Solar Energy 77 (2004) 445–459

www.elsevier.com/locate/solener

Photocatalytic water treatment: solar energy applications

Detlef Bahnemann *

Department of Photocatalysis and Nanotechnology, Institut für Technische Chemie, Universität Hannover,
Callinstraße 3, D-30167 Hannover, Germany

Received 23 January 2004; received in revised form 31 March 2004; accepted 31 March 2004
Available online 26 August 2004

Communicated by: Associate Editor Sixto Malato-Rodriguez

Abstract

During the past 20 years research and development in the area of photocatalysis have been tremendous. One of the
major applications of this technology is the degradation of organic pollutants in water and air streams which is con-
sidered as one of the so-called advanced oxidation processes. This overview briefly describes the basic principles of
photocatalysis, focusing in particular on important mechanistic and kinetic aspects as well as on some requirements
for efficient photocatalysts. Since the comparatively high costs associated with the generation of light from electricity
constitute one of the major drawbacks particularly for the rapid commercialization of photocatalytic water treatment
units, solar photons have been utilized here with great success for many years now. Various solar reactors for the
photocatalytic water treatment are described in detail in the main part of this review including the comparison of their
overall performance.

2004 Elsevier Ltd. All rights reserved.

1. Introduction Considerable public attention has been focused on this

Persistent organic chemicals are present as pollut-
ants in wastewater effluent from industrial manufac-
turers and normal households, and in landfill
leachates. They can be found in ground water wells
and surface waters. In all cases they have to be re-
moved to protect our water resources or to achieve
drinking water quality. Therefore, many processes
have been proposed over the years and are currently
being employed to destroy these toxins. The so-called
photocatalytic detoxification has been discussed as an
alternative method for clean-up polluted water in the
scientific literature since 1976 (Carey et al., 1976).
*
Tel.: +49 511 762 5560; fax: +49 511 762 2774.
E-mail address: bahnemann@iftc.uni-hannover.de
possibility of combining heterogeneous catalysis with
solar technologies to achieve the mineralization of tox-
ins present in water (e.g., Hecht, 1990). Compilations
of substances which can mineralized using photocata-
lysis are available (Legrini et al., 1993; Blake, 1994).
Several reviews have been published discussing the
underlying reaction mechanisms and illustrating exam-
ples of successful laboratory and field studies (Kamat,
1993; Fox and Dulay, 1993; Herrmann et al., 1993;
Bahnemann et al., 1994; Serpone and Pelizzetti,
1989; Pelizzetti and Schiavello, 1991; Ollis and Al-
Ekabi, 1993; Hoffmann et al., 1995). While the overall
stoichiometry of most mineralizations appears to
be understood details of the complex reaction mecha-
nism are, however, still only known in a few selected
cases.

0038-092X/$ - see front matter
2004 Elsevier Ltd. All rights reserved.
doi:10.1016/j.solener.2004.03.031
446 D. Bahnemann / Solar Energy 77 (2004) 445–459
2. Mechanisms of photocatalysis
2.1. Initial processes
Semiconductors (e.g., TiO2, ZnO, Fe2O3, CdS and
ZnS) can act as sensitizers for light-induced redox proc-
esses due to their electronic structure which is character-
ized by a filled valence band and an empty conduction
band. Absorption of a photon of energy greater than
the bandgap energy leads to the formation of an elec-
tron/hole pair. In the absence of suitable scavengers,
the stored energy is dissipated within a few nanoseconds
by recombination (Rothenberger et al., 1985). If a suita-
ble scavenger or surface defect state is available to trap
the electron or hole, recombination is prevented and
subsequent redox reactions may occur. The valence
band holes are powerful oxidants (+1.0 to +3.5 V vs.
NHE depending on the semiconductor and pH), while
the conduction band electrons are good reductants
(+0.5 to
1.5 V vs. NHE) (Grätzel, 1989). Most organic
photodegradation reactions utilize the oxidizing power
of the holes either directly or indirectly; however, to pre-
vent a buildup of charge one must also provide a reduc-
ible species to react with the electrons. In bulk
semiconductor electrodes only one species, either the
hole or electron, is available for reaction due to band
bending (Memming, 1988). However, in very small sem-
iconductor particle suspensions both species are present
on the surface. Therefore, careful consideration of both
the oxidative and the reductive paths is required.
Fig. 1 shows a cartoon which is frequently used to
illustrate photocatalytic processes. It consists of a super-
position of the energy bands of a generic semiconductor
(valence band VB, conduction band CB) and the geo-
metrical image of a spherical particle. Absorption of a
photon with an energy h m greater or equal the bandgap
energy Eg generally leads to the formation of an elec-
Fig. 1. Principle Mechanism of Photocatalysis.
tron/hole pair in the semiconductor particle. These
charge carriers subsequently either recombine and dissi-
pate the input energy as heat, get trapped in metastable
surface states, or react with electron donors and accep-
tors adsorbed on the surface or bound within the electri-
cal double layer.
The exact nature of the main oxidizing species
formed on the surface of the semiconductor particles fol-
lowing the absorption of a photon is yet unclear. A sig-
nificant body of literature exists that the initial oxidation
of a pollutant molecule may either occur by indirect oxi-
dation via a surface-bound hydroxyl radical (i.e., a
trapped hole at the particle surface), directly via the va-
lence band hole before it is trapped either within the par-
ticle or at the particle surface or even via hydroxyl
radicals in solution.
In support of a hydroxyl radical as the principal reac-
tive oxidant in photoactivated TiO2 is the observation
that intermediates detected during the photocatalytic
degradation of halogenated aromatic compounds are
typically hydroxylated structures (Augugliaro et al.,
1988; Ollis et al., 1984; Turchi and Ollis, 1989; Turchi
and Ollis, 1990; Terzian et al., 1991). These intermedi-
ates are consistent with those found when similar aro-
matics are reacted with a known source of hydroxyl
radicals. In addition, ESR studies have verified the exist-
ence of hydroxyl and hydroperoxyl radicals in aqueous
solutions of illuminated TiO2 (Noda et al., 1993; Anpo
et al., 1985; Jaeger and Bard, 1979). Mao et al. have
found that the rate of oxidation of chlorinated ethanes
correlates with the C–H bond strengths of the organics,
which indicates that H atom abstraction by OH is an
important factor in the rate-determining step for oxida-
tion (Mao et al., 1991). The strong correlation between
degradation rates and concentration of the organic pol-
lutant adsorbed to the surface also implies that the hyd-
roxyl radicals are adsorbed at the surface (Moser et al.,
1991; Matthews and McEvoy, 1992; Ohtani and Nishi-
moto, 1993; Tunesi and Anderson, 1991; Tunesi and
Anderson, 1992; Phillips and Raupp, 1992; Stafford
et al., 1993). On the other hand, Mao et al. have ob-
served that trichloroacetic acid and oxalic acid are oxi-
dized primarily by valence band holes on TiO2 via a
photo-Kolbe process. It should be noted that these com-
pounds also have no hydrogen atoms available for
abstraction by OH (Mao et al., 1991).
The primary processes occurring upon bandgap irra-
diation of extremely small titanium dioxide particles
(diameter d = 2.5 nm) have been studied extensively
employing ultra-fast laser flash photolysis equipment
with picosecond or even subpicosecond time-resolution
(Serpone et al., 1995; Colombo and Bowman, 1995; Co-
lombo et al., 1995). While quantitative discrepancies
were apparent which could, for example, be explained
by slightly different preparation techniques of the TiO2
colloids, the authors agreed principally on the underly-
 D. Bahnemann / Solar Energy 77 (2004) 445–459
ing qualitative concepts of photocatalysis. Following
their generation within the short light pulse both, elec-
tron and hole, are extremely rapidly trapped in surface
states of the semiconductor particle. As has been shown
earlier, the short wavelength absorption corresponds to
the absorption of trapped holes and the long wavelength
absorption to trapped electrons (Bahnemann et al.,
1984a,b). Since the trapped electron exhibits this strong
transient optical absorption around 650 nm while the
trapped hole absorbs predominantly at shorter wave-
lengths, i.e., around 430 nm, a very broad, featureless
transient absorption spectrum ranging from 400 nm un-
til 800 nm is consequently observed immediately follow-
ing the laser flash. The recombination kinetics of the
trapped charge carriers have been studied in detail by
both groups of authors (Serpone et al., 1995; Colombo
and Bowman, 1995; Colombo et al., 1995).
Studying the decay kinetics of these transient absorp-
tion spectra in the presence of molecular oxygen, air or
molecular nitrogen, respectively, a rate constant
k1 = 7.6 · 107 L mol
1 s
1 has been determined for the
reaction
e
þ O ! O
ð1Þ
tr 2 2
of the trapped electrons with molecular oxygen (Bahn-
emann et al., 1997). This rate constant is considerably
smaller than that reported for the reaction of hydrated
electrons (e
aq ) with molecular oxygen in homogeneous
aqueous solutions (i.e., k = 1.9 · 1010 L mol
1 s
1 (Bux-
ton et al., 1988)). However, this difference is readily ex-
plained by the much more negative one-electron redox
potential of e
aq (i.e., E =
0.87 V (vs. NHE) (Schwartz,
1984)) compared with the value for the flat-band
potential of a colloidal TiO2 particle (i.e., E = 0.12

0.059 * (pH) V (vs. NHE) (Duonghong et al., 1982)).
Hence, while the hydrated electrons possess a consider-
able driving force for the reduction of molecular oxygen
to form the superoxide radical O
2 (E =
0.33 V (vs.
NHE) (Bielski et al., 1985)), at pH 3 the conduction
band electrons of the colloidal TiO2 particles barely
reach a one-electron potential negative enough to
reduce O2.
2.2. Degradation of pollutants
2.2.1. Reductive pathways
It is interesting to note that while many laboratory
studies have demonstrated the feasibility of the method
of photocatalytic detoxification for almost all classes
of hazardous chemicals, almost all of these studies have
been directed towards the oxidative degradation of or-
ganic pollutants in water. Only in a few cases a reductive
pathway has been discussed as the initiating step of pho-
tocatalytic conversions.
Perhalogenated hydrocarbons, however, often con-
tain the carbon atoms in their highest oxidation state
447
and can therefore not be attacked oxidatively. Following
an early report by Güsten and co-workers on the photo-
catalytic degradation of chlorofluoromethanes on zinc
oxide surfaces in the gas phase (Filby et al., 1981), Ollis
et al. (Hsiao et al., 1983; Ollis, 1985), Bahnemann et al.
(1987a), and Sabin et al. (1992) have shown that tetra-
chloromethane can be photocatalytically degraded in
aqueous TiO2-suspensions. Photonic efficiencies and
mechanistic details have, however, not been investigated
by these authors. In the case of halothane (2-bromo-
2-chloro-1,1,1-trifluoroethane) it could be shown that
reduction processes on colloidal TiO2-particles can be
very efficient with quantum yields for the elimination
of bromide and fluoride ions reaching U = 0.43 when
photoplatinized photocatalysts were used (Bahnemann
et al., 1987b). Moreover, the reaction of conduction
band electrons generated by light absorption in TiO2
particles with tetrabromomethane (Al-Ekabi et al.,
1989a), bromoform (Kuhler et al., 1993) and various
transition metal ions (Butler and Davis, 1993; Prairie
et al., 1993; Herrmann et al., 1986; Herrmann et al.,
1988) has also been studied.
Using tetrachloromethane, CCl4, as a model pollut-
ant, very detailed mechanistic studies have been carried
out (Hilgendorff et al., 1993a, 1996; Choi and Hoffmann,
1995). Hilgendorff et al. (1993a, 1996) observed that pro-
tons and chloride ions are formed when oxygen-contain-
ing aqueous suspensions of bare or platinized TiO2
particles and CCl4 are illuminated with light of sufficient
energy at pH 11, while the concentrations of O2 and
CCl4 are decreasing. The following overall stoichiometry
was reported in these studies:
TiO2 =hm
CCl4 þ 2H2 O ! CO2 þ 4Hþ þ 4Cl
ð2Þ
Apparently, the photocatalytic degradation of tetra-
chloromethane behaves like in an ‘‘ideal’’ photocatalytic
process which is not restricted by adsorption/desorption
phenomena, mass-transfer limitations or diffusion con-
trol, since it obeys zero order kinetics with the rate
increasing linearly with the substrate concentration
(Hilgendorff et al., 1996).
2.2.2. Oxidative pathways
A wide variety of organic compounds have been
tested for their ability to be degraded photocatalytically
with the initial reaction step being an oxidative attack
(Blake, 1994; Hoffmann et al., 1995). In the following
only one model compound, namely 4-chlorophenol,
has been selected to elucidate typical details of the
underlying reaction mechanism of these degradation
processes. Its photocatalytic reaction pathways have
been examined in detail and can be taken as representa-
tive for many of these very complex mechanisms. For a
more comprehensive listing of treatable pollutants the
reader is referred to several review articles which have
448 D. Bahnemann / Solar Energy 77 (2004) 445–459
been published over the years (Legrini et al., 1993;
Blake, 1994; Kamat, 1993; Fox and Dulay, 1993; Herr-
mann et al., 1993; Bahnemann et al., 1994; Serpone and
Pelizzetti, 1989; Pelizzetti and Schiavello, 1991; Ollis and
Al-Ekabi, 1993; Hoffmann et al., 1995).
The photodegradation of chlorophenols in aqueous
solution has received considerable attention because
these compounds are important xenobiotic micro-
pollutants of the aquatic environment originating, for
example, from industrial chemical synthesis. More
specifically, 4-chlorophenol (4-CP) is used for the pro-
duction of quinizarin (a dye), clofibrate (a drug), chlor-
phenesin and dichlorophen (fungicides) (Oudjehani and
Boule, 1992). Therefore, several investigations of the
photocatalytic decomposition of chlorophenols using
metal oxide semiconductors either in aqueous heteroge-
neous suspensions (Al-Sayyed et al., 1991; Sehili et al.,
1989; Barbeni et al., 1984; Stafford et al., 1994; Cunning-
ham and Sedlak, 1994) or in an immobilized form (Al-
Ekabi et al., 1989b; Hofstadler et al., 1994; Vinodgopal
et al., 1994) have been published. The kinetics of the
photocatalytic 4-CP degradation have commonly been
interpreted as being indicative of a Langmuir–Hinshel-
wood-type mechanism in which the limitation of the
4-CP decomposition rate at higher pollutant concentra-
tions is related to the extent of adsorption of the pollut-
ant molecule on the TiO2 surface (Al-Sayyed et al., 1991;
Stafford et al., 1994). However, Cunningham et al.
showed (Cunningham and Sedlak, 1994) that the ob-
served adsorption constants of chlorophenols on TiO2
in the dark differ from equivalent data obtained from
the kinetic analysis of their photocatalytic degradation
employing the same TiO2 particles under UV-
irradiation.
A detailed study of the intermediates formed during
the photocatalytic 4-CP degradation showed that not
only their number but also their concentration is
strongly influenced by the reaction pH (Theurich et al.,
1996). An increasing amount of hydroxylated products
(especially hydroxyhydroquinone, HHQ) was, for exam-
ple, observed for increasing pH values which Theurich
et al. attributed to higher stabilities of polyhydroxylated
benzenes and to a change in the adsorption properties of
these intermediates at alkaline pH values (Theurich
et al., 1996). HHQ was found to be strongly adsorbed
on TiO2 in acidic media while it was poorly adsorbed
at pH 11. It should be noted, however, that at even
higher pH-values a spontaneous oxidation of polyhy-
droxylated phenols and thus a decreasing yield of
HHQ will be the expected result.
Besides a fast TOC-degradation the suppression of
the formation of byproducts (especially when they are
highly toxic) is an important criterion for a good photo-
catalyst. Thus, it is interesting to note that while slightly
higher photonic efficiencies were obtained with TiO2
Degussa P25 markedly lower concentrations of byprod-
ucts were detected when TiO2 Sachtleben Hombikat UV
100 was the photocatalyst (Theurich et al., 1996). A sim-
ilar result was published by Mills et al. (Mills and Saw-
unyama, 1994) who compared Degussa P25 and several
other commercial TiO2-powders with laboratory made
TiO2. Significant smaller amounts of intermediates were
detected when the latter materials were used as the
photocatalyst.
An increasing concentration of the intermediates
benzoquinone (BQ) and hydroquinone (HQ) formed
during its photocatalytic treatment is apparently inhibit-
ing the degradation of the model pollutant 4-chlorophe-
nol (Theurich et al., 1996). Relatively low photonic
efficiencies have therefore been observed for this process
which have mechanistically been explained by the exist-
ence of a photocatalytic balance between HQ and BQ
(Theurich et al., 1996). A fast electron shuttle mecha-
nism between HQ and BQ has been proposed to describe
the underlying reactions. It has been shown that BQ can
act as a very effective electron scavenger over illuminated
TiO2 or ZnO being able to compete successfully with
molecular oxygen for the photogenerated electrons. On
the other hand, the oxidation of hydroquinone will
again result in the formation of benzoquinone thus
resulting in a shortcircuiting of the photocatalyst. Only
when benzoquinone molecules can be oxidized hydroxy-
benzoquinone (HBQ) will eventually be formed which is
a prerequisite for the subsequent ring cleavage leading to
the complete 4-CP degradation (cf. Fig. 2).
A high number of intermediates has been identified
during the photocatalytic 4-CP degradation indicating
a complex reaction mechanism. Fig. 2 illustrates the pro-
posed pathways of the photocatalytic/photolytic decom-
position of 4-CP (Theurich et al., 1996). The primary
oxidation products of 4-CP are 4-chlorocatechol (4-
CC), hydroquinone (HQ) and benzoquinone (BQ). 4-
CC is formed by the addition of a hydroxyl radical to
the ortho position of the hydroxyl group of 4-CP (reac-
tion a), followed by an elimination of a hydrogen atom
to recover the aromatic ring (reaction e). Subsequently,
4-CC is further degraded to hydroxyhydroquinone
(HHQ) by oxidation with another hydroxyl radical (or
a valence band hole) (reaction m) or by a reductive at-
tack of an electron (reaction k) followed by the addition
of molecular oxygen (reaction p) or by Cl-abstraction
(reaction l), both yielding HBQ (reaction w). HQ is
formed when the attack of the hydroxyl radical takes
places in the para position (reaction c) and a chlorine
atom (reaction f) or, if the generated intermediate reacts
first with an electron, a chloride ion is released. HQ can
either be oxidized to BQ (reaction j), resulting in the
photocatalytic short circuit described above, or be trans-
formed to HHQ (reaction n and r) by a second attack of
an hydroxyl radical in analogy to the transformation of
4-CP to 4-CC. HHQ can be further oxidized to hydroxy-
benzoquinone (HBQ) (reaction u) which probably leads
 D. Bahnemann / Solar Energy 77 (2004) 445–459 449

Fig. 2. Initial Steps in the Photocatalytic Degradation of 4-Chlorophenol (adopted from Theurich et al. (1996)).

to a similar photocatalytic short circuit as observed for
the system HQ/BQ, with the difference that HHQ and
HBQ are less stable, especially against direct photolysis,
than HQ and BQ. BQ can either be formed by the oxi-
dation of HQ (reaction j) or directly by the reaction of
molecular oxygen (reaction g) with the hydroxy phenyl
radical (HPR) generated by the abstraction of a Cl-
atom (reaction d), leading to a peroxyl radical which
can disproportionate to HQ and BQ (reactions h and
i) (Al-Ekabi et al., 1989). The formation of HPR as a
short-lived radical intermediate has been suggested in a
study by electron paramagnetic resonance spin trapping
detection during the direct photolysis of 4-CP (Lip-
czynska-Kochany et al., 1991).
BQ is either reduced to HQ (reaction j) or oxidized
(by the attack of a hydroxyl radical) to HBQ (reaction
o and s). Therefore, HHQ and HBQ are formed as sec-
ondary intermediates from each of the three primary
oxidation products (4-CC, HQ and BQ) (see Fig. 2).
Assuming that the ring cleavage is possible only for
450 D. Bahnemann / Solar Energy 77 (2004) 445–459
aromatic compounds which are hydroxylated in position
1 and 2, HHQ and its quinone derivative HBQ are the
last aromatic intermediates in this degradation scheme.
An attack of another hydroxyl radical leads to the ring
cleavage and therefore to non-cyclic intermediates. In
the case of HBQ these intermediates should be 3-hyd-
roxy derivatives of muconic aldehyde or muconic acid,
compounds which should undergo rapid oxidation to
the final mineralization products carbon dioxide and
water. Okamoto et al. detected during their investigation
of the photocatalytic oxidation of phenol many uniden-
tified peaks by HPLC (Okamoto et al., 1985). Because of
the very short retention times of these compounds under
the chosen experimental conditions they assumed that
these compounds are very polar products, i.e., aldehydes
and/or carboxylic acids.
3. Solar reactors and pilot plants
3.1. Reactor types
The artificial generation of photons required for the
detoxification of polluted water is the most important
source of costs during the operating of photocatalytic
wastewater treatment plants. This suggests to use the
sun as an economically and ecologically sensible light
source. With a typical UV-flux near the surface of the
earth of 20–30 W m
2 the sun puts 0.2–0.3 mol photons
m
2 h
1 in the 300–400 nm range at the process disposal
(Bahnemann, 1994). Principally these photons are suita-
ble for destroying water pollutants in photocatalytic
reactors.
Over the last 15 years several reactors for the solar
photocatalytic water treatment have been developed
and tested. In the following, the four most frequently
used reactor concepts will be presented.
Fig. 3. Flow Chart of the PTR installation located at the Plataforma Solar de Almeria in Spain (adopted from Sánchez (1993)).
3.1.1. Parabolic trough reactor (PTR)
A parabolic trough reactor (PTR) concentrates the
parallel (direct) rays of the photocatalytically active
ultra-violet part of the solar spectrum by a factor of
30–50 and can be characterized as a typical plug flow
reactor. This reactor type had been chosen for the first
solar detoxification loops constructed in the USA in
Albuquerque (by Sandia National Laboratories), in
California (by Lawrence Livermoore Laboratories)
and in Spain in Almeria (by the Plataforma Solar de
Almeria, PSA). Parallel to investigations under direc-
tion of NREL and Sandia National Laboratories in
New Mexico (USA) and in California (Alpert et al.,
1991) several research groups from different European
countries, funded by the European Community, have
tested the parabolic trough reactor installed at the
PSA (Spain) for solar wastewater treatment in the early
1990s.
Fig. 3 shows a schematic flow chart of the installation
at the PSA. The reactor located at the Plataforma Solar
de Almeria in Spain contained a total volume of 419 L
and consisted of six parabolic trough modules connected
in series (type ‘‘Helioman’’ from Mannesmann, initially
designed for the conversion of solar radiation into heat)
which were controlled by a two-axial (azimuth and ele-
vation) tracking system. The solar rays were concen-
trated by parabolic trough mirrors (total aperture
area: 192 m2) with an aluminized UV-reflective surface
and focused on borosilicate glass tubes (total length:
108 m, illuminated volume: 242 L) (Sánchez, 1993),
which were filled with the contaminated aqueous TiO2
suspensions moving at flow rates between 250 and
3500 L h
1. Due to losses caused by reflectivity, translu-
cence and system errors the yield of the UV-light pho-
tons reaching the contaminated suspensions is about
58% of the original light intensity entering the aperture
plane.
 D. Bahnemann / Solar Energy 77 (2004) 445–459 451

3.1.2. Thin film fixed bed reactor (TFFBR)

One of the first solar reactors not applying a light-
concentrating system and thus being able to utilize the
diffuse as well as the direct portion of the solar UV-A
irradiation for the photocatalytic process is the thin-
film-fixed-bed reactor (TFFBR) depicted in Fig. 4 (Hil-
gendorff et al., 1993b; Bockelmann et al., 1995; Goslich
et al., 1997). It should be noted that under AM (air
mass) 1.5 conditions the diffuse (Edif (300–
400 nm) = 24.3 W m
2) and direct (Edir (300–
400 nm) = 25.0 W m
2) portion of the solar radiation
reaching the surface of the earth are almost equal (Bird
et al., 1983). This means that a light concentrating sys-
tem, e.g., a parabolic trough reactor can in principal
only employ half of the solar radiation available in this
particular spectral region. The most important part of
the thin-film-fixed-bed reactor is a sloping plate (width
0.6 m · height 1.2 m) coated with the photocatalyst
(e.g., titanium dioxide Degussa P25) (Bockelmann,
1993) and rinsed with the polluted water in a very thin
film (100 lm). The flow rate is controlled by a cassette
peristaltic pump and can be varied between 1 and
6.5 L h
1 (Bockelmann et al., 1995).
Fig. 5 shows one of the first TFFBR test reactors
during its operation at the Plataforma Solar de Almeria
in Spain. The efficiency of its performance was tested
and compared with that of the PTR during several test
campaigns employing both, model compounds dissolved
in pure water and real wastewater samples from various
industrial companies. In every case it was observed that
the efficiency of the TFFBR was superior to that of the
PTR as expected based upon the above described insola-
tion conditions effective in both systems (Goslich et al.,
1997).
Fig. 4. Flow Chart of a TFFBR reactor (adopted from Goslich
et al. (1997)).
Fig. 5. Photograph of the TFFBR reactor installed at the
Plataforma Solar de Almeria in Spain (Photo: D. Bockelmann).
3.1.3. Compound parabolic collecting reactor (CPCR)
A compound parabolic collecting reactor (CPCR) is
a trough reactor without light concentrating properties.
It differs from a conventional parabolic trough reactor
by the shape of its reflecting mirrors. A reflector of a
parabolic trough reactor has a parabolic profile with
the reaction pipe in its focal line. Consequently only par-
allel light entering the parabolic trough can be focused
into the reaction pipe and a sun-tracking system is re-
quired. The shape of a CPCRÕs reflector usually consists
of two half circular profiles side by side, and a parabolic
continuation at both outer sides of the circles. The focal
line is located closely above the connection of the two
circles. This geometry enables light entering from almost
any direction to be reflected into the focal line of the
CPCR, i.e., most of the diffuse light entering the module
can also be employed for the photocatalytic reaction.
Due to this geometry a CPCR exhibits only a small con-
centration factor (<1.2). The CPCRs at the Plataforma
Solar de Almeria in Spain (PSA) manufactured by
Industrial Solar Technology Corporation, Denver,
Colorado (USA), have a concentration factor of 1.15,
i.e., this type of reactor has practically no light concen-
trating properties (Malato et al., 2002).
Moreover, a CPCR must not necessarily track the
sun due to its geometry. The azimuth should be adjusted
to the complementary angle a of the geographical alti-
tude and the pipes should be aligned south and from
top to bottom. A schematic view of one CPCR-module
at the PSA is given in Fig. 6. The angle of incidence a is
adjusted to 37 for all CPCR modules at the PSA.
452 D. Bahnemann / Solar Energy 77 (2004) 445–459

Fig. 6. Schematic View of one CPCR-module installed at the

Plataforma Solar de Almeria in Spain (adapted from Malato
et al. (2002)).
One CPCR-module of the PSA consists of eight par-
allel reflectors made from polished aluminum. This
material has very good reflection properties especially
in the UV-region of the solar spectrum. In the spectral
range of 295–387 nm its reflection efficiency is 83.2%.
A single CPCR reflector has a length of 1.22 m and a
width of 0.152 m, i.e., the effective reflecting area of
one module, consisting of eight reflectors adds up to
1.48 m2. The overall reflecting area of all six modules is
8.9 m2. A reaction pipe made of a transparent fluoropol-
ymer (Teflon) is fixed in the focal line of each CPCR
reflector, through which the suspension containing the
photocatalyst and the pollutant circulates. The transmis-
sivity of Teflon in the spectral range under consideration
is 76.8%, thus limiting the achievable reactor efficiency
to 63.9%. The absorber pipe is as long as the module
(1.22 m), has an inner diameter of 48 mm and conse-
quently an illuminated volume of 2.21 L per pipe. The
connectors between the absorber pipes consist of poly-
propylene; the overall illuminated volume of all 6 mod-
ules adds up to 106 L. The achievable flow rates in the
CPCR-module field at the PSA are between 2250 L h
1
and 8000 L h
1 and the minimum required volume is
248 L (Malato et al., 2002).
Fig. 7. Schematic View of a DSSR reactor showing the inner
structure of the transparent structured box made of PLEXI-
GLAS (adapted from (van Well et al., 1997)).
and the diffuse portion of the solar radiation in analogy
to the CPCR. After the degradation process the photo-
catalyst has to be removed from the suspension either by
filtering or by sedimentation for both reactors.
The DSSR consists of a modified double-skin sheet
(SDP 16/32) manufactured by the Röhm GmbH in
Darmstadt (Germany). One module has a length of
1400 mm, a width of 980 mm and contains 30 channels
each of which being 28.5 mm by 12 mm, and their total
inner volume adds up to 14.4 L in a single sheet. Fig. 8
shows a photograph of an operating DSSR pilot plant
consisting of three modules.
3.2. Solar photocatalytic treatment of real wastewater
Despite its obvious potential for the detoxification of
polluted water, there has been very little commercial or
industrial use of photocatalysis as a technology so far.
According to a review by Goswami (Goswami, 1997)
the published literature shows only two engineering scale

3.1.4. Double skin sheet reactor (DSSR)

A new kind of non-concentrating reactor is the dou-
ble skin sheet reactor (DSSR). It consists of a flat and
transparent structured box made of PLEXIGLAS
(Goswami, 1995; Benz et al., 1996; van Well et al.,
1997). The inner structure of the reactor is schematically
drawn in Fig. 7. The suspension containing the model
pollutant and the photocatalyst is pumped through these
channels. The comparison of the spectral irradiance of
the sun (AM 1.5) with the transmission spectrum of
the PLEXIGLAS used to manufacture the double skin
sheets evinces that the UV-A portion of the solar spec-
trum below 400 nm nicely matches with the onset of
the PLEXIGLAS-transmission (van Well et al., Fig. 8. Photograph of a DSSR pilot plant installed at the ISFH
1997). This type of reactor can utilize both the direct in Hannover/Germany (Photo: R. Goslich, ISFH).
 D. Bahnemann / Solar Energy 77 (2004) 445–459
demonstrations for groundwater treatment in the US
and one industrial wastewater treatment in Spain. Engi-
neering scale field experiments have been conducted by
the National Renewable Energy Laboratory (NREL)
at Lawrence Livermoore National Laboratory (LLNL)
treating ground water contaminated with trichloroethyl-
ene (TCE) (Mehos and Turchi, 1993). This field system
consisted of 158 m2 of parabolic trough reactors and
used Degussa P25 particles (0.1%) as photocatalyst in
a slurry flow configuration. With this relatively low tita-
nium dioxide content the TCE concentration was re-
duced from 200 ppb to less than 5 ppb.
Engineering scale demonstrations of the non-concen-
trating solar reactor technology were conducted at
Tyndall Air Force Base, FL (USA), in 1992 (Goswami
et al., 1993) treating groundwater contaminated with
fuel, oil, and lubricants which was leaking from under-
ground storage tanks. Crittenden and co-workers used
a tubular photoreactor equipped with a reflector and
packed with platinized titanium dioxide supported on
silica gel (Crittenden et al., 1996). This system was used
in a single-pass mode for the photocatalytic degradation
of benzene, toluene, ethylbenzene, and xylene (BTEX).
Without pretreatment fouling of the photocatalyst was
observed after less than one hour of operation. Ionic
species were found to be principally responsible for this
fouling of the photocatalyst and for the resulting inhibi-
tion of the BTEX degradation. A pretreatment unit (fil-
ter, oxygenation unit, ion exchanger) was developed to
remove ionic species and turbidity, and to add oxidants.
By using this pretreatment unit, the photocatalytic treat-
ment plant operated efficiently, and no decrease in the
photocatalytic activity was observed during 25 days of
continuous operation. Even on a rainy day, BTEX com-
pounds (>2 mg L
1 influent concentration) were de-
stroyed within 6.5 min of operation (Zhang et al., 1996).
Goswami and co-workers conducted field tests with a
slurry tubular reactor without reflector (SOLTOXTM
reactor). Five modules connected in series with a total
aperture area of 10 m2 were used. The photoreactor
array was set up facing south and tilted at an angle of
15 from the horizontal. The experiments were per-
formed in a batch mode by recycling the contaminated
groundwater between a tank and the reactor array (Wy-
ness et al., 1994a,b). While these field tests successfully
demonstrated the feasibility of destroying benzene, tolu-
ene, ethylbenzene, and xylenes (BTEX) in the ground-
water, the observed reaction rates were rather low.
However, in a laboratory test with city water spiked with
the same amounts of BTEX as those present in the
groundwater and using the same reactors under similar
sunlight conditions, Goswami et al. observed reaction
rates which were an order of magnitude higher (Wyness
et al., 1994a). These results suggested that a careful site
treatability study and the establishment of an appropri-
ate pretreatment will be very important prerequisites for
453
the successful field deployment of solar photocatalytic
processes (Goswami, 1997). A very similar pilot plant
has been used by Goswami and co-workers for the treat-
ment of a BTEX contaminated groundwater at Gains-
ville, FL (USA) (Goswami et al., 1997; Srinivasan
et al., 1997, 1998). The flat plate (DSSR) type photore-
actor used in this study was developed from a double
skinned acrylic panel. The size of the reactor was
1.20 m · 2.4 m with a skin thickness of 0.32 cm. The reac-
tor had 39 channels (3 cm · 1.3 cm). Headers were made
from 0.64 cm thick cylindrical tubes of 3.8 cm inner
diameter. The total solar treatment loop (six reactors
and a 1.9 m3 tank) was operated in a batch mode. The
titanium dioxide concentration was 0.1%. Experimental
results showed that 1.9 m3 of water contaminated with
1000 ppb of BTEX can be cleaned reaching a final con-
centration of 10 ppb in about three hours provided that
an average solar UV energy flux density of 28 W m
2 is
employed (Goswami et al., 1997).
Other real wastewaters that have been treated suc-
cessfully with solar photocatalysis include wastewater
from 5-Fluorouracil manufacturing (Anheden et al.,
1996) and biologically pretreated distillery effluents
(Zaidi et al., 1995). In the following, one example of a
plant for the after-treatment of biologically pretreated
wastewaters will be described in more detail since this
presents one particularly attractive feature of photocat-
alytic water treatment. On the one hand, these waters
usually do not contain high quantities of pollutants,
i.e., are ideal candidates for photocatalytic clean-up
processes. On the other hand, advanced oxidation proc-
esses have to be employed since the capacity of the bio-
logical processes has been exhausted beforehand.
Dillert et al. have treated biologically pretreated
industrial wastewaters from the factories Wolfsburg
(Germany) and Taubaté (Brazil) of the Volkswagen
AG in laboratory and bench-scale experiments (Dillert
et al., 1999a,b). The results of the experiments, which
have been performed using the DSSR, were so promis-
ing, that a pilot plant was installed in the Wolfsburg fac-
tory during the summer 1998 (Dillert et al., 1999c). The
flowsheet of a more recent version of this pilot plant
which was installed in 2000 is given in Fig. 9 while
Fig. 10 shows a photograph of this plant.
This pilot plant which is operated in a recycle batch
mode consists of twelve double-skin sheet photoreactors
(manufactured by Solacryl) with a total irradiated area
of 27.6 m2. These reactors are connected by a recircula-
tion pipe with a tank (1.1 m3 volume). The water
(500 L) coming from the biological treatment plant is
pumped into the tank and mixed with a catalyst slurry.
The solar photocatalytic treatment starts after the tank
has been filled. The suspension (total volume 1 m3)
recycles between the tank and the DSSRs for 8–11 h dur-
ing the daytime. The flowrate out of the tank can be var-
ied between 5000 and 8000 L h
1 resulting in a flowrate
454 D. Bahnemann / Solar Energy 77 (2004) 445–459
Fig. 9. Flow chart of a DSSR pilot plant installed in the Wolfsburg factory of the Volkswagen AG (adapted from Dillert et al.
(1999c)).
Fig. 10. Photograph of the DSSR pilot plant installed in the
Wolfsburg factory of the Volkswagen AG (Photo: J. Lohmann,
ISFH).
between 830 and 1300 L h
1 in the reactor sheets. After
the desired time of the solar photocatalytic treatment
the suspension is pumped out of the reactors into the
tank. The photocatalyst (5 g L
1 Degussa P 25) is al-
lowed to settle down during the night. After this
sedimentation period the supernatant liquid (500 L) is
pumped out of the tank and the treatment cycle can
be started again by filling the tank with a new batch of
wastewater. The COD and the TOC of the incoming
wastewater were determined to vary between 76.6 and
29.0 mg L
1 and 26.9 and 10.2 mg L
1, respectively. By
mixing the incoming wastewater with the catalyst slurry,
which remained in the pilot plant after the supernatant
water was pumped out, the COD and the TOC de-
creased. The initial values of the COD and the TOC
were determined to vary between 69.5 and 21.1 mg L
1
and 26.9 and 7.3 mg L
1, respectively. During the
solar-catalytic treatment a decrease of the concentration
of the organic pollutants, determined as COD and TOC,
was observed in most cases.
As a consequence of this oxidation process as well as
due to the increase of the reaction temperature, the con-
centration of molecular oxygen in the reflux pipe be-
tween the photoreactors and the tank as well as inside
the tank was decreasing. More than 50% of the organic
pollutants initially present in the mixed water inside the
pilot plant could thus be degraded within eight to eleven
hours of illumination. Naturally, the total mass of the
degraded contaminants was found to depend on the ini-
tial pollutant concentration, the time of illumination,
and, in particular, on the solar UV energy flux density.
Bekbölet et al. have applied the photocatalytic de-
gradation process to the destruction of biologically
 D. Bahnemann / Solar Energy 77 (2004) 445–459
pretreated landfill leachate effluent comparing Sachtle-
ben Hombikat UV100 with Degussa P25 as potential
photocatalyst (Bekbölet et al., 1996). These experiments
were performed in a batch reactor at pH 3. Both materi-
als, especially Hombikat UV100, showed a strong
adsorption of the organic pollutants; the overall TOC-re-
moval was almost the same for both catalysts ( 70%)
after 5 h of irradiation with a high-pressure xenon-lamp,
the BOD5 increase was insignificant indicating that the
water could still not be treated by conventional biologi-
cal methods (BOD5 is defined as the biological oxygen
demand during an incubation period of 5 days at 37 C
and thus represents a convenient measure for the biode-
gradability of a real wastewater). In addition to the
detoxification experiments, the pH dependent adsorptive
properties of the pollutants present in the landfill leac-
hate onto Hombikat UV100 were examined. In good
agreement with the pH dependence of the photocatalytic
degradation of the leachate, a strong adsorption was ob-
served between pH 3 and pH 8 with a maximum at pH 5,
while at pH 11 the adsorption was much lower (Weich-
grebe et al., 1993). Further degradation experiments were
performed with the thin-film fixed-bed reactor (TFFBR)
using Hombikat UV100 coated glass panes irradiated by
fluorescent tubes, resembling the UV(A)-portion of the
solar spectrum. The highest degradation rate was ob-
served between pH 3 and 7 with a maximum at pH 5,
while at pH 9 and 11 the degradation was rather slow.
It could be shown by the investigation by Bekbölet et
al. that photocatalysis is indeed a feasible method for
the successful cleaning-up of landfill effluent water
(Bekbölet et al., 1996). This is especially important as
Weichgrebe (Weichgrebe, 1994; Weichgrebe et al.,
1993) could demonstrate that by photocatalysis higher
rates can be obtained for the degradation of landfill leac-
hate than with other, more conventional wet oxidation
methods (e.g., ozone or hydrogen peroxide coupled with
UV-irradiation).
Fig. 11. Flow chart of a TFFBR pilot plant installed at the site of a textile factory in Tunisia (Menzel Temime).
455
In 1997 Freudenhammer et al. reported their results
from a pilot study using TFFBR reactors which was per-
formed in various Mediterranean countries and showed
that biologically pretreated textile wastewater can be
cleaned by solar photocatalysis with a maximum degra-
dation rate of 3 g COD h
1 m
2 (Freudenhammer et al.,
1997). Moreover, a comparison of reaction rates meas-
ured under artificial and under solar illumination,
respectively, which was also performed during this study
proved the necessity of outdoor experiments. It was con-
cluded that photocatalysis should be a suitable technol-
ogy as the final stage of purification of biologically or
physically pretreated wastewater in particular in sun-
rich areas. However, it was also pointed out that indus-
trial applications of solar wastewater treatment on a
larger scale is in strong demand for cheap photocatalysts
of higher activity to be competitive with treatment meth-
ods already established on the market (Freudenhammer
et al., 1997).
Based on these results, a pilot plant, financed by the
EC (INCO-DC IC18CT980267), has been built at the
site of a textile factory in Tunisia (Menzel Temime).
The TFFBR was chosen because previous studies
showed sufficient degradation rates with the selected tex-
tile wastewater in combination with its simple, low cost
construction and the low energy consumption (Freuden-
hammer et al., 1997). However, to integrate results ob-
tained with suspended catalysts showing in some cases
a higher efficiency than the fixed system, the possibility
of using suspended catalysts has also been considered.
The pilot plant the flow chart of which is shown in
Fig. 11 is characterised by the following main features:
Two TFFBR reactors with a width of 2.5 m and a
length of 10 m each, corresponding to a total illuminated
area of 50 m2, were built in concrete and are oriented to
the south with an inclination angle of 20 (Fig. 12). The
reactors can be operated in parallel or as a cascade flow
and in a continuous or a recycling mode, respectively,
456 D. Bahnemann / Solar Energy 77 (2004) 445–459
Fig. 12. Photograph of the TFFBR pilot plant installed at the site of a textile factory in Tunisia (Photo: D. Hufschmidt, University of
Hannover).
depending on the reaction kinetics. The pumps are de-
signed for a maximum flow of 3 m3 h
1. The plant is
connected to the sewer system of the textile factory for
continuous and long term experiments. Two sequencing
batch reactors (SBR) are connected to the TFFBR for
pre- and posttreatment with a total volume of 15 m3 each
and a membrane aeration system. An automatic injec-
tion system of chemical compounds can be used for
pH regulation or nutrient addition.
The plant is designed to operate the two reactors
independently. This allows a comparison of different cat-
alysts, fixation techniques, hydrodynamics and effluents/
model compounds under identical conditions. The plant
can be operated with suspended and fixed catalysts. A
sedimentation tank is connected to the reactors (via a
storage tank) to separate and recycle the catalyst. The
experiments can be carried out in continuous or recy-
cling mode. The following instrumentation and control
instruments are installed and connected to a computer
system to allow an automatic operation: pumps and
valves control, flow rates, tank levels, UV-A intensity,
dissolved oxygen, pH, conductivity. Experiments are
currently being conducted to test the operation condi-
tions of this newly installed facility.
References
Al-Ekabi, H., Draper, A.M., De Mayo, P., 1989a. Carbon
tetrabromide as an electron acceptor in heterogeneous
photocatalysis. Can. J. Chem. 67, 1061–1064.
Al-Ekabi, H., Serpone, N., Pelizzetti, E., Minero, C., Fox,
R.B., Draper, R.B., 1989b. Kinetic studies in heterogeneous
photocatalysis. 2. TiO2-mediated degradation of 4-chloro-
phenol alone and in a three-component mixture of 4-
chlorophenol, 2,4-dichlorophenol, and 2,4,5-trichlorophe-
nol in air-equilibrated aqueous media. Langmuir 5, 250–
255.
Alpert, D.J., Sprung, J.L., Pacheco, J.E., Prairie, M.R., Reilly,
T.A., Milne, T.A., Nimlos, M.R., 1991. Sandia National
LaboratoriesÕ work in solar detoxification of hazardous
wastes. Solar Energy Mater. 25, 594–607.
Al-Sayyed, G., DÕOliveira, J.C., Pichat, P., 1991. Semiconduc-
tor-sensitized photodegradation of 4-chlorophenol in water.
J. Photochem. Photobiol. A 58, 99–114.
Anheden, M., Goswami, D.Y., Svedberg, G., 1996. Photocat-
alytic treatment of wastewater from 5-fluorouracil manu-
facturing. J. Solar Energy Eng. 118, 2–8.
Anpo, M., Shima, T., Kubokawa, Y., 1985. ESR and photo-
luminescence evidence for the photocatalytic formation of
hydroxyl radicals on small TiO2 particles. Chem. Lett.,
1799–1802.
Augugliaro, V., Palmisano, L., Sclafani, A., Minero, C.,
Pelizzetti, E., 1988. Photocatalytic degradation of phenol
in aqueous titanium dioxide dispersions. Toxicol. Environ.
Chem. 16, 89–109.
Bahnemann, D.W., 1994. Solare Abwasserentgiftung. Nachr.
Chem. Tech. Lab. 42, 378–388.
Bahnemann, D.W., Henglein, A., Lilie, J., Spanhel, L., 1984a.
Flash photolysis observation of the absorption spectra of
trapped positive holes and electrons in colloidal TiO2. J.
Phys. Chem. 88, 703–711.
Bahnemann, D.W., Henglein, A., Spanhel, L., 1984b. Detec-
tion of the intermediates of colloidal TiO2-catalyzed
photoreactions. Faraday Discuss. Chem. Soc. 78, 151–
163.
Bahnemann, D.W., Fischer, C.H., Hoffmann, M.R., Hong,
J., Mönig, J., Kormann, C., 1987a. Mechanistic study of the
photocatalytic decomposition of organic compounds on
 D. Bahnemann / Solar Energy 77 (2004) 445–459
semiconductor particles. Prep. Am. Chem. Soc. Environ.
Chem. 27, 528–534.
Bahnemann, D.W., Mönig, J., Chapman, R., 1987b. Efficient
photocatalysis of the irreversible one-electron and two-
electron reduction of halothane on platinized colloidal
titanium dioxide in aqueous suspension. J. Phys. Chem. 91,
3782–3788.
Bahnemann, D.W., Cunningham, J., Fox, M.A., Pelizzetti, E.,
Serpone, N., 1994. Photocatalytic treatment of waters. In:
Helz, G., Zepp, R., Crosby, D.G. (Eds.), Aquatic and
Surface Chemistry. Lewis Publishers, Boca Raton, FL, pp.
261–316.
Bahnemann, D.W., Hilgendorff, M., Memming, R., 1997.
Charge carrier dynamics at TiO2 particles: reactivity of free
and trapped holes. J. Phys. Chem. B 101, 4265–4275.
Barbeni, M., Pramauro, P., Pelizzetti, E., Borgarello, E.,
Grätzel, M., Serpone, N., 1984. Photodegradation of 4-
chlorophenol catalyzed by titanium dioxide particles. Nouv.
J. Chim. 8, 547–550.
Bekbölet, M., Lindner, M., Weichgrebe, D., Bahnemann,
D.W., 1996. Photocatalytic detoxification with the thin-film
fixed-bed reactor (TFFBR): clean-up of highly polluted
landfill effluents using a novel TiO2-photocatalyst. Solar
Energy 56, 455–469.
Benz, V., Müller, M., Bahnemann, D.W., Weichgrebe, D.,
Brehm, M., 1996. Reaktoren für die photokatalytische
Abwasserreinigung mit Stegmehrfachplatten als Solarele-
mente. Deutsche Offenlegungsschrift DE 195 14 372 A1.
Bielski, B.H.J., Cabelli, D.E., Arudi, R.L., Ross, A.B., 1985.
Reactivity of HO2/O
2 radicals in aqueous solution. J. Phys.
Chem. Ref. Data 14, 1041–1100.
Bird, R.E., Hulstrom, R.L., Lewis, L.J., 1983. Terrestrial solar
spectral data set. Solar Energy 30, 563–573.
Blake, D.M., 1994. Bibliography of work on the photocatalytic
removal of hazardous compounds from water and air.
Report NREL/TP-430-6048, National Renewable Energy
Laboratory, USA.
Bockelmann, D., 1993. Verfahren zur Fixierung von Metallox-
idkatalysator auf einem Träger. Deutsche Offenlegungssch-
rift P 4 237 390.5.
Bockelmann, D., Weichgrebe, D., Goslich, R., Bahnemann,
D.W., 1995. Concentrating vs. non-concentrating reactors
for solar water detoxification. Solar Energy Mater. Solar
Cells 38, 441–451.
Butler, E.C., Davis, A.P., 1993. Photocatalytic oxidation in
aqueous titanium dioxide suspensions: the influence of
dissolved transition metals. J. Photochem. Photobiol. A 70,
273–283.
Buxton, G.V., Greenstock, C.L., Helman, W.P., Ross, A.B.,
1988. Critical-review of rate constants for reactions of
hydrated electrons, hydrogen-atoms and hydroxyl radicals
(.OH/.O
) in aqueous solution. J. Phys. Chem. Ref. Data 17,
513–886.
Carey, J.H., Lawrence, J., Tosine, H.M., 1976. Photodechlo-
rination of PCBÕs in the presence of titanium dioxide in
aqueous suspensions. Bull. Environ. Contam. Toxicol. 16,
697–701.
Choi, W., Hoffmann, M.R., 1995. Photoreductive mechanism
of CCl4 degradation on TiO2 particles and effects of
electron-donors. Environ. Sci. Technol. 29, 1646–
1654.
457
Colombo, D.P., Bowman, R.M., 1995. Femtosecond diffuse-
reflectance spectroscopy of TiO2 powders. J. Phys. Chem.
99, 11752–11756.
Colombo, D.P., Roussel, K.A., Sach, J., Skinner, D.E.,
Cavaleri, J.J., Bowman, R.M., 1995. Femtosecond study
of the intensity dependence of electron-hole dynamics in
TiO2 nanoclusters. Chem. Phys. Lett. 232, 207–214.
Crittenden, J.C., Zhang, Y., Hand, D.W., Perram, D.L.,
Marchand, E.G., 1996. Solar detoxification of fuel-contam-
inated groundwater using fixed-bed photocatalysts. Water
Environ. Res. 68, 270–278.
Cunningham, J., Sedlak, P., 1994. Interrelationships between
pollutant concentration, extent of adsorption, TiO2-sensi-
tized removal, photon flux and levels of electron or hole
trapping additives, 1. Aqueous monochlorphenol-TiO2
(P25) suspensions. J. Photochem. Photobiol. A 77, 255–263.
Dillert, R., Vollmer, S., Schober, M., Theurich, J., Bahnemann,
D., Arntz, D., Pahlmann, K., Wienefeld, J., Schmedding,
G., Sager, G., 1999a. Laboruntersuchungen zur photokat-
alytischen Behandlung eines biologisch behandelten Indus-
trieabwassers. Gwf Wasser Abwasser 140, 293–297.
Dillert, R., Vollmer, S., Schober, M., Theurich, J., Bahnemann,
D., Arntz, D., Pahlmann, K., Wienefeld, J., Schmedding,
G., Sager, G., 1999b. Photokatalytische Behandlung eines
Industrieabwassers im Stegdoppelplattenreaktor. Chem.
Ing. Technol. 71, 396–399.
Dillert, R., Vollmer, S., Gross, E., Schober, M., Bahnemann,
D., Wienefeld, D., Pahlmann, K., Schmedding, T., Arntz,
G., Sager, G., 1999c. Solar-catalytic treatment of an
industrial wastewater. Z. Phys. Chem. 213, 141–147.
Duonghong, D., Ramsden, J., Grätzel, M., 1982. Dynamics of
interfacial electron-transfer processes in colloidal semicon-
ductor systems. J. Am. Chem. Soc. 104, 2977–2985.
Filby, W.G., Mintas, M., Güsten, H., 1981. Heterogeneous
catalytic degradation of chlorofluoromethanes on zinc oxide
surfaces. Ber. Bunsenges. Phys. Chem. 85, 189–192.
Fox, M.A., Dulay, M.T., 1993. Heterogeneous photocatalysis.
Chem. Rev. 93, 341–357.
Freudenhammer, H., Bahnemann, D., Bousselmi, L., Geissen,
S.U., Ghrabi, S.U., Saleh, F., Si-Salah, A., Siemon, U.,
Vogelpohl, A., 1997. Detoxification and recycling of waste-
water by solar-catalytic treatment. Water Sci. Technol. 35,
149–156.
Goslich, R., Dillert, R., Bahnemann, D.W., 1997. Solar water
treatment: principles and reactors. Water Sci. Technol. 35,
137–148.
Goswami, D.Y., 1995. Engineering of solar photocatalytic
detoxification and disinfection. In: Böer, K.W. (Ed.),
Advances in Solar Energy, vol. 10. ASES, pp. 165–210.
Goswami, D.Y., 1997. A review of engineering developments of
aqueous phase solar photocatalytic detoxification and
disinfection processes. J. Solar Energy Eng. 119, 101–107.
Goswami, D.Y., Klausner, J.F., Mathur, G.D., Martin, A.,
Wyness, P., Schanze, K., Turchi, C., Marchand, E., 1993.
Solar photocatalytic treatment of groundwater at Tyndall
AFB: field test results. In: Proceedings of the 1993 Annual
Conference of the American Solar Energy Society, pp. 235–
239.
Goswami, D.Y., Sharma, S.K., Mathur, G.D., Jotshi, C.K.,
1997. Techno-economic analysis of solar detoxification
systems. J. Solar Energy Eng. 119, 108–113.
458 D. Bahnemann / Solar Energy 77 (2004) 445–459
Grätzel, M., 1989. Heterogeneous Photochemical Electron
Transfer. CRC Press, Boca Raton, FL.
Hecht, J., 1990. Sunlight gives toxic waste a tanning. New Sci.
14, 36.
Herrmann, J.-M., Guillard, C., Pichat, P., 1993. Heterogeneous
photocatalysis: an emerging technology for water treatment.
Catal. Today 17, 7–20.
Herrmann, J.M., Disdier, J., Pichat, P., 1986. Photoassisted
platinum deposition on TiO2 powder using various plati-
num complexes. J. Phys. Chem. 90, 6028–6034.
Herrmann, J.M., Disdier, J., Pichat, P., 1988. Photocatalytic
deposition of silver on powder titania: consequences for the
recovery of silver. J. Catal. 113, 72–81.
Hilgendorff, M., Bockelmann, D., Nogueira, R.F.P., Weich-
grebe, D., Jardim, W.F., Bahnemann, D., Goslich, R.,
1993a. Photocatalytic decomposition of aliphatic halogen-
ated hydrocarbons in water: laboratory studies and solar
applications. In: 6th International Symposium of Solar
Thermal Concentrating Technologies, Proceedings, vol. 2.
Editorial CIEMAT, Madrid, pp. 1167–1181.
Hilgendorff, M., Hilgendorff, M., Bahnemann, D.W., 1993b.
Reductive Photocatalytic Elimination of Tetrachlorometh-
ane on Platinized Titanium Dioxide in Aqueous Suspension.
In: Tomkiewicz, M., Haynes, R., Yoneyama, H., Hori, Y.
(Eds.), Environmental Aspects of Electrochemistry and
Photoelectrochemistry, Proceedings of Electrochemical
Society. The Electrochemical Society, Pennington, pp.
112–121.
Hilgendorff, M., Hilgendorff, M., Bahnemann, D.W., 1996.
Mechanisms of photocatalysis: the reductive degradation of
tetrachloromethane in aqueous titanium dioxide suspen-
sions. J. Adv. Oxid. Technol. 1, 35–43.
Hoffmann, M.R., Martin, S.T., Choi, W., Bahnemann, D.W.,
1995. Environmental applications of semiconductor photo-
catalysis. Chem. Rev. 95, 69–96.
Hofstadler, K., Bauer, R., Novalic, S., Heisler, G., 1994. New
reactor design for photocatalytic waste-water treatment
with TiO2 immobilized on fused-silica glass-fibers—photo-
mineralization of 4-chlorophenol. Environ. Sci. Technol. 28,
670–674.
Hsiao, C.Y., Lee, C.L., Ollis, D.F., 1983. Heterogeneous
photocatalysis: degradation of dilute solutions of dichloro-
methane (CH2Cl2), chloroform (CHCl3), and carbon tetra-
chloride (CCl4 ) with illuminated TiO2 photocatalyst. J.
Catal. 82, 418–423.
Jaeger, C.D., Bard, A., 1979. Spin trapping and electron spin
resonance detection of radical intermediates in the photo-
decomposition of water at TiO2 particulate systems. J. Phys.
Chem. 83, 3146–3152.
Kamat, P.V., 1993. Photochemistry on Nonreactive (Semicon-
ductor) Surfaces. Chem. Rev. 93, 267–300.
Kuhler, R.J., Santo, G.A., Caudill, T.T., Betterton, E.A.,
Arnold, R.G., 1993. Photoreductive dehalogenation of
bromoform with titanium-dioxide cobalt macrocycle hybrid
catalysts. Environ. Sci. Technol. 27, 2104–2111.
Legrini, O., Oliveros, E., Braun, A.M., 1993. Photochemi-
cal processes for water treatment. Chem. Rev. 93, 671–
698.
Lipczynska-Kochany, E., Kochany, J., Bolton, J.R., 1991.
Electron paramagnetic resonance spin traping detection of
short-lived radical intermediates in the direct photolysis of
4-chlorophenol in aerated aqueous solution. J. Photochem.
Photobiol. A 62, 229–240.
Mao, Y., Schöneich, C., Asmus, K.-D., 1991. Identification of
organic acids and other intermediates in oxidative degrada-
tion of chlorinated ethanes on TiO2 surfaces en route to
mineralization. A combined photocatalytic and radiation
chemical study. J. Phys. Chem. 95, 10080–10089.
Malato, S., Blanco, J., Vidal, A., Richter, C., 2002. Photocat-
alysis with solar energy at a pilot-plant scale: an overview.
Appl. Catal. B—Environ 37, 1–15.
Matthews, R.W., McEvoy, S.R., 1992. Photocatalytic degra-
dation of phenol in the presence of near—UV illuminated
titanium dioxide. J. Photochem. Photobiol. A 64, 231–246.
Mehos, M.S., Turchi, C.S., 1993. Field testing solar photocat-
alytic detoxification on TCE-contaminated groundwater.
Environ. Prog. 12, 194–199.
Memming, R., 1988. In: Photoelectrochemical Solar Energy
Conversion. In: Steckham, E. (Ed.), Topics in Current
Chemistry, vol. 143. Springer, Berlin, Heidelberg, New
York, pp. 79–111.
Mills, A., Sawunyama, P., 1994. Photocatalytic degradation of
4-chlorophenol mediated by TiO2—a comparative study of
laboratory made and commercial TiO2 samples. J. Photo-
chem. Photobiol. A 84, 305–309.
Moser, J., Punchihewa, S., Infelta, P.P., Grätzel, M., 1991.
Surface complexation of colloidal semiconductors strongly
enhances interfacial electron-transfer rates. Langmuir 7,
3012–3018.
Noda, H., Oikawa, K., Kamada, H., 1993. ESR spin-trapping
study of active oxygen radicals from photoexcited semicon-
ductors in aqueous H2O2 solutions. Chem. Soc. Jpn. 66,
455–458.
Ohtani, B., Nishimoto, S., 1993. Effect of surface adsorptions of
aliphatic alcohols and silver ion on the photocatalytic
activity of TiO2 suspended in aqueous solutions. J. Phys.
Chem. 97, 920–926.
Okamoto, K., Yamamoto, Y., Tanaka, H., Itaya, A., 1985.
Heterogeneous photocatalytic decomposition of phenol
over TiO2 powder. Bull. Chem. Soc. Jpn. 58, 2015–2022.
Ollis, D.F., 1985. Contaminant degradation in water: hetero-
geneous photocatalysis degrades halogenated hydrocarbon
contaminants. Environ. Sci. Technol 19, 480–484.
Ollis, D.F., Al-Ekabi, H., 1993. Photocatalytic Purification and
Treatment of Water and Air, vol. 3. Elsevier, Amsterdam.
Ollis, D.F., Hsiao, C.Y., Budiman, L., Lee, C.L., 1984.
Heterogeneous photoassisted catalysis: conversions of per-
chloroethylene, dichloroethane, chloroacetic acids, and
chlorobenzenes. J. Catal. 88, 89–96.
Oudjehani, K., Boule, P., 1992. Photoreactivity of 4-chloro-
phenol in aqueous solution. J. Photochem. Photobiol. A 68,
363–373.
Pelizzetti, E., Schiavello, M., 1991. Photochemical Conversion
and Storage of Solar Energy. Kluwer Academic Publishers,
Dordrecht.
Phillips, L.A., Raupp, G.B., 1992. Infrared spectroscopic
investigation of gas–solid heterogeneous photocatalytic
oxidation of trichloroethylene. J. Mol. Catal. 77, 297–311.
Prairie, M.R., Evans, L.R., Stange, B.M., Martinez, S.L., 1993.
An investigation of TiO2 photocatalysis for the treatment of
water contaminated with metals and organic chemicals.
Environ. Sci. Technol. 27, 1776–1782.
 D. Bahnemann / Solar Energy 77 (2004) 445–459
Rothenberger, G., Moser, J., Grätzel, M., Serpone, N.,
Sharma, D.K., 1985. Charge carrier trapping and recombi-
nation dynamics in small semiconductor particles. J. Am.
Chem. Soc. 107, 8054–8059.
Sabin, F., Türk, T., Vogler, A., 1992. Photo-oxidation of
organic compounds in the presence of titanium dioxide:
determination of the efficiency. J. Photochem. Photobiol. A
63, 99–106.
Sánchez, M., 1993. Activities performed at the Plataforma
Solar de Almeria (Spain) during the ECDGXII access to
large-scale scientific installations program. Final technical
report PSA TR05/93, CIEMAT/DLR, Madrid, Spain.
Schwartz, S.E., 1984. Gas- and aqueous-phase chemistry of
HO2 in liquid water clouds. J. Geophys. Res. 89, 589–598.
Sehili, T., Boule, P., Lemaire, J., 1989. Photocatalysed Trans-
formation of chloroaromatic derivatives on zinc oxide, III:
chlorophenols. J. Photochem. Photobiol. A 50, 117–127.
Serpone, N., Pelizzetti, E., 1989. Photocatalysis: Fundamentals
and Applications. Wiley, New York.
Serpone, N., Lawless, D., Khairutdinov, R., Pelizzetti, E., 1995.
Subnanosecond relaxation dynamics in TiO2 colloidal sols
(particle sizes R(P) = 1.0–13.4 nm)—relevance to heteroge-
nous photocatalysis. J. Phys. Chem. 99, 16655–16661.
Srinivasan, M., Klausner, J.F., Jotshi, C.K., Goswami, D.Y.,
1997. Solar photocatalytic treatment of BTEX contami-
nated groundwater. In: Proceedings of the ISES 1997 Solar
World Congress.
Srinivasan, M., Goswami, D.Y., Klausner, J.F., Jotshi, C.K.,
1998. Design, construction and testing of a solar photocat-
alytic facility for treatment of btex contaminated ground-
water. In: Solar Engineering 1998, Proceedings of the
ASME International Solar Energy Conference, Albuquer-
que, New Mexico, pp. 299–304.
Stafford, U., Gray, K.A., Kamat, P.V., Varma, A., 1993. An in
situ diffuse reflectance FTIR investigation of photocatalytic
degradation of 4-chlorophenol on a TiO2 powder surface.
Chem. Phys. Lett. 205, 55–61.
Stafford, U., Gray, K.A., Kamat, P.V., 1994. Radiolytic and
TiO2-assisted photocatalytic degradation of 4-chlorophe-
nol—a comparative study. J. Phys. Chem. 98, 6343–6351.
Terzian, R., Serpone, N., Draper, R.B., Fox, M.A., Pelizzetti,
E., 1991. Pulse radiolytic studies of the reaction of penta-
halophenols with OH radicals: formation of pentahalo-
phenoxyl, dihydroxypentahalocyclohexadienyl, and
semiquinone radicals. Langmuir 7, 3081–3089.
Theurich, J., Lindner, M., Bahnemann, D.W., 1996. Photocat-
alytic degradation of 4-chlorophenol in aerated aqueous
titanium dioxide suspensions: a kinetic and mechanistic
study. Langmuir 12, 6368–6376.
459
Tunesi, S., Anderson, M., 1991. Influence of chemisorption on
the photodecomposition of salicylic acid and related com-
pounds using suspended TiO2 ceramic membranes. J. Phys.
Chem. 95, 3399–3405.
Tunesi, S., Anderson, M.A., 1992. Surface effects in photo-
chemistry: an in situ cylindrical internal reflection-Fourier
transform infrared investigation of the effect of ring
substituents on chemisorption onto TiO2 ceramic mem-
branes. Langmuir 8, 487–495.
Turchi, C.S., Ollis, D.F., 1989. Mixed reactant photocatalysis:
intermediates and mutual rate inhibition. J. Catal. 119, 483–
496.
Turchi, C.S., Ollis, D.F., 1990. Photocatalytic degradation of
organic water contaminants: mechanisms involving hyd-
roxyl radical attack. J. Catal. 122, 178–192.
van Well, M., Dillert, R.H.G., Bahnemann, D.W., Benz, V.W.,
Müller, M.A., 1997. A novel non-concentrating reactor for
solar water detoxification. Trans. ASME, J. Solar Energy
Eng. 119, 114–119.
Vinodgopal, K., Stafford, U., Gray, K.A., Kamat, P.V., 1994.
Electrochemically assisted photocatalysis. 2. The role of
oxygen and reaction intermediates in thr degradation of 4-
chlorophenol on immobilized TiO2 particulate films. J.
Phys. Chem. 98, 6797–6803.
Weichgrebe, D., 1994. Beitrag zur Chemisch-Oxidativen
Abwasserbehandlung. Cuvillier Verlag, Göttingen,
Germany.
Weichgrebe, D., Vogelpohl, A., Bockelmann, D., Bahnemann,
D., 1993. Treatment of landfill leachates by photocatalytic
oxidation using TiO2: a comparison with alternative pho-
tochemical technologies. In: Ollis, D.F., Al-Ekabi, H.
(Eds.), Photocatalytic Purification and Treatment of Water
and Air. Elsevier, Amsterdam, pp. 579–584.
Wyness, P., Klausner, J.F., Goswami, D.Y., Schanze, K.S.,
1994a. Performance of nonconcentrating solar photocata-
lytic oxidation reactors. 1. Flat-plate configuration. J. Solar
Energy Eng. 116, 2–7.
Wyness, P., Klausner, J.F., Goswami, D.Y., Schanze, K.S.,
1994b. Performance of nonconcentrating solar photocata-
lytic oxidation reactors. 2. Shallow pond configuration. J.
Solar Energy Eng. 116, 8–13.
Zaidi, A.H., Goswami, D.Y., Wilkie, A.C., 1995. Solar photo-
catalytic post-treatment of anaerobically digested distillery
effluent. In: SolarÕ95, Proceedings of the American Solar
Energy Society Annual Conference, Minneapolis, Minne-
sota, pp. 51–56.
Zhang, Y., Crittenden, J.C., Hand, D.W., Perram, D.L., 1996.
Destruction of organic compounds in water using supported
photocatalysts. J. Solar Energy Eng. 118, 123–129.