Photocatalysis by illuminated titania: Oxidation of hydroquinone and p-benzoquinone

Article abstract of DOI:10.1007/PL00010219

Photocatalytic oxidation of hydroquinonc and p-benzoquinone, the main intermediates which are formed in the process of phenol oxidation on irradiated titania, was studied. The reaction proceeds via several steps. Some of the intermediates were detected by HPLC and GC-MS: ethanedial, glycerol, and 1,2,4-trihydroxybenzene. Under illumination in the presence of TiO₂, p-benzoquinone initially transforms partly to hydroquinone, partly oxidizes to unidentified intermediates. The residue, about 4% of the concentration of hydroquinone, undergoes slow oxidation together with the hydroquinone. The rate of hydroquinone photooxidation shows 1^st order behaviour.

Full text of DOI:10.1007/PL00010219

Monatshefte fuÈr Chemie 130, 377±384 (1999)
Photocatalysis by Illuminated Titania:
Oxidation of Hydroquinone and p-Benzoquinone
Ã
ꢀ
Andrzej Sobczynski , èukasz Duczmal, and Anna Dobosz
ꢀ ꢀ
Faculty of Commodity Science, Poznan University of Economics, PL-60067 Poznan, Poland
Summary. Photocatalytic oxidation of hydroquinone and p-benzoquinone, the main intermediates
which are formed in the process of phenol oxidation on irradiated titania, was studied. The reaction
proceeds via several steps. Some of the intermediates were detected by HPLC and GC-MS:
ethanedial, glycerol, and 1,2,4-trihydroxybenzene. Under illumination in the presence of TiO2,
p-benzoquinone initially transforms partly to hydroquinone, partly oxidizes to unidenti®ed
intermediates. The residue, about 4% of the concentration of hydroquinone, undergoes slow
oxidation together with the hydroquinone. The rate of hydroquinone photooxidation shows 1st order
behaviour.
Keywords. Photodegradation; Hydroquinone; p-Benzoquinone; Intermediates; Kinetics.
Photokatalyse durch belichtetes TiO2: Oxidation von Hydrochinon und p-Benzochinon
Zusammenfassung. Die Photokatalyse von Hydrochinon und p-Benzochinon, den wesentlichsten
Zwischenprodukten bei der Oxidation von Phenol an belichtetem TiO2, wurde untersucht. Die
È
È
Reaktion verlauft uber mehrere Stufen. Einige der Zwischenprodukte konnten mittels HPLC und
GC-MS nachgewiesen werden: Ethandial, Glycerin und 1,2,4-Trihydroxybenzol. Unter Lichtein-
wirkung in Gegenwart von TiO2 setzt sich p-Benzochinon zuerst teilweise zu Hydrochinon um,
teilweise wird es zu nicht identi®zierten Produkten oxidiert. Der RuÈckstand (ca. 4% der Menge an
Hydrochinon) wird zusammen mit Hydrochinon langsam weiter oxidiert. Die Kinetik der
È
Photooxidation von Hydrochinon verlauft nach erster Ordnung.
Introduction
If a semiconductor, e.g. TiO2, is illuminated by light with a wavelength equal or
higher than its forbidden gap, electrons and holes are generated on and near the
surface. Some of the electron-hole pairs recombine, some are trapped into the so-
called surface traps giving highly active surface species (Ti3‡ and Oꢀ or ÁOH) [1, 2].
In aqueous TiO2 suspensions saturated with air or oxygen, hydroxyl radicals are
mainly formed [3, 4]. They attack organic molecules present in the solution
forming organic radicals which can react further with photogenerated electrons or
holes or primary products of their reactions giving highly oxidized species; the
Ã
Corresponding author

Â
A. Sobczynski et al.
378
®nal products are carbon dioxide and water. The trapped electrons react mainly
with molecular oxygen dissolved in water. As a result, surface Oꢀ2 is formed; this
ion can be reduced further to H2O2, ÁOH, and OHꢀ [2]. The produced species may
react with the organic radicals formed in the oxidative route. Therefore, the organic
compounds present in the illuminated titania slurry undergo many chain,
consecutive, and other reactions [2, 4±11]. Finally, in an ideal case, all intermediate
compounds are fully mineralized, i.e. they are oxidized to carbon dioxide and
water.
Although photodecomposition of phenol in aerated aqueous TiO2 suspensions
leads ®nally to CO2 and H2O, a number of aromatic [12] and aliphatic [13]
intermediates are also obtained. Recent investigations have shown a big difference
between phenol and total organic carbon diminution [14], con®rming thus the
existence of a number of reaction intermediates. Most recent studies on
photocatalytic oxidation of phenol on illuminated TiO2 show that several ring
intermediates like di- and trihydroxybenzenes are present in the reaction mixture
[15]. It has also been observed that hydroquinone and p-benzoquinone are the
major intermediates of the process. The present paper deals with the photocatalytic
oxidation of these compounds.
Results and Discussion
As has been mentioned above, hydroquinone and p-benzoquinone are the main ring
intermediates formed in the photocatalytic oxidation of phenol in the presence of
TiO2. Quantitatively, about 15% hydroquinone and 4% p-benzoquinone (initial
phenol concentration: 2 Á 10ꢀ4 M) were detected after 3 hours of the photocatalytic
reaction [15]. Therefore, the photocatalytic phenol oxidation obviously follows
mainly the pathway phenol!hydroquinone (HQ)!p-benzoquinone (BQ)!other
intermediates.
It is well known from the chemistry of hydroquinone and p-benzoquinone that
both compounds can be easily transformed into one another [16, 17]; the normal
redox potential of BQ/HQ amounts to 0.715 V [18]. Also, both compounds may
undergo further hydroxylation [16, 17]. Irradiation of HQ and BQ solutions
(ꢀ ˆ 254 nm) can yield coupled products, e.g. 2,20,5,50-tetrahydroxybiphenyl [17].
The results of the photocatalytic oxidation of 2 Á 10ꢀ4 M solutions of HQ, BQ
and, in addition, an equimolar mixture of both compounds, are shown in Figs. 1±3.
Three techniques were applied in order to follow the disappearance of the
compounds: their total oxidation, i.e. their mineralization to CO2 and H2O, was
studied by means of total organic carbon measurements (TOC), formation of
intermediates, HPLC, and GC-MS.
It can be seen from Figs. 1±3 that in all solutions HQ and BQ are present,
although at different proportions. The HQ solution contains initially about 4% BQ
(Fig. 1). When illuminated, both compounds underwent decomposition; after 3 h
their concentrations were lowered by about 2.5 times (Fig. 1). It is worth
mentioning that under the conditions of the photocatalytic reaction the ratio BQ/
(BQ‡HQ) was nearly constant: the slurry still contained 4±5% BQ. If one assumes
that oxidation of hydroquinone yields primarily p-benzoquinone (see above), it is
clear that the last compound undergoes fast transformations resulting in a constant

Photocatalysis by Illuminated Titania
379
Fig. 1. Changes in the reaction solution vs. time during photocatalytic oxidation of hydroquinone;
the loss of TOC is expressed in units of hydroquinone (and p-benzoquinone) concentration
Fig. 2. Changes in the reaction solution vs. time during photocatalytic oxidation of p-benzoquinone;
the loss of TOC is expressed in units of p-benzoquinone (and hydroquinone) concentration
Fig. 3. Changes in the reaction solution vs. time during photocatalytic oxidation of a 1:1
hydroquinone/p-benzoquinone mixture

Â
A. Sobczynski et al.
380
BQ/(BQ‡HQ) ratio. Mineralization of the reaction mixture, i.e. oxidation to CO2
and H2O, proceeds much slower: after 3 h of illumination, only about 14% of initial
organic matter (BQ‡HQ) disappeared (see the TOC curve in Fig. 1).
The picture of the photocatalytic BQ oxidation is somewhat more complicated
(Fig. 2). Initially, the concentration of BQ in the reaction slurry was about
1.83 Á 10ꢀ4 M, that of HQ 0.17 Á 10ꢀ4 M (about 9% HQ). As before, the total
concentration was 2 Á 10ꢀ4 M. However, after 30 min illumination the concentra-
tions of BQ and HQ were 0.05 Á 10ꢀ4 M and 1.13 Á 10ꢀ4 M. Note that (i) the amount
of BQ decreased drastically, (ii) at the same time the amount of HQ increased
strongly, although to a lesser extent, and (iii) the ratio BQ/(BQ‡HQ) reached a
value close to 0.04. Comparing the BQ and HQ curves in Fig. 2, it can be seen that
during the initial 30 min of illumination only part of BQ is reduced to HQ. More
than 30% of the initial BQ undergoes partial oxidation and forms some
intermediates, a little is fully mineralized (see TOC curve), and the rest forms a
mixture with HQ in which the BQ content is about 4%. The reaction slurry contains
thus a rather huge amount of unknown primary products of the BQ oxidation; they
change the colour of the slurry to pale red. However, the red colour vanishes during
further irradiation. Also, during the next 30 min the reaction of full mineralization
proceeds faster. Prolonged illumination causes changes in HQ and BQ contamin-
ation similar to those observed before for the reaction of HQ (Fig. 1).
Figure 3 shows the variations of HQ and BQ concentrations during illumination
of the equimolar slurry of hydroquinone and p-benzoquinone in the presence of
TiO2. Here again BQ partly transforms to HQ (the concentration of the latter
increased) and partly oxidizes to other compounds. The ratio BQ/(BQ‡HQ) lowers
with time from the initial value of 0.5 to about 0.04±0.05. Upon further
illumination, the changes of the concentrations of HQ and BQ follow those
depicted in Figs. 1 and 2.
The products of the photocatalytic oxidation of HQ were extracted from the
reaction mixture without and with initial acetylation and analyzed by GC-MS.
Before, the hydroquinone solution was illuminated for 2 hours in the presence of
titania. Only two new compounds were detected in addition to those mentioned
above: 1,2,4-benzenetriol triacetate (after acetylation) and glycerol. Another
reaction intermediate is presumable ethanedial as determined by HPLC. The
compound was identi®ed on the basis of its UV/Vis spectrum.
As reported in the literature, photooxidation of HQ and BQ (direct photolysis in
the presence of oxygen without any photocatalyst) yields for HQ the dimer
2,20,5,50-tetrahydroxybiphenyl, for BQ hydroquinone, 1,2,4-benzenetriol, hydro-
xylated p-benzoquinone, and traces of 2,20,5,50-tetrahydroxybiphenyl [17].
Hydroxylated biphenyls were also observed in products of photocatalytic oxidation
of 2,4-dichlorophenol in the presence of ZnO [19]. Other authors claim the
existence of a number of aliphatic intermediates, among them acetate and formate,
during photocatalytic degradation of aromatic water pollutants, including phenols,
in the presence of TiO2 [13]. The results of the present investigations are not
contradictory to these reports: BQ yields mainly HQ and unidenti®ed intermediates
which decompose relatively fast during prolonged illumination in the presence of
TiO2. The photoreaction follows that observed for HQ (cf. Figs. 1±3). Note that the
changes in HQ and BQ concentrations during photocatalytic oxidation of a 1:1

Photocatalysis by Illuminated Titania
381
mixture of hydroquinone and p-benzoquinone con®rm clearly the results obtained
for the photocatalytic oxidation of BQ itself. Higher hydroxylated aromatic
intermediates (1,2,4-benzenetriol), glycerol and, presumable, ethanedial were
determined in the products of photocatalytic HQ oxidation. Acetate and formate as
well as hydroxylated biphenyls may not be detected by the applied methods.
However, a great amount of unidenti®ed organic matter (see TOC curves in Figs. 1
and 2) shows that a number of aliphatic compounds and, possibly, polymers exist.
The kinetics of the photocatalytic reaction of HQ destruction were studied. The
results, shown in Fig. 4, refer to the sum of BQ and HQ. Initial rates of the
photocatalytic reaction, estimated from the data of Fig. 4, were used for further
calculations. The 1/r vs. 1/c dependence (r ˆ dc/dt, i.e. the reaction rate, c:
concentration of the substrate) is depicted in Fig. 5. One can see from the Fig. 5
that, although the points corresponding to the initial reaction rates somewhat
Fig. 4. Kinetics of hydroquinone disappearance; individual curves refer to different hydroquinone
concentrations
Fig. 5. Plot of 1/r vs. 1/c (for de®nition of r and c, cf. text)

Â
A. Sobczynski et al.
382
scatter, a linear dependence exists rather than another kind of relationship.
Therefore, the reaction shows 1st order behaviour, and the kinetic constant can be
calculated [14]: k ˆ 1.5 Á 10ꢀ8 mol Á dmꢀ3 Á sꢀ1, Kads ˆ 4.6 Á 104 dm3 Á molꢀ1. First
order kinetics have already been reported for the photocatalytic phenol oxidation in
the presence of titania (see e.g. Refs [14, 20, 21] and references cited therein). The
k and Kads values also are close to those reported before for 1,3-benzenediol
photooxidation under similar conditions [22]. The use of the initial reaction rates
for kinetic calculations allows to avoid the in¯uence of the reaction intermediates
on the estimated constants. During the photocatalytic studies, illumination and
oxygen content in the reactor solution were kept constant; therefore, these two
variables are not included in the rate expression.
Conclusions
Photocatalytic oxidation of hydroquinone and p-benzoquinone ± the main
intermediates which are formed in the process of phenol photodestruction on
illuminated titania ± proceeds via several reaction steps which involve several
reaction intermediates. Some of them were determined by HPLC and GC-MS:
ethanedial, glycerol, and 1,2,4-trihydroxybenzene. Interestingly, BQ partly under-
goes fast oxidation, partly transformation to HQ when irradiated in solution in the
presence of TiO2. The oxidation reaction follows that of pure HQ. The
photocatalytic oxidation of HQ seems to be of 1st order vs. substrate concentration,
in accordance with data reported in the literature for other compounds [20].
To our knowledge, this is the ®rst study on photocatalytic oxidation of
hydroquinone and p-benzoquinone. It is our opinion that recognition of main
reaction products and rate constants of photocatalytic oxidation of dihydroxy-
benzenes can help to elucidate also the mechanism of the process of phenol
photocatalytic destruction used in waste water treatment.
Experimental
Photocatalytic studies
The experiments were performed in a non-continuous mode (batch reactor) in a photoreactor similar
to that described earlier [14, 21]. The reactor consisted of an external glass tube with a conical
bottom equipped with a septum and a draft tube placed in the center of the external one. The reactor
content was mixed by an air stream applied at the bottom using a needle and ¯owing up through the
draft tube. The dimensions of the reactor were as follows: inner diameter of the external tube,
40 mm; h, 160 mm; inner diameter of the draft tube, 10 mm; h, 120 mm; total volume of the reactor,
ca. 190 ml. The photoreactor was ®lled with 165 ml of a reaction solution and 0.05 g TiO2 anatase
(99.9%, Aldrich). Depending on the experiment, the reaction solution contained hydroquinone (99%,
Aldrich), p-benzoquinone (99%, Aldrich), or a mixture (1:1) of both compounds. The titania powder
was sonicated before illumination with a small volume of the reactor solution. The reactor was
illuminated from a side wall with a 180 W medium pressure Hg lamp. The Pyrex walls of the reactor
determined a lower limit of entering light (about 300 nm cutoff ®lter). The photoreactor was cooled
by an air stream; the reaction temperature was maintained at 28Æ2ꢁC. During the photoreaction, 1 ml
of the slurry was withdrawn every 15 or 30 min using a syringe and, after ®ltration over Millipore
Millex GV3 units, analyzed by HPLC (Waters) using both ¯uorescence (Waters 474) and photodiode

Photocatalysis by Illuminated Titania
383
array (Waters 991) detectors. The reaction products were separated on a Supelcosil LC8 column
using water-methanol with a gradient concentration as an eluent.
For TOC determinations, 2 ml of the irradiated slurry were withdrawn from the photoreactor
every 30 min and, after ®ltration over the Millipore units, analyzed on a CA-10 TOC Shimadzu
Analyser.
GC-MS experiments were performed as follows: ®rstly, 165 ml 2 Á 10ꢀ4 M HQ and 0.05 g TiO2
were illuminated 2 h in the photoreactor; next, the phenols in the slurry were acetylated according to
a procedure described in Refs. [22, 23]. After extraction with CH2Cl2 and partial evaporation of the
extractant, the residue was analyzed on a mass spectrometer AMD-402 (Germany) coupled with a
gas chromatograph (Hewlett Packard Model 5890 Series II). The analyzed products were separated
on a DB-1 column using a temperature program (80±300ꢁC). For MS analysis, an ionization energy
of 70 eVand an acceleration voltage of 8 kV were applied. Simple extraction of the reaction products
(without acetylation) from the slurry after illumination was also performed, and products were
analyzed by GC-MS.
The photocatalyst
The titania ± TiO2 anatase, 99.9%, Aldrich ± possessed a BET speci®c surface area of 9.85 m2 Á gꢀ1
as measured on a Micromeritics ASAP 2010 apparatus. It showed the typical anatase absorption
spectrum with an absorption onset at about 400 nm. The re¯ectance spectrum, not shown in this
paper, was measured on Specord M-40 spectrophotometer. The water used for the photocatalytic
studies was doubly distilled in a quarz water still. All reactants were of p.a. purity or HPLC
speci®ed.
Acknowledgements
The authors are grateful to the Commitee of Scienti®c Research (KBN) in Warsaw, Poland (project 6
P04G 055 12) for funds to perform this work. The authors thank Mrs. Danuta Siatkowska from the
ꢀ
A. Mickiewicz University in Poznan, Poland, for GC-MS analyses.
References
[1] Anpo M (1989) Res Chem Intermed 11: 67
[2] Gerischer H, Heller A (1991) J Phys Chem 95: 5261
[3] Brezova V, Stasko A, Lapcik Jr L (1991) J Photochem Photobiol A Chem 59: 115
[4] Brezova V, Stasko A (1994) J Catal 147: 156
[5] Lawless D, Serpone N, Meisel D (1991) J Phys Chem 95: 5166
[6] Goldstein S, Czapski G, Rabani J (1994) J Phys Chem 98: 6586
[7] Fox MA (1993) The Role of Hydroxyl Radicals in the Photocatalysed Detoxi®cation of Organic
Pollutants: Pulse Radiolysis and Time Resolved Re¯ectance Measurements. In: Ollis DF, Al-
Ekabi H (eds) Photocatalytic Puri®cation and Treatment of Water and Air. Elsevier, Amsterdam
p 163
[8] Drapper RB, Fox MA (1990) Langmuir 6: 1396
[9] Stafford U, Gray KA, Kamat PV (1994) J Phys Chem 98: 6343
[10] Cermenati L, Pichat P, Guillard C, Albini A (1997) J Phys Chem B 101: 2650
[11] Kesselman JM, Weres O, Lewis NS, Hoffmann MR (1997) J Phys Chem B 101: 2637
[12] Okamoto K, Yamamoto Y, Tanaka H, Itaya A (1985) Bull Chem Soc Jpn 58: 2015
[13] Pichat P, Guillard C, Maillard C, Amalric L, D'Oliveira JC (1993) TiO2 Photocatalytic
Destruction of Water Pollutants: Intermediates; Properties ± Degradability Correlation; Effects
of Inorganic Ions and TiO2 Surface Area; Comparisons with H2O2 Processes. In: Ollis DF, Al-

Â
A. Sobczynski et al.: Photocatalysis by Illuminated Titania
384
Ekabi H (eds) Photocatalytic Puri®cation and Treatment of Water and Air. Elsevier, Amsterdam
p 207
ꢀ
[14] Sobczynski A, Gimenez J, Cervera-March S (1997) Monatsh Chem 128: 1109
ꢀ
[15] Sobczynski A, Duczmal è (to be published elsewhere)
[16] Stoddart JF (ed) (1979) Comprehensive Organic Chemistry, vol 1. Pergamon Press, Oxford, p 739
[17] Joshek HJ, Miller SJ (1966) J Am Chem Soc 88: 3273
[18] Fuson FC (1966) Reactions of Organic Compounds. Wiley, New York
[19] Sehili T, Boule P, Lemaire J (1991) Chemosphere 22: 1053
[20] Turchi CS, Ollis DF (1990) J Catal 122: 202
ꢀ
[21] Sobczynski A, Gimenez J, Cervera-March S (1996) Pol J Appl Chem 40: 103
[22] Duczmal è, Sobczynski A (1998) React Kinet Catal Lett (submitted)
ꢀ
[23] Standard Methods for the Examination of Water and Wastewater (1992) 18th edn, American
Public Health Association, p 6±77
[24] Bao ML, Pantani F, Barbieri K, Burrini D, Grif®ni O (1996) Chromatographia 42: 227
Received August 3, 1998. Accepted (revised) September 15, 1998