TiO2-mediated photomineralization of 2-chlorobiphenyl: The role of O2

Article abstract of DOI:10.1016/S0043-1354(00)00009-9

Photocatalytic mineralization of 2-chlorobiphenyl (2-CB) in aqueous TiO₂ suspensions was conducted under four different oxygen partial pressures (Po₂) in closed reactors. The observed O₂ adsorption equilibrium constant in TiO₂ aqueous suspension was 0.88 (kPa)^-1 based on the Langmuir-Hinshelwood rate function. Apart from the conventional electron-scavenging function, the dissolved oxygen was critical for the degradation of the hydroxyl intermediates, i.e., 2-chlorobiphenyl-ols and biphenyl-2-ol, which are potentially toxic. Degradation of 2-CB in the irradiated H₂O₂ aqueous solutions under two different Po₂ was also studied for comparison. For UV/H₂O₂, Po₂ does not affect the removal of 2-CB, while reducing the Po₂ from 21 to 2 kPa resulted in 35% less CO₂ formation after 5 h irradiation. The similar O₂-dependent destruction of the hydroxyl intermediates was also observed in the UV/H₂O₂ system. We propose that molecular O₂ acts as a key reactant following the attack of a second ·OH radical during the ring opening in the degradation of the hydroxyl byproducts. Copyright (C) 2000 Elsevier Science Ltd.

Full text of DOI:10.1016/S0043-1354(00)00009-9

Wat. Res. Vol. 34, No. 10, pp. 2791±2797, 2000
7 2000 Elsevier Science Ltd. All rights reserved
Printed in Great Britain
PII: S0043-1354(00)00009-9
0043-1354/00/$ - see front matter
www.elsevier.com/locate/watres
TiO2-MEDIATED PHOTOMINERALIZATION OF 2-
CHLOROBIPHENYL: THE ROLE OF O2
YONGBING WANG1M and CHIA-SWEE HONG1, 2
*
1Department of Environmental Health and Toxicology, School of Public Health, State University of
New York at Albany, Albany, NY 12201-0509, USA and 2Wadsworth Center, New York State
Department of Health, Albany, NY 12201-0509, USA
(First received 1 March 1999; accepted in revised form 1 October 1999)
AbstractÐPhotocatalytic mineralization of 2-chlorobiphenyl (2-CB) in aqueous TiO2 suspensions was
conducted under four di€erent oxygen partial pressures (Po2) in closed reactors. The observed O2
1
adsorption equilibrium constant in TiO2 aqueous suspension was 0.88 (kPa) based on the Langmuir±
Hinshelwood rate function. Apart from the conventional electron-scavenging function, the dissolved
oxygen was critical for the degradation of the hydroxyl intermediates, i.e., 2-chlorobiphenyl-ols and
biphenyl-2-ol, which are potentially toxic. Degradation of 2-CB in the irradiated H2O2 aqueous
solutions under two di€erent Po2 was also studied for comparison. For UV/H2O2, Po2 does not a€ect
the removal of 2-CB, while reducing the Po2 from 21 to 2 kPa resulted in 35% less CO2 formation
after 5 h irradiation. The similar O2-dependent destruction of the hydroxyl intermediates was also
observed in the UV/H2O2 system. We propose that molecular O2 acts as a key reactant following the
attack of a second ÁOH radical during the ring opening in the degradation of the hydroxyl byproducts.
7 2000 Elsevier Science Ltd. All rights reserved
Key wordsÐrole of oxygen, photocatalytic degradation, 2-chlorobiphenyl (2-CB), titanium dioxide
(TiO2)
INTRODUCTION
O2 levels on the rate of ÁOH radical formation on
the TiO2 surface. Oxygen is known to readily
Since it was ®rst used to destroy aqueous phase
polychlorinated biphenyls (Carey et al., 1976),
titanium dioxide (TiO2-anatase polymorph) photo-
duced conduction-band electrons during which
catalytic degradation of organic contaminants
has attracted much attention as an alternative
for the degradation of environmental contaminants
in air and water. Various classes of organic
compounds have been either partially degraded or
completely mineralized in UV-irradiated aqueous
TiO2 suspensions (Pelizzetti et al., 1990a; Mills
et al., 1993).
The primary reactions involved in TiO2 photoca-
talysis have been well documented (Turchi and
TiO2 photocatalytic degradation of gas phase 4-
Ollis, 1990). Oxygen was found to be essential for
semiconductor photocatalytic degradation of or-
ganic compounds (Al-Ekabi et al., 1991; Okamoto
et al., 1985a,b; Schwarz et al., 1997). Experiments
oxygen is more than that of merely an electron sca-
have demonstrated the dependence of initial reac-
tion rates and total mineralization of organic com-
pounds on O2 in the TiO2 photocatalysis (Al-Ekabi
et al., 1991; Okamoto et al., 1985a,b). Schwarz et
al. (1997) gave evidence of the in¯uence of dissolved
adsorb onto the TiO2 surface (Boonstra and Mut-
saers, 1975) and can be reduced by the photoin-
superoxides are formed (Rothenberger et al., 1985).
Oxygen and its reduced species, such as O2 Á and
ÁO22 , were suggested to play important roles in
TiO2-catalyzed photomineralization (Al-Ekabi et
al., 1991; Gratzel et al., 1990). In gas phase photo-
catalysis of trichloroethylene, instead of ÁOH rad-
icals, the gas phase O2-derived active radical
species, such as HOÁ2, ÁO2 , were suggested to initiate
the oxidation (Fan and Yates, 1996). In a study of
chlorophenol (4-CP), Gray et al. (1993) pointed out
that ÁOH radicals may not be exclusively respon-
sible for the photooxidation of 4-CP and the role of
venger.
The diculties in understanding the complex
roles of oxygen in photocatalysis are mainly due to
the lack of proper reaction intermediates. The 2-CB
used in this study is an ideal compound for this
purpose because it has two aromatic rings with
di€erent electron densities. The advantage of using
*Author to whom all correspondence should be addressed.
Tel.: +1-518-473-7299; fax: +1-518-473-7689; e-mail:
hongc@wadsworth.org
2-CB has been demonstrated by the formation of
stable aromatic intermediates at various stages in
2791

2792
Yongbing Wang and Chia-Swee Hong
Following CO2 measurement, the suspensions were sub-
jected to hexane extraction three times. The extracts were
dehydrated and followed by GC/ECD (HP 5890) analysis.
Details about 2-CB determination can be found elsewhere
(Hong et al., 1998). The hexane-extracted suspensions
were then acidi®ed to pH 01 with H2SO4 and extracted
three times with dichloromethane for more polar inter-
mediates. The hexane and dichloromethane extracts were
the process of TiO2 photocatalysis (Hong et al.,
1998).
The main objectives of this study were to investi-
gate the e€ect of O2 on the degradation of the aro-
matic intermediates and to provide mechanistic
insight into the role of dissolved oxygen in TiO2
photocatalysis. The e€ect of O2 on the evolution of
intermediates and CO2 in the UV/H2O2 system was
concentrated individually to around 1.0 ml with
Kuderna±Danish (K±D) apparatus and further concen-
a
also studied as a comparison to discern the function trated down to 0.1 ml with gentle nitrogen blowing. Five
microliters of each concentrated sample was injected into
GC/FTIR/MS (HP 5890/HP 6968B/HP 5970). The identi-
®cation and quantitation of the intermediates were
described in our previous papers (Hong et al., 1998; Wang
et al., 1998).
of molecular oxygen and that of the superoxide
species.
EXPERIMENTAL
Materials
RESULTS AND DISCUSSION
The nonporous P-25 TiO2 powder (80% anatase, aver-
age diameter 0.02 mm and surface area 050 m2/g) was a
gift from Degussa Corporation (Dublin, OH). 2-CB was
The conventional concept of oxygen function
Near UV irradiation of anatase-TiO2 generates
conduction-band electron (eCB) and valence-band
hole (h+VB) pairs, some of which are trapped on the
particle surface at some defect sites where redox
chemistry can take place (Turchi and Ollis, 1990).
Without a surface-bound reducible adsorbate, the
surface-trapped eCB and h+VB recombine rapidly
(Bahnemann et al., 1997). Dissolved oxygen readily
adsorbs on the TiO2 surface and can be reduced by
eCB to form superoxide ion radicals O2 Á and/or
O22 Á (Bahnemann et al., 1997). The surviving sur-
face-trapped h+VB then oxidizes TiO2 surface-sorbed
H2O or OH to produce ÁOH radicals (Matthews,
1984). Hitherto, conduction-band electron scaven-
ging, which is essential for ÁOH radical formation,
was considered the major function of the dissolved
O2 in semiconductor photocatalysis.
purchased from AccuStandard (New Haven, CT) and was
used without further puri®cation. Hydrogen peroxide
(30%, reagent grade) and Na2CO3ÁH2O (reagent grade)
were obtained from Baker Analyzed. All the organic sol-
vents used were nanograde from Mallinckrodt, Inc. (St
Louis, MO). Ultrapure compressed O2 and N2 used in this
study were provided by Praxair (Albany, NY). The water
used in all experiments was produced by a Milli-Q water
puri®cation system (Millipore).
Photocatalytic degradation
Aqueous 2-CB solutions were produced by the genera-
tor column technique (Hong and Qiao, 1995). Photocata-
lytic degradation of 2-CB was carried out in 100-ml
borosilicate glass bottles. 2-CB aqueous solutions with
TiO2 suspensions were stirred for 1 h in the dark to estab-
lish
a 2-CB adsorption±desorption equilibrium. Then,
50 ml of this solution was transferred into the bottle. The
bottles were sealed with Te¯on-lined septa and aluminum
crimp tops. The light intensity through borosilicate glass
was measured to be 1400 mW/cm2 using a radiometer
(Model DIX-365, Spectronic Corporation). A laboratory-
constructed `solarbox' equipped with eight UVA-340 light
tubes (Q-Panel Company) provided the UV irradiation
source. This UV source has a maximum emission at
340 nm. The reaction bottles were placed up-side down on
a special rack on a multi-magnetic stirrer (Lab-line instru-
ments, Inc.) in the solarbox. A ®xed distance (5 cm) from
the light source to the solution surface was kept for all the
experiments. To better monitor the degradation process, a
relatively low TiO2 loading of 25 mg/l for all the exper-
iments was used. The 2-CB solutions were used at natural
pH (6.8 2 0.1). Hydrogen peroxide concentration for the
UV/H2O2 experiments was 0.01 M. H2O2 was added to
the reaction solutions prior to irradiation.
According to Henry's Law, the dissolved oxygen
concentration is proportional to the gas phase Po2.
To explore the e€ect of dissolved O2, TiO2 photoca-
Oxygen partial pressure
The reactor headspace was purged with pure N2 or O2
to control the oxygen partial pressure (Po2). Throughout
the purging, direct blowing on the solution was carefully
avoided. The headspace Po2 was then determined by an
HP 5890 GC equipped with a packed HP carbosphere col-
umn and a thermal conductivity detector. The detector
was operated at 1508C. The column temperature was kept
constant at 308C. Ambient air was used as the reference.
Fig. 1. E€ect of oxygen partial pressure on 2-CB disap-
pearance and CO2 formation in the UV/TiO2 system.
[TiO2]=25 mg/l. Illumination intensity was 1400 mW/cm2.
(Q q) 186 kPa, (* w) 21 kPa, (T t) 4 kPa, (W r) 0.5
kPa. The solid symbols are for 2-CB and the hollow sym-
bols for CO2.
Analytical determination
CO2 was quantitated by headspace gas chromatographic
analysis with a GC (HP 5890)/FTIR (HP 6968B) equipped
with a 25 m HP-Ultra 2 column. The detailed method was
described previously (Hong et al., 1998).

Photomineralization of 2-chlorobiphenyl
2793
talytic degradation of 2-CB was conducted under and 1/Po2 as shown in Fig. 2 curve b. From the
four di€erent Po2s in the closed reaction vessels. values of the intercept 1/k and the slope 1/(kKo2),
1
Figure 1 shows the e€ect of Po2 on 2-CB disappear- Ko2 ˆ 0:88 (kPa) was derived. This is comparable
1
ance and total mineralization. A higher Po2 was ac- with those of 0.11 (kPa) (Okamoto et al., 1985a)
1
companied with a quicker 2-CB disappearance and and 1.1 (kPa) (Augugliaro et al., 1988) reported
a faster CO2 formation. The molar ratio of CO2 in the studies of the TiO2 photocatalytic degra-
produced to 2-CB removed after 80 min of ir- dation of phenol.
radiation for Po2s of 0.5, 4, 21, and 186 kPa were
E€ect of O2 on intermediate degradation
1.3, 2.7, 6.3, and 5.4, respectively, which are far
lower than the stoichiometric ratio of 12 at com-
Evolution of the aromatic intermediates was
plete mineralization. At the low Po2 of 0.5 kPa, investigated using single ion monitoring mass spec-
only 16% of the 2-CB was mineralized to CO2, trometry. The distribution pro®les of some inter-
compared with mineralization of 53%, 84% and mediates as a function of reaction time are shown
94%, respectively, under increasing Po2s, after 5 h in Figs 3 and 4. There was a general trend towards
illumination.
more accumulation of hydroxyl by-products at a
The initial reaction rate (ro) of 2-CB degradation lower Po2. Meanwhile, the maximum accumulation
®rst increased rapidly and then progressively leveled shifted to longer reaction times (Fig. 3). Noting
o€ with increasing Po2 (Fig. 2 curve a). Recently, that a slower rate of 2-CB disappearance was ac-
Schwarz et al. (1997) reported a similar relationship companied with a lower Po2 (Fig. 1) and the pri-
between Po2 and the observed ÁOH radical for- mary intermediates are directly formed from the
mation rate on the TiO2 surface. The initial rate of loss of H or Cl from 2-CB by ÁOH radical attack
2-CB removal was directly related to the concen- (Hong et al., 1998), the higher accumulations of the
tration of the TiO2 surface-adsorbed ÁOH radical. hydroxyl byproducts under lower O2 levels is
The result demonstrates the function of O2 in pre- obviously the result of their slower degradation
venting electron and hole recombination.
rates. It is also noted that the one-ring aromatic in-
The Langmuir±Hinshelwood reaction kinetics termediates at the Po2 of 0.5 kPa were hardly
model is often used (Okamoto et al., 1985a,b; detectable compared to that in the higher Po2 con-
Augugliaro et al., 1988) to describe the relation of ditions (Fig. 4), which implies that the photocataly-
ro and Po2:
tic activity responsible for the transformation of the
primary by-products was nearly cut o€ under this
low O2 condition. The observed higher accumu-
lations of the primary intermediates under lower
Po2s suggest that O2 was needed for the degra-
dation of the hydroxyl by-products, i.e., the clea-
ro ˆ kKo2Po2=ꢀ1 ‡ Ko2Po2†
where Ko2 is the adsorption equilibrium constant of
O2 on the TiO2 surface under aqueous colloid con-
ditions and k is the reaction rate constant. A
directly transformed form of the above equation is:
1=ro ˆ 1=ꢀkKo2Po2† ‡ 1=k
which predicts a linear relationship between 1/ro
Fig. 3. E€ect of oxygen partial pressure on the evolution
of two primary intermediates in the UV/TiO2 system.
Fig. 2. (a) E€ect of oxygen partial pressure on reaction Experimental conditions are as in Fig. 1. (*) 21 kPa, (W)
rate; (b) 1/ro vs 1/Po2. 0.5 kPa.

2794
Yongbing Wang and Chia-Swee Hong
vage of one of the aromatic rings of 2-CB. The
marginal formation of the secondary and tertiary
by-products observed at the Po2 of 0.5 kPa is
obviously due to the residual oxygen, because there
was no e€ort made to completely remove the oxy-
gen from the system. Once the oxygen is consumed,
the degradation of the primary intermediates may
cease.
For the secondary intermediate, i.e., the aromatic
aldehyde, carboxylation of the branch chains to
form 2-chlorobenzoic acid may be the major path-
way of their oxidation, while direct ring cleavage
should be a much slower route (Augugliaro et al.,
1988). For the aromatic acid, i.e., 2-chlorobenzoic
acid, a slower build-up process and a smaller ac-
cumulation peak (Fig. 4) accompanied a lower Po2,
because the formation was slower under a lower
Po2.
Fig. 5. E€ect of oxygen partial pressure on 2-CB disap-
pearance and mineralization in the irradiated H2O2 sol-
utions. [H2O2]=0.01 M. (* w) 21 kPa, (Q q) 2 kPa. The
solid symbols are for 2-CB and the hollow symbols for
CO2.
The involvement of molecular oxygen in ring cleavage
H2O2-photosensitized degradation of organic
compounds has been well-documented as an ÁOH-
initiated destruction process (Izumi et al., 1981; Ste-
fan et al., 1996). Besides the homogeneous property,
a unique characteristic of the UV/H2O2 system is
that there is no molecular O2 reduction by electrons
as in the UV/TiO2 system (Howe and Gratzel,
1987; Izumi et al., 1981; Liao and Gurol, 1995; Ste-
fan et al., 1996). Therefore, by comparing the in¯u-
ence of Po2 on the intermediate degradation in
these two systems, we can discern whether the mol-
ecular O2 itself or the superoxide species formed
from the reduction of O2 on the TiO2 surface, or
both, are responsible for the observed O2-dependent
destruction of the primary aromatic intermediates
in the process.
Figure 5 shows the disappearance of 2-CB and
the resulting formation of CO2 in the irradiated
H2O2 solutions with Po2s of 21 and 2 kPa. The dis-
appearance of 2-CB was similar in the two cases,
because ÁOH radicals were directly generated
through photolysis of H2O2. However, reducing the
Po2 from the ambient level of 21 kPa to 2 kPa sig-
ni®cantly lowered the CO2 formation. When Po2
was 21 kPa, 2-CB was completely mineralized to
CO2 after 5 h irradiation, in contrast to only 65%
mineralization under the lower O2 level. The ac-
cumulation of some aromatic intermediates as a
Fig. 4. E€ect of oxygen partial pressure on the evolution
of one secondary and one tertiary intermediates in the
UV/TiO2 system. Experimental conditions are as in Fig. 1.
(*) 21 kPa, (W) 0.5 kPa.
Fig. 6. E€ect of oxygen partial pressure on the evolution
of two primary intermediates in the irradiated H2O2 sol-
utions. [H2O2]=0.01 M. (*) 21 kPa, (w) 2 kPa.

Photomineralization of 2-chlorobiphenyl
2795
function of reaction time is depicted in Figs 6 and dence that molecular O2 plays a key role in the
7. As found in the TiO2 photocatalysis, the re- ring-opening process.
duction of Po2 signi®cantly increased the accumu-
Nucleophilic
OH
substitution
remarkably
lation of the hydroxyl byproducts. The times for increases the electron density of the benzene ring,
the highest accumulation for the hydroxyl inter- which normally makes the double bonds more vul-
mediates were shifted about 30 min later under the nerable to an attack from a peroxo group. Gratzal
lower O2 level. Considering the formation rates for et al. (1990) has proposed that ÁO22 formed in the
the primary products were the same, as indicated UV/TiO2 system directly attacks an aromatic
by the same rate of 2-CB disappearance (Fig. 5), double bond. Recently, Bahnemann et al. (1997)
the degradation of the hydroxyl byproducts was reported that most of the trapped charge carriers
slower at the reduced Po2. Therefore, the aromatic (eCB and h+VB) recombine within 200 ns, only a min-
ring-opening process for the hydroxyl intermediates ority survives and is available for the reduction of
showed an O2-dependent behavior in both UV/ surface-sorbed molecular O2. This indicates that
H2O2 and UV/TiO2 systems.
molecular O2 is the dominant oxygen species on the
The secondary intermediate, i.e., the aromatic TiO2 surface. Therefore, instead of a reduced oxy-
aldehyde, in the UV/H2O2 system was found to ac- gen species, molecular oxygen seems to be directly
cumulate at similar levels under two di€erent Po2, involved in the ring-opening reactions for 2-CB in
which is much lower than that in UV/TiO2 system both UV/TiO2 and UV/H2O2 systems.
with similar O2 conditions. This may be due to the
Possible mechanism for O2-dependent ring opening
relatively fast branch-chain carboxylation process in
this homogeneous system.
In the TiO2 photocatalytic degradation of other
In contrast to that in the TiO2 photocatalysis sys- aromatics, dimers or diol was often reported as the
tem, the aromatic acid in the UV/H2O2 system had minor byproducts (Okamoto et al., 1985a; Pelizzetti
signi®cantly higher accumulations at the lower Po2. et al., 1990a,b; Pichat et al., 1993). In this study,
The degradation of the primary intermediates the failure to detect chlorobiphenol dimers or chlor-
became slower under the reduced Po2, which indi- obiphenyl-diol could be due to the analytical limi-
cates a slower formation process for the aromatic tations, however, oxidative ring cleavage should be
acid. Therefore, the degradation of 2-chlorobenzoic the major route for the further oxidation of the hy-
acid was slower under the lower Po2. This suggests droxyl intermediates (Okamoto et al., 1985a; Peliz-
that the ring cleavage for the aromatic acid may zetti et al., 1990a,b; Pichat et al., 1993).
also require O2.
The intermediate pro®le indicates that the single-
The O2-dependent aromatic ring cleavage ring aromatic intermediates were all derived from
observed in the UV/H2O2 system, in which there is
no superoxide oxygen species, provides direct evi-
Fig. 7. E€ect of oxygen partial pressure on the evolution
of one secondary and one tertiary intermediates in the
irradiated H2O2 solutions. [H2O2]=0.01 M. (*) 21 kPa, Fig. 8. Proposed mechanism for the O2-dependent ring-
(w) 2 kPa.
opening process of 2-chlorobiphenyl-ol.

2796
Yongbing Wang and Chia-Swee Hong
the opening of the ring with a hydroxyl group results from this study suggest that dissolved oxy-
(Hong et al., 1998). It becomes clear that the open- gen may play a critical role in the degradation of
ing of the aromatic ring with hydroxyl group con- the hydroxyl aromatic products, which are poten-
sists of the attack of a second ÁOH radical followed tially toxic, in TiO2 photocatalysis of PCBs and
by an involvement of molecular O2.
other halogenated aromatics.
Figure 8 gives a proposed mechanism for the
observed O2-dependent ring-opening process. Rad-
icals resulting from ÁOH radical addition are fre-
quently detected in TiO2 photocatalysis of aromatic
compounds (Harbour and Hair, 1979; Schwarz et
al., 1997; Sehested et al., 1977). Before the hydroxyl
molecules escape from the TiO2 surface, a diol rad-
ical (A) is expected to form from the attack of a
second ÁOH radical. This diol radical was also pro-
posed in an ÁOH-initiated ring cleavage for o-cresol
in atmospheric chemical processes (Chien et al.,
1998) although there was no demonstration. The
TiO2 surface-sorbed molecular O2 is then readily
trapped by radical A to form B (Cermenati et al.,
1997; Okamoto et al., 1985a; Takagi et al., 1985).
The highly reactive `on-site' peroxo radical then
attacks an ortho hydroxyl group. This results in the
formation of a less reactive radical (HOÁ2) and an
epoxide intermediate (C) (Chien et al., 1998; Gratzel
et al., 1990), which is subjected to immediate hy-
drolysis, rearrangement, the opening of the ring,
and formation of a bifunctional species (D). D is
then quickly degraded and proceeds to the for-
mation of the relatively stable aldehyde or ketone
intermediates. Generation of aldehyde or ketone
may depend on the position of ring opening. The
superoxide radicals, h+VB, eCB, ÁOH radicals, and O2
CONCLUSIONS
1. A higher Po2 resulted in faster initial rate of 2-
CB transformation and CO2 formation. The
e€ect of oxygen partial pressure on 2-CB miner-
alization is more signi®cant than that on 2-CB
transformation. At Po2 of 0.5 kPa, 75% of 2-CB
were transformed but only 16% of 2-CB were
mineralized to CO2 after 5 h illumination.
2. The distribution pro®les of aromatic intermedi-
ates indicate that molecular oxygen was needed
for the degradation of the hydroxyl by-products,
which implies the involvement of molecular oxy-
gen in the cleavage of aromatic ring.
3. The similar O2-dependent destruction of the hy-
droxyl intermediates was also observed in the
UV/H2O2 system.
AcknowledgementsÐThis research was supported by
NIEHS Superfund Basic Program under Grant No. ESO
4913-07. The authors wish to thank Dr Rajinder Narang
for his assistance with oxygen analysis.
REFERENCES
as well may all be involved in the oxidative degra- Al-Ekabi H., Safarzadeh-Amiri A., Sifton W. and Story J.
dation of D. In UV/H2O2 system, HOÁ2 and ÁOH
(1991) Advanced technology for water puri®cation by
heterogeneous photocatalysis. Int. J. Environ. Pollut.
1(12), 125±136.
Augugliaro V., Palmisano L., Sclafani A., Minero C. and
radicals may similarly be involved in this degra-
dation process.
The O2-dependent aromatic ring-cleavage of the
hydroxyl intermediates is obviously one of the
major rate-limiting steps in the process of TiO2
photocatalytic mineralization of 2-CB. This may
also be true in UV/H2O2 system.
Pelizzetti E. (1988) Photocatalytic degradation of phenol
in aqueous titanium dioxide dispersions. Toxicol.
Environ. Chem. 16(2), 89±109.
Bahnemann D. W., Hilgendor€ M. and Memming R.
(1997) Charge carrier dynamics at TiO2 particles: reac-
tivity of free and trapped holes. J. Phys. Chem. B
101(21), 4265±4275.
Hydroxyl aromatic products were often found to
be the initial oxidation products for the degradation
of aromatic compounds in semiconductor photoca-
talysis (Cermenati et al., 1997; Theurich et al.,
1996) and other ÁOH-based oxidation processes
(Chien et al., 1998; Moza et al., 1988). The identi®-
cation of the seven 2-chlorobiphenyl-ols and biphe-
nyl-2-ol as the primary byproducts in TiO2
photocatalysis of 2-CB (Hong et al., 1998; Wang et
al., 1998) indicates that hydroxyl PCBs might be
the major preliminary products in TiO2 photocata-
lysis of PCBs. Polychlorinated biphenyl-ol and
polychlorinated biphenyl-diol were also found in
TiO2 photocatalytic degradation of dichlorophenol
as the condensation products (Minero et al., 1995).
These hydroxyl intermediates are similar to those
PCB metabolites in biological systems, which are
potentially toxic or more toxic than the parent PCB
Boonstra A. H. and Mutsaers C. A. H. A. (1975) Relation
between the photoadsorption of oxygen and the number
of hydroxyl groups on a titanium dioxide surface. J.
Phys. Chem. 79(16), 1694±1698.
Carey J. H., Lawrence J. and Tosine H. M. (1976) Photo-
dechlorination of PCBs in the presence of titanium diox-
ide in aqueous suspensions. Bull. Environ. Contam.
Toxicol. 16(6), 697±701.
Cermenati L., Pichat P., Guillard C. and Albini A. (1997)
Probing the TiO2 photocatalytic mechanisms in water
puri®cation by use of quinoline, photo-fenton generated
OH.bul. radicals and superoxide dismutase. J. Phys.
Chem. B 101(14), 2650±2658.
Chien C.-J., Charles M. J., Sexton K. G. and Je€ries H.
E. (1998) Analysis of airborne generated OH.bul.radi-
cals and superoxide dismutase. J. Phys. Chem.
101(14), 2650±2658.
B
Fan J. and Yates Jr J. T. (1996) Mechanism of photooxi-
dation of trichloroethylene on TiO2: detection of inter-
mediates by infrared spectroscopy. J. Am. Chem. Soc.
118(19), 4686±4692.
congeners (Gierthy et al., 1997; Safe, 1994). The Gierthy J. F., Arcaro K. F. and Floyd M. (1997) Assess-

Photomineralization of 2-chlorobiphenyl
2797
ment of PCB estrogenicity in a human breast cancer cell
line. Chemosphere 34(57), 1495±1505.
Gratzel C. K., Jirousek M. and Gratzel M. (1990) De-
composition of organophosphorus compounds on
photoactivated titanium dioxide surfaces. J. Mol. Catal.
60(3), 375±387.
Gray K. A., Sta€ord U., Dieckmann M. S. and Kamat P.
V. (1993) Photocatalytic Puri®cation and Treatment of
Water and Air, eds D. F. Ollis and H. Al-Ekabi, pp.
455±472. Elsevier, Amsterdam.
Harbour J. R. and Hair M. L. (1979) Radical intermedi-
ates in the photosynthetic generation of hydrogen per-
oxide with aqueous zinc oxide dispersions. J. Phys.
Chem. 83(6), 652±656.
Hong C. S. and Qiao H. (1995) Generator column deter-
mination of aqueous solubilities for non-ortho and
mono-ortho substituted polychlorinated biphenyls. Che-
mosphere 31(11/12), 4549±4557.
Hong C. S., Wang Y. and Bush B. (1998) Kinetics and
products of the TiO2 photocatalytic degradation of 2-
chlorobiphenyls in water. Chemosphere 36(7), 1653±
1667.
Howe R. F. and Gratzel M. (1987) EPR study of hydrated
anatase under UV irradiation. J. Phys. Chem. 91(14),
3906±3909.
Izumi I., Fan F.-R. F. and Bard A. J. (1981) Hetero-
geneous photocatalytic decomposition of benzoic acid
and adipic acid on platinized titanium dioxide power.
The photo-kolbe decarboxylative route to the break-
down of the benzene ring and to the production of
butane. J. Phys. Chem. 85(3), 218±223.
Liao C.-H. and Gurol M. D. (1995) Chemical oxidation
by photolytic decomposition of hydrogen peroxide.
Environ. Sci. Technol. 29(12), 3007±3014.
Matthews R. W. (1984) Hydroxylation reactions induced
by near-ultraviolet photolysis of aqueous titanium diox-
ide suspensions. J. Chem. Soc., Faraday Trans. I 80(2),
457±471.
Mills A., Davies R. H. and Worsley D. (1993) Water puri-
®cation by semiconductor photocatalysis. Chem. Soc.
Rev. 22(6), 417±425.
Minero C., Pelizzetti E., Pichat P., Sega M. and Vincenti
M. (1995) Formation of condensation products in
advanced oxidation technologies: the photocatalytic
degradation of dichlorophenols on TiO2. Environ. Sci.
Technol. 29(9), 2226±2234.
Moza P. N., Fytianos K., Samanidou V. and Korte F.
(1988) Photodecomposition of chlorophenols in aqueous
medium in the presence of hydrogen peroxide. Bull.
Environ. Contam. Toxicol. 41(5), 678±682.
Okamoto K., Yamamoto Y., Tanaka H., Tanaka M. and
Itaya A. (1985a) Heterogeneous photocatalytic de-
composition of phenol over anatase powder. Bull.
Chem. Soc. Jpn 58(7), 2015±2022.
(1985b) Kinetics of heterogeneous photocatalytic de-
composition of phenol over anatase titanium dioxide
powder. Bull. Chem. Soc. Jpn 58(7), 2023±2028.
Pelizzetti E., Maurino V., Minero C., Carlin V., Pramauro
E., Zerbinati O. and Tosato M. L. (1990a) Photocataly-
tic degradation of atrazine and other s-triazine herbi-
cides. Environ. Sci. Technol. 24(10), 1559±1565.
Pelizzetti E., Minero C. and Maurino V. (1990b) The role
of colloidal particles in the photodegradation of organic
compounds of environmental concern in aquatic sys-
tems. Adv. Colloid Interface Sci. 32(23), 271±316.
Pichat P., Guillard C., Maillard C., Amalric L. and D'Oli-
veira J.-C. (1993) Photocatalytic Puri®cation and Treat-
ment of Water and Air, eds D. F. Ollis and H. Al-Ekabi,
pp. 207±223. Elsevier Science Publishers B.V, The Neth-
erlands.
Rothenberger G., Moser J., Gratzel M., Serpone N. and
Sharma D. K. (1985) Charge carrier trapping and
recombination dynamics in small semiconductor par-
ticles. J. Am. Chem. Soc. 107(26), 8054±8059.
Safe S. H. (1994) Polychlorinated biphenyls (PCBs): en-
vironmental impact, biochemical and toxic responses,
and implication for risk assessment. Crit. Rev. Toxicol.
24(2), 87±149.
Schwarz P. F., Turro N. J., Bossmann S. H., Braun A.
M., Abdel Wahab A.-M. and Duerr H. (1997) A new
method to determine the generation of hydroxyl radicals
in illuminated TiO2 suspensions. J. Phys. Chem.
101(36), 7127±7134.
B
Sehested K., Holcman J. and Hart E. J. (1977) Conversion
of hydroxycyclohexadienyl radicals of methylated ben-
zenes to cation radicals in acid media. J. Phys. Chem.
81(14), 1363±1367.
Stefan M. I., Hoy A. R. and Bolton J. R. (1996) Kinetics
and mechanism of the degradation and mineralization
of acetone in dilute aqueous solution sensitized by the
UV photolysis of hydrogen peroxide. Environ. Sci. Tech-
nol. 30(7), 2382±2390.
Takagi K., Fujioka T., Sawaki Y. and Iwamura H. (1985)
An oxygen-18 tracer on the titanium dioxide-sensitized
photooxidation of aromatic compounds. Chem. Lett. 7,
913±916.
Theurich J., Lindner M. and Bahnemann D. W. (1996)
Photocatalytic degradation of 4-chlorophenol in aerated
aqueous titanium dioxide suspensions: a kinetic and
mechanistic study. Langmuir 12(26), 6368±6376.
Turchi C. S. and Ollis D. F. (1990) Photocatalytic degra-
dation of organic water contaminants: mechanisms
involving hydroxyl radical attack. J. Catal. 122(1), 178±
192.
Wang Y., Hong C. S. and Bush B. (1998) Decomposition
of 2-chlorobiphenyl in aqueous solutions by UV ir-
radiation with the presence of titanium dioxide. J. Adv.
Oxid. Technol. 3(3), 299±306.
Okamoto K., Yamamoto Y., Tanaka H. and Itaya A.