View Article Online / Journal Homepage / Table of Contents for this issue

FEATURE ARTICLE www.rsc.org/materials | Journal of Materials Chemistry

Wastewater treatment with heterogeneous Fenton-type catalysts based on

porous materials
Martin Hartmann,*a Simon Kullmanna and Harald Kellerb
Received 2nd March 2010, Accepted 7th May 2010
DOI: 10.1039/c0jm00577k

The Fenton reagent which was discovered by H. J. H. Fenton over 100 years ago is still of special
interest for wastewater treatment in the present time. The main objective of heterogeneous Fenton
catalysts is the combination of high activity, low iron leaching and the possibility of employing the
system at pH environments near neutrality. This review discusses the use of iron-containing pillared
Published on 30 July 2010 on http://pubs.rsc.org | doi:10.1039/C0JM00577K

clays as well as micro- and mesoporous (alumino-) silicate catalysts prepared by ion-exchange,
impregnation or direct synthesis for the degradation of organic compounds like phenol and organic
Downloaded by University of Michigan Library on 21 February 2013

dyes including Orange II. Moreover, the treatment of more complex waste streams is reviewed. After
a general introduction into the chemistry of the Fenton and the photo-Fenton reaction as well as the
catalysts employed, specific examples are discussed with respect to the active species present, the
stability of the catalyst and the influence of reaction conditions such as pH, temperature and
the presence of impurities.

Introduction processes. Their main aim is the preparation of the effluents for
subsequent biological or physical cleaning steps.
The increasing worldwide demand for clean water requires the Advanced oxidation processes (AOP) are particularly useful
development of concepts for an outright cleaning of wastewaters. for cleaning biologically toxic or non-degradable materials such
In particular in the so-called developed countries, it is one of the as aromatics, pesticides, petroleum constituents, and volatile
main characteristics of industry to produce a large amount of organic compounds in waste water.1 The contaminant materials
waste in addition to the desired product. Part of the industrial are converted to a large extent into stable inorganic compounds
waste is not biodegradable and requires efficient removal such as water, carbon dioxide and salts, i.e. they undergo
mineralization. A goal of the waste water purification by means
of AOP procedures is the reduction of the chemical contaminants
a
Universit€
at Erlangen-N€
urnberg, Erlangen Catalysis Resource Center, and the toxicity to such an extent that the cleaned waste water
Egerlandstr. 3, D-91058 Erlangen, Germany. E-mail: martin.hartmann@ may be reintroduced into receiving streams or, at least, into
ecrc.uni-erlangen.de
b
a conventional sewage treatment. Ozone, hydrogen peroxide,
BASF SE, D-67056 Ludwigshafen, Germany

Martin Hartmann obtained his Simon Kullmann was born in

Ph.D. in chemistry at the Univer- Frankenthal (Germany) in
sity of Dortmund in 1993. Before 1980. He studied chemistry at
joining the TU Kaiserslautern in the Technical University of
1999 as assistant professor, he Kaiserslautern where he received
worked at the University of his diploma degree in 2006 under
Houston and the University of the supervision of Professor Dr.
Stuttgart. In 2006 he was appoin- Martin Hartmann. In the same
ted Professor of Advanced Mate- year, he started his PhD work in
rials Science at the University of the group of Prof. Hartmann at
Augsburg. In 2009 he became the Institute of Physics, Univer-
Professor of Catalysis at the Simon Kullmann sity of Augsburg and moved with
University of Erlangen-Nurem- him to the Erlangen Catalysis
Martin Hartmann berg and director of the Erlangen Resource Center at the Friedrich
Catalysis Resource Center. Prof. Alexander University in Erlangen Nuremberg in 2009. His current
Hartmann is co-author of more research interests focus on the assembly, stabilization and char-
than 150 journal publications in the area of micro-and mesoporous acterization of bio functionalized surfaces.
materials for catalytic and separation applications. Moreover, he is
interested in the use of solid state NMR and ESR spectroscopy for
the characterization of porous solids.

9002 | J. Mater. Chem., 2010, 20, 9002–9017 This journal is ª The Royal Society of Chemistry 2010
 View Article Online

oxygen, and air are the main oxidants tested and applied to date However, there are also concerns about the influence of Fenton’s
in APOs. These procedures may also be combined with UV reagent on subsurface microorganisms. Miller et al. observed
irradiation and specific catalysts may be used to generate that the direct application of the reagent to microorganisms
hydroxyl radicals. The hydroxyl radical is a very powerful resulted in their death.8 Contact between soils pretreated by
oxidant and can be generated amongst others by the so-called Fenton’s Reagent and microorganisms, however, resulted in
Fenton reaction, that is the reaction of Fe2+ with H2O2. The microbial growth and decreased diversity. A damage to cell
mineralization of the toxic substances is due to the electrophilic components due to radical attack was noticed, too.
attack of the hydroxyl radical (HOc) and simply yields CO2, H2O In the present feature article, the use of heterogeneous Fenton-
and inorganic salts. Thereby the extremely high oxidation type catalysts based on porous supports such as zeolites, clays
potential of the hydroxyl radical (HOc) of 2.8 V vs. NHE is and mesoporous silicas is reviewed. Main emphasis is placed on
utilized. The use of Fenton’s reagent (Fen+/H2O2) is a well known understanding the factors determining the reaction rate, the
example for an advanced oxidation process that requires the use stability of the catalysts and the degradation pathways. Results
of a catalyst. Compared with physically/mechanically based concerning characterization of the active sites are also discussed
processes (e.g. flocculation, precipitation, adsorption on acti- with respect to the activity and selectivity of the heterogeneous
Published on 30 July 2010 on http://pubs.rsc.org | doi:10.1039/C0JM00577K

vated carbon, filtration) the advantage of the Fenton method is Fenton catalysts containing iron.
Downloaded by University of Michigan Library on 21 February 2013

the potentially complete mineralization of the organic

substances, whereas the toxicants are merely separated and
concentrated by physical/mechanical methods. Fenton chemistry
Already in 1968, Bishop et al. studied the catalytic oxidation of
General. At the end of the 19th century, H. J. H. Fenton
refractory organics by H2O2 in municipal waste waters. In their
reported that ferrous ions strongly promote the oxidation of
experimental study, the authors attempted to determine the
tartaric acid by hydrogen peroxide.9 Further research has shown
technical feasibility of using hydroxyl radicals for the decom-
that the combination of H2O2 and a ferrous salt, ‘‘Fenton’s
position of a broad spectrum of organic residues in wastewater.2
reagent’’, is an effective oxidant for a wide range of organic
Fenton and photo-Fenton processes are known to be effective
substrates. In 1934, Haber and Weiss proposed that the active
and inexpensive methods for water and wastewater purification
species is the hydroxyl radical.
from toxic or non-biodegradable organic pollutants.2–4
The Fenton reaction can be written in its general form:
Hydrogen peroxide as the oxidant does not produce toxic by-
products and oxidizes organic substances under mild conditions Mn+ + H2O2 / M(n1)+ + OH + HO$ (M ¼ Fe, Cu, Mn) (1)
in the presence of iron species. In this context, Tarr et al. high-
lighted the effect of natural water matrices on hydroxyl radical Moreover, they also proposed a series of chain reactions
formation and scavenging in a homogeneous system.5 Another involved in the oxidation using Fenton’s reagent.10 Their redox
major field of application of the Fenton reagent is the remedia- chain formulation was put in its most recent form by Barb et al.11
tion of contaminated soils. The degradation of diesel fuel in and was up till now only slightly modified12,13 or extended.14
grounds by six different iron compounds and by naturally- According to the reactions summarized in Table 1, the combi-
occurring minerals was investigated by Watts et al.6 They ach- nation of ferrous irons and hydrogen peroxide induces a series of
ieved conversions >99% in the hydrogen peroxide supported chain reactions initiated by the degradation of peroxide to
oxidation of diesel using iron(III) perchlorate and nitrate within
one hour.
Table 1 Listing of possible Fe(III)-initiated chain reactions
The efficiency of Fenton’s reagent in the treatment of petro-
leum-contaminated sand, which was polluted in an oil spill in Rio Rate constant at
de Janeiro was examined by Millioli et al. It was reported that the Reaction pH 3 (mol1 s1) Ref.
oil removal is increased when the reaction is carried out with
higher H2O2 concentrations and lower Fe2+ concentrations.7 T1.1 Fe(III) + H2O2 / Fe(II) + HO2c/ 2.0  103 15
O2$  + H+
T1.2 Fe(II) + H2O2 / Fe(III) + cOH + 63 11
OH
Dr. Harald Keller received his T1.3 H2O2 + cOH / HO2c/O2$ + H2O 3.3  107 15
T1.4 Fe(III) + HO2c/O2$  / Fe(II) + O2 7.8  105 15
PhD in 1988 from the Institute of + H+
Organic Chemistry at the Univer- T1.5 Fe(II) + cOH / Fe(III) + OH 3.2  108 15
sity of Kaiserslautern, Germany. T1.6 Fe(II) + HO2c/O2$  / Fe(III) + 1.3  106 15
H2O
From 1988 – 1990 he joined the
T1.7 HO2c/O2$  + HO2c/O2$  / H2O2 2.3  106 15
group of Prof. Jean-Marie Lehn at + O2
the University Louis Pasteur in T1.8 cOH + HO2c/O2$  / H2O + O2 7.1  109 15
Strasbourg, France. Since 1990 he T1.9 cOH + cOH / H2O2 5.2  109 15
T1.10 cOH + R–H / H2O + Rc > 108–109 16
has been working for BASF T1.11 cOH + Fe(II) / HO–Fe(III) n.a. 16
research. In 2009 he was appointed T1.12 Rc + Fe(III) / Fe(II) + R+; R+ + n.a. 16
senior scientist in the biopolymer H2O / R–OH + H+
area. T1.13 Rc + Fe(II) / Fe(III) + R; R + n.a. 16
H2O / R–H + OH
T1.14 Rc + Rc / R-R n.a. 16
Harald Keller

This journal is ª The Royal Society of Chemistry 2010 J. Mater. Chem., 2010, 20, 9002–9017 | 9003
 View Article Online

a hydroxyl radical and a hydroxide ion (T1.2), the original

Fenton reaction. In the absence of light, the chain initiation
reaction is the reduction of Fe(III) by hydrogen peroxide (T1.1),
producing two chain-propagating species Fe(II) and HO2c/O2c
(superoxide radical and its conjugate acid). Fe(II) is quickly
oxidized by another H2O2 molecule, and the chain then propa- Fig. 1 Reaction cycle for the photo-Fenton process.
gates by reactions T1.3 and T1.4. Thus, reactions T1.2, T1.3, and
T1.4 form a cycle that sustains itself by regenerating Fe(II).15 The and a ferric ion, and the cycle continues. Thus, two HOc radicals
reactions T1.5, T1.6 and T1.8 will terminate the chain reaction. are generated per photocatalyzed cycle of the ferric/ferrous
In the more general case, when the organic substrate is system.
a generic chain-saturated aliphatic hydrocarbon R–H, its fate is Heterogeneous Fenton systems
mainly governed by the nature of Rc produced by the fast
H-abstraction (T1.10), which competes with step T1.11. In recent years, homogeneous Fenton systems have been widely
 If Rc is a tertiary radical, or adjacent to any group capable of studied and are believed to be promising for wastewater
Published on 30 July 2010 on http://pubs.rsc.org | doi:10.1039/C0JM00577K

stabilizing a positive charge, it allows the starting oxidation state treatment. However, it has to be pointed out that the homoge-
Downloaded by University of Michigan Library on 21 February 2013

of iron to be re-established by an electron-transfer reaction neous Fenton process has major disadvantages; it requires up to
(T1.12), which is followed by the alkyl hydroxylation. 50–80 ppm of iron ions in solution which is well above the
 If Rc is a carbonyl conjugated radical (or adjacent to any European Union (EU) directive that allows only 2 ppm of iron in
group capable of stabilizing a carbanion), the process is stopped treated water to be released directly into the environment.
by a one-electron reduction (T1.13). Moreover, it is difficult to remove the sludge contaminated with
 Finally, if Rc is a primary or secondary alkyl radical, it Fe ions after the treatment. Another disadvantage of the
terminates the chain by the dimerization reaction T1.14.16 homogenous Fenton reaction is that the treatment has to be
The efficiency of the Fenton oxidation process depends, performed at acidic pH preferentially between 2.5 and 3.5. A
among other factors such as temperature, pH, hydrogen further significant drawback of the use of soluble iron salts is the
peroxide and catalyst concentration, on the reduction of Fe3+ to complexation of the iron ions by compounds present in real feeds
Fe2+. Thus, the presence of reaction products that are able to such as EDTA or possible reaction products such as oxalic acid.
reduce Fe3+ and regenerate the catalyst is crucial. However, This will result in reduction of the concentration of free iron ions
certain intermediates form iron complexes and thereby remove in solution (accompanied by a reduced reaction rate) and the
Fe3+ from the Fe2+/Fe3+ cycle, thus delaying or inhibiting the promotion of an unwanted side reaction, viz. the decomposition
oxidation process. of H2O2 to water and oxygen. In order to limit both effects, an
The maximum catalytic activity is typically found at a pH iron salt and H2O2 are added continuously during the course of
around 3. The oxidation rates in the Fe(III)/H2O2 system were the reaction. Consequently, the development of supported Fen-
observed to decrease dramatically at pH values larger than 3.14 ton catalysts is recently gaining importance. The use of solid
Fe3+ starts to precipitate as Fe(OH)3 and H2O2 is preferentially catalysts in principle eliminates or reduces these drawbacks as
decomposed into O2 and H2O. Besides, the formation of Fe(II) outlined by Perathoner and Centi.23 A further motivation for the
complexes at high pH leads to a decrease in Fe2+ concentration. use of solid Fenton catalyst is the potential stabilization of the
On the other hand, Fe2+ regeneration by reaction of Fe3+ with less stable reduced Fe(II) or Cu(I) oxidation states by a suitable
H2O2 (T1.1, T1.4) is inhibited at more acidic pH values. The (porous) support. The stabilization of Fe2+ and Cu+ is well
reaction temperature is another crucial parameter in the Fenton documented in the case of zeolites.24,25
process. In principle, increasing the reaction temperature should The ideal solid Fenton catalyst should not be prone to iron (or
enhance the kinetics of the process, but it also favors the copper) leaching, but most of the heterogeneous Fenton catalysts
decomposition of hydrogen peroxide, whose rate increases by developed up to now still exhibit this problem. Thus iron ions
a factor of 2 each 10  C in the range 20 to 100  C.17 It is still an significantly leach from the catalyst during the reaction, which
ongoing discussion, whether the free hydroxyl radicals18 or some causes a loss in activity with time and generates secondary metal
FeIV ¼ O moieties19 are the reactive species. However, in a recent ion pollution. It is anticipated that leaching below 1% of active
paper spectroscopic measurements ruled out [(H2O)5FeIV ¼ O]2+ Fe(II) species may be sufficient to produce part of the activity.26
as a reactive intermediate in the Fenton reaction in acidic and Therefore, it is well accepted that the most important issue in the
neutral aqueous solutions.20 heterogeneous Fenton process is the development of a heteroge-
neous catalyst with high catalytic activity and long-term stability
Photo-Fenton reaction. It is well known that UV (l > 250 nm) at a reasonable cost. In principle, the heterogeneous catalyst for
light irradiation improves the effectiveness of the homogeneous the photo-Fenton process may contain Fe ions, Fe clusters, or Fe
Fenton reaction due to regeneration of the consumed Fe2+ ions oxides as active sites, while for the Fenton process iron ions are
(i.e. reduction of Fe3+ to Fe2+) and direct H2O2 photolysis.21 considered the prime active sites. The use of other metals, in
During the reaction, the amount of ferrous ions decreases and particular copper,27–30 is also reported, but the discussion of this
the oxidized ferric ions are accumulated in the system. After all work is beyond the scope of this paper.
the Fe2+ ions are consumed, the reaction stops. Under UV
Porous iron-containing catalysts employed in Fenton processes
radiation, photochemical regeneration of ferrous ions by photo-
reduction of ferric ions occurs (Fig. 1).22 The newly generated Pillared clays. Metal oxide pillared clays are a class of
ferrous ions react with H2O2 generating a second HOc radical microporous solid acids derived from the intercalation of

9004 | J. Mater. Chem., 2010, 20, 9002–9017 This journal is ª The Royal Society of Chemistry 2010
 View Article Online

polycations into the galleries of layered silicates with mica-like or reactions involving large molecules that are not capable of
2 : 1 sheet structures. The most commonly used pillaring agent is entering the channels of microporous zeolites. MCM-41 (Mobil
the Keggin-like ion [Al13O4(OH)24(H2O)12]7+, which is intro- Composition of Matter No. 41) and SBA-15 (Santa Barbara
duced in layered materials such as smectite clays, viz. montmo- Amorphous material No. 15) are two representatives of this class
rillonite, beidelite or saponite.31 The elementary sheet of these of materials which posses an ordered hexagonal arrangement
clays contains a central layer of edge-shared aluminium or (space group p6mn) of mesopores with a narrow pore size
magnesium octahedra sandwiched between two layers of corner- distribution. They exhibit very high specific surface areas and
shared silicon tetrahedra. In the case of Al3+, two out of three pore volumes in the order of 1000 m2 g1 and 1.0 cm3 g1.
octahedral positions are occupied, while in the case of Mg2+ all
octahedral positions are occupied. Smectites have a low layer Preparation of iron-containing catalysts. The method of iron
charge (0.25–0.60) and swell easily in water. As a consequence, introduction into a porous support is of tremendous importance
the interlamellar surface is easily available for ion-exchange and for the properties of the resulting catalysts. Iron-containing
pillaring is feasible. Calcination of the air-dried Al13-clay reac- micro- and mesoporous molecular sieves show redox properties,
tion product converts the material to a stable metal oxide pillared which render them interesting for potential applications in
Published on 30 July 2010 on http://pubs.rsc.org | doi:10.1039/C0JM00577K

derivative, which is now microporous and exhibits a surface area selected oxidation processes.36–38 It is generally believed that the
Downloaded by University of Michigan Library on 21 February 2013

in the range between 200 and 400 m2 g1. Pillared clays usually reducibility of active iron species plays a key role for the oxida-
exhibit a broader pore size distribution than zeolites, which is tion process. The reducibility is influenced by various factors,
typically centered around 0.9 to 1 nm. Introduction of iron into which can be fine-tuned during the catalyst preparation and
pillared clays is achieved by addition of an iron salt into the activation before the catalytic reaction. In principle, iron can be
pillaring solution and subsequent calcination of the resulting introduced by ion-exchange in the liquid or the solid state,
material. impregnation or by isomorphous substitution. Generally, three
different kinds of iron species in the molecular sieves have to be
Zeolites and mesoporous silica materials. Iron-containing considered, viz. extra-framework iron on cation positions, iso-
zeolites and mesoporous materials are of particular interest in morphously substituted iron in the zeolite framework and iron
this area because of their unique catalytic properties in various oxide species located in the channels. The preferred techniques to
reactions, e.g. reduction of nitrous oxides,32,33 oxidation of introduce iron species into micro- and mesoporous molecular
benzene to phenol,34 and the selective oxidation of methane.35 sieves will be described below.
Zeolites are three-dimensional, crystalline hydrated aluminosili-
cate materials having cage- or channel-like structures with Ion-exchange (liquid vs. solid state). In zeolites and mesoporous
internal surface areas up to several hundred m2 g1 and large materials, the tetrahedrally coordinated Al atoms introduce
cation exchange capacities. At present more than 190 unique a negative charge into the framework, which has to be balanced
zeolite frameworks have been identified, and at least 41 naturally by protons or other cations. The ion exchange with Fe3+ is
occurring zeolite frameworks are known. Both natural and typically performed by stirring the parent material in an aqueous
synthetic zeolites are used in industry as adsorbents, catalysts, solution of the iron salt at room temperature or slightly above for
soil modifiers and ion-exchangers. The most important zeolites in several hours. When the zeolite crystals are immersed in the
this context are ZSM-5 (MFI, framework code of the Interna- aqueous electrolyte, ion exchange between the solid zeolite phase
tional Zeolite Association (IZA)), Y (FAU) and Beta (BEA). and the solution proceeds. For example, starting from NaY,
ZSM-5 contains an intersecting channel system with straight 10- three sodium ions are exchanged for one iron cation, which is
ring channels (0.53  0.56 nm) and sinusoidal 10-ring channels three times positively charged (Fig. 2).
(0.55  0.51 nm). FAU has a three-dimensional channels system This method is the most simple to be applied for industrial
with a large so-called supercage (d ¼ 1.3 nm), which is accessible practice; however, several challenges still remain as outlined by
via 12-ring windows having a diameter of 0.74 nm. Zeolite Beta Melian-Cabrera et al.:39 (1) reducing the steps during the material
also possesses a three-dimensional pore system with straight and preparation, (2) minimization of waste by replacing inorganic
tortuous 12-ring pore paths with a free aperture of 0.73  0.6 nm salts by e.g. citrates, (3) utilization of the full exchange capacity
and 0.56  0.56 nm, respectively. and (4) understanding the Fe(III) hydrolysis and obtaining the
The large pore diameter (2 to 30 nm) of mesoporous silica optimum pH conditions to perform the exchange. Moreover,
materials makes them potentially attractive as catalysts for a lack of reproducibility of exchange methods in solution has

Fig. 2 Exchange of Fe3+ ions for Na+ in aluminosilicates.

This journal is ª The Royal Society of Chemistry 2010 J. Mater. Chem., 2010, 20, 9002–9017 | 9005
 View Article Online

frequently been reported.40 In the course of their studies, the by the also positively charged silica species in highly acidic
authors developed a novel concept which allows the simulta- media. Vinu et al. attempted the incorporation of iron into the
neous detemplation of the zeolite with simultaneous introduction SBA-15 wall by simple adjustment of the pH of the synthesis gel
of iron based on the Fenton reaction.41 to a value above the isoelectric point of silica (pH  2). Under
Solid state ion exchange has some advantages over the these conditions the negatively charged silica species can easily
conventional liquid ion exchange, viz. (i) handling of large interact with the positively charged iron hydroxo complexes
volumes of salt solutions is not required; (ii) the problem of waste [Fe(OH)2]+, resulting in incorporation of larger quantities of Fe
salt solution can be avoided and (iii) metal cations might be into the SBA-15 silica matrix.54 In addition to pH adjustment,
introduced into narrow pore cavities in cases where the large the addition of NH4F to the reaction mixture promotes precip-
hydration shell of the cations might prohibit ion exchange in itation of the product. The obtained improvement of the struc-
aqueous solution. Depending on the starting material, solid state tural ordering of the Fe-SBA-15 materials is attributed to the
iron ion exchange may be expressed by the two equations given addition of the Fe(NO3)3$9H2O precursor. Generally, the addi-
below: tion of inorganic salts is helpful in order to obtain highly ordered
mesoporous materials, which is due to the specific interactions
Published on 30 July 2010 on http://pubs.rsc.org | doi:10.1039/C0JM00577K

a FeCln + b Na+Z # x Fen+Z + (b-nx)Na+Z + (a-x)FeCln + nx between non-ionic surfactant and metal ions.55 The preparation
Downloaded by University of Michigan Library on 21 February 2013

NaCl (2) of these materials relies on the control of tetramethoxysilane

(TMOS) hydrolysis and silica polymerization in the presence of
a FeCln + b H+Z # x Fen+Z + (b-nx)H+Z + (a-x)FeCln + nx fluoride ions and the amphiphilic block copolymer, which serves
HCl (3) as structure-directing agent.56,57

In eqn (3) it is assumed that the reaction takes place between Impregnation with iron. There are two principally different
iron chloride (or sulfate) and a sodium zeolite for the sake of methods for the impregnation of zeolites or mesoporous mate-
simplicity. In this case the equilibrium is obtained in a similar rials with iron species. The first one is called wet impregnation,
manner to liquid ion exchange. If the hydrogen form of the which means the parent material is suspended in an aqueous iron
zeolite is used (eqn (3)), the equilibrium may be shifted to the far salt solution and heated up for some time to evaporate the water.
right resulting in 100% exchange if the formed HCl is removed in
The second method is the so called ‘‘incipient wetness impreg-
a stream of inert gas or vacuum. The preparation of Fe(II)-ZSM- nation’’ in which a volume of iron salt solution equal to the total
5 through solid state ion exchange has been reported.42 In order pore volume of the material is used. In both cases the impreg-
to prevent formation of Fe(III), the ion exchange can be per- nated samples are dried and calcined prior to further modifica-
formed under anaerobic conditions. tion or their use.58 It has been reported that part of Fe3+ ions
grafted onto the silica surface of SBA-15 is strongly stabilized in
Incorporation of iron during the hydrothermal synthesis. The their respective oxidation state and can only be reduced by
insertion of heteroatoms such as iron into the framework of hydrogen at a temperature exceeding 800  C.59
micro- or mesoporous materials allows the fine tuning of the
properties of these materials. In this case, an iron salt is added
Use of heterogeneous Fenton catalysts based on porous supports
directly to the synthesis mixture and replaces framework atoms
in wastewater treatment
such as aluminium or silicon. Based on the ionic radii, one can
predict that the incorporation of Fe3+ (r ¼ 0.063 nm) is favored Fenton chemistry has been predominately applied to the degra-
over Fe2+ (r ¼ 0.077 nm). It has been reported that Fe3+ located in dation of model compounds such as phenol, Orange II,
the zeolite framework is stable towards reduction (easy reduc- carboxylic acids and others. The effective removal of aromatic
ibility is believed to be important in many catalytic oxidation pollutants present in wastewater streams generated by industrial
processes).43 However, iron tends to be partially removed from processes has become a challenging task due to more restrictive
the framework by thermal treatment especially in the presence of environmental laws and regulations. Aromatic compounds are
water vapor (steaming). This behavior can be explained with the refractory to conventional chemical and biological treatments.
low tendency of Fe3+ to attain tetrahedral coordination. The Phenol is one of the most important representatives of this type
introduction of Fe3+ into the zeolite framework generates of organic pollutants, because it is toxic even at low concentra-
Brønsted acid sites, whereas iron located in extra framework sites tions and its presence in natural waters can lead to the formation
reveals redox character. Despite the fact that a considerable of halogenated compounds during the chlorine disinfection
amount of work has been carried out on zeolites44,45 or related process. Nitrophenols are common toxic, industrial and persis-
systems such as aluminophosphates,46 obtaining solid evidence tent pollutants which have to be degraded to concentrations of
for framework substitution is not a trivial task and requires the less than 20 ppb. Azo dyes, which account for approximately
combination of a plethora of characterization techniques.47 one-half of all known dyes, are commonly used as coloring
Ferrisilicate analogs of many different high-silica zeolites have agents in the food, pharmaceutical, and textile industries. As
been described and framework siting of Fe(III) ions in these a result, they are the most common synthetic colorants released
materials is well established.48 into the environment.60 Due to their intense color, azo dyes are
Several procedures of the direct synthesis of iron-containing readily visible in effluent water and can be the focus of significant
MCM-41 or SBA-15 materials have been published.49–53 While environmental complaints. More than 10 000 dyes are consumed
the synthesis of Fe-MCM-41 is straight forward, the introduc- in the textile processing industry and their concentrations range
tion of iron into SBA-15 is difficult due to the repulsion of Fe3+ from 10 to 10 000 mg L1 depending on the process. Many of the

9006 | J. Mater. Chem., 2010, 20, 9002–9017 This journal is ª The Royal Society of Chemistry 2010
 View Article Online

azo dyes and their intermediate products such as aromatic

amines are toxic to aquatic life, carcinogenic and mutagenic to
humans and have to be removed from textile wastewater before
discharge. For example, genotoxicity cannot be ruled out for
Orange II and the existing data on carcinogenicity are inadequate
for any conclusion.61
Recently more realistic wastewater streams from olive mills,
electronics manufacture and other industries are increasingly
tested with respect to the use of Fenton processes. For the
characterization of real wastewaters, global parameters such as
chemical and biological oxygen demand as well as total organic
content are important because they are related to the organic
load of the water under study. The chemical oxygen demand
(COD) of a water sample is defined as a measure of the oxygen
Published on 30 July 2010 on http://pubs.rsc.org | doi:10.1039/C0JM00577K

equivalent (given in mg O2 L1) of the organic (and inorganic)

Downloaded by University of Michigan Library on 21 February 2013

matter that can be oxidized using strong oxidizing agents. Any

organic material that is able to react with dissolved molecular
oxygen contributes to the biological oxygen demand (BOC). The
five-day biological oxygen demand (BOD5) is defined as the
amount of oxygen that is consumed by aerobic heterotrophic
bacteria during metabolization of organic water constituents
over five days at 20  C. The COD/BOD ratio allows the esti-
Fig. 3 Photo-Fenton degradation of phenol over Fe(III)-HY (after ref.
mation of the biodegradability of organic water pollutants, thus,
64).
it provides useful information for water monitoring concerning
its self-purification process. The total organic content (TOC) is components into flexible films by a simple spray technique using
the most relevant parameter for the global determination of an acrylic emulsion at ambient conditions.65 The addition of the
organic matter in wastewater. It is defined as the sum of all zeolite allows the adsorption of the pollutant molecules and thus
organically bound carbon atoms in dissolved and undissolved assists the photo-degradation. The zeolite not only concentrates
organic compounds present in the sample. Measurements of the the pollutant molecules near the photo-catalyst, but also stabi-
total carbon content (TC), which is the sum of the inorganic lizes Fe(III) and Fe(II) ions through interactions with the zeolite
carbon and the TOC, are usually based on the catalytic framework in case of the Fenton system. Thus, inorganic silica
combustion of the sample in the presence of oxygen. Alterna- fabrics, which are known for their amenability to chemical
tively, the TOC is determined photochemically. For a more modification, are believed to be one of the most promising
detailed discussion the reader is referred to ref. 1. supports for the immobilization of Fenton catalysts.
The ion-exchange of addition metal ions such as Mn2+ and
Phenol and 4-nitrophenol. Initial studies by Fajerweg et al. have Cu2+ resulting in Fe–Cu–Mn–Y catalysts appears to be beneficial
explored the conversion of phenol as a model compound over for the performance in the photo-Fenton degradation of phenol-
iron-exchanged medium-pore zeolite ZSM-5 (Fe-ZSM-5, 2 wt% containing wastewater.66 However, the suggested synergistic
Fe).62,63 Both phenol degradation and iron leaching were found effect of the presence of different metal ions is not further
to depend strongly on the pH of the solution. The optimum pH substantiated and explained as the Fe : Mn : Cu ratio did not
value was around pH ¼ 3.5, while the optimal reaction temper- affect the phenol removal rates very much. Nevertheless, the
ature was found to be 90  C. The catalyst can be recycled three addition of other metals such as Nb5+ is currently investigated in
times with limited loss of activity; however, the activity of the order to improve photo-Fenton processes.67 More work is
catalyst can be restored be calcination at 450  C in air. expected to expand into this area.
Noorjahan et al. studied the wide-pore zeolite Fe(III)-HY Ovejero et al.68 have compared the oxidation of aqueous
catalyst for an efficient degradation of phenol at pH ¼ 6. The solutions of phenol over different iron containing zeolites
photo-Fenton process exhibited on the one hand an increased (Table 2). The amount of leached iron ranges from 10 to 90%
rate of reaction which highlights the synergistic role of the zeolite depending on the synthesis route, the iron environment in the
and the degradation process and on the other hand it was not zeolite and the concentration and strength of the acid sites
limited to the narrow pH range of homogeneous systems.64 The present. While the stability towards iron leaching of most of the
authors propose a catalytic cycle of the photo-Fenton degrada- catalysts is relatively low, the two Fe-TS-1 catalysts with the
tion of phenol over Fe(III)-HY (Fig. 3) based on their activity lower iron content (viz. Fe-TS-1(1) and Fe-TS-1(2)) prepared by
studies and other spectroscopic evidence from UV-Vis, IR and direct synthesis exhibit significantly higher stability. Thus, the
XPS experiments. The results indicate that oxidation of phenol environment of the iron species is believed to play a major role
occurs in several steps. First, hydroxylation intermediates like for the stability and activity of these catalysts. It is believed that
catechol and quinine are formed, which undergo further oxida- both titanium and iron are in tetrahedral coordination in the
tion to yield muconic acid and finally oxalic acid. zeolite framework as evident from UV-Vis spectroscopy.
Kumari and co-workers examined the interplay of HZSM-5 The photo-Fenton process employing H2O2 and Fe-ZSM-5 was
and a titanium oxide thin film by immobilization of the two also used for the degradation of 4-nitrophenol by Pulgarin et al.69

This journal is ª The Royal Society of Chemistry 2010 J. Mater. Chem., 2010, 20, 9002–9017 | 9007
 View Article Online

Table 2 Catalytic properties of different iron-containing materials Wyoming SWy-1 and Tunisia-Gafsa VI (an iron-rich sample)
for phenol degradation (after ref. 68). TR ¼ 100  C, p ¼ 1 MPa, reaction based on the use of a mixture of FeCl3 and chlorhydrol
time ¼ 120 min, starting pH ¼ 5.6
(aluminium trichloride) is described by Ruiz-Hitzky et al.74 They
Iron Iron also indicate that the presence of redox centers in the layers or in
XTOC XH2O2 content/ leached the galleries of the materials, together with a Brønsted acid
Catalyst (%) (%) Final pH nSi/nTi wt% (%) environment in the galleries of the PILCs, (mixed Fe–Al-PILCs
Direct synthesis and Al-PILCs derived from the Tunisian clay) induces the
Fe-TS-1(1) 64.1 100 2.89 60 0.64 15.7 hydroxylation of phenol reaching conversions close to 70%
Fe-TS-1(2) 66.0 100 2.76 62 1.18 18.3 under microwave irradiation, even at low reaction time (5 min).
Fe-TS-1(3) 70.7 100 3.08 43 4.43 42.4 However, the stability of the iron species and their resistance to
Fe-silicalite 79.1 100 3.21 - 3.64 43.1
Fe-ZSM-5 68.2 n.d. 3.11 nSi/nAl ¼ 18 2.48 95.3 leaching was not discussed.
Ion exchange In order to expand the range of available pore sizes, iron-
Fe-NaY 78.5 n.d. 4.74 — 1.29 27.6 containing mesoporous silicates of the MCM-41 and HMS
Fe-USY 67.3 n.d. 4.67 — 1.08 57.8
(hexagonal mesoporous silica) type were prepared by adding iron
Published on 30 July 2010 on http://pubs.rsc.org | doi:10.1039/C0JM00577K

Fe-ZSM-5 54.5 n.d. n.d. — 0.33 73.9

potassium oxalate, iron potassium EDTA complex or Fe(NO3)3
Downloaded by University of Michigan Library on 21 February 2013

to the synthesis solution. It is anticipated that iron is incorpo-

At pH ¼ 3 approximately 3 mol of hydrogen peroxide per mol of rated into the silica framework by isomorphous substitution and
4-nitrophenol are required to remove all aromatic intermediate is in tetrahedral coordination within the pore walls. For all
compounds from the solution. It is suggested that the degradation materials, the conversion of phenol was total and rapid (in less
proceeds through hydroxylation of the ring leading to cleavage than 15 min at 80  C and 1 bar). The removal of total pollution
and mineralization (Fig. 4) as evident from NMR studies. (TOC) reached a plateau between 55 and 85% within three hours
Mixed Al–Fe pillared clays prepared from bentonite and a mineralization (e.g. conversion to CO2 and H2O) selectivity
(bentonite is a generally impure clay consisting mostly of of about 95%.75 Leaching is severe with these samples, the lowest
montmorillonite) were reported to be rather efficient in the value being 6% obtained with a material which is not porous and
treatment of aqueous solutions of phenol.70 Under mild reaction exhibits a low specific surface area (10 m2 g1). A more promising
conditions, viz. 70  C and 1 bar, after two hours about 80% of the approach is the use of iron-containing SBA-15, which allows the
initial phenol concentration of 5  104 mol l1 is mineralized. It complete removal of aromatic compounds with TOC reduction
is reported that the iron leaching is low, amounting to less than up to 55% in only 10 min under mild reaction conditions (100  C
0.2 wt% of the total initial iron content. The catalyst could be and 1 MPa air pressure).76 The final reaction products were
recycled two times with negligible deactivation. ESR data indi- mainly carboxylic acids (viz. maleic, acetic and formic acid).
cate that three different iron species are present in the sample, viz. However, leaching of up to 75% of the incorporated iron species
(i) isolated Fe3+ in highly distorted octahedral symmetry located is reported for this catalyst. More resistant to leaching (< 10% of
in the layer of the clay, (ii) oxide clusters and (iii) isolated iron the total iron content) is a hematite/SBA-15 composite with a Fe-
species in octahedral symmetry probably belonging to the pillars content of 16 wt% and a BET surface area of 468 m2 g1, which
as extra-framework species or by substituting the Al atoms of the removes 70% of the TOC within 10 min.77 The active iron species
pillars. The authors conclude that the latter species are very of this catalyst has not been unambiguously assigned in this
active sites for the phenol oxidation by hydrogen peroxide.71 sample, but the presence of small clusters or microaggregates of
Catalysts based on Al–Fe pillared clays prepared from Fe2O3 is suggested by diffuse reflectance UV-VIS spectroscopy.
commercial mordenite showed a higher activity for phenol The same catalyst is also tested in the photo Fenton process using
conversion and TOC removal at 298 K, whereas the pillared UV near visible irradiation (l > 313 nm).78,79 Compared to
clays with Zr–Fe decompose H2O2 more rapidly.72 Leaching of crystalline hematite and the same catalyst in the dark Fenton
iron was low (< 2 wt%) within the reaction time of practically reaction (Yphenol < 5%), the photocatalytic activity of Fe2O3/
complete phenol conversion. Similar iron-exchanged pillared SBA-15 is significantly higher and yields a phenol conversion
clays were derived from synthetic beidellite.73 At pH ¼ 5, this close to 50% after 4 h at TR ¼ 25  C and initial pH value of 5.5.
catalyst allows the total elimination of phenol and a significant The stability of the catalyst to iron leaching is strongly dependent
COD (chemical oxygen demand) removal (ca. 88% after 3 h at on the oxidant to catalyst ratio and the pH of the reaction
50  C) without generating additional pollution by iron hydroxide mixture, but the iron concentration in the aqueous solution does
sludges. In addition, the catalyst can be reused for successive runs not exceed 8 ppm. The mesoporous support plays a major role in
without significant loss of activity. the adsorption of the organic compounds and subsequent
An alternative method for the preparation of mixed Fe–Al- degradation on the solid surface results in a remarkable
pillared clays (Fe–Al-PILCs) derived from two natural smectites, enhancement in performance compared to the homogeneous
process. Nevertheless, the authors also point out that homoge-
nous reactions due to iron ions dissolved from the catalysts also
contribute to the overall performance.
The application of ultrasound irradiation in addition to the
Fe-SBA-15/H2O2 process is found to enhance the activity of the
catalyst without reducing its stability. Catalyst loadings of 0.6 g
L1 and a concentration of H2O2 close to twice the stoichiometric
Fig. 4 Mineralization of 4-nitrophenol after ref. 69. amount result in TOC removal ranging from 30 to 40%.80

9008 | J. Mater. Chem., 2010, 20, 9002–9017 This journal is ª The Royal Society of Chemistry 2010
 View Article Online

Typically, the degradation of organic pollutants using ultra- (ABET ¼ 691 m2 g1) and a carbon aerogel (ABET ¼ 641 m2 g1) as
sound treatment is ascribed to direct pyrolysis (RH + ultrasound a support for the iron catalyst.85 The catalysts have been
/ degradation products) in or around the collapsing bubbles prepared via wet impregnation using Fe(II) acetate. XPS and
and/or direct oxidation by attack of the formed hydroxyl radicals XRD measurements showed that iron is present in different
in the bulk solution or interface between the collapsing bubbles. oxidation states, viz. Fe(II) and Fe(III). The better dispersion of
In addition ultrasound (US) radiation result in the direct iron in the case of the carbon aerogel is due to the large mesopore
formation of free hydroxyl radicals from hydrogen peroxide volume and the external surface area of this sample and results in
(H2O2 + US / 2 HOc). Hydrogen peroxide also reacts with a higher catalytic activity. It is clearly shown that Orange II
supported Fe(III) active sites to produce the intermediate removal proceeds via two processes, viz. adsorption onto the
Fe(OOH)2+, which dissoziates into a hydroperoxyl radical HO2c support and catalytic degradation; the latter one is the most
and Fe2+ sites. These sites further react with H2O2 in the typical relevant one. Whether the adsorption of the substrate is an
Fenton reaction additionally generating OHc. It is believed that advantage or not is not decided yet. While most authors consider
the limiting step in the Fenton reaction is the slow kinetics of the preconcentration of the substrate in the vicinity of the reac-
Fe(III) reduction to Fe(II) mediated by the Fe(OHH)2+ complex tive sites to be beneficial, Georgi and Kopinke consider it to be
Published on 30 July 2010 on http://pubs.rsc.org | doi:10.1039/C0JM00577K

formation. The dissociation of this complex is enhanced by disadvantageous.86 They claim that the predominant degrada-
Downloaded by University of Michigan Library on 21 February 2013

sonication, which results in an increased efficiency of the US/Fe- tion pathway is the attack of the hydroxyl radicals on the organic
SBA-15/H2O2 system compared to the silent process.80 contaminant fraction that is freely dissolved in the aqueous phase
inside the pores of the activated carbon, whereas the Orange II
Orange II (C.I. Acid Orange 7). A number of papers were adsorbed on the support is nearly unreactive. A major problem
published studying the potential of Fenton systems for the with carbon supports is the attack of the carbon by the generated
decomposition of non-biodegradable azo dyes such as Orange II hydroxyl radicals resulting in leaching of the immobilized iron.
(Fig. 5).81,82,88 Different supports such as nafion membranes, In a recent comparative study, the tested carbon supports show
zeolites, MCM-41, carbon supports and pillared clays were a significantly higher leaching compared to silica-based
employed. supports.81
Fernandez et al. reported the photo-assisted degradation of Hu and co-workers used a bentonite clay based Fe-nano-
Orange II catalyzed by Fe-exchanged nafion membranes in the composite mainly consisting of Fe2O3 (hematite) and SiO2
presence of H2O2 and visible light. It was proposed that surface (quartz) crystallites87 as well as a Laponite clay composite,
iron oxide and iron hydroxide are the catalytically active sites on prepared through a reaction of Fe(NO3)3$9H2O and Laponite
the polymer fabric during the reaction.82 To reduce the cost of the RD.88 Both catalysts display a reasonable photocatalytic activity
Fe3+/Nafion membrane catalyst, Dhananjeyan et al. prepared (TOC removal of ca. 60% after 2 h) and negligible iron leaching
a Fe3+/Nafion/glass fiber catalyst for the photo-assisted Fenton (below 2 mg L1 after 120 min) at an initial solution pH of 6.6.89
degradation of Orange II.83 In another approach, Shi and co- These low cost heterogeneous Fenton catalysts are reported to
workers used immobilized Fe(III) on collagen fibers as a cata- exhibit an excellent long-term stability. Due to their nanometre
lyst.84 Collagen fiber, which is the structural protein of the scale structure and high specific surface area they significantly
connective tissue, sums up over 30% of the total weight of all enhance the degradation of Orange II under UV light irradiation.
proteins in the human body. The catalyst itself has a high A reaction kinetics analysis illustrates that the photo-Fenton
adsorption capacity for the Orange II dye molecule, which discoloration of Orange II obeys a pseudo first order kinetics in
suggests that the oxidation mainly takes place at the surface of the first 15 min of the reaction. The activation energy was
the fiber. Fe ions in solution can be re-adsorbed by the collagen calculated to 9.94 kJ mol1 indicating that this reaction is not
fiber after the oxidation of Fe(II) to Fe(III) in the presence of very sensitive to temperature.
H2O2. Therefore, the total amount of Fe ions leached from the Chen et al. suggested that the catalytic activity of a hydroxyl-
fiber is very small. Ramirez et al. employed active carbon Fe-pillared bentonite (H–Fe–P–B) is due to Fe-polycations
[Fen(OH)m(H2O)x](3nm)+ between the bentonite sheets rather than
Fe3+ or Fe2+ in the tetrahedral or octahedral sheets of bentonite.90
The extended pH range observed for pillared clays in Fenton-type
oxidations is ascribed in part to their surface acidity. Pillared clays
usually contain Brønsted and Lewis acid sites, which come from
the pillars between the bentonite sheets. H–Fe–P–B mainly
contains Brønsted acid sites; after calcination at higher tempera-
ture, the hydroxyl-Fe species will be converted to oxide pillars
through dehydration and dehydroxylation. Moreover, the
Brønsted acid sites of the pillars will change to Lewis acid sites.91
Al-pillared Saponite is a smectite clay characterized by a high
negative charge generated by a high degree of Si–Al substitution
in the tetrahedral sheet. The high occupancy of the octahedral
layer mainly by magnesium together with a certain degree of
substitution by trivalent cations result in a charge close to zero.
The saponite has been intercalated with Al polycations and was
Fig. 5 Structure of Orange II. impregnated with iron after calcination at 500  C. The batch

This journal is ª The Royal Society of Chemistry 2010 J. Mater. Chem., 2010, 20, 9002–9017 | 9009
 View Article Online

reactor results reveal that 99% of dye degradation with 91% of Dyes other than Orange II
TOC reduction is reached at 70  C using only 90 mg of clay
Studies on the degradation of different dyes using Fe-containing
catalyst per litre of solution, when the clay is impregnated with 17
zeolites and clays are compiled in Table 3. The structures of the
wt% iron using iron oxalate.92 Moreover, the amount of iron
reactive dyes are shown in Fig. 6.
released in solution is lower than 1 ppm in the worst case.
Kusic et al.102 compared the degradation of the anthraquinone
a-Fe2O3 supported on zeolite 4A (Fe-4A) was prepared with
dye C.I. Reactive Blue 49 with that of the azo reactive dye C.I.
a hydrothermal calcination method.93 Characterization by
Reactive Blue 137 employing Fenton (Fe2+/H2O2) and Fenton-
powder X-ray diffraction revealed that the iron oxide was highly
‘‘like’’ (Fe3+/H2O2 and Fe0/H2O2) processes. Both the reduction
dispersed on the surface of 4A zeolite. Fe-4A composite effec-
in TOC and AOX is enhanced in the presence of zeolites
tively degraded Orange II under acidic and neutral conditions via
(Fig. 7).103 Zeolite addition resulted in enhancement of miner-
a sono-Fenton process. The authors suggest that the 4A zeolite
alization depending on the type and the amount of zeolite added.
promotes the production of cavitation bubbles under ultrasonic
Imines, imides, iminocarbonates and sulfonamides are suggested
treatment, while a-Fe2O3 catalyzes the homolysis of oxygen
from IR experiments as possible degradation intermediates of
molecule producing hydroxyl radical and high-valent iron-oxo
Published on 30 July 2010 on http://pubs.rsc.org | doi:10.1039/C0JM00577K

RB49 and RB137.104

species. Fe-4A-4 composite with a practical a-Fe2O3 content of
The catalytic performance of an ultrastable Y zeolite (USY)
Downloaded by University of Michigan Library on 21 February 2013

24.9% showed the maximum sono-Fenton activity in the degra-

ion-exchanged with Fe for the degradation of the azo dye
dation of Orange II. Determination of the amount of dissolved
Reactive Yellow 84 (C.I. RY84; Fig. 6),105 indicates that almost
iron revealed that iron leaching from Fe-4A composite is low,
total removal of the dye color and a significant removal of COD
while the recyclability test confirmed that the Fe-4A composite
and TOC is feasible under moderate reaction conditions
possesses a long life-time under neutral condition in the sono-
(Table 4). The formation of carboxylic acids during the degra-
Fenton reaction.
dation of the dye results in a decrease of pH followed by leaching
The comparison of the above mentioned results is difficult
of iron into the reaction mixture. A similar behavior is also
since the reaction conditions are typically rather different. A very
reported in other studies. In particular, iron-exchanged zeolites
recent paper compared a number of different supports including
are prone to back exchange of iron by protons resulting in an
activated carbon, mesoporous silica (MCM-41), zeolite USY and
increased iron concentration in solution.
the clay sepiolite with largely the same initial iron content (ca.
The general question of how the large dye molecules are
4 wt%).81 Sepiolite turned out to be the best catalyst with respect
entering the zeolite micropores is typically not addressed in the
to TOC removal. Moreover, the iron leaching (ca. 14%) was the
above mentioned papers. It has, therefore, to be assumed that the
smallest for all catalysts under study.

Table 3 Survey of dyes tested

Substrate Catalysts Reaction conditions and remarks Ref.

C.I. Reactive Blue 49 NH4ZSM-5, HY Zeolite is added to iron-containing 103

C.I. Reactive Blue 137 NH4ZSM-5, HY solutions 103
C.I. Reactive Yellow 84 Fe2+/USY (ion-exchanged) PR ¼ 1 bar; TR ¼ 50  C, pH ¼ 5, 94
ccatalyst ¼ 1g L1, cH2O2 ¼ 20
mmol L1
Procion Marine H-EXL Fe2+/USY PR ¼ 1 bar; TR ¼ 50  C, pH ¼ 5, 95,96
ccatalyst ¼ 1g L1, cH2O2 ¼ 20
mmol L1, color removal (10
min) ¼ 97%
Acid Blue 74 Fe-ZSM-5 optimal conditions: UV-C lamp, 97
pH ¼ 5, cH2O2 ¼ 21.4 mmol L1,
ccatalyst ¼ 0.5 g L1; cdye ¼ 8.6 
105 mol L1; TOC reduction
(after 120 min): 57.1%
Azo dye X-3B Fe-Bentonite, Al/Fe-Bentonite UV irradiation 98
Acid light yellow H–Fe–P-Bentonite UV-A (l ¼ 365 nm) lamp, pH ¼ 3, 99
cdye ¼ 50 mg L1, ccatalyst ¼ 1 g
L1; TR ¼ 30  C; discoloration
(after 120 min) > 98%, TOC
removal ca. 65%
Reactive Brilliant Blue KNR FeY, FeZSM-5 prepared by ion- pH ¼ 2.5, cdye ¼ 250 mg L1, 100
exchange (ie) or co-precipitation ccatalyst ¼ 4 g L1; cH2O2 ¼ 30
(co) mmol L1; Order of the catalytic
activity:
FeZSM-5co > FeYie > RFeZSM-5ie
> FeYco
C.I. Reactive Orange Clinoptilolite Batch reactor experiments, artificial 101
UV-A as a light source. TR ¼ 35

C, pH ¼ 5.2, ccatalyst ¼ 1 g L1;
cH2O2 ¼ 15 mmol L1;
decoloration > 90% (60 min),
TOC removal: 27%

9010 | J. Mater. Chem., 2010, 20, 9002–9017 This journal is ª The Royal Society of Chemistry 2010
 View Article Online
Published on 30 July 2010 on http://pubs.rsc.org | doi:10.1039/C0JM00577K
Downloaded by University of Michigan Library on 21 February 2013

Fig. 6 Molecular structure of the dyes studied.

observed activity is due to iron leaching into solution or to iron wastewaters.106 Thus, it can be assumed that these materials may
species located at the outer surface of the microporous crystals. be good candidates as support for Fenton-type catalysts. A rare
Given the fact, that ordered mesoporous silicates such as MCM- example is the use of iron oxide nanoparticles supported on silica
41, SBA-15, mesocellular foams and related mesoporous carbons coated SBA-15 for the degradation of Reactive Black 5
with pore diameters exceeding 5 nm are available for more than (Fig. 6).107
10 years, it is surprising that these materials are only sparingly A novel approach is the immobilization of iron on poly-
explored. For example, it has been demonstrated that ordered electrolyte microshells formed through LbL self-assembly 108 The
mesoporous carbons with large pore sizes prepared from the resulting shells have been verified to be effective in the accumu-
surfactant-templating approach are efficient in adsorbing bulky lation of dye pollutants such as rhodamine B, methylene blue and
dye molecules (methylthionine chloride, fuchsin basic, rhoda- acridine orange. Confocal laser scanning microscopy directly
mine B, brilliant yellow, methyl orange, or Sudan G) from demonstrated that the as-synthesized (ALG/CHI)4(ALG/Fe)n

This journal is ª The Royal Society of Chemistry 2010 J. Mater. Chem., 2010, 20, 9002–9017 | 9011
 View Article Online
Published on 30 July 2010 on http://pubs.rsc.org | doi:10.1039/C0JM00577K
Downloaded by University of Michigan Library on 21 February 2013

Fig. 7 Degradation of the azo dyes C.I. RB49 and C.I. RB137 after ref. 103.

(n ¼ 1,2) shells could efficiently accumulate all three dyes in the furthermore suggest that the leaching of iron is related to the
well-defined internal space. Hydrogen peroxide is able to cross amount of oxalic acid formed during the degradation of
the shell walls and reacts with the dyes concentrated in the p-cumaric acid.110
interior of the shells under radiation with visible light. Cinnamic acid (cis-3-phenylpropenoic acid) together with its
substituted derivatives ferulic and p-cumaric acids is considered
Carboxylic acids. Centi et al. have analyzed the performance of to be a model of lignin derivates which are typically found in
Fe-ZSM-5 prepared by liquid ion-exchange in the conversion of wastewaters from agricultural origin. In comparison with
several carboxylic acids such as formic, acetic, propionic and homogenous Fenton-type catalysts, Fe-pillared montmoril-
p-cumaric acid.109,110 The latter is a model compound for lonite111 is found to increase the cinnamic acid removal rates by
phytotoxic phenols present in wastewater from agro-industrial a factor of 1.5 to 2.5 depending on the operation conditions.
processes. The solid catalyst has a higher rate of substrate Moreover, the sensitivity to pH variations is slightly reduced.
conversion as well as a lower sensitivity with respect to pH as Bozzi et al. studied the immobilization of Fe(III) ions on silica
compared to Fe3+ ions in solution. However, a major drawback is fabrics in order to degrade oxalates under visible light.112 The
the higher rate of hydrogen peroxide decomposition over the authors suggested a mechanism where the Fe3+ is extracted from
solid catalyst. Leaching tests suggest that the activity of the Fe- the support to the aqueous medium where it is photo-reduced
ZSM-5 catalyst is not due to leaching of iron, although leaching and catalyzes the Fenton process (Fig. 8). The proposed mech-
of up to 7% of the total amount of iron on the catalyst was found anism is quite complex involving two iron redox states driving
after 3 h at a reaction temperature of 70  C. The authors the conversion of the oxalate using the H2O2 added as an
oxidant. The photoreduction of Fe(III) oxalate complexes is
Table 4 Properties and performance in the degradation of the azo dye believed to play an important part in the overall decomposition
Reactive Yellow 84 of three iron-exchanged ultrastable Y zeolites. Initial of oxalate in contrast to the mechanism with solution phase
reaction conditions: 100 mg L1 C.I. RY84, 1 g L1 USY, 20 mM H2O2, complexes.113 The iron remaining on the fabric surface acts as
pH ¼ 5, T ¼ 50  C, tR ¼ 50 min
crystallization nuclei for the Fe ions returning from the solution
FeY5 FeY11.5 FeY80 by being re-adsorbed onto the silica fabric.

SiO2/Al2O3 5.1 11-5 80 Other decontaminants

Surface area ABET/(m2 g1) 660 730 780 EDTA. Fe-mordenite prepared by ion-exchange in aqueous
Fe content/% 2.76 1.69 0.56
Color removal/% 99.96 97.83 78.87 solution was tested for the photocatalytic degradation of EDTA
COD removal/% > 74 > 74 38.6 (ethylenediaminetetraacetic acid) in the presence of UV irradia-
TOC removal/% 66.8 51.6 10.0 tion (l ¼ 254 nm).114 It was found that the initial rate of degra-
H2O2 decomposition/% 41.7 44.0 23.6
Iron leached/% 8.91 11.54 5.3
dation is very fast due to EDTA adsorption on the catalyst
surface. The reaction follows first order kinetics and formic acid,

9012 | J. Mater. Chem., 2010, 20, 9002–9017 This journal is ª The Royal Society of Chemistry 2010
 View Article Online

and neutral media, H2O2 decomposes via a free radical pathway

with an activation energy of 50 kJ mol1. Under these conditions,
UDMH is rapidly oxidized by hydroxyl radicals (Ea ¼ 41 kJ
mol1). Under optimum reaction conditions, UDMH minerali-
zation amounts to 93% forming formic acid, acetic acid and
nitromethane as the only carbonaceous residues. In strong
alkaline medium, hydrogen peroxide decomposes over FeZSM-5
(Ea ¼ 17 kJ mol1) and a non-radical decomposition mechanism
seems to prevail. In alkaline suspensions, UDMH is oxidized by
aerial oxygen and non-catalytically by H2O2. Pretreatment of the
zeolite with organic or inorganic acids (oxalic, acetic, hydro-
Fig. 8 Suggested photodegradation pathway for oxalates (after ref. chloric or formic acid) may result in more active catalysts due to
112). the formation of active iron oxide or hydroxide particles on the
zeolite surface.116
Published on 30 July 2010 on http://pubs.rsc.org | doi:10.1039/C0JM00577K
Downloaded by University of Michigan Library on 21 February 2013

oxalic acid, oxamic acid, glyoxylic and glycolic acid, IMDA and Diethylnitrophenyl phosphate. FeZSM-5 prepared by direct
EDA are found as prime intermediates (Fig. 8). The photo- synthesis exhibits reduced leaching in the oxidation of a simul-
catalytic degradation of EDTA proceeds through three possible tant of warfare G-agents, diethylnitrophenyl phosphate
pathways depending on the point of attack of the cOH radicals. (DEPN).117 The yield of gaseous CO2 from DEPN amounts to
ED3A, IMDA or glyoxylic acid are formed in the initial step, 80% within 50 min. The involvement of radical species such as
successive OH radical attack in combination with other dark cOH and HO2c was shown in experiments with the radical trap
processes induces step by step cleavage of EDTA as outlined in tetranitromethane.
Fig. 9.
MTBE. Oxygenated compounds such as alcohols (e.g. meth-
1,1-Dimethylhydrazine. The principal component of liquid anol, ethanol, isopropyl alcohol and isobutyl-alcohol) and ethers
rocket fuel, 1,1-dimethylhydrazine (UDMH, unsymmetrical (methyl-tert-butyl ether (MTBE), ethyl tert-butyl ether (ETBE),
dimethylhydrazine), is known as a primary eco-toxicant. UDMH tert-amyl methyl ether (TAME) and diisopropyl ether (DIPE))
is spontaneously oxidized to N-nitrisodimethylamine, a refrac- are added to gasoline in order to improve the fuel cell combus-
tory toxic compound with acute carcinogenic and mutagenic tion efficiency and to lower exhaust emissions of CO and
properties, when exposed to soil or water under natural condi- hydrocarbons. MTBE is the most important fuel oxygenate used
tion. Employing FeZSM-5, Makhotkina et al.115 observed two worldwide and is considered a unique contaminant owing to its
different mechanisms for hydrogen peroxide decomposition ability to move readily throughout various environmental
depending on the initial pH of the reaction suspension. In acidic compartments and its resistance to degradation. Due to its

Fig. 9 Scheme of the possible degradation pathways of EDTA (after ref. 114).

This journal is ª The Royal Society of Chemistry 2010 J. Mater. Chem., 2010, 20, 9002–9017 | 9013
 View Article Online

physico-chemical properties, viz. high water solubility, low Table 5 Characterization of pharmaceutical wastewater after and
Henry’s law constant and low octanol–water partition coeffi- before the treatment with Fe2O3/SBA-15 and H2O2 (after ref. 120)a
cient, MTBE remains in the groundwater for a long time and Inlet Outlet
makes water treatment with conventional water cleaning tech- Parameter stream stream Reduction (%) Difference
nologies difficult.
Two types of iron-containing zeolites, namely Fe-Beta and Fe- COD/(mgO2/L) 1901 360 81 —
TOC/(mg L1) 860 354 59 —
ZSM-5, were tested in the degradation of MTBE in water with BOD5/(mgO2/L) 381 109 71 —
hydrogen peroxide.118 Fe-Beta showed a lower catalytic activity cCl-/(mg L1) 3380 7050 — +3670
for H2O2 decomposition and MTBE degradation. However, the BOD5/COD 0.20 0.30 — +0.1
utilization of H2O2 for MTBE degradation was more efficient for AOS 0.70 2.35 — +1.65
this catalyst. Fe-Beta has a considerably higher adsorption a
BOD5: Five-day biochemical oxygen demand.
capacity for MTBE presumably due to a more favorable fit of
molecule and zeolite pore size. Results of competition experi-
ments using Fe-ZSM-5 and compounds with different adsorp- as phenols, aldehydes, ketones, methanol and carboxylic acids,
Published on 30 July 2010 on http://pubs.rsc.org | doi:10.1039/C0JM00577K

tion tendencies namely, diethyl ether, 1,1,2-trichlorethan (TCE) which severely affect the aqueous ecosystem. Lignin degradation
Downloaded by University of Michigan Library on 21 February 2013

and humic acid, point to a positive effect of adsorption in the and pulp mill wastewater treatment using the Fenton and the
reaction, viz. an enrichment of the contaminant in the vicinity of photo Fenton processes was studied by Makhotkina et al.121 The
the active sites. The presence of high concentrations of humic lignin oxidation rate is lower for the heterogeneous Fenton
acid did not negatively affect the performance of the catalyst. catalyst based on Fe-ZSM-5 as compared to the iron nitrate in
tert-Butyl alcohol (TBA), tert-butyl formate (TBF), methyl solution probably due to steric reasons. The rather poor
acetate and acetone have been identified as the main intermedi- performance of the heterogeneous Fenton system is explained by
ates in the Fenton oxidation of MTBE. While TBA was formed the authors with the predominant oxidation of other wastewater
only in trace amounts over Fe-ZSM-5, TBF represents a signifi- constituents as evident from the decreased COD, color and
cant intermediate, which, however, will not persist and is subject turbidity. A significant improvement of the performance of this
to fast further degradation. system was observed for the UV-assisted reaction, although the
presence of Fe-ZSM-5 resulted in drastic decrease in the treat-
More complex waste streams. As early studies focused ment efficiency compared to oxidation with H2O2/UV.
predominately on the degradation of model compounds, more
recent works are increasingly devoted to the treatment of more Olive mill waste water. A series of iron-containing BEA cata-
complex and more realistic waste streams. lysts with different iron content has been tested in the oxidation
of crude olive mill wastewater under mild conditions (28  C and
Pharmaceutical wastewaters. The treatment of pharmaceutical atmospheric pressure).122 The liquid residue of olive oil extrac-
wastewaters has emerged as an important topic during the last tion is a serious environmental problem due to its high organic
decade. The effective removal of pharmaceutical residues from fraction including sugars, tannins, acids, pectins, lipids and
manufacturing plants (and hospitals) is a challenging task due to especially phenols and polyphenols which are known to be bio-
the wide variety of components and broad range of concentra- calcitrant. The best catalyst with a Fe/Al ratio of 1.2 resulted in
tions in such streams.119 Molero et al. report the treatment of a COD removal of 30% and a 40% removal of phenolic mole-
wastewater obtained from a large scale drug and fine chemical cules. UV-vis spectroscopic studies indicate that the iron cata-
manufacturing plant which is focused on the synthesis of drugs lysts employed in this study contained a range of iron species
for cardiovascular and venous diseases as well as diabetes.120 Iron from isolated ions to bulky iron oxide agglomerates. The
oxide (hematite) supported on SBA-15 was employed as catalyst different Fe environments are closely related to the leaching of
in a fixed-bed reactor. The catalyst powder was extruded with active species. It is assumed that non-tetrahedrally coordinated
bentonite and methylcellulose as binder. The high activity of the iron species are more labile to leaching than tetrahedrally-coor-
catalyst as evident from the large reduction in TOC and chemical dinated Fe ions. Moreover, catalytic activity is increased with
oxygen demand (COD) is accompanied by a remarkable stability rising iron dispersion.
of the supported iron catalyst (Table 5). Moreover, a significant The performance of copper- and iron-based pillared clays has
enhancement of the biodegradability of the outlet effluent was been found to be comparable in the treatment of olive mill waste
achieved. The BOD5/COD ratio was enhanced from 0.2 to 0.3, water containing significant amounts of chlorine-containing
while the average oxidation state (AOS) increased from 0.7 to ions.123 The copper-containing PILC shows a lower formation of
2.3. The enhancement of both parameters increases the biode- the main reaction intermediate oxalic acid compared to the iron-
gradability and makes the wastewater suitable for a conventional containing analogue. This is ascribed to higher rate of conversion
biological treatment. of this intermediate and/or the presence of alternative pathways
for the degradation of the phenolic compounds present in the
Lignin. Lignin is a cell wall constituent of the majority of wastewater. Fe-PLCs exhibit a slightly higher rate of attack on
terrestrial plants, e.g. firword and foliferous wood contain 28–30 the aromatic ring of the phenolic compounds, while nearly the
wt% and 18–24 wt% lignin respectively. In pulp mills lignin is same rate of attack on the carbon side chain is found for both
separated from fibers to be disposed with the wastewater, thus catalysts.
being the main constituent of chemical oxygen demand and Cu3(BTC)2, a porous coordination polymer, (BTC ¼ benzene
color. Decomposition of lignin results in toxic by-products such 1,3,5-tricarboxylic acid) belongs to an emerging class of novel

9014 | J. Mater. Chem., 2010, 20, 9002–9017 This journal is ª The Royal Society of Chemistry 2010
 View Article Online

crystalline materials with high specific pore volume and surface ion-clusters as connectors should be explored, although the
area, high metal content and a crystallographically defined pore instability of many MOFs against water may represent a major
size. The biodegradability of olive mill waste water from the hurdle. This class of materials may also not be suitable for the
wastewater treatment plant of Montalto Uffugo (Calabria, degradation of larger dye molecules. The combination of the
South Italy) was evaluated after treatment with hydrogen Fenton reaction with UV irradiation, ultrasound and maybe
peroxide in the presence of Cu3(BTC)2 at room temperature microwave irradiation looks promising, although such tech-
under atmospheric pressure.124 The treatment increases the niques might be more difficult to apply in large-scale processing.
BCOD (biodegradable COD) fraction from 52 to 84.5% with However, the deficit in fundamental knowledge on the
a polyphenol abatement of 96%. Moreover, a significant reduc- involved chemistry hampers the tailoring of the catalyst with
tion of other nonbiodegradable compounds and a small reduc- respect to its envisaged task. The majority of the studies pub-
tion of the total COD are observed. lished so far are devoted to test a series of catalysts with respect
to its activity in the degradation of model compounds such as
Wine Vinasse. The advanced treatment of wine vinasses using phenol, Orange II or carboxylic acids. Furthermore, optimiza-
the Fenton reaction was investigated by Yang et al.125 Their tion of experimental conditions including temperature, reactant
Published on 30 July 2010 on http://pubs.rsc.org | doi:10.1039/C0JM00577K

results indicate that the chemical oxygen demand (COD), concentration and pH was predominantly addressed. The
Downloaded by University of Michigan Library on 21 February 2013

biological oxygen demand (BOD5), and color level were below obvious lack of careful experiments aiming at a proper design of
250 mg L1, 60 mg L1, and 25 C.U., respectively, after a reaction both catalyst (in particular with respect to the iron species
time of 30 min, when the dosages of FeSO4 and H2O2 were present) and reactor is at present a major hurdle for further
450 mg L1 and 300 mg L1, respectively. Also, the experimental development. Several fundamental questions are still not satis-
results showed that settling prior to pH adjustment was better factorily answered: (1) is adsorption of the reactants on the
than pH adjustment prior to settling after Fenton’s reaction. It is porous support beneficial for the Fenton process? (2) Are the
shown that the destruction of the recalcitrant organics in the large dye molecules adsorbed in the pores of the catalyst or is
winery wastewater was effective and alcohols and aldehydes are leaching of iron species into the solution required for their
the major organic constituents of Fenton-treated effluents. degradation? (3) Can the reaction be performed truly heteroge-
neously?
In order to address the above mentioned points, thorough
Conclusions and outlook
studies devoted to catalyst preparation and rational develop-
The majority of the studies published so far dealing with the use ment, characterization of active species present by means of
of solid Fenton-type catalysts based on porous materials confirm a range of spectroscopic techniques and elucidation of the
the potential of these technology for wastewater treatment using degradation pathways have to be conducted. After the successful
hydrogen peroxide in the presence or absence of light. The use of development of an active and stable catalyst, more efforts have to
a solid allows an easier recovery of the catalyst after the reaction be directed to proper reactor and process design. In particular,
and, thus, reduces the contamination of the effluent with metal studies of more realistic waste streams such as olive mill waste-
(predominantly iron or copper) ions. Moreover, it has been water, mixtures of different dyes or hospital discharge containing
shown that the presence of solids has a marked influence on the a variety of different drugs in high dilution are highly welcome.
concentration of radical species present, which significantly alters Finally, the products obtained during treatment frequently show
the relative rates of the set of radical reactions. higher toxicity than the precursor compounds. When the organic
While initial studies used conventional zeolites such as Y and content is not completely removed, the remediation capacity of
ZSM-5, which have micropores with entrance diameters smaller the process has to be evaluated by determining the acute toxicity
than 1 nm, more recent work also evaluates mesoporous silica e.g. by using the bioluminescence in Vibrio fisheri for primary
materials with pore diameters between 4 and 10 nm. While access estimation.
of large dye molecules to the zeolitic micropores is unlikely, the
mesoporous supports may allow adsorption of these large Acknowledgements
molecules inside the mesopores. In particular, the degradation of
bulky dyes may be possible using these materials as support for Financial support by Fonds der Chemischen Industrie is grate-
Fenton-type catalysts. The large diversity in chemical composi- fully acknowledged.
tion available and the range of pore sizes that can be synthesized
render these materials ideal candidates for the tailor-made References
preparation of heterogeneous Fenton catalysts. The use of other
1 T. Oppenl€ander, Photochemical Purification of Water and Air, Wiley-
transition metal ions such as copper, cobalt, vanadium or VCH, Weinheim, 2003.
niobium should also be considered. Surface modification in order 2 D. F. Bishop, G. Stern, M. Fleischman and L. S. Marshall, Ind. Eng.
to e.g. fine tune the hydrophobic or hydrophilic properties of the Chem. Process Des. Dev., 1968, 7, 110–117.
porous support may also be a promising target. 3 P. Schulte, M. Volkmer, F. Kuhn, US Patent 1994, 5, pp. 348–665.
4 J. Prousek, Chem. Listy, 1995, 89, 11–21.
A still pending task is the preparation of truly heterogeneous 5 M. E. Lindsey and M. A. Tarr, Chemosphere, 2000, 41, 409–417.
catalysts, where leaching of iron species in solution is largely 6 R. J. Watts and S. E. Dilly, J. Hazard. Mater., 1996, 51, 209–224.
suppressed. The use of clay-based catalysts seems to be a step in 7 V. S. Millioli, D. D. C. Freire and M. C. Cammarota, Engenharia
Termica, Edição Especial, 2002, 44–47.
the right direction, but the principle problem is still not satis-
8 C. M. Miller, R. L. Valentine, M. C. Roehl and P. J. J. Alvarez,
factorily solved up till now. The use of novel materials such as Water Res., 1996, 30, 2579–2586.
porous coordination polymers (MOFs) which contain 9 H. J. H. Fenton, J. Chem. Soc., 1894, 65, 899–910.

This journal is ª The Royal Society of Chemistry 2010 J. Mater. Chem., 2010, 20, 9002–9017 | 9015
 View Article Online

10 F. Haber and J. J. Weiss, Proc. R. Soc. London, Ser. A, 1934, 147, 51 P. B. Amama, S. Lim, D. Ciuparu, Y. Yang, L. Pfefferle and
332–351. G. L. Haller, J. Phys. Chem. B, 2005, 109, 2645–2656.
11 W. G. Barb, J. H. Baxendale, P. George and K. R. Hargrave, Trans. 52 Z. Y. Yuan, S. Q. Liu, T. H. Chen, J. Z. Wang and H. X. Li, J. Chem.
Faraday Soc., 1951, 47, 591–616. Soc., Chem. Commun., 1995, 973–974.
12 C. Walling, Acc. Chem. Res., 1975, 8, 125–131. 53 S. Samanta, S. Giri, P. U. Sastry, N. K. Mal, A. Manna and
13 W. J. Cooper, C. K. Duesterberg and T. D. Waite, Environ. Sci. A. Bhaumik, Ind. Eng. Chem. Res., 2003, 42, 3012–3018.
Technol., 2005, 39, 5052–5058. 54 A. Vinu, D. P. Sawant, K. Ariga, K. Z. Hossain, S. B. Halligudi,
14 H. G. Joseph De Laat, Environ. Sci. Technol., 1999, 33, 2726–2732. M. Hartmann and M. Nomura, Chem. Mater., 2005, 17, 5339–5345.
15 W. P. Kwan and B. M. Voelker, Environ. Sci. Technol., 2002, 36, 55 Y. Li, Z. Feng, Y. Lian, K. Sun, L. Zhang, G. Jia, Q. Yang and C. Li,
1467–1476. Microporous Mesoporous Mater., 2005, 84, 41–49.
16 F. Gozzo, J. Mol. Catal. A: Chem., 2001, 171, 1–22. 56 Y. Li, Z. L. Zhang, Q. Yang, Z. Wei, Z. Feng and C. Li, J. Phys.
17 C. W. Jones, Applications of hydrogen peroxide and derivates, The Chem. B, 2004, 108, 9739–9744.
Royal Society of Chemistry, Cambridge, UK, 1999. 57 J. M. Kim, Y.-J. Han, B. F. Chmelka and G. D. Stucky, Chem.
18 C. Walling, Acc. Chem. Res., 1998, 31, 155–157. Commun., 2000, 2437–2438.
19 D. T. Sawyer, A. Sobkowiak and T. Matsushita, Acc. Chem. Res., 58 P. Decyk, M. Trejda and M. Ziolek, C.R. Chimie, 2005, 8, 635–654.
1996, 29, 409–416. 59 P. Kustrowski, L. Chmielarz, J. Surman, E. Bidzinska,
20 O. Pestovsky, S. Stoian, E. L. Bominaar, X. Shan, E. M€ unck, L. Que R. Dziembaj, P. Cool and E. F. Vansant, J. Phys. Chem. A, 2005,
Published on 30 July 2010 on http://pubs.rsc.org | doi:10.1039/C0JM00577K

Jr. and A. Bakac, Angew. Chem., 2005, 117, 7031–7034. 109, 9808–9815.
21 G. Ruppert, R. Bauer and G. Heisler, J. Photochem. Photobiol., A, 60 J. Bumpus, Microbial Degradation of Azo Dyes.
Downloaded by University of Michigan Library on 21 February 2013

1993, 73, 75–78. Biotransformations: microbial degradation of health-risk compounds,

22 B. Faust and J. Hoigne, Atmos. Environ., 1990, 24A, 79–89. Elsevier, New York. 1995.
23 S. Perathoner and G. Centi, Top. Catal., 2005, 33, 207–224. 61 The EFSA Journal, 2005, 263.
24 E. V. Starokon, K. A. Dubkov, L. V. Pirutko and G. I. Panov, Top. 62 K. Fajerweg and H. Debellefontaine, Appl. Catal., B, 1996, 10,
Catal., 2003, 23, 137–143. L229–235.
25 M. Hartmann and B. Boddenberg, Microporous Mater., 1994, 2, 63 K. Fajerweg, J. N. Foussard, A. Perrard and H. Debellefontaine,
127–136. Water Sci. Technol., 1997, 35, 103–110.
26 E. V. Kuznetsova, E. N. Savinov, L. A. Vostrikova and 64 M. Noorjahan, V. D. Kumari, M. Subrahmanyam and L. Panda,
V. N. Parmon, Appl. Catal., B, 2004, 51, 165–170. Appl. Catal., B, 2005, 57, 291–298.
27 J. Barrault, C. Bouchoule, K. Echachoui, N. Frini-Srasra and 65 M. Noorjahan, V. D. Kumari, M. Subrahmanyam and P. Boule,
M. Trabelski, Appl. Catal., B, 1998, 15, 269–274. Appl. Catal., B, 2004, 47, 209–213.
28 S. Hamoudi, F. Larachi and A. Sayari, J. Catal., 1998, 177, 247–258. 66 Z. Zheng, L. Lei, S. Xu and P. Cen, J. Zhejiang Univ., Sci., 2004, 5,
29 F. Larachi, S. Levesque and A. Sayari, J. Chem. Technol. 206–211.
Biotechnol., 1998, 73, 127–130. 67 L. C. A. Oliveira, M. Goncalves, M. C. Guerreiro, T. C. Ramalho,
30 S. Valange, Z. Gabelica, M. Abdellaudi, J. M. Clacens and J. D. Fabris, M. C. Pereira and K. Sapag, Appl. Catal., A, 2007,
J. Barrault, Microporous Mesoporous Mater., 1999, 30, 177. 316, 117–124.
31 J.-R. Butruille, T. J. Pinnavia, in Comprehensive Supramolecular 68 G. Ovejero, J. L. Sotelo, F. Martinez, J. A. Melero and K. Gordo,
Chemistry Vol. 7, ed. G. Alberti, T. Bein, Pergamon, Oxford, pp. Ind. Eng. Chem. Res., 2001, 40, 3921–3928.
219–250. 69 C. Pulgarin, P. Peringer, P. Albers and J. Kimi, J. Mol. Catal. A:
32 J. Perez-Ramırez, M. S. Kumar and A. Br€ uckner, J. Catal., 2004, Chem., 1995, 95, 61–74.
223, 13–27. 70 J. Barrault, M. Abdellaoui, C. Bouchoule, A. Majeste,
33 H.-Y. Chen and W. M. H. Sachtler, Catal. Today, 1998, 42, 73–83. J. M. Tatibouet, A. Louloudi, N. Papayannakos and
34 A. E. Shilov and G. B. Shulpin, Chem. Rev., 1997, 97, 2879–2932. N. H. Gangas, Appl. Catal., B, 2000, 27, L225–235.
35 J. Valyon, W. S. Millman and W. K. Hall, Catal. Lett., 1994, 24, 71 E. Guelou, J. Barrault, J. Fournier and J. M. Tatibouet, Appl. Catal.,
215–225. B, 2003, 44, 1–8.
36 P. Kustrowski, L. Chmielarz, J. Surman, E. Bidzinska, R. Dziembaj, 72 C. B. Mollina, J. A. Casas, J. A. Zazo and J. J. Rodriguez, Chem.
P. Cool and E. F. Vansant, J. Phys. Chem. A, 2005, 109, 9808–9815. Eng. J., 2006, 118, 29–35.
37 G. Sartori and R. Maggi, Chem. Rev., 2006, 106, 1077–1104. 73 C. Catrinescu, C. Teodosiu, M. Macoveanu, J. Miehe-Brendle and
38 A. Corma and H. Garcia, Chem. Rev., 2002, 102, 3837–3892. R. Le Dred, Water Res., 2003, 37, 1154–1160.
39 I. Melian-Cabrera, F. Kapteijn and J. A. Moulijn, Catal. Today, 74 S. Letaief, B. Cascal, P. Aranda, M. A. Martin-Luengo and E. Ruiz-
2005, 110, 255–263. Hitzky, Appl. Clay Sci., 2003, 22, 263.
40 P. Marturano, L. Drozdova, A. Kogelbauer and R. Prins, J. Catal., 75 N. Crowther and F. Larachi, Appl. Catal., B, 2003, 46, 293–305.
2000, 190, 460–468. 76 G. Calleja, J. A. Melero, F. Martinez and R. Molina, Water Res.,
41 I. Melian-Cabrera, F. Kapteijn and J. A. Moulijn, Chem. Commun., 2005, 39, 1741–1750.
2005, 2744–2746. 77 J. A. Melero, G. Calleja, F. Martinez, R. Molina and M. I. Pariente,
42 M. Rauscher, K. Kesore, R. M€ onnig, W. Schwieger, A. Tißler and Chem. Eng. J., 2007, 131, 245.
T. Turek, Appl. Catal., A, 1999, 184, 249–256. 78 F. Martinez, G. Calleja, J. A. Melero and R. Molina, Appl. Catal., B,
43 G. D. Pirngruber, in Ordered Porous Solids–Recent Advances and 2005, 60, 181–190.
Prospects, 2009, ed. V. Valtchev, S. Mintova, M. Tsapatsis, 79 F. Martinez, G. Calleja, J. A. Melero and R. Molina, Appl. Catal., B,
Elsevier, Amsterdam, pp. 749–771. 2007, 70, 452–460.
44 R. Aiello, J. B. Nagy, G. Giordano, A. Katovic and F. Testa, C.R. 80 R. Molina, F. Martinez, J. A. Melero, D. H. Bremner and
Chimie, 2005, 8, 321–329. A. G. Chakinala, Appl. Catal., B, 2006, 66, 198–207.
45 V. Umamaheswari, A. Vinu, W. B€ ohlmann, A. P€ oppl and 81 A. Rodriguez, G. Ovejero, J. L. Sotelo, M. Mestanza and J. Garcia,
M. Hartmann, Microporous Mesoporous Mater., 2006, 89, 47–57. Ind.Eng. Chem. Res., 2010, 49, 598–505.
46 M. Hartmann and L. Kevan, Res. Chem. Intermed., 2002, 28, 625– 82 J. Fernandez, J. Bandara, A. Lopez, Ph. Buffat and J. Kiwi,
695. Langmuir, 1999, 15, 185–192.
47 J. Perez-Ramirez, J. C. Groen, A. Br€ uckner, M. S. Kumar, 83 M. R. Dhananjeyan, J. Kiwi, P. Albers and O. Enea, Helv. Chim.
U. Bentrup, M. N. Debbagh and L. A. Villaescusa, J. Catal., Acta, 2001, 84, 3433.
2005, 232, 318–334. 84 R. Tang, X.-P. Liao, X. Liu and B. Shi, Chem. Commun., 2005,
48 R. Kumar and P. Ratnasamy, Stud. Surf. Sci. Catal., 1991, 60, 43– 5882–5884.
52. 85 J. H. Ramirez, F. J. Maldonado-Hodar, A. F. Perez-Cadenas,
49 A. Vinu, K. U. Nandhini, V. Murugesan, W. B€ ohlmann, C. Moreno-Castillia, C. A. Costa and L. M. Madeira, Appl.
V. Umamaheswari, A. P€ oppl and M. Hartmann, Appl. Catal., A, Catal., B, 2007, 75, 312–323.
2004, 265, 1–10. 86 A. Georgi and F. D. Kopinke, Appl. Catal., B, 2005, 58, 9–18.
50 M. Stockenhuber, M. J. Hudson and R. W. Joyner, J. Phys. Chem. 87 J. Feng, X. Hu and P. L. Yue, Environ. Sci. Technol., 2004, 38, 269–
B, 2000, 104, 3370–3374. 275.

9016 | J. Mater. Chem., 2010, 20, 9002–9017 This journal is ª The Royal Society of Chemistry 2010
 View Article Online

88 J. Feng, X. Hu, P. L. Yue, H. Y. Zhu and G. Q. Lu, Ind. Eng. Chem. 108 X. Tao, J. Su, L. Wang and J.-F. Chen, J. Mol. Catal. A: Chem.,
Res., 2003, 42, 2058–2066. 2008, 280, 186–193.
89 J. Feng, X. Hu and P. L. Yue, Water Res., 2006, 40, 641. 109 G. Centi, S. Perathoner, T. Torre and M. G. Verduna, Catal. Today,
90 J. Chen and L. Zhu, J. Photochem. Photobiol., A, 2007, 188, 56. 2000, 55, 61–69.
91 F. Kooli and W. Jones, Chem. Mater., 1997, 9, 2913. 110 G. Centi, S. Perathoner and G. Romeo, Stud. Surf. Sci. Catal., 2001,
92 J. Herney Ramirez, C. A. Costa, L. M. Madeira, G. Mata, 135, 5156–5163.
M. A. Vicente, M. L. Rojas-Cervantes, A. J. Lopez-Peinardo and 111 D. Tabet, M. Daidi, M. Houari, P. Pichat and H. Khalaf, J. Environ.
R. M. Martin-Aranda, Appl. Catal., B, 2007, 71, 44–56. Manage., 2006, 80, 342–346.
93 F. Chen, Y. Li, W. Cai and J. Zhang, J. Hazardous Mater, 2010, 177, 112 A. Bozzi, T. Yuranova, J. Mielczarski, A. Lopez and J. Kiwi, Chem.
743–749. Commun., 2002, 2202–2203.
94 M. Neamtu, C. Catrinescu and A. Kettrup, Appl. Catal., B, 2004, 51, 113 A. Bozzi, T. Yuranova, J. Mielczarski and J. Kiwi, New J. Chem.,
149–157. 2004, 28, 519–526.
95 C. Catrinescu, M. Neamtu, M. Macoveanu, A. Yediler and 114 M. M. Emara, A. S. M. Tourky and M. M. El-Moselhy, J. Hazard.
A. Kettrup, Environ. Eng. Mange. J., 2002, 1, 177–186. Mater., 2009, 166, 514–522.
96 M. Neamtu, C. Zaharia, C. Catrinescu, A. Yediler, M. Macoveanu 115 O. A. Makhotkina, E. V. Kuznetsova and S. V. Preis, Appl. Catal.,
and A. Kettrup, Appl. Catal., B, 2004, 48, 287–294. B, 2006, 68, 85–91.
97 M. B. Kasiri, H. Aleboyeh and A. Aleboyeh, Appl. Catal., B, 2008, 116 E. V. Parkhomchuk, M. P. Vanina and S. Preis, Catal. Commun.,
Published on 30 July 2010 on http://pubs.rsc.org | doi:10.1039/C0JM00577K

84, 9–15. 2008, 9, 381–385.

98 Y. Liu, Y. Lu and X. Zhu, J. Hazard. Mater., 2006, 132, 196–201. 117 E. V. Kuznetsova, E. N. Savinov, L. A. Vostrikova and
Downloaded by University of Michigan Library on 21 February 2013

99 J. Chen and L. Zhu, Catal. Today, 2007, 126, 463–470. G. V. Echevskii, Water Sci. Technol., 2004, 49, 109–116.
100 A. Chen, X. Ma and H. Sun, J. Hazard. Mater., 2008, 156, 568–575. 118 R. Gonzales-Olmos, U. Roland, H. Toufar and F.-D. Kopinke,
101 M. Tekbas, H. Cengiz Yatmaz and N. Bektas, Microporous Appl. Catal., B, 2009, 89, 356–364.
Mesoporous Mater., 2008, 115, 594–602. 119 P. Bautista, A. F. Mohedano, J. A. Casas, J. A. Zazo and
102 H. Kusic, A. Loncaric Bozic and N. Koprivanac, Dyes Pigm., 2007, J. J. Rodriguez, J. Chem. Technol. Biotechnol., 2008, 83, 1323–1338.
74, 380–387. 120 J. A. Molero, F. Martinez, J. A. Botas, R. Molina and M. I. Pariente,
103 H. Kusic, A. Loncaric Bozic, N. Koprivanac and S. Paic, Dyes Water Res., 2009, 43, 4010–4018.
Pigm., 2007, 74, 388–395. 121 O. A. Makhotkina, S. V. Preis and E. V. Parkhomchuk, Appl. Catal.,
104 W. Feng, D. Nansheng and H. Helin, Chemosphere, 2000, 41, 1233– B, 2008, 84, 821–826.
1238. 122 W. Jajjar, S. Azabou, S. Sayadi and A. Ghorbel, Appl. Catal., B,
105 M. Neamtu, C. Catrinescu and A. Kettrup, Appl. Catal., B, 2004, 51, 2009, 88, 299–304.
149–157. 123 S. Caudo, G. Centi, C. Genovese and S. Perathoner, Appl. Catal., B,
106 X. Zhuang, Y. Wan, C. Feng, Y. Shen and D. Zhao, Chem. Mater., 2007, 70, 437–446.
2009, 21, 706–716. 124 S. De Rosa, G. Giordano, T. Granato, A. Katovic, A. Siliciliano and
107 H. Lim, J. Lee, S. Jin, J. Yoon and T. Hyeon, Chem. Commun., 2006, F. Tripicchio, J. Agric. Food Chem., 2005, 53, 8306–8309.
463–465. 125 J. Yang, M. Xing and X. Zhou, Environ. Eng. Sci., 2008, 25, 1229–1234.

This journal is ª The Royal Society of Chemistry 2010 J. Mater. Chem., 2010, 20, 9002–9017 | 9017