Table 1 Thermal rearrangement of 1,4-endoperoxides 2a to 1,2-dioxetanes
5a and thermal decomposition of 5a to ketoesters 6aab
DH‡/kcal
DS‡/cal K21 DG‡/kcal
Peroxide
mol21
mol21
mol21
t1/2 (25 °C)/yr
cis-2a
trans-2a
5a
29.8
29.1
26.2
13.9
11.9
29.5
25.7
25.5
29.1
0.07
0.05
19.6
Scheme 3
a Rearrangement of 2a to 5a was carried out in toluene-d8 at 40–60 °C,
whereas decomposition of 5a to 6a was performed in toluene-d8 at
90–110 °C. b For all the Arrhenius plots, r > 0.999.
parameters were estimated from Arrhenius plots as summarized
in Table 1, which shows that dioxetanes 5a are far more stable
than endoperoxides 2a. The other endoperoxides 2b and 2c
rearranged thermally also into the corresponding dioxetanes 5b
and 5c, whose thermolysis gave ketoesters 6b and 6c,
respectively. Free energies of activation (DG‡/kcal mol21) for
these peroxides were also estimated as follows; 2b: 24.3–24.4,
2c: 24.4–24.7, 5b: 27.1 and 5c: 28.4.
Various 1,4-endoperoxides have been known to undergo
acid-catalyzed decomposition to carbonyl fragments, pre-
sumably through 1,2-dioxetanes, whose spectroscopic identi-
fication and/or isolation has not been achieved in most cases.
Schaap et al. have reported a successful spectroscopic identi-
fication of a dioxetane formed by the rearrangement of
endoperoxide, and suggested that the relatively unstable
dioxetane as a chemiluminescent substrate can be ‘stored’ as a
precursor endoperoxide and generated when needed.6 In
contrast, the present dioxetanes 5 are far more stable than
endoperoxides 2 so that they can be stored as such rather than be
regarded as precursors.
Scheme 4
the present non-stereospecific 1,4-addition of singlet oxygen.
Since no E–Z isomerization of the double bond in pyridone and
dihydrofuran rings can occur for 1, the isomerization of an
intermediary zwitterion as in Scheme 3 can not proceed. The
most likely explanation is that the initially formed peroxirane or
zwitterion attacks the remaining double bond of the starting
1,3-diene from both p-faces. This process should be operable
when two ethylene units of the 1,3-diene lie far from apart in the
same plane as illustrated in Scheme 4 in which a peroxirane
intermediate is adopted for convenience. Such structure of a
1,3-diene system is most likely feasible for 1 because of steric
repulsion between the pyridone ring and a tert-butyl group in
the dihydrofuran ring.§ Although singlet oxygen attacks
possibly in the first step to a double bond of the pyridone and/or
the double bond of dihydrofuran for 1, the initial addition of
singlet oxygen to the former may be more likely since the
addition to the latter may lead more or less to a dioxetane as in
the case of 4-tert-butyl-3,3-dimethyl-2,3-dihydrofurans bearing
an aryl4 or a styryl group5 at the 5-position.
Although endoperoxides cis and trans-2a were not iso-
merized to each other, both endoperoxides changed gradually
into a dioxetane 5a even at room temperature. On heating in hot
benzene, 2a was transformed selectively into 5a (colorless
needles, mp 114.0–115.0 °C)¶ which was stable enough for
handling at room temperature though it decomposed into a
ketoester 6a in hot toluene (Scheme 5). It should be noted that
the thermolysis of 5a (90 °C, toluene) gave light (lmax = 411
nm) whose spectrum was in good agreement with the fluores-
cence spectrum of 6a. Both reaction rates for isomerization of
endoperoxide 2a to 1,2-dioxetane 5a and for decomposition of
dioxetane 5a to ketoester 6a followed first-order kinetics. Thus,
these reaction rates were measured in toluene-d8 at various
Notes and references
† Similar singlet oxygenation of 1a at 0 °C gave a 57+43 mixture of cis and
trans isomers of 2a.
‡ Selected data: for cis-2a: dH(500 MHz, CDCl3) 1.11 (s, 3H), 1.15 (s, 9H),
1.25 (s, 3H), 3.81 (d, J 8.2 Hz, 1H), 4.41 (d, J 15.1 Hz, 1H), 4.45 (d, J 8.2
Hz, 1H), 4.82 (d, J 15.1 Hz, 1H), 5.13 (d, J 2.3 Hz, 1H), 5.59 (d, J 5.7 Hz,
1H), 6.95 (dd, J 5.7 and 2.3 Hz, 1H), 7.16–7.20 (m, 2H), 7.28–7.38 (m, 3H);
dC(125 Hz, CDCl3) 18.1, 24.9, 26.4, 36.5, 45.3, 47.0, 78.1, 80.6, 82.6,
106.0, 113.1, 128.1, 128.2, 129.0, 134.9, 135.1, 138.3, 167.4; MS (m/z, %)
369 (M+, 46), 325 (4), 313 (9), 285 (12), 284 (16), 230 (33), 212 (60), 91
(100).
For trans-2a: dH(500 MHz, CDCl3) 1.11 (s, 3H), 1.12 (s, 9H), 1.28 (s,
3H), 3.79 (d, J 8.2 Hz, 1H), 4.31 (d, J 15.4 Hz, 1H), 4.49 (d, J 8.2 Hz, 1H),
5.06 (d, J 15.4 Hz, 1H), 5.23 (d, J 2.3 Hz, 1H), 5.61 (d, J 5.5 Hz, 1H), 6.93
(dd, J 5.5, 2.3 Hz, 1H), 7.19–7.23 (m, 2H), 7.30–7.39 (m, 3H); dC(125 Hz,
CDCl3) 18.0, 24.8, 26.7, 36.3, 45.3, 46.9, 78.8, 80.7, 83.2, 105.7, 113.4,
128.0, 128.3, 129.1, 134.5, 135.2, 138.2, 167.0; MS (m/z, %) 369 (M+, 47),
313 (10), 285 (13), 284 (17), 230 (35), 212 (63), 91 (100).
§ An MM2 calculation suggested that the torsion angle between the
pyridone ring and the dihydrofuran ring in 1a was 73.4° at the most stable
conformation.
1
temperatures by H NMR spectroscopy, and their activation
¶ Selected data for 5a: dH(500 MHz, CDCl3) 1.08 (s, 9H), 1.13 and 1.31 (2
s, 6H), 3.80 (d, J 8.3 Hz, 1H), 4.54 (d, J 8.3 Hz, 1H), 5.09 (d, J 14.4 Hz, 1H),
5.20 (d, J 14.4 Hz, 1H), 6.32 (dd, J 7.4, 1.8 Hz, 1H), 6.94 (d, J 1.8 Hz, 1H),
7.23–7.38 (m, 6H); dC(125 Hz, CDCl3) 18.2, 24.9, 27.0, 36.7, 45.5, 51.8,
80.7, 105.3, 105.7, 114.7, 121.4, 128.1, 128.1, 129.0, 135.9, 136.7, 147.8,
161.9; MS (m/z, %) 369 (M+, 51), 337 (M+ 2 O2, 2), 313 (11), 230 (39), 212
(71), 185 (15), 91 (100).
1 K. Gollnick and A. Griesbeck, Tetrahedron Lett., 1983, 24, 3303.
2 K. E. O’Shea and C. S. Foote, J. Am. Chem. Soc., 1988, 110, 7167.
3 J. Motoyoshiya, Y. Okuda, I. Matsuoka, S. Hayashi, Y. Takaguchi and H.
Aoyama, J. Org. Chem., 1999, 64, 493.
4 M. Matsumoto, H. Murakami and N. Watanabe, Chem. Commun., 1998,
2319.
5 M. Matsumoto, T. Ishihara, N. Watanabe and T. Hiroshima, Tetrahedron
Lett., 1999, 40, 4571.
6 A. P. Schaap, P. A. Burns and K. A. Zaklika, J. Am. Chem. Soc., 1977, 99,
1270.
Scheme 5
822
Chem. Commun., 2000, 821–822