Angewandte
Chemie
DOI: 10.1002/anie.201201222
Photochemistry with Visible Light
Molecular Oxygen as a Redox Catalyst in Intramolecular
Photocycloadditions of Coumarins**
Darius Paul Kranz, Axel Georg Griesbeck,* Ronald Alle, Raul Perez-Ruiz,
Jçrg Martin Neudçrfl, Klaus Meerholz, and Hans-Gꢀnther Schmalz*
Nature uses sunlight as a source of energy for various
chemical transformations by exploiting suitable chromo-
phores or light-absorbing photocatalysts. Taking natural
processes as a role model, various new photochemical trans-
formations are currently being developed especially targeting
applications in photovoltaics, solar energy storage, and water
splitting.[1] Moreover, photocatalysis enjoys increasing use in
organic synthesis,[2] most prominently in [2+2] cycloadditions
of enones,[3] in photoredox-mediated enantioselective orga-
nocatalysis,[4] and in the photoreductive generation of radicals
from alkyl bromides and chlorides.[5] Most common are metal-
based visible-light photocatalysts such as the readily acces-
sible complex [Ru(bpy)3Cl2] (bpy = 2,2’-bipyridyl).[6]
However, after a full day of irradiation only very little
conversion of 1 to a new (isomeric) product (rac-2) could be
detected by GC–MS analysis (Table 1, entry 1). By chance, we
Table 1: Intramolecular [2+2] photocycloaddition of the 3-substituted
coumarin derivative 1 under different conditions demonstrating an
accelerating effect of oxygen.
Entry
Conditions[a]
t [days]
1/rac-2[b]
Photochemical [2+2] cycloadditions are of importance
because cyclobutanes represent valuable synthetic intermedi-
ates and are also found as substructures in many bioactive
natural products.[7] The intramolecular [2+2] cycloaddition of
4-substituted coumarins was extensively studied by Bach and
co-workers, who even succeeded in performing such reactions
in an enantioselective fashion using chiral templates or Lewis
acids.[8]
Recently, we became interested in the intramolecular
[2+2] cycloaddition of coumarins bearing an unsaturated
alkyl substituent in the 3-position (such as 1), and we realized
that virtually no examples for this type of transformation have
been reported in the literature.[9] Herein we describe the
results of a study which has led to the discovery that such
transformations can be performed efficiently using visible
light in the presence of oxygen as a redox catalyst. Moreover,
the unique role of molecular oxygen as a promoter of these
(nonoxidative) photochemical transformations was proven by
spectroscopic and electrochemical methods.
1
2
3
4
5
6
degassed under Ar
under Ar
under air
1
4
4
4
4
4
95:5
59:41
10:90
5:95
5:95
55:45
under O2
under O2 + Rose B[c]
under Ar + Rose B[c]
[a] A solution of 1 was irradiated with a 150 W HQI lamp (white light) at
room temperature. [b] The ratio 1/rac-2 (conversion) was determined by
1H NMR spectroscopy. [c] Reaction performed in the presence of
1 mol% of Rose Bengal (Rose B.).
found that the formation of rac-2 was significantly faster when
a non-degassed solution of 1 (in the same solvent) was
employed (Table 1, entry 2). To probe whether oxygen might
promote the photocycloaddition, a third experiment (Table 1,
entry 3) was conducted using a solution of 1 flushed with air.
And indeed, after 4 days 90% conversion of 1 to rac-2 was
observed, which could even be improved by using pure
oxygen gas (Table 1, entry 4). The very clean reaction
afforded the product rac-2 as a single diastereomer (NMR
analysis) the structure of which was unambiguously proven by
X-ray crystal structure analysis (Figure 1). Two additional
experiments (Table 1, entries 5 and 6) showed that the
addition of 1 mol% of Rose Bengal (as a common singlet-
oxygen sensitizer) had at best very little effect on the reaction
outcome both in the presence and absence of oxygen gas.
In a second series of experiments the irradiation of 1 with
white light was investigated in the presence of oxygen and
different additives using a standard irradiation time of 24 h.
The results shown in Table 2 indicated that the presence of
Rose Bengal or other common sensitizers such as TPP
(tetraphenylporphyrin), or electron acceptors such as DCA
(9,10-dicyanoanthracene), DNB (1,4-dinitrobenzene), or
para-benzoquinone (p-BQ) did not result in any increase of
product formation. However, the addition of 5 mol% of the
Our study started with an attempt to synthesize the
cyclobutane rac-2 by irradiating a degassed dichloromethane
solution of the coumarin 1[10] using a 150 W HQI lamp.
[*] Dipl.-Chem. D. P. Kranz, Prof. Dr. A. G. Griesbeck, Dr. R. Alle,
Dr. J. M. Neudçrfl, Prof. Dr. K. Meerholz, Prof. Dr. H.-G. Schmalz
Department fꢀr Chemie, Universitꢁt zu Kçln
Greinstrasse 4, 50939 Kçln (Germany)
E-mail: griesbeck@uni-koeln.de
Dr. R. Perez-Ruiz
Departamento de Quimica
Universidad Politecnica de Valencia (Spain)
[**] This work was supported by the Universitꢁt zu Kçln and the Fonds
der Chemischen Industrie. We thank Dr. S. Neufeind and
Dipl.-Chem. A. M. Heinsch fꢀr stimulating discussions.
Supporting information for this article is available on the WWW
Angew. Chem. Int. Ed. 2012, 51, 1 – 6
ꢀ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
1
These are not the final page numbers!