be-
tween the rate constants. With the values of zR and zi as
well as the values of kO3R previously evaluated (Benitez
et al., 2000) and depicted in Table 2, the rate constant
kO3i were deduced in every experiment conducted, being
its values also reported in Table 2.
As it is seen, the obtained rate constants kO3i for the
reaction between ozone and PCP increase with the pH,
Therefore, the overall ozonation reaction rate of
acidic organic substances, like CPs, can be expressed as
a function of the specific ozonation rate constant of the
un-dissociated and dissociated forms of the studied or-
ganic substance (PCP in this case), as follows:
rO3 ¼ kO3i½O3ꢂ½PCPꢂ
tot
ꢁ
ꢁ
¼ kCP ½O3ꢂ½PCP ꢂ þ kCP½O3ꢂ½PCPꢂ
ð18Þ
where kO3i is the rate constant for the mentioned overall
ozonation reaction based on the total concentration of
PCP present.
For using Eq. (18), it is valuable to determine kCP
ꢁ
,
the specific ozonation rate constant for the ionic form
PCPꢁ, as well as kCP, the specific ozonation rate constant
for the un-dissociated form PCP. Once these specific rate
Fig. 6. Determination of the rate constant for the reaction
between ozone and PCP in the ozonation process, by compe-
tition kinetics at several pHs.

F.J. Benitez et al. / Chemosphere 51(2003) 651–662
constants for the ozonation of both species are known,
659
the application of Eq. (18) provides the global reaction
rate rO3 of the PCP ozonation reaction and the overall
rate constants kO3i during the treatment of natural
waters and wastewaters at any pH.
For this purpose, it must be taken into account that
the ratio of the dissociated form to the un-dissociated
form of an acidic substance is given by the degree of
dissociation (a), which is defined by the following ex-
pression:
½PCPꢁꢂ
½PCPꢁꢂ
tot
a ¼
¼
ð19Þ
½PCPꢁꢂ þ ½PCPꢂ ½PCPꢂ
At the same time, the relation between the degree of
dissociation and the pH is
1
a ¼
ð20Þ
½Hþꢂ
1 þ
Ka
Fig. 7. Validation of the PCP specific rate constants (un-dis-
sociated and dissociated forms) in the ozonation process, ac-
cording to Eq. (21).
From Eqs. (18) and (19) it is obtained
ꢁ
kO3i ¼ akCP þ ð1 ꢁ aÞkCP
ð21Þ
which allows to predict the overall rate constants kO3i for
the ozonation of PCP at whichever pH of the water,
was clearly identified. In addition, the un-identified
compound X1 appeared as well.
ꢁ
once kCP and kCP are known.
For that determination, the procedure used was as
follows: with the pKa ¼ 4:7 of PCP, and the pH values in
every experiment, the values of a were calculated from
Eq. (20). Then, by means of Eq. (21) applied to a couple
of the previously deduced overall rate constants kO3i (see
The formation of this kind of compounds was also
reported previously by several authors. Thus, Trapido
et al. (1997) identified quinones in general and chloride
ions during the ozonation of several CPs, and Kuo and
Huang (1995) detected chlorocatechol in the ozonation
of p-chlorophenol. More specifically, the ozonation
of PCP conducted by Hong and Zeng (2002) resulted in
the generation of tetrachloro-p-benzoquinone and tet-
rachlorohydroquinone, that were further degraded by
ozone into other open-ring products like ketones and
acids, and finally, into oxalic acid with quantitative re-
lease of chloride ions.
ꢁ
Table 2), the unknown rate constants kCP and kCP are
determined for PCP, the average values obtained being
9.1 Â 106 and 1.0 Â 104 l/mol s, respectively. These values
are in the same range than those obtained in our pre-
vious research (Benitez et al., 2000) for the less substi-
tuted aromatic rings.
ꢁ
Finally, the determined specific rate constants kCP
and kCP for PCP must be validated. For this purpose, the
overall rate constants kO3i in the range of pH from 2.5 to
10 were theoretically calculated by means of Eq. (21),
with the values of a previously obtained from Eq. (20) at
each pH. The deduced kO3i values are plotted in Fig. 7
(lines) together with the experimental values (symbols)
which are depicted in Table 2. It can be seen an excellent
agreement between the theoretical and experimental
values, which confirm the goodness of the procedure
followed.
Fig. 8 presents the evolution of the reaction products
identified in this work at pH 4, 5 and 7. At pH 4 and 5,
great amount of tetrachloro-p-benzoquinone was found,
with a maximum value around 5 ppm at pH 5 when the
initial concentration of PCP was 10 ppm. On the other
hand, tetrachlorocatechol also reached a maximum
value after 2–3 min of reaction, to decay later, while
tetrachlorohydroquinone was still increasing after 6 min
of reaction. However, at pH 7 a different behavior was
observed, with higher concentrations of tetrachloro-
hydroquinone, and lower of tetrachlorocatechol and tet-
rachloro-p-benzoquinone. Therefore, the first degradation
products from the ozone attack are tetrachloro-
p-benzoquinone, tetrachlorocatechol and tetrachloro-
hydroquinone, having the un-identified compound X1 a
special importance only at pH 7. Thus, the mass balance
at acid pH is higher than 80% during most of the reac-
tion time.
3.2.3. Reaction products
In a similar way as in the photochemical process, the
main reaction products formed during the ozonation of
PCP were also identified, and their evolution during the
reaction at several pHs was followed. These products
were: tetrachlorocatechol and tetrachlorohydroquinone
again, as well as tetrachloro-p-benzoquinone that now

660
F.J. Benitez et al. / Chemosphere 51(2003) 651–662
of the degradation intermediates is produced, probably
due to the additional presence of OH radicals at neutral
and basic pHs. A second reason for the high reactivity of
intermediates at neutral or basic pH is their acid–base
equilibria, since the ozone attack to the dissociated spe-
cies is favored as commented before for PCP.
The most plausible reason for the formation of these
products is a mechanism that considers an electrophilic
ozone attack to the phenolic compound. Thus, in the first
step the ozone attack tends to occur at the ortho- and
para-positions, which are of higher electron density and
more active compared with other positions because of the
resonance donation of oxygen electrons. Consequently,
the reaction of ozone attack at these positions becomes
more favorable (Bailey, 1958). From the results shown in
Fig. 8, ozone attack to the para-position is predominant
in the present case to the ortho-position, specially at acid
pHs. In these initial reactions chloride ions are released.
The intermediates later would tend to decompose
upon continued ozonation. The subsequent decomposi-
tion could result in the release of chloride ion from the
chlorinated aliphatic compounds, and production of
oxalic acid as final product. Kim and Moon (2000)
confirmed this formation of oxalic acid, and they sug-
gested that it constitutes the final product of the PCP
ozonation because it is not oxidized by ozone possibly
due to the fact that it does not have any carbon–carbon
double bond.
x1
x1
In order to check the mentioned generation of free
chloride ion, its concentration was also measured in this
research, being shown the results obtained in Fig. 9. As
it is seen, the Clꢁ concentration increased even after the
PCP degradation was almost completed at pH 7 (more
than 3 min of reaction), which indicates that interme-
diates containing chloride were continuously degraded
x1
Fig. 8. Profile of the identified intermediate products during
the ozonation of PCP at different pHs. [B] represents any
compound; TeCHqn, TeCCtc and TeCpBqn stand for tetra-
chlorohydroquinone, tetrachlorocatechol and tetrachloro-p-
benzoquinone respectively. PCP concentration and the mass
balance are represented in the left axis, while the degradation
products are in the right axis.
It can be also observed in Fig. 8 that at acid pH 4,
tetrachloro-p-benzoquinone is a relatively stable com-
pound, since its concentration does not decrease even
after complete PCP elimination. However, a conversion
of tetrachloro-p-benzoquinone into tetrachlorohydro-
quinone is deduced at pH 5 after complete PCP de-
gradation, due to redox reactions between these two
compounds and oxygen radicals such as superoxide
radicals. When the pH is increased to 7, a fast elimination
Fig. 9. Evolution of chloride ion concentration (continuous
lines) and the chloride formation yield (dashed lines) during the
ozonation of PCP.

F.J. Benitez et al. / Chemosphere 51(2003) 651–662
661
in the presence of ozone and probably OH radicals,
and thus a chloride formation yield of 4 was reached.
However, the chloride concentration and the chloride
formation yield were almost constant after complete
PCP elimination at pH 4, corroborating the stability of
the intermediates at this pH, as was commented above.
The dechlorination of PCP took place at the same time
as ring cleavage, which led to the formation of the
mentioned oxalic acid. In conclusion, the ozonation of
PCP yields intermediates which are probably more
biodegradable prior to complete mineralization.
Tecnologia of Spain’’ (CICYT), under Project PPQ2001-
0744, and the Junta de Extremadura through Project
2PR01A004.
References
Alfano, O.M., Romero, R.L., Cassano, A.E., 1986. Radiation
field modelling in photoreactors. 1. Homogeneous media.
Chem. Eng. Sci. 41, 421–444.
Bailey, P.S., 1958. The reactions of ozone with organic
compounds. Chem. Rec. 58, 925–1010.
Benitez, F.J., Beltran-Heredia, J., Gonzalez, T., Real, F.J.,
1995. Photooxidation of carbofuran by a polychromatic UV
irradiation without and with hydrogen peroxide. Ind. Eng.
Chem. Res. 34, 4099–4105.
4. Conclusions
Benitez, F.J., Beltran-Heredia, J., Acero, J.L., Rubio, F.J.,
2000. Rate constants for the reactions of ozone with
chlorophenols in aqueous solutions. J. Hazard. Mater. B
79, 271–285.
From the results obtained in the present study it can
be concluded that in the photodegradation of PCP, in-
creasing decomposition rates are reached when the pH is
increased. In a first approximation, this process can be
considered as a first-order reaction, and the values de-
duced for the first-order rate constants are in the range
between 0.16 ꢀ 0.005 minꢁ1 (at pH 3) and 0.26 ꢀ 0.007
minꢁ1 (at pH 9). The more rigorous kinetic study leads to
the determination of the overall quantum yield / of the
photoreaction, which values vary from 200 ꢀ 7 Â 10ꢁ3
mol/Eins (at pH 3) to 22 ꢀ 1.1 Â 10ꢁ3 mol/Eins (at pH 9).
In this photochemical degradation of PCP, tetrachloro-
catechol and tetrachlorohydroquinone have been iden-
tified as major intermediate products.
Benitez, F.J., Beltran-Heredia, J., Acero, J.L., Rubio, F.J.,
2001. Oxidation of several chlorophenolic derivatives by UV
irradiation and hydroxyl radicals. J. Chem. Technol. Bio-
technol. 76, 312–320.
Benoit-Guyod, J.L., Bruckner, C., Benoit-Guyod, M., 1994.
Degradation of chlorophenols by ozone and light. Fresen.
Environ. Bull. 3, 331–338.
Directive 2000/60/EC of the European parliament and on the
Council of 23 October 2000 establishing a framework for
Community action in the field of water policy.
Gurol, M., Nekouinaini, S., 1984. Kinetic behavior of ozone in
aqueous solutions of substituted phenols. Ind. Eng. Chem.
Fundam. 23, 54–60.
In the oxidation of PCP by ozone, the stoichiometric
ratio was first determined, its value being 1.5 moles of
ozone consumed per mol of PCP reacted. As it can be
expected very high reaction rates for this process, the
kinetic study for the reaction between ozone and PCP is
performed by using a competition kinetics model, which
is adequate for the evaluation of high rate constants.
Following this model, the rate constant kO3i is deduced
and it increased with the pH. Values between 0.67 Â 105
l/mol s (at pH 2.5) and 314 Â 105 l/mol s (at pH 9) are
found. In this kinetic study, the specific ozonation rate
Hoigne, J., Bader, H., 1983. Rate constants of reactions of
ozone with organic and inorganic compounds in water. II.
Dissociating organic compounds. Water Res. 17, 185–194.
Hong, J., Kim, D., Cheong, C., Jung, S.-Y., Yoo, M.-R., Kim,
K.-J., Kim, T.-K., Park, Y.-C., 2000. Identification of
Photolytical Transformation products of Pentachlorophe-
nol in Water Analytical Sciences 16, 621–626.
Hong, P.K.A., Zeng, Y., 2002. Degradation of pentachloro-
phenol by ozonation and biodegradability of intermediates.
Water Res. 36, 4243–4254.
Huang, G.L., Xiao, H., Chi, J., Shiu, W.Y., Mackay, D., 2000.
Effects of pH on the aqueous solubility of selected chlori-
nated phenols. J. Chem. Eng. Data 45, 411–414.
Hugul, M., Apak, R., Demirci, S., 2000. Modeling the kinetics
of UV/hydrogen peroxide oxidation of some mono-, di- and
trichlorophenols. J. Hazard. Mater. B 77, 193–208.
Jardim, W.F., Moraes, S.G., Takiyama, M.M.K., 1997. Phot-
ocatalytic degradation of aromatic chlorinated compounds
using TiO2: toxicity of intermediates. Water Res. 31, 1728–
1732.
constant kCP for the ionic form PCPꢁ, as well as the
ꢁ
specific ozonation rate constant kCP for the un-dissoci-
ated form PCP are also determined, being their values
9.1 Â 106 l/mol s and 1.0 Â 104 l/mol s respectively. In the
ozonation process, three main intermediate products
are clearly identified: tetrachlorocatechol, tetrachloro-
hydroquinone and tetrachloro-p-benzoquinone, which
subsequent decompositions by ozone lead to the release
of chloride ion, specially at pH 7.
Keith, L.H., Telliard, W.A., 1979. Priority pollutants:
prospective view. Environ. Sci. Technol. 13, 416–424.
a
Kim, J.Y., Moon, S.H., 2000. A kinetic study on oxidation of
pentachlorophenol by ozone. J. Air Waste Manage. Assoc.
50, 555–562.
Acknowledgements
Ku, Y., Wang, L.-K., 2002. Decomposition of 2-chlorophenol
in aqueous solutions by ozone/UV processes in the presence
of t-butanol. Ozone Sci. Eng. 24, 133–144.
The author wish to gratefully acknowledge financial
support from the ‘‘Comision Interministerial de Ciencia y

662
F.J. Benitez et al. / Chemosphere 51(2003) 651–662
Kuo, C.H., Huang, C.H., 1995. Aqueous phase ozonation of
chlorophenols. J. Hazard. Mater. 41, 31–45.
Staehelin, J., Hoigne, J., 1985. Decomposition of ozone in water
in the presence of organic solutes acting as promoters and
inhibitors of radical chain reactions. Environ. Sci. Technol.
19, 1206–1213.
Legrini, O., Oliveros, E., Braun, A.M., 1993. Photochemical
processes for water treatment. Chem. Rev. 93, 671–698.
Masten, S.J., Davies, S.H., 1994. The use of ozonation to
degrade organic contaminants in wastewaters. Environ. Sci.
Technol. 28, 180–185.
Sundstrom, D.W., Weir, B.A., Klei, H.E., 1989. Destruction of
aromatic pollutants by UV light catalyzed oxidation with
hydrogen peroxide. Environ. Progress 8, 6–11.
Trapido, M., Hirvonen, A., Veressinina, Y., Hentunen, J.,
Munter, R., 1997. Ozonation, ozone/UV and UV/H2O2
degradation of chlorophenols. Ozone Sci. Eng. 18, 75–
96.
Mills, G., Hoffmann, R., 1993. Photocatalytic degradation of
pentachlorophenol on TiO2 particles: identification of
intermediates and mechanism of reactions. Environ. Sci.
Technol. 27, 1681–1689.
Ollis, D.F., Pelizzetti, E., Serpone, N., 1991. Photocatalyzed
destruction of water contaminants. Environ. Sci. Technol.
25, 1523–1529.
Yao, D.C.C., Haag, W.R., 1991. Rate constants for direct
reactions of ozone with several drinking water contami-
nants. Water Res. 25, 761–773.
Shen, Y.S., Ku, Y., Lee, K.C., 1995. The effect of light
absorbance on the decomposition of chlorophenols by UV
radiation and UV/H2O2 processes. Water Res. 29, 907–914.
Yue, P.L., 1993. Modelling of kinetics and reactors for water
purification by photo-oxidation. Chem. Eng. Sci. 48, 1–
11.