Enhancement of the photocatalytic activity of TiO2 through spatial structuring and particle size control: From subnanometric to submillimetric length scale

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

PERSPECTIVE

www.rsc.org/pccp | Physical Chemistry Chemical Physics

Enhancement of the photocatalytic activity of TiO₂through spatial

structuring and particle size control: from subnanometric to

submillimetric length scale

Carmela Aprile, Avelino Corma* and Hermenegildo Garcia*

Received 7th August 2007, Accepted 18th October 2007

First published as an Advance Article on the web 5th November 2007

DOI: 10.1039/b712168g

This review summarizes the physical approaches towards enhancement of the photocatalytic

activity of titanium dioxide by controlling this semiconductor in a certain length scale from

subnanometric to submillimetric distances and provides examples in which the photocatalytic

activity of TiO₂is not promoted by doping or changes in the chemical composition, but rather by

application of physical concepts and spatial structuring of the semiconductor. Thus, encapsulation

inside the micropores and cavities of zeolites (about 1 nm) renders small titanium oxide clusters

with harnessed photocatalytic activity. On the other hand, nanometric titanium particles can be

ordered forming structured periodic mesoporous materials with high speciﬁc surface area and well

deﬁned porosity. Titiania nanotubes of micrometric length, either independent or forming a

membrane, also exhibit unique photocatalytic activity as consequence of the long diﬀusion length

of charge carriers along the nanotube axis. Finally, photonic crystals with an inverse opal

structure and the even more powerful concept of photonic sponges can serve to slow down visible

light photons inside the material, increasing the eﬀective optical path in such a way that light

absorption near the absorption onset of the material is enhanced considerably. All these physical-

based approaches have shown their potential in enhancing the photocatalytic activity of titania,

paving the way for a new generation of novel structured photocatalysts in which physical and

chemical concepts are combined.

tool either to degrade organic pollutants in the gas or liquid

1. Introduction

phase or to perform useful transformations of organic com-

pounds.^1–7,17,19One of the most important limitations is the

Titanium dioxide is the most important photocatalyst because

lack of TiO₂photocatalytic activity with visible light.^1,20,21

it exhibits a high durability, corrosion resistance and high

oxidation potential of the valence band that ensures a general

applicability to activate a broad range of substrates.^1–18

The reason for this is that TiO₂in the anatase form is a wide

band gap semiconductor with a bandgap of 3.2 eV in most

However, there are important drawbacks that severely limit

media, corresponding to an onset of the optical absorption

the application of titanium dioxide photocatalyst as a general

band at about 350 nm. This onset of the TiO₂absorption is

also inadequate to achieve eﬃcient solar light photocatalytic

´

Instituto de Tecnologıa QuımicaCSIC-UPV, Universidad Politecnica

de Valencia, Av. De los Naranjos s/n, 46022 Valencia, Spain. E-mail:

acorma@itq.upv.es, hgarcia@qim.upv.es

activity, since less than 5% of the solar light energy can be

absorbed by TiO₂.

´

Garcıa

full

Hermenegildo

(Herme) has been

Avelino Corma was born in

´

a

Moncofar, Spain in 1951. He

has been the founder and di-

rector of the Instituto de Tec-

Professor and member of the

Institute of Chemical Technol-

ogy at the Technical Univer-

sity of Valencia since 1996. He

has co-authored over 300 pa-

pers and holds 10 international

patents. His main current in-

terests are in green chemistry

and the use of zeolites and

mesoporous materials in supra-

molecular photochemistry and

photocatalysis.

´

nologıa Quımica (UPV-

CSIC) at the Universidad Po-

´

litecnica de Valencia since

1990. One of his research

lines is in zeolites and struc-

tured mesoporous materials,

from the point of view of

synthesis,

characterization

and reactivity. Avelino Corma

is co-author of more than 700

articles and 100 patents,

several of them in commercial use. He is a member of the

PCCP Advisory Board.

ꢀc

This journal is the Owner Societies 2008

Phys. Chem. Chem. Phys., 2008, 10, 769–783 | 769

View Article Online

and the nature of the resulting material is very often not well

deﬁned and, most probably, controversial results can be

obtained depending on the way in which the metal has been

introduced and also depending on the ﬁnal concentration of

the dopant. Thus, it has often been reported that there is an

optimum doping level to achieve the maximum eﬃciency and

beyond this point a decrease in photocatalytic activity is again

observed.^30,46Nevertheless, a consensus is starting to be

achieved about the inappropriateness of metal doping as a

valid solution to enhance the photocatalytic activity of TiO₂, if

not for anything else, just by the fact that the materials

corrode and the dopant is leached.^12–14,31

More recently doping with nitrogen, carbon, sulfur and

other non-metallic elements have also been reported to intro-

duce visible light absorption of titanium dioxide.^{20,22,46–56}

Asahi and co-workers were the ﬁrst to show an absorption

increase in the visible region upon nitrogen doping.⁴⁷This

opened the way to study titania doping with non-metallic

elements.^33,48,57However, due to corrosion and instability of

doped materials, it remains to be seen whether or not non-

metallic element doping can be regarded as a general and valid

approach to increase the photocatalytic eﬃciency of titania

photocatalysts.

Scheme 1 Illustration of photoinduced charge separation (electron

and hole) upon illumination of a titania particle with light of energy

higher than the band gap. Positive hole has suﬃcient oxidation

potential to initiate aerobic oxidation of many organic compounds.

The above comments explain the continued interest in

improving the photocatalytic eﬃciency of TiO₂. To increase

the photocatalytic activity of TiO₂, particularly for visible light

irradiation, two general strategies have been developed com-

prising either using an organic dye as photosensitizer or

doping TiO₂with metallic and non-metallic elements.^20,22–37

The ﬁrst approach (i.e. the use of an organic dye that absorbs

visible light) has worked very well under conditions where

oxygen is excluded and the degradation of the dye is mini-

mized by eﬃcient quenching of the dye oxidation state with an

appropriate electrolyte.^{26–29,38–42}Otherwise, particularly in the

presence of oxygen, the dye becomes rapidly degraded and the

photocatalytic system loses its response towards visible light.

The mechanism of TiO₂dye sensitization has been determined

using time resolved subnanosecond laser ﬂash photolysis

techniques.^{38–41,43,44}These studies have shown that upon light

absorption, the dye in its excited electronic state injects one

electron into the semiconductor conduction band. Scheme 1

illustrates a simple picture of this process.

2. Physical methods to improve the photocatalytic

response

While the above strategies to improve the photocatalytic

eﬃciency have in common that they are based on chemical

concepts which direct the modiﬁcation of the chemical com-

position of the TiO₂material to enhance its photocatalytic

activity, in the present account we will address a diﬀerent

methodology to alter the photocatalytic response that is based

on Physics rather than on Chemistry. In this regard, it is worth

reminding that by reducing the particle size at the nanometric

length scale, a point is reached behind which further decrease

in the size (without altering the chemical composition)

strongly inﬂuences the band gap and the position of the

valence and conducting bands.^20,58This change in the proper-

ties with the particle dimensions is generally described as

‘‘quantum size eﬀect’’ and is a well known example of opera-

tion of physical concepts in the optical properties of a

semiconductor.^59–62

In dye sensitization, the most relevant points are the absorp-

tion spectrum of the dye in the visible region and the energy of

the electron in the excited electronic state of the dye that has to

be high enough to be transferred to the semiconductor con-

duction band. However, electron injection of electron donor

dyes in their excited state onto the TiO₂conduction band is a

quite general phenomenon since the conduction band energy

of TiO₂is not very high. In fact, TiO₂holes are considerably

much more oxidizing than the conduction band electrons are

reducing.

In the present review, we will describe novel methodologies

to achieve the control of the photocatalytic activity of the

materials by structuring the titanium oxide particles at diﬀer-

ent length scales going from subnanometers to submillimeters.

Table 1 summarizes the systems studied and the length scale of

the spatial structuring covered by this Perspective.

A second chemical approach to promote photoresponse of

titanium dioxide into the visible consists in doping the TiO₂

material either with metallic or non metallic elements.^20,22–29

In this case doping introduces occupied or unoccupied orbitals

in the band gap region leading to negative or positive doping,

respectively. However, particularly in the case of metal doping

there are contradictory reports describing either an increase or

a decrease of the catalytic activity.^13,32,34,45

Obviously, physical and chemical concepts are not incom-

patible and there are a few pioneering examples in which a

combination of both, doping and structuring, has been ap-

plied.^20,57,63For the purpose of this review the content has

been organized on the length scale in which TiO₂structuring

has taken place. We will start from those reports in which TiO₂

nanoparticles of nanometric and subnanometric size have been

used as photocatalysts and move towards other reports in

which the spatial structuring of the photocatalyst goes into

micrometric and submillimetric length scales. Although, in this

This controversy arises in part from the diﬃculty to estab-

lish valid comparisons about the photocatalytic activity

among diﬀerent solids testing diﬀerent probe molecules and

employing inconsistent parameters. Also the doping procedure

ꢀc

This journal is the Owner Societies 2008

770 | Phys. Chem. Chem. Phys., 2008, 10, 769–783

View Article Online

Table 1 Examples of spatial structuring of titanium oxides at various

length scales

ITQ-33. Zeolites do not absorb light in the Vis and UV region

and, therefore, it is possible to irradiate guests that are

incorporated inside their micropores. After all, zeolites have

the same chemical composition as quartz and glasses com-

monly used as cells to perform photochemical reactors and

optical tools.

Length scale

Examples

o1 nanometer

Titania clusters encapsulated inside zeolites

Titania nanoparticles forming periodic

mesoporous powders

1–10 nanometers

10 nanometers–10

micrometers

Titania nanotubes

TiO₂can be incorporated into the micropores of zeolites.⁷⁷

There are two diﬀerent strategies to form TiO₂clusters inside

zeolites. Both of them have in common the use of molecular

titanium compounds having a single titanium atom that

subsequently will oligomerize forming Ti–O–Ti bonds. In the

ﬁrst approach, vapors of TiCl₃or TiCl₄are absorbed into

zeolite totally or partially dehydrated and in the second step

hydrolysis and oligomerization to form TiO₂clusters is pro-

moted by thermal treatment in the presence of moisture.^69,70,78

Depending on the level of dehydration of zeolite, the forma-

tion of TiO₂aggregates can take place predominantly on the

exterior (when water ﬁlls the pores during the titanium ab-

sorption) or in the interior (when titanium halides are ad-

sorbed in the interior of the micropores before hydrolysis) of

the zeolite micropores. The main problem of TiCl₃is that it is

an explosive and pyrophoric compound, while both TiCl₃and

TiCl₄form highly corrosive HCl vapors upon hydrolysis.^79–81

These harsh acid conditions can damage to some extent

the zeolite structure particularly producing a partial loss in

crystallinity.

Titania membranes

Photonic crystals

Photonic sponges

4200 micrometers

account we will preferentially discuss the work carried out in

our group, appropriate reference to the work performed by

other groups in the area has also been included to give a

comprehensive view of the ﬁeld. The aim of this review is to

exemplify the potential and opportunities that the use of TiO₂

semiconductor with controlled size or structured in a periodic

way, may have on improving the photocatalytic eﬃciency.

3. Photocatalytic activity of subnanometric

titanium dioxide clusters

The most widely used TiO₂photocatalyst is the Degussa P25

material.^{6,10,15,64–66}The particle size of P25 is about 25 nm and

its surface area is very small (50 m²g^ꢁ1). A reduction of the

particle size, up to a few nanometres, has the beneﬁt of

increasing the external surface area. These small particles tend

to agglomerate by strong interparticle forces when the nano-

metric size region is reached. Further decrease of the particle

size to a few nanometers reaches one point below which

quantum size eﬀects also start to operate and the band gap

of the semiconductor increases, blue-shifting the absorption

Alternatively, TiO₂clusters can be formed inside the zeolite

from the aqueous phase by ion exchanging of Na⁺(the

counter cation normally present in synthetic zeolites) by

(TiQO)²⁺titanyl salts.^82,83These salts are commercially

available, they are non-corrosive and the ion-exchange process

can be carried out in water. Scheme 2 illustrates the two

diﬀerent methodologies used to introduce TiO₂inside the

zeolite.

l_max. In order to avoid the aggregation of these small sub-

After ion exchange, hydrolysis and oligomerization of tita-

nyls (TiQO²⁺) can be carried out simply by treatment of the

TiQO²⁺containing zeolite at mild temperatures of about

150 1C. The main drawback of titanyl is the diﬃculty in

controlling the amount of incorporated titanium, particularly

for low loadings.

nanometric TiO₂clusters, one methodology that has been

developed consists in incorporating these TiO₂clusters inside

the rigid framework of microporous hosts. The most widely

used microporous materials are large pore zeolites.^67–70

Zeolites are crystalline aluminosilicates in which the frame-

work deﬁnes channels and cavities of nanometric and sub-

nanometric dimension.^71–74These empty voids are normally

termed micropores and are open to the exterior allowing mass

transfer from a solution to the internal voids. Faujasite and

zeolite beta are common large pore zeolites used for including

organic and inorganic guest inside cavities^75,76and extra large

pore tridirectional zeolite can oﬀer new possibilities. Fig. 1

shows an illustration of the pores of Faujasite, zeolite beta and

By using vapor phase adsorption of TiCl₄, Schulz-Ekloﬀ

and co-workers have prepared titanium oxide-containing zeo-

lite in which UV/Vis spectroscopic characterization suggests

that at low loadings isolated titanium atoms mono-, bi- or tri-

podally anchored to the zeolite framework are formed.⁶⁷

Higher loading can lead to aggregates in which Ti–O–Ti bonds

are present. It has been proposed that isolated titanium atoms

Fig. 1 Structure of zeolites Faujasite, beta and ITQ-33 in which the micropores are deﬁned by the framework.

ꢀc

This journal is the Owner Societies 2008

Phys. Chem. Chem. Phys., 2008, 10, 769–783 | 771

View Article Online

Scheme 2 Alternative procedures to form subnanometric TiO₂clus-

ters inside the zeolite micropores based on adsorption of compounds

containing a single titanium atom.

in a Si–O–Ti conﬁguration have an absorption l_maxat around

220 nm while clusters with Ti–O–Ti bonds absorb at in the UV

spectra at 240 and higher wavelength.⁶⁷Identical spectro-

scopic criterion has been applied to crystalline titanosilicates

that are widely used as oxidation catalysts.^84–86

In these zeolite samples containing monoatomic Ti^IV, easy

formation of Ti^IIIby reduction of Ti^IVions can occur and this

can simply be detected by a characteristic Ti^IIIabsorption

band at around 630 nm that is responsible for the blue tint of

the powdered sample.^69,87

Fig. 2 Raman spectra of (a) TiO²⁺in water; (b) TiO²⁺@Y zeolite;

(c) sample b after heating at 150 1C for 1 h; (d) TiO₂in anatase phase;

(e) TiO₂in rutile phase.

On the other hand, Thomas et al. have prepared clusters of

TiO₂in mordenite using water soluble titanyl salts.^82,83In-

corporation of titanium in zeolite by using of TiQO²⁺is an

ion exchange process in water that may require consecutive

cycles using increasing TiQO²⁺concentrations in order to

achieve high TiO₂loadings. TiO₂clusters incorporated inside

the zeolite voids exhibit photocatalytic activity for the reduc-

tion of methyl viologen.⁸²Thus, upon TiO₂@mordenite irra-

diation, the corresponding viologen radical cation is generated

and can be detected spectroscopically (Scheme 3).

considered a reﬂection of the operation of quantum size

eﬀects. Also remarkable is that the slope of the absorption

band is more vertical for highly loaded samples suggesting that

in these cases the samples have a more homogeneous size

distribution of encapsulated titanium clusters, probably cor-

responding to the volume of the zeolite cavity.

Photoluminescence is a hallmark of small nanoparticles that

is lost for TiO₂above 20 nm. In this regard it is interesting

that, in accordance with the subnanometric cluster size of the

TiO₂clusters incorporated inside zeolites, TiO₂@zeolite ex-

hibits photoluminescence that is not observed in P25.^89,90

These samples of zeolite encapsulated TiO₂show photo-

catalytic activity for the photooxidation of thianthrene

(Table 2).⁸⁸This heterocyclic compound can be considered

a model for sulfur-containing aromatic compounds present in

fuels and heavy oil fractions that are responsible for the

formation of SO_xduring combustion processes.^91,92There is

a research line to carry out the oxidation of these sulfur

heterocycles to more polar water-soluble sulfoxides and sul-

fones, not only by regular catalytic oxidation but also by

photooxidation.⁹³Importantly, Table 2 shows that the photo-

catalytic activity of zeolite encapsulated TiO₂can be higher

than P25 standard.⁸⁸This photocatalytic enhancement has

been ascribed mainly to the favorable adsorption of

Scaiano and Garcia using the same procedure described by

Thomas, have prepared a large series of TiO₂containing

zeolites in which the zeolite chemical structure and the tita-

nium loading were varied.^88,89High resolution transmission

microscopy as well as XPS (taken as analysis of the external

surface of the zeolite particles) and other characterization

techniques established that in most of the samples, titanium

was incorporated inside the zeolite micropores with only

minor occasional isolated TiO₂nanoparticles detected outside

the zeolite particles.⁸⁸By Raman spectroscopy (Fig. 2) it was

determined that the zeolite encapsulated TiO₂clusters are

present as a mixture of anatase-like and rutile-like structures.

Raman is a powerful spectroscopic technique that together

with X-ray diﬀraction serves to distinguish the anatase or

rutile phases of TiO₂.

By means of diﬀuse reﬂectance UV/Vis spectroscopy

(Fig. 3), it was observed that the TiO₂clusters incorporated

inside zeolites have an onset of the absorption band signiﬁ-

cantly blue shifted as compared to conventional P25 TiO₂

nanoparticles.⁸⁸This blue shift becomes more remarkable

when the titanium loading in the zeolite decreases and is

Fig. 3 Diﬀuse reﬂectance optical spectrum of a series of three zeolite

Y samples containing increasing titanium loadings (titanium concen-

tration increase in the order a o b o c).

Scheme 3 Generation of the methyl viologen radical cation upon

irradiation of a zeolite containing TiO₂clusters (based on ref. 82).

ꢀc

This journal is the Owner Societies 2008

772 | Phys. Chem. Chem. Phys., 2008, 10, 769–783

View Article Online

Table 2 Results of the photocatalyzed TH oxygenation upon 254 nm

UV lamp after 1 h irradiation (data taken from ref. 88)

Photocatalyst

Yield of TH = O%

NaY

2

22

21

4

TiO₂@Y (ﬁrst exchange)

TiO₂@Mordenite (ﬁrst exchange)

P25 TiO₂

thianthrene (TH) inside the zeolite micropores.

Fig. 4 Enzymatic activity of HRP (measured through its ability to

catalyze the oxidation of 2,2⁰-azinobis(3-ethylbenzothiazoline-6-sulfo-

nic acid) diammonium salt that is monitored by the absorbance at 412

nm) as a function of the irradiation time in the presence of a series of

photocatalysts. K and J HRP; m and n HRP + TiO₂@Y;EandB

HRP + TiO₂@MOR; ’ and & HRP + TiO₂anatase. Black

symbols non-irradiated and open symbols irradiated samples.

In the previous examples thianthrene, being a relatively

small molecule, can diﬀuse inside the zeolite micropores and

become oxidized in the photocatalytic process. Incorporation

of TiO₂inside a restricted reaction cavity oﬀers in addition the

possibility to increase the photoactivity for certain substrates

avoiding, at the same time, the degradation of the molecules

that do not enter into the zeolite internal pores. For example,

the same TiO₂@zeolite samples that were able to perform the

photocatalytic oxidation of TH were remarkably ineﬃcient for

the inactivation of horseradish peroxidase (HRP), an enzyme

that is able to decompose hydrogen peroxide. Fig. 4 shows the

HRP inactivation upon irradiation of solutions of this enzyme

in the presence of a series of titanium dioxide photocatalysts

including P25 and TiO₂@Y.

into the TiO₂conduction band, the Ru(bpy)₃²⁺/TiO₂couple

has properties worth exploring in very diﬀerent environments.

In the case of zeolite-encapsulated TiO₂there are several

possible topologies, including those in which the large

2+

Ru(bpy)₃

is size excluded from the zeolite interior and

adsorbed on the zeolite external surface, while the semicon-

ductor oxide is located in the internal voids.⁹⁹In the examples

2+

reported by Bossmann et al., Ru(bpy)₃

(about 1.2 nm

As can been seen in Fig. 4, TiO₂inactivates to a consider-

able extent the ability of HRP to oxidize 2,2⁰-azinobis-

(3-ethylbenzthiazoline-6-sulfonic acid), while due to the large

size of this macromolecule, TiO₂@zeolite produces only a

minor damage of the enzyme, the catalytic activity of the

native enzyme being largely preserved when this enzyme is

irradiated in the presence of TiO₂@zeolite.

diameter) is accommodated inside the cavities of zeolite Y

(spherical cages of about 1.3 nm) but it becomes immobilized

because this metallic complex cannot diﬀuse through the

smaller cavity windows (0.74 nm). This internally entrapped

Ru(bpy)₃²⁺@zeolite can be easily prepared by a ship-in-a-

bottle synthesis starting from Ru(NH₃)₄³⁺@zeolite and add-

ing 2,2⁰-bypyridine.^100–102When a sample of Ru(bpy)₃²⁺@-

This selectivity of TiO₂@zeolite to attack small molecules

can be highly beneﬁcial in those cases in which the photo-

activity of TiO₂plays an adverse eﬀect. This is the case of sun

screens for topical use in cosmetics.⁹⁴For this application in

which the titanium dioxide nanoparticles are going to be

exposed to the solar light, it is necessary to ensure that TiO₂

is avoiding any photocatalytic activity against human skin.⁹⁴

Using crystalline titanosilicate having TiO wires, Zecchina

and co-workers have coined the term ‘‘inverse shape-selective’’

photocatalysis to reﬂect the preferential photocatalytic degra-

dation of large phenols with respect to smaller phenols that

can access to the zeolite interior.⁹⁵In contrast to what is seen

in TiO₂@zeolite (photocatalysis of adsorbed molecules) in the

case of crystalline titanosilicate of the ETS family only Ti–O

wires externally placed in the crystals are photocatalytically

active.⁹⁵

zeolite is modiﬁed by TiO₂, the system contains immobilized

2+

Ru(bpy)₃

ions in close proximity to subnanometric TiO₂

cluster co-incorporated in the same or neighbouring cavity of

the material. Using the photoinduced electron transfer be-

tween zeolite entrapped Ru(bpy)₃²⁺and size-excluded Co(dp-

3+

phen)₃(dpphen = 4,7-diphenyl-1,10-phenantroline) in the

exterior of zeolite particles, Bossmann observed that the

2+

relative eﬃciency of the quenching of internal Ru(bpy)₃

3+

by external Co(dpphen)₃increases along the titanium con-

tent into the zeolite (Scheme 4).⁷⁸This observation indicates

that TiO₂nanoparticles are acting as electron relays, promot-

2+

ing the electron transport from photoexcited Ru(bpy)₃

located inside the zeolite to external Co(dphen)₃³⁺. The

bicomponent

[Ru(bpy)₃²⁺/TiO₂@zeolite]

photocatalyst

shows a high activity for the degradation of 2,4-xylidine

considered as a model pollutant of aromatic amines.¹⁰³An

analogous system also developed by this group and tested for

the photo-Fenton degradation of 2,4-xylidine has been

[Fe(bpy)₃²⁺/TiO₂]@zeolite.¹⁰⁴Given the lack of reusable

photo-Fenton catalysts, this material shows promise as a

recoverable photo-Fenton catalyst.

Following the pioneering lead of Mallouk, who wanted to

develop zeolite-based photocatalyst for water splitting,^96–98

Bossmann and co-workers were the ﬁrst to report a multi

2+

component system containing Ru(bpy)₃and TiO₂simulta-

neously incorporated inside zeolite Y.⁷⁸Due to the importance

of dye sensitized solar cells in which photoexcitation of poly-

pyridyl ruthenium complexes lead to a fast electron injection

Similar samples of Ru(bpy)₃²⁺/TiO₂@Zeolite Y have been

studied by Scaiano and Garcia by means of ﬂuorescence,

ꢀc

This journal is the Owner Societies 2008

Phys. Chem. Chem. Phys., 2008, 10, 769–783 | 773

View Article Online

Scheme 4 Pictorial illustration of photoinduced electron transfer

between encapsulated Ru(bpy)₃²⁺and external Co(dphen)₃³⁺favored

by small TiO₂clusters acting as electron relays (based on ref. 78).

Scheme 5 Illustration of zeolite Y encapsulated Ru(bpy)₃²⁺complex

and TP⁺cation in neighbouring supercages. The intrazeolite space

between the electron donor [Ru(bpy)₃²⁺] and the electron acceptor

(TP⁺) can be occupied by TiO₂or absorbed water (based on ref. 106).

steady-state and time-resolved, as well as laser ﬂash photo-

lysis.¹⁰⁵It was observed that this sample exhibits reduced

2+

and TP⁺, it has been observed that the TiO₂clusters inter-

2+

and TP⁺disfavours the interac-

emission and shorter life time from Ru(bpy)₃excited state

posed between Ru(bpy)₃

triplet and it undergoes prompt photoinduced charge separa-

tion compared to analogous Ru(bpy)₃²⁺@Y materials lacking

TiO₂. This was taken as evidence of the interaction between

Ru(bpy)₃²⁺in its excited state and co-adsorbed TiO₂clusters.

Similarly, a Y zeolite containing 2,4,6-triphenylpyrylium

(TP⁺) and TiO₂has been prepared.¹⁰⁵Fig. 5 shows a mole-

cular model of TP⁺encapsulated within the cavities of zeolite

Y. Spectroscopic characterization using time resolved photo-

luminescence and laser ﬂash photolysis also revealed an inter-

action between the light-absorbing dye in its excited state and

TiO₂clusters. Photocatalytic experiments using HRP deacti-

vation in buﬀered aqueous solution showed that deactivation

of HRP is considerably more important for bifunctional

(Ru(bpy)₃²⁺/TiO₂or TP⁺/TiO₂) systems encapsulated in

zeolites than the photocatalytic activity of each of the indivi-

dual components inside the zeolite.¹⁰⁵

tion between the donor and the acceptor, restoring the

2+

Ru(bpy)₃

emission to 90% of the original value in the

absence of TP⁺quenchers. Experimentally, the photoinduced

2+

electron transfer between excited Ru(bpy)₃

and a good

electron acceptor as TP⁺is more eﬃcient when no TiO₂is

located between these two termini. This has been interpreted

considering that the higher LUMO energy for encapsulated

TiO₂clusters, in which quantum size eﬀects increase the band

gap as compared to TiO₂nanoparticles of 25 nm, disfavors the

ability of TiO₂to act as a semiconductor. Thus, electron

injection is energetically less favourable for encapsulated

TiO₂clusters than for P25 and the injected electrons less

mobile and more trapped that in conventional large TiO₂

particles. Therefore, encapsulated TiO₂behaves, in this case,

more as insulator than as an electron relay. Overall the

previous studies show the ease of preparation of encapsulated

TiO₂clusters inside micropore hosts and the advantages in

terms of high eﬃciency and size control of the photocatalytic

activity. In this regard, Anpo and coworkers have made

substantial contributions towards the application of zeolite-

encapsulated TiO₂clusters as photocatalysts for NO_xand CO₂

ﬁxation.^58,107–116Overall, in addition to the examples of

increased photocatalytic activity, the observed harnessing of

the TiO₂photocatalytic activity and the introduction of some

kind of substrate selectivity is a subject that still requires

considerable attention to be exploited in full.

A more complex system in line with those previously

described is the one containing Ru(bpy)₃²⁺/TiO₂/TP⁺inside

Y zeolite (Scheme 5).¹⁰⁶

This is a three component system in which an electron donor

[Ru(bpy)₃²⁺] and an electron acceptor (TP⁺) are simulta-

neously immobilized within the zeolite cavities and interposed

by TiO₂acting as an electron relay. In this case using

2+

ﬂuorescence to monitor the interaction between Ru(bpy)₃

4. Nanometric structuring: mesoporous titanium

dioxide as photocatalysts

The original report from Mobil researchers describing the

synthesis of structured mesoporous silicas using the surfactant

template methodology constituted a breakthrough in material

chemistry in the 90s.^117–119Surfactant molecules, in concen-

trations above the critical micellar concentration (cmc) under-

go in aqueous media spontaneous self assembly forming

micelles. Further increases in surfactant concentration leads

to rods in which the aqueous colloidal surfactant solution can

be considered as a liquid crystal phase. Concomitant hydro-

lysis, oligomerization and polycondensation of soluble silica

Fig. 5 Molecular model of TP⁺cation occupying the cavities of

zeolite Y.

ꢀc

This journal is the Owner Societies 2008

774 | Phys. Chem. Chem. Phys., 2008, 10, 769–783

View Article Online

photocatalysis. Even more importantly and less evident, struc-

turing can aﬀect the length of the charge migration in the

charge separate state.

One of the advantages of sol–gel synthesis of mesoporous

materials is the possibility to form uniform ﬁlms on a sub-

strate. Using the method studied in detail by Sanchez

et al.,^123,124a uniform ﬁlm of mesoporous titania on glass

substrate can be obtained dipping the glass slide into an acidic

solution of titanium alcoxide in ethanol. Immediately after

dipping the glass, the surfactant concentration is lower than

the cmc, but as the glass is removed from the solution and

ethanol evaporates, the cmc point is reached and the surfac-

tant starts to template the formation of thin layers of a

mesoporous titanium oxide. The key point is to control care-

fully the rate of solvent evaporation that has to be suﬃciently

slow to allow the templation and oligomerization of the

titanium oxide around the self assembled micelles created by

the surfactant in its liquid crystal state. Applying the above

methodology, highly structured materials constituted by ana-

tase nanoparticles (5–10 nm) ordered forming a mesoporous

ﬁlm (mp-TiO₂) perpendicular to the glass slide have been

prepared by Stuky et al.¹²⁵The as-synthesized mp-TiO₂

material is initially structured in the 5–100 nm length scale

forming mesopores, but the walls are formed by an amorphous

TiO₂phase. Calcination of the material produces crystalliza-

tion of the as-synthetized amorphous titanium dioxide into

anatase phase without destroying the mesoporous ordering

of the ﬁlm. Careful control of the calcination temperature

(o550 1C) is crucial to avoid the formation of the rutile phase

that, as expected in view of its lower activity, is signiﬁcantly

less photocatalytic active than the material constituted by the

anatase phase.

Scheme 6 General formation mechanism of a MCM-41 material.

precursors occur at the water–surfactant rod interphase, the

silica polymer acquiring the shape of the rods. Initially the

surfactant rods are not rigid and fairly independent of each

other, but as the silica phase densiﬁes and grows the rods bind

together in a close packing conﬁguration. The resulting solid

after considerable oligomerization/condensation of the silicon

precursor and removal of the template is a material with

extraordinary periodicity formed by channels of strictly reg-

ular dimensions in the nanometric scale (from 2 to 10 nm).

Thus, mesoporous silicas constituted a logical extension of the

zeolite materials. Scheme 6 illustrates the mechanism of for-

mation of MCM-41 mesoporous material.

The regularity of the channels and walls in MCM-41 is so

remarkable that the material shows X-ray ordering at long

distances, even though the material does not have a short

distance order. The terms ‘‘periodic’’ or ‘‘structured’’ have

been used to denote this type of solid that are amorphous at

the subnanometric scale but exhibit order in the nanometric

length scale. Fig. 6 shows transmission electron microscopy

images in which side and frontal views of the channels are

presented. The consequence of mesoporosity is that these

materials rank at the top of the list of the large surface area

solids with values for speciﬁc internal surface and pore volume

as high as 1000 m²g^ꢁ1and 1 ml g^ꢁ1. For comparison common

values of surface area and volume in zeolite can be 400 m²g^ꢁ1

Films of structured mesoporous mp-TiO₂anatase materials

ﬁlled with furfuryl alcohol as the source of polymeric carbon

have been used as photoanodes for water photolysis using a

xenon lamp.¹²⁶Scheme 7 presents the route followed for the

preparation of these carbon containing mp-TiO₂photocata-

lysts. The maximum eﬃciency of photon-to-hydrogen conver-

sion of 2.5% was obtained with carbon doped titanium

dioxide and it was more than four times higher than analogous

mesoporous TiO₂ﬁlms prepared without furfuryl alcohol.¹²⁶

Following an analogous methodology, but impregnating the

ordered mp-TiO₂anatase ﬁlms with silver instead of carbon,

the resulting ﬁlms have been used as photocatalysts for the

oxidative degradation of stearic acid.¹²⁷A control experiment

comparing the activity with analogous mp-TiO₂ﬁlms without

silver nanoparticles has shown that the metal particles notably

increase the initial degradation rate. However at long

and 0.3 ml g^ꢁ1

.

Shortly after the original report from Mobil, there was an

interest in expanding the results initially obtained for silica and

silico-titanate to most of the transition metal oxides. This

subject has been reviewed several times.^120–122Titania is an

obvious example of metal oxide that can beneﬁt from being

prepared as a periodic mesoporous solid. For photocatalytic

applications, the large surface area and pore volume charac-

teristics of mesoporous materials will play a positive role in

Scheme 7 Preparation of a mesoporous anatase ﬁlm ﬁlled on con-

ductive carbon that has been used as photoanode for the production of

hydrogen (based on ref. 126).

Fig. 6 TEM images showing a parallel (left) and perpendicular (right)

view of the channels in MCM-41 material.

ꢀc

This journal is the Owner Societies 2008

Phys. Chem. Chem. Phys., 2008, 10, 769–783 | 775

View Article Online

Fig. 7 Transmission electron microscopy images of side (left) and

front (right) views of mp-TiO₂materials obtained using colloidal TiO₂

nanoparticles as building blocks (taken from ref. 128).

The photocatalytic activity of mp-TiO₂has been studied for

the degradation of phenol in water solutions.¹²⁸It was ob-

served that for those structured mesoporous TiO₂samples

having low titanium content, the turnover frequency (TOF,

mol of degraded phenol ꢂ mol of titania^ꢁ1ꢂ h^ꢁ1) value can be

four times higher than the standard P25. However, as the

titanium content of the material increases (mp-TiO₂with 99%

TiO₂and 1% TEOS), the TOF for phenol degradation

decreases and the material mp-TiO₂with very low silicon

content and hexagonal ordering has a catalytic activity mea-

sured by TOF about 4 times lower than P25. Since TOF

measures the activity per each titanium atom, the inﬂuence

of the titanium content in the TOF value has been interpreted

as reﬂecting the beneﬁcial inﬂuence of accessibility and isola-

tion of the TiO₂domains on the intrinsic activity of titanium

atoms.

Scheme 8 Synthetic procedure of mesoporous titania/silica materials

in which colloidal titania nanoparticles (3–5 nm) have been used as

building blocks (taken from ref. 128).

irradiation times the presence of silver seems to be detrimental

for the catalytic activity of the materials.¹²⁷As one of the

possible reasons to explain this lower photocatalytic activity

upon use, it has been proposed that poisoning or deactivation

of Ag nanoparticles by degradation intermediates may take

place. The previous two examples in which the channels of

mesoporous titania have been used to host conductive carbon

and silver nanoparticles, both species cooperating in the

photocatalytic activity, illustrates the vast potential of these

materials with respect to non-porous conventional anatase

nanoparticles. Encapsulation of guests inside mesoporous

TiO₂should allow the use of host/guest supramolecular

chemistry to increase the photocatalytic activity of TiO₂.

A diﬀerent synthetic strategy in which preformed TiO₂

nanoparticles (3–5 nm) are used as building blocks in combi-

nation with variable proportions of tetraethyl orthosilicate

(TEOS) has also been used to form structured mesoporous

materials.¹²⁸The role of TEOS is to bind the titanium dioxide

nanoparticles through Si–O–Ti bonds in such a way that, on

the one hand, the presence of TEOS holds the TiO₂nanopar-

ticles forming a rigid matrix, while on the other hand TEOS

favors the action of the surfactant templating the material.

Scheme 8 shows the procedure followed to prepare these

mesoporous materials obtained from preformed TiO₂nano-

particle building blocks.

An interesting application of photocatalysis, consists in

pursuing the synthesis of organic compounds rather than the

degradation and destruction of them. Albini has coined the

term ‘‘positive photocatalysis’’ to describe this aspect of

photocatalysis.^130,131In this regard, photoxidation of alkanes

to ketones is one of the most important reactions that can be

performed using titanium dioxide.¹³²One example of alkane

photoxidation that could have a large economical impact is the

conversion of cyclohexane into cyclohexanol and cyclohexa-

none.^132–135The key point in this reaction is to reach high

selectivity to cyclohexanone at high conversions. Although

poorly understood, it has been found that the nature and

structuring of the mp-TiO₂photocatalyst has a remarkable

inﬂuence on the selectivity towards cyclohexanone.¹³⁶Thus,

structured mp-TiO₂exhibits at irradiation wavelength longer

than 290 nm a remarkable higher selectivity towards the

formation of cyclohexanone than P25.¹³⁶This diﬀerence in

product selectivity between the mesoporous and non-meso-

porous titania disappears when the irradiation is carried out

with ﬁltering short wavelength light from the excitation ir-

radiation. Table 3 shows relevant photocatalytic data for the

photooxygenation of cyclohexane in which the higher selec-

tivity towards formation of cyclohexanone of mp-TiO₂with

respect to P25 TiO₂at the same conversion using short

wavelength irradiation can be seen.

As surfactants to template the synthesis of mp-TiO₂by

assembly of TiO₂nanoparticles either CTABr or Pluronic-

123 were used.¹²⁸After calcination, TEM images and powder

XRD showed that the materials have mesoporous ordering.

Raman spectra also indicate that the predominant TiO₂phase

is anatase. Fig. 7 shows selected TEM images corresponding to

these mesostructured mp-TiO₂titania photocatalysts contain-

ing variable proportions of silica.

In contrast to some other reports in which mixed Ti/Si

oxides exhibited a band gap energy, energy levels as well as

photocatalytic activity strongly inﬂuenced by the composition

of the mixed oxide,¹²⁹the previous mp-TiO₂materials have

separate and independent TiO₂and SiO₂domains covalently

bonded through a few Ti–O–Si linkage at the interphase. One

of these domains is formed by anatase nanoparticles for which

silica is not interfering with the band gap levels because it is

present as a separate phase.

ꢀc

This journal is the Owner Societies 2008

776 | Phys. Chem. Chem. Phys., 2008, 10, 769–783

View Article Online

Table 3 Selectivity of cyclohexane oxygenation using two titania

photocatalysts as a function of the excitation light (data taken from

ref. 136)

l excitation/nm

and light

Alcohol

Photocatalyst (%)

Ketone CO₂ꢂ 1/6

intensity/W cm^ꢁ2

(%)

89

(%)

7

4290

mpTiO₂

4

11

3

2.8 ꢂ 10^ꢁ2

4290

Scheme 10 Preparation of ﬁlms of 2D and 3D structured titania

photocatalyst (based on ref. 137).

P-25 TiO₂

mpTiO₂

81

90

8

7

9

2.8 ꢂ 10^ꢁ2

4350

7.8 ꢂ 10^ꢁ3

4350

Table 4 Results of the photocatalytic decoloration of MB adsorbed

on a series of mesoporous (hexagonal, h, or cubic, c) and non-porous

(nanocrystalline, nc) titania samples at the same surface coverage (data

taken from ref. 137)

P-25 TiO₂

1

7.8 ꢂ 10^ꢁ3

It has been proposed that the diﬀerence in product selectiv-

ity arises from the better absorption of peroxyl radicals on the

mesoporous surface that favors its consecutive reduction by

trapped electrons.¹³²Compared to amorphous non-porous

titania, the mesoporous photocatalyst must adsorb the peroxyl

radical stronger and does not desorb this species into the

solution (Scheme 9).

Apparent constant

rate solid

Annealing

Crystallinity

temperature/1C of anatase (%) state/k min^ꢁ1

h-meso-TiO₂400

h-meso-TiO₂450

c-mesoꢁTiO₂400

c-mesoꢁTiO₂450

69

86

84

98

0.07

0.10

0.09

0.13

0.07

nc-TiO₂

450

100

Ozin and coworkers have also prepared ﬁlms of mesoporous

mp-TiO₂on glass.¹³⁷Pluronic-123 (P-123) was used as a

structure directing agent and 1-butanol as the solvent. By

careful control of the preparation conditions, particularly

temperature and relative humidity, two mesoporous mp-

TiO₂ﬁlms having 2D hexagonal (h-mp-TiO₂) and a 3D cubic

(c-mp-TiO₂) structure were obtained (Scheme 10).

In contrast to this measurement in the absence of solvent,

the kinetics of the degradation of 5 mM MB aqueous solution

is signiﬁcantly more complex due to the combination of many

factors including titania crystallinity, surface area, diﬀusion

coeﬃcients, adsorption, etc.¹³⁷Thus, for the degradation of

aqueous solutions it was found that annealing at 400 1C

provides the material with the optimum activity. Overall the

above results establish that the porous structure aﬀects posi-

tively the photocatalytic activity compared to conventional

non-porous titania. It has to be noted, however, that the

degradation of MB using titania photocatalyst constitutes a

speciﬁc mechanistic case in which light initiating photodegra-

dation is absorbed by the dye rather than by the titania. For

many other cases, the substrate does not absorb light and it

has to be titania which should be excited by direct light

absorption.

The channels of h-mp-TiO₂run perpendicularly to the

substrate and therefore they should be accessible for substrate

incorporation. After preparation, the ﬁlms were submitted to

three diﬀerent annealing temperatures up to 459 1C, crystal-

linity increasing with the annealing temperature. Thus, all the

h-mp-TiO₂samples consisted of variable proportions of

anatase nanoparticles in a matrix of amorphous titania.

The photocatalytic activity of these ﬁlms was tested for the

degradation of methylene blue (MB), the photocatalytic reac-

tion being followed by optical spectroscopy.¹³⁷For the photo-

catalytic degradation in dry conditions, a constant surface

coverage of 0.6 was selected. A summary of the results is

shown in Table 4. It was observed that the photocatalytic

activity of mesoporous materials annealed at 450 1C is over 1.5

times higher than the analogous ﬁlm constituted by 100%

anatase prepared in the absence of P-123 (nc-TiO₂in Table 4).

It was concluded from these solid state experiments that the

degree of crystallinity was the main factor governing the

photocatalytic activity and that the morphology of the meso-

pores does not play any role. This conclusion is based on

experiments in which all the samples have been studied at the

same surface coverage and, therefore, mesoporous samples

with higher surface area than non-porous titania contain a

higher loading of MB.

A further development of mesoporous titania as photoca-

talyst has been reported by Li, Lu and coworkers.¹³⁸These

authors have prepared a photocatalyst constituted by meso-

porous titania embedding gold nanoparticles (Au/mp-TiO₂).

Its preparation requires P-123 as structure directing agent, a

mixture of TiCl₄and Ti(OBu)₄and AuCl₃as the source of

gold using ethanol as solvent. The gel is cast on a Petri dish to

form a thin layer that is subsequently aged at 100 1C to form a

homogeneous mesostructured nanocomposite. Calcination at

350 1C in air removes the template while inducing crystal-

lization of TiO₂and formation of gold nanoparticles. Scheme

11 presents the procedure followed to obtain this mesoporous

titania containing gold.

Scheme 9 Proposed mechanism to justify the higher selectivity towards cyclohexanone when radicals remain adsorbed on the TiO₂surface.

ꢀc

This journal is the Owner Societies 2008

Phys. Chem. Chem. Phys., 2008, 10, 769–783 | 777

View Article Online

Table 5 Surface area and photophysical data obtained by laser ﬂash

photolysis of NT-TiO₂and other TiO₂materials

BET surface area/m²g^ꢁ1

f_cs

t_1/2/ms

a

TiO₂

NT-TiO₂-400

Standard TiO₂-1

Standard TiO₂-2

Standard TiO₂-3

a

225

300

50

2.0

7.1

4.8

1.6

3.5 + 0.4

0.6 + 0.2

1.0 + 0.2

0.7 + 0.2

Scheme 11 Preparation of Au/mp-TiO₂(based on ref. 138).

10

Standard TiO₂corresponding to commercial samples of spherical

shaped nanoparticles.

This Au/mp-TiO₂material was tested for the photocatalytic

degradation of phenol and for the reduction of chromium-

(^VI).¹³⁸The results show that the presence of gold nanoparti-

cles at 0.5 wt% increase the photocatalytic activity about 1.5

and 3.5 times for phenol oxidation and chromium(^VI) reduc-

tion, respectively, as compared to analogous mesoporous

titania material lacking gold. It would also be of interest to

have a valid comparison of the photocatalytic activity of these

materials with that of standard P25.

The consequence of the low dimensionality of NT-TiO₂can

also be revealed by steady-state diﬀuse-reﬂectance spectro-

scopy after UV irradiation (365 nm) of a suspension of

nanotubes in aerated acetonitrile containing MPTM in which

the photogeneration of an increase concentration of trapped

electrons is clearly observed (Scheme 12). These trapped

electrons can serve to degrade CCl₄. Importantly, none of

the photocatalysts constituted by nanoparticulate powder

exhibit a similar behavior.¹⁴³Thus, these studies nicely ex-

emplify the advantage of having TiO₂spatially organized as

nanotubes instead of independent nanoparticles.

5. From nanometric to micrometric structuring:

titania nanotubes as photocatalyst

Titania materials of 1D dimensionality such as nanotubes,

nanoﬁbers and nanowires have also attracted attention for

their use in photocatalysis.^139–142In particular titania nano-

tubes being tens to hundreds of nanometers in diameter and

micrometric length have been the subject of intensive study.

Titania nanotubes have a relatively high surface area com-

pared to non-porous titania and time-resolved diﬀuse-reﬂec-

tance spectroscopy has shown that charge recombination is

disfavored by the tubular morphology of the titania. These

titania nanotubes (NT-TiO₂) can be conveniently obtained

starting from titania nanoparticles as, for instance P-25, that

are digested under strong basic conditions in an autoclave at

about 150 1C for several hours.¹⁴³Annealing of these NT-TiO₂

at 400 1C for 3 h renders nanotubes that are composed 100%

by anatase. Laser ﬂash photolysis of the NT-TiO₂compared

with conventional titania nanoparticles has allowed to esti-

mate apparent quantum yield of charge separation (f_cs).

Table 5 lists some of the f_csvalues measured for NT-TiO₂

compared with diﬀerent conventional spherical nanoparticles

powders.

Notably the half life of the photogenerated hole (h⁺) is

signiﬁcantly longer for NT-TiO₂as compared to nanoparti-

cles.¹⁴³Furthermore when 4-(methylthio)phenylmethanol

(MTPM) was added to the slurry as a hole quencher, the

characteristic absorption band of the MTPM⁺radical cation

(around 550 nm) was observed.¹⁴³This MTPM⁺adsorbed on

NT-TiO₂exhibits remarkably longer half-lives compared to

the situation in which MTPM⁺is adsorbed on nanoparticles.

This has been attributed to a longer diﬀusion length of the

charge carriers (4200 nm) as a consequence of the low

dimensionality of the nanotubes. Analogously, a higher diﬀu-

sion length of charge carriers has also been claimed by

Yanagida et al. to justify the higher eﬃciency of electron

transport in NT-TiO₂electrodes compared to analogous

electrodes prepared with nanoparticles.⁴²Fig. 8 illustrates

the proposal to justify the long lifetime of charge separation

in titania nanotubes.

Compared to the use of independent TiO₂nanotubes, it is

even more interesting to have an ordered array of these

nanotubes. Thus, a surface of NT-TiO₂(0.8 cm²) in which

they are ordered perpendicularly to the substrate has been

obtained by anodization of a titanium foil in an electrolyte

consisting in 1 M NH₄H₂PO₄+ 0.5 wt% NH₄F and applying

a voltage of 15 V.¹⁴⁴The surface of these TiO₂nanotubes

(70 nm diameter and 20 nm walls thickness) could be doped

with carbon by thermal treatment between 500 and 800 1C

under controlled CO gas ﬂow.¹⁴⁴The resulting carbon doped

NT-TiO₂exhibits higher photocurrent density and more eﬃ-

cient photocatalytic water splitting under visible light illumi-

nation (l 4 420 nm) than undoped NT-TiO₂and a total

photocurrent more that 20 times higher than that of a ﬁlm of

P25 nanoparticles. Thus, these measurements parallel to those

that have been made with ﬁlms of carbon-doped mp-TiO2 as

photoanodes can cause water splitting.

Analogously, Schmuki and coworkers have prepared large

surfaces of ordered NT-TiO₂of various lengths from 0.5 to

4.5 mm.¹⁴⁵The deepest ﬁlms have tubes of 45 nm diameter,

while a common diameter for most of the ﬁlms is 100 nm.

After preparation by anodization, the nanotubular ﬁlms were

annealed at 450 1C, a calcination temperature that was found

Fig. 8 Pictorial representation of the advantages derived from the use

of titania nanotubes. The aspect ratio favors the photocatalytic

activity.

ꢀc

This journal is the Owner Societies 2008

778 | Phys. Chem. Chem. Phys., 2008, 10, 769–783

View Article Online

light absorption can be extended into the visible region. As

commented on in the Introduction, doping and metal incor-

poration are typical examples of this approach. In addition to

this methodology based on chemistry, one physical approach

could be to increase the optimum light path, producing the

entrapment of the light into the material. This methodology

will promote light absorption, particularly for photons on the

red side of the absorption inset.

Light absorption by the photocatalyst produces charge

separation (Scheme 13). In contrast, those wavelengths that

have lower energy than the band gap cannot be absorbed and

therefore they will be reﬂected scattered and/or transmitted by

the powder. However, on top of this necessary condition

imposed by the relative energy of photons and band gap, it

exists a probability associated with the electron transition that

is related to the speciﬁc absorptivity of the material at a

speciﬁc wavelength. In other words, energetics is a prerequisite

for light absorption but other considerations related to the

probability of the electronic transition also need to be con-

sidered. Photonic crystals serve to promote light absorption by

increasing the eﬀective light path through the material.^146–151

In this way, photonic crystals act by increasing the apparent

speciﬁc absorption coeﬃcient of the material for those wave-

lengths matching the dimensions of the crystal.

Scheme 12 Hole trapping by MTPM and simultaneous formation of

an oxidation product (based on ref. 143).

to be the optimum to induce complete crystallization of the

NT-TiO₂into the anatase phase without traces of rutile. It was

found that the presence of rutile is remarkably detrimental for

the overall photocatalytic activity of the solid. Using as probe

molecules one azo dye like Acid Orange 7 (AO7) and one non

azo dye like MB, the photocatalytic activity of this surface of

ordered arrays of NT-TiO₂was established and compared to

that of a ﬁlm of independent P25 nanoparticles.¹⁴⁵It was

observed that the photocatalytic activity of the P25 ﬁlm was

signiﬁcantly lower than that of analogous ﬁlms of NT-TiO₂.¹⁴⁵

Moreover the photocatalytic activity of the ordered NT-TiO₂

layers increases with the layer thickness.¹⁴⁵Although the

authors suggest that the higher eﬃciency of the ordered tubes,

as compared to P25 ﬁlm, is due to an optimized photocatalyst

geometry that favors substrate diﬀusion and diminishes the

required charge carriers diﬀusion path, no experimental evi-

dence supporting this claim was given. Moreover, the proposal

that the UV light is absorbed exclusively by the photocatalyst

in the region in which the dye exhibits an absorption minimum

is questionable. Normally one would have expected that

photocatalysis is initiated by light absorption from the dye

followed by electron injection from the dye in its excited state

into the semiconductor conduction band.

In a photonic crystal the space is ordered in such a way that

the light that enters in the crystal is trapped inside by a

multitude of reﬂections and diﬀractions making light escape

impossible. In order to produce this eﬀect, the dimensions of

the periodic dielectric constant ﬁeld in the photonic crystal

have to be commensurate with the wavelength of the trapped

light. For this reason, photonic crystals have to be made of

ﬁlms having void dimensions between 0.3–0.8 mm correspond-

ing to any of the visible wavelengths spatially ordered in a

regular way expanding to the millimetric length scale. Thus,

photonic crystals are based on the spatial structuring of titania

in the submillimetric length scale.

Since the trapped light can not escape out of the crystal the

probability that light absorption and charge separation occurs

is considerably favored particularly at the onset of the absorp-

tion band where the speciﬁc absorption of the material is very

low. Instead of allowing the photon to escape from the

material, the photonic crystal arrangement increases the prob-

ability of electronic excitation by increasing the light/photo-

catalyst contact time. The fundamental of a photonic crystal is

the structuring of the space creating a periodic dielectric

constant ﬁeld in the length scale of the visible light wave-

length.^{150,152–154}This spatial structuring produces coherent

Bragg diﬀraction patterns that forbid light of the correspond-

ing wavelength to propagate through the material. At the

Bragg diﬀraction frequencies, photons propagate with

strongly reduced velocity so they are called slow photons. If

Although, the number of examples reporting the speciﬁc

features of photocatalyst having micrometric NT-TiO₂mor-

phology are still quite limited and many more examples are

desirable, from the existing data it can be anticipated that the

potential of these structures in photocatalysis is very large and

promising in order to enhance the photocatalytic activity just

by changing the shape and dimensions of the titania particles

from conventionally spherical.

6. Titania structuring in the submillimetric length

scale: photonic crystals as photocatalyst

Because less than a few percent of the solar photons can be

absorbed by powdered anatase photocatalyst, considerable

eﬀorts have been made from a chemical point of view trying

to modify the composition of TiO₂in such a way that anatase

Scheme 13 Application of the photonic crystal concept in an inverse

opal conﬁguration to increase the eﬀective light path through the

photocatalyst increasing the probability of charge separation.

ꢀc

This journal is the Owner Societies 2008

Phys. Chem. Chem. Phys., 2008, 10, 769–783 | 779

View Article Online

the energy of the slow photons overlaps with the absorption of

the material, (even with very weak absorption) then the

enhancement of the light absorption occurs as a consequence

of the increase of the eﬀective optical path length through the

material.

Scheme 14 Schematic representation of the defects in the photonic

crystal made of latex spheres induced by the introduction of larger

spheres in a perfect opal ﬁlm (based on ref. 156).

Ozin and coworkers have used the physical approach to

slow down photons and increase the photocatalytic activity of

TiO₂.¹⁵⁵Thus, these authors prepared an inverse opal con-

stituted by anatase nanocrystals ordered around monodisperse

empty spheres of dimensions from 280 to 500 nm. The

dimension of the sphere creates a light stop band at the

wavelength corresponding to the dimensions of the spheres.

To test the photocatalytic activity enhancement of this inverse

photonic crystal, MB was used as a probe in the absence of

solvent.¹⁵⁵Studies with monochromatic light but also with

polychromatic visible light were carried out to determine the

inﬂuence of the void diameter in the red or in the blue side of

the anatase optical absorption. Comparative photocatalytic

experiments with unstructured TiO₂ﬁlms were carried out at

similar surface coverage for which the absence of dye multi-

layer or aggregates can be ensured. For monochromatic

370 nm irradiation, the inﬂuence of the void dimensions of

the inverse opals can lead to a decrease in the photocatalytic

activity due to light reﬂection (for ﬁlms with a void size of

370 nm) or to an enhancement of the photocatalytic activity

when 370 nm corresponds to a somewhat longer length than

the void size of the opal (ﬁlms with void size of 340 nm).

Furthermore, the enhancement factor of the photocatalytic

eﬃciency depends on the angle of the incident light following

the expected relationship for the Bragg equation.¹⁵⁵For the

polychromatic white light irradiation, it was also observed that

the optimum void dimension to obtain the maximum enhance-

ment of the photocatalytic activity was 300 nm for which an

enhancement factor of 2.3 was determined for ﬁlms with

photonic crystal structuring with respect to unstructured

TiO₂nanoparticles. Interestingly, the photocatalytic activity

of a mesostructured TiO₂ﬁlm prepared by demolition of an

inverse opal matrix crashing several samples and using the

resulting powder to prepare a non-structured ﬁlm was the

same as an analogous ﬁlm made of conventional TiO₂nano-

particles.¹⁵⁵This indicates that the intrinsic photocatalytic

activity of the TiO₂particles forming the inverse opal structure

is the one that should be expected for conventional TiO₂.

The practicality of inverse opal ﬁlms constituted by anatase

nanoparticles as an eﬃcient photocatalyst depends upon the

level of structural disorder that can be tolerated while main-

taining the enhancement eﬀect. Since structuring, dislocation

and other defects are inherent to a 3D crystal, it is of interest to

determine through a systematic study the inﬂuence of the

degree of disorder on the photocatalytic activity enhancement.

This study has been carried out by Ozin and coworkers using

the photodegradation of MB as a probe to test the photo-

catalytic eﬃciency of titania inverse opal having some disor-

der.¹⁵⁶The disorder in the crystal was purposely introduced by

preparing ﬁlms substituting in the system diﬀerent proportions

of 150 nm spheres by spheres of 180 or 210 nm (Scheme 14).

Comparing the photocatalytic activity of the ﬁlms prepared

with monodisperse 150 nm latex spheres, the enhancement of

the photocatalytic activity decreases gradually with the struc-

tural defect and a point is reached in which no advantages with

respect to the unstructured ﬁlm of conventional monocrystal-

line TiO₂is achieved. For a ﬁlm of polystyrene opal template,

it was observed that doping with larger spheres introduces

defects that destroy the long range order much more signiﬁ-

cantly than when doping with smaller spheres.¹⁵⁶In general,

the band of the inverse opal structure corresponding to the

Bragg diﬀraction is much more sensitive to defects than that of

the polystyrene templates, probably because disorder of the

template is magniﬁed during the replica of the template to

form inverse opal.

Nevertheless, a photodegradation enhancement of about

50% respect to unstructured ﬁlms can be easily achieved even

for inverse opals having a considerable degree of disorder.¹⁵⁶

In any case, defects produced by the presence of larger spheres

(210 nm as compared to 180 nm) inﬂuence strongly reducing

the photocatalytic activity much more drastically than defects

introduced by spheres with sizes more similar to those con-

stituting the perfect opal.

Besides the opal structure, Meseguer et al. have expanded

the concept to the photonic sponge.¹⁵⁷In this case instead of

having a mononodal distribution of spherical voids of a single

diameter, the sponge has an appropriate distribution of dif-

ferent size spheres in such a way that the entire wavelength

range in a particular spectral region can be trapped.¹⁵⁷This

means that, in contrast to the classical photonic crystal, not

only one particular wavelength, but most of the visible spectra

can be trapped inside a photonic sponge. In this way, the

photonic sponge should show enhanced photocatalytic activ-

ity no matter which excitation wavelengths or substrates are

going to be used. These sponges are particularly useful for

polychromatic light irradiation of colorless substrates for

which TiO₂should absorb photons.

A photonic sponge can be obtained by impregnation with

P25 nanoparticles of a ﬁlm constituted by a mixture of

polystyrene latex spheres of diﬀerent size in a proportion that

compensate smaller versus larger diameters.¹⁵⁸This propor-

tion follows a geometry in which smaller spheres occupy the

corners and empty spaces left among larger spheres in a close

packing conﬁguration.

The photocatalytic activity of a photonic sponge has been

tested using succinonitrile as substrate and following the

degradation course by IR spectroscopy.¹⁵⁸Fig. 9 shows the

decrease in the absorption intensity of the characteristic CH₂

stretching vibration band of succinonitrile at about 2990 cm^ꢁ1

upon increase of the irradiation time.

Comparison with an analogous unstructured ﬁlm made by

P25 nanoparticles indicates that structuring the photonic

sponge increases the photocatalytic activity by one order of

magnitude. One interesting observation was that an increase

of the ﬁlm thickness of the photonic sponge can be

ꢀc

This journal is the Owner Societies 2008

780 | Phys. Chem. Chem. Phys., 2008, 10, 769–783

View Article Online

Acknowledgements

Authors thank CICYT (Project MAT2006-14274-C02-01) and

(MAT20070157) for ﬁnancial supports. Carmela Aprile

thanks the Spanish Ministry of Education for a research

associate Juan de la Cierva contract.

References

1 M. Kitano, M. Matsuoka, M. Ueshima and M. Anpo, Appl.

Catal., A, 2007, 325, 1.

2 H. B. Thu, M. Karkmaz, E. Puzenat, C. Guillard and J.-M.

Herrmann, Research on Chemical Intermediates, 2005, 31, 449.

3 M. Schiavello, Heterogeneous Photocatalysis, Kluwer Academic

Publishers, Dordrecht, 1997.

4 M. Schiavello, NATO ASI Series, Series C: Mathematical and

Physical Sciences, 1988, 237, 351.

Fig. 9 Intensity decrease along the irradiation time of the CH₂band

characteristic of succinonitrile incorporated inside a TiO₂photonic

sponge (taken from ref. 158).

5 O. Legrini, E. Oliveros and A. M. Braun, Chem. Rev., 1993, 93,

671.

6 M. A. Fox and M. T. Dulay, Chem. Rev., 1993, 93, 341.

7 D. F. Ollis and H. Al-Ekabi, in Photocatalytic Puriﬁcation and

Treatment of Water and Air, ed. D. F. Ellis and H. Al-Ekabi,

Elsevier, Amsterdam, 1993.

8 O. M. Alfano, D. Bahnemann, A. E. Cassano, R. Dillert and R.

Goslich, Catal. Today, 2000, 58, 199.

unfavorable for the photocatalytic activity of the ﬁlm. This

was interpreted assuming that for thick photonic sponge ﬁlms

the light is not able to penetrate through the whole layer and

therefore the substrate located deep inside the layer can not be

degraded due to the lack of light at this depth. Thus, there is

an optimum larger thickness of about a few microns.

9 D. Bahnemann, Handbook of Environmental Chemistry, 1999, 2,

285.

10 D. Bahnemann, Sol. Energ., 2004, 77, 445.

11 D. Bahnemann, J. Cunningham, M. A. Fox, E. Pelizzetti, P.

Pichat and N. Serpone, Aquat. Surf. Photochem., 1994, 261.

12 J. M. Herrmann, Catal. Today, 1995, 24, 157.

13 J. M. Herrmann, Catal. Today, 1999, 53, 115.

14 J. M. Herrmann, Catal. Sci. Ser., 1999, 1, 171.

15 M. R. Hoﬀmann, S. T. Martin, W. Choi and D. W. Bahnemann,

Chem. Rev., 1995, 95, 69.

16 A. L. Linsebigler, G. Lu and J. T. Yates, Jr, Chem. Rev., 1995, 95,

735.

17 P. Pichat, C. Guillard, C. Maillard, L. Amalric and J. C.

D’Oliveira, Trace Met. Environm., 1993, 3, 207.

18 A. Wold, Chem. Mater., 1993, 5, 280.

7. Conclusions and future prospects

Improvement of the photocatalytic activity of TiO₂is abso-

lutely necessary to implement real life application of photo-

catalysis. While the vast majority of the current approaches to

eﬀect the enhancement of the photocatalytic activity is based

on modiﬁcation of the chemical composition introducing

various degrees of doping (chemical approaches), in the pre-

sent account we have shown that control of the particle size

and spatial organization of TiO₂nanoparticles (physical ap-

proach) is also a powerful methodology to improve the

photocatalytic eﬃciency. Recognition of this fact among the

photocatalytic community will be of large importance to

accelerate further development in this area. In this sense the

independent development of the physical and chemical ap-

proaches to boost the photocatalytic activity is clear at the

present stage. It can be easily anticipated that the near future

developments will combine all the possible methodologies. In

this regard it is remarkable the scarcity of the number of

studies on spatial structuring of TiO₂that use doped or

chemically modiﬁed titania. Doping and spatial structuring

combined together can produce a new generation of materials

with suﬃciently high photocatalytic eﬃciency under visible

and solar light irradiation so that commercial implementation

of photocatalysis can be brought to fruition.

19 J. Cunningham, G. Al-Sayyed, P. Sedlak and J. Caﬀrey, Catal.

Today, 1999, 53, 145.

20 H. Yamashita, M. Takeuchi and M. Anpo, Encyclopedia of

Nanoscience and Nanotechnology, 2004, 10, 639.

21 H. Kisch and W. Macyk, ChemPhysChem, 2002, 3, 399.

22 R. Bacsa, J. Kiwi, T. Ohno, P. Albers and V. Nadtochenko, J.

Phys. Chem. B, 2005, 109, 5994.

23 K. E. Karakitsou and X. E. Verykios, J. Phys. Chem., 1993, 97,

1184.

24 J. Kiwi and M. Gratzel, J. Phys. Chem., 1986, 90, 637.

25 J. Kiwi and C. Morrison, J. Phys. Chem., 1984, 88, 6146.

26 Z. Jin, X. Zhang, Y. Li, S. Li and G. Lu, Catal. Commun., 2007, 8,

1267.

27 O. Ozcan, F. Yukruk, E. U. Akkaya and D. Uner, Appl. Catal.,

B, 2007, 71, 291.

28 Z. Jin, X. Zhang, G. Lu and S. Li, J. Mol. Catal. A: Chem., 2006,

259, 275.

29 R. Abe, K. Sayama and H. Arakawa, J. Photochem. Photobiol.,

A, 2004, 166, 115.

30 L.-C. Chen, C.-M. Huang and F.-R. Tsai, J. Mol. Catal. A:

Chem., 2007, 265, 133.

31 F. Fresno, C. Guillard, J. M. Coronado, J.-M. Chovelon, D.

Tudela, J. Soria and J.-M. Herrmann, J. Photochem. Photobiol.,

A, 2005, 173, 13.

32 L. Ge and M. Xu, Mater. Sci. Eng., B, 2006, 131, 222.

33 N. Serpone, J. Phys. Chem. B, 2006, 110, 24287.

34 J. C. Colmenares, M. A. Aramendia, A. Marinas, J. M. Marinas

and F. J. Urbano, Appl. Catal., A, 2006, 306, 120.

35 S. Rengaraj and X. Z. Li, J. Mol. Catal. A: Chem., 2006, 243, 60.

36 W. Choi, A. Termin and M. R. Hoﬀmann, J. Phys. Chem., 1994,

98, 13669.

Also, it can be expected that host–guest supramolecular

chemistry concepts will be applied on top of spatial structuring

and chemical doping. Thus, one further degree of complexity

will involve systems in which a guest is present inside a porous

titania with appropriate shape, dimensions and spatial order-

ing to promote the photocatalytic eﬃciency of the semicon-

ductor by photosensitization and electron/hole injection. In

this case, the guest has to be persistent enough to be stable

during the operation of the photocatalytic system.

37 W. Choi, A. Termin and M. R. Hoﬀmann, Angew. Chem., 1994,

106, 1148.

ꢀc

This journal is the Owner Societies 2008

Phys. Chem. Chem. Phys., 2008, 10, 769–783 | 781

View Article Online

38 C. Bauer, G. Boschloo, E. Mukhtar and A. Hagfeldt, J. Phys.

Chem. B, 2002, 106, 12693.

39 J. He, J. Zhao, T. Shen, H. Hidaka and N. Serpone, J. Phys.

Chem. B, 1997, 101, 9027.

76 (a) A. Corma, M. J. Diaz-Cabanas, F. Rey, S. Nicolopoulus and

K. Boulahya, Chem. Commun., 2004, 12, 1356; (b) A. Corma, M.

J. Diaz-Cabanas, J. L. Jorda, C. Martinez and Manuel Moliner,

Nature, 2006, 443, 842.

40 P. V. Kamat, S. Das, K. G. Thomas and M. V. George, Chem.

Phys. Lett., 1991, 178, 75.

41 D. I. Kreller and P. V. Kamat, J. Phys. Chem., 1991, 95,

4406.

77 A. Corma and H. Garcia, Chem. Commun., 2004, 1443.

78 S. H. Bossmann, C. Turro, C. Schnabel, M. R. Pokhrel, L. M.

Payawan, Jr, B. Baumeister and M. Woerner, J. Phys. Chem. B,

2001, 105, 5374.

42 Y. Ohsaki, N. Masaki, T. Kitamura, Y. Wada, T. Okamoto, T.

Sekino, K. Niihara and S. Yanagida, Phys. Chem. Chem. Phys.,

2005, 7, 4157.

79 T. Kapias and R. F. Griﬃths, J. Hazard. Mater., 2005, 119, 41.

80 P. K. Roy, A. Bhatt and C. Rajagopal, J. Hazard. Mater., 2003,

102, 167.

43 J. E. Kroeze, T. J. Savenije and J. M. Warman, J. Am. Chem.

Soc., 2004, 126, 7608.

81 E. H. Fochtman, C. Swanstrom and T. Cashen, Environ. Prog.,

1988, 7, 112.

44 A. Kay, R. Humphry-Baker and M. Graetzel, J. Phys. Chem.,

1994, 98, 952.

82 X. Liu, K. K. Iu and J. K. Thomas, J. Chem. Soc., Faraday

Trans., 1993, 89, 1861.

45 Y.-J. Lin, S.-L. Tseng, W.-J. Huang and W.-J. Wu, J. Environ.

Sci. Health, Part B, 2006, 41, 1143.

83 X. Liu, K. K. Iu and J. K. Thomas, Chemical. Phys. Lett., 1992,

195, 163.

46 T. Okato, T. Sakano and M. Obara, Phys. Rev. B: Condens.

Matter Mater. Phys., 2005, 72, 115124/1.

47 R. Asahi, T. Morikawa, T. Ohwaki, K. Aoki and Y. Taga,

Science, 2001, 293, 269.

84 A. Corma and H. Garcia, Chem. Rev., 2002, 102, 3837.

85 A. Zecchina, S. Bordiga, C. Lamberti, G. Ricchiardi, C. Lam-

berti, G. Ricchiardi, D. Scarano, G. Petrini, G. Leofanti and M.

Mantegazza, Catal. Today, 1996, 32, 97.

48 Y. Cong, F. Chen, J. Zhang and M. Anpo, Chem. Lett., 2006, 35,

800.

86 F. Geobaldo, S. Bordiga, A. Zecchina, E. Giamello, G. Leofanti

and G. Petrini, Catal. Lett., 1992, 16, 109.

49 C. Di Valentin, G. Pacchioni and A. Selloni, Chem. Mater., 2005,

17, 6656.

50 D. Li, H. Haneda, N. K. Labhsetwar, S. Hishita and N. Ohashi,

Chem. Phys. Lett., 2005, 401, 579.

51 T. Ohno, T. Tsubota, M. Toyofuku and R. Inaba, Catal. Lett.,

2004, 98, 255.

52 S. Sakthivel, M. Janczarek and H. Kisch, J. Phys. Chem. B, 2004,

108, 19384.

87 M. Wark, G. Grubert, M. Warnken, G. Schulz-Ekloﬀ, M.

Ganschow, Y. Rohlﬁng, T. Bogdahn-Rai and D. Wohrle, Appl.

Mineral., 2000, 1, 253.

88 G. Cosa, M. S. Galletero, L. Fernandez, F. Marquez, H. Garcia

and J. C. Scaiano, New J. Chem., 2002, 26, 1448.

89 S. Corrent, G. Cosa, J. C. Scaiano, M. S. Galletero, M. Alvaro

and H. Garcia, Chem. Mater., 2001, 13, 715.

90 S. Zhang, N. Fujii and Y. Nosaka, J. Mol. Catal. A: Chem., 1998,

129, 219.

91 T. V. Choudhary, Ind. Eng. Chem. Res., DOI: 10.1021/ie061617d.

92 C. Song, Catal. Today, 2003, 86, 211.

53 T. Tachikawa, S. Tojo, K. Kawai, M. Endo, M. Fujitsuka, T.

Ohno, K. Nishijima, Z. Miyamoto and T. Majima, J. Phys.

Chem. B, 2004, 108, 19299.

54 M. Mrowetz, W. Balcerski, A. J. Colussi and M. R. Hoﬀmann, J.

Phys. Chem. B, 2004, 108, 17269.

55 H. Liu and L. Gao, J. Am. Ceram. Soc., 2004, 87, 1582.

56 S. Sakthivel and H. Kisch, ChemPhysChem, 2003, 4, 487.

57 S.-Z. Chu, S. Inoue, K. Wada, D. Li and J. Suzuki, Langmuir,

2005, 21, 8035.

58 M. Anpo, H. Yamashita, Y. Ichihashi, Y. Fujii and M. Honda, J.

Phys. Chem. B, 1997, 101, 2632.

59 G. Lassaletta, A. Fernandez, J. P. Espinos and A. R. Gonzalez-

Elipe, J. Phys. Chem., 1995, 99, 1484.

60 C. Kormann, D. W. Bahnemann and M. R. Hoﬀmann, J. Phys.

Chem., 1988, 92, 5196.

61 S. H. Tolbert, A. B. Herhold, C. S. Johnson and A. P. Alivisatos,

Phys. Rev. Lett., 1994, 73, 3266.

62 A. Henglein, Chem. Rev., 1989, 89, 1861.

63 M. Alvaro, E. Carbonell, V. Fornes and H. Garcia, Chem-

PhysChem, 2006, 7, 200.

93 (a) A. Chica, A. Corma and M. E. Domine, J. Catal., 2006, 242,

299; (b) M. Alvaro, E. Carbonell and H. Garcia, Appl. Catal., B,

2004, 51, 195.

94 M. N. Chretien, Pure Appl. Chem., 2007, 79, 1.

95 F. X. Llabres i Xamena, P. Calza, C. Lamberti, C. Prestipino, A.

Damin, S. Bordiga, E. Pelizzetti and A. Zecchina, J. Am. Chem.

Soc., 2003, 125, 2264.

96 Y. I. Kim, S. W. Keller, J. S. Krueger, E. H. Yonemoto, G. B.

Saupe and T. E. Mallouk, J. Phys. Chem. B, 1997, 101, 2491.

97 Y. I. Kim, R. L. Riley, M. J. Huq, S. Salim, A. N. Le and T. E.

Mallouk, Mater. Res. Soc. Symp. Proceed., 1991, 233, 145.

98 J. S. Krueger, C. Lai, Z. Li, J. E. Mayer and T. E. Mallouk, J.

Inclusion Phenom. Mol. Recognit., 1990, 365.

99 H. Garcia and H. D. Roth, Chem. Rev., 2002, 102, 3947.

100 P. K. Dutta and W. Turbeville, J. Phys. Chem., 1992, 96, 9410.

101 W. Turbeville, D. S. Robins and P. K. Dutta, J. Phys. Chem.,

1992, 96, 5024.

64 D. C. Hurum, K. A. Gray, T. Rajh and M. C. Thurnauer, J. Phys.

Chem. B, 2005, 109, 977.

65 D. C. Hurum, A. G. Agrios, K. A. Gray, T. Rajh and M. C.

Thurnauer, J. Phys. Chem. B, 2003, 107, 4545.

66 Z. Ding, G. Q. Lu and P. F. Greenﬁeld, J. Phys. Chem. B, 2000,

104, 4815.

67 J. Klaas, G. Schulz-Ekloﬀ and N. I. Jaeger, J. Phys. Chem. B,

1997, 101, 1305.

68 X. S. Zhao, G. Q. Lu and G. J. Millar, J. Porous Mater., 1996, 3,

61.

102 J. A. Incavo and P. K. Dutta, J. Phys. Chem., 1990, 94, 3075.

103 S. H. Bossmann, S. Jockusch, P. Schwarz, B. Baumeister, S.

Goeb, C. Schnabel, L. Payawan, Jr, M. R. Pokhrel, M. Woerner,

A. M. Braun and N. J. Turro, Photochem. Photobiol. Sci., 2003, 2,

477.

104 S. H. Bossmann, N. Shahin, H. Le Thanh, A. Bonﬁll, M. Worner

and A. M. Braun, ChemPhysChem, 2002, 3, 401.

105 G. Cosa, M. N. Chretien, M. S. Galletero, V. Fornes, H. Garcia

and J. C. Scaiano, J. Phys. Chem. B, 2002, 106, 2460.

106 M. Alvaro, M. N. Chretien, V. Fornes, M. S. Galletero, H.

Garcia and J. C. Scaiano, J. Phys. Chem. B, 2004, 108,

16621.

69 G. Grubert, M. Stockenhuber, O. P. Tkachenko and M. Wark,

Chem. Mater., 2002, 14, 2458.

70 G. Grubert, M. Wark, N. I. Jaeger, G. Schulz-Ekloﬀ and O. P.

Tkachenko, J. Phys. Chem. B, 1998, 102, 1665.

71 R. M. Barrer, Zeolites and Clay Minerals as Sorbents and

Molecular Sieves, Academic Press, London, 1978.

72 D. W. Breck, Zeolite Molecular Sieves: Structure, Chemistry, and

Use, Wiley-Interscience, New York, 1974.

107 K. Ikeue, H. Yamashita and M. Anpo, Electrochemistry, 2002,

70, 402.

108 H. Yamashita, K. Ikeue and M. Anpo, ACS Symp. Ser., 2002,

809, 330.

109 J. Zhang, Y. Hu, M. Matsuoka, H. Yamashita, M. Minagawa, H.

Hidaka and M. Anpo, J. Phys. Chem. B, 2001, 105, 8395.

110 K. Ikeue, H. Yamashita, M. Anpo and T. Takewaki, J. Phys.

Chem. B, 2001, 105, 8350.

73 A. Corma, F. Rey, J. Rius, M. J. Sabater and S. Valencia, Nature,

2004, 431, 287.

74 B. W. Wojciechowski and A. Corma, Catalytic Cracking: Cata-

lysts, Chemistry, and Kinetics, Khimiya, Moscow, 1990.

75 L. B. McCusker and C. Baerlocher, Stud. Surf. Sci. Catal., 2001,

137, 37.

111 K. Ikeue, H. Yamashita and M. Anpo, Chem. Lett., 1999,

1135.

112 M. Anpo, S. Higashimoto and M. Matsuoka, Eco Ind., 1999, 4,

11.

ꢀc

This journal is the Owner Societies 2008

782 | Phys. Chem. Chem. Phys., 2008, 10, 769–783

View Article Online

113 H. Yamashita, Y. Fujii, Y. Ichihashi, S. G. Zhang, K. Ikeue, D.

R. Park, K. Koyano, T. Tatsumi and M. Anpo, Catal. Today,

1998, 45, 221.

133 A. Maldotti, A. Molinari, G. Varani, M. Lenarda, L. Storaro, F.

Bigi, R. Maggi, A. Mazzacani and G. Sartori, J. Catal., 2002, 209,

210.

114 M. Anpo, H. Yamashita, K. Ikeue, Y. Fujii, Y. Ichihashi, S. G.

Zhang, D. R. Park, S. Ehara, S.-E. Park, J.-S. Chang and J. W.

Yoo, Stud. Surf. Sci. Catal., 1998, 114, 177.

115 H. Yamashita, Y. Ichihashi, S. G. Zhang, Y. Matsumura, Y.

Souma, T. Tatsumi and M. Anpo, Appl. Surf. Sci., 1997, 121/122,

305.

134 A. Molinari, R. Amadelli, L. Andreotti and A. Maldotti, J. Chem.

Soc., Dalton Trans., 1999, 1203.

135 P. Boarini, V. Carassiti, A. Maldotti and R. Amadelli, Langmuir,

1998, 14, 2080.

136 A. Maldotti, A. Molinari, E. Carbonell and H. Garcia, unpub-

lished results.

116 H. Yamashita, Y. Ichihashi, M. Anpo, M. Hashimoto, C. Louis

and M. Che, J. Phys. Chem., 1996, 100, 16041.

117 J. C. Vartuli, C. T. Kresge, M. E. Leonowicz, A. S. Chu, S. B.

McCullen, I. D. Johnson and E. W. Sheppard, Chem. Mater.,

1994, 6, 2070.

137 M. A. Carreon, S. Y. Choi, M. Mamak, N. Chopra and G. A.

Ozin, J. Mater. Chem., 2007, 17, 82.

138 H. Li, Z. Bian, J. Zhu, Y. Huo, H. Li and Y. Lu, J. Am. Chem.

Soc., 2007, 129, 4538.

139 J. T. McCann, M. Marquez and Y. Xia, Nano Lett., 2006, 6, 2868.

140 Z. Zhong, T.-P. Ang, J. Luo, H.-C. Gan and A. Gedanken, Chem.

Mater., 2005, 17, 6814.

118 C. T. Kresge, M. E. Leonowicz, W. J. Roth, J. C. Vartuli and J. S.

Beck, Nature, 1992, 359, 710.

119 J. S. Beck, J. C. Vartuli, W. J. Roth, M. E. Leonowicz, C. T.

Kresge, K. D. Schmitt, C. T. W. Chu, D. H. Olson, E. W.

Sheppard, S. B. McCullen, J. B. Higgins and J. L. Schenkler,

J. Am. Chem. Soc., 1992, 114, 10834.

141 C. Xiong and K. J. Balkus, Jr, Chem. Mater., 2005, 17, 5136.

142 Z.-Y. Yuan and B.-L. Su, Colloids Surf., A, 2004, 241, 173.

143 T. Tachikawa, M. Fujitsuka and T. Majima, J. Phys. Chem. C,

2007, 111, 5259.

120 C. Boissiere, D. Grosso, B. Smarsly, T. Brezesinski, S. Lepoutre,

L. Nicole, J. C. V. Marcos, M. Antonietti and C. Sanchez, Mater.

Res. Soc. Symp. Proceed., 2005, 847, 135.

144 J. H. Park, S. Kim and A. J. Bard, Nano Lett., 2006, 6, 24.

145 J. M. Macak, M. Zlamal, J. Krysa and P. Schmuki, Small, 2007,

3, 300.

121 S. Polarz and M. Antonietti, Chem. Commun., 2002, 2593.

122 C. G. Goltner and M. Antonietti, Adv. Mater., 1997, 9,

431.

123 E. L. Crepaldi, G. J. de Soler-Illia, D. Grosso, F. Cagnol, F.

Ribot and C. Sanchez, J. Am. Chem. Soc., 2003, 125, 9770.

124 D. Grosso, G. J. d. A. A. Soler-Illia, E. L. Crepaldi, F. Cagnol, C.

Sinturel, A. Bourgeois, A. Brunet-Bruneau, H. Amenitsch, P. A.

Albouy and C. Sanchez, Chem. Mater., 2003, 15, 4562.

125 M. H. Bartl, S. W. Boettcher, K. L. Frindell and G. D. Stucky,

Acc. Chem. Res., 2005, 38, 263.

126 J. Tang, Y. Wu, E. W. McFarland and G. D. Stucky, Chem.

Commun., 2004, 1670.

127 M. Andersson, H. Birkedal, N. R. Franklin, T. Ostomel, S.

Boettcher, A. E. C. Palmqvist and G. D. Stucky, Chem. Mater.,

2005, 17, 1409.

146 T. Baba, Nat. Photonics, 2007, 1, 11.

147 D. J. Norris, Nat. Mater., 2007, 6, 177.

148 D. A. B. Miller, Nat. Mater., 2006, 5, 83.

149 M. Egen, R. Zentel, P. Ferrand, S. Eiden, G. Maret and F.

Caruso, Photonic Cryst., 2004, 109.

150 N. Tetreault, H. Miguez and G. A. Ozin, Adv. Mater., 2004, 16,

1471.

151 N. Hall, Chem. Commun., 2003, 2639.

152 A. Arsenault, F. Fleischhaker, G. von Freymann, V. Kitaev, H.

Miguez, A. Mihi, N. Tetreault, E. Vekris, I. Manners, S. Aitch-

ison, D. Perovic and G. A. Ozin, Adv. Mater., 2006, 18, 2779.

153 M. Wegener, K. Arnold, K. Busch, M. Deubel, C. Enkrich, D.

Fenske, G. von Freymann, M. Hermatschweller, S. John, M.

Kappes, A. Kaso, S. Linden, G. A. Ozin, S. Pereira, F. Perez-

Willard, C. M. Soukoulis, N. Tetreault and S. Wong, PMSE

Preprints, 2006, 94, 97.

128 M. Alvaro, C. Aprile, M. Benitez, E. Carbonell and H. Garcia, J.

Phys. Chem. B, 2006, 110, 6661.

129 M. Gaertner, V. Dremov, P. Mueller and H. Kisch, Chem-

PhysChem, 2005, 6, 714.

154 G. A. Ozin and S. M. Yang, Adv. Funct. Mater., 2001, 11, 95.

155 J. I. L. Chen, G. von Freymann, S. Y. Choi, V. Kitaev and G. A.

Ozin, Adv. Mater., 2006, 18, 1915.

130 M. Fagnoni, D. Dondi, D. Ravelli and A. Albini, Chem. Rev.,

2007, 107, 2725.

156 J. I. L. Chen, G. von Freymann, V. Kitaev and G. A. Ozin, J. Am.

Chem. Soc., 2007, 129, 1196.

131 L. Cermenati, C. Richter and A. Albini, Chem. Commun., 1998,

805.

132 A. Maldotti, A. Molinari and R. Amadelli, Chem. Rev., 2002,

102, 3811.

157 F. Ramiro-Manzano, P. Atienzar, I. Rodriguez, F. Meseguer, H.

Garcia and A. Corma, Chem. Commun., 2007, 242.

158 E. Carbonell, F. Ramiro-Manzano, F. Meseguer, A. Corma and

H. Garcia, unpublished results.

ꢀc

This journal is the Owner Societies 2008

Phys. Chem. Chem. Phys., 2008, 10, 769–783 | 783

Article Doi

Enhancement of the photocatalytic activity of TiO2 through spatial structuring and particle size control: From subnanometric to submillimetric length scale

DOI: 10.1039/b712168g

Source and publish data:

Authors:

Article abstract of DOI:10.1039/b712168g

Full text of DOI:10.1039/b712168g

Products guided by the article

R&D Labs maybe for 108-93-0

Relevant to this article

Hot Product