Remarkable enhancement of photo-allylation of aromatic carbonyl compounds with a hypervalent allylsilicon reagent by donor molecules

Article abstract of DOI:10.1016/j.tetlet.2006.11.057

Photo-allylation of various aromatic carbonyl compounds with a penta-coordinated allylsiliconate reagent was remarkably accelerated by the addition of a donor molecule. As the oxidation potential of the allylsiliconate was significantly decreased in the p

Full text of DOI:10.1016/j.tetlet.2006.11.057

Tetrahedron Letters 48 (2007) 211–214
Remarkable enhancement of photo-allylation of aromatic
carbonyl compounds with a hypervalent allylsilicon reagent
by donor molecules
*
Yutaka Nishigaichi, Akira Suzuki and Akio Takuwa
Department of Material Science, Faculty of Science and Engineering, Shimane University, 1060 Nishikawatsu-cho,
Matsue, Shimane 690-8504, Japan
Received 25 September 2006; revised 8 November 2006; accepted 10 November 2006
Abstract—Photo-allylation of various aromatic carbonyl compounds with a penta-coordinated allylsiliconate reagent was remark-
ably accelerated by the addition of a donor molecule. As the oxidation potential of the allylsiliconate was significantly decreased in
the presence of a donor molecule, the more efficient photo-induced electron transfer from the allylsiliconate to the excited substrate
was enabled by the hexa-coordination of the silicon atom.
Ó 2006 Elsevier Ltd. All rights reserved.
5
Nucleophilic allylation of carbonyl compounds is an
important process in organic synthesis. Many organo-
metallic methods have been developed on the basis of
thermal reactions; classically, anionic and thus strongly
basic reagents have been widely used and, recently, less
basic ones have been employed with the aid of Lewis
sponding allyltin reagents, its reactivity was far from
6
satisfaction: in the photoreaction of benzil 2, allylbisca-
techolatosiliconate 1 gave only 22% of the allyl adduct 3
(Scheme 1 and Table 1, entry 2) while allyltrimethyltin
afforded a nearly quantitative yield. Thus, we have been
5
making an effort to improve the reactivity.
1
acid activation to afford various selectivities. In
4
contrast to such thermal allylations, photochemical
carbonyl-allylation has been much less popular though
As mentioned earlier, the present photo-allylation is
considered to proceed via photo-induced single electron
transfer, which is known to be preferred in a polar sol-
vent due to the stabilization of the resulting ion radicals
2
its reactivity and selectivity are very characteristic. In
most cases, allylic tin reagents have been utilized due
to their high reactivity in the photo-induced electron
transfer (PET) process. However, the toxic drawback
of organotin reagents makes the corresponding silicon
reagents preferable instead, while their reported use is
very limited for their lower reactivity. One successful
example is photo-allylation of naphthaldehydes acti-
7
by the solvation. Therefore, methanol was attempted as
N
7
a more polar solvent (ET in Table 1) than acetonitrile
for the photoreaction between allylsiliconate 1 and
benzil 2 and, indeed, found to give a much higher yield
of the allyl adduct (Table 1, entry 3) than acetonitrile
(entry 2). In contrast, less polar solvent such as benzene
did not improve the yield of the adduct along with
the formation of unidentified complex by-products
(entry 1). As another polar solvent, DMF was also
8
9
3
vated by the metal ion coordination.
4
As another promising example, our previous report has
N
revealed that allylsilicon reagents can be activated to-
ward photo-induced allylation of carbonyl compounds
by the hypercoordination of the silicon atom. Though
the reaction mode was very similar to that of the corre-
employed, which has a similar ET value to acetonitrile.
The result was unexpectedly successful; the adduct was
obtained in a quantitative yield (entry 4). These results
suggested that the solvent polarity might not be the
dominant reason for the enhancement of the photo-
allylation.
Keywords: Allylation; Hypervalent elements; Photochemistry; Silicon
and compounds; Solvents and solvent effects.
Accordingly, we next paid attention to the donicity
(Lewis basicity) of the solvent. As shown in Table 1 in
*
Corresponding author. E-mail: nishigai@riko.shimane-u.ac.jp
0040-4039/$ - see front matter Ó 2006 Elsevier Ltd. All rights reserved.
doi:10.1016/j.tetlet.2006.11.057

212
Y. Nishigaichi et al. / Tetrahedron Letters 48 (2007) 211–214
Scheme 1.
a
Table 1. Solvent effect on the allyl adduct (3) formation
N
N
b
Entry
Solvent
/CH
E
DN
Yield of 3 (%)
Recovered 2 (%)
T
b
1
2
3
4
C
6
H
6
3
CN (8/2)
0.111
0.460
0.762
0.404
0.00
0.36
0.49
0.69
19
22
28
64
6
c
CH
CH
3
CN
OH
3
80
99
DMF
0
a
b
c
Light was irradiated for 7 h.
Values are for benzene alone.
Taken from Ref. 4.
N
T
4
comparison to the E value, the solvent donicity,
1.12 V (vs SCE) in acetonitrile. When DMF or DMSO
was added, EOX was decreased to 1.08 or 1.06 V, respec-
tively. More donating amines further decreased the
value to 0.78–0.80 V, which is even lower than that of
allyltrimethyltin (1.06 V). These results apparently
indicate that the added donor molecule facilitates the
electron transfer from 1 to the excited benzil 2.
N 10
DN , appears more consistent with the corresponding
yield. This fact indicated that the electron-donation of
the solvent promoted the photo-allylation. Along this
line, it was likely that a donor molecule would also be
able to activate the photo-allylation even in a small
amount but not as a solvent. As summarized in Table
2
, to obtain a mechanistic insight, various donor mole-
cules were added in only an equivalent amount to the
reaction mixture in acetonitrile. Interestingly, as the
DN increased, the photoreaction exhibited a higher
Therefore, a reaction path is proposed as shown in
Scheme 2. The hexa-coordinated allylsiliconate species
1 is supposed as a reacting intermediate, though such
N
0
yield of the allyl adduct except for triethylamine. Meth-
anol and DMF, which worked well as a solvent, were
not effective as an additive of one equivalent (entries 1
and 2). But 6 equiv of DMF afforded much increased
yield (entry 2). More donating DMSO was more effec-
tive as an additive (entry 3). When nitrogen donors,
which are even more donating, were employed, more
than 70% of the allyl adduct was obtained as shown in
entries 4–7. Against these results, triethylamine, which
has the highest DN among the attempted additives,
showed a very poor yield (entry 8).
highly coordinated species could not be observed spec-
troscopically. In addition, it was reported that the ther-
1
2
mal reaction of penta-coordinated allylsiliconates such
as 1 toward aldehydes proceeded via a cyclic transition
state with a hexa-coordinated silicon center. Further-
more, that the thermal reaction of 1 was disturbed in a
1
2a
donor solvent
strongly indicates the formation of a
hexa-coordinated silicon species from 1 and the donor
solvent. According to this scheme, the inconsistency of
triethylamine can be explained by its oxidation poten-
tial. It showed the lower value (0.74 V) than 1, while
the other donor molecules showed the values higher
than 0.98 V. Thus, uncoordinated triethylamine itself
underwent the preferential electron transfer over 1,
which resulted in poor formation of the allyl adduct.
This additive effect was also supported by the measure-
1
1
ment of the oxidation potential EOX of 1, which is also
shown in Table 2. Without any additive, 1 showed
a
3
Table 2. Additive effect of donor molecules on the allyl adduct formation in CH CN
b
N
c
d
d
Entry
Donor molecule
DN
E
OX of 1 (V)
Yield of 3 (%)
Recovered 2 (%)
1
2
3
4
5
CH OH
DMF
DMSO
Pyridine
Imidazole
n-Butylamine
Piperidine
Triethylamine
3
0.49
0.69
0.77
0.85
—
1.08
1.03
1.57
—
11 (22)
38 (80)
52
70
79
82
79
9
80 (63)
61 (19)
45
30
16
18
11
62
1.08
1.06
0.80
0.80
0.79
0.78
—
e
6
7
e
8
a
b
c
Irradiation time was 7 h unless otherwise noted.
An equimolar amount of the additive toward 1 was used.
See Ref. 11.
The values in the parentheses are for the reactions with 6 equiv of an additive.
Irradiation time was 4 h.
d
e

Y. Nishigaichi et al. / Tetrahedron Letters 48 (2007) 211–214
213
Scheme 2.
Since a donating solvent such as DMF or methanol
remarkably enhanced the photo-allylation as mentioned
above in Table 1, this method was applied to other aro-
matic carbonyl compounds. The results are collected in
Table 3 compared with those in acetonitrile. It should be
noted that donating solvents exhibited better results
than acetonitrile in every case here but DMF was not
necessarily the superior solvent for the substrates other
than 2 because not only donicity but also polarity of
the solvent should affect the reaction efficiency. As
shown in entries 1 and 2, quinones, which are good
electron acceptors, gave high yields of the corresponding
adduct in methanol or DMF. Unsymmetrical 1,2-dione
in entry 3 also underwent remarkable enhancement of
the photo-allylation in methanol. It is interesting that
the benzoyl selectivity was observed as previously re-
ported for the photo-allylation by the allyltin reagent.
This supports the radical mechanism of the present
photo-allylation.
1
3
Mono-carbonyl compounds shown in entries 4–6 also
experienced significant enhancement in a donating sol-
vent, where they afforded moderate and comparable
1
4
yields of allyl adducts to those of allyltin reagent.
The inefficient allylation was mainly due to the forma-
tion of pinacols, which were probably given by the
photoreduction via rapid hydrogen abstraction.
In conclusion, efficient photo-allylation of aromatic car-
bonyl compounds by a hypercoordinated allylsiliconate
Table 3. Photo-allylation of various carbonyl compounds with 1
a
Entry
Substrate
Solvent, reaction time (h), wave length (nm)
Allyl adduct
Yield (%)
1
MeOH, 3, >400
95 (84)
2
DMF, 1, >400
88 (47)
b
3
4
MeOH, 7, >400
MeOH, 7, >330
47/26 (16/5)
58 (28)
5
6
MeOH, 7, >330
MeOH, 7, >330
46 (11)
32 (22)
a
Values in parentheses are yields obtained by 7 h irradiation in acetonitrile as a solvent.
Benzoyl adduct/acetyl adduct.
b

214
Y. Nishigaichi et al. / Tetrahedron Letters 48 (2007) 211–214
was realized in a donating solvent or by the addition of a
donor molecule. The efficiency has been improved to as
high as that of the corresponding tin reagent. These
results possess considerable significance from both
mechanistic and synthetic viewpoints. The hypercoordi-
nation-driven photo-allylation is more efficient with a
hexa-coordinated reagent than with a penta-coordinated
one. Furthermore, it opens up a way to the use of
organosilicon reagents in the photo-allylation instead
of the harmful tin counterpart. The present methodol-
ogy will be applied to other substrates and other hyper-
coordinated siliconate reagents including stereoselective
6. A solution of a substrate (0.2 mmol) and 1 (0.3 mmol) in
an indicated solvent (10 ml) was irradiated by a high
pressure mercury lamp (300 W) through a suitable filter
solution (k >400 nm or 330 nm) at an ambient tempera-
ture under a nitrogen atmosphere. After an indicated
period, the reaction mixture was condensed and chro-
matographed on silica gel to afford the products.
7
. Reichardt, C. Solvents and Solvent Effects in Organic
Chemistry; VCH: Weinheim, 1988.
8. Photo-promotion of the allylation was confirmed by the
control reaction in the dark, where no allyl adduct was
obtained both in methanol and in DMF.
9
. The allylsiliconate 1 was not soluble enough in benzene
1
4b,15
alone, thus 20% of acetonitrile was used as a co-solvent.
0. Marcus, Y. J. Solution Chem. 1984, 13, 599.
1. Redox potentials were measured using a DPV technique at
reaction.
Such projects are now in progress in our
1
1
laboratory and the results will appear in due course.
2
0 mV/s scan rate in acetonitrile. The values are reported
versus SCE.
1
2. (a) Hosomi, A.; Kohra, S.; Ogata, K.; Yanagi, T.;
Tominaga, Y. J. Org. Chem. 1990, 55, 2415; (b) Kira,
M.; Sato, K.; Sakurai, H. J. Am. Chem. Soc. 1988, 110,
4599; (c) Cerveau, G.; Chuit, C.; Corriu, R. J. P.; Reye, C.
J. Organomet. Chem. 1987, 328, C17.
References and notes
1
. Yamamoto, Y.; Asao, N. Chem. Rev. 1993, 93, 2207;
Nishigaichi, Y.; Takuwa, A.; Naruta, Y.; Maruyama, K.
Tetrahedron 1993, 49, 7395.
13. Takuwa, A.; Nishigaichi, Y.; Yamashita, K.; Iwamoto, H.
Chem. Lett. 1990, 1761.
14. (a) Takuwa, A.; Tagawa, H.; Iwamoto, H.; Soga, O.;
Maruyama, K. Chem. Lett. 1987, 1091; (b) Takuwa, A.;
Shiigi, J.; Nishigaichi, Y. Tetrahedron Lett. 1993, 34,
3457.
2
3
. Nishigaichi, Y.; Takuwa, A. Photochemistry 2003, 34, 183.
. Fukuzumi, S.; Okamoto, T.; Otera, J. J. Am. Chem. Soc.
1994, 116, 5503.
4
. Nishigaichi, Y.; Suzuki, A.; Saito, T.; Takuwa, A.
Tetrahedron Lett. 2005, 46, 5149.
5
. Takuwa, A.; Nishigaichi, Y.; Yamashita, K.; Iwamoto, H.
Chem. Lett. 1990, 639.
15. Takuwa, A.; Nishigaichi, Y.; Yamaoka, T.; Iihama, K.
J. Chem. Soc., Chem. Commun. 1991, 1359.