intermediate in the photocatalytic system. Indeed, in one oxidation
?
route, photogenerated OH radicals can abstract formyl hydrogens
1 does not decrease significantly during the 8 h of the experiment;
(iii) the carbamate 2 is obtained with about 85% selectivity.
In this work, p-nitrotoluene has been chosen as a substrate
because, in the given experimental conditions, it undergoes neither
direct photochemical reactions, nor facile redox processes of the
para-substituent methyl group. These requirements set the limits to
the choice of other nitroarenes, and appropriate experimental
conditions for further investigations should comply with these
criteria. Thus, for example, p-nitrobenzaldehyde does not appear
to be a suitable substrate since we verified that both functional
14
from acetaldehyde yielding acyl radicals (first step in eqn. (4)).
These, in turn, are well known to undergo very fast decarbonyla-
tion according to the second step in eqn. (4). Gas chromatographic
analysis provided evidence that addition of CO did not affect
significantly yield and selectivity of the photocatalytic conversion
of 1 to 2, suggesting that the photogenerated CO, if it is involved,
is likely to participate in the carbonylation process as an adsorbed
species derived, in turn, from the adsorbed aldehyde.
groups undergo reduction at photoexcited TiO
2
, thus severely
Some additional experiments indicate that nature and reactivity
of the employed hole-scavenger had a pronounced influence on the
selectivity of the photocatalytic process. In fact, with MeOH as
photooxidizable substrate instead of EtOH, the amine 3 was the
only observable photoproduct of 1. Other experiments showed
that also in the presence of 2-PrOH the photoinduced reduction of
complicating the overall process.
In conclusion, this work describes optimal conditions in which
the reactive species generated both from the conduction-band
electrons and from the valence-band positive holes are precursors
of carbamate formation. The described reaction is a new appealing
synthetic route in applied organic chemistry, since it may produce
very important chemicals such as carbamates with good selectivity
and without using harmful reagents such as phosgene and CO. In
particular, we achieve the goal of generating carbonylating species
in situ from EtOH in mild temperature and pressure conditions
using an inexpensive photocatalyst that combines unique char-
acteristics, such as stability and environmental tolerance. In order
to rationalize the drastic influence of alcohol on the activity of
1
proceeded mainly to amine formation: 85% of 3, 15% of 2. A
plausible explanation is that both MeOH and 2-PrOH produce
lower amounts of CO in comparison to EtOH. Indeed, hydrogen
atom abstraction from formaldehyde (from MeOH oxidation) is
comparatively more difficult than from acetaldehyde (eqn. (4), step
1). Moreover, further oxidation of acetone (from 2-PrOH) can
3
2?
COCH radicals which are known to yield substantial
lead to CH
amounts of ketene (CH
15
2
CO) instead of CO .
2
TiO , more experiments are needed.
This completely new photochemical activity of TiO2 is of
general interest in the field of photocatalysis. In fact, we definitely
confirm that suitable experimental conditions can be found in
ð4Þ
2
order to control the well known powerful reactivity of TiO which,
therefore, can be successfully employed also for synthetic purposes
in addition to environmental decontamination. Further studies are
also in progress to establish the possible employment of other
semiconducting compounds able to utilize the visible light for the
exploitation of solar light. Future developments will also have to
take into account the effect of other substituents on the nitroarene.
The peculiar reactivity of EtOH has prompted us to carry on
photocatalytic experiments in neat EtOH. It is seen in Fig. 1 that
the formation of both 2 and 3 follows a zero-order kinetics,
indicating that consecutive reactions can be ruled out. It is
noteworthy that the use of neat ethanol as dispersing medium
enhances both efficiency and selectivity of the photocatalytic
process. In particular, it is seen that: (i) the overall rate of
conversion of 1 (about 13%) almost doubles with respect to the
experiments carried out in mixed solvent; (ii) the conversion rate of
a
Andrea Maldotti,* Rossano Amadelli, Luca Samiolo,
a
a
b
a
b
Alessandra Molinari, Andrea Penoni, Stefano Tollari and
c
Sergio Cenini
Dipartimento di Chimica, Sezione ISOF –CNR, Universit a` di Ferrara,
a
Dipartimento di Scienze Chimiche e Ambientali, Universita`
b
dell’Insubria, Via Valleggio 11, 22100, Como, Italy
Dipartimento di Chimica Inorganica Metallorganica e Analitica,
c
I.S.T.M.- C.N.R., Universit a` di Milano, Via G. Venezian 21, 20133,
Milano, Italy
Notes and references
1
R. H. Richter, R. D. Priester, Jr, in Kirk-Othmer Encyclopedia of
Chemical Technology, ed. J. I. Kroschwitz and M. Howe-Grant, (Eds.),
vol 14, Wiley, New York, 1995, p. 905.
2
S. Cenini and F. Ragaini, Catalytic Reductive Carbonylation of Organic
Nitro Compounds, Kluwer Academic Publishers, Dordrecht, 1997.
F. Paul, Coord. Chem. Rev., 2000, 20, 269.
A. M. Tafesh and J. Weguny, Chem. Rev., 1996, 96, 2035.
F. Ragaini, C. Cognolato, M. Gasperini and S. Cenini, Angew. Chem.
Int. Ed., 2003, 42, 2886.
3
4
5
6
7
A. Maldotti, A. Molinari and R. Amadelli, Chem. Rev., 2002, 102, 3811.
P. Paola, C. Minero, M. Vincenti and E. Pelizzetti, Catal. Today, 1997,
39, 187.
8
9
A. Maldotti, L. Andreotti, A. Molinari, S. Tollari, A. Penoni and
S. Cenini, J. Photochem. Photobiol., A, 2000, 133, 129.
P. Pichat, M. N. Mozzanega and H Courbon, J. Chem. Soc., Farady
Trans. 1, 1987, 83, 697.
Fig. 1 Formation of 2 and 3 upon irradiation (l . 380 nm) of TiO
22
2
23
)
2
3
(
3 g dm ) dispersed in deaerated p-nitrotoluene (3.0 6 10 mol dm
containing EtOH.
1
750 | Chem. Commun., 2005, 1749–1751
This journal is ß The Royal Society of Chemistry 2005