simulated full beam sunlight. The activity of TaON (Vis) was
found to be 19.3 times higher than that of TiO2 (P25) (UV + Vis)
after 60 min of reaction. The batch mode activity of TaON
photocatalyst reached a plateau at 100 min, since most of the
available O2 in the batch reactor was consumed to form CO2,
within this time. Separate experiments were carried out to ascertain
the involvement of TaON in the reaction. Fig. 1(B) shows typical
data from one such run, where the reaction was allowed to
continue for 500 min and the TaON generated 101 6 1023 mol of
CO2 per gram of catalyst, by the photooxidation of methanol in
water. One gram of TaON corresponds to only 4.7 6 1023 mol.
Therefore, the turnover number recorded within 500 min works
out to be 21.5 (mol(CO2) per mol(TaON)). This is an indication
that the observed methanol oxidation is indeed catalytic and that
TaON is sufficiently stable for photocatalytic applications. Studies
to evaluate the long-term stability of TaON as a catalyst and
methods to immobilize it on solid transparent substrates8 are
currently progressing.
micropores could act as a reservoir for organic molecules, and
the pore volume of TaON originates from mesopores, which can
efficiently transport the reactant and product molecules between
the inside and the outside regions of particles.9 An efficient
transport of this nature should improve the catalytic activity of the
material. Often inefficient molecular transport leads to poor
catalytic activity.10
One of the major factors affecting the activity of a photocatalyst
is its BET surface area, because, in general, the surface interactions
yield sorption isotherms of a Langmuirian nature.1 A photo-
catalyst with large surface area generally shows higher catalytic
activity, provided all other reaction conditions are comparable and
well controlled. However, since the surface area of TaON was
observed to be ten times smaller than that of TiO2 (P25, ca. 55 m2
g21), some other factors must be favorably influencing the activity
of TaON. The first reason is attributed to the larger number of
absorbable photons in the visible light of the simulated sunlight.
The second reason could be its meso–microporous structure: the
micropores could seize the reactant molecules like a molecular
sieve and the mesopores transport them easily to the numerous
active sites available on the catalyst surface.9 Removal of the
reaction products from the active sites is also facile, when the
catalyst is sufficiently porous. Since it was reported that small
pores (10 nm) hindered the transport of molecules (I2),11 the
smaller micropores are believed to be contributing only very little
to the molecular transport. The movement of molecules would be
smooth through the mesopores. The third reason for the relatively
better activity of TaON photocatalyst may be that the quantum
yield of electron–hole recombination is smaller in this new catalyst,
when compared to TiO2. In TiO2, the quantum yield of electron–
hole recombination is substantially larger than that of radical
generation.12 This aspect needs more detailed studies.
Fig. 2 shows the pore-size distribution of Ta2O5 and TaON. The
diameter of channels inside TaON, which were observed as holes
on its surface in SEM images (Fig. 3(B)),6 has in fact a size
distribution between 10 and 100 nm (Fig. 2(A)) with a distribution
maximum centered at 28.6 nm. As the surface of Ta2O5 is smooth
without many pores (Fig. 3(A)), NH3 could only act slowly on the
surface and transform the oxide into an oxynitride. The reaction
proceeds from Ta2O5 to TaON through an intermediate, with an
accompanying molecular volume change, causing pore generation.
As the porosity becomes very high, the material becomes friable
and breaks down to smaller particles (Fig. 3(B)). This mechanism
seems credible when the enlarged surface area of the oxynitride is
also considered. Ta2O5 and TaON showed BET areas of 0.691 and
4.64 m2 g21, respectively. The enhanced surface area of TaON
comes from the newly formed micropores (,2 nm). These
In summary, TaON showed superior photocatalytic activity for
oxidation of methanol under simulated sunlight. To our knowl-
edge, for this reaction, no other visible-light-driven photocatalyst
has surpassed the activity of the benchmark TiO2 (P25) catalyst
under full beam sunlight irradiation. The photocatalytic activity of
this meso-microporous TaON was 19.3 times higher than that of
TiO2 (P25) under the simulated sunlight. The reaction over TaON
has been shown to be photocatalytic and not stoichiometric. This
TaON photocatalyst may become useful in processes related to the
purification of contaminated water. In future, other meso–
microporous nitride materials may become useful materials in
photocatalysis. A reduction in the level of Pt loading or finding an
alternative to Pt will be helpful. Future research will focus on this
aim.
Fig. 2 Pore-size distribution of Ta2O5 and TaON: (A) BJH pore volume
distribution (dV/dD) and (B) pore area distribution (dA/dD).
Funding was provided by the Swiss confederation (CTI/Top
Nano 21: Project No. 5904.1), for a collaborative study with Christ
AG (Switzerland). The authors thank Dr Gvadecak, Mr B. Senior,
Dr G. Rothenberger and Dr R. Bashyam (EPFL) for TEM, SEM,
RDS and XPS, respectively. We also thank Mr Kuze (Graduate
School of Materials Science, Nara Institute of Science and
Technology, Japan) for XRD and Mr Kandavelu (EPFL) for
his help in the laboratory.
Seigo Ito, K. Ravindranathan Thampi,* Pascal Comte, Paul Liska and
Michael Gra¨tzel
Laboratory of Photoniques and Interfaces (LPI), Institute des Sciences
et Ingenierie Chemiques (ISIC), Ecole Polytechnique Fe´de´rale de
Lausanne (EPFL), CH-1015, Lausanne, Switzerland.
Fig. 3 SEM of Ta2O5 (A) and TaON (B).
This journal is ß The Royal Society of Chemistry 2005
Chem. Commun., 2005, 268–270 | 269
Ito, Seigo
