monoperester 1c with Cs2CO3, but not K2CO3 or KOAc. This
disparity drew our attention to the potential importance of
ion pairing, and we turned to n-Bu4NF (TBAF) as a
convenient base which would afford a highly dissociated
peroxyanion. Gratifyingly, treatment of 1c with TBAF led
to extremely rapid reaction and a 39% yield of 1O2 (as 3-O2).
unable to isolate a monoperoxysulfonate, reaction of 1b and
terpinene (3) with toluenesulfonyl chloride (1.0 equiv) and
KOtBu resulted in the rapid disappearance (TLC) of the
dihydroperoxide and the formation of 3-O2 (Scheme 3).
1
Scheme 3
.
Generation of O2 from 1,1-Dihydroperoxide
Table 2. Protocols for Preparative Oxidation
products
perester (equiv) trap promotera time (h) T (°C) (yield, %)b
1c (3)
1c (2)
1c (4)
1c (1.5)
1c (8)
1c (6)
4
7
3
4
5
6
TBAF c
0.5
0.5
0.5
0.5
1
0
0
rt
rt
rt
rt
4-O2 (91)
NR
TBAFc
The unprecedented fragmentation described above could
involve a Grob-like fragmentation24 or, alternatively, de-
composition of an unstable peroxetane derived from 4-exo-
tet attack of the peroxyanion on the activated peroxide
(Scheme 4).25 Regardless of pathway, the fragmentation
CsF, TMA
CsF, TMA
CsF, TMA
CsF, TMA
3-O2 (69)
4-O2 (94)
5-O2 (81)
6-O2 (76)
7-O2/9-O2/8
(91, 58:28:14)
3-O2 (62)
3-O2 (66)
0.5
1c (8)
1e (3)
2c (3)
7
3
3
CsF, TMA
CsF, TMA
CsF, TMA
0.5
0.5
0.5
rt
rt
rt
a Promoter present in 2.4 equiv relative to monoperester. b Based upon
conversion of trap to product. c THF as solvent.
Scheme 4. Potential Mechanisms
Further exploring the TBAF-promoted reaction (Table 2),
we found that the use of excess TBAF and 1c allowed
consumption of furan 4 but failed to oxidize the less reactive
7. Concerned that overly rapid generation of 1O2 might allow
escape from a saturated solution, we investigated the
decomposition of excess (1.5-8 equiv) monoperester in the
presence of CsF and Me4NOAc (TMA). Reactions were
allowed to run for 30 min, but were typically complete (TLC)
within 10 min. Complete consumption of all substrates was
now observed. Citronellol (7) reacted to furnish a 91% yield
of a 58:28:14 mixture of 7-O2, 9-O2, and ketone 8. The
formation of the isomeric hydroperoxides is characteristic
clearly requires both a highly dissociated peroxyanion and
a peroxide activated toward heterolytic O-O scission. For
example, the monoperesters do not generate oxygen in the
absence of base, while we found the 1,1-dihydroperoxides
to be unaffected by the bases employed in these studies.26
The efficiency of 1O2 production from the new fragmentation
compares very favorably with known oxygen-generating
systems.4,6-8
In conclusion, we have developed a new heterolytic
fragmentation that allows efficient and rapid generation of
1O2 in nondeuterated organic solvents from readily available
precursors. The clean regeneration of the parent ketone
suggests an avenue for possible development of solid-
supported or phase-separable reagents while the efficiency
6,7
1
for reactions of O2 with 7; ketone 8 derives from base-
promoted fragmentation of 7-O2.23 The CsF protocol was
also successfully applied to monoperesters 1e and 2c.
1
Finally, O2 can be generated via in situ formation and
decomposition of monoperoxysulfonates. Although we were
(18) Cudden, R. C. P.; Hewlett, C. J. Chem. Soc. C 1968, 298.
Dihydroperoxide monoesters have been investigated as radical initiators:
Van de Bovenkamp-Bouwman, A. G.; Van Gendt, J. W. J.; Meijer, J.; Hogt,
A. H; Van Swieten, A. P. PCT Int. Appl. WO 9932442 A1 19990701, 1999.
(19) Monoester 1c is not detonated by a hammer blow and melts without
decomposition at 37 °C. It is stable for less than 1 day at 60 °C.
(20) Percarbonates of tertiary hydroperoxides and monoperesters of 1,1-
dihydroperoxides are both known to undergo Criegee rearrangement:
Villenave, J. J.; Filliatre, C.; Maillard, B.; Jaouhari, R. Bull. Soc. Chim.
Belg. 1982, 91, 301. Velluz, L.; Amiard, G.; Martel, J.; Warnant, J. Compt.
Rend. 1957, 244, 1937 We are uncertain as to the basis for the increased
stability of 1c vs 1d.
(23) Kornblum, N.; DelaMare, H. E. J. Am. Chem. Soc. 1951, 73, 880.
(24) Grob, C. A. Angew. Chem., Int. Ed. Engl. 1969, 8, 535.
(25) Although 4-exo/tet displacements by peroxyanions to form diox-
etanes have been observed: Kopecky, K.; Filby, J. E.; Mumford, C.;
Lockwood, P. A.; Ding, J.-Y. Can. J. Chem. 1975, 53, 1103 The
corresponding closure to peroxetanes is unknown. The intermediacy of
peroxetanes could in principle be established by the relative fractions of
18O16O formed upon decomposition of dihydroperoxides derived from
mixtures of H18O18OH and H16O16OH. We thank one of the reviewers for
this suggestion.
(21) Reported yields are based upon either isolation or quantitative GC/
MS of oxidation products relative to an internal standard; see the Supporting
Information for details. In general, the ketone byproduct (1a or 2a) was
recovered in good yield from the decomposition reactions.
(22) Due to the facility of self-sensitized oxidation, the use of DPBF
for quantitative experiments should include control reactions or take care
to exclude light and oxygen. See: Owakowsa, M. J. Chem. Soc., Faraday
Trans. 1 1984, 80, 2119.
(26) 1,1-Dihydroperoxides have been successfully bisalkylated in the
presence of Cs2CO3. Kim, H. S.; Nagai, Y.; Ono, K.; Begum, K.; Wataya,
Y.; Hamada, Y.; Tsuchiya, K.; Masuyama, A.; Nojima, M.; McCullough,
K. J. J. Med. Chem. 2001, 44, 2357–61.
4574
Org. Lett., Vol. 11, No. 20, 2009