Table 1 Activation parameters and efficiency of 1O2 generation for
thermolysis of N-substituted 2-pyridone endoperoxides (2a–2e) in
CDCl3
1O2/%
DG{/
DH{/
DS{/
kcal mol21 kcal mol21 cal K21 mol21
Aa
Bb
2a
2b
2c
2d
2e
a
24.3
24.1
24.7
24.6
25.2
22.8
23.0
23.2
24.1
25.2
25.3
23.7
25.0
21.4
0.1
96 ¡ 1 81 ¡ 1
96 ¡ 1 73 ¡ 1
97 ¡ 1 84 ¡ 4
89 ¡ 2 74 ¡ 6
84 ¡ 1 82 ¡ 1
b
Mean of 1O2 yields for , 20% conversion of 2 at 40 uC. Mean of
1O2 yields at 30, 40, and 50 uC. 1O2 yield at each temperature was
estimated from the data at five points for 30–80% conversion of 2.
Scheme 2
Chemical 1O2 sources which are less complicated mechanistically
are quite useful for 1O2 reactions of biological substrates,
mechanistic investigations of 1O2 reactions with organic substrates,
and as an essential chemical species for COIL (chemical oxygen–
iodine laser).10 As described earlier, the present endoperoxides (2)
and 2) positive DS{ values correlate with the relatively low yields of
1O2, and slightly negative or near zero DS{ values correlate with
the high yields of 1O2.8,9 They have also suggested that these
tendencies can be interpreted in terms of a diradical mechanism
1
would be useful as a new chemical source of O2. Comparing 2
1
with 1,4-endoperoxides of alkylnaphthalene derivatives11,12 as
chemical 1O2 sources, one advantageous feature of 2 probably lies
in the easy preparation and modification of their precursors,
namely, N-substituted 2-pyridones (1), since the synthesis of most
substituted naphthalenes is unexpectedly burdensome. A good
example is the preparation of a polymer-supported 1O2 generator.
Functionalized polymer represented by Merrifield’s resin is often
used in a wide variety of reactions in the solid phase. 2-Pyridone
was easily joined to Merrifield’s resinI to give 1f by means of
nucleophilic substitution similar to the synthesis of 1a–1e. The
thus-prepared pyridone supported on polymer (1f) was also easily
converted into the corresponding endoperoxide (2f) by sensitized
photooxygenation similar to the case of 1a–1e except that a
suspension of the substrate was used (Scheme 2).** When
endoperoxide supported on Merrifield’s resin (2f) was heated with
olefin (5) at 40 uC for 1 h, 1O2 was estimated to be liberated in 85%
yield based on produced hydroperoxide (8) and the consumed
endoperoxide.
which leads to a low yield of O2 and a concerted mechanism
1
which leads to high yields of O2, respectively.
DS{ values ranged from slightly positive to negative for
endoperoxides of pyridones (2a–2e). Furthermore, column A in
Table 1 shows that the liberation of 1O2 occurred nearly
quantitatively for most cases of thermolysis of 2 at low conversion
of 2 (, 20%). These relationships between the yield of 1O2 and the
DS{ value are illustrated in Fig. 1. One would realize from Fig. 1
that endoperoxides of pyridones (2) (squares) belong to the same
class which includes endoperoxides of polynuclear aromatics for
the mechanism of 1O2 liberation. On the other hand, the efficiency
1
of O2 generation decreased considerably at high conversion of 2
1
except 2e (y80%), though the absolute yields of O2 stayed at a
high level. When the thermolysis of 2a was carried out in the
1
presence of an equimolar amount of 2-pyridone (1a), O2 yield
decreased to ca. 80% at low conversion of 2a. Thus, the decrease of
1O2 yields is likely to be attributable to the fact that 2-pyridone (1)
produced during the thermolysis of 2 quenches 1O2 physically,
though detailed experiments would be required to confirm such a
conclusion.
In conclusion, we have shown here that N-substituted pyridones
(1) easily undergo singlet oxygenation to give exclusively the
corresponding endoperoxides (2), which decompose to give
pyridones again while liberating 1O2 in high yields.
Endoperoxides (2) have been shown to be a new type of chemical
1O2 source, preparation and structural modification of which are
1
very easy and the efficiency of O2 liberation from which is high.
1
Finally, it should be noted that a water-soluble and neutral O2
generator, such as an amide derivative (2g), was also obtained.
Masakatsu Matsumoto,* Masayo Yamada and Nobuko Watanabe
Department of Materials Science, Kanagawa University, Tsuchiya,
Hiratsuka, Kanagawa, 259-1293, Japan.
E-mail: matsumo-chem@kanagawa-u.ac.jp
Notes and references
{ Selected data for 2a: 1H-NMR (400 MHz, CDCl3): dH 4.42 (d,
J 5 15.3 Hz, 1H), 4.87 (d, J 5 15.3 Hz, 1H), 5.08 (ddd, J 5 5.8, 2.1,
0.6 Hz, 1H), 5.53 (ddd, J 5 5.2, 2.1, 0.6 Hz, 1H), 6.71 (ddd, J 5 7.8, 5.2,
2.1 Hz, 1H), 6.76 (ddd, J 5 7.8, 5.8, 2.1 Hz, 1H), 7.19–7.24 (m, 2H), 7.28–
7.39 (m, 3H) ppm; 13C-NMR (100 MHz, CDCl3): dC 46.9, 78.0, 83.2, 127.8,
127.9, 128.6, 128.8, 134.2, 135.1, 167.9 ppm; IR (KBr): 3089, 3031, 2932,
1691, 1426, 1350, 1158 cm21; Mass (EI, %): 217 (M+, 25), 201 (70), 185
(76), 184 (34), 106 (34), 91 (100), 65 (20).
Fig. 1 Relationship between 1O2 yield and activation entropy for
thermolysis of endoperoxides of polynuclear aromatics and pyridone
endoperoxides (2).
484 | Chem. Commun., 2005, 483–485
This journal is ß The Royal Society of Chemistry 2005