Facile photochemical transformation of alkyl aryl selenides to the corresponding carbonyl compounds by molecular oxygen: Use of selenides as masked carbonyl groups

Article abstract of DOI:10.1021/jo801730j

(Chemical Equation Presented) Alkyl aryl selenides with and without functional groups on the alkyl group were transformed efficiently into the corresponding carbonyl compounds, particulary primary alkyl aryl selenides in good yields, by a simple photolysis in the presence of air or oxygen. This transformation can be conducted without protection of functional groups. The yield of carbonyl compounds was much affected by the solvent viscosity, reaction temperature, concentration of dissolved oxygen in the solvents, wavelength of light, and structure of the aryl substituents. The present study indicates that aryl selenides can be considered as a masked carbonyl group that can be easily converted to a carbonyl group by very mild reaction conditions even in the presence of various unprotected functional groups. Therefore, this functional group transformation can be used as an important tool in organic synthesis due to its simplicity and mild reaction condition.

Full text of DOI:10.1021/jo801730j

Facile Photochemical Transformation of Alkyl Aryl Selenides to the
Corresponding Carbonyl Compounds by Molecular Oxygen: Use of
Selenides as Masked Carbonyl Groups
Takeshi Hyugano,† Suyou Liu,‡,§ and Akihiko Ouchi*,†,‡
Graduate School of Pure and Applied Sciences, UniVersity of Tsukuba, Ibaraki 305-8571, and the National
Institute of AdVanced Industrial Science and Technology (AIST), Tsukuba, Ibaraki 305-8565, Japan
ouchi.akihiko@aist.go.jp
ReceiVed August 4, 2008
Alkyl aryl selenides with and without functional groups on the alkyl group were transformed efficiently
into the corresponding carbonyl compounds, particulary primary alkyl aryl selenides in good yields, by
a simple photolysis in the presence of air or oxygen. This transformation can be conducted without
protection of functional groups. The yield of carbonyl compounds was much affected by the solvent
viscosity, reaction temperature, concentration of dissolved oxygen in the solvents, wavelength of light,
and structure of the aryl substituents. The present study indicates that aryl selenides can be considered
as a masked carbonyl group that can be easily converted to a carbonyl group by very mild reaction
conditions even in the presence of various unprotected functional groups. Therefore, this functional group
transformation can be used as an important tool in organic synthesis due to its simplicity and mild reaction
condition.
Introduction
reactions for the C-C bond formation are those of carbanions
stabilized by selenium atoms and stannyl radical-mediated
reactions,1 but photochemical reactions have also been reported,
in which the reaction proceeds via carbon radicals that are
generated by photochemical PhSe-C bond cleavages.2,3 The
most widely used reaction for the functional group transforma-
tion is that into olefins, which proceeds by oxidation to
selenoxides and successive elimination.1 Although other pho-
Organoselenium compounds have been used as important
tools in organic synthesis, and arylselenyl groups have been
the most commonly used species.1 Major synthetic applications
of the reactions of arylselenyl compounds generally fall into
two categories, i.e., C-C bond formations at the carbon atom
adjacent to the selenium atom and transformations of functional
groups into other functional groups. The most widely used
† University of Tsukuba.
(2) (a) Renaud, P. Top. Curr. Chem. 2000, 208, 81–112. (b) Pandey, G.;
Gadre, S. R. Acc. Chem. Res. 2004, 37, 201–210.
‡ AIST.
§ Present affiliation: Central South University, Changsha, Hunan, People’s
Republic of China.
(3) (a) Byers, J. H.; Lane, G. C. J. Org. Chem. 1993, 58, 3355–3360. (b)
Renaud, P.; Vionnet, J.-P. J. Org. Chem. 1993, 58, 5895–5896. (c) Ouchi, A.;
Adam, W. J. Chem. Soc., Chem. Commun. 1993, 628–629. (d) Curran, D. P.;
Martin-Esker, A. A.; Ko, S.-B.; Newcomb, M. J. Org. Chem. 1993, 58, 4691–
4695. (e) Curran, D. P.; Eichenberger, E.; Collis, M.; Roepel, M. G.; Thoma, G.
J. Am. Chem. Soc. 1994, 116, 4279–4288. (f) Ouchi, A.; Yabe, A.; Adam, W.
Tetrahedron Lett. 1994, 35, 6309–6312. (g) Ryu, I.; Muraoka, H.; Kambe, N.;
Komatsu, M.; Sonoda, N. J. Org. Chem. 1996, 61, 6396–6403. (h) Back, T. G.;
Gladstone, P. L.; Parrez, M. J. Org. Chem. 1996, 61, 3806–3814. (i) Ouchi, A.;
Koga, Y. J. Org. Chem. 1997, 62, 7376–7383. (j) Byers, J. H.; Shaughnessy,
E. A.; Mackie, T. N. Heterocycles 1998, 48, 2071–2077. (k) Ouchi, A.; Sakuragi,
M.; Kitahara, H.; Zandomeneghi, M. J. Org. Chem. 2000, 65, 2350–2357. (l)
Ouchi, A.; Li, Z.; Sakuragi, M.; Majima, T. J. Am. Chem. Soc. 2003, 125, 1104–
1108. (m) Murakami, S.; Kim, S.; Ishii, H.; Fuchigami, T. Synlett 2004, 815–
818. (n) Murakami, S.; Ishii, H.; Fuchigami, T. J. Fluorine Chem. 2004, 125,
609–614.
(1) (a) Patai, S., Rappoport, Z., Eds. The Chemistry of Organic Selenium
and Tellurium Compounds; John Wiley & Sons: Chichester, U.K., 1987; Vol.
2. (b) Krief, A. In ComprehensiVe Organic Synthesis, SelectiVity, Strategy &
Efficiency in Modern Organic Chemistry; Trost, B. M., Fleming, I., Eds.;
Pergamon Press: Oxford, U.K., 1991; Vol. 1, Section 2.6, pp 629-728. (c) Krief,
A. In ComprehensiVe Organic Synthesis, SelectiVity, Strategy & Efficiency in
Modern Organic Chemistry; Trost, B. M., Fleming, I., Eds.; Pergamon Press:
Oxford, U.K., 1991; Vol. 3, Section 1.3, pp 85-191. (d) Krief, A. In
ComprehensiVe Organometallic Chemistry II, A ReView of the Literature 1982-
1994; Abel, E. W., Stone, F. G. A., Wilkinson, G., Eds.; Pergamon, Elsevier
Science Ltd.: Oxford, U.K., 1995; Vol. 11, Section 13, pp 515-569. (e) Kataoka,
T.; Yoshimatsu, M. In ComprehensiVe Organic Functional Group Transforma-
tions; Katritzky, A. R., Meth-Cohn, O., Rees, C. W., Eds.; Pergamon, Elsevier
Science Ltd.: Oxford, U.K., 1995; Vol. 2, Section 2.04, pp 277-295.
10.1021/jo801730j CCC: $40.75
Published on Web 10/15/2008
2008 American Chemical Society
J. Org. Chem. 2008, 73, 8861–8866 8861

Hyugano et al.
FIGURE 1. Absorption spectra of 1a (green line),10 1b (blue line), 1c
(red line), and 1d (purple line). Conditions: concentration, 10-5 M in
hexane; optical path length, 10 mm.
FIGURE 2. Decrease of 1a and yields of the photoproducts 2-7 as a
function of the irradiation time. Symbols: 1a (9), 2 (b), 3 (2), 4 ([),
5a (0), 6a (O), and 7a (4). Photolysis conditions: concentration of
1a, 1 mM in hexane; light source, 15 W low-pressure mercury lamp
fitted with a UV-25 cutoff filter (0.42 mW ·cm-2); optical path length,
10 mm; in air; room temperature. Yields of the products are based on
the consumed starting material. The results are the average of three
independent runs.
SCHEME 1. Photolysis of Pentyl Aryl Selenides
aldehyde 2 was observed with the progress of the photolysis
for 1a, 1b, and 1c, which gave a final yield for 2 of 69%, 55%,
and 76% for 1a, 1b, and 1c, respectively.
Scheme 2 shows a plausible reaction mechanism for the
photolysis of 1, in which the reaction in the absence of oxygen
is confirmed by photolysis of 1a under a nitrogen atmosphere.10
Although details of the reaction mechanism for the formation
of the oxygenated compounds 2, 4, and ArOH compounds 6a
and 6b are still not clear at the moment, comparison of
photolysis in the presence and in the absence of oxygen indicates
clearly that these oxygenated compounds were formed by
reaction(s) with oxygen, most probably via peroxy radicals. The
mechanism in Scheme 2 implies that (i) facilitation of the escape
of pentyl radicals from solvent cages and (ii) the increase in
the ratio of path A to path B are two important factors for the
increase in the yield of 2 in the photolysis.
Solvent Effect in the Photolysis of 1a. It is reported that
the escape of pentyl radicals from solvent cages shown in
Scheme 2 is facilitated by using solvents with low viscosity.10
Therefore, the solvent viscosity effect on the distribution of
photoproducts was studied by conducting the photolysis of 1a
in different solvents. Figure 3a shows the results for a variety
of hydrocarbon solvents. As seen in the figure, the yield of
oxygenated photoproducts 2, 4, and 6a increased with a decrease
of the solvent viscosity, which is consistent with the order of
the facility of radical escape from solvent cages.
tochemical functional group transformations into alcohols,2a,4
methyl ether,2b and hydroperoxide5 have been reported, simple
functional group transformation into carbonyl compounds, which
is more useful for synthetic purposes, has not been reported so
far.
We report here a facile direct transformation of alkyl aryl
selenides to the corresponding carbonyl compounds by a simple
photooxidation using air or oxygen, which proceeds without
the need for protection of various functional groups. This
transformation can extend the scope of the utilization of
arylselenyl groups in organic synthesis and can be used as an
important tool in organic synthesis.
Results and Discussion
Absorption Spectra of Alkyl Aryl Selenides 1. To determine
the optimal irradiation wavelength for the photolyses, the UV
spectra of the pentyl aryl selenides 1 used in this study were
measured. Figure 1 shows the absorption spectra of pentyl
phenyl selenide (1a), 2-naphthyl pentyl selenide (1b), 1-naphthyl
pentyl selenide (1c), and pentyl 1-pyrenyl selenide (1d). The
UV absorption bands of 1a-d correspond to π-π* excitations
of arylselenyl groups6 and an n-π* excitation of the selenium
atom.7
Photolysis of 1a, 1b, and 1c by a Low-Pressure Mercury
Lamp. Photolysis of 1a, 1b, and 1c in hexane was conducted
with a low-pressure mercury lamp in air. The major products
obtained were pentanal (2), 1-pentene (3), 1-pentanol (4), ArH
compounds 5, ArOH compounds 6, and diaryl diselenides 7
(Scheme 1).
When polar solvents were used, the yield of 2 was lower
than that in hydrocarbons at the same solvent viscosity (Figure
3b). This result is rationalized by the difference in the solubility
of oxygen gas in the solvents; the solubility of oxygen in polar
solvents is lower than that in hydrocarbons,11 which reduces
the probability of the reaction between oxygen and pentyl
radicals. These results indicate that the photolysis should be
conducted in solvents that have low viscosity and high solubility
of oxygen gas to obtain 2 in high yields.
Effect of Aryl Substituents. The photolysis of 1a under a
nitrogen atmosphere showed the formation of 3 via abstraction
of hydrogen from pentyl radicals by phenylselenyl radicals in
Figure 2 shows the results of photolysis of 1a in hexane. Most
of 1a was consumed after 40 min of irradiation by a low-
pressure mercury lamp (Figure 2), whereas the consumption
was much slower for 1b and 1c (Figure S2, Supporting
Information).8 A slight increase in the yield9 of the desired
(10) Ouchi, A.; Liu, S.; Li, Z.; Kumar, S. A.; Suzuki, T.; Hyugano, T.;
Kitahara, H. J. Org. Chem. 2007, 72, 8700–8706.
(11) (a) Battino, R., Ed. Oxygen and Ozone; IUPAC Solubility Data Series,
Vol. 7; Pergamon Press: Oxford, U.K., 1981. (b) Wilhelm, E.; Battino, R. Chem.
ReV. 1973, 73, 1–9.
(4) Hong, I. S.; Greenberg, M. M. Org. Lett. 2004, 6, 5011–5013.
(5) Giese, B.; Beyrich-Graf, X.; Erdmann, P.; Giraud, L.; Imwinkelried, P.;
Mu¨ller, S. N.; Schwitter, U. J. Am. Chem. Soc. 1995, 117, 6146–6147.
(6) Jaffe´, H. H.; Orchin, M. Theory and Application of UltraViolet Spec-
troscopy; John Wiley & Sons: New York, 1962; Chapter 12, Section 12.2.
(7) Mason, W. R. J. Phys. Chem. 1996, 100, 8139–8143.
(12) (a) Fossey, J.; Lefort, D.; Sorba, J. Free Radicals in Organic Chemistry;
John Wiley & Sons: Chichester, U.K., 1995; p 33. (b) Fossey, J.; Lefort, D.;
Sorba, J. Free Radicals in Organic Chemistry; John Wiley & Sons: Chichester,
U.K., 1995; pp 152-154.
(8) Original data are given as Supporting Information.
(13) In the case of the photolysis in decane, the yield of 6a was not obtained
due to the overlapping of the peaks of decane and 6a in the GLC analysis.
(9) Yields of the products are based on the consumed starting material.
8862 J. Org. Chem. Vol. 73, No. 22, 2008

Transformation of Selenides to Carbonyl Compounds
SCHEME 2. Plausible Reaction Mechanism for the Photolysis of Pentyl Aryl Selenides 1
TABLE 1. Yield of 2-7 in the Photolysis of Aryl Pentyl Selenides 1
yielda (%)
run substrate wavelength (nm) irradiation timea(min) consumptiona (%)
2
3
4
5
6
7
2 + 3 + 4
5 + 6
1
2
3
4
5
6
1a
1b
1c
1b
1c
1d
254b
40 (1)
40 (5)
40 (5)
40 (5)
20 (5)
95 (8)
69 13
55 11
9
6
7
24 (22) 17 (0) 36 (38)
91
72
92
96
98
95
41 (22)
8 (7)
12 (13)
1 (1)
5 (5)
8 (0)
80 (18)
89 (28)
87 (22)
98 (29)
98 (13)
7 (7)
10 (10)
1 (1)
1 (0) 71 (79)
2 (3) 54 (62)
0 (0) 89 (98)
0 (0) 59 (79)
0 (0) 54 (9)
76
9
>330c
63 23 10
78 15
76 13
5
6
5 (5)
320 (10)
8 (0)
a Yields of the products are based on the consumed starting material. The results are the average of two independent runs. The values in parentheses
show those at the early stage of the photolysis. b Photolysis conditions: substrate concentration, 1 mM in hexane; light source, 15 W low-pressure
mercury lamp fitted with a Toshiba UV-25 cutoff filter (0.42 mW ·cm-2); optical path length, 10 mm; in air; at room temperature. c Photolysis
conditions: substrate, 1 mM in hexane; light source, 500 W xenon short-arc lamp fitted with an 18 cm water filter and a Toshiba UV-33 cutoff filter
(0.14 mW ·cm-2); optical path length, 10 mm; in air; at room temperature.
FIGURE 3. Consumption of 1a and distribution of photoproducts
FIGURE 4. Consumption of 1c and yield of photoproducts 2-7c using
different cutoff filters. Symbols: 1c (9), 2 (b), 3 (2), 4 ([), naphthalene
(5c; 0), 1-naphthol (6c; O), and bis(1-naphthyl) diselenide (7c; 4).
Photolysis conditions: concentration of 1c, 1 mM in hexane; optical
path length, 10 mm; light source, 500 W xenon short-arc lamp fitted
with an 18 cm water filter and a cutoff filter; irradiation time, 20 min;
in air; room temperature. Yields of the products are based on the
consumed starting material. The results are the average of two
independent runs.
2-7 (a) in hydrocarbon and (b) in polar solvents as a function of
the solvent viscosity. Symbols:13 1a (9), 2 (b), 3 (2), 4 ([), 5a
(0), 6a (O), and 7a (4). Photolysis conditions: concentration of
1a, 1 mM; light source, 15 W low-pressure mercury lamp fitted
with a UV-25 cutoff filter (0.42 mW · cm-2); irradiation time, 40
min; optical path length, 10 mm; in air; room temperature. Yields
of the products are based on the consumed starting material. The
results are the average of two independent runs.
solvent cages (cf. Scheme 2).10 To suppress such hydrogen
abstractions through steric hindrance of aryl groups, photolysis
using the 1-naphthyl group was conducted and compared with
those of phenyl and 2-naphthyl groups. Table 1, runs 1-3,
shows the effect of the aryl substituents in low-pressure mercury
lamp photolysis. As seen in the table, the yield of 2 from 1c
was higher than that from 1a or 1b, and the yield of 3 was the
lowest among 1a-c. The mass balance of the products formed
by the pentyl-Se bond cleavage, 2 + 3 + 4, was 92% for 1c,
which was comparable to that of 1a (91%) and much higher
than that of 1b (72%). These results indicate that the 1-naphthyl
group (1c) was the best aryl group among the three aryl
substituents.
in the photolysis of 1c, which showed the best yield of 2 among
1a-c, was investigated. The photolysis was conducted in hexane
solutions using a xenon lamp fitted with various cutoff filters.
Figure 4 shows the wavelength effect on the photolysis of
1c using different cutoff filters. The yield of pentanal 2 and
diselenide 7c increased with increasing irradiation wavelength8
up to 330 nm (UV-33 cutoff filter). However, when longer
wavelengths (UV-35 cutoff filter) were used, the yield of 2
decreased together with a drastic decrease in the consumption
of 1c; the latter can be explained by a considerable decrease in
the absorption of 1c between the light using UV-33 and UV-
35 filters. From these results, irradiation using a UV-33 filter
was found to be optimal for the photolysis. A similar trend was
observed for 1b,8 but with slower consumption of the starting
compound and lower yield of 2.
Wavelength Dependence in the Photolysis of 1c. The
photolysis of 1a under a nitrogen atmosphere showed that
the ratio of path A in Scheme 2 increased with an increase of
the irradiation wavelength. 10 Therefore, the wavelength effect
J. Org. Chem. Vol. 73, No. 22, 2008 8863

Hyugano et al.
SCHEME 3. Photolysis of 8
SCHEME 4. Photolysis of 9
FIGURE 5. Consumption of 1c and yield of photoproducts 2-7c as a
function of the reaction temperature. Symbols: 1c (0, 9), 2 (O, b), 3
(4,2), 4 (open and solid right-pointing triangles), 5c (open and solid
left-pointing triangles), 6c (3, 1), and 7c (], [). Photolysis conditions:
concentration of 1c, 1 mM in hexane; light source, 500 W xenon short-
arc lamp fitted with an 18 cm water filter and a UV-33 cutoff filter
(0.14 mW ·cm-2); irradiation time, 20 min, optical path length, 10 mm;
in air. The reaction temperature was controlled by a variable-temperature
liquid nitrogen cryostat (open symbols, 183 - 313 K) or by a constant-
temperature water bath (closed symbols, 293 - 333 K). Yields of the
products are based on the consumed starting material. The results are
the average of two independent runs.
The photolysis was conducted with 10 and 20 mM hexane
solutions of 1c, both in air and under an oxygen atmosphere.8
The photolysis required irradiation for a longer time with
increased concentration of 1c. When photolysis was conducted
in air, the yield of the desired 2 decreased with progress of the
reaction in solutions with a high concentration of 1c, although
the yield of 2 was almost the same at the initial stage of the
photolysis, irrespective of the concentration of 1c. However,
when the photolysis was conducted in oxygen-saturated hexane
solutions, the yield of the desired 2 in the solutions with a high
concentration of 1c was comparable to that of dilute solutions
even at the final stage of the photolysis. These results indicate
that the decrease of the yield of 2 at the final stage of the
photolysis in air is probably due to the lack of dissolved oxygen
for the reaction. This result indicates also that the same yield
of 2 can be obtained in solutions with a high concentration of
substrate if sufficient oxygen is supplied to the reaction mixture.
Photolysis of Alkyl 1-Naphthyl Selenides. Table 2 shows a
summary of the photolysis of various alkyl 1-naphthyl selenides
having different functional groups in the alkyl group.8 The
photolysis was conducted using the optimal conditions estab-
lished for 1c under an oxygen atmosphere. As seen in Table 2,
the transformation to aldehydes proceeded in good yields when
the alkyl group was a primary alkyl group (runs 1-5). A
preparative photolysis of 10 using a larger reaction vessel with
a higher concentration of substrate showed almost the same yield
as that observed in the small-scale photolysis (run 4).
In the case of selenide 8, photochemically generated terminal
octyl radical 8′ may form internal octyl radical 8′′ by an
intramolecular 1,5-H shift because the bond dissociation
energy12a of homolytic cleavage of primary C-H bonds is
slightly higher than that of secondary C-H bonds (Scheme 3).
From radical 8′′, formation of 4-octanone (16a) and 4-hydroxy-
octane (16b) is expected from successive reaction with molec-
ular oxygen. However, 16a and 16b were not detected by GLC
analyses. This result indicates that it is not necessary to consider
the occurrence of radical shifts within alkyl chains.
As for selenide 9, photochemically generated hexenyl radical
9′ is known to undergo fast intramolecular exo cyclization to
give cyclopentylmethyl radical 9′′,12b from which cyclopentan-
ecarboxaldehyde (17a) and cyclopentylmethyl alcohol (17b) are
expected to be formed (Scheme 4). However, to our surprise,
17a and 17b were not detected by GLC analyses. This result
indicates that oxygen trapping proceeded faster than intramo-
lecular radical cyclization. The reason for this phenomenon is
Temperature Effect in the Photolysis of 1c. The photolysis
of 1c was conducted at different temperatures to optimize the
reaction conditions. As shown in Figure 5, the consumption of
1c and the yield of the photoproducts were much affected by
the reaction temperature. The consumption of 1c and the yield
of 2, 4, and 5c increased with increased temperature, whereas
the yield of 3 and 7c decreased with increased temperature. This
temperature dependence can be explained by the decrease of
solvent viscosity with an increase of reaction temperature, which
facilitates the escape of pentyl radicals from solvent cages (cf.
Scheme 2).10 This is consistent with the results obtained by
experiments on the viscosity effect. However, the consumption
of 1c and the yield of 2 leveled off at about 300 K, which
indicates that room temperature is optimal for the photolysis.
The optimal conditions for the photolysis of 1c were to
conduct the photolysis in hexane at room temperature using a
xenon lamp fitted with a UV-33 cutoff filter. The same
conditions were applied to the photolysis of 1b and 1d, whose
time profiles of the consumption of 1 and the yields of
photoproducts 2-7 were monitored.8 Some of these results are
summarized in Table 1, runs 4-6. As seen in the table, the rate
of consumption of 1 increased in the order 1d < 1b < 1c, and
the yield of the desired 2 increased in the order 1b < 1d e 1c.
The yield of 2 in the photolysis of 1b and 1c was increased
compared with that using a low-pressure mercury lamp. The
total yield of 2, 3, and 4 formed by the pentyl-Se bond cleavage
was almost quantitative. The yield of 5 and 6 was decreased
compared with that from photolysis using a low-pressure
mercury lamp, particularly at the initial stage of the photolysis.
This result indicates considerable suppression of the Ar-Se
bond cleavage, which is consistent with significant suppression
of path B in Scheme 2 using light of longer wavelength.10
Effect of the Oxygen Concentration in Solutions. In the
experiments on solvent effect, we have observed that the yield9
of photoproducts was much affected by the amount of dissolved
oxygen in solution. Therefore, to increase the yield of 2, we
have conducted the photolysis of 1b and 1c in air and under an
oxygen atmosphere for comparison.8 When the photolysis was
conducted under an oxygen atmosphere instead of in air, the
yield of 2 increased from 63% to 78% in the case of 1b and
from 78% to 82% for 1c.
Effect of the Substrate Concentration. The effect of the
substrate concentration was studied because it is preferable to
conduct the reaction with a higher concentration of substrate.
8864 J. Org. Chem. Vol. 73, No. 22, 2008

Transformation of Selenides to Carbonyl Compounds
TABLE 2. Photochemical Conversion of Alkyl 1-Naphthyl Selenides to the Corresponding Carbonyl Compoundsa
a Yields of the products are based on the consumed starting material. The results are the average of two independent runs. Photolysis conditions:
concentration of substrates, 1 mM in hexane; light source, 500 W xenon short-arc lamp fitted with an 18 cm water filter and a UV-33 cutoff filter (0.14
mW·cm-2); irradiation time, 15 min; optical path length, 10 mm; under oxygen; room temperature. b A 1 mM concentration in heptane was used to
avoid overlapping of the solvent and product peaks in GLC analysis. c Isolated yield. Photolysis conditions: concentration of 10, 10 mM in pentane;
reaction vessel, Pyrex glass test tubes (i.d. ) 20 mm); light source, 500 W xenon short-arc lamp fitted with an 18 cm water filter; irradiation time, 6 h;
under oxygen; room temperature. d Rearranged product 23 was obtained as a major product.
not clear at the moment, but this result indicates that we do not
have to consider participation of competitive radical cyclization
in the photochemical conversion of selenides to carbonyl
compounds.
considerable amount of olefins, which is most probably due to
the larger number of hydrogen atoms on the carbons adjacent
to the carbon radical, which increases the probability of
hydrogen abstraction to form olefins.
For selenides with primary alkyl groups, some corresponding
olefins and alcohols were formed together with the desired
aldehydes. The olefins were formed by abstraction of hydrogen
adjacent to the carbon radical by arylselenyl radicals within the
solvent cage.10 Therefore, to avoid the formation of olefins, the
reaction was tested with selenides having no ꢀ-hydrogen atoms
to the selenium atom (runs 6 and 7), in which the formation of
olefins is not possible. To our surprise, the yield of the
corresponding aldehydes was slightly lower than those of
primary alkyl groups despite the fact that a significant side
reaction, formation of olefins, was completely blocked.
Although 2-methylallyl selenide 13 gave corresponding
aldehyde 21 in a good yield (run 7), allylic radicals are known
to react at both its ends. Therefore, the reactivity of unsym-
metrical allylic radical was tested using trans-2-butenyl selenide
14; the yield of the expected trans-2-butenal (22) was low and
showed a preferential formation of methyl vinyl ketone (23)
(run 8). This result indicates that, in the case of unsymmetrical
allylic radicals, the carbonyl group is preferentially introduced
at the more substituted end of the allylic radicals.
Conclusion
Alkyl aryl selenides with and without functional groups on
the alkyl group were transformed efficiently into the corre-
sponding carbonyl compounds, particulary primary alkyl aryl
selenides in good yields, by a simple photolysis in the presence
of air or oxygen. The yield of carbonyl compounds was much
affected by the solvent viscosity, reaction temperature, concen-
tration of dissolved oxygen in the solvents, wavelength of light,
and structure of the aryl substituents. The best photolysis
condition was to conduct the photolysis in hexane at room
temperature under an oxygen atmosphere using a 1-naphthyl
group as an aryl group with a xenon lamp fitted with a UV-33
cutoff filter. The present study indicates that aryl selenides can
be considered as a masked carbonyl group that can be easily
converted to a carbonyl group by very mild reaction conditions
even in the presence of various unprotected functional groups.
Therefore, this functional group transformation can be used as
an important tool in organic synthesis due to its simplicity and
mild reaction conditions.
The yield of ketone from the secondary alkyl group was lower
than that of the primary alkyl group due to formation of a
J. Org. Chem. Vol. 73, No. 22, 2008 8865

Hyugano et al.
solutions of authentic samples, and the yields of each product were
calculated from the following equation: 100 × [absolute yield (%)]/
[consumption of starting materials (%)].
Experimental Section
General Aspects. Selenides 1a-d and 8-15, diselenides 7c and
7d, aldehydes 17-19, and olefins 29 and 31 were synthesized by
general procedures.8 The solvents used for the photolyses were
spectral grade methanol, ethanol, 2-propanol, acetonitrile, and 1,4-
dioxane, guaranteed reagent grade heptane, decane, dodecane, and
cyclohexane, and fractionally distilled spectral grade hexane.
Authentic samples of 2, octanal (16), benzaldehyde (20), 2-methyl-
2-propenal (21), 22, 23, 3-pentanone (24), 16a, 17a, pentanol,
octanol, 5-hexenol, ethyl 6-hydroxyhexanoate, 4-oxopentanol, ben-
zyl alcohol, 2-methyl-2-propenol, 3-buten-2-ol, trans-2-butenol,
3-pentanol, 16b, 17b, 1-pentene, 1-octene, 1,5-hexadiene, trans-
2-pentene, cis-2-pentene, benzene, naphthalene, pyrene, phenol,
1-naphthol, 2-naphthol, and pentane were purchased and used as
received.
Photolysis of Alkyl Aryl Selenides. The photolysis was
conducted in 1.5 mL of solution in air or under an oxygen
atmosphere using a synthetic quartz cell of 10 mm width and 10
mm optical path length fitted with a Teflon/silicon rubber septum.
The solutions for the photolysis of 1a-d and 8-15 (1 mM) were
saturated with air or with oxygen before photolysis. Caution:
general precautions haVe to be taken for handling pure oxygen,
and it is desirable to put a safety glass in front of the reaction
Vessel! The photolysis was conducted with a 500 W xenon lamp
fitted with an 18 cm water filter and a cutoff filter (UV-29, UV-31,
UV-33, or UV-35) or with a 15 W low-pressure mercury lamp fitted
with a UV-25 cutoff filter. In the experiments on the temperature
effect, the temperature was controlled by a variable-temperature
liquid nitrogen cryostat (183-313 K) or by a constant-tempera-
ture water bath fitted with a recirculating chiller (293-333 K). The
light intensity of the xenon lamp was measured by a spectroradi-
ometer. The consumption of 1 and the yield of product were
analyzed by comparison with authentic samples by capillary GLC
[capillary columns, NEUTRA BOND-1 (60 m, 0.25 mm i.d., 1.5
µm thickness) or INERT CAP-1 (30 m, 0.25 mm i.d., 1.5 µm
thickness)] and HPLC [RP-8e column (250 mm, 4 mm i.d.)] fitted
with a UV detector (detected at 254 nm). For all GLC and HPLC
analyses, absolute yields of products were determined using standard
Preparative Photolysis of 5-(Ethoxycarbonyl)pentyl 1-Naph-
thyl Selenide (10). A 20 mL sample of an oxygen-saturated hexane
solution of 10 [10 mM, 70 mg (0.2 mmol)] was transferred into
two 20 mL Pyrex test tubes (i.d. 20 mm) fitted with silicon rubber
septa that were connected to a balloon filled with oxygen with
syringe needles. Caution: general precautions haVe to be taken for
handling pure oxygen, and it is desirable to put a safety glass in
front of the reaction Vessel!. The photolysis was conducted for 6 h
with a 500 W xenon lamp fitted with an 18 cm water filter. The
solvent was evaporated in vacuo, and the crude products were
purified by silica gel column chromatography (n-hexane:ethyl
acetate ) 50:1), yielding 24 mg (76%) of the desired aldehyde 18
as a colorless oil. 1H NMR (δ): 1.26 (3H, t, J ) 7.2 Hz), 1.62-1.70
(4H, m), 2.27-2.36 (2H, m), 2.43-2.50 (2H, m), 4.08-4.16 (2H,
q, J ) 7.2 Hz), 9.76 (1H, s) ppm. 13C NMR (δ): 14.2, 21.5, 24.4,
34.0, 43.5, 60.4, 173.2, 202.0 ppm. GLC analysis showed that the
product was identical to an authentic sample of 18.
Supporting Information Available: Synthetic procedure of
alkyl aryl selenides 1a-d, 8-15, diselenides 7c and 7d,
aldehydes 17-19, and olefins 29 and 31, results on the
consumption of 1b and 1c and yield of photoproducts using a
low-pressure Hg lamp (UV-25 cutoff filter), consumption of
1b and yield of photoproducts using a Xe lamp (different cutoff
filters), consumption of 1b-d and 8-15 and yield of photo-
products using a Xe lamp (UV-33 cutoff filter), and consumption
of 1c and yield of photoproducts (with high substrate concentra-
tion in air and under an oxygen atmosphere) using a Xe lamp
(UV-33 cutoff filter), and emission spectra of a xenon lamp
fitted with an 18 cm water filter and different cutoff filters. This
material is available free of charge via the Internet at
http://pubs.acs.org.
JO801730J
8866 J. Org. Chem. Vol. 73, No. 22, 2008